EPA Document #740-Rl-8006
September 2020
United States Office of Chemical Safety and
Environmental Protection Agency Pollution Prevention
Risk Evaluation for
Cyclic Aliphatic Bromide Cluster
(HBCD)
CASRN: 25637-99-4
CASRN: 3194-55-6
CASRN: 3194-57-8
September 2020
-------
TABLE OF CONTENTS
TABLE OF CONTENTS 2
ACKNOWLEDGEMENTS 22
ABBREVIATIONS 23
EXECUTIVE SUMMARY 28
1 INTRODUCTION 44
1.1 Physical and Chemical Properties 46
1.2 Uses and Production Volume 47
1.2.1 Data and Information Sources 47
1.2.2 Domestic Manufacture of HBCD 47
1.2.3 Importation of HBCD 50
1.2.4 Toxics Release Inventory Data on HBCD 51
1.2.5 Uses of HBCD 53
1.2.5.1 Automobile Replacement Parts 53
1.2.5.2 Expanded Polystyrene (EPS) and Extruded Polystyrene (XPS) Foam 54
1.2.5.3 Flux/Solder Paste 54
1.2.6 Recycling of EPS and XPS Foam 55
1.2.7 Recycling of Electronics Waste (E-Waste) Containing HIPS 56
1.2.8 Legacy Activities and Uses 56
1.2.9 Historical Activities Resulting in Continued Exposures 57
1.2.10 Summary 57
1.2.11 List of Conditions of Use 58
1.3 Regulatory and Assessment History 58
1.4 Scope of the Evaluation 60
1.4.1 Conditions of Use Included in the Risk Evaluation 60
1.4.2 Exposure Pathways and Risks Addressed by other EPA Administered Statutes 66
1.4.2.1 TSCA Authorities Supporting Tailored Risk Evaluations and Intra-agency Referrals.. 66
1.4.2.2 EPA-administered Statutes and Regulatory Programs that Address Specific Exposure
Pathways and/or Risks 69
1.4.3 Conceptual Models 70
1.5 Systematic Review 76
1.5.1 Data and Information Collection 76
2 EXPOSURES 83
2.1 Fate and Transport 83
2.1.1 Fate and Transport Approach and Methodology 83
2.1.2 Summary of Fate and Transport 84
2.1.2.1 Air 86
2.1.2.2 Water 86
2.1.2.3 Soil and Sediment 87
2.1.2.4 Wastewater Treatment Plants 87
2.1.2.5 Persistence 88
2.1.2.6 Bioaccumulation/Bioconcentration 88
2.1.2.7 PBT Characterization 89
2.1.3 Assumptions and Key Sources of Uncertainty for Fate and Transport 90
Page 2 of 723
-------
2.2 Releases to the Environment 97
2.2.1 Release Assessment Approach and Methodology 98
2.2.2 Repackaging of Import Containers 102
2.2.3 Compounding of Polystyrene Resin to Produce XPS Masterbatch 108
2.2.4 Processing to Produce XPS Foam using XPS Masterbatch 113
2.2.5 Processing of HBCD to Produce XPS Foam 118
2.2.6 Processing to Produce EPS Foam from Imported EPS Resin Beads 123
2.2.7 Processing to Produce SIPs and Automobile Replacement Parts from XPS/EPS Foam .... 129
2.2.8 Use: Installation of Automobile Replacement Parts 134
2.2.9 Use: Installation of XPS/EPS Foam Insulation in Residential, Public, and Commercial
Buildings, and Other Structures 134
2.2.10 Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures 140
2.2.11 Processing: Recycling of EPS Foam and Reuse of XPS Foam 144
2.2.12 Formulation of Flux/Solder Pastes 148
2.2.13 Use of Flux/Solder Pastes 150
2.2.14 Recycling of Electronics Waste (E-Waste) Containing HIPS 156
2.2.15 Sensitivity Analysis - Process Volume 160
2.2.16 Assumptions and Key Sources of Uncertainties for Environmental Releases 166
2.3 Environmental Exposures 168
2.3.1 Approach and Methodology 168
2.3.2 Aquatic Environment - Surface Water and Sediment 173
2.3.2.1 Non-Scenario Specific Approach 173
2.3.2.1.1 Surface Water Concentrations 174
2.3.2.1.2 Sediment Concentrations 175
2.3.2.2 Scenario Specific Approach 176
2.3.2.2.1 E-FAST: Predicted Flowing Surface Water Concentrations (First Tier Modeling)
177
2.3.2.2.2 VVWM-PSC: Predicted Flowing Surface Water Concentrations (Second Tier
Modeling) and Sediment Concentrations 179
2.3.2.2.3 IIOAC: Predicted Pond Water and Sediment Concentrations 183
2.3.3 Terrestrial Environment - Soil 185
2.3.3.1 Non-Scenario Specific Approach - Air Deposition and Biosolid Application 185
2.3.3.2 Scenario Specific Approach - Air Deposition 187
2.3.3.3 Combined Soil Concentration - Air Deposition, Background, Biosolid Application . 189
2.3.4 Assessment of Exposure in Targeted Wildlife 189
2.3.5 Summary of Results for Environmental Exposure Assessment 190
2.3.6 Sensitivity Analysis - Environmental Exposure 192
2.3.6.1 Modeled Sediment 192
2.3.6.2 Monitoring Data (General) 192
2.3.6.3 Fish Tissue 192
2.3.6.4 Scenario Inputs (product amount, WWTR%) 192
2.3.7 Assumptions and Key Sources of Uncertainty in Environmental Exposure Assessment... 196
2.4 Human Exposures 197
2.4.1 Occupational Exposures 200
2.4.1.1 Occupational Exposures Approach and Methodology 200
2.4.1.2 Repackaging of Import Containers 211
Page 3 of 723
-------
2.4.1.3 Compounding of Polystyrene Resin to Produce XPS Masterbatch 215
2.4.1.4 Processing of HBCD to Produce XPS Foam using XPS Masterbatch 221
2.4.1.5 Processing of HBCD to Produce XPS Foam Using HBCD Powder 224
2.4.1.6 Processing of HBCD to Produce EPS Foam from Imported EPS Resin Beads 226
2.4.1.7 Processing of HBCD to Produce SIPs and Automobile Replacement Parts from
XPS/EPS Foam 230
2.4.1.8 Use: Installation of Automobile Replacement Parts 232
2.4.1.9 Use: Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial
Buildings, and Other Structures 232
2.4.1.10 Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public
and Commercial Buildings, and Other Structures 234
2.4.1.11 Recycling of EPS Foam and Reuse of XPS Foam 237
2.4.1.12 Formulation of Flux/Solder Pastes 239
2.4.1.13 Use of Flux/Solder Paste 241
2.4.1.14 Recycling of Electronics Waste (E-Waste) Containing HIPS 242
2.4.1.15 Assumptions and Key Sources of Uncertainties for Occupational Exposures 247
2.4.1.15.1 Number of Workers 247
2.4.1.15.2 Estimation of Inhalation Exposure Concentration and Average Daily Dose 247
2.4.1.15.3 Modeling Dermal Exposures 248
2.4.1.15.4 Occupational Non-User (ONU) Potential Inhalation Exposure 248
2.4.1.15.5 Firefighter Potential Occupational Exposure 248
2.4.1.15.6 Summary of Occupational Exposures 249
2.4.2 General Population (Background) Exposures 255
2.4.2.1 General Population Exposure Approach and Methodology 255
2.4.2.2 Indirect Estimation Using Environmental Monitoring Data and Exposure Factors 256
2.4.2.2.1 Diet — Ingestion 259
2.4.2.2.2 Dust and Soil — Incidental Ingestion 262
2.4.2.2.3 Air — Inhalation 265
2.4.2.2.4 Dermal 267
2.4.2.2.5 Aggregate General Population Exposure and Dose 269
2.4.2.2.6 Occupational Microenvironments 271
2.4.2.3 Exposure Reconstruction Using Human Biomonitoring Data and Reverse Dosimetry272
2.4.2.4 Comparison of General Population Approaches 274
2.4.2.5 General Population Subsistence Fisher Exposures 276
2.4.3 Highly Exposed General Population Exposures 277
2.4.3.1 Approach and Methodology 277
2.4.3.2 Near Facility Dietary (Fish) — Ingestion 278
2.4.3.3 Near Facility Suspended Particulates in Air — Inhalation 283
2.4.3.4 Aggregate Highly Exposed Population Exposure and Dose 288
2.4.4 Consumer Exposures 293
2.4.4.1 Approach and Methodology 293
2.4.4.2 XPS/EPS Insulation In Residences — Indoor Air and Settled Dust 293
2.4.4.3 Automobile Components that Contain HBCD — Indoor Air and Settled Dust 296
2.4.4.4 Recycled Consumer Articles that Contain HBCD — Mouthing 297
Page 4 of 723
-------
2.4.5 Qualitative Exposure Scenarios 299
2.4.5.1 Emissions to Ambient Air from EPS and XPS Insulation in Residences 299
2.4.5.2 HBCD Sent to Landfill Across the Lifecycle 300
2.4.5.3 Occupational Exposure Associated with the Condition of Use of Land Disposal of
Formulated Products and Articles 301
2.4.6 Sensitivity Analysis - Human Exposure 304
2.4.6.1 Sensitivity Analysis - Infant Exposures 304
2.4.6.2 Sensitivity Analysis - Variation in Production Volume 306
2.4.7 Assumptions and Key Sources of Uncertainty in the General Population, Highly Exposed,
and Consumer Exposure Assessment 307
2.4.8 Potentially Exposed or Susceptible Subpopulations 310
3 HAZARDS 313
3.1 Environmental Hazards 313
3.1.1 Approach and Methodology 313
3.1.2 Hazard Identification 313
3.1.2.1 Aquatic Toxicity 315
3.1.2.2 Terrestrial Organisms 319
3.1.3 HBCD Trophic Transfer in the Environment 321
3.1.4 Weight of the Scientific Evidence 328
3.1.5 Concentrations of Concern 329
3.1.6 Summary of Environmental Hazard 331
3.1.7 Assumptions and Key Sources of Uncertainty for the Environmental Hazard Assessment 332
3.2 Human Health Hazards 334
3.2.1 Approach and Methodology 334
3.2.2 Toxicokinetics 337
3.2.2.1 ADME 337
3.2.2.1.1 Absorption 337
3.2.2.1.2 Distribution 338
3.2.2.1.3 Metabolism 339
3.2.2.1.4 Elimination 341
3.2.2.2 Description of Toxicokinetic Models 343
3.2.3 Hazard Identification 344
3.2.3.1 Non-Cancer Hazards 344
3.2.3.1.1 Thyroid Effects 344
3.2.3.1.2 Liver Effects 345
3.2.3.1.3 Reproductive Effects 345
3.2.3.1.4 Developmental Effects 346
3.2.3.1.5 Neurological Effects 346
3.2.3.1.6 Immune System Effects 347
3.2.3.1.7 Overt Toxicity Following Acute/Short Term Exposure 347
3.2.3.1.8 Sensitization/Irritation 347
3.2.3.2 Genotoxicity and Cancer Hazards 348
3.2.4 Weight of the Scientific Evidence 350
Page 5 of 723
-------
3.2.4.1 Non-Cancer Hazards 350
3.2.4.1.1 Thyroid Effects 350
3.2.4.1.2 Liver Effects 353
3.2.4.1.3 Reproductive Effects 353
3.2.4.1.4 Developmental Effects 354
3.2.4.1.5 Neurological Effects 355
3.2.4.1.6 Immune System Effects 356
3.2.4.1.7 Overt Toxicity Following Acute/Short Term Exposures 357
3.2.4.1.8 Sensitization/Irritation 357
3.2.4.2 Genotoxicity/Carcinogenicity 358
3.2.4.3 Summary of Human Health Hazards Used to Evaluate Acute and Chronic Exposures
358
3.2.5 Dose-Response Assessment 358
3.2.5.1 Selection of Studies for Non-Cancer Dose-Response Assessment 358
3.2.5.2 Derivation of Points of Departure and Uncertainty Factors 360
3.2.5.2.1 PODs for Acute Exposure 361
3.2.5.2.2 PODs for Chronic Exposures 364
3.2.5.2.3 Human Equivalent Doses 368
3.2.5.2.4 Uncertainty Factors 369
3.2.5.3 Points of Departure for Human Health Hazard Endpoints 370
3.2.6 Assumptions and Key Sources of Uncertainties for the Human Health Hazard Assessment
377
3.2.6.1 Toxicokinetics 377
3.2.6.2 Human Health Endpoints 377
3.2.7 Potentially Exposed or Susceptible Subpopulations 379
RISK CHARACTERIZATION 381
. 1 Environmental Ri sk 381
4.1.1 Environmental Risk Estimation 384
4.1.1.1 Environmental Effect Levels of HBCD 384
4.1.1.2 Acute and Chronic Concentrations of Concern 384
4.1.2 Calculation of Risk Quotient (RQ) Values for HBCD 386
4.1.3 Risk Estimation Approach 386
4.1.3.1 Risk Estimation Based on HBCD Surface Water and Sediment Concentrations using
Environmental Monitoring Data and Modeling Results 389
4.1.3.1.1 Risk Estimation Based on Surface Water and Sediment Monitoring Data 389
4.1.3.1.2 Risk Estimation Based on Surface Water and Sediment Modeling Data 391
4.1.3.1.3 Risk Estimation for the Recycling of Electronics Waste Containing HIPS 393
4.1.3.2 Risk Estimation based on HBCD Soil Concentrations using Environmental Monitoring
and Modeling Data 393
4.1.3.2.1 Risk Estimation Based on Soil Monitoring Data 393
4.1.3.2.2 Risk Estimation Based on Soil Modeling Data 393
Page 6 of 723
-------
4.1.3.2.3 Risk Estimation for the Recycling of Electronics Waste Containing HIPS 394
4,1,3,3 Risk Estimation based on Exposure via Trophic Transfer 395
4,1.4 Environmental Risk Results 396
4.1.4.1 Risk Characterization for Aquatic and Terrestrial Ecosystems based on Environmental
Monitoring Data 397
4.1.4.2 Risk Characterization for Aquatic and Terrestrial Ecosystems based on Modeled
Surface Water and Sediment Concentrations 398
4.1.4.3 Risk Characterization for Aquatic and Terrestrial Ecosystems based on Exposure via
Potential Trophic Transfer of HBCD 404
4.1.4.4 Targeted Sensitivity Analysis 405
4.1.4.4.1 Summary of Ranges of RQs: Production Volume 406
4.2 Human Health Risk 410
4.2.1 Risk Estimation Approach 410
4,2.1.1 Representative Points of Departure for Use in Risk Estimation 412
4.2.2 Risk Estimation for Workers 413
4.2.2.1 Occupational Risk Estimation for Non-Cancer Effects Following Acute Inhalation
Exposures 416
4.2.2.2 Occupational Risk Estimation for Non-Cancer Effects Following Chronic Inhalation
Exposures 418
4.2.2.3 Occupational Risk Estimation for Non-Cancer Effects Following Acute Dermal
Exposures 422
4.2.2.4 Occupational Risk Estimation for Non-Cancer Effects Following Chronic Dermal
Exposures 423
4.2.2.5 Occupational Risk Estimation for the Recycling of Electronics Waste Containing HIPS
424
4.2.3 Risk Estimation for General Population and Consumers 425
4.2.3.1 General Population Risk Estimation for Non-Cancer Effects - Background Exposure
425
4.2.3.1.1 Occupational Microenvironments 426
4.2.3.2 General Population Risk Estimation for Non-Cancer Effects - Subsistence Fishers .. 427
4.2.3.3 General Population Risk Estimation for Non-Cancer Effects - Highly Exposed
Populations 428
4.2.3.3.1 General Population Risk Estimation for Non-Cancer Effects Following Acute
Exposures - Highly Exposed Populations 429
4.2.3.3.2 General Population Risk Estimation for Non-Cancer Effects Following Chronic
Exposures - Highly Exposed Populations 434
4.2.3.4 Targeted Sensitivity Analysis 438
4.3 Assumptions and Key Sources of Uncertainty for the Risk Characterization 439
4.3.1 Assumptions and Key Sources of Uncertainties for the Environmental Risk Characterization
439
4.3.1.1 Confidence in Risk Estimates 443
4.3.2 Assumptions and Key Sources of Uncertainties for the Human Health Risk Characterization
444
4.3.2.1 Physical-Chemical Properties and Toxicokinetics Considerations 444
4.3.2.2 Human Health Hazard Considerations 445
4.3.2.3 Occupational Exposure Considerations and Confidence Statements 446
4.3.2.4 PPE Considerations 447
Page 7 of 723
-------
4.3.2.5 General Population/Consumer Exposure Considerations and Confidence Statements 447
4.3.2.6 Considerations of Historical Production Volumes and Activities 448
4.4 Other Risk Related Considerations 449
4.4.1 Potentially Exposed or Susceptible Subpopulations 449
4.4.2 Aggregate and Sentinel Exposures 450
4.5 Risk Conclusions 451
4.5.1 Environmental Risk Conclusions 451
4.5.1.1 Summary of Risk Estimates for Aquatic Organisms 451
4.5.1.2 Summary of Risk Estimates for Terrestrial Organisms 453
4.5.2 Human Health Risk Conclusions 459
4.5.2.1 Summary of Risk Estimates for Workers 459
4.5.2.2 Summary of Risk Estimates for General Population and Consumers 464
UNREASONABLE RISK DETERMINATION 468
5.1 Overview 468
5.1.1 Human Health 468
5.1.1.1 Non-Cancer Risk Estimates 469
5.1.1.2 Cancer Risk Estimates 469
5.1.1.3 Determining Unreasonable Risk of Injury to Health 469
5.1.2 Environment 470
5.1.2.1 Determining Unreasonable Risk of Injury to the Environment 471
5.2 Detailed Unreasonable Risk Determinations by Condition of Use 472
5.2.1 Human Health 473
5.2.1.1 Manufacturing - Import - (Import) 473
5.2.1.2 Processing - Incorporated into Formulation, Mixture or Reaction Product - Flame
Retardants used in Custom Compounding of Resin (e.g., compounding in XPS
masterbatch) and in Solder Paste 475
5.2.1.3 Processing - Incorporation into an Article - Flame Retardants used in Plastics Product
Manufacturing (manufacture of XPS and EPS foam; manufacture of structural
insulation panels (SIPS) and automobile replacement parts from XPS and EPS foam)
477
5.2.1.4 Processing - Recycling - Recycling of XPS and EPS Foam, Resin, Panels containing
HBCD 478
5.2.1.5 Processing - Recycling - Recycling of electronics waste containing HIPS that contain
HBCD 479
5.2.1.6 Distribution in Commerce - Distribution - Distribution 480
5.2.1.7 Commercial/Consumer Use - Building/Construction Materials - Plastic Articles (hard)
Construction and Building Materials covering Large Surface Areas (e.g., EPS/XPS
foam insulation in residential, public and commercial buildings, and other structures)
and Solder Paste 480
5.2.1.8 Commercial/Consumer Use - Other - Automobile Replacement Parts and Plastic and
Other Articles 483
5.2.1.9 Commercial/Consumer Use - Other - Formulated Products and Articles 485
5.2.1.10 Disposal - Other Land Disposal (e.g. construction and demolition waste) - Demolition
and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and
Commercial Buildings and Other Structures 486
5.2.1.11 Disposal -Disposal of Formulated Products and Articles 488
5.2.2 Environment 490
5.2.2.1 Manufacturing - Import - (Import) 490
Page 8 of 723
-------
5.2.2.2 Processing - Incorporated into Formulation, Mixture or Reaction Product - Flame
retardants used in Custom Compounding of Resin (e.g., compounding in XPS
masterbatch) and in Solder Paste 491
5.2.2.3 Processing - Incorporation into an Article - Flame Retardants used in Plastics Product
Manufacturing (manufacture of XPS and EPS foam; manufacture of structural
insulation panels (SIPS) and automobile replacement parts from XPS and EPS foam)
492
5.2.2.4 Processing - Recycling - Recycling of XPS and EPS Foam, Resin, Panels Containing
HBCD 492
5.2.2.5 Processing - Recycling - Recycling of Electronics Waste Containing HIPS that
Contain HBCD 493
5.2.2.6 Distribution in Commerce - Distribution - Distribution 493
5.2.2.7 Commercial/Consumer Use - Building/Construction Materials - Plastic Articles (hard)
Construction and Building Materials Covering Large Surface Areas (e.g., EPS/XPS
foam insulation in residential, public and commercial buildings, and other structures)
and Solder Paste 494
5.2.2.8 Disposal - Other Land Disposal (e.g. construction and demolition waste) - Demolition
and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and
Commercial Buildings and Other Structures 494
5.2.2.9 Disposal - Land Disposal of Formulated Products and Articles 495
5.3 Changes to the Unreasonable Risk Determination from Draft Risk Evaluation to Final Risk
Evaluation 496
5.4 Unreasonable Risk Determination Conclusion 496
5.4.1 No Unreasonable Risk Determinations 496
5.4.2 Unreasonable Risk Determinations 497
REFERENCES 498
APPENDICES 537
Appendix A REGULATORY HISTORY 537
A.l Federal Laws and Regulations ...........537
A.2 State Laws and Regulations.... ...538
A.3 International Laws and Regulations 540
Appendix B LIST OF SUPPLEMENTAL DOCUMENTS 541
Appendix C FATE AND TRANSPORT 543
C.l Biodcgradation ...543
C.2 Bioconccntration/Bioaccumulation 545
C.3 Calculation of Lipid Normalized Bioaceumulation Factors for HBCD .548
Appendix D RELEASES TO THE ENVIRONMENT 550
D.l 2017 TRI Releases Not Used in this Assessment....... .....550
D.2 Evaluation of Environmental Release Data Sources ...551
Appendix E OCCUPATIONAL EXPOSURES 556
E.l Inhalation. Monitoring Data Summary....... 556
E.2 Summary of Other Assessment Approaches .....................................563
E.3 Equations for Calculating Acute and Chronic (Non-Cancer) Inhalation Exposures 571
E.4 Sample Calculations for Calculating Acute and Chronic (Non-Cancer) Inhalation Exposure ..575
Page 9 of 723
-------
E.5 Approaches for Estimating Number of Workers..,.....,.....,,,.,.,.,,,,....,.....,.,.,.... ,..,.,576
E.6 Evaluation of Occupational Exposure Data Sources 579
E.7 Data Integration Strategy for Occupational Exposure and Release Data/ Information 588
E.8 Information on the Age of Employed Persons .........593
Appendix F ENVIRONMENTAL EXPOSURES 595
F. 1 Modeled Exposure Scenarios Across Conditions of Use 595
F.l.l Water Releases 595
F.1.2 Air Releases 604
F.2 E-FAST and VVWM-F8C Modeling.. ......610
F.3 IIOAC Modeling ...........614
Appendix G GENERAL POPULATION, HIGHLY EXPOSED AND CONSUMER
EXPOSURES 617
G. 1 Exposure Factors for General Population, Highly Exposed, and Consumer Exposure
Calculations 617
G.2 Scenario Gl: General Population 622
G.3 Scenario 111: Near Facility Dietary (Fish) — Ingestion., 624
G.4 Scenario H2; Near Facility Suspended Particulates in Air — Inhalation 635
G.5 Scenarios CI and C2: Consumer Exposure to XPS/EPS Insulation in Residences and
Automobiles Calculations 646
G.5.1 General Mass Balance Equation Used in IECCU 646
G.5.2 Typical Residential House 647
G.5.3 Typical Passenger Vehicle 647
G.5.4 Estimation of Key Parameters 648
G.5.5 Model Parameters 650
G.5.6 Simulation Results 651
G.5.7 Discussion 654
Appendix H ENVIRONMENTAL HAZARDS 658
H,1 Supplemental Environmental Hazard Information., ...........658
H.2 Calculations Used to Evaluate the Potential Trophic Transfer of HBCD ...658
H.3 KABAM Outputs for Aquatic HBCD Bioaccumulation and Bioconcentration.. ..659
H.3.1 10th Percentile Surface and Pore Water Concentrations 659
H.3.2 50th Percentile Surface and Pore Water Concentrations 662
Appendix I BMD MODELING RESULTS FOR SELECTED PODs 665
I.1 Noncancer Endpoints for BMD Modeling ..................................665
I.1.1 Thyroid Effects 667
1.1.2 Liver Effects 670
1.1.3 Reproductive Effects 674
LI.4 Developmental Effects 676
Appendix J ENVIRONMENTAL RISK 686
J.l Aquatic Environment,..., ,686
J. 1.1 Risk Quotients based on a Production Volume of 100,000 lbs/yr and 0% Removal from
Direct Releases 686
J. 1.1.1 E-FAST Initial Screening for Surface Water Concentrations 686
J. 1.1.2 PSC Predicted Surface Water and Sediment Concentrations 691
Page 10 of 723
-------
J. 1.2 Targeted Sensitivity Analysis 701
J. 1.2.1 Exposure Scenario 1: Repackaging of Import Containers 701
J. 1.2.2 Exposure Scenario 3: Processing of HBCD to produce XPS Foam using XPS
Masterbatch 705
J. 1.2.3 Exposure Scenario 5: Processing of HBCD to Produce EPS Foam from Imported EPS
Resin Beads 710
J.l.2.4 Trophic Transfer: Risk Quotients for Terrestrial Mammals based on KABAM 715
J.1.3 Terrestrial Environment 717
J.1.3.1 IIOAC Predicted Soil Concentrations via Air Deposition 717
Appendix K Human Health Risk 719
K.l Targeted Sensitivity Analysis .719
Appendix L Dermal Absorption Estimate Method Comparison 721
L.l Fraction Absorbed Method As Used in Risk Evaluation...................,.........,..........,.„................721
L.2 Permeability Method ....................................721
L.3 Method Comparison ......722
L.3.1 Occupational Exposure Using Flux 722
L.3.2 General Population Considerations 723
Page 11 of 723
-------
LIST OF TABLES
Table 1-1. Physical and Chemical Properties of HBCD 47
Table 1-2. Production Volume (Manufacture and Import) of HBCD in CDR Reporting Period (2012 to
2015)a 48
Table 1-3. Conditions under Which a Company Must Report to CDR (shaded area applies to HBCD
reporting specifically and "x" indicates broad conditions requiring reporting) 49
Table 1-4. U.S. Volume of Imports of HBCD, 2016 through July 2020 50
Table 1-5. Summary of HBCD TRI Production-Related Waste Managed from 2017-2018 (lbs) 52
Table 1-6. Summary of HBCD TRI Releases to the Environment from 2017-2018 (lbs) 53
Table 1-7. Assessment History of HBCD 59
Table 1-8. Categories and Subcategories of Conditions of Use and Corresponding Exposure Scenario
Included in the Scope of the Risk Evaluation for HBCD a 62
Table 2-1. Summary of Environmental Fate and Transport Properties for HBCD 84
Table 2-2. HBCD Biodegradation Half-Lives Selected for Use in Risk Evaluation 91
Table 2-3. HBCD Biodegradation Half-lives (days) Reported and Representative Half-lives Calculated
Using OPP/NAFTA Guidance 93
Table 2-4. Impact of the Use of the Range of Biodegradation Half-lives (days) Reported and
Representative Half-lives Calculated Using OPP/NAFTA Guidance on PSC-VVWM
Concentration Estimates21 93
Table 2-5. HBCD Bioaccumulation and Bioconcentration Factors Reviewed for Use in the Risk
Evaluation 96
Table 2-6. Summary of HBCD Release Sources During Repackaging of Import Containers 104
Table 2-7. Repackaging of Import Containers - HBCD Release Data Source Evaluation 105
Table 2-8. Input Variables to Equation 2-1 for Repackaging of HBCD Import Containers 105
Table 2-9. Summary of HBCD Releases from Repackaging of Import Containers 107
Table 2-10. HBCD Release Data Reported in the EURAR for XPS Masterbatch Production 109
Table 2-11. Compounding of Polystyrene to Produce XPS Masterbatch Release Data Source Evaluation
110
Table 2-12. Input Variables to Equation 2-1 for XPS Masterbatch Production 110
Table 2-13. Summary of HBCD Releases from XPS Masterbatch Production 112
Table 2-14. HBCD Release Data Reported in the EURAR for Manufacturing of XPS Foam from XPS
Masterbatch 114
Table 2-15. XPS Foam Manufacturing Using XPS Masterbatch Release Data Source Evaluation 115
Table 2-16. Input Variables to Equation 2-1 for XPS Foam Manufacturing Using XPS Masterbatch.. 115
Table 2-17. Summary of HBCD Releases from XPS Foam Manufacturing Using XPS Masterbatch .. 117
Table 2-18. HBCD Release Data Reported in the EURAR for Manufacturing of XPS Foam using HBCD
Powder 119
Table 2-19. Manufacturing of XPS Foam Using HBCD Powder Release Data Source Evaluation 120
Table 2-20. Input Variables to Equation 2-1 for XPS Foam Manufacturing Using HBCD Powder 121
Table 2-21. Summary of HBCD Releases from XPS Foam Manufacturing Using HBCD 122
Table 2-22. Summary of HBCD Releases from XPS Foam Manufacturing Using HBCD from 2017 TRI
Data 122
Table 2-23. Summary of HBCD Releases During Manufacturing of EPS Foam from the 2009 OECD
ESD on Plastics Additives and Standard EPA/OPPT Models 125
Table 2-24. Manufacturing of EPS Foam from Imported EPS Resin Beads Release Data Source
Evaluation 126
Table 2-25. Input Variables to Equation 2-1 for EPS Foam Manufacturing from EPS Resin Beads .... 126
Table 2-26. Summary of HBCD Releases from EPS Foam Manufacturing from EPS Resin Beads 128
Page 12 of 723
-------
Table 2-27. Summary of HBCD Release Sources During the Manufacturing of SIPs and Automobile
Replacement Parts from XPS/EPS Foam 130
Table 2-28. Manufacturing of SIPs and Automobile Replacement Parts from XPS/EPS Foam Release
Data Source Evaluation 131
Table 2-29. Input Variables to Equation 2-1 for the Manufacturing of SIPs and Automobile Replacement
Parts from XPS/EPS Foam 131
Table 2-30. Summary of HBCD Releases from the Manufacturing of SIPs and Automobile Replacement
Parts from XPS/EPS Foam 133
Table 2-31. Particle Generation Factors Reported in the EURAR for Sawing or Cutting of XPS/EPS
Foam Prior to Installation 136
Table 2-32. Summary of HBCD Release Sources During Installation of XPS/EPS Foam Insulation in
Residential, Public, and Commercial Buildings, and Other Structures 136
Table 2-33. Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings,
and Other Structures Release Data Source Evaluation 137
Table 2-34. Input Variables to Equation 2-1 for the Installation of XPS/EPS Foam Insulation in
Residential, Public and Commercial Buildings, and Other Structures 137
Table 2-35. Summary of HBCD Releases from Installation of XPS/EPS Foam Insulation in Residential,
Public and Commercial Buildings, and Other Structures 139
Table 2-36. Particle Generation Factors for the Demolition of XPS and EPS 141
Table 2-37. Demolition of XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings,
and Other Structures Release Data Source Evaluation 142
Table 2-38. Summary of HBCD Releases from Demolition of XPS/EPS Foam Insulation in Residential,
Public and Commercial Buildings, and Other Structures 142
Table 2-39. Summary of HBCD Releases from Demolition of XPS/EPS Foam Insulation in Residential,
Public and Commercial Buildings, and Other Structures 143
Table 2-40. Recycling of EPS Foam Release Data Source Evaluation 145
Table 2-41. Input Variables to Equation 2-1 for the Recycling of EPS Foam 145
Table 2-42. Summary of HBCD Releases from the Recycling of EPS Foam 147
Table 2-43. Formulation of Flux/Solder Pastes Release Data Source Evaluation 149
Table 2-44. Summary of HBCD Releases from Flux/Solder Paste Formulation Sites from 2017 TRI Data
150
Table 2-45. Summary of HBCD Release Sources During Use of Flux and Solder Pastes 153
Table 2-46. Use of Flux and Solder Pastes Release Data Source Evaluation 154
Table 2-47. Input Variables to Equation 2-1 for Use of Flux and Solder Pastes 154
Table 2-48. Summary of HBCD Releases from Use of Flux and Solder Pastes 155
Table 2-49. Values, References for, and Overall Confidence Ratings of Input Variables of Equations
HBCD Release Rate from E-Waste Recycling Sites 159
Table 2-50. Summary of HBCD Releases from Sensitivity Analysis of Repackaging of Import
Containers 162
Table 2-51. Summary of HBCD Releases from Sensitivity Analysis of XPS Foam Manufacturing Using
XPS Masterbatch 163
Table 2-52. Summary of HBCD Releases from Sensitivity Analysis of EPS Foam Manufacturing from
EPS Resin Beads 164
Table 2-53. Overview of Approaches Used in HBCD Environmental Exposure Assessment 170
Table 2-54. Summary of Subscenarios Used Across Conditions of Use for Water Releases of HBCD 171
Table 2-55. Summary of Scenarios Used Across Conditions of Use for Air Releases of HBCD 172
Table 2-56. Summary of Central Tendency and High-End Estimated Surface Water Concentrations from
Monitoring Data 175
Page 13 of 723
-------
Table 2-57. Summary of Central Tendency and High-End Estimated Sediment Concentrations from
Monitoring Data 176
Table 2-58. Receiving Stream Flow Values 178
Table 2-59. Estimated HBCD Surface Water (|ig/L) Concentrations Using E-FAST 178
Table 2-60. Inputs for Modeling HBCD Sediment Concentration using VVWM-PSC 180
Table 2-61. Estimated HBCD Surface Water Concentrations (|ig/L) and Sediment Concentrations
(|ig/kg) Using VVWM-PSC with 50th Percentile 7Q10 Flows 181
Table 2-62. Estimated HBCD Surface Water Concentrations (|ig/L) and Sediment Concentrations
(|ig/kg) Using VVWM-PSC with 10th Percentile 7Q10 Flows 182
Table 2-63. Summary of Annualized Deposition and Estimated Pond Surface Water and Sediment
Concentration from Air Deposition 185
Table 2-64. Summary of Central Tendency and High-End Estimated Soil Concentrations from
Monitoring Data 186
Table 2-65. Summary of Annualized Deposition and Estimated Soil Concentration from Air Deposition
189
Table 2-66. Comparison of Published Literature and Modeling Results for Concentrations of HBCD in
Surface Water, Sediment, and Soil 190
Table 2-67. Sensitivity Analysis of Central Tendency Estimate Assumptions in Monitoring Data 192
Table 2-68. Summary of HBCD Surface Water Concentrations from Sensitivity Analysis: Varying
Production Volume and Waste Water Treatment Removal- Environmental Exposures 195
Table 2-69. Exposure Scenarios Descriptions for Receptor Groups 198
Table 2-70. Summary of Inhalation Exposure Assessment Approaches 203
Table 2-71. A Summary for Each of the 12 Occupational Exposure Scenarios (OESs) 208
Table 2-72. Assigned Protection Factors for Respirators in OSHA Standard 29 CFR Section 1910.134
210
Table 2-73. Inhalation Monitoring Data for Manufacturing of HBCD 213
Table 2-74. Summary of Inhalation Monitoring Data for Handling of HBCD 217
Table 2-75. Summary of Inhalation Monitoring Data for the Manufacture of XPS Foam Using XPS
Masterbatch Containing HBCD 223
Table 2-76. Summary of Inhalation Monitoring Data for Handling of XPS and EPS Foam Containing
HBCD 229
Table 2-77. U.S. Number of Establishments and Employees for Installation of XPS/EPS Foam
Insulation in Residential, Public and Commercial Buildings, and Other Structures 233
Table 2-78. U.S. Number of Establishments and Employees for Formulation of Solder Flux 240
Table 2-79. Inhalation Monitoring Data for HBCD at Electronics Recycling Sites 244
Table 2-80. Acute and Chronic Inhalation Exposure Estimates, Worker Occupational Scenarios a 251
Table 2-81. Acute and Chronic Dermal Exposure Estimates, Worker Occupational Scenarios 252
Table 2-82. Summary of Monitoring Studies Identified and Used in the General Population Exposure
Assessment 258
Table 2-83. Summary of Concentrations and Ingestion Rates Used in General Population Diet Exposure
Estimate 261
Table 2-84. Summary of Dust and Soil Inputs Used in Estimating Dust and Soil Ingestion Dose for
HBCD 265
Table 2-85 Inputs for Estimation of HBCD Inhalation Dose 266
Table 2-86. Age Specific ADD for Dermal Exposure from Dust, Soil, and Materials 269
Table 2-87. General Population Central Tendency HBCD Exposure by Pathway and Age Group -
(mg/kg/day) 269
Page 14 of 723
-------
Table 2-88. General Population Central Tendency Source Contribution by Pathway and Age Group (%
Contribution to Total HBCD Exposure) 270
Table 2-89. General Population High-End HBCD Exposure by Pathway and Age Group (mg/kg/day)270
Table 2-90. General Population High-End Source Contribution by Pathway and Age Group (%
Contribution to Total HBCD Exposure) 270
Table 2-91. Occupational Microenvironments Doses as a Percentage of Aggregate General Population
Exposure 272
Table 2-92. Human HBCD Biomonitoring Data Sets by Population, Type and Number 272
Table 2-93. Measured HBCD Concentrations From Various Species and Locations in (Chen et al. 2011)
276
Table 2-94. Aggregate Central Tendency Exposure Comparison for Subsistence Fishers 277
Table 2-95. Summary of HBCD Fish Concentration Data for Estimating Fish Ingestion Dose 279
Table 2-96. Highly Exposed Group: Range of High-End HBCD Fish Ingestion Dose by Scenario and
Age Group (mg/kg/day) 281
Table 2-97. Highly Exposed Group: Range of Central HBCD Fish Ingestion by Scenario and Age Group
(mg/kg/day) 282
Table 2-98. Overall Summary of HBCD Averaged Indoor and Outdoor Air Concentrations for 12
Emission Scenarios 284
Table 2-99. Highly Exposed Group: Range of HBCD Inhalation Dose by Scenario and Age Group,
Acute Dose Rate (mg/kg/day) 286
Table 2-100. Range of HBCD Inhalation Dose by Scenario and Age Group, Average Daily Dose
(mg/kg/day) 287
Table 2-101. Range of HBCD Aggregate Exposure Acute Dose Rate (mg/kg/day) - Background and
Modeled Fish Dose by Scenario and Age 289
Table 2-102. Range of HBCD Aggregate Exposure Average Daily Dose (mg/kg/day): Background and
Modeled Fish Dose by Scenario and Age 289
Table 2-103. Range of HBCD Aggregate Exposure Acute Dose Rate (mg/kg/day): Background and
Modeled Inhalation Dose by Scenario and Age 290
Table 2-104. Range of HBCD Aggregate Exposure Average Daily Dose (mg/kg/day): Background and
Modeled Inhalation Dose by Scenario and Age 291
Table 2-105. Age Specific ADR Associated with Residential Insulation Scenario CI 295
Table 2-106. Age Specific ADD Associated with Residential Insulation Scenario CI 295
Table 2-107. Age Specific ADR Associated with HBCD in Automobile Component Scenario C2 297
Table 2-108. Age Specific ADD Associated with HBCD in Automobile Component Scenario C2 297
Table 2-109. Estimated Exposure from Mouthing of Articles Containing HBCD 298
Table 2-110. Sensitivity Analysis of Upper End Monitoring Distribution Assumptions in Monitoring
Data 306
Table 2-111. Sensitivity Analysis of Central Tendency Estimate Assumptions in Monitoring Data .... 306
Table 2-112. Summary of Surface Water Concentrations from Sensitivity Analysis: Varying HBCD
Production Volume and Waste Water Treatment Removal -Human Exposures (Fish
Ingestion) 307
Table 2-113. Qualitative Assessment of the Uncertainty and Variability Associated with General
Population Assessment 308
Table 2-114. Qualitative Assessment of the Uncertainty and Variability Associated with Highly Exposed
Population Assessment 309
Table 3-1. Environmental Hazard Characterization of HBCD to Aquatic and Terrestrial Organisms .. 314
Page 15 of 723
-------
Table 3-2. Potential Trophic Transfer of HBCD in Aquatic and Terrestrial Ecosystems Using the U.S.
EPA Final Water Quality Guidance for Great Lakes System and U.S. EPA Wildlife
Exposure Factors Handbook 325
Table 3-3. Potential Trophic Transfer of HBCD in Aquatic and Terrestrial Ecosystems using the ECHA
Guidance on Information Requirements and Chemical Safety Assessment (Environmental
Exposure Assessment) 327
Table 3-4. Concentrations of Concern (COCs) for Aquatic Toxicity 330
Table 3-5. Terrestrial Effect Concentrations (Hazard) used to Evaluate Toxicity to Terrestrial Organisms
330
Table 3-6. Study Design Features of Developmental Toxicity Studies 362
Table 3-7. Study Design Features of Studies that Examined T4 Levels 365
Table 3-8. Study Design Features of Studies that Examined Liver Weight 367
Table 3-9. Summary of BMD Modeling Results and Derivation of HEDs for HBCD 370
Table 3-10. PODs and Benchmark MOEs for Effects Following Acute Exposure to HBCD 373
Table 3-11 PODs and Benchmark MOEs for Effects Following Chronic Exposure to HBCD 374
Table 3-12. PODs Selected for Risk Estimation for Each Target Organ/System 376
Table 4-1. Concentrations of Concern (COCs) Derived to Evaluate Toxicity to Aquatic Organisms for
HBCD 385
Table 4-2. Hazard Effect Concentrations used to Evaluate Toxicity to Terrestrial Organisms 385
Table 4-3. Calculated Risk Quotients based on HBCD Surface Water (|ig/L) Concentrations as Reported
in Environmental Monitoring Studies 389
Table 4-4. Calculated Risk Quotients based on HBCD Sediment Concentrations (|ig/kg) as Reported in
Environmental Monitoring Studies 390
Table 4-5. Range of Risk Quotients for Modeled Surface Water and Sediment HBCD Concentrations for
Each Condition of Use Using a Production Volume of 100,000 lbs/yr (0% removal for
direct release) 391
Table 4-6. Calculated Risk Quotients based on HBCD Soil Concentrations (|ig/kg) as Reported in
Environmental Monitoring Studies 393
Table 4-7. Calculated Risk Quotients based on HBCD Soil Concentrations (|ig/kg) as Reported in
Environmental Monitoring Studies and Calculated using Modeling Data 394
Table 4-8. Calculated Risk Quotients based on Potential Trophic Transfer of HBCD in Aquatic and
Terrestrial Ecosystems Using the U.S. EPA Final Water Quality Guidance for Great
Lakes System and U.S. EPA Wildlife Exposure Factors Handbook 395
Table 4-9. Calculated Risk Quotients based on Potential Trophic Transfer of HBCD in Aquatic and
Terrestrial Ecosystems using the ECHA Guidance on Information Requirements and
Chemical Safety Assessment (Environmental Exposure Assessment) 396
Table 4-10. Range of Risk Quotients for Modeled Surface Water and Sediment HBCD Concentrations
for Three Conditions of Use Scenarios Using a Production Volume of 100,000, 50,000,
and 25,000 lbs/yr 407
Table 4-11. Use Scenarios, Populations of Interest and Toxicological Endpoints Used for Acute and
Chronic Exposures 410
Table 4-12. Most Sensitive Endpoints From Each Health Domain Used for Risk Estimation 413
Table 4-13. Inhalation Exposure Data Summary and Respirator Use Determination 414
Table 4-14. Risk Estimation for Non-Cancer Effects Following Acute Inhalation Exposures 417
Table 4-15. Risk Estimation for Non-Cancer Effects Following Chronic Inhalation Exposures 419
Table 4-16. Risk Estimation for Non-Cancer Effects Following Acute Dermal Exposures 422
Table 4-17. Risk Estimate for Workers - Non-Cancer Effects Following Chronic Dermal Exposures 423
Table 4-18. Risk Estimates for Recycling of Electronics Waste Containing HIPS 424
Page 16 of 723
-------
Table 4-19. General Population Risk Estimation for Non-Cancer Effects - Background Exposure 426
Table 4-20. General Population Risk Estimation for Non-Cancer Effects - Subsistence Fishers 427
Table 4-21. Risk Estimation for Non-Cancer Effects Following Acute Exposure to Highly Exposed
Population - Fish Ingestion 430
Table 4-22. Risk Estimation for Non-Cancer Effects Following Acute Exposure to Highly Exposed
Population - Inhalation 432
Table 4-23. Risk Estimation for Non-Cancer Effects Following Acute Exposure to Highly Exposed
Populations - Consumer Articles 433
Table 4-24. Risk Estimation for Non-Cancer Effects Following Chronic Exposure to Highly Exposed
Population - Fish Ingestion 436
Table 4-25. Risk Estimation for Non-Cancer Effects Following Chronic Exposure to Highly Exposed
Populations - Consumer Articles 438
Table 4-26. Summary of Risk for Aquatic Organisms 454
Table 4-27. Occupational Risk Summary Table 461
Table 4-28. Highly Exposed General Population/Consumer Risk Summary Table 465
Table 4-29. Risk Summary for Additional PESS Groups 467
LIST OF FIGURES
Figure 1-1. HBCD Life Cycle Diagram 65
Figure 1-2. HBCD Conceptual Model for Industrial and Commercial Activities and Uses: Worker and
Occupational Non-User Exposures and Hazards 72
Figure 1-3. HBCD Conceptual Model for Consumer Activities and Uses: Consumer Exposures and
Hazards 73
Figure 1-4. HBCD Conceptual Model for Environmental Releases and Wastes: General Population
Exposures and Hazards 74
Figure 1-5. HBCD Conceptual Model for Environmental Releases and Wastes: Ecological Exposures
and Hazards 75
Figure 1-6 HBCD Literature Flow Diagram for Environmental Fate and Transport Data Sources 78
Figure 1-7. HBCD Literature Flow Diagram for Environmental Releases and Occupational Exposure
Data Sources 79
Figure 1-8. Literature Flow Diagram for General Population, Consumer and Environmental Exposure
Data Sources for HBCD 80
Figure 1-9. Literature Flow Diagram for Environmental Hazard Data Sources for HBCD 81
Figure 1-10. Literature Flow Diagram for Human Health Hazard Key/Supporting Data Sources for
HBCD 82
Figure 2-1. Abiotic Reduction of HBCD to 5,6,9,10-tetrabromocyclododec-l-ene (TBCD), 87
Figure 2-2. Overview of receptor groups considered within the Risk Evaluation 198
Figure 2-3 Overview of General Population Exposure Assessment 255
Figure 2-4 Two Exposure Assessment Approaches used to Estimate General Population Exposure to
HBCD 256
Figure 2-5. Source Contribution by Pathway for Aggregate General Population Exposures 271
Figure 2-6. Comparison of HBCD Exposure via Environmental Monitoring/Exposure Factor and Human
Biomonitoring/Reverse Dosimetry Approaches 275
Figure 2-7. Overview of exposure assessment method for highly exposed scenarios 278
Figure 2-8. Comparison of Potential HBCD Fish Ingestion Dose based on Modeled Surface Water
Concentrations, Fish Tissue Monitoring Data, and Surface Water Monitoring Data 283
Figure 2-9. Overview of exposure assessment method for consumer exposure scenarios 293
Page 17 of 723
-------
Figure 2-10. Comparison of HBCD Dose for Infants in the General Population from Different
Sensitivity Analyses 305
Figure 3-1. EPA Approach to Hazard Identification, Data Integration, and Dose-Response Analysis for
HBCD 335
Figure 3-2 Proposed Pathways for Metabolism of HBCD in Rats 341
LIST OF APPENDIX TABLES
Table_Apx A-l. Federal Laws and Regulations 537
Table_Apx A-2. State Laws and Regulations 538
Table_Apx A-3. International Laws and Regulations 540
Table_Apx D-l. 2017 TRI Data Not Used in this Assessment 550
Table_Apx D-2. Summary of Release Data and Systematic Review Results 552
TableApx E-l. Inhalation Monitoring Data for Handling of HBCD 557
TableApx E-2. Inhalation Monitoring Data For Handling of XPS and EPS Foam Containing HBCD561
Table Apx E-3. Summary of HBCD Occupational Inhalation Exposure Assessment Results and the
Associated Assessment Basis and Assessment Approach that are Reported in EU (2008)
564
Table Apx E-4. Summary of HBCD Occupational Exposure Assessment Results and the Associated
Assessment Basis and Approach that are Reported in NICNAS (2012) 566
Table Apx E-5. Summary of Approaches from Other Risk Assessment Reports (RARs) 568
Table Apx E-6 Parameter Values for Calculating Inhalation Exposure Estimates 571
Table_Apx E-7. Overview of Average Worker Tenure from U.S. Census SIPP (Age Group 50+) 574
Table_Apx E-8. Median Years of Tenure with Current Employer by Age Group 574
Table Apx E-9. SOCs with Worker and ONU Designations for All Exposure scenarios 576
Table Apx E-10. Estimated Number of Potentially Exposed Workers and ONUs under NAICS 325991
577
Table Apx E-l 1. Summary of Inhalation Monitoring Data and Systematic Review Results 580
Table Apx E-12. Hierarchy guiding integration of occupational exposure data/information 591
Table Apx E-13. Hierarchy guiding integration of environmental release data/information 592
Table Apx E-14. Percentage of Employed Persons by Age, Sex, and Industry Sector 593
TableApx E-15. Percentage of Employed Persons Age 16-19 Years by Detailed Industry Sector 593
TableApx F-l. Scenarios Used Across Conditions of Use for Water Releases of HBCD 595
Table Apx F-2. Scenarios Used Across Conditions of Use for Air Releases of HBCD 604
Table Apx F-3. Estimated HBCD Surface Water (|ig/L) Concentrations Using E-FAST 611
Table Apx F-4. Total Annual Particle Deposition from Facility Air Releases 614
Table Apx F-5. Estimated Soil Concentrations from Facility Air Releases 615
Table_Apx G-l. Body Weight by Age Group 617
Table Apx G-2. Central Tendency (Mean) Dietary Ingestion Rates by Age Group- Fruit, Vegetables,
Grains, Meats, Dairy, Fats, Consumers Only 617
Table Apx G-3. High-end (95th Percentile) Dietary Ingestion Rates (Consumers Only) by Age Group-
Fruit, Vegetables, Grains, Meats, Dairy, Fats 618
Table_Apx G-4. Fish Ingestion Rates by Age Group 618
Table_Apx G-5. Breastmilk Ingestion Rates 618
Table_Apx G-6. Inhalation Rate by Age Group 619
Table_Apx G-l. Dust and Soil Ingestion Rate by Age Group 619
Table_Apx G-8. Generic Activity Patterns for Time Spent While Awake 619
Table_Apx G-9. Generic Activity Patterns for Time Spent in a Day (24 hours) 620
Table_Apx G-10. Surface Area to Body Weight Ratios (cm2/kg) By Age Group a 620
Page 18 of 723
-------
TableApx G-l 1. Dermal Adherence Factors for Dust By Age Group 621
TableApx G-12. Dermal Adherence Factors for Soil By Age Group 621
Table_Apx G-l3. Surface Area of Object Mouthed (cm2) 622
Table_Apx G-14. Hourly Mouthing Duration (min) 622
Table Apx G-l5. Estimated Average Daily Dose (ADD) by Age Group for Diet 622
Table Apx G-16. Percent of Dietary ADD by Food Group 623
Table Apx G-17. Estimated Average Dose Rate (ADR) by Age Group for Diet 623
Table Apx G-18. Percent of Dietary ADR by Food Group 624
Table Apx G-19. Surface Water Concentrations from VVWM-PSC Modeling and Calculated Fish
Tissue Concentrations 624
Table Apx G-20. Highly Exposed Acute Dose Rate and Average Daily Doses (mg/kg/day) for Modeled
Fish Ingestion Only 627
Table_Apx G-21. Highly Exposed Aggregate Acute Dose Rate and Average Daily Doses (mg/kg/day)
for Modeled Fish Ingestion and Background 631
Table_Apx G-22. Highly Exposed Acute Dose Rate (mg/kg/day) for Modeled Air Only 636
Table_Apx G-23. Highly Exposed Average Daily Dose (mg/kg/day) for Modeled Air Only 638
Table_Apx G-24. Highly Exposed Aggregate Acute Dose Rate (mg/kg/day) for Modeled Air and Non-
Air and Background 641
Table_Apx G-25. Highly Exposed Aggregate Average Daily Dose (mg/kg/day) for Modeled Air and
Non-Air and Background 644
Table_Apx G-26. Zone Names, Volumes, and Baseline Ventilation Rates 647
Table_Apx G-27. Parameters for the HBCD sources 650
Table_Apx G-28. Parameters for the HBCD sinks 650
Table_Apx G-29. Parameters for Settled Dust 651
Table Apx G-30. Mass Balance Results for HBCD in the Simulated Home at 100 Elapsed Days 653
Table Apx G-31. Parameters Used in Comparing EPS and XPS Foams 655
Table Apx 1-1. Noncancer Endpoints Selected for Dose-response Modeling for HBCD 665
Table Apx 1-2. Summary of BMD modeling results for T4 in F0 parental male CRL Sprague-Dawley
rats exposed to HBCD by diet for 18 weeks (Ema et al. 2008); BMR = 10% RD from
control mean 667
Table Apx 1-3. Summary of BMD modeling results for relative liver weight (g/100 g BW) in male F1
CRL rats exposed to HBCD on GD 0-PND 26, dose TWA gestation through lactation
(Ema et al. 2008); BMR = 10% RD from control mean and 1 SD change from control
mean 670
Table Apx 1-4. Summary of BMD modeling results for relative liver weight (g/100 g BW) in male CRL
Sprague-Dawley rats exposed to HBCD by gavage for 13 weeks (WIL Research 2001);
BMR = 10% RD from control mean and 1 SD change from control mean 672
Table Apx 1-5. Summary of BMD modeling results for relative liver weight (g/100 g BW) in female
CRL Sprague-Dawley rats exposed to HBCD by gavage for 13 weeks (WIL Research
2001); BMR = 10% RD from control mean and 1 SD change from control mean 673
Table Apx 1-6. Summary of BMD modeling results for primordial follicles in F1 parental female CRL
Sprague-Dawley rats exposed to HBCD by diet for 18 weeks (Ema et al. 2008); BMR =
1% RD from control mean, 5% RD from control mean, and 10% RD from control mean
674
Table Apx 1-7. Summary of BMD modeling results for offspring loss from PND 4 through PND 21 in
F2 offspring CRL Sprague-Dawley rats; lactational doses of F1 dams (Ema et al. 2008);
BMR = 1% ER and 5% ER 676
Page 19 of 723
-------
TableApx 1-8. Summary of BMD modeling results for pup weight during lactation in F2 male offspring
CRL Sprague-Dawley rats (PND 21) exposed to HBCD by diet for 3 weeks, lactational
dose (Ema et al. 2008); BMR = 5% RD from control mean, 10% RD from control mean,
0.5 SD change from control mean, and 1 SD change from control mean 680
Table Apx 1-9. Effect of Dose on Potential Litter-Specific Covariates 682
Table Apx 1-10. Summary of BMD modeling results for delayed eye opening F2 female offspring CRL
Sprague-Dawley rats (PND 14); F2 generation doses (Ema et al. 2008); BMR = 5% ER
and 10% ER 683
Table Apx 1-11. Summary of BMD modeling results for delayed eye opening F2 female offspring CRL
Sprague-Dawley rats (PND 14); F2 generation doses (Ema et al. 2008); BMR = 5% ER
and 10% ER 685
Table Apx J-l. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations
(|ig/L) Using E-FAST (0% Removal) 686
Table Apx J-2. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations
(|ig/L) Using PSC (0%> Removal) 691
Table Apx J-3. Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (|ig/kg)
Using PSC (0%> Removal) 696
Table Apx J-4. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations
(|ig/L) Using PSC (Targeted Sensitivity Analysis Parameter: Production Volume) 701
Table Apx J-5. Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (|ig/kg)
Using PSC (Targeted Sensitivity Analysis Parameter: Production Volume) 703
Table Apx J-6. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations
(|ig/L) Using PSC (Targeted Sensitivity Analysis Parameters: Production Volume).... 705
Table Apx J-7. Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (|ig/kg)
Using PSC (Targeted Sensitivity Analysis Parameters: Production Volume) 707
Table Apx J-8 Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations
(|ig/L) Using PSC (Targeted Sensitivity Analysis Parameters: Production Volume).... 710
Table Apx J-9 Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (|ig/kg)
Using PSC (Targeted Sensitivity Analysis Parameters: Production Volume) 713
Table Apx J-10. Chemical Properties: Input Parameters for KABAM (vl) based on Estimated HBCD
Surface Water and Sediment Concentrations (|ig/kg) Using PSC 715
Table Apx J-l 1. HBCD Hazard Data: Input Parameters for KABAM (vl) 715
Table Apx J-12. Calculated Risk Quotients based on KABAM (vl) based on Estimated HBCD Surface
Water and Sediment Concentrations (|ig/kg) Using PSC 716
Table Apx J-13. Calculated Risk Quotients based on Estimated HBCD Soil Concentrations (|ig/kg)
Using IIO AC 717
Table Apx K-l. Targeted Sensitivity Analysis Based on Production Volume for the Highly Exposed
Population Following Acute Exposure 720
Page 20 of 723
-------
LIST OF APPENDIX FIGURES
FigureApx F-l. Depiction of the Chemical Processes in the Point Source Calculator 611
FigureApx G-l. The three-zone configuration for a generic residential setting and baseline ventilation
and interzonal air flows 647
Figure Apx G-2. Predicted Gas-phase HBCD Concentration in Living Area 651
Figure Apx G-3. Predicted HBCD Concentration in Airborne PM in Living Area 652
Figure Apx G-4. Predicted HBCD Concentration in Settled Dust 652
Figure Apx G-5. Predicted HBCD Emission Rates from Polystyrene Foam Boards in Attic and
Crawlspace 653
Figure Apx G-6. Rate of HBCD Sorption by Gypsum Board Walls 653
Figure Apx G-l. Predicted HBCD Concentrations in Vehicle's Cabin 654
Figure Apx G-8. Predicted HBCD Concentrations in the Settled Dust in Vehicle's Cabin. The Dust
Contained no HBCD Initially 654
Figure Apx G-9. Simulated HBCD Concentrations with Different Solid-phase Diffusion Coefficients
656
Figure_Apx G-10. Comparison of Normalized Emission Rates 657
Figure Apx 1-1. Plot of mean response by dose, with fitted curve for Exponential 4 Model, for T4 in FO
parental CRL Sprague-Dawley male rats exposed to HBCD by diet for 18 weeks (Ema et
al. 2008) 668
Figure Apx 1-2. Plot of mean response by dose with fitted curve for Exponential (M4) model with
constant variance for relative liver weight (g/100 g BW) in F1 weanling male CRL
Sprague-Dawley rats exposed to HBCD on GD 0-PND 26, dose TWA gestation through
lactation (Ema et al. 2008) 670
Figure Apx 1-3. Plot of mean response by dose, with fitted curve for Exponential M4, for primordial
follicles in F1 parental female CRL Sprague-Dawley rats exposed to HBCD by diet for
18 weeks (Ema et al. 2008) 675
Figure Apx 1-4. Plot of incidence rate by dose, with fitted curve for the nested logistic model where the
litter specific covariate was not used and the intra-litter correlations were estimated, for
incidence of offspring loss from PND 4 through PND 21 in F2 offspring CRL Sprague-
Dawley rats; lactational doses of F1 dams (Ema et al. 2008) 677
Figure Apx 1-5. Plot of mean response by dose with fitted curve for Exponential (M4) model with
constant variance for pup weight during lactation in F2 male offspring CRL Sprague-
Dawley rats (PND 21) exposed to HBCD by diet multigenerationally, lactational dose
(Ema et al. 2008) 681
Figure Apx 1-6 and Figure Apx 1-7. Plot of mean response by dose with fitted curve for Frequentist
Nested Logistic Model without litter-specific covariate and with intra-litter correlation;
and 0.95 Lower Confidence Limit for the BMDL in F2 female offspring CRL Sprague-
Dawley rats (PND 14) exposed to HBCD multigenerationally (Ema et al. 2008). Plots
display results for BMRs of 10% and 5% ER, respectively 684
Figure Apx 1-8. Plot of mean response by dose with fitted curve for Frequentist Nested Logistic Model
without litter-specific covariate and with intra-litter correlation; and 0.95 Lower
Confidence Limit for the BMDL in F2 male offspring CRL Sprague-Dawley rats (PND
14) exposed to HBCD multigenerationally (Ema et al. 2008). Plot displays results for
BMRs of 10% and 5% ER 685
Figure Apx L-l. Excerpt of Dermal Exposure Results from Repackaging of Import Containers 721
-------
ACKNOWLEDGEMENTS
This report was developed by the United States Environmental Protection Agency (U.S. EPA), Office of
Pollution Prevention and Toxics (OPPT), Office of Chemical Safety and Pollution Prevention (OCSPP).
Acknowledgements
The OPPT Assessment Team gratefully acknowledges participation or input from intra-agency
reviewers that included multiple offices within EPA, inter-agency reviewers that included multiple
Federal agencies, and assistance from EPA contractors GDIT (Contract No. CIO-SP3,
HHSN316201200013W), ERG (Contract No. EP-W-12-006), Versar (Contract No. EP-W-17-006), ICF
(Contract No. EPC14001), and SRC (Contract No. EP-W-12-003). The human health hazard section was
developed as a cooperative effort between OPPT and the Office of Research and Development (ORD).
EPA also gratefully acknowledges the contribution of individual animal data on physical development
from (Ema et at. 2008) provided by Dr. Sakiko Fujii from the Safety Research Institute for Chemical
Compounds Co. Ltd. In Japan.
Docket
Supporting information can be found in the public docket: EPA-HQ-OPPT-2019-0237.
Disclaimer
Reference herein to any specific commercial products, process or service by trade name, trademark,
manufacturer or otherwise does not constitute or imply its endorsement, recommendation or favoring by
the United States Government.
Authors
Stan Barone (Deputy Division Director), Nhan Nguyen (Management Lead), Karen Eisenreich
(Management Lead), James Bressette (Staff Lead), Eva Wong (Staff Lead), Kathy Anitole, Sarah Au,
Charles Bevington (formerly with EPA), Dan DePasquale, Majd El-Zoobi, Zaida Figueroa, Whitney
Hollinshead, Keith Jacobs, Amuel Kennedy, Tim Lehman, David Lynch, Chantel Nicolas, Sue Slotnick,
Jennifer Wills (formerly with EPA), Liang Zhang
Contributors
The human health hazard section was developed in collaboration with EPA's Office of Research and
Development (ORD). The hazard section improved and expanded upon a draft IRIS assessment. The
IRIS assessment has been discontinued, and a new/updated assessment will not be added to the IRIS
database at this time (https://cfpub.epa.gov/ncea/iris2/chernicalLanding.cfrn?&substance nrobr=1035).
EPA updated the original draft based on TSCA risk assessment practices, incorporating results of
systematic review and relying on the best available science.
The IRIS assessment team for HBCD included April Luke, MS (Assessment Manager), Kathleen
Newhouse, MS (Assessment Manager), Laura Dishaw, Ph.D. (Co-assessment Manager), Xabier
Arzuaga, Ph.D., Christine Cai, MS, Geniece Lehmann, Ph.D., Zheng (Jenny) Li, Ph.D., and James
Andre Weaver, Ph.D.
EPA would also like to acknowledge contributions by visiting scholar from Japan, Asako Hotta.
Page 22 of 723
-------
ABBREVIATIONS
°c
Degrees Celsius
7Q10
Lowest expected weekly flow over a ten-year period
atm
Atmosphere(s)
AAD
Acute Absorbed Dose
ACC
American Chemistry Council
ADC
Average Daily Concentration
ADME
Absorption, Distribution, Metabolism, and Excretion
ADR
Acute Dose Rate
AERMOD
AMS (American Meteorological Society)/EPA Regulatory Model
AF
Assessment Factor
AIC
Akaike Information Criterion
AIHA
American Industrial Hygiene Association
ALT
Atlanine Aminotransferase
APF
Assigned Protection Factors
AT SDR
Agency for Toxic Substances and Disease Registry
AUC
Area Under the Curve
BAF
Bioaccumulation Factor
BALF
Bronchoalveolar lavage fluid
BCF
Bioconcentration Factor
BDE209
3,3',4,4',5,'5',6,6'-decabromodiphenyl ether
bdwt
Body Weight
BLS
Bureau of Labor Statistics
BMD
Benchmark Dose Modeling
BMDL
Lower Confidence limit on the BMD
BMR
Benchmark Response
BW3'4
Body Weight Scaling to the 3/4 Power
C&D
Construction and Demolition
CAA
Clean Air Act
CAD
Chronic Absorbed Dose
CASRN
Chemical Abstracts Service Registry Number
CBI
Confidential Business Information
CCL
Candidate Contaminant List
CDR
Chemical Data Reporting
CDT
1,5,9-cyclodecatriene
CEPA
The Center for European Policy Agency
CFR
Code of Federal Regulations
CHAD
Consolidated Human Activity Database
COC
Concentration of Concern
COU
Condition of Use
CPSC
Consumer Product Safety Commission
CSCL
Chemical Substance Control Law
CT
Central Tendency
DAF
Dosimetric Adjustment Factor
Page 23 of 723
-------
DBCD 9,10-dibromocyclododeca-1,5-diene
DEE Data Extraction and Evaluation
DMF Dimethylformamide
DMSO Dimethyl sulfoxide
Dwt Dry weight
EASE Estimation and Assessment of Substance Exposure
EC European Commission
EC50 Median Effective Concentration
ECB European Chemicals Bureau
ECHA European Chemicals Agency
EC/HC Environment Canada / Health Canada
ECOTOX ECOTOXicology knowledgebase
E-FAST Exposure and Fate Assessment Screening Tool
EINECS European Inventory of Existing Commercial Substances
EPCRA Emergency Planning and Community Right-to-Know Act
EPS Expanded Polystyrene
EPS-IA Expanded Polystyrene Industry Alliance
ER Extra Risk
ESD Emission Scenario Document
EU European Union
EURAR European Union Risk Assessment Report
FR Federal Register
FOB Functional Occupational Battery
g Gram(s)
GI tract Gastrointestinal tract
GM Geometric Mean
GS Generic Scenario
GSH Glutathione
GST Glutathione-S-transferase
HAP Hazardous Air Pollutant
HBCD/HBCDD Hexabromocyclododecane
HED Human Equivalent Dose
HERO Health and Environmental Research Online
HE High-End
HIPS High Impact Polystyrene
HPLC High Performance Liquid Chromatography
HQ Headquarters
hr Hour
IECCU Indoor Environmental Concentrations in Buildings with Conditioned and Unconditioned
Zones
IIOAC Integrated Indoor-Outdoor Air Calculator
Ind Industrial
KABAM KOW(based) Aquatic BioAccumulation Model
KLH Keyhole limpet Hemocyanin
kg Kilogram(s)
Page 24 of 723
-------
Koa Octanol: Air Partition Coefficient
L Liter(s)
lb Pound
LADC Lifetime Average Daily Concentration
LCD Liquid-Crystal Display
LC/MS Liquid Chromatography-Mass Spectrometry
LOQ Level of Quantitation
LOAEL Lowest Observed Adverse Effect Level
LOEC Lowest Observed Effect Concentration
Log Koc Logarithmic Organic Carbon:Water Partition Coefficient
Log Kow Logarithmic Octanol:Water Partition Coefficient
LPO Lipid Peroxidation
m3 Cubic Meter(s)
MATC Maximum Acceptable Toxicant Concentration
MFG Manufacture
MLD Million Liters per Day
mmHg Millimeter(s) of Mercury
MO A Mode of Action
MOE Margin of Exposure
MOEJ Ministry of Environment Government in Japan
MSW Municipal Solid Waste
MSWLF Municipal Solid Waste Landfills
MT Metric Tons
N/A Not Applicable
NAICS North American Industry Classification System
ND No Data
NICNAS National Industrial Chemicals Notification and Assessment Scheme
NIOSH National Institute for Occupational Safety and Health
NITE National Institute of Technology and Evaluation
NOAEL No Observable Adverse Effect Level
NOEC No Observed Effect Concentration
NR Not Reported
NRC National Research Council
OARS Occupational Alliance for Risk Science
OECD Organisation for Economic Co-operation and Development
OEL Occupational Exposure Limits
OES Occupational Exposure Scenario
ONU Occupational Non-User
OPPT Office of Pollution Prevention and Toxics
OSHA Occupational Safety and Health Administration
P Persistence
P&CB Public and Commercial Buildings
PAPR Power Air-Purifying Respirator
PBDE Polybrominated Diphenyl Ether
PBPK/PD Physiologically Based Pharmacokinetic / Pharmacodynamic
Page 25 of 723
-------
PBZ
Personal Breathing Zone
PDM
Probabilistic Dilution Model
PEC
Predicted Environmental Concentration
PECO
Populations, Exposures, Comparators and Outcomes
PESO
Pathways and Processes, Exposure, Setting or Scenario, and Outcomes
PESS
Potentially Exposed or Susceptible Subpopulations
PM
Particulate Matter
PND
Post-Natal Day
PNOR
Particles Not Otherwise Regulated
POD
Point of Departure
POPs
Stockholm Convention on Persistent Organic Pollutants
POTW
Publicly Owned Treatment Works
ppm
Part(s) per Million
PQL
Practical Quantitation Limit
PTF
Post Fertilization
PV
Production Volume
QC
Quality Control
RAR
Risk Assessment Report
RCRA
Resource Conservation and Recovery Act
RD
Relative Deviation
REACH
European Union's Registration, Evaluation, Authorisation and Restriction of Chemicals
RESO
Receptors, Exposures, Setting or Scenario, and Outcomes
ROS
Reactive Oxygen Species
SAR
Supplied-Air Respirator
SCBA
Self-Contained Breathing Apparatus
SCCH
Stockholm Convention Clearing House
SD
Standard Deviation
SHGB
Sex Hormone Binding Globulin
SIAP
Screening Information Dataset Initial Assessment Profile
SIC
Standard Information Panels
SIDS
Screening Information dataset
SIPS
Structural Insulated Panels
site-yr
Site-year
SNUN
Significant New Use Notice
SNUR
Significant New Use Rule
SOC
Standard Occupational Classification
SOD
Superoxide dismutase
SPF
Spray polyurethane foam
SUSB
Statistics of U.S. Businesses
SVHC
Substance of Very High Concern
SWC
Surface Water Concentration
T
Toxicity
TBCD
5,6,9,10-tetrabromocyclododec-l-ene
TGD
Technical Guidance Document
TLV
Threshold Limit Value
Page 26 of 723
-------
TOC Total Organic Carbon
TRI Toxics Release Inventory
TSCA Toxic Substances Control Act
TSH Thyroid Stimulating Hormone
TURA Toxics Use Reduction Act
TWA Time-Weighted Average
UF Uncertainty Factor
U.S. United States
UNEP United Nations Environment Programme
vB Very Bioaccumulative
VVWM-PSC Variable Volume Water Model - Point Source Calculator
WEEE Waste Electrical and Electronic Equipment
WEEL Workplace Environmental Exposure Level
WOE Weight of the Scientific Evidence
WSDE Washington State Department of Ecology
WWT/WWTP Wastewater Treatment Plant
XPS Extruded Polystyrene {i.e., Extruded Polystyrene foam)
XPSA Extruded Polystyrene Association
Yr Year
Page 27 of 723
-------
EXECUTIVE SUMMARY
This Risk Evaluation for cyclic aliphatic bromide cluster chemicals, including hexabromocyclododecane
(HBCD), was performed in accordance with the Frank R. Lautenberg Chemical Safety for the 21st
Century Act and is being issued following public comment and peer review. The Frank R. Lautenberg
Chemical Safety for the 21st Century Act amended the Toxic Substances Control Act (TSCA), the
nation's primary chemicals management law, in June 2016. Under the amended statute, EPA is required,
under TSCA Section 6(b), to conduct risk evaluations to determine whether a chemical substance
presents an unreasonable risk of injury to health or the environment, under the conditions of use, without
consideration of costs or other non-risk factors, including an unreasonable risk to potentially exposed or
susceptible subpopulations identified as relevant to the Risk Evaluation. Also, as required by TSCA
Section (6)(b), EPA established, by rule, a process to conduct these Risk Evaluations, Procedures for
Chemical Risk Evaluation Under the Amended Toxic Substances Control Act (82. FR 33726) (Risk
Evaluation Rule). This Risk Evaluation is in conformance with TSCA Section 6(b) and the Risk
Evaluation Rule, and is to be used to inform risk management decisions. In accordance with TSCA
Section 6(b), if EPA finds unreasonable risk from a chemical substance under its conditions of use in
any final Risk Evaluation, the Agency will propose actions to address those risks within the timeframe
required by TSCA. However, any proposed or final determination that a chemical substance presents
unreasonable risk under TSCA Section 6(b) is not the same as a finding that a chemical substance is
"imminently hazardous" under TSCA Section 7. The conclusions, findings, and determinations in this
final Risk Evaluation are for the purpose of identifying whether the chemical substance presents
unreasonable risk under the conditions of use, in accordance with TSCA Section 6, and are not intended
to represent any findings under TSCA Section 7.
TSCA Section 26(h) and (i) require EPA, when conducting Risk Evaluations, to use scientific
information, technical procedures, measures, methods, protocols, methodologies and models consistent
with the best available science and base its decisions on the weight of the scientific evidence.1 To meet
these TSCA Section 26(h) science standards, EPA used the TSCA systematic review process described
in the Application of Systematic Review for TSCA Risk Evaluations document ( )). The
data collection, data evaluation and data integration stages of the systematic review process are used to
develop the exposure, fate and hazard assessments for the risk evaluations.
The cyclic aliphatic bromide cluster chemicals, including HBCD (Chemical Abstracts Service Registry
Number [CASRN] 25637-99-4), 1,2,5,6,9,10-hexabromocyclododecane (1,2,5,6,9,10-HBCD; CASRN
3194-55-6) are flame retardants. Conditions of use for 1,2,5,6-tetrabromocyclooctane (CASRN 3194-57-
8), another chemical in the cyclic aliphatic bromide cluster, were not identified. For the purposes of this
final Risk Evaluation document, the use of "HBCD" refers to either CASRN 25637-99-4 or 3194-55-6,
or both. The primary use of HBCD has been as a flame retardant in expanded polystyrene and extruded
polystyrene; however, EPA identified other uses including use as a component of solder and use in
automobile replacement parts.
HBCD is a persistent, bioaccumulative and toxic (PBT) substance that exists as a non-volatile solid
(Section 1.1). HBCD released to the environment remains unchanged for months or longer and
accumulates in aquatic and terrestrial organisms including humans. Because of these characteristics,
even low levels of HBCD move through aquatic and terrestrial food chains from lower to higher levels
1 Weight of the scientific evidence is defined in EPA regulations as a systematic review method, applied in a manner suited to
the nature of the evidence or decision, that uses a pre-established protocol to comprehensively, objectively, transparently, and
consistently identify and evaluate each stream of evidence, including strengths, limitations, and relevance of each study and
to integrate evidence as necessary and appropriate based upon strengths, limitations, and relevance. 40 CFR 702.33.
Page 28 of 723
-------
and result in increasing concentrations in aquatic and terrestrial life higher in the food chain (Section
2.1). In contrast to chemicals that do not exhibit PBT characteristics, ecological impacts due to trophic
level transfer of HBCD and human dietary exposure pathways to HBCD including fish ingestion are
considered. Background levels of HBCD have been measured in a variety of environmental media and
biota, in indoor air and dust, and in human milk, blood, and urine. Due to HBCD's persistence, humans
and environmental organisms can be exposed to background levels that stem from past activities at the
five stages in the life of the chemical, i.e., manufacture (including import), processing, distribution, use,
and disposal. Releases of HBCD could have resulted from activities that still occur or from releases
associated with uses that phased out of all life stages. These characteristics and their impacts on
environmental and human exposure to HBCD were important considerations in the HBCD Risk
Evaluation. EPA considered a variety of exposure pathways for HBCD to workers, general population,
consumers, and the environment, although certain pathways may have undergone minimal evaluation
based on assessment of physical-chemical properties or other considerations such as existing EPA
regulations (see Section 1.4).
The production (domestic manufacturing and importation) and use of HBCD has rapidly declined in the
U.S. and globally over the past 10 years due to international regulation and the availability of
substitutes. Annual production volumes were consistently 10-50 million lbs from 2007 to 2011. From
2012 to 2015, production fell to 1-10 million lbs/year. Additional communications with industry
representatives indicate that, as of 2018, domestic manufacture of HBCD has ceased and there are
currently no U.S. manufacturers of the chemical. Use of stockpiles and exportation from the United
States was completed at the end of 2017 and is further discussed in Section 1.2.2 of this final Risk
Evaluation. Under the United Nations Stockholm Convention on Persistent Organic Pollutants, 171 of
the 188 Party countries have agreed to ban the production, use, import, and export of HBCD, consistent
with the obligations of the Convention. The United States is not a signatory to the Convention.
Furthermore, substitutes have been adopted in the market. For example, Dow Chemical developed the
polymeric flame retardant that replaced HBCD for use in insulation boards used in construction. The
product is licensed to other manufacturers including Albemarle, Chemtura, and Bromine Compounds
Limited (part of ICL Industrial Products); these companies sell the chemical under different trade
names.
EPA has not identified reasonably available information to suggest that HBCD is currently domestically
manufactured in any quantity. Consideration of the status of manufacturing, availability of viable
substitutes and the strong international regulatory focus on phasing out of manufacturing, use and
international trade in HBCD has led EPA to believe the domestic manufacturing of HBCD is not known,
intended or reasonably foreseen to occur.
Based on information received by industry associations and member companies, historic major
importers have since 2017 ceased importation of the chemical. It is reasonably foreseen, however, that
foreign manufacturers in countries that have not agreed to the Stockholm ban or are non-signatories of
the Convention are or will be in the future producing HBCD that could be imported in quantities below
CDR reporting thresholds. For these reasons, EPA has included the import of HBCD in the final Risk
Evaluation.
The primary use of HBCD in the United States historically has been as a flame retardant in XPS/EPS
insulation foam used in construction. This use had accounted for 95% of all HBCD applications in the
past decade. Based on information from a market report, HBCD was used primarily in construction
materials, which may have included structural insulated panels (SIPS).
Page 29 of 723
-------
Although some of the industry comments on the draft Risk Evaluation indicate with more certainty than
previous comments that the phaseout of HBCD for XPS/EPS insulation foam is complete, the industry
associations do not represent every possible importer and processor of HBCD. Taking into account the
high percentage of HBCD production volume dedicated to these two uses in previous years, and the fact
that companies have the ability to import the chemical in low volumes below the CDR reporting
threshold, EPA believes that it is reasonably foreseen that EPS and XPS Association non-members
currently are or will in the future be using imported HBCD-containing resins in their processes. EPA
therefore included the processing and use of HBCD in XPS and EPS insulation in the final Risk
Evaluation.
In addition to the major use of HBCD in insulation, much smaller quantities have been processed into
products and articles including automotive replacement parts, solder paste, electrical and electronic
products, textiles, adhesives, and coatings. These six products and articles are considered conditions of
use (COUs). As the chemical has declined in importance, the only remaining processing of HBCD into
products and articles is for automotive replacement parts and solder paste. Manufacture, processing, use,
and distribution of HBCD for the other four products and articles have phased out, although commercial/
consumer use and disposal still occur. For the four minor products for which manufacturing, processing,
use and distribution have been phased out, the final RE adds two COUs: Use in other formulated
products and articles (e.g., textiles, electrical and electronic products, adhesives, and coatings) and
Disposal of other formulated products and articles (e.g., textiles, electrical and electronic products,
adhesives, and coatings). All six minor use products and articles are included as COUs in this final Risk
Evaluation.
Reused and recycled EPS and XPS foam insulation board, siding, roof membrane and roofing ballast
material are available in the United States. Two companies were identified that directly reuse (e.g., reuse
without reforming) and recycle (e.g., melting and inserting into the manufacturing process) XPS and
EPS foam insulation. Once processed, recycled EPS roofing insulation is taken to polystyrene product
manufacturers, notably picture frame manufacturers, mostly in China. Recycled roofing material is also
sent to other EPS recycling plants that may use different processes. XPS roofing material is reused due
to the special equipment needed to recycle XPS. The recycling of HB CD-containing EPS and reuse of
XPS insulations boards for use in construction materials is included as a COU in this final Risk
evaluation.
While only anecdotal information is available indicating HBCD use in high impact polystyrene (HIPS)
in electronics occurred in the United States (Section 1.2), there are more substantial data from the EU
indicating a range of between 2 and 7 percent of HBCD production volume in Europe was historically
used in HIPS and that the majority of HIPS was used in electronics. This makes it likely that electronics
products with HBCD-containing HIPS have been imported into the United States in past years. EPA
believes that it is reasonably foreseen that HBCD may be present in recycling of electronics waste and
therefore included this condition of use, called recycling of electronics waste containing HIPS that
contain HBCD, in this final Risk Evaluation. Previously, EPA inadvertently omitted recycling of
electronics waste in the draft Risk Evaluation.
The draft Risk Evaluation contained a COU for consumer use of recycled consumer articles which was
inadvertently left off the list of COUs in Table 1-8. The COU is inserted into Table 1-8 of this final Risk
Evaluation.
EPA previously described three specific scenarios under which the Agency could determine to exclude
certain conditions of use from chemical risk evaluations: legacy uses, associated disposal, and legacy
Page 30 of 723
-------
disposal. By legacy use, EPA referred to circumstances associated with activities that do not reflect
ongoing or prospective manufacturing, processing, or distribution. By associated disposal, EPA referred
to future disposal from legacy uses. By legacy disposal, EPA referred to disposals that have already
occurred. In the rule, "EPA interpreted] the mandates under Section 6(a)-(b) to conduct risk evaluations
and any corresponding risk management to focus on uses for which manufacturing, processing, or
distribution in commerce is intended, known to be occurring, or reasonably foreseen to occur (i.e., is
prospective or on-going), rather than reaching back to evaluate the risks associated with legacy uses,
associated disposal, and legacy disposal, and interprets the definition of 'conditions of use' in that
context." (82 FR 33730) As a result, EPA did not include any legacy uses, associated disposals, or
legacy disposals as conditions of use within the scope of the Risk Evaluations for the first 10 chemicals
undergoing the new TSCA Risk Evaluation process.
However, some stakeholders disagreed with this interpretation and challenged the final Risk Evaluation
Rule in court. In 2019, the Ninth Circuit Court of Appeals ruled that EPA cannot categorically exclude
"legacy use" and "associated disposal" from the definition of "conditions of use" (Safer Chemicals,
Healthy Families v. U.S. Envtl. Prot. Agency, 943 F.3d 397, 425 (9th Cir. 2019)). As a result of the
court's opinion, EPA will no longer exclude legacy use or associated disposal from the definition of
conditions of use for chemical risk evaluations. Rather, when these activities are intended, known, or
reasonably foreseen, they will be considered uses and disposal, respectively, within the definition of
conditions of use. Thus, in conducting a Risk Evaluation, certain parts of the lifecycle for a given COU
may not be evaluated because those parts are not intended, known, or reasonably foreseen. For example,
if the manufacture (including import), processing and distribution parts of the life cycle are not intended,
known, or reasonably foreseen, then the evaluation will only consider the uses and disposal stages of the
lifecycle to be COUs. The court did not rule against EPA's exclusion of legacy disposal from scopes of
risk evaluations.
Prior to the court ruling on Safer Chemicals, Healthy Families v. U.S. Envtl. Prot. Agency, at the
beginning of the Risk Evaluation process for HBCD, EPA had information indicating that a small
percentage of the chemical's production volume (less than 5%) had been used in the past in the
processing of four products and articles. The items were adhesives, coatings, electronics, and textiles.
HBCD is no longer manufactured, processed, or distributed in commerce as part of the four products and
articles. In accordance with the final Risk Evaluation Rule, EPA considered activities involving these
products and articles to be "legacy uses" and "associated disposal" and excluded the activities from the
scope of the August 2019 draft HBCD Risk Evaluation. Later that year, the court made its ruling in
Safer Chemicals Healthy Families v. U.S. Envtl. Prot. Agency. Because of the court ruling, as well as
public and SACC review comments, EPA is no longer excluding the four products and articles in the
Risk Evaluation. Although manufacturing, processing, and distribution in commerce of HBCD in the
products and articles has ended, commercial/consumer use and associated disposal are still occurring
and these activities are COUs in the final risk evaluation. EPA evaluated exposure to these use and
disposal activities and has made a determination for each COU on whether exposure presents
unreasonable risk. Legacy disposal of HBCD, i.e. disposal that occurred in the past, is not a COU.
Likewise, other activities in the HBCD lifecycle stages that occurred in the past are not COUs, although
EPA has evaluated exposure to background levels of HBCD resulting from past activities that left
HBCD in environmental media and indoor air and dust. EPA did not exclude any activity determined to
be a COU.
The conditions of use evaluated for HBCD, as further described in Section 1.4.1 of the final Risk
Evaluation for HBCD, include:
Page 31 of 723
-------
• Importation of HBCD
• Processing of flame retardants: use in custom compounding of resin and solder paste
• Processing of flame retardants: use in manufacture of XPS and EPS foam; use in manufacture of
structural insulated panels; use in automobile replacement parts from XPS and EPS foam
• Processing: recycling of XPS and EPS foam, resin, panels containing HBCD; electronics waste
• Processing: recycling of electronics waste containing HIPS that contains HBCD
• Distribution: activities related to distribution
• Use in building and construction materials
• Use in automobile replacement parts
• Use in plastic and other articles
• Use in other formulated products and articles, e.g., adhesives, coatings, textiles, and electronics
• Disposal of construction and demolition waste
• Disposal of other formulated products and articles, e.g., adhesives, coatings, textiles, and
electronics
Approach
EPA used reasonably available information2 in a fit-for-purpose approach, to develop a risk evaluation
that relies on the best available science and is based on the weight of the scientific evidence. EPA used
previous assessments as a starting point for identifying key and supporting studies to inform the
exposure, fate, and hazard assessments. EPA also evaluated other studies published since the publication
of any previous analyses. EPA reviewed reasonably available information and evaluated the quality of
the methods and reporting of results of the individual studies using the evaluation strategies described in
Application of Systematic Review in TSCA Risk Evaluations (U.S. EPA. 2.018b). To satisfy requirements
in TSCA Section 26(j)(4) and 40 CFR 702.51(e), EPA has provided a list of studies considered in
carrying out the Risk Evaluation, and the results of those studies are included in the Systematic Review
Data Quality Evaluation/Extraction Documents (see Appendix B, items 6 and 7).
In the problem formulation, EPA identified the conditions of use within the scope of the risk evaluation
and presented three conceptual models and an analysis plan for this Risk Evaluation (U.S. EPA. 2018e).
These have been carried into the Risk Evaluation where EPA has quantitatively evaluated the risk to the
environment and human health, using both monitoring data and modeling approaches, for the conditions
of use (identified in Section 1.4.1 of this risk evaluation). EPA quantitatively evaluated the risk to
aquatic (pelagic and benthic) and terrestrial organisms from exposure to surface water, sediment and soil
(via air deposition) as a result of the manufacturing, processing, use, or disposal of HBCD. EPA
evaluated risk to workers, from inhalation and dermal exposures (EPA was unable to quantitatively
evaluate risk to occupational non-users (ONUs))3, by comparing the estimated acute and chronic
exposures to human health hazards (e.g., thyroid effects, liver effects, reproductive effects,
developmental effects). EPA also evaluated the risk to the general population and consumers from acute
and chronic inhalation, dermal, and oral exposures.
EPA used environmental fate parameters, physical-chemical properties, monitoring data and modeling
approaches to assess exposure to aquatic organisms, and sediments and soil exposure to terrestrial
species. The exposure and environmental hazard analyses for these environmental release pathways was
conducted based on a quantitative assessment of predicted environmental concentrations of HBCD in
2 Defined in 40 CFR 702.33 in part as "information that EPA possesses, or can reasonably obtain and synthesize for use in
risk evaluations, considering the deadlines ...for completing the evaluation...".
3 ONUs are workers who do not directly handle HBCD but perform work in an area where HBCD is present.
Page 32 of 723
-------
surface water, sediment, and soil. These exposure analyses are detailed in Section 2.1 through 2.3.5 and
environmental hazards are discussed in Section 3.1 for aquatic and terrestrial organisms.
EPA evaluated potential occupational exposures to HBCD that result from the conditions of use that are
in the scope of this Risk Evaluation as listed in Section 1.4 (Scope of the Evaluation). EPA evaluated
potential acute and chronic inhalation and dermal exposures to workers. EPA estimated potential
inhalation exposure concentrations based on HBCD worker inhalation monitoring data that EPA
obtained via a systematic review of the literature. In the case of some occupational exposure scenarios,
EPA's systematic review of the literature did not result in worker inhalation monitoring data and EPA
estimated potential inhalation exposure concentrations in accordance with other estimation methods.
EPA's systematic review did not result in any HBCD worker dermal monitoring data and EPA estimated
dermal exposures in accordance with modeling approaches. EPA did not quantitatively evaluate
inhalation exposures of ONUs to HBCD due to lack of adequate, reasonably available, worker
monitoring data and lack of relevant mathematical models. The occupational exposure evaluation is
described in detail in Section 2.4.1. In this Risk Evaluation, consumer exposures were evaluated for
individuals who have articles containing HBCD in their homes or automobiles. The consumer exposure
assessment also includes the mouthing of consumer articles that contained HBCD. The consumer
exposure evaluation is described in detail in Section 2.4.4.
HBCD is present and persistent in various environmental media such as surface water, sediment, soil
and air. EPA quantitatively evaluated inhalation, ingestion and dermal exposures to the general
population via exposure to indoor and ambient air; dermal contact with soil and dust and oral exposures
via ingestion of food, breast milk, soil, dust and fish. While HBCD is released to surface water, EPA
determined during problem formulation that no further analysis beyond what was presented in the
problem formulation document would be done for the drinking water exposure pathway in this Risk
Evaluation. While this exposure pathway remains in the scope of the risk evaluation, EPA found no
further analysis was necessary. Further analysis was not conducted for the drinking water pathway based
on a qualitative assessment of the physical chemical properties and fate of HBCD in the environment as
well as the absence of any detection of HBCD in monitored water samples.
While environmental exposures are expected to decline as importing and processing of the chemical are
being phased out, based on past production volumes (millions of pounds per year) and the fact that
cessation of domestic manufacturing is recent, reductions in environmental and biological
concentrations will likely occur gradually over a period of time for this persistent and bioaccumulative
compound. The time scales for this are dependent on the age of the products, their useful service lives
and timelines for replacement.
EPA also evaluated background exposures in calculating risk estimates for the environment and general
population, representing chronic, steady-state risks from sustained background exposure in the
environment due to HBCD's persistence. These exposures cannot be associated with any particular COU
or past use and it is unknown which combination of potential sources associated with evaluated COUs or
past uses contribute to this background exposure. These background exposures were considered
independently of COU-specific releases within exposure routes but were also aggregated across different
exposure routes when applicable {i.e., for human health). The totality of background exposure includes
steady-state environmental exposures from ongoing releases not associated with a particular COU, and
releases stemming from historical activities (Section 1.2.9) due to HBCD's persistence in the
environment. Historical activities are past activities that may have released HBCD but no longer occur
{e.g., releases from a manufacturing plant before it stopped producing HBCD, residual indoor dust from
formerly owned HBCD-containing products, legacy disposal).
Page 33 of 723
-------
In the absence of reasonably available information on product-specific releases, cumulative background
exposure was also used as a surrogate for assessing the COUs for use and disposal of formulated
products (e.g., adhesives and coatings) and articles (e.g., textiles, electrical and electronic products),
minor-use products, and articles which are no longer manufactured, processed, or distributed. While
EPA cannot determine COU-specific releases or exposures for these products/articles, they are expected
to contribute to background exposure and would therefore comprise a subset of the total background
exposure levels. The amount of HBCD in the subset of background exposure levels is unknown, but the
risk estimation of exposure to the total background levels of HBCD is an upper bound and therefore
constitute a conservative risk characterization for the two COUs.
EPA reviewed the environmental hazard data using the data quality review evaluation metrics and the
rating criteria described in the Application of Systematic Review in TSCA Risk Evaluations (U.S. EPA.
2.018b). EPA concluded that HBCD poses a hazard to environmental aquatic and terrestrial receptors.
Hazard thresholds for aquatic organisms were derived using algae, fish and invertebrates, as a result of
acute and chronic exposures. Similarly, maize, earthworms, kestrel, osprey and rats were used to derive
hazard thresholds for terrestrial organisms due to both acute and chronic exposures to HBCD. The
results of the environmental hazard assessment are in Section 3.1.
In the human hazard section, EPA evaluated reasonably available information and identified hazard
endpoints including acute/chronic toxicity, non-cancer effects, associated with inhalation, oral and
dermal exposures. EPA used an approach based on the Framework for Human Health Risk Assessment
to Inform Decision Making (U.S. EPA. 2014e) to evaluate, extract and integrate HBCD's human health
hazard and dose-response information. EPA reviewed key and supporting information from previous
hazard assessments as well as the existing body of knowledge on HBCD's human health hazards. These
data sources included the TRI Technical Review of HBCD ( 16e). the TSCA Work Plan
Problem Formulation and Initial Assessment, ( a), Preliminary Materials for the IRIS
Toxicological Review of HBCD ( Q14f) as well as other publications (U.S. EPA. 2016e.
2014d: NICNAS 2.012a: EC/HC 2.011: EIN.ECS 2008: U.S. EPA 2008a: OECD 20071
EPA considered adverse effects for HBCD across organ systems. EPA considered data on toxicity
following acute and chronic exposures, for irritation, sensitization, genotoxicity, reproductive,
developmental and other systemic toxicity and carcinogenicity. From these effects, the EPA selected
endpoints supported by the evidence for non-cancer that were amenable to quantitative analysis for
dose-response assessment as discussed in more detail in Section 3.2.5. Based on the weight of the
scientific evidence evaluation, four health effect domains were selected for non-cancer dose-response
analysis: (1) thyroid; (2) liver; (3) female reproductive; and (4) developmental. These hazards were
carried forward for dose-response analysis. Given the different HBCD exposure scenarios considered
(both acute and chronic), different endpoints were considered for risk estimation based on the expected
exposure durations. The results of the human hazard assessment are in Section 3.2.
Risk Characterization
Environmental Risk: For environmental risk, EPA utilized a risk quotient (RQ) to compare the
environmental concentration to the effect level to characterize the risk to aquatic and terrestrial
organisms. As described in Section 3.1.5, the environmental hazard thresholds are based on
environmental hazard concentrations reported for both aquatic and terrestrial organisms. The algae
concentration of concern (COC) is based on observed reductions in growth rate as a result of a 72-hour
exposure to HBCD. The acute COC is based on delayed zebrafish embryo hatching as a result of a 96-
Page 34 of 723
-------
hour exposure to HBCD. Finally the chronic COC for pelagic (water flea) and benthic (California
blackworm) invertebrates are based on reduced growth in surviving young and a reduction in worm
number, respectively. Hazard thresholds used to characterize risk for terrestrial soil organisms include
effects regarding reproduction and mortality in earthworms exposed to HBCD for 56 days.
HBCD is a persistent, bioaccumulative and toxic (PBT) chemical, and is expected to be present in
surface water, sediment and soil. To characterize HBCD exposure in aquatic and terrestrial
environments, both environmental modeling and monitoring data were used to provide media-specific
concentrations of HBCD. To characterize environmental risk associated with a COU, models were used
to estimate environmental concentrations of HBCD (e.g., surface water, sediment, soil via air
deposition), where the predicted HBCD releases associated with exposure scenarios depend on many
factors (e.g., days of release, HBCD half-life). Environmental monitoring information provides time
and geographically-specific snapshots of measured concentrations of HBCD that are not linked to a
specific, identified current or past industrial or commercial activity. Environmental monitoring
information supports the estimated HBCD concentrations associated with various conditions of use, and
both types of environmental exposure estimates are used to derive environmental risk. The results of the
risk characterization are in Section 4.1.4, and Table 4-25 summarizes the RQs for aquatic and terrestrial
organisms.
EPA identified the expected environmental exposures for aquatic and terrestrial species under the
conditions of use in the scope of the risk evaluation. Estimated releases from specific exposure scenarios
result in modeled surface water and sediment concentrations that exceed the aquatic benchmark (RQ >
1) for acute, chronic and/ or algae COC for every COU except for disposal, where only the chronic COC
for benthic invertebrates was not exceeded (acute, chronic and algae COCs for pelagic organisms were
exceeded and environmental risks were indicated). Furthermore, surface water and sediment HBCD
concentrations measured near industrial facilities also exceeded acute, chronic and/or algae COCs,
whereas those measured near general population sites did not. In regard to the characterization of risk to
terrestrial organisms, there were no HBCD soil concentrations attained from modeled or monitoring data
that exceeded the chronic COC. Details of these estimates are in Section 4.5.1.
Risks to aquatic organisms were identified for every COU with water releases, based on exceedances of
COCs for pelagic and/or benthic organisms. EPA found it unlikely that there may be risks of concern for
terrestrial soil organisms based on the air releases of HBCD associated with the conditions of use.
Human Health Risks: Risks were estimated for all human receptors following both acute and chronic
exposure for representative endpoints from every hazard domain carried through to dose-response
analysis. Risks for acute exposures were only evaluated for developmental endpoints, while all
endpoints were evaluated for chronic risks. Risk conclusions were based on the most robust and
sensitive acute (offspring loss) and chronic (thyroid hormone effects) endpoints. Thyroid hormone
changes (both acute and chronic) are considered the primary effect resulting from HBCD exposure, as
they are associated with all of the other observed downstream endpoints.
EPA estimated potential non-cancer risks resulting from acute and chronic inhalation and dermal
exposures using a Margin of Exposure (MOE) approach. EPA estimated risks for workers under
several occupational exposure scenarios using scenario-specific assumptions regarding the expected
use of personal protective equipment (PPE) for respiratory and dermal exposures for workers directly
handling HBCD. More information on respiratory and dermal protection, including EPA's approach
regarding the occupational exposure scenarios for HBCD, is in Section 2.4.1.
Page 35 of 723
-------
For acute and chronic exposures via inhalation without PPE {i.e., no respirators), risks are indicated for
workers relative to the benchmarks for multiple occupational exposure scenarios (OES). There are risks
at both high-end and central tendency exposure levels relative to benchmark for four OES, and there are
risks based on high-end inhalation exposure levels for another six OES. With use of PPE during relevant
conditions of use, worker exposures were estimated to be reduced. This resulted in fewer conditions of
use with estimated acute, chronic non-cancer, or cancer inhalation or dermal risks. With use of
respiratory protection, non-cancer risks were not indicated for any conditions of use within the scope of
the risk evaluation. Specifically, when respirators are worn (APF 5, 10, or 50), risks are not indicated for
both acute and chronic exposure durations at both high-end and central tendency exposure levels.
Workers exposed through Installation or Demolition of XPS/EPS Foam Insulation are unlikely to wear
respiratory protection. Therefore, when considering assumed use of PPE, risks are indicated only for
those two OES. Risks were not indicated at either high-end or central tendency exposure levels for
Processing of HBCD to produce XPS foam using XPS Masterbatch, Occupational microenvironments,
and Recycling of electronics waste containing HIPS. Occupational non-users (ONUs) are assumed to
have lower exposure levels than workers in most instances but exposures could not be quantified.
Exposures are site-specific and are depended on several site-specific factors including engineering
controls, work practices, and particle size. Also, EPA did not identify any peer-reviewed models that can
be used to estimate exposures for ONUs for these specific scenarios.
For acute and chronic exposures via dermal contact without PPE {i.e., no gloves) risks are indicated for
workers relative to the benchmark for multiple OES, with risks at both high-end and central tendency
exposure levels for five OES. Risks are indicated based only on chronic exposure at the high-end
exposure level for a single OES, Use offlux/solder paste. EPA does not expect any level of dermal
exposure to HBCD following proper use of impervious gloves (Section 2.4.1.1). Therefore, risk
estimates are not provided, and risks are not identified for any exposure scenario when impervious
gloves are worn and used appropriately. EPA did not evaluate ONU dermal exposure to HBCD since
they are not expected to handle the chemical. ONUs are potentially exposed to HBCD dermally through
contact with surfaces where HBCD dust has settled but EPA did not quantify these risks due to minimal
exposure.
For the general population, EPA estimated non-cancer risks resulting from chronic aggregate
background exposure via all relevant pathways including dust, soil, indoor air, diet, and dermal
pathways (Section 4.2.3.1). Risks were also estimated based on a subset of aggregate background
exposures for workers in occupational microenvironments (Section 4.2.3.1.1). For the most sensitive
highly exposed general population (a Potentially Exposed or Susceptible Subpopulations (PESS) group
who are expected to live close to facility or residential HBCD sources, see Section 2.4.3), EPA
estimated non-cancer risks resulting from acute or chronic exposures via inhalation or fish ingestion
(Section 4.2.3.3). For highly exposed general population risk estimation, EPA incorporated summed
exposures from representative fish ingestion or air inhalation modeled exposures and the aggregate
central tendency general population biomonitoring-based exposures (representing background exposure)
for all other exposure routes. Risks were estimated based on the highest and representative moderate
exposure sub-scenarios representing variability in estimated releases and wastewater treatment. Risks
were also estimated for consumers based on indoor air and dust exposure and aggregated background
exposures from other routes. Risks were indicated relative to benchmark only for a single OES at the
highest exposure sub-scenario, via acute fish ingestion (Table 4-21). For all other exposure scenarios,
risk estimates were several fold above the benchmark and risk is not expected. Based on qualitative
consideration of the physical-chemical and fate characteristics as well as low concentrations in surface
Page 36 of 723
-------
water and the absence of any monitored levels in drinking water, HBCD is not expected to be present in
drinking water. Therefore, risks were not identified for HBCD via drinking water exposure.
EPA additionally calculated distinct risk estimates for various PESS groups including subsistence fishers
(a group that ingests elevated levels of fish compared to the general population) and newborns less than
1 year old (who are not expected to ingest fish) based on chronic high-end aggregate background
exposure (Table 4-29). Additional details on risk considerations for all PESS groups are described in
Section 4.4.1. Risk estimates did not indicate risk relative to the benchmark for either of these two
highly-exposed receptor groups.
Uncertainties: Key assumptions and uncertainties in the environmental risk estimation are related to
data used for the characterization of environmental exposure (e.g., model input parameters, inability to
directly relate monitoring sites to conditions of use) and environmental hazard (e.g., selection of
representative organisms, allometric-scaling to estimate hazard thresholds for other organisms).
Additionally, the reasonably available environmental monitoring data was limited temporally and
geographically. Assumptions and key sources of uncertainty in the risk characterization are detailed in
Section 4.3.1.
For the human health risk estimation, key assumptions and uncertainties are related to the toxicokinetics
of HBCD, including high-end assumptions about dermal absorption and uncertainty whether existing
UFs sufficiently account for bioaccumulation in human tissues. Additional sources of uncertainty
related to human health hazard include the application of adult rodent thyroid hormone changes to
humans in a developmental context and the absence of reliable dose-response information for
developmental neurotoxicity endpoints. EPA also considered differing assumptions about PPE usage
for each OES which strongly influences the risk conclusions. Important assumptions and key sources of
uncertainty in the risk characterization are described in more detail in Section 4.3.2.
EPA's assessments, risk estimations, and risk determinations account for uncertainties throughout the
risk evaluation. EPA used reasonably available information, in a fit-for-purpose approach, to develop a
risk evaluation that relies on the best available science and is based on the weight of the scientific
evidence. For instance, systematic review was conducted to identify reasonably available information
related to HBCD hazards and exposures. If no applicable monitoring data were identified, exposure
scenarios were assessed using a modeling approach that requires the input of various chemical
parameters and exposure factors. When possible, default model input parameters were modified based
on chemical-specific inputs available in literature databases. The consideration of uncertainties supports
the Agency's risk determinations, each of which is supported by substantial evidence, as set forth in
detail in later sections of this final Risk Evaluation.
Potentially Exposed Susceptible Subpopulations
TSCA Section 6(b)(4) requires that EPA conduct Risk Evaluations to determine whether a chemical
substance presents unreasonable risk under the conditions of use, including unreasonable risk to a
potentially exposed or susceptible subpopulation identified as relevant to the Risk Evaluation. TSCA
Section 3(12) defines "potentially exposed or susceptible subpopulation as a group of individuals within
the general population identified by the Administrator who, due to either greater susceptibility or
greater exposure, may be at greater risk than the general population of adverse health effects from
exposure to a chemical substance or mixture, such as infants, children, pregnant women, workers, or the
elderly."
Page 37 of 723
-------
In developing the Risk Evaluation, EPA analyzed reasonably available information to ascertain whether
some human receptor groups may have greater exposure than the general population to the hazard posed
by HBCD. In consideration of the most highly exposed groups, EPA considered workers using HBCD
and ONUs in the vicinity of HBCD to be PESS groups based on higher exposures than the general
population. Exposures of HBCD would also be expected to be higher amongst individuals exposed to
scenario-specific exposures, from releases to water, air, and consumer articles as compared to the
general population. These include the highly exposed general population, or individuals who are
expected to live close to facility sources (Section 2.4.3).
Based on the bioaccumulation of HBCD and partitioning to lipid, subpopulations with elevated body fat
or on a high-fat diet are of increased susceptibility and represent an important PESS group. Pregnant
women and women of reproductive age are another potentially exposed or susceptible subpopulation
based on the possibility of reproductive and developmental effects following exposure. Humans with
pre-existing health conditions or genetic predispositions related to any of the affected health domains are
also susceptible subpopulations, as they may experience HBCD toxicity at lower doses than the general
population.
EPA accounted for PESS in risk estimation by providing risk conclusions (Section 4.5.2) based on
the most sensitive receptor or lifestage (ie., female workers of reproductive age for occupational
risk, the youngest relevant lifestage for general population and consumer risk) and consideration
of high end exposures (Table 4-27;
Page 38 of 723
-------
Table 4-28). Estimated risks to additional highly-exposed PESS groups were also separately calculated
(Table 4-29).
Aggregate and Sentinel Exposures
Section 6(b)(4)(F)(ii) of TSCA requires the EPA, as a part of the Risk Evaluation, to describe whether
aggregate or sentinel exposures under the conditions of use were considered and the basis for their
consideration. The EPA has defined aggregate exposure as "the combined exposures to an individual
from a single chemical substance across multiple routes and across multiple pathways (40 CFR Section
702.33)." Exposures to HBCD were evaluated by inhalation and dermal routes separately for workers
and consumers. Inhalation and dermal exposures are assumed to occur simultaneously for workers and
consumers. EPA chose not to employ simple additivity of exposure pathways at this time within a COU
because of the uncertainties present in the current exposure estimation procedures that may lead to an
overestimate of exposure without the use of a PBPK model available for determining the effect on
internal dose estimates. For all general population exposure routes, background aggregate exposures for
all exposure routes were combined with specific modeled exposures for the pathway of interest {i.e., fish
ingestion, air inhalation, dust/indoor air, mouthing). Aggregating general population exposures is
appropriate because these background exposures are based on monitoring data and account for the
persistence of HBCD in biological tissues.
The EPA defines sentinel exposure as "the exposure to a single chemical substance that represents the
plausible upper bound of exposure relative to all other exposures within a broad category of similar or
related exposures" (40 CFR Section 702.33). In this Risk Evaluation, the EPA considered sentinel
exposure the highest exposure given the details of the conditions of use and the potential exposure
scenarios. EPA characterized high-end exposures in evaluating both modeled and monitored exposures
to various receptors. Sentinel exposures for workers are the high-end exposure levels with assumptions
of no PPE within each OES. In cases where sentinel exposures result in MOEs greater than the
benchmark, indicating that risk is not likely, EPA did no further analysis to refine the risk estimates
because sentinel exposures represent the worst-case scenario.
For additional discussion on incorporation of aggregate and sentinel exposures into the Risk Evaluation,
see Section 4.4.2.
Unreasonable Risk Determination
In each Risk Evaluation under TSCA Section 6(b), EPA determines whether a chemical substance
presents an unreasonable risk of injury to health or the environment, under the conditions of use. The
determination does not consider costs or other non-risk factors. In making this determination, EPA
considers relevant risk-related factors, including, but not limited to: the effects of the chemical substance
on health and human exposure to such substance under the conditions of use (including cancer and non-
cancer risks); the effects of the chemical substance on the environment and environmental exposure
under the conditions of use; the population exposed (including any potentially exposed or susceptible
subpopulations, as determined by EPA); the severity of hazard (including the nature of the hazard, the
irreversibility of the hazard); and uncertainties. EPA also takes into consideration the Agency's
confidence in the data used in the risk estimate. This includes an evaluation of the strengths, limitations,
and uncertainties associated with the information used to inform the risk estimate and the risk
characterization. The rationale for the unreasonable risk determination is in Section 5.2. The Agency's
determinations are supported by substantial evidence, as set forth in detail in later sections of this final
Risk Evaluation. EPA did not exclude any activity determined to be a COU, and a risk determination
was made on all identified COUs.
-------
Unreasonable Risk of Injury to the Environment
Listed below are EPA's determinations of unreasonable risk for specific conditions of use of HBCD
based on risks of exposure for aquatic and terrestrial organisms. To characterize the exposures to HBCD
by aquatic and terrestrial organisms, EPA considered modeled data to represent surface water and
sediment concentrations near facilities actively releasing HBCD to surface water, and soil
concentrations due to facilities actively releasing HBCD through air releases and deposition. Monitored
concentrations to represent ambient water, sediment and soil concentrations of HBCD were also
considered to characterize exposure of aquatic and terrestrial organisms to HBCD. EPA considered the
biological relevance of the species to determine the environmental hazard thresholds, as well as
frequency and duration of the exposures, and uncertainties given the different sources of information
used to characterize the hazard and exposure and derive risk quotients (RQ). For pelagic organisms,
EPA evaluated unreasonable risk of delayed hatching and reduced growth of juvenile organisms due to
acute and chronic exposures to HBCD, respectively. EPA evaluated algae risk separately from the
categorization of an acute or chronic exposure, and unreasonable risk of reduced algae growth was
evaluated. Based on the physical-chemical properties, HBCD partitions to sediment and soil. For benthic
organisms, EPA evaluated unreasonable risk of reduced reproduction due to chronic exposure to HBCD.
EPA also evaluated unreasonable risks of reduced reproduction and survival of soil organisms due to
chronic exposure to HBCD. EPA determined that the evaluation supports an unreasonable risk
determination to aquatic organisms (pelagic and benthic) for each condition of use of HBCD within the
scope of the Risk Evaluation but does not support unreasonable risk determinations for terrestrial soil
organisms.
Unreasonable Risk of Injury to Aquatic Organisms: EPA made unreasonable risk determinations for
risks to pelagic and benthic species due to HBCD exposures at high-end concentrations in both surface
water and sediment. The unreasonable risk determination applies to six of twelve conditions of use
within the scope of the Risk Evaluation.
No Unreasonable Risk of Injury to Terrestrial Organisms: The hazard endpoints for terrestrial organisms
in the Risk Evaluation are growth, reproduction, and thyroid hormone effects. Results of the evaluation
support a determination of no unreasonable risk to terrestrial organisms for all conditions of use of
HBCD within the scope of the Risk Evaluation.
No Unreasonable Risk of Injury to Health: EPA's determinations of unreasonable risk for specific
conditions of use of HBCD listed below are based on health risks of exposure to HBCD for workers, the
general population, the highly exposed general population (fish consumption, air inhalation, and worst-
case aggregate infant (less than 1 year old) exposure), consumers, and other PESS. The hazard endpoint
for acute exposures is offspring loss and for chronic exposures, the endpoint is non-cancer thyroid
effects. Risks for cancer were not evaluated based on inadequate weight of scientific evidence for cancer
hazard (Section 3.2.4.2).
No Unreasonable Risk of Injury to Health of the General Population: As part of the problem formulation
for HBCD, EPA found that exposures to the general population may occur from the conditions of use
due to releases to air, water or land, and evaluated the risk of HBCD exposures to the general population
from multiple routes. EPA found no unreasonable risk for the general population or the highly exposed
general population from any of the conditions of use via exposures from ambient air, surface water,
biosolids, or sediments. Similarly, EPA determined that the evaluation does not support an unreasonable
risk determination to the general population of exposure to HBCD via drinking water based on a
Page 40 of 723
-------
qualitative assessment of the physical chemical properties and fate of HBCD in the environment as well
as the absence of any detection of HBCD in monitored water samples.
In addition, EPA found that exposures to general population via disposal pathways fall under the
jurisdiction of other environmental statutes administered by EPA, i.e., SDWA (Safe Drinking Water
Act). As explained in more detail in Section 1.4.2, EPA believes it is both reasonable and prudent to
tailor TSCA risk evaluations when other EPA offices have expertise and experience to address specific
environmental media, rather than attempt to evaluate and regulate potential exposures and risks from
those media under TSCA. EPA believes that coordinated action on exposure pathways and risks
addressed by other EPA-administered statutes and regulatory programs is consistent with statutory text
and legislative history, particularly as they pertain to TSCA's function as a "gap-filling" statute, and also
furthers EPA aims to efficiently use Agency resources, avoid duplicating efforts taken pursuant to other
Agency programs, and meet the statutory deadline for completing risk evaluations. EPA has therefore
tailored the scope of the risk evaluation for HBCD using authorities in TSCA Section 6(b) and 9(b)(1).
Unreasonable Risk of Injury to Health of Workers: The Risk Evaluation of non-cancer effects from
acute and chronic dermal and inhalation occupational exposures was the basis for EPA's determination
of no unreasonable risk to workers' health for eight conditions of use within the scope of the Risk
Evaluation. For two other COUs within the scope of the Risk Evaluation (Commercial/Consumer Use of
Building Materials (Installation) and Disposal (Demolition)), EPA determined there is unreasonable risk
to workers from inhalation exposure.
EPA generally assumes compliance with OSHA requirements for protection of workers, including the
implementation of the hierarchy of controls. In support of this assumption, EPA used reasonably
available information, including public comments, indicating that some employers, particularly in the
industrial setting, are providing appropriate engineering or administrative controls or PPE to their
employees consistent with OSHA requirements. EPA does not have reasonably available information to
support this assumption for each COU; however, EPA does not believe that the Agency must presume,
in the absence of such information, a lack of compliance with existing regulatory programs and
practices. Rather, EPA assumes there is compliance with worker protection standards unless case-
specific facts indicate otherwise, and therefore existing OSHA regulations for worker protection and
hazard communication will result in use of appropriate PPE in a manner that achieves the stated APF or
PF. EPA's decisions for unreasonable risk to workers are based on high-end exposure estimates, in order
to account for the uncertainties related to whether or not workers are using PPE. EPA believes this is a
reasonable and appropriate approach that accounts for reasonably available information and professional
judgment related to worker protection practices, and addresses uncertainties regarding availability and
use of PPE.
For each COU of HBCD with an identified risk for workers, EPA assumes, as a baseline, the use of a
respirator with an APF of 5, 10, or 50. Similarly, EPA assumes the proper use of impervious gloves,
which is expected to completely prevent dermal exposures to HBCD. However, EPA assumes that for
some conditions of use, the use of appropriate respirators is not a standard industry practice, based on
best professional judgment given the burden associated with the use of supplied-air respirators,
including the expense of the equipment and the necessity of fit-testing and training for proper use. EPA
does not assume that as a standard industry practice that workers installing or demolishing XPS/EPS
insulation in buildings and structures wear respirators.
The unreasonable risk determinations incorporate consideration of the PPE that EPA assumes that
Page 41 of 723
-------
workers use. A full description of EPA's unreasonable risk determination for each condition of use is in
Section 5.2.
Unreasonable Risk of Injury to Health of Occupational Non-Users (ONUs): EPA expects that ONUs
have lower exposure levels than workers in most instances (Section 4) but exposures could not be
quantified, and EPA did not make unreasonable risk determinations for ONUs in most cases. For the two
conditions of use encompassing installation or demolition of building insulation for which EPA expects
that worker and ONU exposure are similar, EPA found unreasonable risk from these exposures to
HBCD.
No Unreasonable Risk of Injury to Health of Consumers: EPA evaluated non-cancer risks to consumers
from acute and chronic inhalation and ingestion exposures to indoor air and dust. These exposures were
associated with consumer use of products and articles in buildings and vehicles. In addition, EPA
assessed the risk to children from mouthing of articles made from recycled plastic containing HBCD.
EPA did not find unreasonable risk from this consumer exposure to HBCD.
Unreasonable Risk of Injury to Health of Potentially Exposed or Susceptible Subpopulations Not
Associated with Any Particular CPU: Based on risk estimates of exposure to HBCD for various PESS
including subsistence fishers and newborns less than 1 year old, EPA did not find unreasonable risk of
exposure to HBCD.
Summary of Unreasonable Risk Determinations:
In conducting Risk Evaluations, "EPA will determine whether the chemical substance presents an
unreasonable risk of injury to health or the environment under each COU within the scope of the Risk
Evaluation..." 40 CFR 702.47. Pursuant to TSCA Section 6(i)(l), a determination of "no unreasonable
risk" shall be issued by order and considered to be a final agency action. Under EPA's implementing
regulations, "[a] determination by EPA that the chemical substance, under one or more of the conditions
of use within the scope of the risk evaluation, does not present an unreasonable risk of injury to health or
the environment will be issued by order and considered to be a final Agency action, effective on the date
of issuance of the order." 40 CFR 702.49(d).
EPA evaluated 12 conditions of use. EPA has determined that the following conditions of use of HBCD
do not present an unreasonable risk of injury to health or the environment. These determinations are
considered final agency action and is being issued by order pursuant to TSCA Section 6(i)(l). The
details of these determination are in Section 5.2 and the TSCA Section 6(i)(l) order is contained in
Section 5.4.1 of this final Risk Evaluation.
Conditions of Use that Do Not Present an Unreasonable Risk
• Processing: Recycling (of electronics waste containing high impact polystyrene (HIPS) that
contains HBCD)
• Distribution
• Commercial/Consumer Use: Other - Replacement Automobile Parts
• Commercial/Consumer Use: Other - Plastic and Other Articles
• Commercial/Consumer Use: Other - Formulated Products and Articles
• Disposal of Formulated Products and Articles
Page 42 of 723
-------
EPA has determined that the following conditions of use of HBCD present an unreasonable risk of
injury to health and/or the environment. There is unreasonable risk of injury to the environment for the
six conditions of use below as well as unreasonable risk of injury to the health of workers for
commercial/consumer use of building/construction materials and for disposal (demolition). EPA will
initiate TSCA Section 6(a) risk management actions on these conditions of use as required under TSCA
Section 6(c)(1). Pursuant to TSCA Section 6(i)(2), the unreasonable risk determinations for these
conditions of use are not considered final agency action. The details of these determinations are in
Section 5.2.
Manufacturing That Presents an Unreasonable Risk to the Environment
• Import
Processing that Presents an Unreasonable Risk to the Environment
• Processing: Incorporation into a Formulation, Mixture, or Reaction Products
• Processing: Incorporation into Article
• Processing: Recycling (of XPS and EPS foam, resin, and panels containing HBCD)
Commercial/Consumer* Use that Presents an Unreasonable Risk to Human Health and the
Environment
• Commercial/Consumer Use: Building/Construction Materials (Installation)
*Note: While commercial and consumer use was assessed as part of the same exposure scenario,
risks were quantified separately and no unreasonable risks to consumers were identified.
Disposal that Presents an Unreasonable Risk to Human Health and the Environment
• Disposal (Demolition)
Page 43 of 723
-------
1 INTRODUCTION
This document is the final Risk Evaluation for HBCD under the Frank R. Lautenberg Chemical Safety
for the 21st Century Act. The Frank R. Lautenberg Chemical Safety for the 21st Century Act amended
the Toxic Substances Control Act (TSCA), the nation's primary chemicals management law, on June 22,
2016.
The Agency published the Scope of the Risk Evaluation for HBCD ( )17d) in June 2017, and
the Problem Formulation for Cyclic Aliphatic Bromide Cluster (HBCD) in June 2018 (
2018g)., which represented the analytical phase of Risk Evaluation in which "the purpose for the
assessment is articulated, the problem is defined, and a plan for analyzing and characterizing risk is
determined" as described in Section 2.2 of the Framework for Human Health Risk Assessment to Inform
Decision Making. EPA received comments on the published Problem Formulation (U.S. EPA. 2018g) for
HBCD and has considered the comments specific to HBCD, as well as more general comments
regarding EPA's chemical Risk Evaluation approach for developing the Risk Evaluations for the first 10
chemicals EPA is evaluating.
The problem formulation identified the conditions of use and presented a conceptual model and an
analysis plan. Based on EPA's analysis of the conditions of use, physical-chemical and fate properties,
environmental releases, and exposure pathways, the problem formulation preliminarily concluded that
further analysis was necessary for exposure pathways to environmental receptors, workers, consumers
and the general population. The mouthing of articles pathway was added to the conceptual model after
the published Problem Formulation based on review of reasonably available information. Further
analysis was not conducted for the drinking water pathway based on a qualitative assessment of the
physical chemical properties and fate of HBCD in the environment. EPA subsequently published a
draft Risk Evaluation for HBCD and has taken public and peer review comments.
At the beginning of the Risk Evaluation process for HBCD, EPA had information that a small
percentage of the chemical's production volume was used in the processing of several products and
articles, including electronics (Use Document, EPA-HO-OPPT-2Q16-0735-0003). Further
investigation led EPA to conclude that HBCD was no longer manufactured, processed, or distributed
for use in such products and articles. The uses of HBCD in such products and articles and the disposal
of those products and articles were therefore excluded from the evaluation as "legacy uses" and
"associated disposal," respectively. In August 2019, EPA completed its draft Risk Evaluation on the
narrowed scope, and later that year, the court made its ruling in Safer Chemicals Healthy Families v.
U.S. Envtl. Prot. Because of the court ruling, as well as public and SACC review comments, EPA
conducted additional assessments on what would have been termed "legacy use": general population
exposure to HBCD in dust and indoor air released from HBCD-containing products and articles that
are still in use but for which the manufacture, processing, and distribution for such use has ceased.
The conclusions, findings, and determinations in this final Risk Evaluation are for the purpose of
identifying whether the chemical substance presents unreasonable risk or no unreasonable risk under
the conditions of use, in accordance with TSCA Section 6, and are not intended to represent any
findings under TSCA Section 7.
As per EPA's final rule, Procedures for Chemical Risk Evaluation Under the Amenc w
Substances Control Act (82 FR 33726 (July 20, 2017)), this Risk Evaluation was subject to both public
Page 44 of 723
-------
comment and peer review, which are distinct but related processes. EPA provided 60 days for public
comment on any and all aspects of this Risk Evaluation, including the submission of any additional
information that might be relevant to the science underlying the Risk Evaluation and the outcome of the
systematic review associated with HBCD. This satisfies TSCA Section 6(b)(4)(H), which requires EPA
to provide public notice and an opportunity for comment on a draft Risk Evaluation prior to publishing a
final Risk Evaluation.
Peer review was conducted in accordance with EPA's regulatory procedures for chemical Risk
Evaluations, including using the EPA Peer Review Handbook and other methods consistent with the
science standards laid out in Section 26 of TSCA {See 40 CFR 702.45). As explained in the Risk
Evaluation Rule (82 FR 33726 (July 20, 2017)), the purpose of peer review is for the independent
review of the science underlying the risk assessment. Peer review will therefore address aspects of the
underlying science as outlined in the charge to the peer review panel such as hazard assessment,
assessment of dose-response, exposure assessment, and risk characterization.
As EPA explained in the Risk Evaluation Rule (82 FR 33726 (July 20, 2017)), it is important for peer
reviewers to consider how the underlying risk evaluation analyses fit together to produce an integrated
risk characterization, which forms the basis of an unreasonable risk determination. EPA believed peer
reviewers were most effective in this role if they received the benefit of public comments on draft risk
evaluations prior to peer review. For this reason, and consistent with standard Agency practice, the
public comment period preceded peer review. The final risk evaluation changed in response to public
comments received on the draft risk evaluation and/or in response to peer review, which itself may be
informed by public comments. EPA responded to public and peer review comments received on the
draft risk evaluation and explained changes made in response to those comments in this final risk
evaluation and the associated response to comments document.
In this final Risk Evaluation, Section 1 presents the basic physical-chemical characteristics of HBCD, as
well as a background on regulatory history, conditions of use, and conceptual models, with particular
emphasis on any changes since the publication of the draft Risk Evaluation. Section 1 also includes a
discussion of the systematic review process utilized in this final Risk Evaluation. Section 2 provides a
discussion and analysis of the exposures, both human and environmental, that can be expected based on
the conditions of use for HBCD. Section 3 discusses environmental and human health hazards of
HBCD. Risk characterization is presented in Section 4, which integrates and assesses the best available
science and "reasonably available information"4 on environmental and human health hazards and
exposures, as required by TSCA (15 U.S.C. 2605(b)(4)(F)). This section also includes a discussion of
any uncertainties and how they impact the Risk Evaluation. In Section 0, the Agency presents the risk
determination of whether risks posed by the chemical substance under the conditions of use are
"unreasonable" as required under TSCA (15 U.S.C. 2605(b)(4)).
EPA also solicited input on the first 10 chemicals as it developed use documents, scope documents, and
problem formulations. At each step, EPA has received information and comments specific to individual
chemicals and of a more general nature relating to various aspects of the Risk Evaluation process,
technical issues, and the regulatory and statutory requirements. EPA has considered comments and
4 "Reasonably available information means information that EPA possesses or can reasonably generate, obtain, and
synthesize for use in risk evaluations, considering the deadlines specified in TSCA Section 6(b)(4)(G) for completing such
evaluation. Information that meets the terms of the preceding sentence is reasonably available information whether or not the
information is confidential business information, that is protected from public disclosure under TSCA Section 14."
Page 45 of 723
-------
information received at each step in the process and factored in the information and comments as the
Agency deemed appropriate and relevant including comments on the published problem formulation of
HBCD.
1.1 Physical and Chemical Properties
Physical and chemical properties influence the environmental behavior and the toxic properties of a
chemical, thereby informing the potential conditions of use, exposure pathways and routes and hazards
that EPA intends to consider. For scope development, EPA considered the measured or estimated
physical and chemical properties set forth in Table 1-1. EPA found no additional information throughout
the development of the Risk Evaluation that would change these values. Data evaluation results for
physical and chemical properties studies can be found in [Risk Evaluation for Cyclic Aliphatic Bromide
Cluster (HBCD) Systematic Review Supplemental File: Data Quality Evaluation of Physical-Chemical
Properties Studies ( ) |.
HBCD is a white odorless non-volatile solid that is used as a flame retardant. Technical HBCD is often
characterized as a mixture of mainly three diastereomers, a-, P- and y-HBCD with the y-HBCD as main
component (>70%). The fate and biological effects of these compounds are stereoselective, and there is
limited data for the diastereomers. Technical HBCD may contain some impurities, such as
tetrabromocyclododecene or other isomeric HBCDs fUNEP 2010a). which are not included in this Risk
Evaluation. The density of HBCD is greater than that of water (2.24 g/cm3 at 20°C). It has low water
solubility (66 |ig/L at 20°C) and a log octanol:water partition coefficient (log Kow) of 5.62.
Page 46 of 723
-------
Table 1-1. Physical and Chemical Properties of HBCD
Property
Value a
References
Molecular formula
Ci2Hi8Br6
Molecular weight
641.7 g/mole
Physical form
White solid; odorless
CEINECS 2008)
Melting point
Ranges from approximately:
172-184°C to 201-205°C
CEINECS 2008) citing
(Smith et al. 2005)
Boiling point
>190°C (decomposes)
(Peled et al. 1995)
Density
2.24 g/cm3
CEINECS 2008)
Vapor pressure
4.7E-07 mmHg at 21°C
(Wildlife Intl 1997c)
Vapor density
Not readily available
CEINECS 2008)
Water solubility
66 |ig/L at 20°C
CEINECS 2008) citing
(MacGregor and Nixon
2004)
Octanol:water partition coefficient
(log Kow)
5.625 at 25°C
(Wildlife Intl 1997a)
Henry's Law constant
7.4E-06 atm-m3/mole (calculated)
(U.S. EPA 2012b)
Flash point
Not readily available
Autoflammability
Decomposes at >190°C
CEINECS 2008)
Viscosity
Not readily available
Refractive index
Not readily available
Dielectric constant
Not readily available
a Measured unless otherwise noted.
1.2 Uses and Production Volume
1.2.1 Data and Information Sources
The summary of use and production volume information for HBCD presented below is based on
research conducted for the Problem Formulation Document for Cyclic Aliphatic Bromide Cluster
(HBCD) and any additional information that was obtained since the publication of that document. This
research was based on reasonably available information, including the Use and Market Profile for
HBCD, (EPA-HQ-OPPT-2016-0735-00491 public meetings, and meetings with companies, industry
groups, chemical users and other stakeholders to aid in identifying and verifying the conditions of use
(COUs) included in this final Risk Evaluation. The information and input received from the public,
stakeholder meetings and the additional contacts were incorporated into this section, as applicable.
1.2.2 Domestic Manufacture of HBCD
Domestic manufacture of HBCD ceased as of 2017 and is not intended, known, or reasonably foreseen
to occur, and is therefore not considered a COU in this final Risk Evaluation.
Page 47 of 723
-------
A shown in Table 1-2, data reported for the CDR period for 2016 for HBCD indicate that between 1 and
10 million pounds of each CASRN were manufactured in or imported into the United States in 2015; the
national production volume is CBI (U.S. EPA 2016d). These are the most recent CDR data available.
The data provides an overview of the historic trends in production volume of HBCD. For both CASRNs,
site-specific production volumes for the 2015 reporting year were withheld as TSCA CBI. Six firms
comprising nine sites are identified by the 2016 CDR as manufacturers or importers of HBCD:
Chemtura Corporation, Albemarle Corporation, Dow Chemical Company, Campine NV, BASF
Corporation, and Styropek USA, Inc (U.S. EPA 2016d). ICL-IP previously manufactured an HBCD-
containing flame retardant marketed as FR-1206. This product has been discontinued, and ICL-IP has
reportedly ceased production of products containing HBCD (Anon. 2015). The 2016 CDR reporting data
for HBCD from EPA's CDR database (U.S. EPA 2016d) are provided in Table 1-2. CDR data collection
occurs every four years (next reporting period will be in 2020); this information has not changed from
that provided in the 2018 HBCD Problem Formulation.
Table 1-2. Production Volume (Manufacture and Import) of HBCD in CDR Reporting Period
(2012 to 2015)a
Reporting Year
2012
2013
2014
2015
Total Aggregate
Production Volume (lbs)
CASRN 25637-99-4
1-10 million
1-10 million
1-10 million
1-10 million
CASRN 3194-55-6
10-50 million
10-50 million
1-10 million
1-10 million
a The CDR data for the 2016 reoortine oeriod is available via ChemView (httt>s://iava.er>a.eov/chemview) (U.S. EPA
2016d).
U.S. manufacturers have indicated complete replacement of HBCD in their product lines (U.S. EPA
2017i) and that depletion of stockpiles and cessation of export was completed in 2017 based on
communications with recent manufacturers. According to the North American Flame Retardant Alliance
(NAFRA), "HBCD is no longer domestically manufactured or imported and NAFRA members have
worked with downstream users to transition to newer technologies that have an improved environmental,
health, and safety profile while also providing critical fire safety benefits" (ACC/North American Flame
Retardant Alliance. 2019).
Communication with Chemtura (now called Lanxess Solutions, US) indicates that the company has not
manufactured HBCD since 2015, and that there are currently no U.S. manufacturers of the chemical.
The company does not intend to manufacture, import, or export HBCD in the future and has no existing
stockpiles (LANXESS 2017). Albemarle Corporation, another historic manufacturer of HBCD,
indicated that they stopped manufacturing HBCD flame retardants in 2016 and do not intend to resume
the manufacture of HBCD-based flame retardants. In 2017, Albemarle exported its entire inventory of
approximately 57 metric tons (MT) of HBCD to Mexico and Turkey for use in construction (XPS/EPS)
applications. Albemarle does not intend to import HBCD in the future (Albemarle 2017). Dow Chemical
developed the polymeric flame retardant that replaced HBCD for use in insulation boards used in
construction. It is licensed to other manufacturers including Albemarle, Chemtura, and Bromine
Compounds Limited (part of ICL Industrial Products); these companies sell the chemical under different
trade names. Consideration of the status of manufacturing, availability of viable substitutes and the
international regulatory focus on phasing out of domestic manufacturing, use and international trade in
HBCD has led EPA to conclude that domestic manufacturing of HBCD is not known, intended, or
reasonably foreseen to occur.
Page 48 of 723
-------
In their public comment on the draft Risk Evaluation, the North American Flame Retardants Alliance
(NAFRA) of the American Chemistry Council stated that HBCD is no longer domestically
manufactured or imported and NAFRA members have worked with downstream users to transition to
newer technologies that have an improved environmental, health, and safety profile while also providing
critical fire safety benefits. NAFRA represents the former major manufacturers and importers of HBCD
and the possibility remains that small businesses are importing the chemical.
Table 1-3 below presents the various conditions under which a company must report to CDR ("x"
indicates reporting required) for the 2016 reporting period. Typically, a manufacturer is required to
report any volume above 25,000 pounds, while small manufacturers5 are only required to report any
volume above 100,000 pounds. Since HBCD is subject to a TSCA Section 5(a)(2) Significant New Use
Rule (SNUR), the reporting threshold has been reduced to 2,500 pounds for large size firms. For small
manufacturers, however, the threshold remains at 100,000 pounds. EPA has no indication that small
manufacturers are manufacturing HBCD and concludes that manufacturing of HBCD is not reasonably
foreseen and therefore is excluded as a Condition of Use in this final Risk Evaluation.
Table 1-3. Conditions under Which a Company Must Report to CDR (shaded area applies to
IBCD reporting specifically and "x" indicates broad conditions requiring reporting)
Obligation to Report to CDR Information When Subject to TSCA Action as
Indicated in Left column
TSCA Action
Subject to 25,000 lb.
reporting threshold
Subject to 2,500 lb
reporting threshold
Not eligible for
certain full or
partial exemptions
from reporting
Not eligible for
small manufacturer
exemption
Not subject to TSCA action
X
TSCA section 4 rules
(proposed or promulgated)
X
X
X
Enforceable Consent
Agreements (ECAs)
X
X
TSCA section 5(a)(2) SNURs
(proposed or promulgated)
X
X
TSCA section 5(b)(4) rules
(proposed or promulgated)
X
X
X
TSCA section 5(e) orders
X
X
X
TSCA section 5(f) orders
X
X
TSCA section 5 civil actions
X
X
X
TSCA section 6 rules
(proposed or promulgated)
X
X
X
TSCA section 7 civil actions
X
X
X
5 The definition of a small manufacturer varies depending on the sector.
Page 49 of 723
-------
1.2.3 Importation of HBCD
In 2011, the total global production of HBCD was estimated at approximately31,000 metric tons in
2011, of which about 13,000 tons were produced in EU countries and in the United States, and 18,000
tons in China fUNEP 2011). This volume is expected to have decreased following the agreement by
parties to the United Nations Stockholm Convention on Persistent Organic Pollutants (POPs), in May
2013. Parties to the Convention will develop inventories of HCBD in the future fUNEP 201 1).
The companies that previously reported HBCD import volumes to CDR have stated to EPA that they
permanently stopped their import activity in 2016 or 2017. The Dow Chemical Company imported 19
metric tons (MT) of HBCD in 2016 and roughly 48 MT in 2017. Dow possessed roughly 41 MT of
HBCD in stockpiles as of September 2017, which the company then used to produce XPS foam. By
November 2017, Dow had stopped using HBCD at all of its plants and had no intention of importing
HBCD in the future (Dow Chemical 2017). As noted above, Dow developed the polymeric flame
retardant called BlueEdge for use in construction insulation boards that replaced HBCD. It is licensed to
other manufacturers including Albemarle, Chemtura, and Bromine Compounds Limited (part of ICL
Industrial Products); these companies sell the chemical under different trade names.
Similarly, Campine NV indicated in a correspondence with EPA that they had ceased importation of
HBCD in 2016 (Campine 2017). BASF has indicated in a correspondence with EPA (BASF 2017) that
the company ceased importing HBCD in 2016 and has no remaining stockpiles of the chemical.
Styropek, another historic importer of HBCD based on CDR, has also indicated in its correspondences
with EPA that the company phased out the use of HBCD as a flame retardant in 2016.
Datamyne (http://www.datamyne.com) collects import data on shipments into the United States and
provides information on each shipment. Datamyne is a commercial searchable trade database that covers
the import and export data and global commerce of more than 50 countries across 5 continents
(approximately 76% of the world's import trade by value) and includes the cross-border commerce of
the United States with over 230 trading partners. EPA queried the database for bills of lading related to
HBCD. Due to the nature of Datamyne data, some shipments containing the chemical of concern may be
excluded due to being categorized under other names that were not included in the search terms.
Datamyne does not include articles/products containing the chemical unless the chemical name is
included in the description of the article/product. Datamyne indicates that there was import of HBCD in
2016 and 2017, however, shows no import in 2018 to July 2020 when the last search was conducted for
this assessment as shown in Table 1-4. .
Table 1-4. U.S. Volume of Imports of HBCD, 2016 through July 2020
Year
Total Import Volume (kgs)
Number of Unique Consignees
2016
399,315
5a
2017
46,096
1
2018-2020 (July)
0
0
a One consignee did not declare their name.
Source(s): httD://\vww.datamvnc.com
Although there are a number of possible source countries for importation of HBCD to the United States,
under the United Nations Stockholm Convention on Persistent Organic Pollutants (POPs), 171 of the
188 Parties (countries) have agreed to ban the production, use, import, and export of HBCD, consistent
with the obligations of that Convention (SCCH 2018a. b). The Convention does include a process by
Page 50 of 723
-------
which a party can apply for a time limited exemption to continue production and/or use of a listed
chemical, however, that exemption is limited to the specific use(s) identified in the Convention. In
accordance with Article 4, specific exemptions expire five years after the date of entry into force of the
Convention with respect to that particular chemical, unless an additional five-year extension is granted
by the Conference of the Parties (SCCH2018b). For HBCD, the specific uses for which a Party can
register a production or use exemption is limited to use "in EPS and XPS in buildings." According to the
Register of Specific Exemptions for the Convention when accessed in 2018, there were three Parties
registered for production for those uses and six Parties registered for use. In July 2020, two exemptions
remained in effect. The United States and approximately six other countries are not Parties to the
Convention (SCCH 2018c).
EPA has no direct evidence of current import of HBCD, however, there are several countries that have
not agreed to the Stockholm Convention HBCD ban or are not Parties to the Convention and therefore
can still export HBCD legally to the United States. Domestic firms could import quantities of up to
100,000 lbs of HBCD per year without reporting to the CDR. Given these facts, EPA is considering the
import of HBCD to be known and/or reasonably foreseen and is including it as a COU in this final Risk
Evaluation.
1.2.4 Toxics Release Inventory Data on HBCD
Following the publication of the Problem Formulation in 2018, information became available for HBCD
as reported by facilities to the Toxics Release Inventory (TRI) program. Under the Emergency Planning
and Community Right-to-Know Act (EPCRA) Section 313, HBCD is a TRI-reportable category6
effective January 1, 2017 and EPA has finalized the addition of the HBCD category to the list of
chemicals with special concern (see 40 CFR 372.28(a)(2)) and established a 100 lb reporting threshold.
Four facilities reported HBCD for the 2017 TRI reporting year; follow-up with the companies indicates
that only one facility is involved in ongoing processing of HBCD. Two facilities belong to Dow
Chemical, which said it stopped producing HBCD by 2018 ( ). A third facility, owned
by Flame Control Coatings, said in 2018 that it had stopped using HBCD for manufacture of coatings
(Flame Control Coatings. 2.018). The fourth facility. Indium Corporation of America, continues to
process HBCD for use in the manufacture of solder paste (see more about this use in Section 1.2.5.3).
Table 1-5 provides production-related waste management data for HBCD reported by subject facilities
to the TRI program for reporting years 2017 and 20187. In reporting year 2017, four facilities reported a
total of approximately 724 lbs of HBCD waste managed. Of this total, zero lbs were recycled, 51 lbs
were recovered for energy, 82 lbs were treated, and 591 lbs were disposed of or otherwise released into
the environment. In reporting year 2018, only one facility (Indium Corporation of America) reported to
the TRI program for HBCD.
6 The HBCD category covers HBCD as identified through two primary Chemical Abstracts Service Registry Numbers
(CASRNs): 3194-55-6 (1,2,5,6,9,10-hexabromocyclododecane) and 25637-99-4 (hexabromocyclododecane).
7 Reporting year 2017 was the first year available for HBCD and reporting year 2018 is the most recent TRI data year. Data
presented in Table 1-5 and Table 1-6 were queried using TRI Explorer and uses the 2018 National Analysis data set (released
to the public in November 2019).
Page 51 of 723
-------
Table 1-5. Summary of HBCD TRI Production-Related Waste Managed from 2017-2018 (lbs)
Year
Number of
Facilities
Recycling
Energy
Recovery
Treatment
Releases a'b'c
Total Production
Related Waste
2017
4 d
0
51
82
591
724
2018
1
0
0
0
3.6
3.6
Data source: 2017-2018 TRI Data (Undated November 2019) (U.S. EPA 2017h).
a Terminology used in these columns may not match the more detailed data element names used in the TRI public data and
analysis access points.
b Does not include releases due to one-time events not associated with production such as remedial actions or earthquakes.
0 Counts all releases including release quantities transferred and release quantities disposed of by a receiving facility
reporting to TRI.
d Reporting facilities include: The Dow Chemical Company (2 locations). Flame Control Coatings LLC, and Indium
Corporation of America.
Table 1-6 provides a summary of HBCD TRI releases to the environment for the same reporting years as
Table 1-5. There were zero pounds of HBCD reported as released to water via surface water discharges,
and a total of 79 lbs of air releases from collective fugitive and stack air emissions reported in 2017. The
majority of HBCD was disposed of to landfills other than Resource Conservation and Recovery Act
(RCRA) Subtitle C (511 lbs), and there was one pound of HBCD transferred to a waste broker for
disposal. In reporting year 2018, Indium Corporation of America reported one pound of stack air
emissions of HBCD and 2.6 lbs of HBCD sent off-site to a waste broker for disposal.
Page 52 of 723
-------
Table 1-6. Summary of HBCD TRI Releases to the Environment from 2017-2018 (lbs)
Number
of
Facilities
Air Releases
Water
Releases
Land Disposal
Other
Releasesa
Total On- and
Off-Site Disposal
or Other
Releases b'c
Stack
Air
Releases
Fugitive
Air
Releases
Class I
Under-
ground
Injection
RCRA
Subtitle C
Landfills
All other
Land
Disposala
Totals
2017
4 e
77
2
0
0
0
511
1
591
79 d
511 d
Totals
2018
1
1
0
0
0
0
0
2.6
3.6
1
0
Data source: 2017-2018 TRI Data (Undated November 2019) (U.S. EPA 2017h).
a Terminology used in these columns may not match the more detailed data element names used in the TRI public data and
analysis access points.
b These release quantities do include releases due to one-time events not associated with production such as remedial
actions or earthquakes.
0 Counts release quantities once at final disposition, accounting for transfers to other TRI reporting facilities that ultimately
dispose of the chemical waste.
d Value shown may be different than the summation of individual data elements due to decimal rounding.
e Reporting facilities include: The Dow Chemical Company (2 locations). Flame Control Coatings LLC, Indium
Corporation of America.
While production-related waste managed shown in Table 1-5 excludes any quantities reported as
catastrophic or one-time releases (TRI section 8 data), release quantities shown in Table 1-6 include
both production-related and non-routine quantities (TRI section 5 and 6 data) from 2017-2018. As a
result, release quantities may differ slightly and may further reflect differences in TRI calculation
methods for reported release range estimates (U.S. EPA 2017h).
1.2.5 Uses of HBCD
Descriptions of the industrial, commercial and consumer use categories identified from the 2016 CDR
(U.S. EPA 2016d) and included in the life cycle diagram are summarized in Section 1.4.1. The
descriptions provide a brief overview of uses by life cycle stage in Figure 1-1. The descriptions provided
below are primarily based on the corresponding industrial function category and/or commercial and
consumer product category descriptions from the 2016 CDR and can be found in EPA's Instructions for
Reporting 2016 TSCA Chemical Data Reporting (U.S. EPA 2016b).
1.2.5.1 Automobile Replacement Parts
EPA received a public comment from the Global Automakers Association stating that HBCD is no
longer used in new automobile manufacturing and is only present in replacement parts manufactured
prior to the date of the EPA HBCD Scoping Document (EPA-HQ-QPPT-2016-073 5-0027). Major
automobile manufacturers have phased out use of HBCD in U.S. automobile and part production but
continue to use it in 155 replacement parts, according to a list provided to EPA by the Alliance of
Automobile Manufacturers in November 2018 after publication of the Problem Formulation. For
approximately 80% of the automobile replacement parts, the HBCD is in polystyrene headliners; most
of the remaining 20% are other parts made with HBCD-containing polystyrene or other plastics. A total
of five parts have HBCD in solder (Alliance of Automobile Manufacturers. 2018). A public comment by
Page 53 of 723
-------
the Alliance of Automobile Manufacturers the following year (EP A-HQ-OPPT-2019-023 7-0049) stated
that "data collected by Alliance members and submitted previously to EPA confirmed that HBCD is
present only in automotive replacement parts and is not found in production parts used in new vehicle
assembly. Our data also shows that HBCD is being phased out (or has already been phased out) of
replacement parts." EPA includes the processing and use of HBCD in automobile replacement parts in
this final Risk Evaluation.
1.2.5.2 Expanded Polystyrene (EPS) and Extruded Polystyrene (XPS) Foam
Use in EPS and XPS foam had historically accounted for 95% of all HBCD applications(u a iir /i
2014d; LINEP 2010a). Based on information from market reports (U.S. EPA. 2017i). HBCD was used
primarily in construction materials, which may include structural insulated panels (SIPS).
"Building/Construction Materials" include products containing HBCD as a flame retardant primarily in
XPS and EPS foam insulation products that are used for the construction of residential, public,
commercial or other structures (UNEP 2010a; Weil and Levchik 2009). The building and construction
industry has used EPS and XPS foam thermal insulation boards and laminates for sheathing products.
HBCD is added to EPS and XPS foam in the form of a resin. EPS foam prevents freezing, provides a
stable fill material and creates high-strength composites in construction applications. XPS foam board is
used mainly for roofing applications and architectural molding. HBCD is used in both types of foams
because it is highly effective at levels less than 1% and, therefore, maintains the insulation properties of
EPS and XPS foam (Morose 2006). EPS foam boards contain approximately 0.5% HBCD by weight in
the final product and XPS foam boards contain 0.5-1% HBCD by weight (Public comment, EPA-HQ-
QPPT-2016-073 5-0017; XPS A. 2017b; U.S. EPA. 2014d; Morose 2006).
According to the EPS Industry Alliance (EPS-IA), an estimated 80-85%) of EPS rigid foam insulation
manufactured in the United States is molded from EPS resins supplied by EPS-IA member companies,
none of whom use HBCD.
The XPS Association (XPSA) stated that its members, who are the major producers of XPS resin, supply
the resin for more than 95% of the XPS foam insulation products manufactured for the North American
market and that the remaining small percentage is probably made using imported resin (XPSA.: ).
This imported resin may contain HBCD, however, the extent to which EPA does not know.
Although some of the industry comments on the draft Risk Evaluation indicate more certainty than
previous comments that the phaseout of HBCD is complete, the associations do not represent every
possible importer and processor of HBCD. There is a potential for import of HBCD in the form of a
resin for use in the manufacture of EPS and XPS foam insulation. Taking into account the high
percentage of HBCD production volume dedicated to these two uses in previous years, and the fact that
small quantities of HBCD could be imported at volumes below the CDR reporting threshold leaves open
the possibility that EPS and XPS manufacturers that are not members of the EPS-IA and XPSA may
currently be using imported HBCD resins in their processes. EPA includes the processing and use of
HBCD in XPS and EPS insulation in the final Risk Evaluation.
1.2.5.3 Flux/Solder Paste
Following the publication of the HBCD Problem Formulation document (U.S. EPA. 2018g). EPA
learned of an ongoing use of HBCD from newly available TRI data reported by the Indium Corporation.
As indicated in Table 1-5. and Table 1-6, the company submitted TRI reporting forms to the TRI
program for HBCD in reporting years 2017 and 2018. In follow-on communications with EPA, Indium
said it processes and uses HBCD as a fluxing aid in solder paste, which it supplies to electronics
manufacturers for use on circuit boards (Indium 2018b). While the quantity of HBCD is unknown, EPA
Page 54 of 723
-------
assumes it is greater than the TRI reporting threshold of 100 lbs per year for HBCD. According to the
company, the amount of HBCD used varies depending on demand from customers. The company
purchased HBCD in a formulated mixture from a single supplier to manufacture flux and solder paste
(Indium 2.018a). The supplier (HM Royal) informed EPA that they no longer sell HBCD. Indium
reported in 2017 to TRI that the maximum amount of HBCD on-site at any one point during the calendar
year was between 1000 to 9,999 lbs. In 2018, the amount reported was 10,000 to 99,000 lbs.
In an email to EPA, Indium said they ship their products to their overseas facilities for the final mixing
step and for sales to electronics manufacturers in China and the United States. They said the company
does not sell directly to consumers, although the final consumer electronics products might be imported
into the United States. Also according to the representative, Indium no longer ships the HBCD-
containing products to the Ell (Indium 2.018a. b). Kester, another company, used HBCD in the past to
manufacture solder paste, but in a phone conversation with EPA indicated that they have discontinued
use (Kester 2018).
Based on the information above, EPA includes the processing of HBCD in the manufacture of solder
paste in this final Risk Evaluation.
1,2.6 Recycling of EPS and XPS Foam
There is limited information about the recycling of EPS and XPS products containing HBCD.
Schlummer et al. (2017) notes that EPS and XPS foam in construction insulation materials may not be
frequently recycled for numerous reasons, including that insulation waste is typically not separated from
mixed waste stream and most insulation containing HBCD is still in place. Schlummer et al. (2017)
describe technologies available only on a small scale to separate HBCD from insulation panels and
recycled polystyrene.
Reuse and recycling of EPS and XPS foam insulation board, siding, roof membrane and roofing ballast
material are available in the United States for consumers. Two companies were identified that directly
reuse (e.g., reuse without reforming) and recycle (e.g., melting and inserting into the manufacturing
process) XPS and EPS foam insulation.
• Green Insulation Group: http://www.greeninsulationgroup.com/products/
• Nationwide Foam Recycling: http://nationwidefoam.com/what-vou-can-recvcle.cfm
Nationwide Foam Recycling, which is owned by Conigliaro Industries, Inc., indicated that their plant
recycles all EPS insulation and reuses all XPS insulation ( 017i). Once processed, their
recycled EPS roofing insulation is taken to polystyrene product manufacturers, notably picture frame
manufacturers, mostly in China. The company also delivers recycled roofing material to other local EPS
recycling plants that may use different processes. Nationwide Foam Recycling processes 90,000 lbs/year
of EPS standard packaging and 10,000 lbs/year of EPS roofing material and estimated only about 10-
20% of EPS roofing material is recycled nationally (U.S. EPA. 2017i). It is not clear what happens to the
remaining volume of waste. The company also reuses XPS roofing material due the special equipment
needed to recycle XPS and indicated that XPS is rarely recycled in the United States. It was estimated
that the majority (>50%) of XPS roofing material is sent to landfills or waste energy plants. Processing
estimates for XPS material were not provided by the company.
The recycling of HBCD-containing EPS and reuse of XPS insulations boards for use in construction
materials is a COU in this final Risk evaluation. Recycling of a product containing a chemical
constitutes processing of the chemical, which is a COU. HBCD was broadly used in EPS and XPS
insulation boards historically, and recycled construction material would typically be required to meet
Page 55 of 723
-------
fire resistant construction codes. EPA believes that this recycling of insulation materials occurs such that
the flame-retardant attributes of the insulation boards is maintained. EPA includes this recycling and the
use of HBCD in the recycled boards in the scope of this final Risk Evaluation. EPA also includes
consumer articles made from recycled HBCD-containing insulation boards based on experimental
product-testing information on HBCD content in consumer articles, and recognition that this as an
important pathway for infants and young children who may exhibit mouthing behaviors.
1.2.7 Recycling of Electronics Waste (E-Waste) Containing HIPS
While only anecdotal information is available indicating HBCD use in high impact polystyrene (HIPS)
in electronics occurred in the United States (Section 2.2), there are more substantial data from the EU
indicating a range of between 2 and 7 percent of HBCD production volume in Europe was historically
used in HIPS and that the majority of HIPS was used in electronics (Leisewitz et at.. 2001; ECHA
2008b). This makes it likely that electronics products with HBCD-containing HIPS were imported into
the United States in past years. EPA believes that it is reasonably foreseen that HBCD will be finding its
way into recycling of electronics waste and therefore included a COU to Table 1-8: Processing -
Recycling - Recycling of electronics waste containing HIPS that contain HBCD.
1.2.8 Legacy Activities and Uses
For the first 10 risk evaluation chemicals (including HBCD), EPA initially excluded chemical uses for
which ongoing and prospective manufacturing, processing, and distribution had ceased; such uses were
referred to as "legacy use," a term no longer used for risk evaluations. EPA also excluded "associated
disposal," which meant "future disposal of a chemical substance that is no longer manufactured,
processed, or distributed for use." (Risk Evaluation Rule, 82 FR at 33729.) In the final risk
determination, EPA did not exclude any activity determined to be a COU.
In developing the scope for HBCD, EPA learned that HBCD was no longer used to manufacture four
minor-use products or articles: adhesives, coatings, HIPS in electronics, and textiles8 (evidence for use
in HIPS in electronics was anecdotal).9 These so-called "legacy uses" were excluded from the scope
along with related activities or disposal in later stages of the chemical life cycle, such as commercial/
consumer use or disposal of HBCD-containing products and articles for which HBCD manufacture,
processing, and distribution for use in such products/articles has ceased. (Problem Formulation for
Cyclic Aliphatic Bromides Cluster, Section 1.2.7). The designation was published in the Problem
Formulation (Section 2.2.2.1) and draft Risk Evaluation (Section 1.2.7). EPA received public comments
stating that the HBCD risk evaluation should include "legacy use." In 2019, the Ninth Circuit Court of
Appeals ruled that EPA cannot categorically exclude "legacy use" and "associated disposal" from risk
evaluations {Safer Chemicals, Healthy Families v. U.S. Envtl. Prot. Agency, 943 F.3d 397, 425 (9th Cir.
2019)).
8 Available information indicates that only a small amount of HBCD was used for these and other minor products and
articles. At least 95% of the total production volume was processed to manufacture XPS/EPS insulation. (CDR 2012). By
2018, a single company was identified as having processed HBCD in the manufacture of adhesives in the past and only one
company was found that had processed HBCD for coatings manufacturing. The evidence of past processing of HBCD
processed in HIPS for electronics articles was antecdotal. Use of HBCD to process consumer textiles had phased out by
2011.
9 The draft Risk Evaluation also erroneously included other articles as no longer being manufactured with the use of HBCD.
These were children's products (including toys and car seats) and furniture (such as bean bag chairs) (Draft Risk Evaluation
Section 2.2.2. l). In fact, HBCD's search returned no reliable information that HBCD ever was used in the processing for these
articles (Problem Formulation, Section 2.2.2.1).
Page 56 of 723
-------
Due to the court ruling, EPA reconsidered the HBCD-containing products and articles to determine if
additional COUs needed to be evaluated in the Risk Evaluation. The four minor-use products and
articles could still be in service, for example textiles containing HBCD may be in seating in public
buildings, and conveyances and electronics products or components in aircraft, office buildings,
residences, or other indoor environments. Migration of the HBCD from the products and articles can
expose occupants to HBCD in indoor air or dust. In addition, some items may be in the process of
disposal. So although they are no longer made and sold, the HBCD in the products and articles are still
conditions of use.
1.2.9 Historical Activities Resulting in Continued Exposures
In addition to the possibility of releases from use and disposal of products and articles that are no longer
manufactured, processed, or distributed, exposure can occur from historical activities not associated with
a current COU. This is due to HBCD's expected persistence in the environment (Section 2.1.2.5).
HBCD may continue in environmental media and indoor dust long after the conclusion of a COU or a
product's life cycle Exposure from these historical releases are accounted for in the background
exposure assessments performed for the environment (Section 2.3.2.1) and general population (Section
2.4.2). The measured levels of HBCD are not linked to specific sources and it is not reasonable to
attempt to estimate the quantity of HBCD, if any, that originated solely from HBCD-containing products
and articles which are no longer known, intended, or reasonably foreseen to occur.
1.2.10 Summary
Domestic manufacture of HBCD had ceased as of 2017 and is not intended, known, or reasonably
foreseen, and is therefore not a COU in this final Risk Evaluation.
Available import data indicate that there was import of HBCD in 2016 and 2017. Under the United
Nations Stockholm Convention on Persistent Organic Pollutants (POPs), 171 of the 188 Parties
(countries) have agreed to ban the production, use, import, and export of HBCD; however, time-limited
exemptions for certain uses exist. Given these exemptions and the possibility that small firms could
import quantities of up to 100,000 lbs of HBCD per year without being required to report to the CDR.
EPA includes the import of HBCD as a COU in this final Risk Evaluation.
Major automobile manufacturers have phased out use of HBCD in U.S. production of new automobiles
and parts but continue to use it in 155 replacement parts, according to a list provided to EPA by the
Alliance of Automobile Manufacturers. The Association was unable to confirm whether the 155 parts
are domestically manufactured or imported. EPA includes the use of HBCD in automobile replacement
parts in this final Risk Evaluation.
HCBD was extensively used in EPS and XPS foam insulation products used in the construction of
residential, public, commercial or other structures. Based on industry association data, manufacturers in
the United States are no longer using HCBD but a small percentage of EPS and XPS is probably made
using imported resin that could contain HCBD. Therefore, EPA includes the use of HBCD in XPS and
EPS insulation using imported HBCD in this final Risk Evaluation.
HBCD is used as a fluxing aid in solder paste, which is supplied to electronics manufacturers for use on
circuit boards. Therefore, EPA includes the processing of HBCD in the manufacture of flux/solder paste
in this final Risk Evaluation.
Based on current practices identified, the recycling of HBCD-containing EPS and XPS insulations
boards for use in construction materials is included as a COU in this final Risk Evaluation. EPA includes
Page 57 of 723
-------
consumer articles made from recycled HBCD-containing insulation boards based on experimental
product-testing information on HBCD content in consumer articles.
HIPS for electrical and electronic appliances may have been imported into the United States in past
years. EPA believes that the use of HBCD in HIPS in electronics is an ongoing use and that these
products may still be present in recycling facilities or may be otherwise currently being used in a manner
that creates exposure potential. Recycling of HBCD containing HIPS in waste electronics is included as
a COU in this final Risk Evaluation. Aggregate exposure was applied for the general population and
consumers, incorporating background aggregate exposures for all exposure routes combined with
specific modeled exposures for the pathway of interest.
1.2.11 List of Conditions of Use
The four COUs added for the final Risk Evaluation are shown in bold.
1. Manufacture - Import
2. Processing - Incorporated into formulation, mixture, or reaction product - Flame retardants used
in custom compounding of resin (e.g., compounding in XPS masterbatch) and in solder paste
3. Processing - Incorporation into article - Flame retardants used in plastics product manufacturing
(manufacture of XPS and EPS foam; manufacture of structural insulated panels (SIPS) and
automobile replacement parts from XPS and EPS foam)
4. Processing - Recycling - Recycling of XPS and EPS foam, resin, panels containing HBCD
5. Processing - Recycling - Recycling of electronics waste containing HIPS that contain
HBCD
6. Distribution
7. Commercial/Consumer Use - Building/construction materials - Plastic articles (hard):
construction and building materials covering large surface areas (e.g., XPS/EPS foam insulation
in residential, public and commercial buildings, and other structures) and solder paste
8. Commercial/Consumer Use - Other - Automobile replacement parts
9. Commercial/Consumer Use - Other - Plastic and other articles
10. Commercial/Consumer Use - Other - Formulated products and articles
11. Disposal - Disposal— Other land disposal (e.g., Construction and demolition waste)
12. Disposal - Disposal— Other land disposal (e.gFormulated products and articles)
1.3 Regulatory and Assessment History
EPA conducted a search of existing domestic and international laws, regulations and assessments
pertaining to HBCD. EPA compiled this summary from data available from federal, state, international
and other government sources, as cited in Table 1-7.
Page 58 of 723
-------
Federal Laws and Regulations
HBCD is subject to federal statutes or regulations, other than TSCA, that are implemented by other
offices within EPA and/or other federal agencies/departments. A summary of federal laws, regulations
and implementing authorities is provided in Appendix A. 1.
State Laws and Regulations
HBCD is subject to state statutes or regulations implemented by state agencies or departments. A
summary of state laws, regulations and implementing authorities is provided in Appendix A.2.
Laws and Regulations in Other Countries and International Treaties or Agreements
HBCD is subject to statutes or regulations in countries other than the United States and/or international
treaties and/or agreements. A summary of these laws, regulations, treaties and/or agreements is provided
in Appendix A. 3.
EPA has identified assessments conducted by other EPA Programs and other organizations. Depending
on the source, these assessments may include information on conditions of use, hazards, exposures and
potentially exposed or susceptible subpopulations. Table 1-7. shows the assessments that have been
conducted.
Table 1-7. Assessment History of HBCD
Authoring Organization
Assessment
EPA assessments
EPA, Office of Chemical Safety and Pollution
Prevention (OCSPP), Office of Pollution Prevention and
Toxics (OPPT)
Initial Risk Based Prioritization of High
Production Volume Chemicals.
C hemi cal/C ategory:
Hexabromocyclododecane (HBCD) (U.S.
EPA 2008a)
EPA, OCSPP, OPPT
Hexabromocvclododecane (HBCD) Action
Plan (U.S. EPA 2010)
EPA, OCSPP, OPPT
Flame Retardant Alternatives for
Hexabromocvclododecane (HBCD) (U.S.
EPA 2014d)
EPA, OCSPP, OPPT
Toxic Chemical Work Plan Problem
Formulation and Initial Assessment for
HBCD, Cvclic Aliphatic Bromide Cluster
(U.S. EPA 2015a)
EPA, OCSPP, OPPT
Scope of the Evaluation for Cvclic Aliphatic
Bromide Cluster (HBCD) (U.S. EPA 2017
EPA, OCSPP, OPPT
Problem Formulation for Cvclic Aliphatic
Bromide Cluster (HBCD) (U.S. EPA 2018a)
Other U.S.-based Organizations
Page 59 of 723
-------
Authoring Organization
Assessment
Consumer Product Safety Commission (CPSC)
CPSC Staff Exposure and Risk Assessment of
Flame Retardant Chemicals in Residential
Upholstered Furniture (CPSC 2001)
National Research Council
National Academy of Sciences Report:
Toxicological Risks of Selected Flame
Retardant Chemicals (NRC 2000a)
International
Organisation for Economic Co-operation and
Development (OECD), Screening Information Data Set
(SIDS)
OECD SIDS Initial Assessment Profile
(SIAP) (OECD 2007)
European Commission (EC), European Chemicals
Bureau
European Union Risk Assessment Report,
Hexabromocvclododecane CASRN 25637-
99-4. EINECS No: 247-148-4 (EINECS
2008)
United Nations Environment Programme (UNEP);
United Nations Stockholm Convention on Persistent
Organic Pollutants (POPs)
Hexabromocvclododecane Draft Risk Profile
(UNEP 2010a)
Hexabromocvclododecane Risk Management
Evaluation (2011) (UNEP 2011)
Environment Canada and Health Canada
Draft Screening Assessment of
Hexabromocvclododecane (EC/HC 2011)
Australian Government Department of Health, National
Industrial Chemicals Notification and Assessment
Scheme (NICNAS)
Priority Existing Chemical Assessment
Report, Hexabromocvclododecane (NICNAS
2012a)
1.4 Scope of the Evaluation
1.4.1 Conditions of Use Included in the Risk Evaluation
TSCA Section 3(4) defines the conditions of use as '' the circumstances, as determined by the
Administrator, under which a chemical substance is intended, known, or reasonably foreseen to be
manufactured, processed, distributed in commerce, used, or disposed of" To determine the conditions
of use of HBCD and inversely, activities that do not qualify as conditions of use, EPA conducted
extensive research and outreach, as described in detail in Problem Formulation Document for Cyclic
Aliphatic Bromide Cluster (HBCD) (U.S. EPA 2018g). Section 1.2 above summarizes these findings and
provides any additional information that was obtained since the publication of that document. EPA did
not evaluate activities that EPA concluded do not constitute conditions of use - for example, because
EPA has insufficient information to find certain activities are circumstances under which the chemical is
actually "intended, known, or reasonably foreseen to be manufactured, processed, distributed in
commerce, used, or disposed of." EPA did not exclude from this final risk evaluation any use constituted
to be COU and a determination was made on each COU.
Page 60 of 723
-------
Based on the information described in Section 1.2, EPA evaluated the importation of HBCD; processing
of HBCD into automobile replacement parts and use of HBCD in such parts; processing of HBCD into
solder paste and use of HBCD in solder paste; incorporation into formulation, mixture or reaction
product (e.g., compounding of masterbatch XPS); the processing of HBCD for incorporation into
articles (e.g., manufacture of EPS and XPS and the manufacture of structural insulated panels from EPS
and XPS); the industrial, commercial and consumer use of EPS and XPS in construction materials (e.g.,
insulation boards) and in plastic and other articles; distribution; disposal (demolition); and recycling of
XPS and EPS foam, resin, and panels containing HBCD and recycling of electronic waste containing
HIPS that contains HBCD.
Table 1-8 presents the conditions of use and associated exposure scenarios that are considered within the
scope of the Risk Evaluation during various life cycle stages including manufacturing, processing, use
(industrial, commercial, and consumer), distribution and disposal. The information is grouped according
to Chemical Data Reporting (CDR) processing codes and use categories (including functional use codes
for industrial uses and product categories for industrial, commercial and consumer uses), in combination
with other data sources (e.g., published literature and consultation with stakeholders) to provide an
overview of conditions of use. EPA notes that some subcategories of use may be grouped under multiple
CDR categories.
Use categories include the following: "industrial use" means use at a site at which one or more
chemicals or mixtures are manufactured (including imported) or processed. "Commercial use" means
the use of a chemical or a mixture containing a chemical (including as part of an article) in a commercial
enterprise providing saleable goods or services. "Consumer use" means the use of a chemical or a
mixture containing a chemical (including as part of an article, such as furniture or clothing) when sold to
or made available to consumers for their use ( I6d).
To understand conditions of use relative to one another and associated potential exposures under those
conditions of use, Figure 1-1 depicts the life cycle diagram and includes the production volume
associated with each stage of the life cycle. The life cycle diagram for HBCD does not include specific
production volumes because the information was claimed as confidential business information (CBI) in
the 2016 CDR reporting ( 016d).
Page 61 of 723
-------
Table 1-8. Categories and Subcategories of Conditions of Use and Corresponding Exposure Scenario Included in the Scope of the
Risk Evaluation for HBCD a
Life Cycle
Stage
Category a
Subcategory b
Occupational/ Environmental
Exposure Scenario c
Consumer Exposure
Scenario
References
Manufacture
Import
Import
Section 2.4.1.2 - Repackaging of
Import Containers (1)
N/A
(U.S. EPA 2016d)
Processing
Incorporated into
formulation, mixture
or reaction product
Flame retardants used in custom
compounding of resin (e.g.,
compounding in XPS
masterbatch) and in solder paste
Section 2.4.1.3 - Compounding of
Polystyrene Resin to Produce XPS
Masterbatch (2)
Section 2.4.1.12- Formulation of
Flux/Solder Pastes (11)
N/A
(EINECS 2008);
(U.S. EPA 2017a)
Incorporated into
article
Flame retardants used in plastics
product manufacturing
(manufacture of XPS and EPS
foam; manufacture of structural
insulated panels (SIPS) and
automobile replacement parts
from XPS and EPS foam)
Section 2.4.1.4 - Processing of
HBCD to produce XPS Foam using
XPS Masterbatch (3)
Section 2.2.5 - Processing of
HBCD to produce XPS Foam using
HBCD Powder (4)
Section 2.4.1.6 - Processing of
HBCD to produce EPS Foam from
Imported EPS Resin Beads (5)
Section 2.4.1.7 - Processing of
HBCD to produce SIPs and
Automobile Replacement Parts
from XPS/EPS Foam (6)
N/A
Use Document. EPA-
HO-OPPT-2016-
0735-0003; Market
Profile. EPA-HO-
OPPT-2016-0735-
0049; (Alliance of
Automobile
Manufacturers 2018a).
Recycling
Recycling of XPS and EPS
foam, resin, panels containing
HBCD
Section 2.4.1.11 - Recycling of
EPS Foam and Reuse of XPS Foam
(10)
N/A
Use Document. EPA-
HO-OPPT-2016-
0735-0003
Recycling of electronics waste
containing HIPS that contain
HBCD
Section 2.4.1.14- Recycling of
electronics waste containing HIPS
(13)
N/A
Use Document. EPA-
HO-OPPT-2016-
0735-0003
Distribution
Distribution
Distribution
Activities related to distribution (e.g., loading, unloading) are considered throughout
the life cycle, rather than using a single distribution scenario.
Page 62 of 723
-------
Life Cycle
Stage
Category a
Subcategory b
Occupational/ Environmental
Exposure Scenario c
Consumer Exposure
Scenario
References
Commercial/
consumer Use
Building/construction
materials
Plastic articles (hard):
construction and building
materials covering large surface
areas (e.g., XPS/EPS foam
insulation in residential, public
and commercial buildings, and
other structures) and solder paste
Section 2.4.1.9 - Installation of
XPS/EPS Foam Insulation in
Residential, Public, and
Commercial Buildings, and Other
Structures (8)
Section 2.4.1.13 - Use of
Flux/Solder Pastes (12)
Section 2.4.2.3 -
Consumer Exposures
during Use of HBCD in
XPS/EPS Insulation in
Residences and Auto
Components
Use Document. EPA-
HO-OPPT-2016-
0735-0003:(U.S.
EPA 2016d): (U.S.
EPA 2014d)
Other
Automobile replacement parts
Section 2.4.1.8 - Installation of
Automobile Replacement Parts (7)
Section 2.4.2.3 -
Consumer Exposures
During Use of HBCD in
XPS/EPS Insulation in
Residences and Auto
Components
N/A
(Alliance of
Automobile
Manufacturers 2018a)
Plastic and other articles'1
Section 2.4.4.4 -
Mouthing of Articles
Containing HBCD
(Abdallah et al. 2018:
Voita et al. 2017)
Formulated products (e.g.,
adhesives and coatings) and
articles (e.g., textiles, electrical
and electronic products)
Section 2.4.2.2.6 - Occupational
Microenvironments (Workers);
Section 2.4.2 - General Population
Background Exposure (General
Population and Consumers)
Section 2.3.2.1 - Non-Scenario
Specific Approach (Environmental;
Aquatic organisms);
Section 2.3.3.1 and 2.3.3.3 - Non-
Scenario Specific Approach
(Environmental;
Terrestrial organisms)
Section 2.4.2 - General
Population Background
Exposure (Consumers)
Disposal
Disposal
Land disposal of construction
and demolition waste
Section 2.4.1.10- Demolition and
Disposal of XPS/EPS Foam
Insulation Products in Residential,
Public and Commercial Buildings,
and Other Structures (9)
N/A
(EINECS 2008)
Page 63 of 723
-------
Life Cycle
Stage
Category a
Subcategory b
Occupational/ Environmental
Exposure Scenario c
Consumer Exposure
Scenario
References
Land disposal of formulated
products (e.g., adhesives and
coatings) and articles (e.g.,
textiles, electrical and electronic
products)
Section 2.4.5.3 - Occupational
Exposure Associated with Land
Disposal of Formulated Products
and Articles (Workers);
Section 2.4.2 - General Population
Background Exposure (General
Population and Consumers);
Section 2.3.2.1 - Non-Scenario
Specific Approach (Environmental)
a These categories of conditions of use appear in the Life Cycle Diagram, reflect CDR codes, and broadly represent conditions of use of HBCD in industrial and/or commercial
settings.
b These subcategories reflect more specific uses of HBCD.
0 Exposure scenarios are numbered in parentheses. This numbering will be referred to throughout the document, including for exposure subscenarios (e.g., 3.1, 3.2, etc.)
d This COU was inadvertently omitted from Table 1-8 in the draft Risk evaluation.
Page 64 of 723
-------
MFG/IMPORT PROCESSING INDUSTRIAL, COMMERCIAL, CONSUMER USES RELEASES and WASTE DISPOSAL
Demolition
<- Import of HBCD
Recycling (EPS foam, HIPS)
Solder/Flux Paste
Automobile Replacement Parts
Plastic and other articles
Formulated products and articles
(e.g., textiles, adhesives, coatings
electrical and electronic products)
Building/Construction Materials
insulation material for residential, public and
commercial buildings or other structures
Incorporated into Article
(2016 CDR Volume CBI)
manufacture of XPS and EPS,
manufacture of SIPs and automobile
replacement parts from XPS and EPS
Incorporated into Formulation,
Mixture, or Reaction Product
(2016 CDR Volume CBI)
• compounding of XPS masterbatch
• formulation of solder/flux paste
Disposal
Reuse
~ Manufacture (Includes Import) ~ Processing ~ Uses: Industrial, Commercial or Consumer.
Figure 1-1. HBCD Life Cycle Diagram
The life cycle diagram depicts the conditions of use that are within the scope of the Risk Evaluation during various life cycle stages including
manufacturing, processing, use (industrial, commercial, consumer), distribution and disposal. Activities related to distribution (e.g., loading,
unloading) will be considered throughout the HBCD life cycle, rather than using a single distribution scenario.
-------
1.4.2 Exposure Pathways and Risks Addressed by other EPA Administered Statutes
In its TSCA section 6(b) risk evaluations, EPA is coordinating action on certain exposure pathways and
risks falling under the jurisdiction of other EPA-administered statutes or regulatory programs. More
specifically, EPA is exercising its TSCA authorities to tailor the scope of its risk evaluations, rather than
focusing on environmental exposure pathways addressed under other EPA-administered statutes or
regulatory programs or risks that could be eliminated or reduced to a sufficient extent by actions taken
under other EPA-administered laws. EPA considers this approach to be a reasonable exercise of the
Agency's TSCA authorities, which include:
• TSCA section 6(b)(4)(D): "The Administrator shall, not later than 6 months after the initiation of
a risk evaluation, publish the scope of the risk evaluation to be conducted, including the hazards,
exposures, conditions of use, and the potentially exposed or susceptible subpopulations the
Administrator expects to consider.
• TSCA section 9(b)(1): "The Administrator shall coordinate actions taken under this chapter with
actions taken under other Federal laws administered in whole or in part by the Administrator. If
the Administrator determines that a risk to health or the environment associated with a chemical
substance or mixture could be eliminated or reduced to a sufficient extent by actions taken under
the authorities contained in such other Federal laws, the Administrator shall use such authorities to
protect against such risk unless the Administrator determines, in the Administrator's discretion,
that it is in the public interest to protect against such risk by actions taken under this chapter."
• TSCA section 9(e): "... [I]f the Administrator obtains information related to exposures or releases
of a chemical substance or mixture that may be prevented or reduced under another Federal law,
including a law not administered by the Administrator, the Administrator shall make such
information available to the relevant Federal agency or office of the Environmental Protection
Agency."
• TSCA section 2(c): "It is the intent of Congress that the Administrator shall carry out this chapter
in a reasonable and prudent manner, and that the Administrator shall consider the environmental,
economic, and social impact of any action the Administrator takes or proposes as provided under
this chapter."
• TSCA section 18(d)(1): "Nothing in this chapter, nor any amendment made by the Frank R.
Lautenberg Chemical Safety for the 21st Century Act, nor any rule, standard of performance, risk
evaluation, or scientific assessment implemented pursuant to this chapter, shall affect the right of a
State or a political subdivision of a State to adopt or enforce any rule, standard of performance,
risk evaluation, scientific assessment, or any other protection for public health or the environment
that— (i) is adopted or authorized under the authority of any other Federal law or adopted to
satisfy or obtain authorization or approval under any other Federal law.
1.4.2.1 TSCA Authorities Supporting Tailored Risk Evaluations and Intra-agency
Referrals
TSCA section 6(b)(4)(D)
TSCA section 6(b)(4)(D) requires EPA, in developing the scope of a risk evaluation, to identify the
hazards, exposures, conditions of use, and potentially exposed or susceptible subpopulations the Agency
"expects to consider" in a risk evaluation. This language suggests that EPA is not required to consider
all conditions of use, hazards, or exposure pathways in risk evaluations. As EPA explained in the
Page 66 of 723
-------
"Procedures for Chemical Risk Evaluation Under the Amended Toxic Substances Control Act" ("Risk
Evaluation Rule"), EPA may, on a case-by-case basis, tailor the scope of the risk evaluation "to focus its
analytical efforts on those exposures that are likely to present the greatest concern, and consequently
merit an unreasonable risk determination." 82 FR 33726, 33729 (July 20, 2017).
In the Problem Formulation documents for many of the first 10 chemicals undergoing risk evaluation,
EPA applied the same authority and rationale to certain exposure pathways, explaining that "EPA is
planning to exercise its discretion under TSCA 6(b)(4)(D) to focus its analytical efforts on exposures
that are likely to present the greatest concern and consequently merit a risk evaluation under TSCA, by
excluding, on a case-by-case basis, certain exposure pathways that fall under the jurisdiction of other
EPA-administered statutes." This approach is informed by the legislative history of the amended TSCA,
which supports the Agency's exercise of discretion to focus the risk evaluation on areas that raise the
greatest potential for risk. See June 7, 2016 Cong. Rec., S3519-S3520. Consistent with the approach
articulated in the Problem Formulation documents, and as described in more detail below, EPA is
exercising its authority under TSCA to tailor the scope of exposures evaluated in TSCA risk evaluations,
rather than focusing on environmental exposure pathways addressed under other EPA-administered,
media-specific statutes and regulatory programs.
TSCA section 9(b)(1)
In addition to TSCA section 6(b)(4)(D), the Agency also has discretionary authority under the first
sentence of TSCA section 9(b)(1) to "coordinate actions taken under [TSCA] with actions taken under
other Federal laws administered in whole or in part by the Administrator." This broad, freestanding
authority provides for intra-agency coordination and cooperation on a range of "actions." In EPA's
view, the phrase "actions taken under [TSCA]" in the first sentence of section 9(b)(1) is reasonably read
to encompass more than just risk management actions, and to include actions taken during risk
evaluation as well. More specifically, the authority to coordinate intra-agency actions exists regardless
of whether the Administrator has first made a definitive finding of risk, formally determined that such
risk could be eliminated or reduced to a sufficient extent by actions taken under authorities in other
EPA-administered Federal laws, and/or made any associated finding as to whether it is in the public
interest to protect against such risk by actions taken under TSCA. TSCA section 9(b)(1) therefore
provides EPA authority to coordinate actions with other EPA offices without ever making a risk finding
or following an identification of risk. This includes coordination on tailoring the scope of TSCA risk
evaluations to focus on areas of greatest concern rather than exposure pathways addressed by other
EPA-administered statutes and regulatory programs, which does not involve a risk determination or
public interest finding under TSCA section 9(b)(2).
In a narrower application of the broad authority provided by the first sentence of TSCA section 9(b)(1),
the remaining provisions of section 9(b)(1) provide EPA authority to identify risks and refer certain of
those risks for action by other EPA offices. Under the second sentence of section 9(b)(1), "[i]f the
Administrator determines that a risk to health or the environment associated with a chemical substance
or mixture could be eliminated or reduced to a sufficient extent by actions taken under the authorities
contained in such other Federal laws, the Administrator shall use such authorities to protect against such
risk unless the Administrator determines, in the Administrator's discretion, that it is in the public interest
to protect against such risk by actions taken under [TSCA]." Coordination of intra-agency action on
risks under TSCA section 9(b)(1) therefore entails both an identification of risk, and a referral of any
risk that could be eliminated or reduced to a sufficient extent under other EPA-administered laws to the
EPA office(s) responsible for implementing those laws (absent a finding that it is in the public interest to
protect against the risk by actions taken under TSCA).
Page 67 of 723
-------
Risk may be identified by OPPT or another EPA office, and the form of the identification may vary. For
instance, OPPT may find that one or more conditions of use for a chemical substance present(s) a risk to
human or ecological receptors through specific exposure routes and/or pathways. This could involve a
quantitative or qualitative assessment of risk based on reasonably available information (which might
include, e.g., findings or statements by other EPA offices or other federal agencies). Alternatively, risk
could be identified by another EPA office. For example, another EPA office administering non-TSCA
authorities may have sufficient monitoring or modeling data to indicate that a particular COU presents
risk to certain human or ecological receptors, based on expected hazards and exposures. This risk
finding could be informed by information made available to the relevant office under TSCA section
9(e), which supports cooperative actions through coordinated information-sharing.
Following an identification of risk, EPA would determine if that risk could be eliminated or reduced to a
sufficient extent by actions taken under authorities in other EPA-administered laws. If so, TSCA
requires EPA to "use such authorities to protect against such risk," unless EPA determines that it is in
the public interest to protect against that risk by actions taken under TSCA. In some instances, EPA may
find that a risk could be sufficiently reduced or eliminated by future action taken under non-TSCA
authority. This might include, e.g., action taken under the authority of the Safe Drinking Water Act
(SDWA) to address risk to the general population from contaminants in drinking water, particularly if
the Office of Water has taken preliminary steps such as listing the subject chemical substance on the
Contaminant Candidate List. This sort of risk finding and referral could occur during the risk evaluation
process, thereby enabling EPA to use a more relevant and appropriate authority administered by another
EPA office to protect against hazards or exposures to affected receptors.
Legislative history on TSCA section 9(b)(1) supports both broad coordination on current intra-agency
actions, and narrower coordination when risk is identified and referred to another EPA office for action.
A Conference Report from the time of TSCA's passage explained that section 9 is intended "to assure
that overlapping or duplicative regulation is avoided while attempting to provide for the greatest
possible measure of protection to health and the environment." S. Rep. No. 94-1302 at 84. See also H.
Rep. No. 114-176 at 28 (stating that the 2016 TSCA amendments "reinforce TSCA's original purpose of
filling gaps in Federal law," and citing new language in section 9(b)(2) intended "to focus the
Administrator's exercise of discretion regarding which statute to apply and to encourage decisions that
avoid confusion, complication, and duplication"). Exercising TSCA section 9(b)(1) authority to
coordinate on tailoring TSCA risk evaluations is consistent with this expression of Congressional intent.
Legislative history also supports a reading of section 9(b)(1) under which EPA coordinates intra-agency
action, including information-sharing under TSCA section 9(e), and the appropriately-positioned EPA
office is responsible for the identification of risk and actions to protect against such risks. See, e.g.,
Senate Report 114-67, 2016 Cong. Rec. S3522 (under TSCA section 9, "if the Administrator finds that
disposal of a chemical substance may pose risks that could be prevented or reduced under the Solid
Waste Disposal Act, the Administrator should ensure that the relevant office of the EPA receives that
information"); H. Rep. No. 114-176 at 28, 2016 Cong. Rec. S3522 (under section 9, "if the
Administrator determines that a risk to health or the environment associated with disposal of a chemical
substance could be eliminated or reduced to a sufficient extent under the Solid Waste Disposal Act, the
Administrator should use those authorities to protect against the risk"). Legislative history on section
9(b)(1) therefore supports coordination with and referral of action to other EPA offices, especially when
statutes and associated regulatory programs administered by those offices could address exposure
pathways or risks associated with conditions of use, hazards, and/or exposure pathways that may
otherwise be within the scope of TSCA risk evaluations.
Page 68 of 723
-------
TSCA sections 2(c) and 18(d)
Finally, TSCA sections 2(c) and 18(d) support coordinated action on exposure pathways and risks
addressed by other EPA-administered statutes and regulatory programs. Section 2(c) directs EPA to
carry out TSCA in a "reasonable and prudent manner" and to consider "the environmental, economic,
and social impact" of its actions under TSCA. Legislative history from around the time of TSCA's
passage indicates that Congress intended EPA to consider the context and take into account the impacts
of each action under TSCA. S. Rep. No. 94-698 at 14 ("the intent of Congress as stated in this
subsection should guide each action the Administrator takes under other sections of the bill").
Section 18(d)(1) specifies that state actions adopted or authorized under any Federal law are not
preempted by an order of no unreasonable risk issued pursuant to TSCA section 6(i)(l) or a rule to
address unreasonable risk issued under TSCA section 6(a). Thus, even if a risk evaluation were to
address exposures or risks that are otherwise addressed by other federal laws and, for example,
implemented by states, the state laws implementing those federal requirements would not be preempted.
In such a case, both the other federal and state laws, as well as any TSCA section 6(i)(l) order or TSCA
section 6(a) rule, would apply to the same issue area. See also TSCA section 18(d)(l)(A)(iii). In
legislative history on amended TSCA pertaining to section 18(d), Congress opined that "[t]his approach
is appropriate for the considerable body of law regulating chemical releases to the environment, such as
air and water quality, where the states have traditionally had a significant regulatory role and often have
a uniquely local concern." Sen. Rep. 114-67 at 26.
EPA's careful consideration of whether other EPA-administered authorities are available, and more
appropriate, for addressing certain exposures and risks is consistent with this Congressional intent to
maintain existing federal requirements and the state actions adopted to locally and more specifically
implement those federal requirements, and to carry out TSCA in a reasonable and prudent manner. EPA
believes it is both reasonable and prudent to tailor TSCA risk evaluations in a manner reflective of
expertise and experience exercised by other EPA and State offices to address specific environmental
media, rather than attempt to evaluate and regulate potential exposures and risks from those media under
TSCA. This approach furthers Congressional direction and EPA aims to efficiently use Agency
resources, avoid duplicating efforts taken pursuant to other Agency programs, and meet the statutory
deadline for completing risk evaluations.
1.4.2.2 EPA-administered Statutes and Regulatory Programs that Address Specific
Exposure Pathways and/or Risks
HBCD is not classified as a RCRA hazardous waste. HBCD containing solid wastes are not expected to
be sent to Subtitle C incinerators, because HBCD is not a hazardous waste and due to the higher cost of
such incineration as compared with MSW or other incinerators; therefore, emissions from hazardous
waste incinerators were not evaluated. However, it is possible that HBCD containing solid wastes could
be sent to Subtitle C incinerators due to other characteristics of an HBCD containing solid waste
mixture.
EPA did not evaluate on-site releases to land that go to underground injection or associated exposures to
the general population or terrestrial species in its risk evaluation. Environmental disposal of HBCD
injected into Class I well types are covered under the jurisdiction of SDWA and disposal of HBCD via
underground injection is not likely to result in environmental and general population exposures. See 40
CFR part 144.
Page 69 of 723
-------
HBCD solid wastes are not required to be disposed of in Subtitle C hazardous waste landfills, however it
is possible that HBCD wastes could be disposed this way due to other characteristics of an HBCD
containing solid waste mixture. Design standards for Subtitle C landfills require double liner, double
leachate collection and removal systems, leak detection system, run on, runoff, and wind dispersal
controls, and a construction quality assurance program. They are also subject to closure and post-closure
care requirements including installing and maintaining a final cover, continuing operation of the leachate
collection and removal system until leachate is no longer detected, maintaining and monitoring the leak
detection and groundwater monitoring system. Bulk liquids may not be disposed in Subtitle C landfills.
Subtitle C landfill operators are required to implement an analysis and testing program to ensure
adequate knowledge of waste being managed, and to train personnel on routine and emergency
operations at the facility. Hazardous waste being disposed in Subtitle C landfills must also meet RCRA
waste treatment standards before disposal. See 40 CFR part 264.
EPA did not evaluate on-site releases to land from RCRA Subtitle C hazardous waste landfills or
exposures of the general population or terrestrial species from such releases in the evaluation. Hazardous
waste landfill design and management controls such as coverings, liners, and leachate collection and
treatment are expected to adequately mitigate HBCD exposure, therefore, on-site releases to land and
exposures of the general population or terrestrial species were not evaluated.
As HBCD is not classified as a RCRA hazardous waste, HBCD containing solid waste may be sent to
RCRA Subtitle D municipal solid waste (MSW) landfills particularly for construction demolition
disposal. The bulk of the HBCD containing solid waste is due to demolished XPS/EPS foam insulation
materials, which would be considered demolition waste. Demolition waste can be sent to MSW landfills,
but is expected to be primarily sent to C&D landfills. EPA is not evaluating on-site releases to land from
RCRA Subtitle D municipal solid waste (MSW) landfills or exposures of the general population or
terrestrial species from such releases in the TSCA evaluation. While permitted and managed by the
individual states, municipal solid waste landfills are required by federal regulations to implement some
of the same requirements as Subtitle C landfills. MSW landfills generally must have a liner system with
leachate collection and conduct groundwater monitoring and corrective action when releases are
detected. MSW landfills are also subject to closure and post-closure care requirements and must have
financial assurance for funding of any needed corrective actions. MSW landfills have also been designed
to allow for the small amounts of hazardous waste generated by households and very small quantity
waste generators (less than 220 lbs per month). Bulk liquids, such as free solvent, may not be disposed
of at MSW landfills. See 40 CFR part 258.
RCRA Subtitle D municipal solid waste (MSW) landfill design and management controls are expected
to adequately mitigate HBCD exposure, therefore, on-site releases to land and exposures of the general
population or terrestrial species were not evaluated. A qualitative assessment of leachate was conducted
to account for potential releases and exposures from disposal of demolition materials containing HBCD.
Since demolition waste can be sent to MSW landfills, but is expected to be primarily sent to C&D
landfills the qualitative assessment of leachate covers the disposal of HBCD to landfills including C&D.
1,4.3 Conceptual Models
The conceptual models for this Risk Evaluation are shown below in Figure 1-2, Figure 1-3, Figure 1-4
and Figure 1-5. EPA considered the potential for hazards to human health and the environment resulting
from exposure pathways outlined in the preliminary conceptual models of the HBCD scope document
G . !*. J h.\ ~sli Ll)- The conceptual models indicate potential exposures resulting from consumer
activities and uses, industrial and commercial activities, and environmental releases and wastes. The
Page 70 of 723
-------
problem formulation documents refined the initial conceptual models and analysis plans that were
provided in the scope documents (' v 1 1 ^ 0
For the purpose of this assessment, EPA considered workers and occupational non-users, which includes
men and women of reproductive age (Figure 1-2). Consumer exposure was assessed for various
pathways for all age-groups, including adults and children (Figure 1-3). Also, EPA considered exposures
to the general population for all age-groups, as well as additional considerations for other exposed
groups (Figure 1-3 and Figure 1-4).
EPA has made four modifications to the conceptual model since the publication of the problem
formulation document. The first was the addition of the solder/flux paste as a COU based on information
reported to the TRI, as discussed in Section 1.2.5.3.
The second change was made to include exposure to liquids for workers associated with solder/flux
paste as this use is expected to be in liquid formulations.
The third change was to more fully describe the use of HBCD in recycled products via the mouthing
pathway. EPA identified information in the open literature that describes articles which contain HBCD,
and recognizes this as an important pathway for infants and young children who may mouth articles.
EPA considered mouthing of recycled plastic products using experimental product-testing information
on HBCD content in consumer articles. See Section 2.4.4.4. for a more detailed discussion of this
exposure scenario.
The last change was the addition of the formulated products and articles as a COU as discussed in
Section 1.2.8.
These changes are reflected in the life cycle diagram and conceptual models.
Page 71 of 723
-------
INDUSTRIAL AND COMMERCIAL EXPOSURE PATHWAY EXPOSURE ROUTE RECEPTORS a HAZARDS
ACTIVITIES/USES
Workers b,
Workers b,
Occupational
Non-Users
Workers b,
Occupational
Non-Users
Wastewater, Solid Wastes, Air Emissions
(see Figures 1-4 and 1-5)
Dust
Emissions a
Dermal
Oral
Solid Contact, Dust
Inhalation
Liquid Contact
Indoor Air
Solid Contact
Dermal, Oral,
Inhalation
Indoor Air of
Commercial Buildings
Recycling
Import of HBCD
Dust emissions from
product installation,
reuse, and demolition.
Outdoor Air
(see Emissions to Air,
Figures 1-4 and 1-5)
Hazards Potentially Associated with
Acute and/or Chronic Exposures
Waste Handling,
Treatment and Disposal
Incorporated into Article:
manufacture of XPS/EPS,
manufacture of SIPs and
automobile replacement parts
from XPS/EPS
Incorporated into Formulation,
Mixture, or Reaction Product:
compounding of XPS
masterbatch; formulation of
solder/flux paste
Formulated products and articles
Building/Construction Materials
Solder/Flux Paste
Figure 1-2. HBCD Conceptual Model for Industrial and Commercial Activities and Uses: Worker and Occupational Non-User
Exposures and Hazards
The conceptual model presents the exposure pathways, exposure routes and hazards to human receptors from industrial and commercial
activities and uses of HBCD.
a Receptors include potentially exposed or susceptible subpopulations.
b EPA also considers the effect that engineering controls and personal protective equipment have on occupational exposure levels.
Page 72 of 723
-------
CONSUMER ACTIVITIES / USES EXPOSURE PATHWAY EXPOSURE ROUTE RECEPTORS3 HAZARDS
-~ Wastewater, Solid Wastes
/ Consumers/ ^
General Population;
v Bystanders
Oral
Dermal
Mouthing
Solid Wastes and
Recycling
Reused-recycled products
Reused-recycled EPS
and XPS
Automobile
Replacement Parts
Hazards Potentially Associated with
Acute and/or Chronic Exposures
Building/Construction
Materials-Primary
(building panels)
Indoor Air, Settled and
Suspended Dust
(in buildings and
automobiles)
Figure 1-3. HBCD Conceptual Model for Consumer Activities and Uses: Consumer Exposures and Hazards
The conceptual model presents the exposure pathways, exposure routes and hazards to human receptors from consumer activities and uses
HBCD.
a Receptors include potentially exposed or susceptible subpopulations.
Page 73 of 723
-------
RELEASES AND WASTES FROM EXPOSURE PATHWAY EXPOSURE ROUTE RECEPTORS'1 HAZARDS
INDUSTRIAL / COMMERCIAL / CONSUMER USES
Direct
discharge
Watera
Sediment
Indirect
discharge
Biosolids
General Population
• General Population
(background)
* Subsistence Fishers
• Highly Exposed
General Population
Soil
Air
—~ Breast Milk
Dermal
Oral
Inhalation
Solid Wastes
Aquatic Biota
Emissions to Air
Terrestrial
Biota
POTW
Wastewater
or Liquid Waste
Industrial Pre-Treatment or
Industrial WWT
Hazards Potentially
Associated with
Acute and/or Chronic
Exposures
Disposal
(e.g., Construction and
Demolition Waste)
Figure 1-4. HBCD Conceptual Model for Environmental Releases and Wastes: General Population Exposures and Hazards
The conceptual model presents the exposure pathways, exposure routes and hazards to human receptors from releases and wastes from
industrial and commercial uses of HBCD.
a Industrial wastewater or liquid wastes may be treated on-site and then released to surface water (direct discharge), or pre-treated and released to POTW
(indirect discharge). For consumer uses, such wastes may be released directly to POTW (i.e., down the drain).
b Receptors include potentially exposed or susceptible subpopulations.
Page 74 of 723
-------
RELEASES AND WASTES FROM EXPOSURE PATHWAY RECEPTORS HAZARDS
INDUSTRIAL/ COMMERCIAL / CONSUMER USES
Direct
discharge
Water a
Sediment
Indirect
discharge
Biosolids
\7
^ ¦=
i I
Terrestrial
Species
Soil
Air
Emissions to Air
POTW
Wastewater
or Liquid Waste
Industrial Pre-Treatment
or Industrial WWT
Disposal
(e.g., Construction and
Demolition Waste)
Hazards Potentially Associated
with Acute and/or Chronic
Exposures
Figure 1-5. HBCD Conceptual Model for Environmental Releases and Wastes: Ecological Exposures and Hazards
The conceptual model presents the exposure pathways and hazards for environmental receptors from industrial and commercial uses of
HBCD.
a Industrial wastewater or liquid wastes may be treated on-site and then released to surface water (direct discharge), or pre-treated and released to POTW
(indirect discharge). For consumer uses, such wastes may be released directly to POTW {i.e., down the drain).
Page 75 of 723
-------
1.5 Systematic Review
TSCA requires EPA to use scientific information, technical procedures, measures, methods, protocols,
methodologies and models consistent with the best available science when making science-based
decisions under Section 6 and base decisions under Section 6 on the weight of scientific evidence.
Within the TSCA Risk Evaluation context, the weight of the scientific evidence is defined as "a
systematic review method, applied in a manner suited to the nature of the evidence or decision, that uses
a pre-established protocol to comprehensively, objectively, transparently, and consistently identify and
evaluate each stream of evidence, including strengths, limitations, and relevance of each study and to
integrate evidence as necessary and appropriate based upon strengths, limitations, and relevance(40
C.F.R. 702.33).
To meet the TSCA Section 26(h) science standards, EPA used the TSCA systematic review process
described in the Application of Systematic Review in TSCA Risk Evaluations document (U.S. EPA.
2.018b. c). The process complements the Risk Evaluation process in that the data collection, data
evaluation and data integration stages of the systematic review process are used to develop the exposure
and hazard assessments based on reasonably available information. EPA defines "reasonably available
information" to mean information that EPA possesses, or can reasonably obtain and synthesize for use in
Risk Evaluations, considering the deadlines for completing the evaluation (RiskEvaluation Rule. 82 FR
33726).
EPA is implementing systematic review methods and approaches within the regulatory context of the
amended TSCA. Although EPA will make an effort to adopt as many best practices as practicable from
the systematic review community, EPA modified the process to ensure that the identification, screening,
evaluation and integration of data and information can support timely regulatory decision making under
the timelines of the statute.
1.5.1 Data and Information Collection
EPA planned and conducted a comprehensive literature search based on key words related to the
different discipline-specific evidence supporting the Risk Evaluation (e.g., environmental fate and
transport; environmental releases and occupational exposure; exposure to general population, consumers
and environmental exposure; and environmental and human health hazards). EPA then developed and
applied inclusion and exclusion criteria during the title/abstract screening to identify information
potentially relevant for the Risk Evaluation process. The literature and screening strategy as specifically
applied to HBCD is described in the Strategy for Conducting Literature Searches for Cyclic Aliphatic
Bromine Cluster (HBCD): Supplemental Document to the TSCA Scope Document ( 017f) and
the results of the title and abstract screening process were published in the Cyclic Aliphatic Bromide
Cluster (HBCD) (CASRN: 25637-99-4; 3194-55-6; 3194-57-8) Bibliography: Supplemental File for the
TSCA Scope Document ( > 17a. b). The screening strategy served to identify relevant studies
and exclude only those that were not pertinent to risk assessment of the chemical. No studies were
excluded at this step based on data quality evaluation, because only relevant studies were carried
forward for data quality evaluation.
For studies determined to be on-topic (or relevant) after title and abstract screening, EPA conducted a
full text screening to further exclude references that were not relevant to the Risk Evaluation. Screening
decisions were made based on eligibility criteria documented in the form of the populations, exposures,
Page 76 of 723
-------
comparators, and outcomes (PECO) framework or a modified framework10. Data sources that met the
criteria were carried forward to the data evaluation stage. The inclusion and exclusion criteria for full
text screening for HBCD are available in Appendix E of the Problem Formulation Document (
2018gY
Although EPA conducted a comprehensive search and screening process as described above, EPA
generally used previous chemical assessments11 to identify key and supporting information that would
be influential in the Risk Evaluation, in other words, information supporting key analyses, arguments,
and/or conclusions in the Risk Evaluation. When applicable, EPA also considered newer information not
considered in the previous chemical assessments and identified during the comprehensive search. Using
this pragmatic approach, EPA evaluated the confidence of the key and supporting data sources as well as
newer information instead of evaluating the confidence of all the underlying evidence ever published on
HBCD's fate and transport, environmental releases, and environmental and human exposure and
hazards. This allowed EPA to maximize the scientific and analytical efforts of other regulatory and non-
regulatory agencies by accepting for the most part the scientific knowledge gathered and analyzed by
others except for influential information sources that may have an impact on the weight of the scientific
evidence and ultimately the risk findings. The influential information (i.e., key/supporting) came from a
smaller pool of sources subject to the rigor of the TSCA systematic review process to ensure that the
Risk Evaluation uses the best available science and the weight of the scientific evidence.
Although EPA conducted a comprehensive search and screening process as described above, EPA made
the decision to leverage the literature published in previous assessments when identifying relevant key
and supporting data12 and information for developing the HBCD Risk Evaluation. This is discussed in
the Strategy for Conducting Literature Searches for Cyclic Aliphatic Bromine Cluster (HBCD):
Supplemental Document to the TSCA Scope Document (U.S. EPA. 2Q17D. In general, many of the key
and supporting data sources were identified in the comprehensive Cyclic Aliphatic Bromide Cluster
(HBCD) (CASRN: 25637-99-4; 3194-55-6; 3194-57-8) Bibliography: Supplemental File for the TSCA
Scope Document (U.S. EPA. 2017a. b). However, there were instances in which EPA missed relevant
references that were not captured in the initial categorization of the on-topic references. EPA found
additional relevant data and information using backward reference searching, which was a technique that
will be included in future search strategies. This issue was discussed in Section 4 of the Application of
Systematic Review for TSCA. Risk Evaluations. Other relevant key and supporting references were
identified through targeted supplemental searches to support the analytical approaches and methods in
the HBCD Risk Evaluation (e.g., to locate specific information for exposure modeling) or to identify
new data and information published after the date limits of the initial search.
EPA used previous chemical assessments to quickly identify relevant key and supporting information as
a pragmatic approach to expedite the quality evaluation of the data sources, but many of those data
sources were already captured in the comprehensive literature as explained above. EPA also considered
10 A PESO statement was used during the full text screening of environmental fate and transport data sources. PESO stands
for Pathways and Processes, Exposure, Setting or Scenario, and Outcomes. A RESO statement was used during the full text
screening of the engineering and occupational exposure literature. RESO stands for Receptors, Exposure, Setting or Scenario,
and Outcomes.
11 Examples of existing assessments are EPA's chemical assessments (e.g., previous work plan risk assessments, problem
formulation documents), ATSDR's Toxicological Profiles, EPA's IRIS assessments and ECHA's dossiers. This is described
in more detail in the Strategy for Conducting Literature Searches for Cyclic Aliphatic Bromine Cluster (HBCD):
Supplemental Document to the TSCA Scope Document (U.S. EPA 20.1.7f).
12 Key and supporting data and information are those that support key analyses, arguments, and/or conclusions in the risk
evaluation.
Page 77 of 723
-------
newer information not taken into account by previous chemical assessments as described in the Strategy
for Conducting Literature Searches for Cyclic Aliphatic Bromine Cluster (HBCD): Supplemental
Document to the TSCA Scope Document (U.S. EPA 2017f). EPA then evaluated the confidence of the
key and supporting data sources as well as newer information instead of evaluating the confidence of all
the underlying evidence ever published on a chemical substance's fate and transport, environmental
releases, environmental and human exposure and hazards. This allowed EPA to maximize the scientific
and analytical efforts of other regulatory and non-regulatory agencies by accepting for the most part the
relevant scientific knowledge gathered and analyzed by others except for influential information sources
that may have an impact on the weight of the scientific evidence and ultimately the risk findings. The
influential information (i.e., key/supporting) would come from a smaller pool of sources subject to the
rigor of the TSCA systematic review process to ensure that the Risk Evaluation uses the best available
science and the weight of the scientific evidence.
Figure 1-6 to Figure 1-10 depict literature flow diagrams illustrating the results of this process for each
scientific discipline-specific evidence supporting the Risk Evaluation. Each diagram provides the total
number of references at the start of each systematic review stage (i.e., data search, data screening, data
evaluation, data extraction/data integration) and those excluded based on criteria guiding the screening
and data quality evaluation decisions.
EPA made the decision to bypass the data screening step for data sources that were highly relevant to the
Risk Evaluation as described above. These data sources are depicted as "key/supporting data sources" in
the literature flow diagrams. The number of "key/supporting data sources" were excluded from the total
count during the data screening stage and added, for the most part, to the data evaluation stage
depending on the discipline-specific evidence. The exception was the environmental releases and
occupational exposure data sources that were subject to a combined data extraction and evaluation step
(Figure 1-7).
Data Screening (n=l,796)
Data Extraction/Data Integration (n=71)
*Key/S up porting
Data Sources (n=3)
Excluded References
-------
* These are key and supporting studies from existing assessments (e.g., EPA IRIS assessments, ATSDR
assessments, ECHA dossiers) that were considered highly relevant for the TSCA Risk Evaluation. These
studies bypassed the data screening step and moved directly to the data evaluation step.
n=93
Key/supporting
data sources
(n=35)
Excluded References (n=1754)
'Data Sources that were not
integrated (n=56)
Data Search Results (n=1847)
Excluded Ret that are
unacceptable based on
evaluation criteria (n=39)
Data Integration (n=33)
Data Extraction/Data Evaluation (n=128)
Data Screening (n=1847)
•The quality of data in these sources (n=56) were acceptable for risk evaluation purposes, but they were ultimately
excluded from further consideration based on EPAs integration approach for environmental release and occupational
exposure data/information EPA's approach uses a hierarchy of preferences that guide decisions about what types of
data/information are included for further analysis, synthesis and integration into the environmental release and
occupational exposure assessments EPA prefers using data with the highest rated quality among those in the higher
level of the hierarchy of preferences (i.e., data > modeling > occupational exposure limits or release limits). If warranted.
EPA may use data/information of lower rated quality as supportive evidence in the environmental release and
occupational exposure assessments
Figure 1-7. HBCD Literature Flow Diagram for Environmental Releases and Occupational
Exposure Data Sources
Literature search results for environmental release and occupational exposure yielded 1,847 data sources.
Of these data sources, 93 were determined to be relevant for the risk evaluation through the data screening
process. These relevant data sources were entered into the data extraction/evaluation phase. After data
extraction/evaluation, EPA identified several data gaps and performed a supplemental, targeted search to
fill these gaps (e.g. to locate information needed for exposure modeling). The supplemental search yielded
35 relevant data sources that bypassed the data screening step and were evaluated and extracted in
accordance with Appendix D: Data Quality Criteria for Occupational Exposure and Release Data of the
Application of Systematic Review for TSCA Risk Evaluations document.
Page 79 of 723
-------
Key Trusted
Studies (n= 0)
Data Extraction/Data
Integration (n=345)
Data Search Results (n= 11,208)
Data Screening (n= 11,208)
Excluded References(n= 9,512)
Unacceptable or excluded
based on evaluation
criteria(n=l,205)
Acceptable but not in scope
(n= 146)
Data Evaluation (n= 1,696)
Figure 1-8. Literature Flow Diagram for General Population, Consumer and Environmental
Exposure Data Sources for HBCD
EPA conducted a literature search to determine relevant data sources for assessing exposures for HBCD
within the scope of the Risk Evaluation. This search identified 11,208 data sources including relevant
supplemental documents. Of these, 9,512 were excluded during the screening of the title, abstract,
and/or full text and 1,696 data sources were recommended for data evaluation across up to five major
study types in accordance with Appendix E: Data Quality Criteria for Studies on Consumer, General
Population and Environmental Exposure of the Application of Systematic Review for TSCA Risk
Evaluations document.(U.S. EPA 2018c). Following the evaluation process, 345 references were
forwarded for further extraction and data integration.
Page 80 of 723
-------
Excluded References due to
ECOTOX Criteria
(n = 92)
Excluded References due to
ECOTOX Criteria
(n = 484)
Data Extraction I Data Integration (n - 48)
Data Evaluation (n = S4)
Full Text Screening (n = 144)
Excluded References that are
unacceptable based
on evaluation criteria and/or are
out of scope
-------
(1=28
Keyfeup porting data
sources
(n = 25)
Exduded References{n = 1.837)
Data Search Results {n = 1,890)
Excluded: Ref thatare
unacceptable based on
evaluation criteria
-------
2 EXPOSURES
This section describes EPA's approach to assessing environmental and human exposures. First, the fate
and transport of HBCD in the environment is characterized. Then, releases of HBCD into the
environment are assessed. Last, this information is integrated into an assessment of occupational,
general population (including highly exposed subpopulations), and environmental exposures for HBCD.
For all exposure-related disciplines, EPA screened, evaluated, extracted, and integrated available
empirical data. In addition, EPA used models to estimate exposures. Both empirical data and modeled
estimates were considered when selecting values for use in the exposure assessment.
Exposure equations and selected values used in the exposure assessment are presented in the following
sections. More specific information is provided in Risk Evaluation for Cyclic Aliphatic Bromide Cluster
(HBCD), Supplemental Information on General Population, Environmental, and Consumer Exposure
Assessment.
Following the inclusion of HBCD on EPA's workplan list in 2012, EPA published a 2015 problem
formulation prior to passage of the Lautenberg amendments, and published an updated scope in 2017
and problem formulation document in 2018. EPA has incorporated the following refinements based on
public comments and review of data since initial work began on HBCD.
• More complete assessment of human dietary exposure from multiple sources (estimates for all
food groups and more specific estimates for breast milk ingestion and fish ingestion) for the
general population,
• Inclusion of dermal pathway,
• Inclusion of refined models used to estimate surface water and ambient air as well as sediment
and indoor dust,
• Inclusion of additional contextual information from monitoring data to determine which data is
likely more applicable to exposure scenarios of interest, and
• Assessment of bioaccumulation and wildlife as part of environmental exposure assessment.
2.1 Fate and Transport
The environmental fate studies considered for this Risk Evaluation are summarized in Appendix C. This
information is based on studies published in (U.S. EPA. 2015a. 2014d; NICNAS 2012a; EC/HC 2011;
EI.NECS 2008; U.S. EPA. 2008a; OECD 2007) and was supplemented by an updated literature search
following problem formulation.
2.1.1 Fate and Transport Approach and Methodology
EPA gathered and evaluated environmental fate information according to the process described in the
Application of Systematic Review in TSCA Risk Evaluations (U.S. EPA. 2018b). Reasonably available
environmental fate information was used in the current evaluation. Furthermore, EPA used previous
regulatory and non-regulatory chemical assessments of HBCD to inform the environmental fate and
transport information discussed in this section and Appendix C. EPA had confidence in the information
used in the previous assessments to describe the environmental fate and transport of HBCD based on
scientific review of the methodologies and quality of the data presented and thus used it to make risk
evaluation decisions.
EPA also used the previous assessment to identify key and supporting fate information that would be
influential in the Risk Evaluation, as described in Section 1.5.1. For instance, EPA assessed the quality
Page 83 of 723
-------
of an HBCD aerobic freshwater sediment biodegradation study (Davis et al. 2006) based on the data
quality criteria described in the Application of Systematic Review in TSCA Risk Evaluations (U.S. EPA
2018b) and the study was rated 'high' confidence. The atmospheric oxidation half-life fate estimate was
based on modeling results from EPI Suite™ (U.S. EPA 2012b). a predictive tool for physical/chemical
and environmental fate properties. The data evaluation table describing the review of these studies as
well as other studies included in Table 2-lean be found in the supplemental document, Data Quality
Evaluation of Environmental Fate and Transport Studies (U.S. EPA 20191).
The HBCD environmental fate characteristics and physical-chemical properties used in fate assessment
are presented in Table 2-1 EPA used EPI Suite™ estimations and reasonably available fate information
to characterize the environmental fate and transport of HBCD. As part of problem formulation, EPA
also analyzed the fate of HBCD in air, water, soil, sediment, and bioaccumulation. The results of the
analyses are described in the 2018 problem formulation for HBCD (U.S. EPA 2018g) and presented
again in Appendix C. This section and Appendix C may also cite other data sources as part of the
reasonably available information on the fate and transport properties of HBCD. EPA did not subject
these other data sources to the later phases of the systematic review process (i.e., data evaluation and
integration) as explained in Section 1.5.1.
2.1.2 Summary of Fate and Transport
Environmental fate includes both transport and transformation processes. Environmental transport is the
movement of the chemical within and between environmental media. Transformation generally occurs
through the degradation or reaction of the chemical with other species in the environment. Hence,
knowledge of the environmental fate of the chemical informs the determination of the specific exposure
pathways and potential human and environmental receptors EPA analyzed in the Risk Evaluation.
Table 2-1 provides a summary of a subset of the environmental fate data that EPA identified, evaluated
and considered in the Risk Evaluation for HBCD. A full list of data considered, identified and evaluated
is provided in Appendix C.
Table 2-1. Summary of Environmental Fate and Transport Properties for HBCD
Property
Value
Reference
Study Quality
Indirect Photolysis
Half-life 2.1 days in air (estimated)
(U.S. EPA 2015a)
NA
Hydrolysis
Not expected due to lack of functional groups
that hydrolyze under environmental conditions
and low water solubility (estimated)
(ECHA 2008b)
NA
Aerobic
Biodegradation in
Water
No biodegradation observed in 28-day closed-
bottle test Organisation for Economic Co-
operation and Development (OECD) Guideline
30ID, EPA OTS 796.3200
(Wildlife Intl 1996)
Medium
Aerobic
Biodegradation in
Sediment
Half-life: 128, 92, and 72 days for a-, y-, and (3-
HBCD, respectively (estimated), based on a 44%
decrease in total initial radioactivity in viable
freshwater sediment of 14C-labeled HBCD (4.67
mg/kg dry weight) after 112 days; method based
on OECD 308
(Davis et al. 2006)
High
Half-life: >120 days (estimated), based on a 15%
decrease in total initial radioactivity in abiotic
freshwater sediment of 14C-labeled HBCD (4.67
mg/kg dry weight) after 112 days; method based
on OECD 308
High
Page 84 of 723
-------
Property
Value
Reference
Study Quality
Half-life: 11 and 32 days (estimated) in viable
sediment collected from Schuylkill River and
Neshaminy creek, respectively, using nominal
HBCD concentrations of 0.034-0.089 mg/kg;
method based on OECD 308
(Davis et al. 2005)
High
Half-life: 190 and 30 days (estimated) in abiotic
sediment collected from Schuylkill River and
Neshaminy creek, respectively, using nominal
HBCD concentrations of 0.034-0.089 mg/kg;
method based on OECD 308
High
Half-life: 92 days (estimated), based on a 61%
decrease in total initial radioactivity in viable
freshwater sediment of 14C-labeled HBCD
(4.31 mg/kg dry weight) after 113 days; method
based on OECD 308
(Davis et al. 2006)
High
Half-life: >120 days (estimated), based on a 33%
decrease in total initial radioactivity in abiotic
freshwater sediment of 14C-labeled HBCD (4.31
mg/kg dry weight) after 113 days; method based
on OECD 308
High
Half-life: 1.5 and 1.1 days (estimated) in viable
sediment collected from Schuylkill River and
Neshaminy creek, respectively, using nominal
HBCD concentrations of 0.063-0.089 mg/kg;
method based on OECD 308
(Davis et al. 2005)
High
Half-life: 10 and 9.9 days (estimated) in abiotic
sediment collected from Schuylkill River and
Neshaminy creek, respectively, using nominal
HBCD concentrations of 0.063-0.089 mg/kg;
method based on OECD 308
High
Aerobic
Biodegradation in
Soil
Half-life: >120 days (estimated), based on a
10% decrease in total initial radioactivity in
viable soil of 14C-labeled HBCD after 113 days;
method based on OECD 307 using HBCD at
3.04 mg/kg dry weight
(Davis et al. 2006)
High
Half-life: >120 days (estimated), based on a 6%
decrease in total initial radioactivity in abiotic
soil of 14C-labeled HBCD after
113 days; method based on OECD 307 using
HBCD at 3.04 mg/kg dry weight
High
Half-life: 63 days (estimated) in viable soil
amended with activated sludge using a nominal
HBCD concentration of 0.025 mg/kg dry weight;
method based on OECD 307
(Davis et al. 2005)
High
Half-life: >120 days (estimated) in abiotic soil
using a nominal HBCD concentration of 0.025
mg/kg dry weight; method based on OECD 307
High
Half-life: 6.9 days (estimated) in viable soil
amended with activated sludge using a nominal
HBCD concentration of 0.025 mg/kg dry weight;
method based on OECD 307
High
Half-life: 82 days (estimated) in abiotic soil
using a nominal HBCD concentration of 0.025
mg/kg dry weight; method based on OECD 307
High
Page 85 of 723
-------
Property
Value
Reference
Study Quality
Soil organic
carbon:water
partition coefficient
(log Koc)
Log Koc = 4.9 (79,433) estimated
(U.S. EPA 2015a)
NA
Log Koc > 5 (> 100,000) OECD Guideline 121
Estimation of the Adsorption Coefficient (Koc) on
Soil and on Sewage Sludge using High
Performance Liquid Chromatography (HPLC)
(ECHA 2017a)
High
Field Measured
Bioaccumulation
Factor (BAF)
Upper trophic level lipid normalized BAF for
total HBCDs of approximately 90,090,000
calculated from the mean HBCD lipid normalized
fish tissue concentration and the HBCD dissolved
water concentration.
Wet weight BAF 290,880
(He etal. 2013)
High
Upper trophic level lipid normalized BAF for
total HBCDs of approximately 3,120,000
calculated from the mean HBCD lipid normalized
fish tissue concentration and the HBCD dissolved
water concentration.
Wet weight BAF 46,488
(Wuetal. 2011)
High
Bioconcentration
Factor (BCF)
fathead minnow 18,100 (whole body)
(Veithetal. 1979)
High
aBCF (steady state, edible portion) rainbow trout
4650 at 1.8 ug/L exposure concentration)
BCF rainbow trout (kinetic, edible portion)
14,039 calculated at 0.18 ug/L exposure
concentration)
Drottar (Wildlife Intl
2000) as cited in
(ECHA 2008b)
High
aHBCD exposure concentrations 1.8 and 0.18 ug/L. Steady state acliieved at 1.8 ug/L but not at 0.18 ug/L
2.1.2.1 Air
HBCD is not expected to undergo significant direct photolysis since it does not absorb radiation in the
environmentally available region of the electromagnetic spectrum that has the potential to cause
molecular degradation (HSDB 2008). HBCD in the vapor phase will be degraded by reaction with
photochemically produced hydroxyl radicals in the atmosphere. A half-life of 2.1 days was calculated
from an estimated rate constant of 5.01 xlO"12 cm3/molecules-second at 25 °C, assuming an atmospheric
hydroxyl radical concentration of 1.5><106 molecules/cm3 and a 12-hour day (U.S. EPA 2011a. 1993a).
Based on an estimated octanol air partition coefficient (Koa) of 1.6 x 109, HBCD is expected to associate
strongly with airborne particulates. HBCD associated with particulates is expected to be less subject to
hydroxy radical oxidation in the atmosphere and primarily removed from the atmosphere through wet or
dry deposition.
2X2,2 Water
HBCD is not expected to undergo hydrolysis in environmental waters because of its lack of
hydrolyzable functional groups. Based on a measured soil organic carbon:water partition coefficient
(K OC ) of >100,000, HBCD is expected to partition from the water column, bind strongly to and be
transported with suspended and benthic sediments. A Henry's Law constant of 6><10"6 atm-m3/mol at 25
°C, calculated based on a vapor pressure of 4.70x 10"7 mm Hg at 21 °C and a water solubility of 66 |ig/L
at 25 °C, indicates that HBCD may volatilize slowly from moist soil and water surfaces. However,
adsorption to suspended solids and sediment will reduce the rate of volatilization from water. An OECD
301D ready biodegradability study (aerobic aqueous medium) on HBCD resulted in no observed
biodegradation in 28 days, suggesting that aerobic biodegradation in the water column may not be rapid
(Wildlife Intl 1996).
Page 86 of 723
-------
2.1.2.3 Soil and Sediment
Based on a measured K0c value of >100,000 HBCD is expected to bind strongly to soil, sediment, and
suspended organic matter. It may undergo abiotic and microbial degradation while associated with
solids. Tests with viable microbes demonstrated increased HBCD degradation compared to the
biologically inhibited control studies. In combination, these studies suggest that HBCD will degrade
slowly in the environment, although faster in sediment than in soil, faster under anaerobic conditions
than aerobic conditions, faster with microbial action than without microbial action, and at different rates
for individual HBCD diastereomers (slower for a-HBCD than for the y- and P- stereoisomers). The
biodegradation half-lives for aerobic sediment and aerobic soil calculated from (Davis et al. 2006) and
(Davis et al. 2005) were used for the assessment. HBCD has been reported to undergo abiotic
degradation in aerobic and anaerobic sediment and aerobic soil (ECHA 2008b; Davis et al. 2006) (see
Figure 2-1). The degradation was attributed to abiotic reductive dehalogenation which can form
tetrabromo and dibromocyclododecane and 1,5,9-cyclododecatriene. Further degradation of 1,5,9-
cyclododecatriene was not observed.
HBCD
TBCD
-2Br"
-2Br"
DBCD
Figure 2-1. Abiotic Reduction of HBCD to 5,6,9,10-tetrabromocyclododec-l-ene (TBCD),
9,10-dibromocyclododeca-l,5-diene (DBCD), and 1,5,9-cyclodecatriene (CDT) in Aerobic and
Anaerobic Sediments (Davis et al. 2006).
2.1.2.4 Wastewater Treatment Plants
No information was found on the removal of HBCD in Publicly Owned Treatment Works (POTWs) in
the United States. However, a study on the removal of HBCD in sewage treatment systems in the Yodo
river basin in Japan was identified and reviewed. (Ichihara et al. 2014) measured influent and effluent
concentrations of HBCD diastereoisomers in 12 sewage treatment plants in the river basin. The range of
removal rates was 80 - 99% with an average of 93% removal. Considering the low volatility and
biodegradability of HBCD, the removal was most likely due to sorption to activated sludge solids. The
Page 87 of 723
-------
EPA EPISuite STP (Sewage Treatment Plant) model was run for HBCD to provide additional
information on HBCD removal. The model simulates an activated sludge wastewater treatment system
and includes the processes of volatilization, adsorption to sludge and biodegradation. The model was run
using the physical-chemical properties reported in Section 1.1, Table 1-1 The biodegradation half-life
was set at 10,000 hours, a default for a non-biodegradable substance. The model calculated
approximately 90% removal of HBCD by adsorption to sludge with less than 1% removed by
biodegradation and volatilization. No information on the treatability of HBCD bound to plastic particles
was found. However, based on the density of these particles a qualitative assessment of their fate in
activated sludge systems can be made. Considering the low volatility and biodegradability of HBCD
these processes are not likely important. Dense particulate HBCD and HBCD associated with
polystyrene beads are expected to be removed with sludge during the sludge settling process. Less dense
HBCD associated with polystyrene foam may be removed in clarification by skimmers designed to
remove floating matter. Based on these findings, HBCD entering activated sludge wastewater treatment
systems is expected to be removed with a treatment efficiency in the range of 90% primarily by
adsorption to sludge. Volatilization and biodegradation of HBCD are not expected to be important
removal processes. Sludge bound HBCD may be further processed or disposed of by several methods
including land application.
2.1.2.5 Persistence
Based on the studies described later in this section HBCD is expected to be persistent in soil, surface
water and groundwater. It may biodegrade slowly under aerobic and anaerobic conditions with half-lives
on the order of months.
2.1.2.6 Bioaccuimilation/Bioconcentratioii
Bioaccumulation and bioconcentration in aquatic and terrestrial organisms, including humans, are
important environmental processes for HBCD. Bioconcentration is the net accumulation of a chemical
by an aquatic organism as a result of uptake directly from the ambient water, through gill membranes or
other external body surfaces. Bioaccumulation is the net accumulation of a chemical by an aquatic
organism as a result of uptake from all environmental sources. For hydrophobic chemicals such as
HBCD, aquatic organisms are exposed via both the diet and ambient water. Thus, bioaccumulation
measurements for HBCD more accurately reflect the contribution of all the routes by which aquatic
organisms are exposed.
Bioaccumulation factors were calculated for freshwater food webs in industrialized areas of Southern
China in two separate field studies. He et al. (He et al. 2013) calculated lipid normalized log BAFs of 4.8
- 7.7 (corresponding to BAFs of 63,000 - 50,000,000) for HBCD diastereomer in carp, tilapia, and
catfish, and found higher BAFs for a-HBCD than P- and y-HBCD. Wu et al. (Wu et al.! ) calculated
log BAFs of 2.85 - 5.98 for the total of all HBCD diastereomers (corresponding to BAFs of 700 -
950,000) in a freshwater food web. Log BAFs for each diastereomer in this study were comparable to
one another (see Appendix C.2). La Guardia et al. (La Guardia et al. 2012) calculated log BAFs in
bivalves and gastropods collected downstream of a textile manufacturing outfall; these ranged from 4.2
to 5.3 for a- and P-HBCD (BAFs of 16,000 - 200,000), and from 3.2 to 4.8 for y-HBCD (BAFs of 1,600
- 63,000).
Drottar and Kruger, (Wildlife Intl LTD 2000) as cited in (ECHA 2008b) measured BCF values ranging
from 8,974 to 13,085 for HBCD in rainbow trout. Veith et al. (Veith et al. 1979) measured a BCF of
18,100 for HBCD in fathead minnows. These BCF values indicate that HBCD exhibits very high
bioconcentration in fish. Widespread detection of this substance in aquatic organisms is further evidence
that HBCD bioconcentrates (Marvin et al. 201 I, K_H \ ,^08b; Covaci et al. 2.006). HBCD has also
Page 88 of 723
-------
been shown to biomagnify. Based on measurements of HBCD in invertebrates, fish, birds, and marine
mammals, biomagnification of HBCD in the aquatic food web is evident, with the highest levels of
HBCD measured in seals and porpoises (Shaw et al. 2012; Letcher et al. 2009; ECHA. 2008b; Covaci et
al. 2006; De Boer et al. 2002). Terrestrial food chain bioaccumulation has also been demonstrated. In a
study using breeding peregrine falcon populations in northern and southwestern Sweden, HBCD
concentrations were measured in the eggs of two groups of wild falcons and one group of captive
falcons fed only domestic chickens not exposed to HBCD. HBCD was not detected in the eggs of the
captive falcons but 150 and 250 ng/g lipid was measured in the eggs of the northern and southwestern
populations, respectively, indicating that HBCD bioaccumulation in terrestrial food chains may also be
important (Lindberg et al. 2004).
2,1.2,7 PBT Characterization
HBCD has been found to meet the criteria for Persistent, Bioaccumulative and Toxic (PBT) chemicals in
assessments conducted by EPA's TRI Program (U.S. EPA. 2016e). ECB (European Chemicals Bureau)
(ECHA. 2008b). Environment Canada/Health Canada (EC/I tC :< * I i) and NICNAS (NICNAS 2012a).
In 2016, EPA finalized a rule adding a hexabromocyclododecane (HBCD) category to the Toxics
Release Inventory (TRI) list of reportable chemicals with a 100-pound reporting threshold. EPA set
reporting threshold for the Toxics Release Inventory (TRI) HBCD category after determining that it
meets the criteria for a PBT chemical. For purposes of EPCRA section 313 reporting, EPA established
persistence half-life criteria for PBT chemicals of 2 months in water/sediment and soil and 2 days in air,
and established bioaccumulation criteria for PBT chemicals as a bioconcentration factor (BCF) or
bioaccumulation factor (BAF) of 1,000 or higher.
In its HBCD risk assessment the European Chemicals Bureau determined that while HBCD does not
unequivocally fulfill the specific P (persistence) criterion, with some reliable studies indicating that
biodegradation can occur, it does not degrade rapidly, and monitoring data indicate a significant degree
of environmental transport and overall stability. The HBCD BCF of 18,100 selected for use in the risk
assessment met the vB (very bioaccumulative) criterion. T (toxicity) criterion was found to be fulfilled
according to available data. The risk assessment further noted that HBCD is ubiquitous in the
environment, being also found in remote areas far away from point sources. The presence of the highest
concentrations of HBCD in marine top-predators such as porpoise and seals provides evidence that
HBCD bioaccumulates up the food chain. Based on an overall assessment it was concluded that HBCD
has PBT properties according to the PBT criteria of the Technical Guidance Document (TGD; ECB
2003).
Environment Canada/Health Canada in its Screening Assessment Report on Hexabromocyclododecane
determined HBCD meets the criteria for persistence in water, soil, and sediment as outlined in the
Persistence and Bioaccumulation Regulations under CEP A 1999 (i.e., half-life in water and soil of 182
days or more, and half-life in sediment of 365 days or more). Additionally, HBCD meets the criteria for
persistence in air set out in the same regulations (i.e., half-life of two days or more, or being subject to
atmospheric transport from the source to a remote area), and the criteria for bioaccumulation as specified
in the Persistence and Bioaccumulation Regulations under CEPA 1999 (i.e., bioaccumulation factors
[BAFs] or bioconcentration factors [BCFs] of 5000 or more).
The Australian Government Department of Health, National Industrial Chemicals Notice and
Assessment Scheme (NICNAS) compared the PBT characteristics of HBCD to Australian PBT criteria
and POPs criteria described in the United Nations Stockholm Convention on Persistent Organic
Pollutants. Based on laboratory data and international environmental monitoring data, sufficient
Page 89 of 723
-------
evidence was found to conclude that HBCD will persist in the environment and meets both Australian
and POPs criteria for persistence. Data provided through both laboratory testing and environmental
sampling of biota show the chemical (particularly the a isomer) is highly bioaccumulative and can be
biomagnified through the food chain. HBCD meets both Australian and POPs criteria for
bioaccumulation.
2.1.3 Assumptions and Key Sources of Uncertainty for Fate and Transport
Biodegradation Half-Lives
A range of aerobic and anaerobic biodegradation half-lives and bioaccumulation and bioconcentration
values have been reported for HBCD. The range of biodegradation half-lives reported were measured in
laboratory studies based on OECD methods for biodegradation in water, soil and sediment. These
studies are subject to several sources of variability including the specific microbial populations used,
water, soil and sediment chemistry, oxygen concentration/redox potential of the collected samples used
in the study, temperature and test substance concentration as well as variability inherent in the
methodology and interlaboratory variability. No single value of bioconcentration or bioaccumulation is
universally applicable as it is influenced by these variables and possibly others. However, the results of
these studies do inform the range of environmental half-lives HBCD might exhibit.
Media specific biodegradation half-lives selected for use in the Risk Evaluation are used as input to the
VVWM-PSC environmental exposure model discussed further in Section 2.3.2.2.2. Due to the
partitioning properties of HBCD its major pathway is expected to be partitioning to sediments where it is
subject to biodegradation. The use of a range of half-lives for aerobic sediment are recommended below.
The selection of shorter half-lives in the range as input to the model will result in lower concentrations
of HBCD in sediments and lower exposures to sediment dwelling organisms, possibly reducing risk
estimates for benthic organisms compared to using half-lives at the longer end of the range.
Half-lives estimated from studies ranged from days to greater than 6 months. Taken as a whole, the
studies demonstrate that under some conditions HBCD may undergo some degree of biodegradation
(complete biodegradation has not been reported) while under other conditions it does not appreciably
biodegrade. When this information is combined with environmental monitoring showing the presence of
HBCD in dated sediment cores it can be concluded that HBCD is persistent in the environment.
Furthermore, multiple jurisdictions have agreed, based on the available scientific evidence, that HBCD
meets criteria for persistence under their regulatory schemes (see Section 2.1.2.7 PBT Characterization)
Although a broad range of biodegradation half-lives for HBCD have been reported in laboratory studies
using aerobic and anaerobic soils and sediments and a single study of the biodegradation of HBCD in
water has been reviewed, a limited number of quantitative half-life ranges were selected for use in the
environmental and general population exposure assessments. Three studies (Davis et al. 2.006; Davis et
at 2005; Wildlife Ii 5) were used to assess the biodegradation half-lives of HBCD. Studies were
selected for use in the Risk Evaluation based of their relevance to the routes of entry of HBCD into the
environment. Releases of HBCD in particulate form to air and water are expected from several industrial
activities. Based on the environmental transport properties of HBCD, releases to air are expected to be
subject to wet and dry deposition to water bodies and soil. HBCD entering water bodies is not expected
to be to present at high levels in solution, but to sorb to suspended solids and ultimately deposit to
sediments. HBCD deposited to soil is expected to sorb strongly with little movement through the soil
column. Soil bound HBCD can enter water through run-off Thus, half-lives for water, soil and sediment
were determined to be most relevant for the Risk Evaluation. The assumption that HBCD enters aerobic
sediments leads to the use of aerobic sediment biodegradation half-lives for this medium. As discussed
further below, HBCD aerobic biodegradation half-lives are longer than anaerobic half- lives for soil (63
Page 90 of 723
-------
to greater than 120 days aerobic vs 6.9 days anaerobic) and sediment (11 to 128 days aerobic vs 1.1 to
92 days anaerobic). The use of the longer aerobic sediment biodegradation half-lives as input to the
environmental exposure model used in the Risk Evaluation will result in higher concentrations of HBCD
in sediments, possibly increasing risk estimates for benthic organisms compared to using anaerobic
sediment biodegradation half-lives at the shorter end of the range. Soil biodegradation half-lives were
not used as input to exposure models because monitored soil concentrations were available and were
used to assess soil related exposure. Thus, the selection of a particular soil biodegradation half-life did
not impact the exposure or Risk Evaluation.
An OECD 301D Closed Bottle Ready Biodegradability test (aerobic aqueous medium) on HBCD
resulted in no observed biodegradation in days. This result suggests that aerobic biodegradation in the
water column will not be rapid. Adsorption to suspended solids with subsequent deposition to the upper
layer of sediment is likely a more rapid process than biodegradation in the water column. Thus, sediment
half-life in the upper sediment layer is more relevant than the water column half-life. It is assumed that
the upper layer of sediments is aerobic. HBCD released to air and deposited on soil surfaces is assumed
to sorb strongly and remain in the surface layer where aerobic conditions prevail. Thus, aerobic soil
biodegradation half-lives are considered most relevant for the soil compartment.
Two studies (Davis et al. 2006; Davis et al. 2005) were selected to assess the biodegradation half-life of
HBCD in aerobic soils and aerobic sediments. Davis et al. (2005) and Davis et al. (2006). reported
aerobic soil biodegradation half-lives ranging from 63 days to greater than 120 days in viable test
systems. Aerobic sediment biodegradation half-lives ranging from 11 days for an HBCD mixture to 128,
92 and 72 days for a-, y-, and P - HBCD, respectively, were reported. From these studies, half-life
values of 2 to 6 months for aerobic soils and 11 days to 4 months for aerobic sediments were chosen.
For aerobic soils these values represent the range reported for biodegradation half-lives of HBCD
mixtures. For aerobic sediments these values represent the shortest half-life reported for an HBCD
mixture and the longest half-life reported for a diastereomer (a- HBCD).
Table 2-2. HBCD Biodegradation Half-Lives Selected for Use in Risk Evaluation
Property
Value
Reference
Study
Quality
Aerobic
Biodegradation in
Water
No biodegradation observed in 28-day closed-bottle test
Organisation for Economic Co-operation and
Development (OECD) Guideline 30ID, EPA OTS
796.3200
(Wildlife Intl
1996) as cited in
(EC 2008)
Medium
Aerobic
Biodegradation in
Sediment
Half-life: 128, 92, and 72 days for a-, y-, and (3 -HBCD,
respectively (estimated), based on a 44% decrease in total
initial radioactivity in viable freshwater sediment of re-
labeled HBCD (4.67 mg/kg dry weight) after 112 days;
method based on OECD 308
(Davis et al. 2006)
High
Half-life: 11 and 32 days (estimated) in viable sediment
collected from Schuylkill River and Neshaminy creek,
respectively, using nominal HBCD concentrations of
0.034-0.089 mg/kg; method based on OECD 308
(Davis et al. 2005)
High
Aerobic
Biodegradation in
Soil
Half-life: >120 days (estimated), based on a 10% decrease
in total initial radioactivity in viable soil of 14C-labeled
HBCD after 113 days; method based on OECD 307 using
HBCD at 3.04 mg/kg dry weight
(Davis et al. 2006)
High
Half-life: 63 days (estimated) in viable soil amended with
activated sludge using a nominal HBCD concentration of
0.025 mg/kg dry weight; method based on OECD 307
(Davis et al. 2005)
High
Page 91 of 723
-------
Biodegradation half-lives for the water column and sediments are required as input to the PSC-VVWM
model. The model is used to estimate water column and sediment concentrations for the Environmental
Risk Characterization described in Section 4.1. EPA used the biodegradation half-life ranges as reported
in or derived from the studies discussed in Sections 2.1.2 and 2.1.3 and as an alternative, the Office of
Pesticide Programs approach to calculating the 90th percentile confidence bound on the mean
biodegradation half-life value, and the Standard Operating Procedure for Using the NAFTA Guidance to
Calculate Representative Half-life Values and Characterizing Pesticide Degradation (U.S. EPA. 2015b)
which provides tools to determine the appropriate kinetics and associated half-lives for biodegradation
studies.
The 90th percentile confidence bound on the mean biodegradation half-life value is calculated according
to the equation below:
Equation 1:
tinput = t 1/2 + [(t90,n-lS) / II'|
where,
tinput = half-life input value (time)
~t"i/2= mean of sample half-lives (time)
s = sample standard deviation (time)
n = number of half-lives available (-)
t9o,n-i = one-sided Student's t value at a = 0.1 (i.e., 1.0-0.9) (-)
This equation does not calculate the 90th percentile of the distribution of half-life values.
The rate of transformation of organic chemicals in the environment is commonly described using first-
order kinetics, often referred to as single first-order (SFO). The first-order representation is convenient
because the rate is summarized with a single parameter (the rate constant, k), and the rate of
transformation is independent of the initial concentration. The half-life, VA =ln(2)/k, indicates the time
required to reduce the concentration by 50% from any concentration point in time. It is an intuitive way
to express the rate of decline of a first-order degradation. In contrast, the DT50 is the time required for
the concentration to decline to half of the initial value. For non-first-order decay, the time to reach half
the concentration from any other concentration point on the curve will be different.
The VVWM-PSC model requires first-order inputs for the modeled chemical's transformation processes
even though a chemical's transformations in aquatic systems often does not follow a single exponential
decline pattern. For this reason, the NAFTA guidance introduces a "representative half-life (trep)" to
estimate an SFO half-life for model input from a degradation curve that does not follow the SFO
equation. The procedure takes into consideration the frequent observation that chemicals can degrade
fast initially and then slowly as time passes, much more so than a first-order representation would
predict. The representative half-life considers both the initial and the slower portions of the decline
curve and is not necessarily numerically similar to the value of the DT50.
Page 92 of 723
-------
Table 2-3. HBCD Biodegradation Half-lives (days) Reported and Representative Half-lives
Calculated Using OPP/NAFTA Guidance
Reference
Medium
Reported
90th Percentile
Confidence Bound
OPP/NAFTA
Guidance
(Davis et al.
2005)
Aerobic Sediment
11
112
6
(Davis et al.
2005)
Aerobic Sediment
32
8
(Davis et al.
2006)
Aerobic Sediment
128
100
The PestDF program calculates and selects the representative half-life value based on the NAFTA
guidance. The tool considers three transformation models: SFO, double first-order in parallel (DFOP),
and indeterminate order rate equation (IORE) and a set criteria for selecting parameters. Based on the
number of fitted parameters, SFO is the simplest of the three models, while DFOP is the most complex.
OPP guidance also allows for a 3X factor to be used to account for uncertainty and variability where
only 1 half-life value is available. In this evaluation the 3X factor was used with the longest reported
half-life from Davis et al. (2006) to give a half-life of 384 days.
In order to demonstrate the effect of changes in benthic half-lives on estimated porewater, water column
and sediment HBCD concentrations estimated by VVWM-PSC, a limited sensitivity analysis was
conducted. All environmental parameters, loading and abiotic half-lives were held constant. Multiple
runs of VVWM-PSC were executed varying only the benthic half-life using the values reported in Table
2-3 above. The results are shown in Table 2-4 below.
Table 2-4. Impact of the Use of the Range of Biodegradation Half-lives (days) Reported and
Representative Half-lives Calculated Using OPP/NAFTA Guidance on PSC-VVWM
Concentration Estimates3
Benthic
Half-life
Days
Water Column
Concentration
21 Day Average
(Jig/L)
Water Column
Concentration
28 Day Average
(Jig/L)
Sediment Pore Water
Concentration
28 Day Average
(Jig/L)
Total Benthic
Concentration
28 Day Average
(jug/kg)
6
19.9
29.3
3.91
15600
8
20.2
29.5
4.6
18400
11
20.5
29.9
5.63
22500
32
21.9
31.2
9.55
38200
100
23.3
32.6
13.5
54200
112
23.4
32.7
13.9
55400
128
23.5
32.8
14.2
56900
384
24.2
33.6
16.3
65200
Page 93 of 723
-------
a Standard Operating Procedure for Using the NAFTA Guidance to Calculate Representative Half-life Values and
Characterizing Pesticide Degradation (U.S. EPA 2015b")
As can be seen from the results of the modeling, benthic half-lives over the ranges discussed in the final
Risk Evaluation have a negligible effect on water column concentrations. Thus, the half-life chosen for
use in PSC-VVWM will not generally result in changes in ecological Risk Quotients for a given
scenario. In contrast, sediment pore water and total benthic concentrations increase approximately 4 to 5
times as benthic half-lives increase from six to 384 days. The impact of half-life on benthic Risk
Quotients are further discussed in Section 4.1 Environmental Risk.
Bioconcentration/Bioaccumulation Factors
A range of bioconcentration/bioaccumulation values have been reported for HBCD and separately for
the three stereoisomers. The range of reported values were measured in laboratory studies or estimated
from field collected data. These studies are subject to several sources of variability including variability
inherent in the methodology, interlaboratory variability and variability due to factors such as the test
species used, test substance concentration, as well as temporal and spatial factors in collection of field
samples. No single value is universally applicable as it is influenced by these variables and possibly
others. However, taken as a whole, studies indicate HBCD is subject to bioconcentration,
bioaccumulation and trophic magnification.
A field measured bioaccumulation factor (BAF) selected for use in the Risk Evaluation (Wu et al.! )
was used as input to the estimation of highly exposed general population fish ingestion exposure
discussed further in Section 2.4.3. Initially, EPA considered two BAF values, one higher and one lower.
Both studies were rated high for data quality. The differences in reported BAFs could be due to a
number of factors including the metabolic differences in the test species selected. The selection of the
higher BAF as input to the estimation of general population fish ingestion exposure will result in higher
fish tissue concentrations of HBCD and higher exposures to general population via fish ingestion. This
will lead to estimates of higher risk for this population compared to using the lower BAF value. Due to
the small number of field derived fish BAF studies found (2) it was not possible to assess the variability
in field derived BAFs across field conditions, dissolved HBCD concentrations, species and trophic
levels. In the studies EPA identified, the reported dissolved HBCD concentrations in Chinese water
bodies were in the range of 0.04 to 0.06 ng/L. These are about an order of magnitude lower than the
range of dissolved HBCD surface water concentrations reported in surface water monitoring studies.
The range of HBCD surface water concentrations biota are assumed to be exposed to for the Risk
Evaluation was determined using monitoring data and model estimates. After consideration of factors
including the edibility and palatability of the species, an upper trophic level lipid normalized field
measured BAF for the northern snakehead was selected for use as a surrogate species for the fish
ingestion exposure assessment. The use of lipid normalized field measured BAF data for an upper
trophic level species incorporates results of dietary exposure and biomagnification in the food web.
However, the small number of BAF values, the limited number of species and field conditions add to
uncertainty associated with the use of these BAFs in estimating human exposure to HBCD via fish
ingestion.
For the purposes of the Risk Evaluation, lipid normalized bioaccumulation factors in whole fish
consumed by humans, and bioconcentration factors in species in aquatic and terrestrial food webs were
used. These values are converted to wet weight BAF values (BAFww) for use in dietary exposure
calculations using the following formula:
Page 94 of 723
-------
BAFww = BAFlw * lipid fraction
See Appendix C for underlying data and calculations of BAFs for HBCD.
Field-measured bioaccumulation factors for HBCD were preferentially used over bioconcentration
factors for the Risk Evaluation. A BAF derived from data obtained from field-collected samples of tissue
and water is the most direct measure of bioaccumulation. A field-measured BAF is determined from
measured chemical concentrations in an aquatic organism and the ambient water collected from the same
field location. Because the data are collected from a natural aquatic ecosystem, a field-measured BAF
reflects an organism's exposure to a chemical through all relevant exposure routes (e.g., water, sediment,
diet). A field-measured BAF also reflects factors that influence the bioavailability and metabolism of a
chemical that might occur in the aquatic organism or its food web. Therefore, field-measured BAFs are
appropriate for all chemicals, regardless of the extent of chemical metabolism in biota (U.S. EPA. 2003).
Specifically, the field measured BAFs reported by (Wu et at 20101 and (He et al. 2013) were reviewed.
These studies scored high using data quality metrics for environmental fate studies. In addition, the
studies reported BAF values in upper trophic level (i.e., piscivorous fish). BAFs in organisms occupying
higher trophic levels in food webs may better reflect exposure due to dietary uptake than organisms in
lower trophic levels. Using data from (Wu et al.: ), an upper trophic level lipid normalized BAF for
total HBCDs of approximately 3,120,000 was calculated from the mean HBCD lipid normalized fish
tissue concentration and the HBCD dissolved water concentration. Using data from (He et al. 2013). an
upper trophic level lipid normalized BAF for total HBCDs of approximately 9,090,000 was calculated
from the mean HBCD lipid normalized fish tissue concentration and the HBCD dissolved water
concentration. It should be noted that in both the studies, sample sizes for fish were small (n= 6-15) and
variability in tissue concentrations for a single species of fish was as high as 3 times the mean value.
While this variability leads to uncertainty in the use of the data, the preference for the use of upper
trophic level field measured BAFs and lack of other similar studies was considered in the decision to use
the study. The steady-state BCF values in rainbow trout edible portions (Wildlife Intl LTD 2000). as
cited in (ECH.A. 2.008b). were used to supplement the Risk Evaluation. A kinetic BCF value of 14,039 for
the 0.18 |ig/L exposure concentration was calculated to address the possibility that steady state was not
reached (EC 38b). The study received a high confidence score based on evaluation metrics for fate
studies.
Due to the small number of field derived fish BAF studies found (2) it was not possible to assess the
variability in field derived BAFs. EPA did not have a sufficient number of bioaccumulation studies to
follow the Office of Water methodology for deriving bioaccumulation factors intended to develop BAFs
for setting national water quality criteria (U.S. EPA. 2000). The methodology is generally used with
large sets of BAF data for multiple trophic levels and species from studies reflecting a range of
geochemical and biological conditions. However, using the approach for chemicals classified in the
Office of Water methodology as nonionic organic chemicals with moderate to high hydrophobicity (log
Kow > 4) and low metabolism to calculate baseline and national BAF values yielded upper trophic level
(TL 4) BAF values approximately two times greater than the field measured values reported for northern
snakehead (Wu et al. 2010). The differences are due, in part, to the differences between site specific and
species-specific variables in the field study (e.g., the particulate organic carbon levels and the lipid
fraction in fish) which impact bioaccumulation factors and the default values for those variables used in
the Office of Water methodology to derive the upper trophic level (TL 4) BAF.
EPA identified two BCF studies and two BAF studies on HBCD. BAF studies are preferred over BCF
studies because they represent exposure of the organism to HBCD via all routes, including diet which is
Page 95 of 723
-------
important for a hydrophobic chemical such as HBCD. The BAF studies (He et al. 2013) reported data
EPA used to calculate upper trophic level lipid normalized BAFs for several trophic levels, however, the
species reported were native to China. With limited available data EPA chose to use the upper trophic
level species (northern snakehead) (Wu et al. 2010) as a surrogate for an upper trophic level species
native fish and assumed its lipid normalized BAF was equivalent to that of an upper trophic level native
fish. Because a single BAF from a single species is used, impacts of factors including lipid content,
organism size, spatial and temporal variability in exposure concentrations, sample size, trophic position
and differences in food webs and ecosystems cannot be considered. The absence of this information
creates uncertainty in how representative the BAF may be and if its use will under or overpredict fish
tissue concentrations and human exposure via fish ingestion.
Table 2-5. HBCD Bioaccumulation and Bioconcentration Factors Reviewed for Use in the Risk
Evaluation
Property
Value
Reference
Study Quality
Field Measured
Bioaccumulation
Factor (BAF)
Upper trophic level lipid normalized BAF for total
HBCDs of approximately 3,120,000 calculated from
the mean HBCD lipid normalized fish tissue
concentration and the HBCD dissolved water
concentration. —northern snakehead
Wet weight BAF 46,488
(Wu et al. 2010)
High
Upper trophic level lipid normalized BAF for total
HBCDs of approximately 9,090,000 calculated from the
mean HBCD lipid normalized fish tissue concentration
and the HBCD dissolved water concentration, -catfish
Wet weight BAF 290,880
(He et al. 2013)
High
fathead minnow 18,100 (whole body)
(Veith et al. 1979)
High
Bioconcentration
Factor (BCF)
rainbow trout 4650 - 6531 (edible portion)
14039 (kinetic BCF 0.18 |ig/L exposure concentration)
(Wildlife Intl LTD
2000) as cited in
(ECHA 2008b)
High
HBCD in Microplastics
HBCD incorporated into EPS and XPS may enter air, water and soil environments as particulates as a
result of its processing, use, and demolition and disposal of building material containing EPS and XPS
insulation. (See Section 2.2 Releases to the Environment). HBCD containing particulates may be
produced during insulation board cutting and building demolition. HBCD containing insulation may
generate particles from physical abrasion and weathering. These particles may include a size range
similar to that of microplastics (i.e., items < 5 mm diameter) (Lambert et al.. 2014). In the aquatic
environment, the ingestion of plastics by biota establishes a potential exposure pathway for chemical
contaminants that may be incorporated into the plastics during manufacture or metals, and persistent,
bioaccumulative, and toxic contaminants that may be sorbed from the water column to plastic. (Engler
2012).
Scientific research including field studies (e.g., Yamashita et al. 2011; Layers et al.. 2014; Rochman et
al.. 2014) and laboratory studies (e.g., Teuten et al.. 2009; Besseling et al.. 2013; Rochman et al.. 2013)
suggests that several groups of aquatic or aquatic-dependent organisms (invertebrates, fish, and birds)
can accumulate chemicals associated with plastics once ingested. Experimental studies investigating the
effects of chemicals associated with plastics on invertebrates and fish indicate that there are negative
sublethal effects on these organisms from chemicals associated with plastics as well as the plastic itself
(e.g., Rochman et al., 2013. 2014; Avio et al.. 2015). However, some bioaccumulation modeling
Page 96 of 723
-------
approaches attempting to simulate environmentally realistic scenarios of exposure provide indirect
evidence that the role of plastics in contributing to body burdens and effects of chemical pollutants may
be relatively small compared with other exposure pathways, such as direct chemical exposure via water,
sediment, or ingestion of contaminated prey (Koetmans et at.. 2016; Bakir et at.. ^4 % flccardi et at..
2016V
EPS particles in the microplastics size range (<5 mm) have been implicated as potential vectors for
HBCD in the marine environment. Such particles can be generated when larger EPS objects in the ocean
are subjected to biodegradation, ultraviolet radiation, temperature, and the mechanical forces associated
with wave action (Rani et at. 2017). In one study, EPS buoys were identified as the source of elevated
HBCD concentrations in sediments off the coast of South Korea (Al-Odaini et at. 2015). Further
investigation found that mussels inhabiting EPS substrates in the same region had higher HBCD body
burdens than those inhabiting high-density polyethylene, metal, and rock Dane et at. 2016). These
findings appear to indicate a potential exposure pathway for ecological and human receptors due to
bioaccumulation of HBCD from microplastics. However, it is not currently feasible to quantify the
exposure of upper trophic level organisms to microplastic-associated HBCD. This is generally true of all
microplastic-associated pollutants due to the large number of variables controlling their uptake and
potential bioaccumulation/biomagnification (Au et at... 2017; Ziccardi et at. 2016). In the specific case of
HBCD, there is currently not sufficient data on the distribution of the chemical in microplastics across
geographic regions (Jane et at. 2017). nor its ability to leach from ingested microplastic particles and
become available for distribution, metabolism, and excretion (Lohmann et at. 2017). If microplastic-
associated HBCD is readily bioavailable, its behavior may be similar to that of pure particulate HBCD.
However, it is more likely that association with microplastics has complex and opposing influences on
HBCD exposure. While they can serve as a vector, microplastics may also reduce bioavailability and
potentially scavenge free HBCD. In the absence of data needed to parameterize a model, this complexity
cannot currently be resolved.
2.2 Releases to the Environment
EPA assessed environmental releases of HBCD for the following HBCD exposure scenarios:
1) Repackaging of Import Containers
2) Compounding of Polystyrene Resin to Produce XPS Masterbatch
3) Processing to Produce XPS Foam using XPS Masterbatch
4) Processing of HBCD to Produce XPS Foam
5) Processing to Produce EPS Foam from Imported EPS Resin Beads
6) Processing to Produce SIPs and Automobile Replacement Parts from XPS/EPS Foam
7) Use: Installation of Automobile Replacement Parts
8) Use: Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings,
and Other Structures
9) Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures
10) Recycling of EPS Foam and Reuse of XPS foam
11) Formulation of Flux/Solder Pastes
12) Use of Flux/Solder Pastes
13) Recycling of Electronics Waste (E-Waste) Containing HIPS
As discussed in Section 1.2.8, HBCD is no longer used to manufacture, process, or distribute four
minor-use products or articles: textiles, HIPS in electronics, adhesives, and coatings. The four minor-use
products and articles are expected to be currently already installed or in service. The processing of these
Page 97 of 723
-------
products during disposal at landfills and waste transfer stations may result in fugitive air releases of dust
containing HBCD. These releases are not quantified in this section. EPA believes exposures to general
population and environmental receptors are accounted for in the assessment of background exposure
which is discussed in Section 2.4.2 for general population and Section 2.3.3.1 for terrestrial receptors.
Components of the Environmental Release Assessment
The environmental release assessment of each exposure scenario is comprised of the following
components:
1. Process Description: A description of the exposure scenario, including the role of the chemical
in the use; process vessels, equipment, and tools used during the exposure scenario; and
descriptions of the worker activities, including an assessment for potential points of worker
exposure and environmental releases.
2. Facility Estimates / Processing or Use Volume and Number of Sites: An estimate of the
quantity of HBCD imported, processed, or otherwise used for each exposure scenario. An
estimate of the number of sites that use the chemical for the given exposure scenario.
3. Environmental Releases: Estimates of chemical released into the environment (air, surface
water, land) and wastes disposed to treatment methods (incinerators, wastewater treatment
plants).
2.2.1 Release Assessment Approach and Methodology
Process Description
EPA performed a literature search to find descriptions of processes involved in each exposure scenario
to identify worker activities that could potentially result in releases to the environment. Where process
descriptions were unclear or not available, EPA referenced relevant emission scenario documents
(ESDs) and generic scenarios (GSs), specifically the 2009 OECD ESD on Plastic Additives, the 2014
Draft OECD ESD on Use of Additives in Plastics Compounding, and the 2010 OECD ESD on
Chemicals Used in the Electronics Industry. The process description for each exposure scenario will be
discussed in each section.
Processing or Use Volume and Number of Sites
As indicated in Section 1.2.2 and 1.2.3, EPA has determined that the import of HBCD constitutes an
intended, known and reasonably foreseen activity. The companies identified by the 2016 CDR as
importers of HBCD have ceased importing, processing and using HBCD. The possibility exists that
small firms could import quantities of up to 100,000 lbs/year per site without reporting to CDR. For the
purpose of this Risk Evaluation, EPA used the CDR reporting threshold for small manufacturers
(importers) of 100,000 pounds per year as the volume of HBCD imported by a possible unidentified site.
EPA believes this volume is not unreasonable considering the recent relatively high volumes of HBCD
manufactured / imported, processed and used through 2017 for XPS/EPS foam as shown in Table 1-2
and Table 1-4. EPA does note, however, that 100,000 pounds per year is an upper bound for the import
volume for the unknown site, otherwise, the importer would be out of compliance with CDR reporting
requirements. The lifecycle of the imported HBCD and more specifically the percentage of the volume
used for each of the exposure scenarios is uncertain, and therefore, EPA uses the volume basis of
100,000 pounds per site per year to estimate environmental releases and exposures of each of the
following exposure scenarios that entail the processing of HBCD for products and formulations
containing HBCD:
• Repackaging of Import Containers
• Compounding of Polystyrene Resin to Produce XPS Masterbatch
Page 98 of 723
-------
• Processing to Produce XPS Foam using XPS Masterbatch
• Processing of HBCD to Produce XPS Foam
• Processing to Produce EPS Foam from Imported EPS Resin Beads
• Processing to Produce SIPs and Automobile Replacement Parts from XPS/EPS Foam (the
processing volume for each exposure scenario is 100,000 pounds/year)
The import volume of 100,000 pounds per year is also used for assessing releases, number of sites, and
exposures for the following exposure scenarios and will be further described in Sections 2.2.8 and 2.2.9,
respectively:
• Use: Installation of Automobile Replacement Parts
• Use: Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings,
and Other Structures
EPA performed a sensitivity analysis for selected exposure scenarios using import volumes of 50,000
lbs/yr-site and 25,000 lbs/yr-site to examine the effect of process volume on environmental releases and
resulting general population and environmental exposures. This is discussed in Section 2.2.15.
Environmental Release Assessment
EPA assessed, where applicable, releases to fugitive or stack air, discharges to on-site wastewater
treatment (WWT), Publicly Owned Treatment Works (POTWs), or surface water, disposal to landfill,
and treatment via incineration. EPA refers to these as methods of release, disposal, treatment, or
discharge in the remainder of this section. All releases assessed are of solid HBCD or solid mixtures
containing HBCD.
EPA assessed releases to landfill for Repackaging of Import Containers, Processing to Produce EPS
Foam from Imported EPS Resin Beads, Processing: Recycling of EPS Foam and Reuse of XPS Foam,
Processing to Produce SIPs and Automobile Replacement Parts from XPS/EPS Foam, and Use:
Installation of XPS/EPS Foam Insulation in Residential, Public, and Commercial Buildings, and Other
Structures in accordance with the 2009 OECD ESD on Plastic Additives. EPA assessed releases to
landfill for Demolition and Disposal in accordance with data from (Townsend et al. 2019: U.S. EPA.
2018: TCEQ 2017) and for Use of Flux/Solder Pastes in accordance with the 2010 OECD ESD on
Chemicals Used in the Electronics Industry. The landfill types are not specified in these sources. As
discussed in Section 1.4.2.2, EPA is not evaluating releases to RCRA Subtitle C hazardous waste
landfills and RCRA Subtitle D municipal solid waste landfills (MWSLFs). Hazardous waste and
municipal waste landfill design and management controls such as coverings, liners, and leachate
collection and treatment are expected to adequately mitigate HBCD exposure, therefore, releases were
not evaluated. HBCD is not designated as a RCRA hazardous waste because it is not specifically listed
as a known hazardous waste and does not exhibit the characteristics of a hazardous waste (ignitability,
corrosivity, reactivity or toxicity) (40 CFR 261). HBCD waste could be sent to industrial non-hazardous
landfills, which are described here: https://www.epa.gov/landfills/industrial-and-construction-and-
demolition-cd-landfills. Therefore, EPA assessed releases to these types of landfills.
EPA gathered and evaluated environmental release information according to the process described in the
Application of Systematic Review in TSCA Risk Evaluations (i__S iOj Sb). The key data sources
resulting from this process that were used to assess releases include TRI data, the European Union Risk
Assessment Report (EURAR), and ( 008b). The TRI data has an overall confidence rating of
medium. The EURAR and ( 08b) have overall confidence ratings of high.
Page 99 of 723
-------
Where available, EPA used 2017 TRI data to provide a basis for estimating releases. Facilities are only
required to report to TRI if the facility has 10 or more full-time employee equivalents, is included in an
applicable NAICS code, and manufactures, processes, or otherwise uses the chemical in quantities
greater than a certain threshold in a given year (100-pound threshold for HBCD). Due to these
limitations, some sites that use HBCD may not report to TRI and are not included in these datasets. EPA
did not use some of the TRI data based on additional information gathered about current uses and
reported releases. Specifically, EPA did not use the 2017 releases reported by Flame Control Coatings,
LLC. The company indicated that they have ceased the use of HBCD in coatings.
TRI reporting by subject facilities is required by law to provide information on releases and other waste
management activities of Emergency Planning and Community Right-to-Know Act (EPCRA) Section
313 chemicals {i.e., TRI chemicals) to the public for informed decision making and to EPA to assist the
Agency in determining the need for future regulations. Section 313 of EPCRA and Section 6607 of the
Pollution Prevention Act (PPA) require certain facilities to report release and other waste management
quantities of TRI-listed chemicals annually when a reporting threshold is triggered, but these statutes do
not impose any monitoring burden for determining the quantities.
TRI data are self-reported by the subject facility where some facilities are required to measure or
monitor emissions or other waste management quantities due to regulations unrelated to the TRI
program, or due to company policies. These existing, readily available data are often used by facilities
for TRI reporting purposes. When measured {e.g., monitoring) data are not "readily available," or are
known to be non-representative for TRI reporting purposes, the TRI regulations require that facilities
determine release and other waste management quantities of TRI-listed chemicals by making
"reasonable estimates." Such reasonable estimates include a variety of different approaches ranging
from published or site-specific emission factors {e.g., AP-42), mass balance calculations, or other
engineering estimation methods or best engineering judgment. TRI reports are then submitted directly to
EPA on an annual basis and must be certified by a facility's senior management official that the
quantities reported to TRI are reasonable estimates as required by law.
Where releases are possible, but TRI data were not available, releases were mostly estimated using
release data from the European Union Risk Assessment Report (EURAR). EPA rated the release data
from the EURAR an overall confidence rating of High during the systematic review process. This rating
takes into account the reliability of the data (EPA considers the European Chemicals Agency [ECHA] to
be a reliable source), the representativeness of the data, the accessibility / clarity of the data, and the
variability and uncertainty of the data.
Where the above data were not available, EPA used relevant OECD Emission Scenario Documents
(ESDs) or EPA Generic Scenarios (GSs from the 2009 OECD ESD on Plastic Additives, the 2018 Draft
GS on the Application of Spray Polyurethane Foam, and the 2010 OECD ESD on Chemicals Used in the
Electronics Industry). ESDs and GSs are standard sources used by EPA/OPPT for engineering
assessments. These documents provide information on particular processes, including release sources,
emission factors, and method of release, disposal, treatment, or discharge.13 EPA attempts to address
variability in releases estimated with EURAR, OECD ESD, or EPA GS data by estimating ranges of
emission factors and release days, as further described below.
13 Additional information on OECD ESDs can be found at: http://www.oecd.org/chemicalsafetv/risk-
assessment/introductiontoeniissionscenariodocuments.htm. Additional information on EPA GSs can be found at:
https://www.epa.gov/tsca-screemng-tools/chemsteer-chemicai-screen.ing-tool-exposures-and-environ.mentai-reieases.
Page 100 of 723
-------
Specifically, for each exposure scenario, EPA estimated daily and annual quantities of HBCD released,
where applicable using the following parameters:
• The annual importation, processing, or use volume per site.
• The number of importation, processing, or use sites.
• The emission factors for releases of HBCD.
• The number of days of HBCD releases.
The general approach for determining annual importation, processing, or use volume and the associated
number of sites for each exposure scenario is discussed above.
An emission factor is the fraction of material emitted or released per unit volume {i.e., kg released/kg
throughput) during a specific activity or exposure scenario {e.g., import, processing, or use). EPA
determined emission factors either from EURAR data or from ESDs and GSs. Where available, EPA
used EURAR release data, which is available as annual site-specific HBCD release quantities. The
associated HBCD processing volumes at these sites were not provided in the EURAR. The EURAR only
provided the combined HBCD processing volume for all the sites for which release data was provided.
EPA could not calculate site-specific emission factors due to the lack of site-specific HBCD processing
volumes. Using EURAR data, EPA calculated overall emission factors for an exposure scenario by
dividing the total amount of HBCD released for all sites by the total HBCD processing volume for all
the sites. For the purpose of this Risk Evaluation, EPA refers to these emission factors as average
emission factors. In some cases, the EURAR provided what they call "worst-case" emission factors,
described as being derived from the site with the highest release estimates. In these cases, EPA used
these "worst-case" emission factors as they were reported by the EURAR because EPA could not
calculate them without the site-specific HBCD processing volumes. EPA used both the average and
"worst-case" emission factors from the EURAR to provide a range of emission factors and release
quantities.
Where EURAR data were not available, EPA used emission factors that were reported in OECD ESDs
or EPA GSs. Where there were multiple approaches for estimating emission factors in the ESDs or GSs,
such as from assuming different types of containers or vessels are being cleaned, EPA assessed a range
of emission factors. The information provided in ESDs and GSs generally do not have statistical
characterization of the emission factors.
EPA calculated a range of annual release quantities for each exposure scenario by multiplying the range
of emission factors and the annual throughput of HBCD at a site. EPA calculated daily release quantities
by dividing the range of annual release quantities by the estimated number of release days. For most
exposure scenarios, EPA estimated a range of release days to generate a range of daily release estimates.
In general, EPA used the lowest estimated value and the highest estimated value of number of release
days to develop a range. EPA does not know the statistical characterization {e.g., mean, maximum, 95th
percentile) of these ranges because EPA did not find a comprehensive dataset of release days from
which these statistics could be calculated. In order to develop estimates of release days in support of
determining these ranges, EPA used one or a combination of the following approaches, in order of
priority:
• Where available, EPA used the number of release days reported in the EURAR for the sites with
HBCD release days. The number of release days is based on industry data for sites that perform
the same operations as those being assessed.
Page 101 of 723
-------
• Where data on release days reported by industry was not available, EPA estimated the number of
release days using ESDs or GSs.
• Where data were limited using the above two approaches, EPA estimated the number of release
days using the European Communities Technical Guidance Document ( B). This
technical guidance document contains methodology for estimating the number of release days
using the industrial category (i.e., polymer industry, electronics), use category/function within
the industry(/'.e., flame retardant), lifecycle stage (i.e., manufacturing, formulation, or use), and
the production volume (tons/yr) of the chemical of interest (i.e., HBCD importation, processing,
or use volume). EPA estimated the number of release days using the most applicable industry
category, which was the polymer processing industry in most, but not all, exposure scenarios.
EPA then selected the most applicable use category/function within the industry for the exposure
scenario and used the assessed HBCD processing or use volume solely to determine number of
release days. In some cases, where the above two approaches could not be used, EPA developed
ranges of release days using this method by determining the lowest and highest number of
potential release days by varying the function and HBCD processing or use volume within an
industry category.
Using the HBCD volume, number of sites, a range of emission factors, and a range or release days, EPA
calculated a range of daily releases per site for each exposure scenario using Equation 2-1:
Equation 2-1. R = [(T -h Ns) x /] -h Nd
Where:
R =
V =
Ns =
f =
Nd =
the amount of HBCD released per day to water, air, or landfill from a site (kg per day per
site)
annual U.S. HBCD importation, processing, or use volume (kg per yr)
the number of U.S. importation, processing, or use sites (sites)
emission factor for release of HBCD to water, air, or landfill from a process (kg of HBCD
released to water or air or landfill per kg of HBCD imported, processed or used)
the number of release days per year from a site (days)
Specific details related to the use of release data or models and the calculation of ranges of emission
factors and release days for each exposure scenario are further described below.
Releases to air were assessed as hourly rates to enable the modeling of these releases for the assessment
of general population exposure. EPA assumes the industrial processes that are associated with the
exposure scenario are operated at least 8 hours/day. Furthermore, air release sources such as unloading
and addition into processing equipment may occur throughout a day, so EPA assumes air releases may
occur over the entire operation time of 8 hours/day. This may result in underestimation or
overestimation of the hourly rate of releases to air.
2.2.2 Repackaging of Import Containers
In the United States, HBCD was manufactured in three grades: fine powder, standard grade powder, and
granules (EC SO8b). HBCD particle size distribution in HBCD products varied depending on the
producer and is summarized as follows (NICNAS 2012b: ECHA. 2008b):
• For fine grade powder, the mean particle size was 2 to 19 |im.
• For standard grade powder, the mean particle size was 20 to 150 |im.
• For granules, the mean particle size was 560 to 2,400 |im.
Page 102 of 723
-------
HBCD was manufactured at a purity of 90% to 100% HBCD (NICNAS 2012b: KemI 20091 EPA
expects that HBCD would also be imported into the United States at this purity in standard grade
powder or granular form as specified above. HBCD may also be imported in EPS resin beads at a
concentration of 0.7% or in XPS masterbatch at a concentration of 40-70% (NICNAS 2012b; ECHA
2008b). Micronized (fine grade) powder is typically used in textile and adhesive formulations (NICNAS
2012b; ECHA. 2008b). which EPA has determined are no longer exposure scenarios in the United States
and are not assessed in this Risk Evaluation.
EPA has not identified information on the importation and repackaging of HBCD within the United
States. However, EPA expects that importation activities described in risk assessments performed by
other countries are similar to those performed in the United States.
The Australian Priority Existing Chemical Assessment Report on HBCD indicates that powder or
granular HBCD was imported into Australia in 25-kg polylined paper bags and states that this took place
prior to 2010. The report also indicates that EPS resin beads containing HBCD were imported in 25-kg
polylined paper bags and 700-kg lined meshed plastic bags (NICNAS 2012b). The European Union Risk
Assessment Report (EURAR) on HBCD indicates that HBCD powder was packaged in 850-kg boxes
(ECHA. 2008b). Based on information from the Australian Report and the EURAR, EPA evaluated
releases from repackaging assuming HBCD may be imported in 700-kg bags or 850-kg boxes, which
may be repackaged into differently sized containers, depending on customer demand, and quality control
(QC) samples may be taken for analyses.
Once imported into the United States, HBCD powder is used to produce XPS masterbatch or to directly
produce XPS foam.14 Imported EPS resin beads are used to produce EPS foam. Repackaging of import
containers occurs on an as-needed basis, driven by customer demand. Exposures and releases are not
expected if repackaging of HBCD into smaller containers does not occur.
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.1, EPA estimates environmental releases based on a processing volume of
100,000 pounds per site per year and estimates a single unidentified site for this exposure scenario.
Release Sources
Based on the process description, EPA infers that releases may occur from dust generation during the
transfer of HBCD powder, granules, or masterbatch from import containers into new containers and
from residual HBCD in the emptied import containers that are disposed of. NICNAS (2012b) includes
information from one company that repackaged HBCD in an open or semi-closed process. EPA does not
know the prevalence of closed repackaging systems in the United States and estimates dust releases as
described below. Repackaging of HBCD into smaller containers may involve the use of equipment, such
as hoppers. However, EPA believes that the cleaning of such equipment would be infrequent (e.g., done
for maintenance purposes only) and there would be minimal residual material in the equipment prior to
cleaning because such equipment would be designed for gravity flow of solid particulates. Therefore,
EPA did not assess releases from equipment cleaning in this exposure scenario. NICNAS (2.012b) and
14 In this Risk Evaluation, EPA refers to EPS and XPS foam articles, including insulation, as EPS and XPS foam. The
Problem Formulation for Cyclic Aliphatic Bromides Cluster (HBCD) prepared prior to this Risk Evaluation often referred to
these foam articles simply as EPS and XPS.
Page 103 of 723
-------
Environment Canada (EC/HC 2011) did not assess release from equipment cleaning. The EURAR
(ECHA 2008b) did not assess repackaging as a exposure scenario.
Emission Factors
EPA used the emission factors given in the 2009 OECD ESD on Plastic Additives (OECD 2009).
specifically for flame retardants used in activities expected to occur during this exposure scenario, as
described below. The 2009 OECD ESD on Plastics Additives estimates releases by applying emission
factors to the throughput of the chemical of interest, in this case HBCD (OECD 2009). For dust releases,
the OECD ESD estimates an emission factor of up to 0.5% for fine particles (<40 jam) and 0.1% for
coarse particles (>40 pm). EPA uses this range of emission factors to estimate dust releases. Per the
OECD ESD, the initial release is to air, with particles eventually settling and being disposed of as solid
waste or discharged in wastewater from cleaning of surfaces onto which the particles have settled
(OECD 2009). The specific method of release, disposal, treatment, or discharge is dependent on site-
specific factors, such as any pollution controls that are implemented at that site, as well as other factors
such as the equipment used and size of the importation site. EPA does not know the prevalence of dust
capture and control technologies at importation sites in the United States. Depending on site-specific
conditions, HBCD may be released to stack air or fugitive air, discharged to POTW or onsite WWT,
disposed of to landfill, or treated via incineration (OECD 2009).
For container residue, the OECD ESD on Plastics Additives uses an emission factor of 1%. The OECD
ESD indicates that containers are likely to be disposed of to landfill. EPA uses this emission factor to
estimate release of solid HBCD from container disposal to landfill. Although there is no statistical
characterization of this emission factor, EPA believes the 1% emission factor is in the upper end of the
distribution based on EPA's experience. No other release sources are identified in the OECD ESD or
expected by EPA, based on the process description, for this exposure scenario.
A summary of the release sources assessed by EPA is presented in Table 2-6.
Table 2-6. Summary of H
3CD Release Sources During Repackaging of Import Containers
Release Source
Emission Factor used in this
Risk Evaluation
Method of Release,
Disposal, Treatment, or
Discharge Assessed in
this Risk Evaluation
Basis or Source
Dust generation from
unloading solid standard
grade powder from import
containers into new
containers
0.001-0.005 kg HBCD
released/kg HBCD handled
Uncertain:
Stack air, or Fugitive
Air, POTW, Onsite
WWT, Landfill, or
Incineration
(OECD 2009)
Disposal of import
containers (bags) containing
solid HBCD
0.01 kg HBCD released/kg
HBCD in containers
100% Landfill
(OECD 2009)
Number of Release Days
EPA estimated the number of release days based on information in the European Communities
Technical Guidance Document (ECB 2003). EPA estimated the lowest and highest possible number of
release days per year using data from the basic chemicals industry category in the European
Communities Technical Guidance Document. EPA calculates a lower value of 29 days/year and an
upper value of 300 days/year. This range of number of release days per year seems reasonable in
Page 104 of 723
-------
comparison to information from the Australian risk assessment (NICNAS 2012b) which indicates that
one company in Australia infrequently repackaged HBCD imported in 25-kg bags into 15-kg bags at a
rate of one metric ton of HBCD repackaged every three months over a period of five days per
repackaging campaign. Using this repackaging rate of one metric ton (2,205 pounds) over five days and
EPA's production volume of 100,000 pounds HBCD/year, EPA calculates a United States repackaging
frequency of approximately 227 days/year. The estimate of 227 day/year falls within the range of 29 to
300 days/year. Based on these data, EPA estimated a range of release days for this exposure scenario of
29 to 300 days/year.
The data sources used to estimate releases in this section are listed in Table 2-7 with the data quality
score. See Appendix D for more details about data source evaluation.
Table 2-7. Repackaging of Import Containers - HI
»CD Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence Rating
of Data
(ECB 2003)
Days of Release
29 to 300 days/year for all
releases
Medium
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-8.
Table 2-8. Input Variables to Equation 2-1 for Repackaging of HBCD Import Containers
Input Variable
V
(of HBCD)
Ns
(sites)
f
(kg HBCD released/kg HBCD imported)
Nd
(days/yr)
Lower value of emission factors
Upper value of emission factors
100,000 pounds/year =
45,359 kg/yeara
1
0.001 to Stack air. Fugitive Air,
POTW, Onsite WWT, Landfill,
and/or Incineration
0.01 to Landfill
0.005 to Stack air. Fugitive Air,
POTW, Onsite WWT, Landfill, and/or
Incineration
0.01 to Landfill
29-300
a CDR rcDortina threshold for small manufacturers (U.S. EPA 2016b)
The results of these calculations for all methods of release, disposal, treatment, or discharge are
summarized in Table 2-9. EPA presents a range of release estimates from the 2009 OECD ESD on
Plastic Additives (OECD 2009). varied over a range of release days, as previously discussed. The
repackaging of import containers may result in releases to air, discharge to POTW, and/or disposal to
landfill. Overall, disposal to landfill exceeds air releases and wastewater discharges, largely due to the
disposal of the bags in which HBCD is imported.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed range of daily release rates presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
The result of EPA's systematic review is data pertaining to the number of release days with an overall
confidence rating of medium; the quality of the emission factor data was not evaluated because this data
was obtained from an OECD ESD.
Page 105 of 723
-------
The strength of the assessment approach is the estimation of HBCD emission factors and number of
release days as ranges of values to account for variability in the values of these two parameters that EPA
obtained. Furthermore, the strength of the assessment approach is the estimation of the daily release of
HBCD per site as a range of values which encompasses the range of emission factors and the number of
release days that EPA obtained.
There is uncertainty about the extent to which the emission factor data and the data on number of days
of release per year are applicable to the HBCD processing that would occur in the U.S. Based on the
strength and uncertainty of the assessment, EPA has medium confidence in the assessment results.
Page 106 of 723
-------
Table 2-9. Summary of HBCD Releases from Repackaging of Import Containers
Release Source
Method of Release,
Disposal, Treatment,
or Discharge (a)
Releases calculated from lower value of range of
emission factors b
Releases calculated from upper value of range of
emission factors b
Number
of Sites
Hours of
Release
per Day
(hr/day)
Total
Annual
Release for
All Sites
(kg/yr)
Annual
Release per
Site
(kg/site-yr)
Daily Release
(kg/site-day)
Total
Annual
Release for
All Sites
(kg/yr)
Annual
Release Per
Site
(kg/site-yr)
Daily Release
(kg/site-day)
Number of
release days:
29 days/year
Number
of release
days: 300
days/year
Number of
release days:
29 days/year
Number of
release days:
300 days/year
Dust release
during
unloading of
HBCD
May go to one or
more: stack air,
fugitive air, on-site
WWT, POTW,
landfill, or incineration
45.4
45.4
1.56
0.15
227
227
7.82
0.756
1
8
hours/day
Disposal of
transport bags
containing solid
HBCD residual
Landfill
454
454
15.64
1.51
454
454
15.64
1.51
1
8
hours/day
a The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment
use, size of the site, and waste handling practices, including any pollution controls used.
b Release estimates are quantities of HBCD. The physical form of these releases is solid HBCD.
Page 107 of 723
-------
2.2.3 Compounding of Polystyrene Resin to Produce XPS Masterbatch
Imported HBCD powder or granules may be compounded into an XPS masterbatch prior to being sold
to XPS foam manufacturers, who then convert the XPS masterbatch into XPS foam. Imported HBCD
powder may be sent to XPS masterbatch compounding sites in 25-kg bags or supersacks (ECHA. 2008b).
HBCD is unloaded into a hopper and pre-blended with polystyrene in the hopper or else transferred
directly to mixing equipment. From the mixer, the mixture is then fed into an extruder where it is
extruded through a die to produce pellets or granules (NICNAS 2012b). The pellets or granules are air-
cooled or cooled in a water bath, dried, and then packaged (EC 08b). The HBCD content in the
XPS masterbatch is up to 40-70% of the pellets (NICNAS 1 ' . , * ^ 2008b). The packaged XPS
masterbatch is then sent to converting sites, where it is turned into XPS foam.
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.1, EPA estimates environmental releases based on a processing volume of
100,000 pounds per site per year and estimates a single unidentified site for this exposure scenario.
Release Sources
Based on the process description, EPA infers that releases may occur from: dust generation during
unloading of the HBCD powder or granules from the bags in which they were received and during the
compounding process; disposal of the bags in which the HBCD powder is received; and cleaning of
process equipment.
Emission Factors
EPA estimated emission factors based on site-specific release data reported in the EURAR (ECHA
2008b). The EURAR identified 14 sites in the EU that compound polystyrene to produce XPS
masterbatch that is flame retarded with HBCD (ECHA. 2008b). Site-specific annual release rates of solid
HBCD were reported for three of the sites, indicating releases to wastewater and air, which are
summarized in Table 2-10. To maintain confidentiality, the EURAR did not provide site-specific HBCD
processing volumes with which site-specific emission factors could be calculated. However, the EURAR
provided the total HBCD processing volume for the three sites for which release data is available. EPA
calculated overall average emission factors to air and water by dividing the total HBCD release to air or
water from all three sites by the total HBCD processing volume for the three sites. EPA calculated
overall average emission factors of 3.22xl0"5 kg HBCD discharged/kg HBCD processed to water and
6.12 xlO"6 kg HBCD released/kg HBCD processed to air.
The EURAR also provided emission factors of 7.42xl0"5 kg HBCD discharged/kg HBCD processed to
water and 7.3 lxlO"6 kg HBCD released/kg HBCD processed to air, indicating that these are the "worst-
case" factors that the EURAR calculated using the site-specific release and HBCD processing volume
data from the three sites. Because site-specific HBCD processing volume data were not provided, EPA
could not calculate these "worst-case" emission factors. EPA used both the "worst-case" emission
factors as they were reported in the EURAR and the average emission factors calculated by EPA to
provide a range of release estimates during this exposure scenario.
The EURAR indicates that wastewater discharges are to wastewater treatment. EPA did not identify
information about the prevalence of wastewater treatment at these types of processing sites in the United
States and hence assumed that water discharges from this exposure scenario can be to surface water,
POTW, and/or onsite wastewater treatment. The EURAR does not specify if the reported air releases for
Page 108 of 723
-------
these three sites are to stack or fugitive air. Sites may implement dust capture technologies that may
determine whether this release is to stack or fugitive air. EPA did not identify information on the
prevalence of dust capture technologies at processing sites in the United States and assesses this release
may include stack air and/or fugitive air.
Table 2-10. HBCD Release Data Reported in the EURAR for XPS Masterbatch Production
Site-Specific Release Data
Process Volume
Site Identity
Release to Water
Release to Air
kg/yr
kg/yr
Site 1
0.12
2.6
The EURAR identifies a total of 1,160
metric tons of HBCD is processed at the 3
sites with site-specific release data.
Site 2
0.27
1.2
Site 3
37
3.3
Number of Release Days
EPA estimated the number of release days based on information reported in the European Communities
Technical Guidance Document (ECB 2003) because the actual number of release days associated with
the site-specific annual release rates discussed above is not reported in the EURAR. Instead, the number
of release days reported in the EURAR are defaults recommended in the European Communities
Technical Guidance Document (ECB 2003). The Environment Canada assessment also estimated
emission days for compounding with the same methodology (EC/HC 2011). HBCD compounding
occurs once per day at a site for the production of polystyrene masterbatch according to the Australian
risk assessment. EPA did not use this information because the HBCD processing volume is not reported.
Using the European Communities Technical Guidance Document (ECB 2003) and the defaults for
formulation within the polymer industry, EPA estimated 60 emission days/year for an HBCD processing
volume of 100,000 pounds (45.3 metric tons). EPA used the 2014 Draft OECD ESD on Use of
Additives in the Plastics Compounding to estimate the number of release days during this exposure
scenario. The OECD ESD indicates that, based on EPA new chemical submissions from industry, that
the lowest number of operating days reported was 10 days/year (U.S. EPA 2014a). Based on these data,
EPA estimated a range of release days of 10 to 60 days/year.
The data sources used to estimate releases in this section are listed in Table 2-11 along with the data
quality score. See Appendix D for more details about data source evaluation.
Page 109 of 723
-------
Table 2-11. Compounding of Polystyrene to Produce XPS Masterbatch Release Data Source
Evaluation
Source Reference
Data Type
Value
Overall Confidence Rating of
Data
(ECHA 2008b)
Site-Specific Release
Data
See Table 2-10
High
(ECHA 2008b)
"Worst-Case" Emission
Factors
7.42xl0"5 to water
and 7.31 xlO"6 to air
High
(ECB 2003)
Release Days
10 to 60 days/year
for all releases
Medium
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-12.
Table 2-12. Input Variables to Equation 2-1 for XPS Masterbatch Production
Input Variable
V
(of HBCD)
Ns
(sites)
f
(kg HBCD released/kg HBCD processed)
Nd
(days/yr)
Average calculated from
EURAR data
"Worst-case" given in EURAR
100,000 pounds/year
= 45,359 kg/year
1
6.12E-06 to stack air and/or
fugitive air
3.22E-05 to surface water, onsite
WWT. and/or POTW
7.31E-06 to stack air and/or
fugitive air
7.42E-05 to surface water, onsite
WWT, and/or POTW
10-60
a CDR reporting threshold for small manufacturers (U.S. EPA 2016b)
The daily amount of solid HBCD released per site from compounding of polystyrene to produce XPS
masterbatch was calculated with Equation 2-1. The results of these calculations are summarized in Table
2-13.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium to high confidence in the assessed range of daily release rates presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
As detailed in Table 2-11, the result of EPA's systematic review is data with an overall confidence
rating of high or medium, which is a strength of the assessment. In particular, the overall confidence
rating of the data pertaining to the number of release days is medium.
Another strength of the assessment approach is the estimation of HBCD emission factors and number of
release days as ranges of values to account for variability in the values of these two parameters that EPA
obtained or estimated. Furthermore, the strength of the assessment approach is the estimation of the
daily release of HBCD per site as a range of values which encompasses the range of emission factors
and the number of release days that EPA obtained or estimated.
Page 110 of 723
-------
There is uncertainty about the extent to which the emission factor data, including the emission factors
calculated from release and processing volume data, and the data on number of days of release per year
are applicable to the HBCD processing that would occur in the U.S. Based on the strength and
uncertainty of the assessment, EPA has medium to high confidence in the assessment results.
Page 111 of 723
-------
Table 2-13. Summary of HBCD Releases from XPS Masterbatch Production
Release Source
Method of
Release,
Disposal,
Treatment, or
Dischargea
Releases calculated from average emission factor based
on EURAR release data b
Releases calculated from worst case
emission factor as it was reported in the
EURARb
Number
of Sites
Hours of
Release
per Day
(hr/day)
Total Annual
Release for All
Sites
(kg/yr)
Annual
Release Per
Site
(kg/site-yr)
Daily Release (kg/site-day)
Total
Annual
Release for
All Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release
(kg/site-day)
Number of
release days:
10 days/year
Number of
release days:
60 days/year
Number of
release
days: 10
days/year
Number
of release
days: 60
days/year
Unknown - these data
were reported by EU
sites in the EURAR
as total annual release
per site
May go to one or
more: Stack air or
fugitive air
0.278
0.278
0.028
4.63E-03
0.332
0.332
0.033
5.53E-03
1
8 hours/day
Unknown - these data
were reported by EU
sites in the EURAR
as total annual release
per site
May go to one or
more: Surface
Water, Onsite
WWT, orPOTW
1.46
1.46
0.15
2.44E-02
3.37
3.37
0.337
5.61E-02
1
8 hours/day
a The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment
use, size of the site, and waste handling practices, including any pollution controls used.
b Release estimates are quantities of HBCD. The physical form of these releases is solid HBCD or solid mixtures containing polystyrene and HBCD.
Page 112 of 723
-------
2.2.4 Processing to Produce XPS Foam using XPS Masterbatch
XPS masterbatch is used to make XPS foam. The HBCD content in the XPS masterbatch ranges from 40
to 70 weight percent within the XPS masterbatch pellets or granules flSflCNAS 2012b; ECHA 2008b).
Once received at XPS foam production sites, the XPS masterbatch, along with additional polystyrene
and other additives such as dyes, are charged to an extruder (ECHA. 2008b). In the extruder, the
polystyrene is melted, allowing the HBCD and other additives to become suspended in a polymer gel.
Blowing agent is added to the gel, the gel is cooled, and it is then extruded through a die where the
blowing agent volatilizes. This volatilization within the plastic gel causes the plastic to become a foam
as it is extruded (ECHA. 2008b). HBCD content in XPS foam ranges from 0.5 to 3 wt% (U.S. EPA.
2015a: Takieami etal. u, U HC^< n, , I bi \ 2008b).
Once the XPS foam is made, it may be cut, sawed, or machined into various shapes (often referred to as
secondary processing), shrink-wrapped, palleted, and shipped to structural insulated panels (SIPs) and
automobile replacement part production sites or directly to end users for installation into structures such
as buildings (E( 108b). Additionally, XPS foam scraps from secondary processing or off-
specification products may be ground and recycled back into the XPS foam production process (often
referred to as reclamation) (EC )08b).
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.1, EPA estimates environmental releases based on a processing volume of
100,000 pounds per site per year and estimates a single unidentified site for this exposure scenario.
Release Sources
Based on the process description, EPA infers that HBCD releases may occur from: dust generation
during unloading the XPS masterbatch from the bags in which they were received; disposal of the bags
in which the XPS masterbatch is received; and periodic cleaning of process equipment.
Foam manufacturing sites may also generate dust and scraps from cutting or trimming of XPS foam into
panels or other shapes for shipment to end users. However, both the EU and Australian risk assessments
specify that industry provided information indicated that generated dust and trimmings may be recycled
back into the foam molding process, thereby reducing or eliminating waste from the cutting and
trimming process (NICNAS 2012b: ECHA. 2008b). EPA does not know the extent that these practices
are used in the United States and the assessed EURAR data is expected to account for any releases from
this source (ECHA 2008b).
Emission Factors
EPA estimated emission factors based on site-specific solid HBCD release data reported in the EURAR
(ECHA 2008b). The EURAR identified 17 sites in the EU that produce XPS foam using XPS
masterbatch that is flame retarded with HBCD (ECHA 2008b). Site-specific release quantities are
provided for four of these sites, which are summarized in Table 2-14. The EURAR indicates that these
sites did not provide air releases and that these air emissions were calculated using emission factors from
a study on emissions at three European XPS foam manufacturing plants (ECHA 2008b). To maintain
confidentiality, the EURAR did not provide site-specific HBCD process volumes with which site-
specific emission factors could be calculated. However, the EURAR provided the total production
Page 113 of 723
-------
volume for the four sites for which release data are available. EPA calculated overall average emission
factors to air and water by dividing the total HBCD releases to air or water from all four sites by the
total HBCD processing volume for the four sites. From these calculations, EPA estimated average
emission factors of 1.07xl0"5 kg HBCD discharged/kg HBCD processed to water and 5.79 xlO"5 kg
HBCD released/kg HBCD processed to air.
The EURAR also calculated estimates of releases to wastewater and air from 13 sites that did not
provide release data by using "worst-case" emission factors that the EURAR calculated from the
available site-specific HBCD release and processing volume data. However, the EURAR did not
provide the "worst-case" emission factors used to determine these estimates. EPA calculated "worst-
case" emission factors by using the total "worst-case" release estimates calculated by the EURAR for
the 13 sites and the HBCD processing volume identified in the EURAR for these 13 sites, as presented
in Table 2-14. EPA calculated "worst-case" emission factors to be 2.63xl0"5 kg HBCD discharged/kg
HBCD processed to water and 5.80xl0"5 kg HBCD released/kg HBCD processed to air. The "worst-
case" air emission factor and average air emission factor are the same because the EURAR used the
same emission factor from a study of three European XPS foam manufacturing plants, as described
above (ECHA 2008b).
The EURAR indicates that wastewater discharges are to wastewater treatment. EPA did not find
information about the prevalence of wastewater treatment at processing sites in the United States and
hence assumed that wastewater discharges from this exposure scenario can be to surface water, POTW,
and/or onsite wastewater treatment. The EURAR does not specify if the reported air releases for these
three sites are to stack or fugitive air. Sites may implement dust capture technologies that affect if this
release is to stack or fugitive air. EPA did not find information about the prevalence of dust capture
technologies at processing sites in the United States and hence assumed this release may include stack
air and/or fugitive air.
Table 2-14. HBCD Release Data Reported in the EURAR for Manufacturing of XPS Foam from
XPS Masterbatch
Site
Release to Water
Release to Air a
Process Volume
kg/yr
kg/yr
Site 1
2.2
0.31
The EURAR identifies a
total of 719 metric tons of
HBCD is processed at the
4 sites with site-specific
release data.
Site 2
0
18
Site 3
1.3
14
Site 4
4.2
9.3
Total "worst-case"
emissions calculated in
the EURAR for 13 sites
without release data
26.67
58.617
The EURAR identifies a
total of 1,011 metric tons
of HBCD is processed at
the 13 sites without release
data.
a These air releases were not reported by the sites by were estimated in the EURAR using emission factors from a study on
emissions from three European XPS foam manufacturing sites (EC HA 2008b).
Page 114 of 723
-------
Number of Release Days
The site-specific data in the EURAR indicates wastewater discharges occur over 1 to 15 days/year,
which are values reported by the sites. Only one site reported emission days for air releases, reporting 15
days/year. Based on these data, EPA estimated wastewater discharges over a range of 1 to 15 days/year.
The remaining three sites did not report emission days for air releases and the EURAR estimated 300 air
emission days for all the sites using defaults in the European Communities Technical Guidance
Document for industrial use in the polymers industry and processing volume at the individual sites (ECB
2003). Using this same European guidance and EPA's HBCD processing volume of 100,000 pounds
HBCD/year (45.4 metric tons), EPA estimated 16 days of emission per year. In lieu of using a range of
15 to 16 days of air emission per year, EPA used 1 day/year as the lower bounding estimate, using the
same low-end of emission days as that reported by the EU sites for wastewater discharges, and 16
days/year based on the European Communities Technical Guidance Document.
The data sources used to estimate releases in this section are listed in Table 2-15 along with the data
quality score. See Appendix D for more details about data source evaluation.
Table 2-15. XPS Foam M
anufacturing Using XPS Masterbatch Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence Rating
of Data
(ECHA 2008b)
Site-Specific Release Data
See Table 2-14
High
(ECHA 2008b)
"Worst-Case" Emissions for
Sites without Release Data
2.63xl0 5 to water and
5.80xl0~5 to air
High
(ECHA 2008b)
Release Days
1 to 15 days/year for water
releases; 15 days/year for air
releases
High
(ECB 2003)
Release Days
16 days/year for all releases
Medium
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-16.
Table 2-16. Input Variables to Equation 2-1 for XPS Foam Manufacturing Using XPS
Masterbatch
Input Variable
V (of HBCD)
Ns
(sites)
f
(kg HBCD released/kg HBCD processed)
Nd
(days/yr)
Average calculated from
EURAR data
"Worst-Case" calculated
from EURAR data
100,000 pounds/year
= 45,359 kg/yeara
1
5.79E-05 to stack air and/or
fugitive air
1.08E-05 to surface water, onsite
WWT, and/or POTW
5.80E-05 to stack air and/or
fugitive air
2.63E-05 to surface water,
onsite WWT, and/or POTW
1-15 (wastewater
discharge), 1-16 (air
release)
a CDR rcDortina threshold for small manufacturers (U.S. EPA 2016b)
The daily amount of solid HBCD released per site from XPS foam manufacturing from XPS
masterbatch was calculated with Equation 2-1. The results of these calculations are summarized in Table
2-17.
Page 115 of 723
-------
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium to high confidence in the assessed range of daily release rates presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
As detailed in Table 2-15, the result of EPA's systematic review is data with an overall confidence
rating of high or medium, which is a strength of the assessment.. In particular, the overall confidence
rating of the data pertaining to the number of release days is high or medium.
The strength of the assessment approach is the estimation of HBCD emission factors and number of
release days as ranges of values to account for variability in the values of these two parameters that EPA
obtained or estimated. Furthermore, the strength of the assessment approach is the estimation of the
daily release of HBCD per site as a range of values which encompasses the range of emission factors
and the number of release days that EPA obtained or estimated.
There is uncertainty about the extent to which the emission factor data, including the emission factors
calculated from release and processing volume data, and the data on number of days of release per year
are applicable to the HBCD processing that would occur in the U.S. Based on the strength and
uncertainty of the assessment, EPA has medium to high confidence in the assessment results.
Page 116 of 723
-------
Table 2-17. Summary of HBCD Releases from XPS Foam Manufacturing Using XPS Masterbatch
Method of
Release,
Discharge,
Treatment, or
Disposala
Releases calculated from average emission factor
based on EURAR release data b
Releases calculated from "worst case" emission
factor based on EURAR release data b
Total
Annual
Release
Per Site
(kg/site-
yr)
Daily Release (kg/site-day)
Total
Annual
Release
Per Site
(kg/site-
yr)
Daily Release (kg/site-day)
Hours of
Release per
Day (hr/day)
Release Source
Annual
Release
for All
Sites
(kg/yr)
Number of
release days:
1 day/year
(water and
air)
Number of
release days:
15 day/year
(water) and 16
day/year (air)
Annual
Release
for All
Sites
(kg/yr)
Number of
release days:
1 day/year
(water and
air)
Number of
release days:
15 day/year
(water) and 16
day/year (air)
Number
of Sites
Unknown - these
data were reported
by EU sites in the
EURAR as total
annual release per
site
May go to one or
more: Stack air
or fugitive air
2.63
2.63
2.63
0.164
2.63
2.63
2.63
0.164
1
8 hours/day
Unknown - these
data were reported
by EU sites in the
EURAR as total
annual release per
site
May go to one or
more: Surface
Water, Onsite
WWT, orPOTW
0.486
0.486
0.486
3.24E-02
1.19
1.19
1.19
0.080
1
8 hours/day
a The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
b Release estimates are quantities of HBCD. The physical form of these releases is solid HBCD or solid mixtures containing polystyrene and HBCD.
Page 117 of 723
-------
2.2.5 Processing of HBCD to Produce XPS Foam
XPS foam can be produced from either XPS masterbatch, as described in Section 2.2.4, or from HBCD
powder or granules. The process for producing XPS foam from HBCD powder is similar to that for
production of HBCD foam from XPS masterbatch. Polystyrene, HBCD powder, and other additives are
fed into an extruder, where the contents are melted to produce a plastic gel. Blowing agent is added to
the gel, which is then sent through a die where the blowing agent volatilizes, producing the extruded
plastic foam. The foam may be cut into shapes, packaged, and shipped to customers. HBCD content in
XPS foam ranges from 0.5 to 3 weight percent (U.S. EPA. 2015a; Takieami e> it Ji'i (, S 1 ;HC 2011;
ECHA 2008bY
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.1, EPA estimates environmental releases based on a processing volume of
100,000 pounds per site per year and estimates a single unidentified site for this exposure scenario.
Release Sources
Based on the process description, EPA infers that releases may occur from: dust generation during
unloading the HBCD powder from the bags in which they were received; disposal of the bags in which
the HBCD powder is received; and periodic cleaning of process equipment.
Foam manufacturing sites may also generate dust and scraps from cutting or trimming of XPS foam into
panels or other shapes for shipment to end users. However, both the EU and Australian risk assessments
specify that industry provided information indicating that generated dust and trimmings may be captured
and recycled back into the foam molding process, thereby reducing or eliminating waste from the cutting
and trimming process (NICNAS 2012b; ECHA. 2008b). EPA does not know the extent to which these
practices are used in the United States and the assessed TRI and EURAR data is expected to account for
any releases from this source (EC )8bY
EPA estimated releases from this exposure scenario using 2017 TRI data and emission factors calculated
from release data from the EURAR. EPA assessed both approaches because the company that reported
to 2017 TRI indicated that they no longer conduct operations with HBCD, as discussed below and did
not report to 2018 TRI as indicated in Section 1.2.4.
TRI Data
The Dow Chemical Company reported releases for two sites that manufacture XPS foam with HBCD.
The company has since indicated that operations with HBCD have ceased. The Dow Chemical
Company communicated with EPA that they imported roughly 48 metric tons in 2017 as discussed
earlier in Section 1.2, which is similar to the importation and processing volume of HBCD that EPA
uses to estimate releases for this exposure scenario (approximately 45.4 metric tons) with the EURAR
data. EPA assessed the 2017 TRI releases as they were reported by Dow. These releases are deemed to
be representative of the potential releases that may occur from sites in the United States that would
manufacture XPS foam with HBCD because the processed volume associated with these releases is
approximately equal to the assessed processing volume. The reported releases are summarized in the
next section along with the releases EPA calculated from the EURAR data. As discussed, the HBCD
processing volume associated with the releases reported in the 2017 TRI (48 metric tons HBCD,
provided through communication with Dow and discussed in Section 1.2) is slightly different than the
volume EPA used to estimate releases from the EURAR data (45.4 metric tons).
Page 118 of 723
-------
Emission Factors
Although TRI data are available for this exposure scenario, EPA also estimated emission factors based
on site-specific release data reported in the EURAR (ECHA 2008b). The EURAR identified 18 sites in
the EU that produce XPS foam using HBCD powder (ECHA 2008b). Site-specific solid HBCD release
quantities are provided for 17 of these sites and a calculated release estimate was provided for the
remaining site. To maintain confidentiality, the EURAR did not provide site-specific HBCD processing
volumes with which site-specific emission factors could be calculated. The EURAR only provided the
total HBCD processing volume for all 18 sites (ECHA 2008b).
EPA calculated overall average emission factors to water and air with this data by dividing the total
HBCD releases for water or air for all sites by the total HBCD processing volume for all sites. The
average emission factors are presented in Table 2-18.
The EURAR indicates that the HBCD release estimates to water presented in Table 2-18 may be
estimated quantities either directly from process operations or from onsite wastewater treatment at these
sites. The EURAR does not specify this detail for the individual sites, thus EPA is uncertain of the
prevalence of onsite wastewater treatment at these European sites. For this Risk Evaluation, EPA
assessed that wastewater discharges estimated using the emission factor determined from the EURAR
data may be entirely to on or offsite wastewater treatment or to surface water. Depending on site-
specific pollution controls, wastewater discharges can be to surface water, POTW, and/or onsite
wastewater treatment and air releases may include stack air and/or fugitive air.
Table 2-18. HBCD Release Data Reported in the EURAR for Manufacturing of XPS Foam using
IBCD Powder
Site-Specific
Release to Water
Release to Air
Process Volume
Release Data
kg/yr
kg/yr
Site 1
4.4
1.5
Site 2
1.2
1.4
Site 3
0.055
3.7
Site 4
3.7
1.5
Site 5
0.0024
1.1
Site 6
0
0.73
Site 7
6
0.54
Site 8
0.0029
0.7
The EURAR identifies a total of 3,232 metric tons of HBCD are
Site 9
0.0019
0.15
processed into XPS masterbatch by 18 sites.
Site 10
0
0.4
Site 11
0
1.8
Site 12
0
1.8
Site 13
0.11
1.2
Site 14
15
1.5
Site 15
0.00004
0.59
Site 16
0.0004
0.91
Page 119 of 723
-------
Site 17
0.021
3.8
Site 18
2.5
0.23
Number of Release Days
The site-specific data in the EURAR indicates wastewater discharges occur over 1 to 12 days/year,
which are values reported by the EU sites. Based on these data, EPA estimated wastewater discharges
over a range of 1 to 12 days/year. None of these sites reported emission days for air releases. For these
sites, the EURAR estimated 42 to 300 air emission days using defaults in the European Communities
Technical Guidance Document for industrial use in the polymers industry and processing volume (ECB
2003). Using this same European guidance and a processing volume of 100,000 pounds HBCD/year
(45.4 metric tons), EPA estimated 16 days of emission per year. EPA used 1 day/year for air emissions
as the lower bounding estimate, using the same low-end of emission days as that reported by the EU
sites for wastewater discharges, and 16 days/year based on the European Communities Technical
Guidance Document.
The data sources used to estimate releases in this section are listed in Table 2-19 along with the data
quality score. See Appendix D for more details about data source evaluation.
Table 2-19. Manufacturing of XPS Foam Using I
BCD Powder Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence Rating
of Data
(ECHA 2008b)
Site-Specific Release Data
See Table 2-18. HBCD
Release Data Reported in the
EURAR for Manufacturing
of XPS Foam using HBCD
Powder
High
(U.S. EPA 2017e)
Site-Specific Release Data
See Table 2-22. Summary of
HBCD Releases from XPS
Foam Manufacturing Using
HBCD from 2017 TRI Data
Medium
(ECHA 2008b)
Release Days
1 to 12 days/year for
wastewater discharges
High
(ECB 2003)
Release Days
16 days/year for all releases
Medium
Page 120 of 723
-------
Environmental Release Assessment Results
The releases reported by the Dow Chemical Company in the 2017 TRI for sites that manufacture XPS
articles with HBCD are presented in Table 2-22. The data in 2017 TRI is reported for the calendar year.
EPA calculated daily releases with the TRI data using the same estimates for days per year that is
discussed above. EPA also calculated releases using Equation 2-1 and the EURAR data discussed above,
and the input variables for this calculation are given in Table 2-20. The results of these calculations are
summarized in Table 2-21.
Table 2-20. Input Variables to Equation 2-1 for XPS Foam Manufacturing Using HBCD Powder
Input Variable
Volume (of
HBCD)
Ns
(sites)
f
(kg HBCD released/kg HBCD processed)
Nd
(days/yr)
Average calculated from EURAR data
100,000
pounds/year =
45,359 kg/year
1
7.29E-06 to stack air and/or fugitive air
1.02E-05 to surface water, onsite WWT, and/or POTW
1-12 (water),
1-16 (air)
a CDR rcDortina threshold for small manufacturers (U.S. EPA 2016b)
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium to high confidence in the assessed range of daily release rates. EPA also assessed
releases using TRI data which EPA assigned an overall confidence rating of medium using systematic
review. EPA considered the quality of the data, the assessment approach, and uncertainties in
assessment results to determine the level of confidence.
As detailed in Table 2-19, the result of EPA's systematic review is data with an overall confidence
rating of high or medium, which is a strength of the assessment.. In particular, the overall confidence
rating of the data pertaining to the number of release days is high or medium.
The strength of the assessment approach is the estimation of number of release days as ranges of values
to account for variability in parameters that EPA obtained or estimated. Furthermore, the strength of the
assessment approach is the estimation of the daily release of HBCD per site as a range of values which
encompasses the number of release days and different sources of release data that EPA obtained or
estimated.
There is uncertainty about the extent to which the emission factor data, including the emission factors
calculated from release and processing volume data, and the data on number of days of release per year
are applicable to the HBCD processing that would occur in the U.S. Based on the strength and
uncertainty of the assessment, EPA has medium to high confidence in the assessment results.
Page 121 of 723
-------
Table 2-21. Summary of HBCD Releases from XPS Foam Manufacturing Using HBCD
Release
Source
Method of
Release,
Disposal,
Treatment,
or Discharge
a
Releases calculated from average emission factor based
on EURAR release data b
Number
of Sites
Hours of
Release
per Day
(hr/day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release (kg/site-day)
Number of
release days:
1 day/year
(water and
air)
Number of release
days: Over 12
day/year (water)
and 16 day/year
(air)
Unknown -
these data
were reported
by EU sites in
the EURAR
as total annual
release per
site
May go to one
or more: Stack
air or fugitive
air
0.331
0.331
0.331
2.07E-02
1
8 hours/day
Unknown -
these data
were reported
by EU sites in
the EURAR
as total annual
release per
site
May go to one
or more:
Surface
Water, Onsite
WWT, or
POTW
0.463
0.463
0.463
0.039
1
8 hours/day
3 The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-
specific conditions, including type of equipment use, size of the site, and waste handling practices, including any pollution
controls used.
b Release estimates are quantities of HBCD. The physical form of these releases is solid HBCD or solid mixtures
containing polystyrene and HBCD.
Table 2-22. Summary of HBCD Releases from XPS Foam Manufacturing Using HBCD from 2017
TRI Data
Site
identity
2017 TRI
Hours of
Release
per Day
(hr/day)
Annual Quantities per Site
(kg/year)
Daily Release (kg/site-day)
Assuming low-end of 1
day/year
Assuming high-end of 16
days/year
Dow
Chemical
Company,
Pevely MO
Stack air3: 1.81
Off-site transfer for
Incineration b: 30.8
Off-site transfer for disposal
to landfill °: 123
Stack air3: 1.81
Off-site transfer for
Incineration b: 30.8
Off-site transfer for
disposal to landfill °: 123
Stack air3: 0.113
Off-site transfer for
incineration b: 1.93
Off-site transfer for disposal
to landfill °: 7.68
8
hours/day
Dow
Chemical
Company,
Dalton GA
Stack air3: 21.3
Off-site transfer for disposal
to landfill °: 109
Off-site transfer for
incinerationd: 23.1
Stack air3: 21.3
Off-site transfer for
disposal to landfill °: 109
Off-site transfer for
incineration d: 23.1
Stack air3: 1.33
Off-site transfer for disposal
to landfill °: 6.80
Off-site transfer for
incineration d: 1.45
8
hours/day
3 These stack air releases were reported under Section 5.2 of the TRI Form R, which correspond to on-site stack or point
air emissions.
bThis incineration quantity was reported under Section 6.2 of the TRI Form R, which corresponds to code M50, which is
off-site transfer for incineration/thermal treatment.
Page 122 of 723
-------
Site
identity
2017 TRI
Hours of
Release
per Day
(hr/day)
Annual Quantities per Site
(kg/year)
Daily Release (kg/site-day)
Assuming low-end of 1
day/year
Assuming high-end of 16
days/year
0 This landfill quantity was reported under Section 6.2 of the TRI Form R, which corresponds to code M64, which is off-
site transfer for disposal to other landfills.
d This incineration quantity was reported under Section 6.2 of the TRI Form R, which corresponds to code M56, which is
off-site transfer for energy recovery. EPA assumes this is to incineration.
2.2.6 Processing to Produce EPS Foam from Imported EPS Resin Beads
To manufacture EPS, EPS beads are first pre-expanded by heating with steam, which causes the beads to
soften and expand to the desired density, as the temperature of the steam exceeds that of the blowing
agent (such as pentane) incorporated in the beads (NICNAS 2012b; ECHA 2008b). Once pre-expansion
is completed, the beads are dried, then placed in shape or block molds. In the molds, the pressure is
dropped with a vacuum pump, eliminating air and water and causing the expanded beads to fuse and
take the shape of the mold (NICNAS 2012b). The EPS foam is then removed from the molds and
cooled.
The shapes or blocks may be cut into smaller sizes and trimmings may be recycled back into the foam
production process (i.e., secondary processing) (ECHA 2008b). The EPS foam is then wrapped for
transport and shipped either to customers who may further process the foam into SIPs or automobile
replacement parts or directly to end users for installation in structures such as buildings and cars. HBCD
content in the EPS foam is typically from 0.5 to 0.7 weight percent, with the usual content being 0.7
weight percent (ECHA 2017c: NICNAS 2012b: ECHA 2009b: Thomsen et al. 2007).
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.1, EPA estimates environmental releases based on a processing volume of
100,000 pounds per site per year and estimates a single unidentified site for this exposure scenario.
Release Sources
Based on the process description, EPA infers that releases may occur from: dust generation during
unloading the EPS resin beads from the bags in which they were received; disposal of the bags in which
the EPS resin beads are received; and periodic cleaning of process equipment.
Foam manufacturing sites may also generate dust and scraps from cutting or trimming of EPS foam into
panels or other shapes for shipment to end users. However, both the EU and Australian risk assessments
specify that industry provided information indicating that generated dust and trimmings may be captured
and recycled back into the foam molding process, thereby reducing or eliminating waste from the cutting
and trimming process (NICNAS 2012b; ECHA 2008b). EPA does not know the extent that these
practices are used in the United States and assessed these release sources as described below.
Page 123 of 723
-------
Emission Factors
EPA used emission factors given in the 2009 OECD ESD on Plastics Additives, as summarized in Table
2-23. Per the OECD ESD, unloading of EPS resin beads is not expected to generate dust. However,
there may be residual resin in the transport containers. The OECD ESD estimates an emission factor of
1% from the disposal of transport containers, which the OECD ESD indicates are disposed of as solid
waste to landfills. Although there is no statistical characterization of this emission factor, EPA believes
the 1% emission factor is in the upper end of the distribution based on EPA's experience. The OECD
ESD indicates that the converting process may result in dust generation at a loss rate of 0.1 to 0.5%,
which is initially released to air, with particles eventually settling and being disposed of as solid waste or
discharged as wastewater (OECD 2009). Per the EPA/OPPTSolids Transfer Dust Loss Model, dust
releases are similarly estimated with a 0.5% emission factor and initial release to air with subsequent
treatment via incineration, disposal to landfill, or discharge as wastewater from wiping and cleaning of
surfaces onto which particles have settled ( Ilia). The method of release, disposal, treatment,
or discharge is dependent on any pollution controls that are implemented at that site, as well as other
factors such as the equipment used and size of the site. EPA did not find information about the
prevalence of dust capture and control technologies at importation sites in the United States. EPA
estimated dust releases with a range of release from 0.1 to 0.5%. The method of release, disposal,
treatment, or discharge may be some or all of the following: stack air, fugitive air, onsite wastewater,
POTW, landfill, or incineration, per the OECD ESD and EPA/OPPT model.
The OECD ESD identifies trimming of produced foam as a release source, estimating a release of 2.5%
to solid waste or water from grinding or machining of the foam. EPA also identified foam trimming
release of 1% to solid waste for closed-cell spray polyurethane foam (SPF). These data were reported by
industry for the development of the draft generic scenario on SPF application (U.S. EPA. 2018d). While
this foam is different than that in this exposure scenario, EPA uses this emission factor of 1% to present
a range of potential releases from the trimming of foam. EPA assessed this release via disposal to
landfill or treatment via incineration, as the foam scraps are likely disposed of as solid waste (U.S. EPA.
2018d; OECD 2009). The method of release, disposal, treatment, or discharge is dependent on any
pollution controls that are implemented at that site, as well as other factors such as the equipment used
and size of the site. EPA did not find information on waste handling procedures at these sites. HBCD
may be disposed of to landfill and/or treated via incineration.
Based on the process description for this exposure scenario, EPA expects that equipment cleaning may
be another source of release. EPA estimated this release using the OECD ESD, which estimates an
emission factor of 1% for all other operations than previously discussed, which EPA assumes includes
equipment cleaning (OECD 2009). In addition, the EPA/OPPT Solid Residuals in Transport Containers
Model also estimates a loss of 1% of processed material. Although there is no statistical characterization
of this emission factor, EPA believes the 1% emission factor is in the upper end of the distribution based
on EPA's experience. The method of release, disposal, treatment, or discharge is dependent on any
pollution controls that are implemented at that site, as well as other factors such as the equipment used
and size of the site. EPA did not identify information on waste handling procedures at these sites. The
method of release, disposal, treatment, or discharge may include some or all of the following depending
on site-specific conditions: surface water, POTW, onsite WWT, POTW, landfill, or incineration.
Page 124 of 723
-------
Table 2-23. Summary of HBCD Releases During Manufacturing of EPS Foam from the 2009
OECD ESD on Plastics Additives and Standard EPA/OPPT Models
Release Source
Emission factor used in this Risk
Evaluation
Method of Release, Disposal,
Treatment, or Discharge
Assessed in this Risk
Evaluation a
Basis or
Source
Dust generation from
unloading EPS resin beads
from transport containers
N/A - HBCD dust generation from
unloading EPS resin beads is expected to be
minimal. Additionally, HBCD is entrained
within the polymer matrix.
(NICNAS
2012b; ECHA
2008b)
Disposal of transport
containers (bags) containing
solid HBCD residual
0.01 kg HBCD released/kg HBCD in
containers
Landfill
(OECD 2009)
Dust / volatilization releases
at elevated temperatures
during converting process
0.001-0.005 kg HBCD released/kg HBCD
processed
Uncertain: Stack air. Fugitive
Air, surface water, onsite
WWT, POTW, Landfill,
Incineration
(OECD 2009)
Equipment cleaning losses
of residual HBCD solids
from compounding
equipment
0.01 kg HBCD released/kg HBCD
processed
Uncertain-
Surface water, onsite WWT,
POTW, Landfill, Incineration
(OECD 2009)
Trimming of foama
0.01 to 0.025 kg HBCD released/kg HBCD
processed
Uncertain Incineration
Landfill
(U.S. EPA
2018d; OECD
2009)
N/A = Not applicable
a Trimmed foam may be reintroduced into the process and not disposed of based on the information in the EURAR and
Australian risk assessment (NICNAS 2012b: ECHA 2008b). EPA includes this release to present a ranee if release
estimates.
EPA's method of assessing emission factors and the methods of assessing the emission factors
pertaining to releases from the manufacture of EPS foam from EPS resin beads as reported in EURAR
and NICNAS (NICNAS. 2012b; ECHA 2008b) are similar because in all cases emission factors were
obtained from an OECD ESD or other similar method. The EURAR and NICNAS only assessed dust
releases during the converting process, and did not assess releases from unloading, disposal of transport
containers and equipment cleaning. Accordingly, EPA's overall emission factor is considerably greater
than the emission factors used in these assessments, and EPA's assessment may be conservative.
Number of Release Days
EPA estimated the number of release days based on information given in the European Communities
Technical Guidance Document (ECB 2003) and in the Australian risk assessment. EPA estimated 16
release days per year using the European Communities Technical Guidance Document for industrial use
in the polymers industry and a processing volume of 100,000 pounds HBCD/year (45.4 metric tons),
The Australian risk assessment includes one estimate of the number of operational days per year at an
EPS foam production plant. This plant reports producing EPS products containing HBCD 8 to 10 times
per year, with each production lasting up to 14 days. This results in production for 112 to 140 days per
year. In conclusion, EPA estimated a range of 16 to 140 days/year.
Page 125 of 723
-------
The data sources used to estimate releases in this section are listed in Table 2-24 along with the data
quality score. See Appendix D for more details about data source evaluation.
Table 2-24. Manufacturing of EPS Foam from Imported EPS Resin Beads Release Data Source
Evaluation
Source Reference
Data Type
Value
Overall Confidence Rating
of Data
(NICNAS 2012b)
Release Days
112 to 140 days/year for all
releases
High
(ECB 2003)
Release Days
16 days/year for all releases
Medium
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-25 below.
Table 2-25. Input Variables to Equation 2-1 for EPS Foam Manufacturing from EPS Resin Beads
Input Variable
V
(of HBCD)
Ns
(sites)
f
(kg HBCD released/kg HBCD processed)
Nd
(days/yr)
Lower value of emission factors
Upper value of emission factors
100,000
pounds/year =
45,359 kg/year
1
0.01 to landfill
0.001 to stack air, fugitive air, surface
water, onsite WWT, POTW, landfill,
and/or incineration
0.01 to surface water, onsite WWT,
POTW, landfill, and/or incineration
0.001 to incineration and/or landfill
0.01 to landfill
0.005 to stack air, fugitive air, surface
water, onsite WWT, POTW, landfill,
and/or incineration
0.01 to surface water, onsite WWT,
POTW, landfill, and/or incineration
0.025 to incineration and/or landfill
16-140
a CDR rcDortina threshold for small manufacturers (U.S. EPA 2016b)
The daily amount of HBCD released per site from EPS foam manufacturing from EPS resin beads was
calculated with Equation 2-1. The results of these calculations are summarized in Table 2-26.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed range of daily release rates presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
As detailed in Table 2-24, the result of EPA's systematic review is data pertaining to the number of
release days with an overall confidence rating of high or medium, which is a strength of the assessment.
EPA did not find release data in TRI or the EURAR that are applicable to this exposure scenario. EPA
estimated releases at EPS foam production sites using emission factors from the 2009 OECD ESD on
Plastic Additives (OECD 2009). the draft generic scenario on SPF application (U.S. EPA 2018d). and an
EPA/OPPT model available in ChemSTEER (U.S. EPA 2013a). The higher emission factor in the ESD
for dust releases corresponds to the same factor used in the EPA/OPPT Solids Transfer Dust Loss
Page 126 of 723
-------
Model, which is based on U.S. release data ( £013a). Additionally, the emission factor from
the draft generic scenario on SPF application ( I) is based on industry input. The
representativeness of these data toward the true distribution of environmental releases for this use is
uncertain and EPA notes that those from the ESD and EPA/OPPT model are likely on the higher end of
the distribution. There is uncertainty in the estimate of the range of release days that is based on industry
data that are included in the Australian risk assessment (NICNAS 2012b). The data from the Australian
risk assessment is not correlated to an HBCD throughput, so EPA could not adjust the number of days
by the assessed production volume (i.e., 100,000 pounds HBCD/year). Based on the strengths and
uncertainties of the assessment, EPA has medium confidence in the assessment results.
Page 127 of 723
-------
Table 2-26. Summary of HBCD Releases from EPS Foam Manufacturing from EPS Resin Beads
Release
Source
Method of
Release, Disposal,
Treatment, or
dischargea
Releases calculated from lower value of range of
emission factors b
Releases calculated from upper value of
range of emission factors b
Number of
Sites
Hours of
Release per
Day (hr/day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release Per
Site
(kg/site-yr)
Daily Release (kg/site-day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Daily Release (kg/site-day)
Number of
release days:
16 days/year
Number of
release days:
140 days/year
Per Site
(kg/site-
yr)
Number of
release days:
16 days/year
Number of
release days:
140
days/year
Dust release
during
converting
process
May go to one or
more: Stack air.
Fugitive Air,
surface water,
onsite WWT,
POTW, Landfill,
or Incineration
45.4
45.4
2.83
0.324
227
227
14.17
1.62
1
8 hours/day
Equipment
cleaning
May go to one or
more: surface
water, onsite
WWT, POTW,
landfill, or
Incineration
454
454
28.3
3.24
454
454
28.3
3.24
1
8 hours/day
Disposal of
transport
containers
Landfill
454
454
28.3
3.24
454
454
28.3
3.24
1
8 hours/day
Trimming
foam scrap
May go to one or
more: Incineration
or landfill
454
454
28.35
3.24
1134
1134
70.87
8.10
1
8 hours/day
a The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
b Release estimates are quantities of HBCD. The physical form of these releases is solid mixtures containing polystyrene and HBCD.
Page 128 of 723
-------
2.2.7 Processing to Produce SIPs and Automobile Replacement Parts from XPS/EPS
Foam
After XPS and EPS foam is produced, the foam may be subsequently sent to specialty fabricators to
produce structural insulated panels (SIPs) or automobile replacement parts.
To manufacture SIPs, the XPS and EPS foam is cut into the desired size panel, either with saws or
thermal wires (NICNAS 2012b). The panels are then adhered to steel, plastic, concrete, plasterboard, or
other sheathing material on either side, forming a sandwich, which is why these panels are also referred
to as sandwich panels (NICNAS 2.012b). Once the SIPs are produced, they are shipped to construction
sites for installation.
Major automobile manufacturers have phased out use of HBCD in U.S. production but continue to use it
in replacement parts, according to information provided by the Alliance of Automobile Manufacturers
(Alliance of Automobile Manufacturers 2018b; Reee 2017; Tatman 2017). Manufacturers identified 155
replacement parts containing HBCD: these include absorbers and two types of insulator panels (Tatman
2.017). For the purpose of this Risk Evaluation , EPA assumes that EPS and XPS foam containing
HBCD is used in these replacement parts (U.S. EPA. 2018f. g).
EPA did not identify specific information regarding the process for manufacturing of automobile parts
containing XPS or EPS foam. EPA believes this process likely involves the molding and cutting of parts,
similar to the manufacturing of panels and boards for construction purposes. Additionally, this process
may include the bonding of the insulation with metal or plastic surfaces. After fabrication, the
automobile replacement parts containing foam are likely shipped to automobile assemblers who install
the parts without further cutting, shaping, or other handling of the parts.
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.1, EPA estimates environmental releases based on a processing volume of
100,000 pounds per site per year. This processing volume is for any one site, and this section covers two
exposure scenarios, Manufacturing of SIPs and Automobile Replacement Parts, so EPA developed
estimates for two modeled sites, one that processes EPS and XPS foam to produce SIPs and one that
processes XPS and EPS foam to produce automobile replacement parts, with 100,000 pounds
HBCD/year at each site.
Release Sources
Based on the process description, EPA infers that releases likely occur at SIPs and automobile
replacement part manufacturing shops from the cutting of EPS and XPS foam to produce parts of
specific dimensions. Specifically, release would occur during the formation of dust during the
fabrication process and from the disposal of foam scraps. Once the parts are fabricated and shipped to
end-users, they are not likely to be further processed or handled in such a way that subsequent release
would occur. EPA estimated releases during this exposure scenario from the cutting or sawing of foam
and the subsequent disposal of foam scraps.
Emission Factors
The emission factor for particles generated by cutting XPS and EPS foam are presented in Table 2-27
(ECHA. 2008b). The method of release, disposal, treatment, or discharge for generated particles
containing HBCD during sawing and cutting is dependent on any pollution controls that are
Page 129 of 723
-------
implemented at that site, as well as other factors such as the equipment used and size of the site. EPA
did not identify information on waste handling procedures at these sites. The method of release,
disposal, treatment, or discharge may include some or all of the following depending on site-specific
conditions: stack air, fugitive air, surface water, POTW, onsite WWT, landfill, and/or incineration.
EPA used the same emission factors for the trimming of XPS and EPS foam that were used in Section
2.2.6 for the manufacturing of EPS foam from EPS resin beads. Specifically, EPA uses a range of loss
fractions of 1 to 2.5% of foam containing HBCD to estimate disposal of foam scrap to landfill or
treatment via incineration, depending on the site's disposal practices. EPA did not identify information
on waste handling procedures at these sites. Part or all of this release could be disposed of to landfill or
treated via incineration. Refer to Section 2.2.6 for additional information on this release.
The emission factors for the manufacture of SIPs and automobile replacement parts are given in Table
2-27. Summary of HBCD Release Sources During the Manufacturing of SIPs and Automobile
Replacement Parts from XPS/EPS Foam
Table 2-27. Summary of HBCD Release Sources During the Manufacturing of SIPs and
Automobile Replacement Parts from XPS/EPS Foam
Release Source
Emission factor used in this Risk
Evaluation (kg HBCD released/kg HBCD
processed)
Method of Release,
Disposal,
Treatment, or
Discharge Assessed
in this Risk
Evaluation
Basis or Source
Lower value of
emission factors
Upper value of
emission factors
Dust generation from
thermal cutting or
sawing of 10% of
XPS (50%) and EPS
(50%) boards
5.06E-05
2.25E-04
Uncertain: Stack air,
Fugitive Air, surface
water, onsite WWT,
POTW, Landfill,
and/or Incineration
(ECHA 2008b)
Trimming disposal
0.01
0.025
Uncertain:
Incineration and/or
landfilla
(OECD 2009) (lower
fraction); (U.S. EPA
2018d)(unner
fraction)
a EPA assumed solid trimming waste disposal is to incineration and/or landfill.
Number of Release Days
EPA estimated range of emission days per year based on the European Communities Technical
Guidance Document for industrial use in the polymer industry (ECB 2003). Specifically, EPA
determined a range of potential emission days by calculating the lowest and highest possible emission
days from the applicable defaults for industrial use in the polymer industry. With this method and the
HBCD processing volume for each exposure scenario (100,000 pounds [45.4 metric tons]), EPA
estimated 16 days/year. The highest number of emission days for industrial use in the polymer industry
is 300 days/year. Based on these values, EPA estimated a range of 16 to 300 emission days/year.
The data sources used to estimate releases in this section are listed in Table 2-28. Manufacturing of SIPs
and Automobile Replacement Parts from XPS/EPS Foam Release Data Source Evaluation
along with the data quality score. See Appendix D for more details about data source evaluation.
Page 130 of 723
-------
Table 2-28. Manufacturing of SIPs and Automobile Replacement Parts from XPS/EPS Foam
Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence
Rating of Data
(ECHA 2008b)
Particle Generation
Factor
See Table 2-31. Particle
Generation Factors
Reported in the EURAR
for Sawing or Cutting of
XPS/EPS Foam Prior to
Installation
High
(ECB 2003)
Release Days
16 to 300 days/year for
all releases
Medium
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-29.
Table 2-29. Input Variables to Equation 2-1 for the Manufacturing of SIPs and Automobile
Replacement Parts from XPS/EPS Foam
Input Variable
V
Ns
F
(kg HBCD released/kg HBCD processed)
Nd
(of HBCD)
(sites)
Lower value of emission
factors
Upper value of emission
factors
(days/yr)
200,000 pounds/year =
90,718 kg/yeara
2
(1 for SIPs
and 1 for
auto parts)
5.06E-05 to Stack air. Fugitive
Air, surface water, onsite WWT,
POTW, landfill, and/or
incineration
0.01 to landfill and/or
incineration
2.25E-04 to Stack air. Fugitive
Air, surface water, onsite
WWT, POTW, landfill, and/or
incineration
0.025 to landfill and/or
incineration
16-300
a CDR reporting threshold volume for small manufacturers were used for each exposure scenario.
The daily amount of solid HBCD released per site from cutting of XPS and EPS foam to manufacture
SIPs and automobile replacement parts was calculated with Equation 2-1. The results of these
calculations are summarized in Table 2-30.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium to high confidence in the assessed range of daily release rates presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
As detailed in Table 2-28, the result of EPA's systematic review is data with an overall confidence
rating of high or medium, which is a strength of the assessment. In particular, the overall confidence
rating of the data pertaining to the number of release days is medium.
Page 131 of 723
-------
The strength of the assessment approach is the estimation of HBCD emission factors and number of
release days as ranges of values to account for variability in the values of these two parameters that EPA
obtained or estimated. Furthermore, the strength of the assessment approach is the estimation of the
daily release of HBCD per site as a range of values which encompasses the range of emission factors
and the number of release days that EPA obtained or estimated.
There is uncertainty about the extent to which the emission factor data and the data on number of release
days are applicable to the HBCD use activities that would occur in the U.S. Based on the strengths and
uncertainty of the assessment, EPA has medium to high confidence in the assessment results.
Page 132 of 723
-------
Table 2-30. Summary of HBCD Releases from the Manufacturing of SIPs and Automobile Replacement Parts from XPS/EPS Foam
Release Source
Method of Release,
Disposal, Treatment, or
Dischargea
Releases calculated from lower value of range of
emission factors b
Releases calculated from upper value of
range of emission factors b
Number
of Sites
Hours of
Release per
Day
(hr/day)
Total
Annual
Release for
All Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release (kg/site-
day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release
(kg/site-day)
Number of
release days:
16 days/year
Number of
release days:
300
days/year
Number of
release
days: 16
days/year
Number of
release
days: 300
days/year
Dust release
during sawing /
cutting of foam
May go to one or more:
Stack air. Fugitive Air,
surface water, onsite
WWT, POTW, Landfill,
or Incineration
4.59
2.29
0.143
7.64E-03
20.4
10.21
0.638
3.40E-02
2
8 hours/day
Trimming foam
scrap
May go to one or more:
Incineration or landfill
907
454
28.3
1.512
2268
1134
70.9
3.78
2
8 hours/day
3 The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
3 Release estimates are quantities of HBCD. The physical form of these releases is solid mixtures containing polystyrene and HBCD.
Page 133 of 723
-------
2.2.8 Use: Installation of Automobile Replacement Parts
EPA did not identify specific process information regarding the installation of automobile replacement
parts containing HBCD. Manufacturers identified 155 replacement parts containing HBCD, these
include absorbers and insulator panels (Alliance of Automobile Manufacturers 2018b). For the purpose
of this Risk Evaluation, based on CDR reporting that showed the vast majority of use of HBCD was for
XPS and EPS, EPA assumes that HBCD in these replacement parts is incorporated into XPS and EPS
foam and that the XPS and EPS foam containing HBCD is used to make the replacement parts.
EPA estimated releases and exposures for the manufacturing of automobile replacement parts from XPS
and EPS foam in Section 2.2.7. Once manufactured, the foam automobile replacement parts are shipped
to automobile assemblers who likely install the parts without further cutting, shaping, or other handling
of the parts. The installation of automobile replacement parts is likely to involve removal of old parts
and insertion of the replacement parts within the vehicle, which EPA does not expect to generate dusts
or other sources of release. Thus, EPA does not expect releases or exposures will occur at automobile
repair sites.
2.2.9 Use: Installation of XPS/EPS Foam Insulation in Residential, Public, and
Commercial Buildings, and Other Structures
Fabricated SIPs or XPS and EPS foam from XPS and EPS foam manufacturing sites are installed at
construction sites for continuous insulation applications such as in walls and roofs on the exterior of
buildings, ceilings and sub floor systems insulation (ECHA. 2008b). Specifically, these materials are used
for insulation within the walls of buildings, as exterior sheathing, and in ceilings, roofs, and subfloors
(NICNAS 2012b). The building and construction industry use XPS and EPS foam thermal insulation
boards and laminates for sheathing products. EPS foam prevents freezing, provides a stable fill material
and creates high-strength composites in construction applications (U.S. EPA. 2018f). XPS foam board is
used mainly for roofing applications and architectural molding. HBCD is used in both types of foams
because it is highly effective at levels less than 1% and maintains the insulation properties of XPS and
EPS foam (Morose 2006).
During installation of the SIPs and XPS and EPS foam that was not previously formed into SIPs, these
materials may be cut or sawed at the construction site to fit into the building structure. Cutting is likely
to be done manually but may be done with thermal wires at large construction sites (EC )8b). The
EURAR assumes that one in every 10 foam boards is cut at construction sites (i.e., 10%). Due to lack of
additional information, EPA estimated releases and exposures from the cutting of 10% of the amount of
HBCD used for construction purposes.
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.1, EPA evaluated this exposure scenario assuming an import volume of
100,000 pounds/year (45,359 kg/year) ( ). EPA does not estimate releases and exposures
for one site for this exposure scenario, as EPA expects this exposure scenario is more widespread. EPA
calculates a range of 34 to 2,696 construction sites for this exposure scenario based on 100,000
pounds/year import volume, as described below.
The Chemical Safety Report on HBCD prepared by the European Chemicals Agency (ECHA) assesses
XPS and EPS foam use rate at a large construction site as approximately 2,440 m3 of foam (ECHA.
2017b). which equates to an applied surface area of 40,733 nr based on an insulation thickness of 0.06
Page 134 of 723
-------
meters (ECHA 2008b). With this use volume, and assuming an average foam density of 40 kg/m3 based
on the average of XPS density (35 kg/m3) and EPS density (45 kg/m3), and an HBCD content of
approximately 1.35 wt% based on the average of HBCD concentration in XPS (2 wt%) and EPS (0.7
wt%) (ECHA. 2.008b). this results in a use rate by a professional contractor of 1,320 kg HBCD/job site.
EPA assumed this HBCD use rate at large construction sites based on ECHA data is representative of
large construction sites in the United States and uses this use rate for this Risk Evaluation. With this use
rate of 1,320 kg HBCD/job site and a total construction use volume of 100,000 pounds/year (45,359
kg/year), EPA calculates 34 sites. EPA used 34 sites as the lower value in a range of the number of
potential affected construction sites.
EPA also calculated the number of potential smaller residential construction sites by assuming a floor
surface area of 2,169 ft2 from U.S. Census Bureau data
(https://www.census.gov/const/C25 Ann/sftotalmedavesqft.pdf). EPA calculated the total applied surface
area to be 519 m2 and the total volume of insulation to be 31.2 m3, assuming a square house with one
layer of insulation on three 10-foot tall stories (including basement and two above ground stories) and a
foam thickness of 0.06 meters (EC ?08b). Using the same density and HBCD concentration as
described above, EPA calculated a use rate of 16.82 kg HBCD/job site. With this use rate of 16.82 kg
HBCD/job site and a total construction use volume of 45,359 kg/year, EPA calculates 2,696 sites. EPA
uses 2,696 sites as the upper value in a range of the number of potential affected construction sites. EPA
provides an estimated range of construction sites depending on the use of HBCD-containing XPS and
EPS foam between commercial and residential sites.
Release Sources
Based on the process description, EPA infers that there are releases from sawing or thermal cutting of
XPS or EPS foam and disposal of trimmings at construction sites. EPA does not expect dust generation
during travel and unloading of the foam slabs at the construction sites (OECD 2009).
Emission Factors
The quantities of particles generated by cutting XPS and EPS foam were measured and are presented in
Table 2-31 (ECHA. 2008b). These data pertain to the methods of cutting of foam in the construction
industry which are cutting with mechanical saws in the case XPS and EPS, and thermal cutting with hot
wires or cutting with a knife and breaking in the case of EPS only. EPA estimated a particle generation
factor for the thermal cutting with hot wires or cutting with a knife and breaking of XPS as described in
Table 2-31.
The proportions of HBCD used for XPS and EPS are similar (ECHA. 2009b). EPA assumes 50 percent
of the HBCD processing volume is used to produce XPS and 50 percent is used to produce EPS. EPA
calculated weighted emission factors for cutting and sawing of foam containing HBCD from the particle
generation factors for XPS and EPS foams given in Table 2-31 and these shares of HBCD used in XPS
and EPS. The calculated emission factors are given in Table 2-32.
Page 135 of 723
-------
Table 2-31. Particle Generation Factors Reported in the EURAR for Sawing or Cutting of
XPS/EPS Foam Prior to Installation
Foam Type
Method of Cutting
Particle Generation Factor a
XPS boards
Sawing
5.0 g of XPS particles /metric ton XPS used b
XPS boards
Cutting with a knife and
then breaking or hot
wire cutting
1.12 g of XPS particles/metric ton XPS used c
EPS boards
Sawing
445 g of EPS particles/metric ton EPS used b
EPS boards
Cutting with a knife and
then breaking or hot
wire cutting
100 g of EPS particles/metric ton EPS used b
a Quantity of particles generated per quantity of foam used assuming that only a tenth of the quantity used is cut and boards
are 6 cm x 60 cm x 125 or 104 cm, and the boards are cut along the short side.
b Measured values as reported in the EU RAR.
0 Calculated by EPA using the same ratio as that for EPS foam. Particle generation factor for cutting = 5.0 g XPS
particles/metric ton XPS sawed x (100 g EPS particles/metric ton EPS cut ^ 445 g EPS particles/metric ton EPS sawed) =
1.12 g XPS particles/metric ton XPS cut.
EU RAR estimated thar one half of the generated particles are released to water while the other half are
released to air. EPA assumes that all generated particles are released to air. EPA expects that
construction sites are not likely to implement dust controls that would result in releases to stack air. EPA
expects that dust releases are initially to fugitive air, with the possibility that the particles may settle and
be discharged in wastewater to surface water or sewers (which lead to either surface water or POTWs).
EPA does not expect that these dust releases will end up in landfills or be incinerated.
In addition to dust release, there may be release from disposal of scrap foam from cutting or sawing of
the foam boards EPA uses the same emission factor for trimming of foam as described in Section 2.2.7.
Table 2-32. Summary of HBCD Release Sources During Installation of XPS/EPS Foam Insulation
in Residential, Public, and Commercial Buildings, and Other Structures
Release Source
Emission factor used in this Risk
Evaluation
(kg HBCD released/kg HBCD processed)
Method of Release,
Disposal,
Treatment, or
Discharge Assessed
in this Risk
Evaluation
Basis or Source
Lower value of
emission factors
Upper value of
emission factors
Dust generation from
thermal cutting or
sawing of XPS 10% of
(50%) and EPS (50%)
boards
5.06E-05
2.25E-04
Uncertain: Fugitive
Air, surface water,
and/or POTW
(ECHA 2008b)
Trimming disposal
0.01
0.025
Uncertain:
Incineration and/or
landfilla
(OECD 2009)
(lower fraction);
(U.S. EPA
2018d)(upper
fraction)
Page 136 of 723
-------
Release Source
Emission factor used in this Risk
Evaluation
(kg HBCD released/kg HBCD processed)
Method of Release,
Disposal,
Treatment, or
Discharge Assessed
in this Risk
Evaluation
Basis or Source
Lower value of
emission factors
Upper value of
emission factors
a EPA assumed solid trimming waste disposal is to incineration and/or landfill.
Number of Release Days
Based on the Draft Application of Spray Polyurethane Foam (SPF) generic Scenario (U.S. EPA 2018d).
EPA estimated that workers install insulation over one day per residential job site and three days for
commercial job sites. These estimates are based on the length of time for application of foam, the size of
the building in which foam is installed, and judgment on additional time needed for set-up, tear-down,
and maintenance activities at the job site.
The data sources used to estimate releases in this section are listed in Table 2-33 along with the data
quality score. See Appendix D for more details about data source evaluation.
Table 2-33. Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial
buildings, and Other Structures Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence Rating
of Data
(ECHA 2008b)
Particle Generation Factor
See Table 2-31. Particle
Generation Factors
Reported in the EURAR
for Sawing or Cutting of
XPS/EPS Foam Prior to
Installation
High
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-34.
Table 2-34. Input Variables to Equation 2-1 for the Installation of XPS/EPS Foam Insulation in
Residential, Public and Commercial Buildings, and Other Structures
Input Variable
V
(of HBCD)
Ns
(sites)
f
(kg HBCD released/kg HBCD processed)
Nd
(days/yr)
Lower value
(Commercia
1 sites)
Upper value
(Residential
sites)
Lower value of
emission factors
(residential)
Upper value of
emission factors
(commercial)
100,000
pounds/year =
45,359 kg/year
(with 10% of
boards assumed to
be cut)
34
2,696
5.06E-05 to Fugitive Air,
surface water, and/or
POTW
0.01 to landfill and/or
incineration
2.25E-04 to Fugitive
Air; surface water,
and/or POTW
0.025 to landfill and/or
incineration
1 (residential) to 3
(commercial sites)
Page 137 of 723
-------
The daily amount of solid HBCD released per site from cutting of XPS and EPS foam at construction
sites was calculated with Equation 2-1. The results of these calculations are summarized in Table 2-35.
EPA presents the lower and upper values of the range of release estimates calculated from varying the
emission factors (lower and upper emission factors), number of sites (residential and commercial), and
number of days per year (one day/year for residential sites and 3 days/year for commercial sites).
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium to high confidence in the assessed range of daily release rates that are presented above.
EPA considered the quality of the data, the assessment approach, and uncertainties in assessment results
to determine the level of confidence.
As shown in Table 2-33, EPA used emission factor data from the EURAR with an overall confidence
rating of high, which is a strength of the assessment..
The strength of the assessment approach is the estimation of HBCD emission factors, amount of HBCD
per construction site and number of release days as ranges of values to account for variability in the
values of these parameters that EPA obtained. Furthermore, the strength of the assessment approach is
the estimation of the daily release of HBCD per site as a range of values which encompasses the range
of parameters that EPA obtained.
The uncertainty of the assessment is the extent to which the emission factor data and the data on number
of release days are applicable to the HBCD use activities that would occur in the U.S. Based on the
strength and uncertainty of the assessment, EPA has medium to high confidence in the assessment
results.
Page 138 of 723
-------
Table 2-35. Summary of HBCD Releases from Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial
buildings, and Other Structures
Release Source
Method of Release,
Disposal, Treatment,
or Discharge a
Releases calculated from lower value of range of emission
factors b
Releases calculated from upper value of range of
emission factors b
Hours of
Release per
Day (hr/day)
Total Annual
Release for
All Sites
(kg/yr)
Annual
Release Per
Site
(kg/site-yr)
Daily
Release
(kg/site-
day)
Days of
Release
(day/year)
Number
of Sites
Total
Annual
Release for
All Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily
Release
(kg/site-
day)
Days of
Release
(day/year)
Number
of Sites
Dust release
during sawing /
cutting of foam
May go to one or
more: Fugitive Air,
surface water, or
POTW
2.3
8.5E-04
8.5E-04
1
2,696
10.2
0.30
0.10
3
34
8 hours/day
Trimming foam
scrap
May go to one or
more: Incineration or
landfill
454
0.168
0.168
1
2,696
1134
33.4
11.1
3
34
8 hours/day
3 The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
3 Release estimates are quantities of HBCD. The physical form of these releases is solid mixtures containing polystyrene and HBCD.
Page 139 of 723
-------
2.2,10 Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential,
Public and Commercial Buildings, and Other Structures
XPS and EPS foam insulation products are removed from buildings through demolition or remodeling of
buildings. The demolition may be accomplished with many methods, including the use of explosives, a
wrecking ball, or manual deconstruction (EC 08b). EPA expects the demolition process is likely to
involve the breaking of XPS and EPS foam insulation products into smaller pieces for subsequent
recycling or disposal at construction and demolition waste landfills or waste to energy facilities.
Environmental Release Assessment Methodology
Total Volume of HBCD in the Buildings Demolished Annually
EPA estimated this volume as a fraction of the amount of HBCD in XPS and EPS currently in use in
buildings of all types in the United States. The Environmental Health Strategy Center estimated that
about 100 million pounds of HBCD existed in use in the "built environment" (EPA interprets this to
mean in buildings of all types) in the United States as of 2010 (comment on Docket ID Number: EPA-
HQ-OPPT-2016-0735-0008, (Safer Chemicals 2017). The number of houses of all types demolished
between 2011-2013, including as a result of disaster, is equal to 0.36% of the number of houses present
in 2011 (HUD 2016). Accordingly, EPA estimates 0.18% of houses are demolished annually. Also,
more than one quarter of the buildings that existed in the year 2000 are expected to be replaced by the
year 2030 in the U.S. ( J008b). Therefore, the number of buildings demolished each year in
the U.S. on average as a fraction of the total number of buildings that existed in the year 2000 is equal to
0.83%). EPA is uncertain whether buildings of all types, including small structures such as houses, are
accounted for in the data obtained from U.S. EPA (2008b). Accordingly, EPA conservatively assessed
the number of buildings of all types demolished each year in the U.S. as a fraction of the total number of
existing buildings of all types to be equal to the sum of 0.18%> and 0.83%> or approximately 1%>.
Approximately 1.7%> of the in-service volume of HBCD in Japan is disposed of each year (Managaki et
al. 2009). but EPA did not use this data because it pertains to Japan and data pertaining to the U.S. is
available as discussed above. In conclusion, 1% of the in-service volume of HBCD in the United States
(100 million pounds) is estimated to be demolished each year. This results in one million pounds/year
(-458,000 kg/year) as the total volume of HBCD in buildings demolished annually.
Number of Demolition Sites
EPA estimated the number of demolition sites to be proportional to the number of installation sites. As
discussed in Section 2.2.9, EPA estimated a lower value of 34 commercial sites and an upper value of
2,696 residential sites for EPS or XPS foam insulation containing HBCD installed based on a processing
volume of 100,000 pounds HBCD/year. Scaling for the larger demolition volume of one million pounds
HBCD/year, EPA estimated a lower value of 343 commercial sites or an upper value of 27,230
residential sites with HBCD-containing insulation are demolished each year. The following is a sample
calculation:
Low-end number of demolition sites = 34 installation sites X (1 million lbs of HBCD /100,000
lb/yr of HCBD) = 343 sites.
Release Sources
During demolition, releases are likely to occur from the generation of XPS and EPS particles resulting
from the breaking of XPS and EPS insulation boards.
Page 140 of 723
-------
Emission Factors
XPS and EPS particle generation factors for cutting and/or manually breaking XPS and EPS boards are
reported in the EU RAR or estimated by EPA. EPA estimated emission factors for releases from
demolition as a range of values based on these various particle generation factors to account for various
demolition methods as discussed below.
The quantities of particles generated by manually breaking XPS and EPS foam were measured and are
presented in Table 2-36. Particle Generation Factors for the Demolition of XPS and EPS
(ECHA 2008b). These factors were used in the EU RAR to assess releases from manual deconstruction
of XPS and EPS boards for the purposes of recycling. For material demolished for disposal instead of
recycling, the emission factor reported was 0.1% kg of HBCD released per kg of HBCD in EPS and
XPS that is demolished (ECHA 2008b). EPA rated this emission factor as unacceptable with regard to
systematic review overall confidence because the EU RAR did not include a reference for this value. To
assess releases from demolition by means other than manual deconstruction, EPA assumed particle
generation factors in the case of such demolition are equivalent to the particle generation factors for
cutting with a knife and manually breaking that EPA used to assess releases from construction in the
U.S. as discussed in Section 2.2.9. These particle generation factors pertain to cutting XPS and EPS with
a hot wire or with a knife and manually breaking the boards, and are presented in Table 2-31. Particle
Generation Factors Reported in the EURAR for Sawing or Cutting of XPS/EPS Foam Prior to
Installation
The values given in Table 2-31 are based on the assumption that only 10% of XPS and EPS boards are
sawed or cut. In contrast, EPA assumed that every board is affected during demolition and therefore
multiplied these particle generation factors by 10. The adjusted particle generation factors are given in
Table 2-36. Particle Generation Factors for the Demolition of XPS and EPS
Table 2-36. Particle Generation Factors for the Demolition of XPS and EPS
Method of Cutting
Type of Foam
Particle Generation Factor
Manual breaking
XPS boards
0 g of XPS particles/metric ton EPS
broken a
EPS boards
90 g of EPS particles/metric ton EPS
broken a
Cutting with a knife
and then manual
breaking
XPS boards
11.2 g of EPS particles /metric ton XPS
cut and broken b
EPS boards
1000 g of XPS particles/metric ton XPS
cut and broken b
a Measured values that are used in the EU RAR to assess releases from manual deconstruction of XPS and EPS
boards for the purpose of recycling.
b These values were determined by multiplying the corresponding values in Table 2-31. Particle Generation
Factors Reported in the EURAR for Sawing or Cutting of XPS/EPS Foam Prior to Installation
by 10 to account for the breaking of every board.
EPA used a weighted average of the XPS and EPS particle generation factors pertaining to manual
breaking to calculate an emission factor for demolition by manual deconstruction. EPA also used a
Page 141 of 723
-------
weighted average of the XPS and EPS particle generation factors pertaining to cutting with a knife and
manual breaking to calculate an emission factor for demolition by means other than manual
deconstruction. EPA assumed the share of HBCD used in either XPS or EPS is 50% to calculate the
weighted averages, and the rationale for this assumption is given in Section 2.2.9. These calculated
emission factors are presented in Table 2-38. EPA assessed the emission factor for demolition as a range
of values with these emission factors as the lower- and higher-end of this range.
Number of Release Days and Media of Release
EPA assumed that demolition at any site occurs during a single day and therefore releases occur during a
single day. The size of the generated foam particles is not reported in the EU RAR and EPA assumed
that all generated particles are sufficiently small to be emitted to ambient air initially. Dust controls at
demolition sites are unlikely and EPA expects that dust generated during demolition is released to
ambient air and may subsequently settle and be released in wastewater, surface water or sewers (which
lead to either surface water or POTWs). EPA does not expect that these dust releases will end up in
landfills or be incinerated.
The data sources used to estimate releases in this section are listed in Table 2-37 along with the data
quality score. See Appendix D for more details about data source evaluation.
Table 2-37. Demolition of XPS/EPS Foam Insulation in Residential, Public and Commercial
buildings, and Other Structures Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence
Rating of Data
(HUD 2016)
Fraction of houses of all
types demolished
0.18%
High
(U.S. EPA 2008b)
Fraction of all buildings
demolished
0.83%
High
(ECHA 2008b)
Particle Generation Factor
See Table 2-36
High
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-38.
Table 2-38. Summary of HBCD Releases from Demolition of XPS/EPS Foam Insulation in
Residential, Public and Commercial Buildings, and Other Structures
Input Variable
V
(of HBCD)
Ns
(sites)
F
(kg HBCD released/kg HBCD processed)
Nd
(days/yr)
Lower value
(Commercia
1 sites)
Upper value
(Residential
sites)
Lower value of
emission factors
Upper value of
emission factors
1 million
pounds/year =
458,128 kg/year
343
27,230
4.50E-05 to Fugitive Air,
surface water, and/or
POTW
5.06E-04 to Fugitive
Air; surface water,
and/or POTW
1
Page 142 of 723
-------
The amount of HBCD released from demolition was calculated with Equation 2-1. The results of these
calculations are summarized in Table 2-39.
Table 2-39. Summary of HBCD Releases from Demolition of XPS/EPS Foam Insulation in
Residential, Public and Commercial Buildings, and Other Structures
Release Source
Method of
Release,
Disposal,
Treatment, or
Discharge a
Releases calculated from lower value of range
of emission factors b
Releases calculated from upper value of
range of emission factors b
Hours of
Release per
Day
(hr/day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-yr)
Daily
Release
(kg/site-
day)
Days of
Release
(day/year)
Number
of Sites
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily
Release
(kg/site-
day)
Days of
Release
(day/year)
Number
of Sites
Generation of
foam particles
during
demolition
May go to one
or more:
Fugitive Air,
surface water,
orPOTW
20.6
7.57E-04
7.57E-04
1
27,230
232
0.675
0.675
1
343
8
hours/day
3 The method of release, disposal treatment, or discharge may include some or all of those listed depending on site-specific
conditions, including waste handling practices.
3 Release estimates are quantities of HBCD in particles of XPS and EPS.
Disposal of HBCD That is Part of Construction and Demolition Waste
Approximately 64 to 70% of construction and demolition (C&D) waste in the United States is disposed
of in landfills and the remaining 30 to 36% is processed for reuse, recycling, or energy recovery (e.g., at
waste energy recovery incinerators) (Townsend et al. 2019; U.S. EPA 2018; Tceq 2017). The C&D
waste that is disposed of in landfills is sent mainly to C&D landfills, but a portion is sent to municipal
solid waste landfills (U.S. EPA 1998; U.S. EPA 2003). The EPA Incident Waste Decision Support Tool
(I-Waste DST) estimated that there were 1,577 C&D landfills in the United States in 2015 (U.S. EPA
2015c) and the Waste Business Journal estimated that there were 1,120 C&D landfills in the United
States in 2019 (Waste Business Journal 2019). There have historically been between 75 and 97 waste-
to-energy facilities in the United States between 2001 and 2018 (Energy Recovery 2018) and there were
108 waste-to-energy facilities in the United States in 2019 (Waste Business Journal 2019).
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed range of daily release rates that are presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
EPA implemented this approach using emission factor data from the EURAR and estimated volume
using HUD (2016) and U.S. EPA (2008b). The data from these sources both have overall confidence
ratings of high, which is a strength of the assessment.
The strength of the assessment approach is the estimation of HBCD emission factors and amount of
HBCD per demolition site as ranges of values to account for variability in the values of these two
parameters that EPA obtained.
The uncertainty of the assessment is the extent to which the emission factor data and the data on
demolition rate are applicable to the HBCD use activities that would occur in the U.S. In particular, the
particle generation for demolition is expected to vary depending on the destructive method of
demolition. There is uncertainty with the use of cutting of XPS/EPS foam particle generation factor as a
Page 143 of 723
-------
surrogate for the higher value emission factor for dust generation during demolition activities. Based on
the strength and uncertainties of the assessment, EPA has medium confidence in the assessment results.
2,2,11 Processing: Recycling of EPS Foam and Reuse of XPS Foam
Schlummer et al. (2.017) reported that XPS and EPS foam in construction insulation materials are rarely
recycled for numerous reasons, including that insulation waste is typically not separated from mixed
waste stream and most insulation containing HBCD is still in place.
To recycle EPS foam, the EPS boards are grinded, melted, and introduced into the EPS molding process
with virgin EPS (ECHA 2008b). Thus, EPS recycling is likely to occur at sites with similar operations to
those described for EPS foam manufacturing in Section 2.2.6. XPS insulation may be reused but is
rarely recycled due to the specialized equipment needed to do so ( E018D. Reuse of XPS may
involve the cutting of the XPS insulation into different sizes, as needed. Based on reasonably available
information, as discussed in the 2018 HBCD Problem Formulation Document, EPA assessed the reuse
of XPS, but not the recycling of XPS ( 18g).
Environmental Release Assessment Methodology
Facility Estimates
EPA identified two companies in the 2018 HBCD Problem Formulation Document that directly reuse
(e.g., reuse without reforming) and recycle (e.g., melting and inserting into the manufacturing process)
XPS and EPS foam insulation (l_ I ^ \ _018g). One of these companies indicated that they recycle
EPS roofing material at a rate of 10,000 pounds/year of EPS and reuse XPS roofing material at an
unknown rate (but does not recycle it due the special equipment needed to recycle XPS). Details on the
operations of the other recycling / reuse company were not provided (U.S. EPA. 2018f). but EPA expects
this company may perform both recycling and reuse of XPS and EPS foam.
EPA estimated releases for two EPS recycling and XPS reuse sites (one site) per company identified in
the 2015 HBCD Problem Formulation document ( )15a) for this exposure scenario) and uses
the same known throughput (10,000 pounds of EPS insulation recycled per year) for both sites. EPA did
not identify data to characterize the statistical representativeness of this assessment. With a typical
HBCD concentration of 0.7 weight percent in EPS insulation (ECHA 11 , INEOS Styrenics 201. ;
115a; ECH.A 2009a. 2008b; Thorn sen et al. 2007). each company processes 70 pounds
HBCD/year in EPS insulation (31.8 kg HBCD/site-year, or 63.5 kg HBCD/year for both sites).
One of the above companies estimates that 10-20% of EPS roofing material is recycled nationally (U.S.
EPA. 2018g). thus the number of sites that perform EPS recycling in the United States is likely greater
than the two sites.
Release Sources
Based on the process description, EPA infers that releases for recycling of EPS foam for this exposure
scenario are similar to those for Manufacturing of EPS Foam from Imported EPS Resin Beads, as
described in Section 2.2.6, with the removal of the trimming release, as EPA does not expect that there
will be waste disposal due to trimming at a EPS recycling site.
Emission Factors
EPA expects that EPS foam is likely to be transported in trucks or other bulk containers for this
exposure scenario, as opposed to the transport of EPS resin beads in bags for the Manufacturing of EPS
Page 144 of 723
-------
Foam from Imported EPS Resin Beads. For this exposure scenario, EPA estimates releases from the
cleaning of bulk containers used to transport the EPS foam to the converting site. The method of release,
disposal, treatment, or discharge is dependent on any pollution controls that are implemented at that site,
as well as other factors such as the equipment used and size of the site. The method of release, disposal,
treatment, or discharge may include some or all of the following depending on site-specific conditions:
surface water, POTW, onsite WWT, POTW, landfill, or incineration.
EPA additionally estimated releases from dust and equipment cleaning residue in accordance with the
methodology described in Section 2.2.6 for the Manufacturing of EPS Foam from Imported EPS Resin
Beads.
Number of Release Days
Using the European Communities Technical Guidance Document for industrial use in the polymers
industry and a processing volume of 140 pounds HBCD/year (<1 metric ton), EPA estimated 1 day of
emission per year (ECB 2003). Based on these data, EPA used a lower bounding estimate of one
day/year, as the number of emission days cannot be lower than this estimate. Because EPS recycling
may occurs at similar sites as EPS foam manufacturing from EPS resin, EPA uses the same upper value
of the range of days determined in Section 2.2.6, which is 140 days/year, which accounts for variability
in the number of days a recycling facility may process HBCD containing EPS foam.
The data sources used to estimate releases in this section are listed in Table 2-40 along with the data
quality score. See Appendix D for more details about data source evaluation.
Table 2-40. Recycling of EPS Foam Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence
Rating of Data
CNICNAS 2012b)
Release Days
140 days/year for all
releases
High
(ECB 2003)
Release Days
1 day/year for all
releases
Medium
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-41. Input
Variables to Equation 2-1 for the Recycling of EPS Foam
Table 2-41. Input Variables to Equation 2-1 for the Recycling of EPS Foam
Input Variable
V
Ns
f
(kg HBCD released/kg HBCD processed)
Nd
(of HBCD)
(sites)
Lower value of emission factors
Upper value of emission factors
(days/yr)
20,000 pounds of EPS
foam/year =140 pounds
HBCD/yr (0.7% HBCD in
foam) = 63.5 kg HBCD/year
2
Container cleaning: 0.01 to
uncertain (could go to surface
water, onsite WWT, POTW,
landfill, and/or incineration)
Container cleaning: 0.01 to
uncertain (could go to surface
water, onsite WWT, POTW,
landfill, and/or incineration)
1-140
Page 145 of 723
-------
Equipment cleaning: 0.01 to
uncertain (could go to surface
water, onsite WWT/POTW,
landfill, and/or incineration)
Dust: 0.001 to uncertain (could go
to stack air, fugitive air, surface
water, onsite WWT, POTW,
landfill, and/or incineration)
Equipment cleaning: 0.01 to
uncertain (could go to surface
water, onsite WWT/POTW,
landfill, and/or incineration)
Dust: 0.005 to uncertain (could go
to stack air, fugitive air, surface
water, onsite WWT, POTW,
landfill, and/or incineration)
The amount of solid HBCD released annually was calculated with Equation 2-1 by multiplying the
processing volume of HBCD by the emission factors. The daily amount of HBCD released from
recycling was calculated by dividing this annual release by the number of days of emission. The results
of these calculations are summarized in Table 2-42.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed range of daily release rates that are presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
EPA used emission factor data from the 2009 OECD ESD on Plastic Additives and other EPA/OPPT
models. The emission factor data were not evaluated because these data were obtained from an ESD or
GS. EPA used data on the number of release days from the European Communities Technical Guidance
Document (ECB 2003) and Australian risk assessment (NICNAS 2012b). The data from the technical
guidance document has an overall confidence rating of medium and the data from the Australian risk
assessment has an overall confidence rating of high; these ratings were assigned using EPA's systematic
review process, as discussed in Section 1.5.
The strength of the assessment approach is the estimation of HBCD emission factors and number of
release days as ranges of values to account for variability in the values of these two parameters that EPA
obtained. Furthermore, the strength of the assessment approach is the estimation of the daily release of
HBCD per site as a range of values which encompasses the range of emission factors and the number of
release days that EPA obtained.
The uncertainty of the assessment is the extent to which the emission factor data and the data on number
of release days are applicable to the HBCD recycling activities that would occur in the U.S. Based on
the strength and uncertainty of the assessment, EPA has medium confidence in the assessment results.
Page 146 of 723
-------
Table 2-42. Summary of HBCD Releases from the Recycling of EPS Foam
Release
Source
Method of
Release,
Disposal,
Treatment, or
dischargea
Releases calculated from lower value of range of
emission factors b
Releases calculated from upper value of range
of emission factors b
Number
of Sites
Hours of
Release per
Day
(hr/day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release (kg/site-
day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release
(kg/site-day)
Number
of release
days: 1
day/year
Number of
release
days: 140
days/year
Number
of release
days: 1
day/year
Number of
release
days: 140
days/year
Dust
release
from
grinding of
foam
May go to one or
more: Stack air.
Fugitive Air,
surface water,
onsite WWT,
POTW, Landfill,
or Incineration
6.35E-02
3.18E-02
3.18E-02
2.27E-04
0.318
0.159
0.159
1.13E-03
2
8 hours/day
Container
cleaning
residual
May go to one or
more: surface
water, onsite
WWT, POTW,
Landfill, or
Incineration
1.270
0.635
0.635
4.54E-03
1.27
0.635
0.635
4.54E-03
2
8 hours/day
a The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment
use, size of the site, and waste handling practices, including any pollution controls used.
b Release estimates are quantities of HBCD. The physical form of these releases is solid mixtures containing polystyrene and HBCD.
Page 147 of 723
-------
2,2.12 Formulation of Flux/Solder Pastes
EPA identified from the TRI data one site that processed HBCD as a formulation component. As
discussed in Section 1.2, communication with this company indicates that this site formulates HBCD
into flux/solder pastes. The TRI data does not specify the physical form of HBCD that is processed as a
formulation component. Based on the process description below, EPA expects HBCD powder is likely
used for this exposure scenario. This exposure scenario represents only the incorporation of HBCD into
formulations of soldering materials.
In communication with EPA, the flux and solder paste formulation company explained that flux/solder
paste components are processed in the U.S. and sent to China for final formulation and sale. The final
solder flux formulations containing HBCD are sold to both international and U.S. customers who use the
formulations primarily for electronics, such as circuit boards.
Incorporation into a formulation, mixture, or reaction product refers to the process of mixing or blending
several raw materials to obtain a single product or preparation (OECD 2010b). First, the components of
the product formulation are unloaded from transport containers, either directly into the mixing
equipment or into an intermediate storage vessel (OECD 2010b). Transfer from transport containers may
be manual or automated using a pumping system. An automated dispenser may be used to feed
components into the mixing vessel to ensure that precise amounts are added at the proper time during
the mixing process. Once in the mixing vessel, the components are then mixed in either a batch or
continuous system. Depending on the specific product, the formulation may be further processed
through filtering. Once the formulation is completed, it is sampled for quality control. The final
formulation is then filled into containers, either through manual dispensing from transfer lines or
through an automatic system. Automatic filling systems are generally used for the filling of smaller
containers that are intended for consumer and commercial applications, whereas manual filling is done
for larger containers (e.g., tank trucks, totes, drums) which are typically used in an industrial setting
(OECD 2010b).
Environmental Release Assessment Methodology
Facility Estimates
EPA expects that the amount of HBCD used in flux/solder paste is significantly less than the amount
used for insulation in buildings, as these uses were not reported by the former manufacturers and
importers of HBCD to the 2016 CDR. Use in EPS and XPS foam has accounted for 95 percent of all
HBCD applications in the past decade (U.S. EPA. 2014d; UNEP 2010a). Due to lack of additional
information, for the purposes of this Risk Evaluation, EPA estimated that the remaining five percent of
HBCD applications are in solder flux formulations. With an importation volume equal to the CDR
threshold of 100,000 pounds/year and 5 percent, EPA used a throughput of 5,000 pounds HBCD/year
(2,268 kg/year) to estimate releases and exposures for this exposure scenario. Indium reported in 2017 to
TRI that the maximum amount of HBCD on-site at any one point during the calendar year was between
1000 to 9,999 lbs. Indium increased the reported maximum amount of HBCD on-site to 10,000 to
99,000 lbs, but with overall reduced releases than 2017 TRI. Therefore EPA assessed the exposure
scenario using 2017 TRI data. EPA assessed one solder formulation site based on TRI data (U.S. EPA.
2} ).
Release Sources
Based on the process description, EPA infers releases may occur from dust generation during the
transfer of HBCD powder from transport containers into blending vessels, residual HBCD in the
Page 148 of 723
-------
emptied transport containers from the direct disposal of the emptied containers, and the periodic
cleaning of blending equipment.
Emission Factors
EPA estimated releases from this exposure scenario using release information reported by the
solder/flux formulation site to the 2017 TRI. As indicated by the 2018 TRI data given in Section 1.2.4,
the releases from this site during 2018 are much lower and therefore EPA assessed releases
conservatively.
Number of Release Days
EPA estimated a range of emission days per year based on the European Communities Technical
Guidance Document for formulation in the electronics industry, as the flux/solder formulations in this
exposure scenario are used for electronics applications (ECB 2003). Specifically, EPA determined a
range of potential emission days by calculating the lowest and highest possible emission days from the
applicable defaults for formulation within the electronics industry. With this method and the HBCD
processing volume for this exposure scenario (5,000 pounds or 2.25 metric tons), EPA estimated 5
days/year. The highest number of emission days for formulation within the electronics industry is 300
days/year. Based on this, EPA estimated a range of 5 to 300 emission days/year.
The data sources used to estimate releases in this section are listed in Table 2-43 along with the data
quality score. See Appendix D for more details about data source evaluation.
Table 2-43. Formulation of Flux/Solder Pastes Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence
Rating of Data
(U.S. EPA 2017a)
Site-Specific Release
Quantities
See Table 2-44. Summary
of HBCD Releases from
Flux/Solder Paste
Formulation Sites from
2017 TRI Data
Medium
(ECB 2003)
Release Days
5 to 300 days/year for all
releases
Medium
Environmental Release Assessment Results
The releases, as they were reported to 2017 TRI, are summarized in Table 2-44. Summary of HBCD
Releases from Flux/Solder Paste Formulation Sites from 2017 TRI Data
The flux/solder paste formulation site reports off-site transfers to a waste broker for disposal (disposal as
defined at 40 CFR 372.3 is "any underground injection, placement in landfills/surface impoundments,
land treatment, or other intentional land disposal") and for treatment via solidification/stabilization (EPA
assumes this disposal is to landfill).
Page 149 of 723
-------
Table 2-44. Summary of HBCD Releases from Flux/Solder Paste Formulation Sites from 2017 TRI
Data
Site
identity
Exposure
scenario
2017 TRI
Hours of
Release
per Day
(hr/day)
Annual Release Per
Site
(kg/site-yr)
Daily Release (kg/site-day)3
Over 5 day/year
Over 300 day/year
INDIUM
CORP OF
AMERICA,
Clinton,
NY
Formulation
of Solder
Fugitive aira: 0.454
Stack airb: 6.350
Unknown disposal
0.454
Off-site landfilld:
6.350
Fugitive aira: 0.091
Stack air b: 1.27
Unknown disposal
0.091
Off-site landfilld:
1.27
Fugitive aira: 0.0015
Stack airb: 0.021
Unknown disposal 0.0015
Off-site landfilld: 0.021
8 hours/day
a These fugitive air releases were reported under Section 5.1 of the TRI Form R, which correspond to on-site fugitive or
non-point air emissions.
b These stack air releases were reported under Section 5.2 of the TRI Form R, which correspond to on-site stack or point
air emissions.
0 This unknown disposal quantity was reported under Section 6.2 of the TRI Form R, which corresponds to code M94,
which is off-site transfer to waste broker for disposal. Disposal (as defined at 40 CFR 372.3) is 'any underground
injection, placement in landfills/surface impoundments, land treatment, or other intentional land disposal'.
d This off-site landfill quantity was reported under Section 6.2 of the TRI Form R, which corresponds to code M40, which
is off-site transfer for treatment via solidification/stabilization. No additional details were provided. EPA assumes the
final method of disposal is landfill.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed range of daily release rates that are presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
EPA used release data from 2017 TRI data, which has an overall confidence rating of medium, assigned
using EPA's systematic review process, as discussed in Section 1.5. EPA used data on number of release
days from the European Communities Technical Guidance Document (ECB_2003), which has an overall
confidence rating of medium.
The strength of the assessment approach is the estimation of number of release days and daily release of
HBCD as ranges of values to account for potential variability in the release days associated with the
annual release amounts.
The uncertainty of the assessment is the extent to which the annual release data is reflective of the full
distribution of release rates and the extent to which the data on number of release days are applicable to
the HBCD processing activities that would occur in the U.S. Based on the strengths and uncertainty of
the assessment, EPA has medium confidence in the assessment results.
2.2.13 Use of Flux/Solder Pastes
As described in Section 1.2.5.3, EPA identified that HBCD is used specifically in solder/flux pastes that
are used in electronics manufacturing. The solder/flux paste formulator indicated that the final
formulations are used both overseas for electronics manufacturing and domestically. EPA did not find
information on the fraction of the solder/flux pastes that are used domestically. EPA assumes that the
entire amount is used in the United States. Additionally, for the purpose of this Risk Evaluation, EPA
assumes that they are used similarly as they are used overseas, specifically in electronics manufacturing.
Page 150 of 723
-------
Within the electronics industry, solder/flux pastes are used to attach components to printed circuit
boards. EPA expects that the use of solder in other industries involve similar release sources and
quantities as those assessed in this Risk Evaluation.
Solder pastes are comprised of solder, which is a metal alloy, predominantly tin mixed with other metals
such as lead and silver, suspended within flux pastes that typically contains rosin, wetting agents,
viscosity modifiers, and other fluxing aids (OECD 2010a). Soldering is a process in which two or more
substrates, or parts (usually metal), are joined together by melting solder paste into the joint and
allowing it to cool, thereby joining the independent parts. Solder paste is first applied in the area
between the substrates to be joined, then heat is applied to the solder paste, which causes the solder to
melt and join the two substrates together once cooled. The solder has a lower melting point than the
adjoining metal substrates, allowing it to be melted during the soldering process without melting the
substrates. The function of flux within the solder paste is to prevent oxidation during the soldering
process, which ensures that soldered joints are secure (OECD 2010a). Soldering differs from welding in
that soldering does not involve melting the substrates being joined.
Solder paste can be applied to metal substrates with a variety of methods. The website of the site that
processes HBCD as a formulation component, identified from TRI, depicts solder paste formulations as
syringe/bead applied to circuits to be soldered. Based on this information, EPA expects the use of
syringe application on circuit boards during this exposure scenario.
Solder pastes are largely made up of metal solder (at least 90 percent), flux (around 5 percent), with the
remainder as solvent and other additives (these specialty chemicals are generally less than one percent of
the composition of the solder paste) (OECD 2010a). HBCD serves as a fluxing aid within solder/flux
paste formulations.
Environmental Release Assessment Methodology
Facility Estimates
As discussed in Section 2.2.12, EPA estimated a throughput of 5,000 pounds HBCD/year (2,268
kg/year) for the formulation of solder flux. EPA uses this same HBCD volume for this exposure
scenario. EPA estimated that the entire throughput is used in the United States, as the portion that is used
internationally is unknown, as discussed above.
EPA uses the OECD ESD on Chemicals Used in the Electronics Industry (OECD 2010a). To calculate
the number of solder use sites as described below. Since the OECD ESD estimates other additives are
generally less than one percent of the composition of the solder paste, EPA used an HBCD composition
of one weight percent for this exposure scenario.
The OECD ESD includes default annual facility use rates for non-aqueous (paste) solder paste
formulations of less than 1,000 kg/site-year for small scale use sites and greater than 1,000 kg/site-year
for large scale use sites. To calculate the number of sites for this exposure scenario, EPA uses a
throughput of 1,000 kg solder formulation/site-year. The number of sites is equal to the HBCD use
volume (2,268 kg/year), divided by the solder paste formulation use rate (1,000 kg/site-year) and HBCD
content in the formulation (0.01). This calculation results in 227 sites.
Page 151 of 723
-------
Release Sources
Based on information in the OECD ESD, EPA infers that releases may occur from: disposal of
containers used to ship the flux/solder paste formulations containing HBCD, cleaning of soldering
equipment and soldered components, and overapplied solder (OECD 2010a).
EPA estimated releases from this exposure scenario using the 2010 OECD ESD on Chemicals used in
the Electronics Industry (OECD 2010a). as the formulator of the solder and flux pastes containing
HBCD indicates that these formulations are used for circuits and other electrical components. Table
2-45 summarizes the release sources assessed by EPA. The methodology used for this assessment is
explained below.
Emission Factors
The OECD ESD on Chemicals Used in the Electronics Industry indicates that the total loss from use of
flux and solder in the electronics industry is typically 10 percent (OECD 2010a). The OECD ESD
specifies that releases contributing to this overall loss may include washing of equipment used for
soldering, washing of components that have been soldered, and from disposal of unused solder by either
solvent washings that occur throughout the electronics manufacturing process or disposal of scrap
components containing solder formulations.
While the OECD ESD does not specifically call out releases from disposal of containers used to ship the
flux and solder paste formulations, EPA expects this release is a part of the total 10 percent loss
estimated by the OECD ESD. The website of the flux and solder formulator identified in TRI indicates
that these formulations are frequently supplied in small containers, such as syringes, from which
application onto substrates may be conducted directly from the containers, without unloading into
separate application equipment. EPA expects that these containers are most likely disposed of as solid
waste to landfill or treated via incineration, as opposed to being cleaned (which may result in liquid
wastes). Thus, EPA estimated release from container residual disposed of to landfill or treated via
incineration, using the EPA/OPPT Small Container Residual Model, which indicates a loss of 0.6
percent from residue inside containers ( 313a). The method of release, disposal, treatment, or
discharge is dependent on any pollution controls that are implemented at that site, as well as other
factors such as the equipment used and size of the site. EPA did not find information on waste handling
procedures at these sites. The method of release, disposal, treatment, or discharge may include disposal
to landfill, treatment via incineration, or both.
The OECD ESD on Chemicals Used in the Electronics Industry indicates that release may occur from
cleaning of equipment or components (such as solder equipment, which is distinguished from
application equipment) (OECD 2010a). The OECD ESD estimates that this release is up to 2 percent of
the use volume discharged in wastewater to on-site WWT or POTW.
The final release that is defined in the OECD ESD is loss of unused flux and solder paste formulations.
This may occur when unused formulation on soldered components {i.e., overapplied solder) is washed
off components in some of the solvent washings that are customary in the electronics manufacturing
process (OE ). This release may also occur from the disposal of scrap components that have
been soldered or that contain unused flux and solder formulation. While the OECD ESD does not
specify an exact loss percentage for this release, it does estimate a total loss of 10 percent, which EPA
used to determine this release fraction by subtracting the upstream losses of container disposal (0.6%)
and equipment cleaning (1 to 2%). Thus, EPA estimated a loss of 7.4 to 8.4 percent for this release. The
OECD ESD indicates that generated process solvents are disposed of as hazardous waste (which EPA
Page 152 of 723
-------
assumes includes incineration or hazardous waste landfill disposal) and that scrap components are
disposed of as solid waste. Thus, EPA assessed disposal to landfill or treatment via incineration. The
method of release, disposal, treatment, or discharge is dependent on any pollution controls that are
implemented at that site, as well as other factors such as the equipment used and size of the site. EPA
did not identify information on waste handling procedures at these sites. The method of release,
disposal, treatment, or discharge may include disposal to landfill, treatment via incineration, or both.
The total loss from this exposure scenario is 10% per the OECD ESD, with variation in the amount of
release for each method of release, disposal, treatment, or discharge (wastewater, landfill, or
incineration).
Table 2-45. Summary of HBCD Release Sources
Juring Use of Flux and Solder Pastes
Release Source
Emission Factor used in
this Risk Evaluation
Method of Release,
Disposal, Treatment, or
Discharge Assessed in this
Risk Evaluation
Basis or Source
Disposal of used transport
container containing solid
HBCD residuals
0.006 kg HBCD released/kg
HBCD in containers
Uncertain: landfill,
incineration
Due to the small container
size (syringes), EPA assumes
containers are disposed of
from the sites as solid waste
to either landfill or
incineration
EPA/OPPT Small Container
Residual Model (U.S. EPA
2013a)
Equipment Cleaning release
of solid HBCD residuals
0.01 to 0.02 kg HBCD
released/kg HBCD used
100% to Onsite WWT/
POTW
(OECD 2010a). - The OECD
ESD indicates that up to 2% of
total releases may be to
wastewater from cleaning of
equipment or components.
Unused flux remaining on
components, which are likely
removed in subsequent
solvent washes
0.084 to 0.074 (10% minus
upstream losses, see above)
kg HBCD released/kg HBCD
used
Uncertain: landfill,
incineration
Solvent washings treated as
hazardous waste. EPA
assessed to incineration or
landfill.
(OECD 2010a).-Perthe
OECD ESD a total of 10%
loss is expected; accounting
for upstream losses, this loss is
7.4%
Number of Release Days
EPA estimated a range of emission days per year based on the European Communities Technical
Guidance Document for use in the electronics industry, as the solder formulations in this exposure
scenario are used for electronics applications (ECB 2003). Specifically, EPA determined a range of
potential emission days by calculating the lowest and highest possible emission days from the applicable
defaults for use within the electronics industry. With this method and the HBCD processing volume for
this exposure scenario (5,000 pounds or 2.25 metric tons), EPA estimated 4 days/year. The highest
number of emission days for use within the electronics industry is 300 days/year. Based on these values,
EPA estimated a range of 4 to 300 emission days/year.
The data sources used to estimate releases in this section are listed in Table 2-46 along with the data
quality score. See Appendix D for more details about data source evaluation.
Page 153 of 723
-------
Table 2-46. Use of Flux and Solder Pastes Release Data Source Evaluation
Source Reference
Data Type
Value
Overall Confidence
Rating of Data
(ECB 2003)
Release Days
4 to 300 days/year for
all releases
Medium
Environmental Release Assessment Results
The variables used for calculating releases with Equation 2-1 are summarized in Table 2-47
Table 2-47. Input Variables to Equation 2-1 for Use of Flux and Solder Pastes
Input Variable
V
(kg HBCD
imported/yr)
Ns
(sites)
f
(kg HBCD released/kg HBCD used)
Nd
(days/yr)
Lower values of emission factors
Upper values of emission factors
5,000 pounds/yr
= 2,268 kg/yr'
227
0.09 to landfill and/or incineration
0.01 to Onsite WWT and/or POTW
0.08 to landfill and/or incineration
0.02 to Onsite WWT and/or POTW
4-300
The amount of solid HBCD released from use of flux and solder pastes was calculated with Equation
2-1. The results of these calculations are summarized in Table 2-48. The use of flux and solder pastes
results in releases to wastewater, municipal landfill, and incineration. The largest source of release is
from unused formulations that are disposed of to landfill or incineration.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed range of daily release rates that are presented above. EPA
considered the quality of the data, assessment approach, and uncertainties in assessment results to
determine the level of confidence.
EPA used emission factor data from the 2010 OECD ESD on Chemicals Used in the Electronics
Industry. The quality of the emission factor data was not evaluated because this data was obtained from
an ESD. EPA used data on number of release days from the European Communities Technical Guidance
Document (ECB_2003), which has an overall confidence rating of medium, assigned using EPA's
systematic review process, as discussed in Section 1.5.
The strength of the assessment approach is the estimation of HBCD emission factors and number of
release days as ranges of values to account for variability in the values of these two parameters that EPA
obtained. Furthermore, the strength of the assessment approach is the estimation of the daily release of
HBCD per site as a range of values which encompasses the range of emission factors and the number of
release days that EPA obtained.
The uncertainty of the assessment is the extent to which the emission factor data and the data on number
of release days are applicable to the HBCD use activities that would occur in the U.S. Based on the
strength and uncertainty of the assessment, EPA has medium confidence in the assessment results.
Page 154 of 723
-------
Table 2-48. Summary of HBCD Releases from Use of Flux and Solder Pastes
Release Source
Method of
Release,
Disposal,
Treatment, or
Dischargea
Higher landfill and incineration releases b
Higher onsite wastewater, POTW releases b
Number
of Sites
Hours of
Release per
Day (hr/day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release (kg/site-
day)
Total
Annual
Release
for All
Sites
(kg/yr)
Annual
Release
Per Site
(kg/site-
yr)
Daily Release (kg/site-
day)
Number of
release
days: 4
day/year
Number of
release
days: 300
day/year
Number of
release
days: 4
day/year
Number of
release
days: 300
day/year
Equipment cleaning release
of solid HBCD residuals
May go to one
or more: Onsite
WWTorPOTW
22.7
0.100
2.50E-02
3.33E-04
45.4
0.200
5.00E-02
6.66E-04
227
8 hours/day
Disposal of transport
containers containing solid
HBCD residual and
overapplied/unused solder
May go to one
or more:
Incineration or
landfill
204
0.899
2.25E-01
3.00E-03
181
0.799
0.200
2.66E-03
227
8 hours/day
a The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
b Release estimates are quantities of HBCD. The physical form of these releases is solid or paste mixtures containing HBCD and other solder / flux formulation
components.
Page 155 of 723
-------
2.2.14 Recycling of Electronics Waste (E-Waste) Containing HIPS
HBCD was used in the production of HIPS, which can be found in television sets, computers, phones,
and other electronic products (Morf et al. 2005). EPA estimated HBCD releases from e-waste recycling
sites to be equal to the following values as discussed below:
• central tendency estimate: 0.008 to 0.024 kg/day-site;
• high-end estimate: 0.12 to 0.38 kg/day-site.
EPA is uncertain of the media of release and hence assesses these rates as rates of releases to air or
landfill or incineration or to some combination of these methods of release, disposal, or treatment
methods.
EPA calculated the HBCD release rates in accordance with the following equations:
Equation 2-2: HBCD release rate from an e-waste recycling Site
R = X x (Vy -h Nd) x £ fj
Equation 2-3: average recycling rate of consumer electronics per e-waste recycling site
Vy = Vt -h Ns
Equation 2-4: total number of e-waste recycling sites
Ns = Nsc x a
Where:
R = the amount of HBCD released per day from an e-waste recycling site to the environment or to
disposal or treatment (kg per day per site)
X = the amount of HBCD contained in recycled consumer electronics (kg HBCD per kg of recycled
electronics)
Vy = annual recycling rate of consumer electronics per site (kg of recycled electronics per year per
site)
Nd = the number of HBCD release days per year from a site (days per year)
fi = emission factor for release of HBCD to the environment or to disposal or treatment from a
particular source at an e-waste recycling site (kg of HBCD released per kg of HBCD contained
in the recycled electronics)
Vt = annual recycling rate of consumer electronics in the U.S. (kg of recycled electronics per year)
Ns = the total number of e-waste recycling sites in the U.S. (sites)
Nsc = the number of certified e-waste recycling sites in the U.S. (sites)
a = the ratio of total number of e-waste recycling sites and number of certified e-waste recycling
sites
EPA calculated the central tendency and high-end HBCD releases rates from central tendency and high-
end values of the annual recycling rate of consumer electronics per site (Vy), respectively. To account
for measurement error in the values of the amount of HBCD contained in recycled consumer electronics
(X) and the values of the various emission factors for release of HBCD to the environment or to disposal
or treatment (fi), EPA calculated each of the central tendency and the high-end release rates as a range of
values. The values of the input variables of Equation 2-2, Equation 2-3, and Equation 2-4, that EPA
chose, references for these values and the overall confidence rating of these values is presented in Table
2-49.
EPA determined the values of the input variables of Equation 2-2, Equation 2-3, and Equation 2-4, as
follows:
Page 156 of 723
-------
1. The Amount of HBCD Contained In Recycled Consumer Electronics. HBCD Emission Factors.
Environmental Media of Release and Treatment and Disposal Methods:
Morf et al. (2005) prepared a mass balance of HBCD in a "modern state-of-the-art" waste electrical and
electronic equipment (WEEE) recycling facility located in Switzerland. They accomplished this by
measuring (a) the mass of WEEE fed to the facility, (b) the masses of the output streams, and (c) the
concentrations of HBCD in all relevant output streams of the facility. They calculated the mass of
HBCD per kg of the recycled WEEE on average, including the parts of the WEEE that are not flame
retarded, to be equal to 17 ± 4 mg of HBCD/kg of WEEE. The WEEE fed to the facility consisted of
"small household appliances (e.g., toasters and vacuum cleaners), office and communication appliances
(e.g., personal computers and monitors, printers, phones, and fax and photocopy machines),
entertainment electronics (e.g., television (TV) sets, videos, camcorders, radios, HiFis, and portable
compact disk (CD players), and small size electrical and electronic (E&E) equipment (e.g., plugs and
mobile phones)."
Morf et al. (2.005) reported the ratio of the mass of HBCD in an output steam to the mass of HBCD in
the WEEE feed to the facility. These mass ratios and the output steams associated with them are as
follows: the fine-grained plastic fractions (0.574 ± 18%), plastics and wooded castings (PC/TV) (0.277
± 81%), fine-grained metal fractions (0.074 ± 24%), dust collected in bag filters (0.04 ± 44%), Cu cables
(0.025 ± 45%), printed circuit boards (0.010 ± 25%) and air emitted from these bag filters (0.002%).
EPA's assessment is that the HBCD contained in the following output streams is released to the
environment or to disposal or treatment: the dust collected in bag filters, the air emitted from the bag
filters, and the fine-grained metal fractions. EPA's rationale for assessing the release, disposal or
treatment of the HBCD in the fine-grained metal fractions is that e-waste recycling in the U.S. may
include metal extraction (NIQSH. 2.014a) and this output stream contains plastic impurities (Morf et al.
2005) which may be separated and/or emitted during the processing of this output steam for the purpose
of metal extraction.
EPA's expectation is that waste streams comprising solid material in filters are disposed of in landfills or
treated via incineration. Also, there may be significant releases to the environment from e-waste
recycling processes that do not include efficient air pollution control devices (Morf et al. 2005). Hence,
EPA's assessment conservatively is that the dust collected in bag filters, and the fine-grained metal
fractions are released to air, to landfill or to incineration or to some combination of this environmental
medium or disposal or treatment methods. EPA expects that releases to water directly from e-waste
recycling sites is unlikely. At the vast majority of sites surveyed by NIOSH, e-waste is disassembled and
separated (NIOSH. 2014b). and these processes do not include aqueous process streams. For example,
the facility examined by Morf et al. (2005) does not include aqueous process streams. Cleaning of
equipment with water between batches is unlikely because contamination is not a problem. Cleaning
surfaces such as floors to remove settled dust is done by vacuuming or compressed air (NIOSH. 2014b)
or dry brushing (Rosenberg et al.: ) although wet mopping and wet brushing are superior to the use
of compressed air or dry brushing as cleaning methods for industrial hygiene reasons (NIQSH. 2014b:
Rosenberg et al. 2011).
2. Consumer Electronics Recycling Rates:
The annual recycling rate of selected consumer electronics during 2015 in the U.S. was equal to 1,230
xlO3 U.S. tons (U.S. EPA. 2019o). Selected consumer electronics "includes products such as TVs,
VCRs, DVD players, video cameras, stereo systems, telephones and computer equipment." EPA
selected the value pertaining to the year 2015 because this is largest reported value. The number of
Page 157 of 723
-------
certified e-waste recycling facilities in the U.S. was 550 during 2015 (U.S. EPA. 2016a). Currently there
are 716 certified sites and 29 non-certified sites ("e-Stev. 20; Sustainable Electronics Recycling
2020). and EPA calculated a, or the ratio of total number of e-waste recycling sites and number of
certified e-waste recycling sites, from these values. The capacity of a state-of-the-art WEEE recycling
facility in Switzerland is 30,000 metric tons per year (Morf et al. 2005) and the capacity of a state-of-
the-art e-waste recycling facility in Canada is also 30,000 metric tons per year (Toroko and Mcdonald
2013). Accordingly, EPA assumed the high-end value of the rate of recycle of consumer electronics per
site in the U.S. to be equal to 30,000 metric tons/year. EPA assumes that an e-waste recycling facility is
operates 5 days a week and is shutdown a total of two weeks during the year for maintenance and hence
estimate the number of operating days to be 250 days/year.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed release rates presented above. EPA considered the quality
of the data, the assessment approach, and uncertainties in assessment results to determine the level of
confidence.
The result of EPA's systematic review is data with an overall confidence rating of medium or high,
which is a strength of the assessment.
The strength of the assessment approach is the estimation of releases based on measurements of HBCD
concentrations in e-waste recycling output streams.
There is uncertainty in the assessed HBCD release rates because the HBCD concentration data, which
pertain to a facility in Switzerland, and the maximum annual e-waste recycling rate per site, which
pertains to Switzerland and Canada, may not represent data that pertain to e-waste recycling facilities in
the U.S.
Page 158 of 723
-------
Table 2-49. Values, References for, and Overall Confidence Ratings of Input Variables of Equations HBCD Release Rate from E-
Waste Recycling Sites
Input Variables
of Equation 2-2,
Equation 2-3,
and Equation 2-4
Values of Input Variables
Reference
Overall
Confidence
Rating
Value Chosen to Calculate the
Central Tendency Release Rate
Value Chosen to
Calculate the High-End
Release Rate
Values Chosen for Calculating the Central
Tendency and High-End Release Rates as a
Range of Values
X
17 ± 4 mg of HBCD / kg of recycled electronics
low-end of range:
21 mg HBCD/kg of recycled electronics
(Morfetal. 2005)
medium
high-end of range:
13 mg HBCD/kg of recycled electronics
Vy
This parameter was calculated from
the values for Vt and Nsc and a given
below in accordance with Equation
2-3, and Equation 2-4.
30,000 metric tons/year
not applicable
(Tomko and
Mcdonald 2013;
Morfetal. 2005)
medium, high
Nd
250 days/year
value assumed by EPA
not applicable
not applicable
f (dust in bag
filter)
0.04 ± 44% kg HBCD/kg HBCD
low-end of range:
0.0224 kg HBCD/kg HBCD
(Morfetal. 2005)
medium
high-end of range:
0.0576 kg HBCD/kg HBCD
f (air emitted from
bag filter)
0.00002 kg HB CD/kg HBCD
not applicable
f (fine grain metal
fractions)
0.074 ± 24% kg HB CD/kg HBCD
low-end of range:
0.0562 kg HBCD/kg HBCD
high-end of range:
0.0918 kg HBCD/kg HBCD
V,
1,230 xlO3 US tons in 2015
not applicable
not applicable
(U.S. EPA 2019o)
medium
Nsc
550 sites in 2015
not applicable
not applicable
(U.S. EPA 2016a)
high
a
1.04 (calculated by EPA from the
current number of certified and non-
certified sites)
not applicable
not applicable
(e-Steward 2020;
Sustainable
Electronics
Recvclins 2020)
medium
Page 159 of 723
-------
2.2.15 Sensitivity Analysis - Process Volume
In Section 2.2.2 through Section 2.2.7, EPA provided release estimates using the CDR reporting
threshold volume of 100,000 lbs/yr-site. EPA selected 100,000 lbs/yr as a conservative process volume
in an effort to account for the uncertainty in the current HBCD import volume. As discussed in Section
1.2.3, EPA determined that the previously high volume HBCD importers (as identified by the 2016
CDR) have permanently stopped importing HBCD. EPA's review of a widely used import database
(Datamyne) identified 5 companies in 2016 importing a total of 399,315 kg/yr (880,339 lbs/yr) of
HBCD, and 1 company importing 46,096 kg/yr (101,624 lbs) in 2017. The 101,624 lbs of import in
2017 were from one consignee in two equal shipments of 23,048 kgs (50,812 lbs). The import of HBCD
has been steadily declining since the United Nations Stockholm Convention on Persistent Organic
Pollutants (POPs) has caused many processors to shift to alternative flame retardants. Due to the
uncertainty with the imported volume, EPA performed a targeted sensitivity analysis of process volume
for select exposure scenarios.
EPA performed the sensitivity analyses for three exposure scenarios at process volumes per site of
50,000 lbs/yr and 25,000 lbs/yr to examine the effect of process volume on environmental releases and
the resulting general population and environmental exposures. EPA selected 50,000 lbs/yr based on the
imported volume reported in one shipment for HBCD (2017), and to account for the declining use of
HBCD, EPA also considered a lower volume of 25,000 lbs/yr. The exposure scenarios considered in the
sensitivity analysis represent the exposure scenarios that resulted in the highest estimates of releases on
a daily basis and include scenarios that rely on both industry data and OECD ESDs. As shown in
equation 2.1, the daily releases of HBCD are estimated based on four parameters: process volume(F),
number of sites (Ns), emission factor (/), and number of release days (Nd). The last parameter, number
of release days (Nd), was estimated by either using industry data, days provided in relevant ESDs/GSs or
European Communities Technical Guidance Document (ECB 2003). Depending on the source, the
selected range of release days may vary based on the expected process volume and was adjusted
accordingly. The determination of release days for each exposure scenario is discussed in their
respective sections: Section 2.2.2, Section 2.2.4, and Section 2.2.6. For all of the selected exposure
scenarios, the estimated total annual release per site decreased by the same factor as the decrease in the
process volume {i.e., annual releases based on 50,000 lbs/yr decreased by a factor of 2; annual releases
based 25,000 lbs/yr decreased by a factor of 4).
Repackaging of Import Containers
For repackaging of import containers, quantities of releases are estimated from dust emissions during the
transfer of HBCD powder from import containers into new containers and from residual HBCD in the
emptied import containers that are disposed of. The quantities of releases at the different process
volumes are presented in Table 2-50. Summary of HBCD Releases from Sensitivity Analysis of
Repackaging of Import Containers
An explanation of the emission factors for this exposure scenario are presented in Section 2.2.2. The
daily quantities of releases into the environment at different process volumes are relatively unchanged as
the range of the daily throughput volume (process volume /site- day) for this exposure scenario did not
significantly change. The lower value of the number of release days {i.e., operating days for this
exposure scenario) were estimated using B-tables from the basic chemicals industry category in the
European Communities Technical Guidance Document (ECB 2003). which calculates a number of
release days using the total import volume of the chemical substance. The changes in process volumes
Page 160 of 723
-------
adjust proportionally the number of release days, the effect was similar daily releases. EPA also deemed
that the higher value of release days, 300 days, should be adjusted to stay within a reasonable range of
daily throughputs based on the expected repackaging process and the reported daily throughput given by
a repackaging site (NICNAS 2012b).
Processing to Produce XPS Foam from XPS Masterbatch
For the manufacturing of XPS foam from XPS Masterbatch, releases are estimated from: dust generation
during unloading the HBCD powder from the bags in which they were received; disposal of the bags in
which the HBCD powder is received; and periodic cleaning of process equipment. An explanation of the
emission factors for this exposure scenario are presented in Section 2.2.4. The releases at the different
process volumes are presented in Table 2-51. The decrease in daily releases into the environment
between process volume is directly proportional to the decrease in the process volume. The release days
specified by site-specific emission data in the EURAR are used for the range of release days.
Processing to Produce EPS Foam from EPS resins
For Manufacturing of EPS foam from EPS resins, releases are estimated from dust generation during
unloading the EPS resin beads from the bags in which they were received and from the converting
process; disposal of the bags in which the EPS resin beads are received; and periodic cleaning of process
equipment. An explanation of the emission factors for this exposure scenario are presented in Section
2.2.6. The releases at the different process volumes are presented in Table 2-52. The changes in daily
release into the environment varies depending on the estimated number of release days For the lower
value of release days that were generated using the Ell TGD- Polymer Industry (ECB 2003). the
adjustment to the release days was proportional to the decrease in process volume. This resulted in little
change for the calculated daily releases at the lower value of release days. The higher value of release
days was reported by a EPS foam manufacturer (NICNAS 2012b). The process volume of the reported
site was not included, so it is uncertain if the lower process volume is applicable to the reported release
days. However, EPA believes given the small percentage of HBCD in EPS resins beads (<1%), 140 days
is still within a reasonable range of release days for EPS foam manufacturing for both 50,000 lbs/yr and
25,000 lbs/yr of HBCD.
Page 161 of 723
-------
Table 2-50. Summary of HBCD Releases from Sensitivity Analysis of Repackaging of Import Containers
Release Source
Method of Release, Disposal, Treatment, or
Dischargea
Releases calculated from lower value
of range of emission factors b
Releases calculated from upper value
of range of emission factors b
Annual
Release
Per Site
(kg/site-
yr)c
Daily Release
(kg/site-day)
Annual
Release
Per Site
(kg/site-
yr)c
Daily Release
(kg/site-day)
Lower
Number of
release days d
Upper
Number of
release dayse
Lower
Number of
release days d
Upper
Number of
release days e
Annual import volume = 100,000 pounds HBCD/year
Dust release during unloading
of HBCD
May go to one or more: Stack air. Fugitive Air, on-site
WWT, POTW, landfill, or incineration
45.4
1.56
0.15
227
7.82
0.756
Disposal of transport bags
Landfill
454
15.64
1.51
454
15.64
1.51
Annual import volume = 50,000 pounds HBCD/year
Dust release during unloading
of HBCD
May go to one or more: Stack air. Fugitive Air, on-site
WWT, POTW, landfill. Incineration
22.7
1.51
0.15
113
7.56
0.756
Disposal of transport bags
Landfill
227
15.12
1.51
227
15.12
1.51
Annual import volume = 25,000 pounds HBCD/year
Dust release during unloading
of HBCD
May go to one or more: Stack air. Fugitive Air, on-site
WWT, POTW, landfill. Incineration
11.3
1.62
0.15
57
8.10
0.756
Disposal of transport bags
Landfill
113
16.20
1.51
113
16.20
1.51
3 The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
3 Release estimates are quantities of HBCD. The physical form of these releases is solid HBCD.
: Based on the assumption of one given site.
dThe lower number of release days is 29 days/yr (100,000 lb/yr), 15 days/yr (50,000 lb/yr), 7 days/yr (25,000 lb/yr). Release days were calculated using the new process
volume usins EU TGD B-tables (ECB 2003). which reauired roundins to the nearest inteser for release davs. While the process volumes were scaled bv 2. due to
rounding, the daily releases are not directly scaled by the same factor.
3 The upper number of release days is 300 days/yr (100,000 lb/yr), 150 days/yr (50,000 lb/yr), 75 days/yr (25,000 lb/yr).
Page 162 of 723
-------
Table 2-51. Summary of HBCD Releases from Sensitivity Analysis of XPS Foam Manufacturing Using XPS Masterbatch
Release Source
Method of Release,
Disposal, Treatment, or
Dischargea
Releases calculated from lower value of
range of emission factors b
Releases calculated from upper value
of range of emission factors b
Annual
Release Per
Site
(kg/site-yr)c
Daily Release
(kg/site-day)
Annual
Release Per
Site
(kg/site-yr)c
Daily Release
(kg/site-day)
Lower
Number of
release days d
Upper
Number of
release days e
Lower
Number of
release days d
Upper
Number of
release days
e
Annual import volume = 100,000 pounds HBCD/year
Unknown - these data were reported by EU sites in
the EURAR as total annual release per site
May go to one or more:
Stack air or fugitive air
2.63
2.63
0.164
2.63
2.63
0.164
Unknown - these data were reported by EU sites in
the EURAR as total annual release per site
May go to one or more:
Surface Water, Onsite
WWT, orPOTW
0.486
0.486
3.24E-02
1.19
1.19
0.080
Annual import volume = 50,000 pounds HBCD/year
Unknown - these data were reported by EU sites in
the EURAR as total annual release per site
May go to one or more:
Stack air, fugitive air
1.31
1.31
0.082
1.31
1.31
0.082
Unknown - these data were reported by EU sites in
the EURAR as total annual release per site
May go to one or more:
Surface Water, Onsite
WWT, POTW
0.243
0.243
1.62E-02
0.60
0.60
0.040
Annual import volume = 25,000 pounds HBCD/year
Unknown - these data were reported by EU sites in
the EURAR as total annual release per site
May go to one or more:
Stack air, fugitive air
0.66
0.66
0.041
0.66
0.66
0.041
Unknown - these data were reported by EU sites in
the EURAR as total annual release per site
May go to one or more:
Surface Water, Onsite
WWT, POTW
0.121
0.121
8.10E-03
0.30
0.30
0.020
3 The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
3 Release estimates are quantities of HBCD. The physical form of these releases is solid mixtures containing polystyrene and HBCD.
: Based on the assumption of one given site.
d The lower number of release days is 1 day/year (for all releases and all annual import volumes).
3 The upper number of release days is 15 day/year (wastewater discharges) and 16 day/year (air releases) for all annual import volumes.
Page 163 of 723
-------
Table 2-52. Summary of HBCD Releases from Sensitivity Analysis of EPS Foam Manufacturing from EPS Resin Beads
Release Source
Method of Release, Disposal, Treatment, or Dischargea
Releases calculated from lower value
of range of emission factors b
Releases calculated from upper
value of range of emission factors b
Annual
Release
Per Site
(kg/site-
yr)c
Daily Release
(kg/site-day)
Annual
Release
Per Site
(kg/site-
yr)c
Daily Release
(kg/site-day)
Lower
Number of
release days d
Upper
Number of
release days e
Lower
Number of
release days d
Upper
Number of
release days e
Annual import volume = 100,000 pounds HBCD/year
Dust release during
converting process
May go to one or more: Stack air. Fugitive Air, surface water,
onsite WWT, POTW, Landfill, or Incineration
45.4
2.83
0.324
227
14.17
1.62
Equipment cleaning
May go to one or more: surface water, onsite WWT, POTW,
landfill, or Incineration
454
28.3
3.24
454
28.3
3.24
Disposal of transport
containers
Landfill
454
28.3
3.24
454
28.3
3.24
Trimming foam scrap
May go to one or more: Incineration or landfill
454
28.35
3.24
1134
70.87
8.10
Annual import volume = 50,000 pounds HBCD/year
Dust release during
converting process
May go to one or more: Stack air. Fugitive Air, surface water,
onsite WWT, POTW, Landfill, Incineration
22.7
2.83
0.162
113
14.17
0.81
Equipment cleaning
May go to one or more: surface water, onsite WWT, POTW,
landfill. Incineration
227
28.3
1.62
227
28.3
1.62
Disposal of transport
containers
Landfill
227
28.3
1.62
227
28.3
1.62
Trimming foam scrap
May go to one or more: Incineration; landfill
227
28.35
1.62
567
70.87
4.05
Annual import volume = 25,000 pounds HBCD/year
Dust release during
converting process
May go to one or more: Stack air. Fugitive Air, surface water,
onsite WWT, POTW, Landfill, Incineration
11.3
2.83
0.081
57
14.17
0.40
Equipment cleaning
May go to one or more: surface water, onsite WWT, POTW,
landfill. Incineration
113
28.3
0.81
113
28.3
0.81
Disposal of transport
containers
Landfill
113
28.3
0.81
113
28.3
0.81
Trimming foam scrap
May go to one or more: Incineration; landfill
113
28.35
0.81
283
70.87
2.02
3 The method of release, disposal, treatment, or discharge may include some or all of those listed depending on site-specific conditions, including type of equipment use,
size of the site, and waste handling practices, including any pollution controls used.
3 Release estimates are quantities of HBCD. The physical form of these releases is solid mixtures containing polystyrene and HBCD.
: Based on the assumption of one given site.
dThe lower number of release days is 16 days/yr (100,000 lb/yr), 8 days/yr (50,000 lb/yr), 4 days/yr (25,000 lb/yr).
Page 164 of 723
-------
Release Source
Method of Release, Disposal, Treatment, or Dischargea
Releases calculated from lower value
of range of emission factors b
Releases calculated from upper
value of range of emission factors b
Annual
Release
Per Site
(kg/site-
yr)c
Daily Release
(kg/site-day)
Annual
Release
Per Site
(kg/site-
yr)c
Daily Release
(kg/site-day)
Lower
Number of
release days d
Upper
Number of
release days e
Lower
Number of
release days d
Upper
Number of
release days e
3 The upper number of release days is 140 days/year (all annual import volumes).
Page 165 of 723
-------
2.2.16 Assumptions and Key Sources of Uncertainties for Environmental Releases
Processing Volume and Number of Sites
This evaluation estimates a processing volume and number of sites for each exposure scenario of HBCD
based on information provided by industry, information from literature or assumes maximum import
volume set at the CDR reporting threshold. For the exposure scenarios involving processing of HBCD
into XPS and EPS foam (discussed in Section 2.2.2 through Section 2.2.7), EPA utilizes a processing
volume of up to 100,000 pounds per year for an unknown site as discussed in Section 2.2.1. There are
uncertainties with the number of possible small firms currently importing HBCD and their import
volumes. This could lead to an overestimation of total annual releases at any given site, if HBCD is
imported, processed, or used at a lower volume. The impact of the processing volume on daily releases
can vary with site-specific variables such as the number of batches (if it's not a continuous process), the
frequency of cleaning or the number of release days also influencing daily releases rates. EPA evaluated
the exposure scenarios related to XPS and EPS foam manufacturing only at 100,000 pounds per year,
however, EPA used a range of release days and emission factors to develop a reasonable range of daily
releases to the environment.
For the use of XPS and EPS foam as insulation building materials, EPA used the total HBCD import
volume of 100,000 pounds for all sites that install XPS and EPS foam insulation (Sections 2.2.9). As
discussed above, there is uncertainty as to the number of small firms importing HBCD and their import
volumes, which leads to uncertainty in the overall volume of HBCD that may be used for XPS and EPS
foam insulation in buildings. To determine the number of sites that install XPS and EPS foam in
buildings, EPA used XPS and EPS foam properties (i.e., density, thickness, and HBCD concentration in
the foam) and assumed building sizes to calculate an HBCD throughput at each construction site, from
which the number of sites could be determined. For this HBCD throughput calculation, EPA used
averaged foam properties between XPS and EPS foam insulation. However, these properties may vary
depending on the type of insulation (i.e., interior wall, exterior wall, or roofing), which results in
uncertainty in this throughput and number of sites estimates. In addition, EPA used assumed building
sizes for residential and commercial sites to develop lower and upper estimates of HBCD throughput
and number of sites. The actual building size and associated HBCD throughput is expected to vary
widely, resulting in additional uncertainty in this estimate. The lower and upper estimates of HBCD
throughput and number of sites may underestimate and overestimate releases, respectively. However,
EPA developed these upper and lower estimates in an effort to capture the possible range of number of
sites and associated releases. For demolition and disposal of XPS/EPS foam insulation (Section 2.2.10),
EPA used the same assumptions to estimate number of demolition sites based on volume information on
the amount of HBCD in the built environment.
For the recycling of EPS foam (Section 2.2.11), EPA estimated HBCD processing volume and number
of sites based on information identified from industry in the HBCD Problem Formulation (TJ.S. EPA.
2018g). There is uncertainty in the extent to which this information captures the full number of sites that
recycle or reuse XPS/EPS building insulation containing HBCD. This could lead to underestimation of
total annual releases for all sites for this exposure scenario; however, EPA believes the estimates of
releases on a per site basis are reasonable because the HBCD processing volume per site is based on
industry data.
For the use of flux/solder pastes containing HBCD (Section 2.2.13), EPA assumed that 5% of 100,000
pounds of HBCD was used for this exposure scenario based on historical data that indicated 95% or
more of HBCD is used in building insulation. As described above, the use of 100,000 pounds is a source
of uncertainty. In addition, there is uncertainty as to whether this historical proportion is still reflective
Page 166 of 723
-------
of the current usage of HBCD in United States. Using this total HBCD volume, EPA calculated the
number of sites and processing volume at each site using the 2010 OECD ESD on Chemicals Used in
the Electronics Industries (OE ). The basis of these calculations is an assumed solder paste
throughput (and associated HBCD content) reported in the OECD ESD to distinguish small scale from
large scale sites that conduct soldering. The solder throughput and HBCD content likely vary between
sites and the use rate in the United States may differ from that reported in the OECD ESD. A major
electronics site may utilize more HBCD-containing flux/solder paste than the assumed solder paste
throughput, which could lead to an underestimation of releases at the site. The uncertainties in these
estimates may result in either underestimation or overestimation of releases on a total and per site basis.
EPA did not estimate the number of sites for the installation of automobile replacement parts (Section
2.2.8). EPA used 2017 TRI data to estimate the number of sites and associated releases for the
formulation of HBCD into solder/flux pastes (Section 2.2.12), rather than estimating these values.
Emission Factors
This report uses existing release data from 2017 TRI data, the EURAR, or modeling approaches from
relevant ESDs or GSs to estimate emission factors during each exposure scenario. For certain exposure
scenarios (Section 2.2.3 through Section 2.2.5), discrete HBCD release quantities provided in the
EURAR were used; however, the EURAR did not provide HBCD throughput {i.e., HBCD processing
volumes) for the specific sites from which emission factors could be calculated. The EURAR only
provided combined HBCD processing volumes for all the sites for which release data were available.
EPA calculated emission factors from EURAR data by dividing the total annual HBCD release
quantities for all sites by the total HBCD processing volume for all sites. There is uncertainty from using
the total HBCD release quantities and total HBCD throughput to calculate emission factors, as this does
not account for variability in the actual HBCD throughput at the site (higher or lower), which would
result in different emission factors for each site.
In some instances, EPA used the reported emission factors in the EURAR. Although EPA expects that
activities described in risk assessments performed by the EURAR are similar to those performed in the
United States, EPA could not verify these values. In particular, uncertainty arises from the geographic
origin of the release data. The data reported in the EURAR pertains to HBCD releases at sites in Europe
and the extent to which this data is applicable to HBCD releases in the U.S. is uncertain. There is also
uncertainty about the extent to which the release data in the EURAR is applicable to the evaluated
exposure scenarios in this Risk Evaluation. Despite potential differences in practices of the European
sites from which data was collected in the EURAR and sites in the United States, these data have an
overall confidence rating of High from the systematic review process.
In cases where there was no release data in the EURAR for the exposure scenario in this risk
assessment, EPA used modeling approaches from relevant ESDs or GSs, specifically the 2009 OECD
ESD on Plastic Additives, and the 2010 OECD ESD on Chemicals Used in the Electronics Industry.
While these ESDs or GSs are applicable to the industries of the exposure scenarios, they are not
necessarily specific to the use of HBCD within these industries. In some cases, OECD ESDs or GSs use
modeling approaches listed in EPA ChemSTEER User Guide (U.S. EPA. 2013a). Although there is no
statistical characterization of the emission factors from these models, EPA believes the emission factors
are in the upper end of the distribution based on EPA's experience. For dust releases in Sections 2.2.2,
2.2.6, and 2.2.11, EPA used emission factors from the 2009 OECD ESD on Plastic Additives, which
provides two discrete emission factors, one for particulates <40 |im and one for particles >40 |im. EPA
Page 167 of 723
-------
expects a distribution of particle sizes and associated emission factors but does not have these data. The
use of the two discrete emission factors from the ESD is a source of uncertainty.
Release Days
EPA estimated the number of release days using industry data from the EURAR, information from
ESDs or GSs, and from the European Communities Technical Guidance Document (ECB 2003). Where
available, EPA used the number of release days reported in the EURAR for sites with specific release
data. The EURAR did not report site-specific HBCD processing volume from which EPA could scale
these release days to account for HBCD throughput at the sites . There is uncertainty in the extent to
which the HBCD throughput and HBCD processing activities and frequency is similar to that assessed
by EPA. EPA also estimated release days using GSs and ESDs. There is uncertainty whether the GSs
and ESDs are reflective of the sites and operations that are included in this Risk Evaluation. As stated
earlier, while ESDs or GSs are applicable to the industries of the exposure scenarios, they are not
necessarily specific to the use of HBCD within these industries. EPA evaluated potential environmental
releases using a range of release days in an effort to address the uncertainty and variability in release
days.
Additionally, EPA estimated release days from the European Communities Technical Guidance
Document (ECB 2003). There is uncertainty in the applicability of this methodology for HBCD use in
the United States. However, EPA evaluated potential environmental releases using a range of release
days in an effort to address the large variability in release days.
2.3 Environmental Exposures
2.3.1 Approach and Methodology
HBCD has been detected in a wide variety of environmental and biological media, as expected based on
its environmental fate properties such as high persistence in soil, surface water, and groundwater, and its
bioaccumulation and bioconcentration tendencies. This environmental exposure assessment focuses on
HBCD concentrations in surface water, sediment, and soil, as these are the media which were evaluated
to determine risks to aquatic (pelagic and benthic) and terrestrial organisms (refer to Section 3 and
Section 4 on hazard and risk characterization, respectively). Ambient air was only assessed for its
contribution via deposition to these media. Levels in wildlife were examined, but were not brought
forward to the environmental risk estimation due to the incompatibility of the hazard and wildlife
biomonitoring data available, as will be explained in Section 4.
Releases from industrial facilities, indoor sources (building materials and dust), and long-range transport
all contribute to levels in the environment. However, source attribution and temporal trends from these
disparate sources is complex. As such, EPA used two main approaches to estimate environmental
exposures. A non-scenario specific approach was used to estimate environmental exposures based on
media concentrations not related to a specific COU release estimate; whereas, a scenario specific
approach was used to estimate environmental exposures that are based specifically on the COU release
estimates. The non-scenario specific approach is generally more applicable to background or away from
facility estimates, but may also be used to represent exposures in industrial areas that contain facilities
relevant to the COUs or other facilities. The approaches used a variety of data types as appropriate,
including:
Page 168 of 723
-------
1) Monitoring data: Measured concentrations from the analysis of primary source monitoring
data (direct use of monitoring data),
2) Modeling data: Predicted concentrations from EPA modeling (modeling data), and/or
3) Concentrations from the interpretation or scaling of monitoring or modeled data {i.e., use of
meta-analysis results, scaling of modeling work by others, etc.).
A summary of the approaches is provided in Table 2-53 and described further below.
Non-Scenario Specific Approach
For the non-scenario specific exposure approach, EPA screened, evaluated, and extracted monitoring
data for surface water, sediment, soil, and targeted wildlife biota. All studies with available monitoring
data and passing evaluation scores were considered for determining environmental concentrations and
overall trends. EPA characterized the data by proximity to industrial facilities based on contextualizing
information provided in the data source. Sampling locations described as industrial, downstream of a
facility, or in proximity of a facility were characterized as "near facility" (or point source). All
remaining data, often with sampling locations described as background, urban, suburban, or rural, were
characterized as "away from facility" (or non-point source). Characterization based on distance between
the sampling location and industrial facility or source attribution is typically not feasible for open source
literature studies because they generally do not provide this information. Additionally, studies do not
always provide the industrial sector of the nearby industrial facilities, which would help to further
characterize the source of HBCD. While primary source monitoring data is the preferred data type for
the non-scenario specific approach, EPA also evaluated monitoring and modeling data provided in
completed assessments.
For the non-scenario specific approach, EPA carried forward for risk estimation an overall central
tendency concentration and high-end concentration for near facility and away from facility datasets.
Since only limited U.S. data was identified through systematic review, data from the U.S. as well as
other high-income countries as classified by the World Bank (June 2019) were included in the final
analysis (https://datahelpdesk.worldbank.ore/knowledgebase/articles/906519-world-bank-coimtrv-and-
lending-groups). High-income countries were selected as surrogate countries based on the assumption
that these countries have manufacturing, processing, and use characteristics that are most likely to
resemble those in the United States. A description of the statistical approach to estimating the central
and high-end concentrations can be found in Risk Evaluation for Cyclic Aliphatic Bromide Cluster
(HBCD), Supplemental Information on General Population, Environmental, and Consumer Exposure
Assessment (U.S. EPA. ). In short, EPA estimated an arithmetic mean and 90th percentile value for
each dataset based on its distribution type (lognormal or normal), and from these values calculated an
overall central tendency (mean of means) and high-end value (average of 90th percentile). The
distribution type was determined from the type and combination of statistical parameters available in the
study {i.e., geometric mean, arithmetic mean, median, geometric standard deviation, standard deviation,
minimum, and/or maximum). Most combinations were assigned a lognormal distribution type, unless
mean estimates were outside the range of reported data. A normal distribution type was assigned to
datasets with only a mean and standard deviation or when the mean and medians were the same.
Datasets were excluded from the final analysis dataset when not enough parameters were available to
estimate a mean or 90th percentile {i.e., only a range of values or only a minimum or maximum value
was reported). The Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on General Population, Environmental, and Consumer Exposure Assessment (U.S. EPA.
2019d) also contains charts that summarize all extracted data and tables with metadata (number of
Page 169 of 723
-------
samples, country, location type, sample years, detection limit, detection frequency, and the data
evaluation score).
Scenario Specific Approach
For the scenario specific exposure approach, no monitoring data specific to U.S. facilities that
manufacture, process, and/or dispose of HBCD related to the conditions of use being assessed were
identified. Therefore, EPA relied on modeling potential releases from facilities using release information
discussed in Section 2.2. The models used in this assessment include: the Exposure Fate Assessment and
Screening Tool (E-FAST), the Variable Volume Water Model Point Source Calculator (VVWM-PSC),
and the Integrated Indoor-Outdoor Air Calculator (IIOAC). A tiered modeling approach was
implemented for surface water concentrations. E-FAST, a simple dilution based model, was first used to
estimate total chemical surface water concentrations in streams. As E-FAST does not consider chemical
partitioning into various media due to a physico-chemical properties (Kow, Koc), it tends to over-
estimate total surface water concentrations and under-estimate the chemical concentration that is sorbed
to soil. Since HBCD's physico-chemical properties lends it to potentially partitioning into various media
(Section 2.1), E-FAST-derived exposures that were greater than the most conservative environmental- or
human health- relevant PoD were triaged for further modeling using the VVWM-PSC model which
incorporates partitioning and degradation. The VVWM-PSC model was also used to estimate settled
sediment in the benthic region of streams. As discussed in Section 2.3.6, a sensitivity analysis was
conducted on select inputs used in the aquatic modeling. Finally, EPA used IIOAC to estimate air
deposition from facility releases, and calculate resulting soil concentrations near the facilities. IIOAC
uses pre-run results from a suite of AERMOD dispersion scenarios at a variety of meteorological and
land-use settings, as well as release emissions, to estimate particle deposition at different distances from
sources that release chemical substances to the air. For contextual purposes only, the IIOAC deposition
results were applied to a generic farm pond setting to calculate concentrations of HBCD in pond surface
water and pond sediment.
For the scenario specific approach, EPA carried forward to risk determination all surface water and
sediment concentrations calculated using VVWM-PSC (second tier model), as well as surface water
concentrations from scenarios modeled in E-FAST (first tier model) that were not triaged for further
modeling in VVWM-PSC.
Table 2-53. Overview of Approaches Used in HBCD Environmental Exposure Assessment
Non-Scenario Specific
Scenario Specific
Primary Data Type
• Monitoring
• Modeling
Characterization
• Near industrial facility (point source) or
away from industrial facility (non-point
source)
• Not specific to a COU
• Near industrial facility (point source)
• Specific to COUs
Facility Estimates/
Releases
• Not applicable
• COU specific (refer to Section 2.2). Releases
were not modeled for a specific facility,
rather hypothetical subscenarios with in each
COU.
Variability
• Central and high-end values
• Lower and upper of days of release/yr and
emission factors, and different release media
types (refer to Section 2.2)
Surface
Water
• Direct use of monitoring data (near and
away from facility)
• Modeling of water releases to rivers (Tiered
approach using E-FAST and WWM-PSC)
Page 170 of 723
-------
Media
Specific
Data Types
• Modeling data from completed assessment
(near and away from facility)
• Modeling of air deposition to ponds (IIOAC)
(contextual purposes only)
Sediment
• Direct use of monitoring data (near and
away from facility)
• Modeling data from completed assessment
(near and away from facility)
• Meta-analysis of monitoring data in
completed assessments
• Modeling of water releases to rivers
(WWM-PSC)
• Modeling of air deposition to ponds (IIOAC)
(contextual purposes only)
Soil3
• Background: Direct use of monitoring data
• Biosolids: Interpretation of monitoring and
model data
• Modeling of air deposition to soil (IIOAC)
Wildlifeb
• Direct use of monitoring data
• Meta-analysis of monitoring data in
completed assessments
• Not applicable
a For soil, the background soil and biosolid soil concentrations were combined with air deposition to soil concentrations for
an overall soil concentration value.
b For wildlife, concentrations were not brought forward to risk estimation.
Water and Air Release Condition of Use Subscenarios for Scenario-Specific Approach
Modeling was conducted by EPA for conditions of use with water and/or air releases, assuming a
conservative process volume of 100,000 pounds/year/site based on the CDR reporting threshold (Section
2.2.1) for exposure scenarios 1 through 6. Lower, more refined facility estimates were used for exposure
scenarios 8 through 12. Up to twelve sub-scenarios per each exposure scenario were created to describe
the range of potential exposure by combining the different identified release types (surface water, on-site
WWT and/or POTW for water releases; stack, fugitive and/or incineration for air releases) with upper
and lower limits (if available) of the number of days of release and emission factors.
Table 2-54 summarizes the water release subscenarios that were used in the E-FAST and VVWM-PSC
models and Table 2-55 summarizes the air release subscenarios that were used in the IIOAC model.
Detailed subscenario tables are provided in Appendix E.
Table 2-54. Summary of Subscenarios Used Across Conditions of Use for Water Releases of
HBCD
Water
Scenarios
COU
Type of Water
Releaseab
Facility
Estimate
(lb/site/yr)
Emission
Factor0
Number of
Release
Daysd
Range of
Daily
Release
(kg/site/day)
Wl.l to
W1.8
1. Repackaging of
Import Containers
On-site,
POTW
100,000
Low: 0.001
High: 0.005
Low: 29
High: 300
1.5E-01 to
7.8E+00
W2.1 to
W2.12
2. Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
Surface Water,
On-site,
POTW
100,000
Low: 3.22E-05
High: 7.42E-05
Low: 10
High: 60
2.4E-02 to
3.4E-01
W3.1 to
W3.12
3. Manufacturing of
XPS Foam using XPS
Masterbatch
Surface Water,
On-site, POTW
100,000
Low: 1.08E-05
High: 2.63E-05
Low: 1
High: 15
3.2E-02 to
1.2E+00
W4.1 to
W4.6
4. Manufacturing of
XPS Foam using
HBCD Powder
Surface Water,
On-site, POTW
100,000
Average: 1.02E-
05
Low: 1
High: 12
3.9E-02 to
4.6E-01
Page 171 of 723
-------
Water
Scenarios
COU
Type of Water
Releaseab
Facility
Estimate
(lb/site/yr)
Emission
Factor0
Number of
Release
Daysd
Range of
Daily
Release
(kg/site/day)
W5.1 to
W5.12
5. Manufacturing of
EPS Foam from
Imported EPS Resin
beads
Surface Water,
On-site, POTW
100,000
Low: 0.011
(combined)
High: 0.015
(combined)
Low: 16
High: 140
3.6E+00 to
4.2E+01
W6.1 to
W6.12
6. Manufacturing of
SIPs and Automobile
Replacement Parts
Surface Water,
On-site, POTW
100,000
Low: 5.06E-05
High: 2.25E-04
Low: 16
High: 300
7.6E-03 to
6.4E-01
W8.1 to
W8.4
8. Installation of
Insulation in Buildings
Surface Water,
POTW
Residential:
37;
Commercial:
2,941
Low: 5.06E-05
High: 2.25E-04
Low: 1
High: 3
8.5E-04 to
1.0E-01
W9.1 to
W9.4
9. Generation of foam
particles during
demolition
Surface Water,
POTW
Low: 37
High: 2,945
Low: 4.5E-05
High: 5.06E-04
1
7.57E-04 to
0.675
W10.1 to
W10.12
10. Recycling of EPS
Foam
Surface Water,
On-site, POTW
70
Low: 0.021
(combined)
High: 0.025
(combined)
Low: 1
High: 140
4.8E-03 to
7.9E-01
W12.1 to
W12.8
12. Use of Solder
On-site, POTW
22
Low: 0.01
High: 0.02
Low: 4
High: 300
3.3E-04 to
5.0E-02
a For each release source, water releases were modeled depending on the potential for the release to go directly to surface
water [Surface Water], to on-site wastewater treatment [On-site], and/or to publicly owned treatment works [POTW]. The
type of release influences two modeling input parameters: 1) Stream flow (million liters per day) and 2) wastewater
removal rates (%). For surface water and on-site WWT release types, the E-FAST default stream flow of "POTW All"
was assigned to COU 8 and the default stream flow of "Plastic Resins" was assigned for all other COUs. For POTW
release types, the E-FAST stream flow default for "Industrial POTWs" was used.
b A water removal rate of 90% was applied to the on-site WWT and POTW releases and no treatment was assumed for
surface water.
e Where identified in literature, EPA utilized a low and high emission factor, with the characterization of those emission
factor described in further details in Section 2.2. If multiple emission factors were identified for the same type of release
media the emission factors were combined.
d Where identified in literature, EPA utilized a range of release days based on the specific condition of use as discussed
further in Section 2.2.
Table 2-55. Summary of Scenarios Used Across Conditions of Use for Air Releases of HBCD
Air
Scenarios
COU
Type of Air
Release
Facility
Estimate
(lb/site/yr)
Emission
Factor
Number of
Release
Days
Range of Daily
Release (kg/site/day)
Al.l to
A1.12
1. Import/Repackaging
Fugitive, Stack,
Incineration
100,000
Low: 0.001
High: 0.005
Low: 29
High: 300
1.5E-01 to 7.8E+00
A2.1 to
A2.8
2. Compounding of
Polystyrene Resin to
Produce XPS Masterbatch
Fugitive, Stack
100,000
Low: 6.12E-06
High: 7.31E-06
Low: 10
High: 60
4.6E-03 to 3.3E-02
A3.1 to
A3.4
3. Manufacturing of XPS
Foam using XPS
Masterbatch
Fugitive, Stack
100,000
Low: 5.79E-05
High: 5.80E-05
Low: 1
High: 16
1.6E-01 to 2.6E+00
A4.1 to
A4.12
4. Manufacturing of XPS
Foam using HBCD Powder
Fugitive, Stack,
Incineration
100,000
Average:
7.29E-06
Low: 1
High: 16
2.1E-02 to 2.3E+01
Page 172 of 723
-------
Air
Scenarios
COU
Type of Air
Release
Facility
Estimate
(lb/site/yr)
Emission
Factor
Number of
Release
Days
Range of Daily
Release (kg/site/day)
A5.1 to
A5.12
5. Manufacturing of EPS
Foam from Imported EPS
Resin beads
Fugitive, Stack,
Incineration
100,000
Low: 0.021
(combined)
High: 0.04
(combined)
Low: 16
High: 140
3.2E-01 to 1.1E+02
A6.1 to
A6.12
6. Manufacturing of SIPs
and Automobile
Replacement Parts
Fugitive, Stack,
Incineration
100,000
Low: 5.06E-05
High: 2.25E-04
Low: 16
High: 300
7.6E-03 to 7.2E+01
A8.1 to
A8.4
8. Installation of Insulation
in Buildings
Fugitive,
Incineration
Residential:
37;
Commercial:
2,941;
Low: 5.06E-05
High: 2.25E-04
Low: 1
High: 3
8.5E-04 to 1.1E+01
W9.1 to 9.2
9. Generation of foam
particles during demolition
Fugitive
Low: 37
High: 2,945
4.5E-05
1
7.57E-04 to 0.675
A10.1 to
A10.12
10. Recycling of EPS Foam
Fugitive, Stack,
Incineration
70
Low: 0.021
(combined)
High: 0.025
(combined)
Low: 1
High: 140
2.3E-04 to 7.9E-01
All.l to
A11.4
11. Formulation of solder
Fugitive, Stack
d
d
Low: 5
High: 300
1.5E-03 to 1.3E+00
A12.1 to
A12.4
12. Use of Solder
Incineration
22
Low: 0.08
(combined)
High: 0.09
(combined)
Low: 4
High: 300
2.7E-03 to 2.2E-01
a For each release source, air releases were modeled depending on whether the releases were from fugitive, stack or
incineration emissions.
b Where identified in literature, EPA utilized a range of emission factors with the characterization of those emission factor
described in further details in Section 2.2.
°Where identified in literature, EPA utilized a range of release days based on the specific condition of use as discussed
further in Section 2.2.
d Daily release estimates were based on releases reported to 2017 TRI
2.3.2 Aquatic Environment - Surface Water and Sediment
2.3.2.1 Non-Scenario Specific Approach
The non-scenario specific approach uses measured media-specific monitoring data to characterize
background exposure to HBCD where releases attributed to historical and current conditions of use may
be encompassed. As described below in Section 2.3.2.2, all exposure scenarios with surface water
releases have predicted surface water and sediment HBCD concentrations, except for land disposal of
other formulated products and articles (e.g., adhesives, coatings, textiles, and electronics) via the
potential leaching capacity of HBCD from these facilities (not through the disposal process of these
formulated products and articles) or runoff. In lieu of having media-specific release information for this
condition of use via leaching or surface runoff, background information (monitoring data) is used as a
proxy to characterize the risk to aquatic organisms.
EPA first evaluated environmental exposures to aquatic organisms based on environmental monitoring
data as well as modeled site-specific exposures or exposure scenarios. This non-scenario-specific
approach estimates background exposure from a multitude of different sources. The totality of
background exposure includes steady-state environmental exposures ongoing releases not associated
with a particular COU, background/indirect exposures from minor use products (e.g., textiles, electrical
Page 173 of 723
-------
and electronic products, adhesives, and coatings) (Section 1.2.8), and releases stemming from historical
activities (Section 1.2.9) due to HBCD's persistence in the environment.
2.3.2.1.1 Surface Water Concentrations
EPA identified and extracted measured concentrations of HBCD in surface water from thirteen primary
source studies. This dataset includes samples collected between 2006 and 2016 from rivers and lakes
located in the United States (Great Lakes area), Antarctica, Canada, China, Denmark, England, Japan,
Korea, Netherlands, Poland, and South Africa. A summary of occurrence of HBCD in surface water is
presented in the Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on General Population, Environmental, and Consumer Exposure Assessment (U.S. EPA.
2019d).
Near facility
Following data aggregation and statistical analysis procedures, three studies were included in the
surrogate country dataset for the near facility (point source) characterization. The central tendency and
high-end surface water concentrations were 0.84 and 0.99 |ig/L, respectively, with a maximum reported
concentration of 3.1 |ig/L. Overall, a tight range of values was reported. Concentrations at the higher
end of the range were detected in Poland in 2014 near industrial facilities that recycle plastic materials
(Kowalski and Mazur 2014) and in Japan in 2011 near dyeing and textile factories (EC/HC ). No
U.S. near facility monitoring data was identified. A review of completed assessments shows similar
results; with a maximum of 1.52 |ig/L reported in the European Commission risk assessment (EINECS
2008) for a small tributary receiving surface water from a production facility estate in the UK.
Modeled site-specific and generic near facility estimates were also compiled from various international
sources. In fresh or seawater, concentrations ranged from 4.8E-5 to 370 |ig/L (EINECS 2008; EC/HC
2011; NICNAS 2012; ECHA.: ). The highest concentration represents a worst case generic scenario
of an intermittent (single day) release from the industrial use of XPS (EC ). Ilyina and
Hunziker 2010 predicted concentration in the North Sea using the Fate and Transport Ocean Model
(FANTOM). Using estimated annual emissions for EU industrial sites, they estimated that HBCD
concentrations in the surface water layer ranged from 10"6 to 0.1 |ig/L. The modeling indicates that
concentrations decline steeply with increasing distance from point sources and respond immediately to
changes in emission, however, a product might be transported to remote environments depending on its
half-life in the atmosphere.
Away from Facility
Following data aggregation and statistical analysis procedures, four studies were included in the
surrogate country dataset for the away from facility (non-point source) characterization. The central
tendency and high-end surface water concentrations were 4.1E-04 and 8.0E-04 |ig/L, respectively, with
a maximum reported concentration of 0.0067 |ig/L. The highest concentration was reported in Japan
from a study which collected samples from 19 sampling locations in the Yodo River basin consisting of
forest, paddy field, and city areas, as well as highly urbanized and industrialized areas (Ichihara et al.
2014). This study reported flow rates and as well as estimated pollutant loads. It is noteworthy that the
lowest flow river, the Yamato River, had the highest HBCD concentration. In the only U.S. study,
(Venier et al. 2014) measured HBCD in surface water samples from the Great Lakes. Overall
concentrations ranged from 2.0E-7 ug/L to 4.4E-6 |ig/L, with an average of 1.2E-6 |ig/L in detected
samples (detection frequency of 14 out of 23 samples). Similar low concentrations were observed in
nine lakes in the UK, with average concentrations ranging from 8.0E-5 ug/L to 2.7E-4 |ig/L (Harrad et
al. 2009).
Page 174 of 723
-------
Table 2-56. Summary of Central Tendency and High-End Estimated Surface Water
Concentrations from Monitoring Data
Site Characterization
Number of
Studies
Identified
Number of
Studies
Included in
Final Dataset
Estimated Concentrations (jig/L)
Central Tendency
High-End
Near Industrial
Facility3 (Point Source)
5
3a
0.84
0.99
Away from Facility13
(Non-Point Source)
9
4b
4.1E-04
8.0E-04
aNear industrial facility studies: (Ichihara et al. 2014: Kowalski and Mazur 2014: Oh et al. 2014)
bAway from facility studies: (Law et al. 2006: Harrad et al. 2009: Ichihara et al. 2014: Venier et al. 2014)
2.3.2.1.2 Sediment Concentrations
EPA identified and extracted measured concentrations of HBCD in surface water from 55 primary
source studies. This dataset includes samples collected between 1974 and 2016 from freshwater and
seawater in the United States, Australia, Canada, China, Czech Republic, England, Italy, Japan, Korea,
Kuwait, Netherlands, Norway, Singapore, South Africa, South Korea, Spain, and Switzerland. A
summary of occurrence of HBCD in sediment is presented in the Risk Evaluation for Cyclic Aliphatic
Bromide Cluster (HBCD), Supplemental Information on General Population, Environmental, and
Consumer Exposure Assessment (U.S. EPA 2019d).
Near Facility
Following data aggregation and statistical analysis procedures, six studies were included in the surrogate
country dataset for the near facility (point source) characterization. The central tendency and high-end
surface water concentrations were 3,443 and 5,073 |ig/kg, respectively, with a maximum reported
concentration of 85,000 |ig/kg. The final surrogate country dataset included only one U.S. study, La
Guardia (La Guardia et al. 2010). sediment samples were collected in 2009 in the vicinity of a municipal
wastewater treatment plant (WWTP) in North Carolina that was likely receiving waste from a textile
manufacturer. Total HBCD concentrations ranged from 3.1 to 42.9 |ig/kg dw downstream of the outfall
(0 to 44.6 km) and was not detected upstream from the outfall. Although not included in the final dataset
due to incomplete statistical data reported, similar but lower concentrations ranging from non-detect to
3.7 |ig/kg were reported in Marvin (Marvin et al. 2006). for suspended sediment samples collected in
urban/industrial areas of the Detroit River in 2001. The higher concentrations (2.6 to 3.7 |ig/kg) were
reported in areas of contemporary industrial activity and the lower concentrations were associated with
areas of historical industrial activity. The maximum concentration in the dataset is from Haukas (Haukas
et al. 2010b). a Norwegian study that sampled sediment from a highly contaminated fjord with the likely
source of HBCD from a local polystyrene production plant. The next highest reported concentration was
7,800 |ig/kg from a Japanese study (Oh et al. 2014) that collected sediment from a river receiving
effluents from textile industries. Two studies in Spain, (Guerra et al. 2009) and (Guerra et al. 2010).
reported a trend with higher sediment concentrations located near point sources and decreasing sediment
concentrations downstream from point sources, and non-detects upstream or further away from point
sources.
Away from Facility
Following data aggregation and statistical analysis procedures, 14 studies were included in the surrogate
country dataset for the away from facility (non-point source) characterization. The central tendency and
high-end surface water concentrations were 6.2 and 19.8 |ig/kg dw, respectively, with a maximum
Page 175 of 723
-------
reported concentration of 1,680 |ig/kg dw. However, many studies in this dataset are based on statistical
summaries for a range of land use patterns, thus the highest concentration were often from areas
reported to be industrial in nature. HBCD concentrations from studies in which industrial activity was
not reported tended be less than 10 |ig/kg dw. The two U.S. studies included in the surrogate country
dataset included (Yang et al. 2012) which reported surface sediment concentrations ranging from 0.04 to
3.1 |ig/kg dw collected from the five Great Lakes in 2007 and (Klosterhaus et al. 2012) which reported
surface sediment concentrations ranging from 0.01 to 2 |ig/kg dw collected from the San Francisco Bay
estuary in 2007. As mentioned above, (Marvin et al. 2006) reported that suspended sediment in the
Detroit River in areas of historical industry activity were less than 2.6 |ig/kg.
The EC (EINECS 2008) assessment characterized sediment concentrations both near point sources and
away from point sources in a meta-analysis of 16 studies encompassing locations in Belgium (Scheldt
Basin), Switzerland, Spain, Ireland, Norway, Sweden, and United Kingdom. Reported concentrations
ranged from 0.05 to 511 |ig/kg for areas not impacted by point sources. Overall the data set is skewed
with median HBCD concentration of 1.5 |ig/kg, lower than the mean HBCD concentration of 31 |ig/kg.
The 90th percentile HBCD concentration was estimated as 100 |ig/kg. When considering pollution by
industrial activities, the maximum observed concentrations were more than 30,000 |ig/kg, but were
associated with production of HBCD and the textile industry.
Modeled site-specific and generic near facility estimates were also compiled from various international
sources. In fresh or marine sediment, concentrations ranged from 1.0E-3 to 4.0E+6 |ig/kg (EINECS
2008; EC/HC. 2011; NICNAS 2012b; ECHA 2017b) The highest concentration represents a worst case
generic scenario of an intermittent (single day) release from the industrial use of XPS (ECHA 2017b).
Ilyina (Ilyina and Hunziker 2010) predicted concentration in the North Sea using the Fate and Transport
Ocean Model (FANTOM). Using estimated annual emissions for EU industrial sites, they estimated
HBCD concentrations in the surface water layer ranged from 10 E -4 to 10 |ig/kg.
Table 2-57. Summary of Central Tendency and High-End Estimated Sediment Concentrations
'rom Monitoring Data
Site Characterization
Number of
Studies
Identified
Number of
Studies
Included in
Final Dataset
Estimated Concentrations (jig/kg)
Central Tendency
High-End
Near Industrial Facility
(Point Source)
15
6a
3,443
5,073
Away from Facility
(Non-Point Source)
45
14b
6.2
19.8
aNear industrial facility studies: (Sellstromet al. 1998; La Guardia et al. 2012; Haukas et al. 2010b; Ohet al. 2014; Al-
Odaini et al. 2015; Stiborova et al. 2017)
bAway from facility studies: (Ramu et al. 2010; Klosterhaus et al. 2012; Yang et al. 2012; Harrad et al. 2009; Haukas et al.
2009; Haukas et al. 2010b; Kohler et al. 2008; Minliet al. 2007; Morris et al. 2004; Remberger et al. 2004; Jeong et al. 2014;
Luigi et al. 2015; Lyons et al. 2015; Al-Odaini et al. 2015; Anim et al. 2017)
2.3.2.2 Scenario Specific Approach
This section describes the method and results from the scenario specific approach to estimating
concentrations in the aquatic environment, when water releases are estimated to occur. E-FAST was
used as a first-tier model to identify where modeled surface water column concentrations did or did not
exceed aquatic hazard values. Since the E-FAST model incorporates defaults that encompass either a
combination of upper percentile and mean exposure parametric values, or all upper percentile parametric
Page 176 of 723
-------
values, the resulting model predictions represent high-end exposures estimates. EPA acknowledges the
conservative nature of this approach, and used the VVWM-PSC, to further describe environmental
exposures as described later in this section. The VVWM-PSC model was then used to identify 1-day and
21-day average dissolved and suspended sediment water concentration as well as 28-day sediment
concentrations. Appendix G contains the daily release amounts and environmental concentrations for
each subscenario modeled.
EPA's Exposure and Fate Assessment Screening Tool (E-FAST), Version 2.0, was specifically
developed to support EPA assessments of potential environmental exposures. The E-FAST model
contains default parameter values that allow for exposure estimations of a chemical in the surface water
after a source emits the chemical into a water body by considering simple dilution. EPA uses Equation
2-5 to estimate surface water concentrations in E-FAST.
Equation 2-5
Where:
SWC = Surface water concentration in |ig/L
R = Release kg/site/day
CF1 = Conversion factor (109 |ig/kg)
T= Percent removal, typically from wastewater treatment
SF = Flow of receiving river (million liters per day)
CF2 = Conversion factor (106 L/day/MLD)
Release (kg/site/day)
As discussed in Section 2.2, the daily release values (kg/site/day) were calculated using a production
volume of 100,000 lbs/yr/site (or another lower facility specific estimate), emission factors (kg HBCD
released/kg HBCD handled), and number of release days per year. Refer to Table 2-54 for a summary of
the release values by COU and Appendix G for subscenario specific release values.
Removal from wastewater treatment (%)
Removal from wastewater treatment is the percentage of the chemical removed from wastewater during
treatment before discharge to a body of water. As discussed in Section 2.1.2.4, removal from wastewater
treatment for HBCD was estimated at 90%. EPA assumed that treatment occurs for "on-site WWT" and
"POTW" release types, and that 90% removal was achieved. EPA assumed that direct releases to water
did not receive wastewater treatment and no wastewater treatment removal was applied. This is a
conservative assumption that results in the total amount of HBCD released to wastewater treatment at a
direct discharging site being released to surface water. This assumption reflects the uncertainty of the
type of wastewater treatment that may be in use at a direct discharging facility and the HBCD removal
efficiency in that treatment.
2.3.2.2.1 E-FAST: Predicted Flowing Surface Water Concentrations (First Tier
Modeling)
SWC =
SF X CF2
Inputs
Page 177 of 723
-------
Flow of receiving river (million liters per day)
E-FAST requires the selection of a receiving stream flow from the E-FAST 2014 database. For site-
specific assessments, the stream flow is selected by searching for a facility's National Pollutant
Discharge Elimination System (NPDES) permit number, name, or the known discharging waterbody
reach code. As no specific facilities were identified for the HBCD assessment for water releases, stream
flows were selected using the "SIC Code Option" within E-FAST. This option uses the 10th and 50th
percentile stream flows of all facilities in a given industry sector, as defined by the Standard Industrial
Classification (SIC) codes of the industry sector. For all "surface water" and "on-site WWT" release
types, the sector based stream flows used were "POTW All" for subscenarios in COU 8 (installation of
insulation into building scenario) and "Plastic Resins" for subscenarios in all other COUs. For the
"POTW" release type, the SIC based stream flow of "Industrial POTWs" was used. These SIC Code
stream flows were selected because they were thought to best represent the industrial activity associated
with the conditions of use and release type.
The flow of rivers is highly variable and is dependent on many factors such as weather patterns and
effluent released from different facilities. The volume of a river varies over time with different flows
expected seasonally and from year to year. The 10th and 50th percentile 7Q10 flows, which represent the
lowest expected weekly flow over a ten-year period, were selected for use in the ecological risk
assessment. In general, the 10th percentile flow values are approximately a factor of 10 lower than 50th
percentile flows. The flows for the selected industry sector/SIC Code are shown in Table 2-58. Although
not used in the ecological assessment, harmonic means are also shown since they were used to calculate
surface water concentrations for the scenario specific fish ingestion scenario in the highly exposed
human exposure assessment. Harmonic mean flow values represent long-term average flow conditions.
Table 2-58. Receiving Stream Flow Values
Sector Within EFAST
7Q10 Flow MLD
50th percentile
7Q10 Flow MLD
10th percentile
Harmonic Mean
Flow MLD
50th percentile
Harmonic Mean
Flow MLD
10th percentile
SIC Code- Plastic Resins
4.0E+02
8.0E+00
1.3E+03
4.5E+01
SIC Code- Industrial POTW
7.8E+01
7.8E+00
2.9E+02
4.0E+01
SIC Code- All POTW
2.7E+01
1.1E+00
1.3E+02
1.1E+01
Outputs
Overall, surface water concentrations ranged from 8.30E-05 to 1.10E+02 |ig/L using the 50th percentile
7Q10 flows and 4.20E-03 to 5.30E+03 using the 10th percentile 7Q10 flows. Refer to Table 2-59 for a
summary of modeled surface water estimates by condition of use, and Appendix E.7 for results by sub-
scenario.
Table 2-59. Estimated HBCD Surface Water (iig/L) Concentrations Using E-FAST
Water Scenarios
7Q10 SWC
50th percentile
7Q10 SWC
10th percentile
Wl.l to W1.8
3.7E-02 - 1.0E+01
1.9E+00 - 1.0E+02
W2.1 to W2.12
6.1E-03 - 8.4E-01
3.0E-01 - 4.2E+01
W3.1 to W3.12
1.0E-02 - 3.0E+00
4.0E-01 - 1.5E+02
W4.1 to W4.6
9.7E-03 - 1.2E+00
4.9E-01 - 5.8E+01
W5.1 to W5.12
8.8E-01 - 1.1E+02
4.4E+01 - 5.3E+03
Page 178 of 723
-------
Water Scenarios
7Q10 SWC
50th percentile
7Q10 SWC
10th percentile
W6.1 to W6.12
1.9E-03 - 1.6E+00
9.5E-02 - 8.0E+01
W8.1 to W8.4
3.2E-04 - 3.7E-01
8.0E-03 - 9.4E+00
W9.1 to W9.4
2.8E-03 - 2.5E+01
7.2E-02 - 6.4E+02
W10.1 to W10.12
1.2E-03 - 2.0E+00
5.9E-02 - 9.9E+01
W12.1 to W12.8
8.3E-05 - 6.4E-02
4.2E-03 - 6.4E-01
Bold = concentration above water solubility of 66 (ig/L
Advantages to the E-FAST model are that it requires minimal input parameters and it has undergone
extensive peer review by experts outside of EPA. The limitations associated with use of the E-FAST
model relate to the assumptions made regarding use of sector-based flow information as a surrogate for
site-specific flow information, as well as lack of partitioning (between dissolved and suspended
sediment within the water column or between the water column and the benthic environment) and
degradation parameters that were employed in the PSC model. Additionally, low-flow stream inputs
combined with high-release estimates may yield overly conservative surface water concentrations
greater than the water solubility of HBCD.
Site-specific parameters influence how partitioning occurs over time. For example, the concentration of
suspended sediments, water depth, and weather patterns all influence how a chemical may partition
between compartments. Physical-chemical properties of the chemical itself also influence partitioning
and half-lives into environmental media. HBCD has a Koc of 100,000, indicating a high potential to
sorb to suspended particles in the water column and settled sediment in the benthic environment. Canada
(EC/HC. 2011) considered these parameters when estimating surface water and sediment concentrations
of HBCD in rivers receiving HBCD from two types of point sources (raw material handling and
compounding). Surface water and sediment concentrations were estimated at 100 m from the facility and
5,000 m from the facility using a fugacity-based model with 10 downstream boxes each with water and
sediment compartments. The model is based on the principles described by Cahill et al. (2003). and
more generally Mackav (1991). The Canadian modeled estimates ranged from 0.04 to 15 |ig/L in surface
water at 100 m from the facility, which is within the range of the E-FAST estimated values. At 5 km
from the facility, the modeled concentrations ranged from 0.03 to 10 |ig/L. In sediment, the Canadian
model predicted concentrations from 230 to 108,200 |ig/kg. The Canadian estimates were modeled using
a quantity of 10,000 kg/year with 60 days of release and 100,000 kg/year with 200 days of release,
combined with worst-case emission factors of 0.055% (raw material handling) and 0.6%
(compounding), and treatment removal rates of 0, 57, and 90%. Stream discharge was set to 0.85 m3/s
(73 MLD) to represent the 25th percentile of observed rates in Southern Ontario. This resulted in 6
subscenarios per point source. It is noteworthy that this modeling was conducted when releases to
surface water from uses of HBCD were likely higher than they are today.
2.3.2.2.2 VVWM-PSC: Predicted Flowing Surface Water Concentrations (Second Tier
Modeling) and Sediment Concentrations
As a second tier approach, EPA used the Variable Volume Waterbody Model (VVWM) - Point Source
Calculator (PSC) (U.S. EPA 2019q) to model dissolved water and settled sediment concentrations
separately, using the same surface water release estimates used in E-FAST (refer to Table 2-54 for the
daily release estimates). The PSC is a tool designed to estimate time-varying surface water
concentrations of a chemical directly applied to a water body, including but not limited to river
segments. Loading into the river can be varied daily, set up to be discrete one-time events, or repetitive
Page 179 of 723
-------
events over most or all of the year. The PSC is a graphical user interface which gathers the user's inputs
and runs USEPA's VVWM. Required inputs are the same as those for the VVWM, but the PSC
graphical interface facilitates user interaction for the direct-application and allows model inputs to be
defined by the user. Time-varying surface water concentrations can be averaged over variable time
periods for comparison to concentrations of concern. For example, 21-day average surface water
concentrations and 28-day average sediment concentrations were used for EPA's modeling assessment.
Inputs
More information on the equations used to estimate surface water and sediment concentrations are
available in the PSC user guide (U.S. EPA 2019q). In short, daily releases and daily flow values are
used along with other model inputs to solve mass-balance equations for the water column and for the
benthic region.
Surface water flow can be set up to be constant flow or use time-varying flows. Since site-specific
information is not available for the HBCD assessment, constant flows matching the SIC-based flow
values used in E-FAST were selected (refer to Table 2-58). Suspended sediment values are highly
variable and are influenced by stream flow, land cover, and river conditions. A Koc value of 100,000
was chosen based on measured data. A weather file is also needed to run VVWM-PSC. This
incorporates variable flow volume through precipitation events. However, variation through
precipitation alters stream flow much less than variations in stream flow from other factors. Use of a
constant flow which varied across scenarios was chosen. Table 2-60 displays the inputs used to run the
VVWM-PSC for HBCD.
Table 2-60. Inputs for Modeling HBCD Sediment Concentration using VVWM-PSC
Input
Type of
Input
Value
Units, Comments
Reference
Sorption Coefficient (Koc)
Chemical
100,000
ml/g
(ECHA 2017b)
Water Column,
Hydrolysis, and Photolysis
Half-lives
Chemical
365
Days
Benthic Half-Live
Chemical
11 to 128
Days
(Davis et al. 2005)
(Davis et al. 2006)
Molecular weight
Chemical
641.7
g/mol
Henry's Law Constant
Chemical
7.4E-6
atm-m3/mole
(U.S. EPA 2012b)
Heat of Henry
Chemical
41570
J/mol
(U.S. EPA 2019a)
Loading schedule
Chemical
Varies can add separate
table and/or add
combinations here.
Offset, number of days
on and off
River width
Environment
8
Meters
(EC/HC 2011)
River depth
Environment
2
Meters
River length
Environment
100
Meters
Flow rate
Environment
Varies
See Table 2-37
(U.S. EPA 2014c)
DFAC
Environment
1.19
Photolysis parameter:
Represents the ratio of
(U.S. EPA 2019a)
Page 180 of 723
-------
Input
Type of
Input
Value
Units, Comments
Reference
vertical path lengths to
depth
Water Column Suspended
Sediment
Environment
50
mg/L
(Dodds and Oakes
2004)
Chlorophyll
Environment
0.005
mg/L
Water Column Fraction
Environment
0.04
Fraction
Organic Content
Water Column Dissolved
Oxygen Content
Environment
5.0
mg/L
Water Column Biomass
Environment
0.4
mg/L
Benthic Depth
Environment
0.05
M
Benthic Porosity
Environment
0.5
(U.S. EPA 2019a)
Bulk Density
Environment
1.35
g/cm3
Benthic Fraction Organic
Content
Environment
0.04
Benthic Dissolved Oxygen
Content
Environment
5.0
mg/L
Benthic Biomass
Environment
0.006
g/m2
Mass Transfer Coefficient
Environment
le-8
m/s
Outputs
A summary of the estimated surface water and sediment concentrations from VVWM-PSC by condition
of use is provided in Table 2-61 based on 7Q10 50th percentiles and in Table 2-62 based on 7Q10 10th
percentiles. Sediment concentrations were calculated for both all and 128 day benthic half-life to
account for the large range of values. The 1-day average overall surface water column concentrations are
similar to estimated surface water concentrations from E-FAST because the same flow values were
used. Further, the PSC was only run for scenarios where the estimated surface water concentration from
E-FAST exceeded an acute or chronic aquatic hazard value (discussed in Section 3.1). See Section 2.3.6
regarding the qualitative sensitivity analysis associated with these results.
Table 2-61. Estimated HBCD Surface Water Concentrations (jig/L) and Sediment Concentrations
(ug/kg) Using VVWM-PSC with 50th Percentile 7Q10 Flows
Water
Scenarios
Water
Column 1
Day
average
Water
Column
Dissolved 1
Day jig/L
Water
Column
Suspended 1
Day jig/L
Water
Column ng/L
21 day
average
Water
Column 21
day Dissolved
Hg/L
Water
Column
21 day
Suspended
Hg/L
Sediment
Mg/kg
28 day
average
(128)a
Sediment
M«/kg
28 day
average
(ii)a
Wl.l to
W1.8
3.7E-02 -
9.7E+00
2.8E-02 -
7.3E+00
5.6E-03 -
1.5E+00
3.0E-02 -
9.4E-01
2.3E-02 -
7.1E-01
4.6E-03 -
1.4E-01
7.7E+01 -
2.0E+03
3.4E+01 -
8.7E+02
W2.1 to
W2.12
3.7E-02 -
8.3E-01
2.8E-02 -
6.3E-01
5.6E-03 -
1.3E-01
1.8E-03 -
4.0E-02
1.3E-03 -
3.0E-02
2.7E-04 -
6.0E-03
2.8E+00 -
6.3E+01
1.3E+00 -
3.0E+01
W3.1 to
W3.12
8.0E-03 -
2.9E+00
6.0E-03 -
2.2E+00
1.2E-03 -
4.4E-01
3.8E-04 -
1.4E-01
2.9E-04 -
1.1E-01
5.8E-05 -
2.1E-02
8.9E-01 -
1.2E+02
4.0E-01 -
8.9E+01
W4.1 to
W4.6
9.6E-03 -
1.1E+00
7.3E-03 -
8.6E-01
1.5E-03 -
1.7E-01
4.6E-04 -
5.4E-02
3.5E-04 -
4.1E-02
6.9E-05 -
8.2E-03
8.2E-01 -
4.6E+01
3.7E-01 -
3.5E+01
Page 181 of 723
-------
Water
Scenarios
Water
Column 1
Day
average
Water
Column
Dissolved 1
Day jig/L
Water
Column
Suspended 1
Day jig/L
Water
Column ng/L
21 day
average
Water
Column 21
day Dissolved
Hg/L
Water
Column
21 day
Suspended
Hg/L
Sediment
Mg/kg
28 day
average
(128)a
Sediment
Mg/kg
28 day
average
air
W5.1 to
W5.12
8.8E-01 -
1.1E+02
6.6E-01 -
7.9E+01
1.3E-01 -
1.6E+01
2.9E-01 -
5.0E+00
2.2E-01 -
3.8E+00
4.5E-02 -
7.6E-01
7.6E+02 -
1.2E+04
3.3E+02 -
5.5E+03
W6.1 to
W6.12
8.4E-03 -
1.6E+00
6.4E-03 -
1.2E+00
1.3E-03 -
2.4E-01
1.7E-03 -
7.5E-02
1.3E-03 -
5.7E-02
2.6E-04 -
1.1E-02
4.1E+00 -
1.8E+02
1.9E+00 -
8.3E+01
W8.1 to
W8.4
2.9E-03 -
3.4E+00
2.2E-0 -3
2.6E+00
4.4E-04 -
5.1E-01
1.4E-04-
4.9E-01
1.1E-04 -
3.7E-01
2.1E-05 -
7.4E-02
1.2E-01 -
1.6E+02
8.9E-02 -
1.1E+02
W9.1 to
W9.4
2.3E+00 -
2.3E+01
1.7E+00-
1.7E+01
3.5E-01 -
3.5E+00
1.1E-01 -
1.1E+00
8.5E-02 -
8.5E-01
1.7E-02 -
1.7E-01
9.4E+01 -
9.4E+02
7.0E+01 -
7.0E+02
W10.1 to
W10.12
1.2E-02-
2.0E+00
8.9E-03 -
1.5E+00
1.8E-03 -
3.0E-01
3.9E-03 -
9.3E-02
3.0E-03 -
7.1E-02
6.0E-04 -
1.4E-02
6.6E+00 -
9.6E+01
4.5E+00 -
6.0E+01
W12.1 to
W12.8
6.2E-03 -
6.2E-02
4.7E-03 -
4.7E-02
9.3E-04 -
9.4E-03
2.9E-04 -
3.0E-03
2.2E-04 -
2.3E-03
4.5E-05 -
4.5E-04
2.9E-01 -
2.9E+00
1.9E-01 -
1.9E+00
a sediment benthic half-life (days)
Bold = concentration above the water solubility of 66 ng/L
Table 2-62. Estimated HBCD Surface Water Concentrations (jig/L) and Sediment Concentrations
lig/kg) Using VVWM-PSC with 10th Percentile 7Q10 Flows
Water
Scenarios
Water
Column 1
Day
average
Water
Column
Dissolved 1
Day ng/L
Water Column
Suspended 1
Day jig/L
Water
Column |ig/L
21 day
average
Water
Column 21
day Dissolved
Hg/L
Water
Column
21 day
Suspended
Hg/L
Sediment
Mg/kg
28 day
average
(128)a
Sediment
Mg/kg
28 day
average
air
Wl.l to
W1.8
1.7E+00-
7.6E+01
1.3E+00 -
5.7E+01
2.6E-01 -
1.1E+01
1.5E+00 -
8.9E+00
1.1E+00 -
6.7E+00
2.2E-01 -
1.3E+00
3.6E+03 -
1.9E+04
1.4E+03 -
7.2E+03
W2.1 to
W2.12
1.4E+00-
3.1E+01
1.1E+00 -
2.4E+01
2.1E-01 -
4.7E+00
7.9E-02 -
1.8E+00
5.9E-02 -
1.3E+00
1.2E-02-
2.7E-01
1.3E+02 -
2.9E+03
5.4E+01 -
1.2E+03
W3.1 to
W3.12
3.0E-01 -
1.1E+02
2.3E-01 -
8.3E+01
4.6E-02 -
1.7E+01
1.8E-02 -
5.7E+00
1.4E-02 -
4.3E+00
2.7E-03 -
8.6E-01
4.1E+01 -
4.7E+03
1.6E+01 -
3.5E+03
W4.1 to
W4.6
3.6E-01 -
4.3E+01
2.7E-01 -
3.2E+01
5.5E-02 -
6.5E+00
2.1E-02 -
2.2E+00
1.6E-02 -
1.7E+00
3.1E-03 -
3.3E-01
3.9E+01 -
1.8E+03
1.5E+01 -
1.4E+03
W5.1 to
W5.12
3.6E+01 -
4.0E+03
2.7E+01 -
3.0E+03
5.4E+00 -
6.0E+02
1.4E+01 -
2.4E+02
1.1E+01 -
1.8E+02
2.1E+00 -
3.6E+01
3.6E+04 -
5.7E+05
1.4E+04 -
2.3E+05
W6.1 to
W6.12
3.9E-01 -
6.0E+01
2.9E-01 -
4.5E+01
5.9E-02 -
9.0E+00
7.9E-02 -
3.5E+00
6.0E-02 -
2.7E+00
1.2E-02-
5.3E-01
1.9E+02 -
8.5E+03
7.6E+01 -
3.4E+03
W8.1 to
W8.4
2.0E-02 -
2.4E+01
1.5E-02 -
1.8E+01
3.0E-03 -
3.6E+00
1.3E-03 -
1.7E+00
9.8E-04 -
1.3E+00
2.0E-04 -
2.6E-01
1.1E+00 -
2.0E+03
7.6E-01 -
9.0E+02
W9.1 to
W9.4
1.6E+01 -
1.6E+02
1.2E+01 -
1.2E+02
2.4E+00 -
2.4E+01
1.0E+00 -
1.0E+01
7.7E-01 -
7.7E+00
1.5E-01 -
1.5E+00
8.4E+02 -
8.4E+03
6.0E+02 -
6.0E+03
W10.1 to
W10.12
4.8E-01 -
7.3E+01
3.6E-01 -
5.5E+01
7.3E-02 -
1.1E+01
1.9E-01 -
3.8E+00
1.4E-01 -
2.8E+00
2.8E-02 -
5.7E-01
2.6E+02 -
3.1E+03
1.8E+02 -
2.3E+03
W12.1 to
W12.8
2.3E-01 -
4.7E-01
1.7E-01 -
3.6E-01
3.5E-02 -
7.2E-02
1.2E-02-
2.5E-02
9.2E-03 -
1.9E-02
1.8E-03 -
3.8E-03
1.3E+01 -
2.6E+01
7.4E+00 -
1.5E+01
a sediment benthic half-life (days)
Bold = concentration above the water solubility of 66 ng/L
Page 182 of 723
-------
2.3.2.2.3 IIOAC: Predicted Pond Water and Sediment Concentrations
With an estimated half-life in air of more than two days, and having been detected in Arctic
environmental media, there is strong evidence of HBCD's potential for long-range transport (TJNEP
2010b). EPA calculated the concentration of HBCD in pond water and sediment resulting from air
deposition using a two-step process.
In the first step, near-facility HBCD annual deposition rates were modeled using EPA's Integrated
Indoor-Outdoor Air Calculator (IIOAC) for 11 condition of use-exposure scenarios with air releases.
Under each scenario, multiple model runs were performed to include different source types and high-end
and central tendency release estimates, as summarized in Table 2-55. For scenarios with site-specific
information, this information was used in the IIOAC model runs to determine the meteorological station
and population setting. When site-specific information was not known, representative central tendency
and high-end meteorological stations were used, along with other default parameters in Appendix G.
Table Apx F-4 in the Environmental Exposure appendix presents the modeled range of total annual
particle deposition for each exposure scenario by source type (fugitive, stack, incineration) and by
receptor (fenceline, community). Fenceline estimates were defined as 100-meter from the source while
community-averaged estimates were within 100 to 1,000-meter from the facility. From the table, the
highest total annual particle deposition amongst all exposure scenarios was:
• 2.28E-05 g/m2 at the fenceline (100 m from the source); and
• 1.75E-06 g/m2 at "community" receptors beyond the fenceline (100 to 1,000 m from the source).
Background deposition rates of HBCD were also reported in a recent study near the Great Lakes and
ranged from non-detectable levels up to 82 ng/m2/d, with an average of 2.3 ng/m2/d. These values
corresponded to wet deposition of HBCD as detected with automated wet-deposition samplers located at
sites ranging from remote to peri-urban (Robson et al. 2.013). Observed HBCD deposition values varied
by location (perhaps due in part to meteorological conditions) and, to a lesser extent, by time, though
sampling time was limited to four years at some sites. For comparison to the IIOAC-modeled values,
EPA assumed that the observed per-day fluxes from (Robson et al. 2013) were held constant for a year,
resulting in:
• 2.99E-05 g/m2/y for maximum deposition; and
• 8.40E-07 g/m2/y for average deposition
Using the deposition rates estimated by IIOAC and the background deposition rates reported by (Robson
et al. ), the total annual deposition and resulting surface water and sediment concentrations were
calculated for a generic farm pond scenario. The scenario is based off of EPA's Office of Pesticides
(OPP) standard farm scenario as described in various models such as the EXAMS model and
GENEEC2. Equation 2-6 was used to calculate the total annual deposition to the water body (|ig) and
the HBCD surface water and sediment concentrations were calculated using Equation 2-7 and Equation
2-8, respectively.
Equation 2-6
AnnDep = TotDep xArxCF
Where
AnnDep = Total annual deposition to water body catchment (|ig)
TotDep = Annual deposition flux to water body catchment (g/m2)
Page 183 of 723
-------
Ar
CF
Equation 2-7
Area of water body catchment (m2)
Conversion of grams to micrograms
PondWaterConc =
AnnDep
Ar x Pond Depth
Where
PondWaterConc
AnnDep
Ar
Pond Depth
CF
Annual-average concentration in water body (|ig/kg)
Total annual deposition to water body (|ig)
Area of water body (m2); default = 10,000 m2 from EPA OPP
standard farm pond scenario
Depth of pond; default = 2 m from EPA OPP standard farm pond
scenario
Conversion of cubic meters to liters
Equation 2-8
PondSedimentConc =
AnnDep
Where
PondWaterConc =
AnnDep =
Ar =
Pond Depth =
Mix =
Dens =
Ar x Mix x Dens
Annual-average concentration in water body (|ig/kg)
Total annual deposition to water body (|ig)
Area of water body (m2); default = 10,000 m2 from EPA OPP
standard farm pond scenario
Depth of pond; default = 2 m from EPA OPP standard farm pond
Scenario
Mixing depth (m); default = 0.1 m
Density of sediment; default = 1,300 kg/m3 from the European
Commission Technical Guidance Document (ECB 2003)
The highest estimated surface water and sediment concentrations amongst all exposure scenarios is
provided in Table 2-63. Summary of Annualized Deposition and Estimated Pond Surface Water and
Sediment Concentration from Air Deposition for fenceline receptors (100 m from the source) and
"community" receptors beyond the fenceline (100 to 1,000 m from the source). For comparison, the
concentrations calculated from the average and high-end deposition from (Robson et al. 2013) is also
provided. The concentrations were in the same order of magnitude as the surface water and sediment
concentrations estimated using VVWM-PSC.
Page 184 of 723
-------
Table 2-63. Summary of Annualized Deposition and Estimated Pond Surface Water and Sediment
Concentration from Air Deposition
Scenario Name
Annualized
Deposition
(g/m2/y)
Estimated
Concentration in
Pond Water
(Jig/L)
Estimated
Concentration in
Pond Sediment
(jug/kg)
Highly Exposed Population - High-end
IIOAC-modeled fenceline
2.3E-05
1.1E-02
1.8E-01
Highly Exposed Population - High-end
IIOAC-modeled community
1.8E-06
8.7E-04
1.4E-02
Background - Hish-end from (Robson et
al. 2013)
3.0E-05
1.5E-02
2.3E-01
Backeround - Average from (Robson et
al. 2013)
8.4E-07
4.2E-04
6.5E-03
2.3.3 Terrestrial Environment - Soil
2.3.3.1 Non-Scenario Specific Approach - Air Deposition and Biosolid Application
This non-scenario specific approach uses measured media-specific monitoring data to characterize
background exposure to HBCD where releases attributed to historical and current conditions of use may
be encompassed. As described below in Section 2.3.3.2, all exposure scenarios with air releases have
predicted soil HBCD concentrations, except for the recycling of electronics waste containing HIPS and
land disposal of other formulated products and articles (e.g., adhesives, coatings, textiles, and
electronics). In regards to the recycling of electronics waste containing HIPs, a semi-quantitative
screening approach was used to compare industrial releases associated with this exposure scenario to
those of other exposure scenarios with air releases; the release days and amount of HBCD released were
factors considered to determine whether this exposure scenario will likely have soil concentrations of
HBCD that may exceed the chronic hazard threshold for earthworms. In regards to the land disposal of
textiles, electrical and electronic products, adhesives and coatings, in lieu of having media-specific
release information for this condition of use via leaching or surface runoff, background information
(monitoring data) is used as a proxy to characterize the risk to aquatic organisms.
EPA first evaluated environmental exposures to terrestrial organisms from soil based on environmental
monitoring data as opposed to modeled site-specific exposures or exposure scenarios. This non-
scenario-specific approach estimates background exposure from a multitude of different sources. The
totality of background exposure includes steady-state environmental exposures to ongoing releases not
associated with a particular COU, background/indirect exposures from minor use products (e.g., textiles,
electrical and electronic products, adhesives, and coatings) (Section 1.2.8), and releases stemming from
historical activities (Section 1.2.9) due to HBCD's persistence in the environment. For the non-scenario
specific approach, EPA estimated soil concentrations from two sources: air deposition and biosolids
application.
Air deposition
For air deposition, EPA identified and extracted measured concentrations of HBCD in soil from 21
primary source studies. This dataset includes samples collected between 1999 and 2015 in Belgium,
Cambodia, China, Indonesia, Sweden, and Vietnam. No U.S. studies were identified. A summary of
Page 185 of 723
-------
occurrence of HBCD in soil is presented in the Risk Evaluation for Cyclic Aliphatic Bromide Cluster
(HBCD), Supplemental Information on General Population, Environmental, and Consumer Exposure
Assessment (U.S. EPA 2019cT).
Following data aggregation and statistical analysis procedures, the surrogate country datasets included
only one study for near facility (point source) characterization and two studies for background (non-
point source) characterization. The near facility study was (Remberger et al. 2004). in which three soil
samples were collected between 300 and 700 m from a flame retardant XPS plastic production facility in
Sweden, with concentrations ranging from 140 to 1,300 |ig/kg dw (calculated central and high-end
values of 1,016 and 1,254 |ig/kg dw). The two background studies were Covaci et al. 2009 and Newton
et al. 2015. In Covaci et al. 2009. soil samples were collected in the perimeter of a home chicken run in
Belgium. In Newton et al. 2015. soil samples were collected in undisturbed rural and urban areas in
Stockholm, Sweden. The estimated central and high-end values from these studies are 1.4 and 3.0 |ig/kg
dw, respectively.
Most soil studies were collected in China. (Wu et al. 2016b) reported soil concentrations ranging from
0.3 to 249 |ig/kg dw, with a median of 5.14 |ig/kg dw, from samples collected in 2012 in areas that
represented a wide variety of land-use types. The soil concentration was influenced by the sample depth
as well as proximity to facilities, with higher concentrations reported near industrial areas. In another
Chinese study, (Tang et al. 2014) collected 90 samples across in residential and agricultural areas across
the Ningbo Region of China. The overall range of soil concentrations reported was ND (<0.068 |ig/kg)
in farmland areas to 103 |ig/kg in industrial areas; land-use highly influenced the overall magnitude of
reported soil concentrations.
Table 2-64. Summary of Central Tendency and High-End Estimated Soil Concentrations from
Monitoring Data
Site Characterization
Number of
Studies
Identified
Number of
Studies
Included in
Final Dataset
Estimated Concentrations (jig/kg)
Central Tendency
High-End
Near Industrial Facility
(Point Source)
9
V
1,016
1,254
Away from Facility
(Non-Point Source)
17
2b
1.4
3.0
a Near industrial facility studies: (Remberger et al. 2004)
bAway from facility studies: ("Covaci et al. 2009: Newton et al. 2015)
Biosolid application
EPA assumes that HBCD that may be deposited to soil through application of biosolids to agricultural
lands. EPA identified and extracted sludge concentrations from 17 studies. Overall, samples were
collected between 2000 and 2016 from Australia, Canada, China, Czech Republic, Indonesia,
Netherlands, South Korea, Spain, Sweden, Switzerland, and the United States. A summary of
occurrence of HBCD in biosolids is presented in the Risk Evaluation for Cyclic Aliphatic Bromide
Cluster (HBCD), Supplemental Information on General Population, Environmental, and Consumer
Exposure Assessment (U.S. EPA 2019d).
Two U.S. studies were identified. Venkatesan (Venkatesan and Halden 2014) reported a concentration
of 19.8 (J,g/kg dw in a single composite sewage sludge sample representing 94 WWTP in 32 U.S. states.
Page 186 of 723
-------
The samples were collected for EPA's 2001 national sewage sludge survey (NSSS). La Guardia (La
Guardia et al. 2010) collected secondary sewage sludge samples from a drying lagoon in 2002, 2005,
2007, and 2008 from one publicly owned WWTP in the Mid-Atlantic U.S. The facility treated domestic
and industrial waste, including discharges from an automobile interior manufacturer, although the
manufacturer relocated from the area in mid-2006. Only one sample, consisting of several grab samples
combined, was analyzed each year. Total HBCD concentrations corrected for TOC content (7 to 28%)
were 324, 400,000, 23125, and 3,171 |ig/kg dw, with a geometric mean concentration of 100,000 |ig/kg
dw (10 mg/kg). These concentrations are several orders of magnitude higher than the levels reported in
Venkatesan (Venkatesan and Hal den 2.014). presumably due to the industrial nature of the waste
received at the WWTP.
To assess soil concentrations resulting from biosolid applications, EPA relied upon modeling work
conducted in Canada (EC/HC ), which used Equation 60 of the European Commission Technical
Guidance Document (TGD) (ECB 2003). The equation in the TGD is as follows:
Equation 2-9
___ _ ^sludge ^ sludge
SOil ~ Dsoli x BDsoll
where:
PECsoii = Predicted environmental concentration (PEC) for soil (mg/kg)
Csludge = concentration in sludge (mg/kg)
ARsiudge = application rate to sludge amended soils (kg/m2/yr); default = 0.5 from Table A-l 1 of
TGD
Dsoii = depth of soil tillage (m); default = 0.2 m in agricultural soil and 0.1 m in pastureland
from Table A-l 1 of TGD
BDsoii = bulk density of soil (kg/m3); default = 1,700 kg/m3 from Section 2.3.4 of TGD
The concentration in sludge was assumed to 10 mg/kg dw based on (La Guardia et al. 2012). which was
the value also used in the Canadian assessment (EC/HC 2011). Using these assumptions, the estimated
soil concentrations after the first year of application were 15 |ig/kg in tilled agricultural soil and 30
|ig/kg in pastureland.
A limitation of Equation 2-9 is that it assumes no losses from transformation, degradation, volatilization,
erosion or leaching to lower soil layers. Additionally, it is assumed there is no input of HBCD from
atmospheric deposition and there are no background HBCD accumulations in the soil.
2,3.3,2 Scenario Specific Approach - Air Deposition
Soil concentrations from air deposition were also estimated for the condition of use scenarios with air
releases. The air deposition modeling was conducted using IIOAC. A description of the modeling and
the deposition results is provided above in Section 2.3.2.2.3. For comparison, EPA also reviewed
deposition from (Robson et al. 2013). assuming that the observed per-day fluxes from were held
constant for a year. Using the deposition rates, the HBCD concentration in soil was calculated with the
following equations:
Equation 2-10
AnnDep = TotDep xArxCF
Page 187 of 723
-------
Where
AnnDep
TotDep
Ar
CF
Total annual deposition to soil (|ig)
Annual deposition flux to soil (g/m2)
Area of soil (m2)
Conversion of grams to micrograms
Equation 2-11
SollConc =
AnnDep
Ar x Mix x Dens
Where
SoilConc
AnnDep
Mix
Annual-average concentration in soil (|ig/kg)
Total annual deposition to soil (|ig)
Mixing depth (m); default = 0.1 m from the European Commission
Technical Guidance Document (ECB 2003)
Area of soil (m2)
Density of soil; default = 1,700 kg/m3 from the European Commission
Technical Guidance Document (ECB 2003)
Ar
Dens
The above equations assume instantaneous mixing with no degradation or other means of chemical
reduction in soil over time and that HBCD loading in soil is only from direct air-to-surface deposition
(i.e., no runoff).
Table Apx F-5 in the Environmental Exposure appendix presents the range of calculated soil
concentrations corresponding to the emission scenarios considered. From the table, the highest estimated
soil concentration amongst all exposure scenarios was:
• 1.34E-01 |ig/kg at the fenceline (100 m from the source); and
• 1.03E-02 |ig/kg at "community" receptors beyond the fenceline (100 to 1,000 m from the
source).
These soil concentrations can be compared to results obtained when background deposition rates from
(Robson et al. 2013) are used:
• 1.76E-01 |ig/kg based on the maximum background deposition from (Robson et al.: ); and
• 4.94E-03 |ig/kg based on the average background deposition from (Robson et al. 2013).
Among the deposition scenarios modeled with IIOAC, the community receptors are likely more
appropriate for typical exposure-assessment purposes, which consider locations where the public would
have regular access (the IIOAC community receptors are within 1 kilometer from the facility). The
spatial averages provided by the community receptors are also more appropriate to use for deposition to
areas of soil since they cover a larger surface area. The highest IIOAC-modeled deposition at the
community receptors is nearly a factor of 5 above the average "background" value observed in the
monitoring study of (Robson et al. 2013). Differences in HBCD concentrations in soil are proportional
to differences in deposition. It is logical that the high-end modeled values of deposition and soil
concentrations near a facility, averaged over a year, are substantially higher than long-term-averaged
values resulting from general transport. Remaining IIOAC deposition rates are comparable with the
Page 188 of 723
-------
reported by (Robson et al. 2013). Table 2-65 summarizes the total annual deposition rates and
corresponding soil concentrations.
Table 2-65. Summary of Annualized Deposition and Estimated Soil Concentration from Air
deposition
Scenario Name
Annualized
Deposition (g/m2/y)
Estimated
Concentration in Soil
(jug/kg)
Highly Exposed Population - High-end IIOAC-
modeled fenceline
2.3E-05
1.3E-01
Highly Exposed Population - High-end IIOAC-
modeled community
1.8E-06
1.0E-02
Background - High-end from (Robson et al. 2013)
3.0E-05
1.8E-01
Background - Average from (Robson et al. 2013)
8.4E-07
4.9E-03
Screening Approach Used to Characterize Exposure for the Recycling of Electronics Waste Containing
HIPs
EPA estimated central tendency and high-end air releases of HBCD from electronic recycling sites to be
0.024 and 0.38 kg/site-d, respectively, for a duration of 250 days. EPA compared the air release
estimates for electronic recycling sites to those that were previously used to quantify HBCD soil
concentration (via air deposition) for releases associated with other conditions of use (Appendix F.1.2).
The daily release amounts of HBCD and number of release days estimated for electronic recycling sites
fall within the range as those used to characterize and estimate soil HBCD concentrations from air
deposition for other conditions of uses. Specifically, in comparison to exposure scenario 6.12, where the
daily release of HBCD (3.8 kg/site-d) and number of release days (300 days) are both higher than those
predicted for electronic recycling sites, the resulting soil HBCD concentration for exposure scenario
6.12 is 3.66E-03 |ig/kg for fenceline communities (near industrial facilities). This exposure scenario's
estimated soil concentration of HBCD does not surpass the hazard threshold for soil organisms (173,000
|ig /kg). Due to the unlikelihood that the lower release amounts and days for electronic recycling sites
will surpass those used for any of the other conditions of use, soil concentrations of HBCD due to air
deposition were not estimated using methods outlined above in Appendix F.1.2 for the exposure
scenario regarding the recycling of electronics waste containing HIPs.
2.3.3.3 Combined Soil Concentration - Air Deposition, Background, Biosolid
Application
The overall magnitude of the contribution of air deposition to soil concentrations is generally low, <1
|ig/kg for all scenarios considered. Further, background soil concentrations based on the soil monitoring
were below 10 |ig/kg. Therefore, an estimated high-end soil concentration of HBCD from all sources,
including biosolids application (30 |ig/kg), air deposition (1 |ig/kg), and background (10 |ig/kg) would
be slightly higher (41 |ig/kg) than potential soil concentrations from any of these individual sources.
2.3.4 Assessment of Exposure in Targeted Wildlife
There are several biomonitoring studies examining the occurrence of HBCD in a wide range of wildlife
biota across multiple trophic levels. Most of the wildlife biomonitoring samples report HBCD in lipid
weight, but some are reported in wet weight. Some studies describe temporal, spatial (Esslinger et al.
2011a). and trophic level (Poma et al. 2014) trends of HBCD concentrations in biota. A summary of
occurrence of HBCD in aquatic and terrestrial biota is presented in Sections 4.1.1 and 4.2.1 of the Risk
Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on General
Page 189 of 723
-------
Population, Environmental, and Consumer Exposure Assessment. Monitoring data was extracted for a
variety of wildlife, including amphibians, aquatic invertebrate, aquatic mammals, birds, fish, terrestrial
mammals, and vegetation.
Certain studies demonstrate that wildlife are more highly exposed when they are close to point sources
i.e., certain species that live near effluent discharge sites (Haukas et al. 2010a). Due to HBCD's
persistence and potential for long-range transport (TJNEP 2010b). exposure to wildlife is expected, at
some level, to continue even as current releases to the environment decline.
2.3.5 Summary of Results for Environmental Exposure Assessment
The monitoring and modeling data presented in the preceding sections is summarized in Table 2-66.
Values with an asterisk indicate that the value was carried forward to risk estimation. A comparison of
the near-facility monitoring concentrations with the scenario-specific modeled concentrations based on
estimated release data indicate general agreement of data. While a meta-analysis using raw data would
have provided a more robust approach, raw data was generally not available for most studies.
Table 2-66. Comparison of Published Literature and Modeling Results for Concentrations of
HBCD in Surface Water, Sediment, and Soil
Data Type
Environmental Media
Point
Source
Proximity
Surface Water
Concentration
(Hg/L)
Sediment Concentration
(Mg/kg)
Soil Concentration
(Mg/kg)
Modeled Estimates
E-FAST modeled
estimates (50th low flow)
8.3E-05 - 1.1E+02
NA
NA
Near
(Scenario-
Specific)
E-FAST modeled
estimates (10th low flow)
4.2E-03 - 5.3E+03
NA
NA
Near
(Scenario-
Specific)
WWM-PSC modeled
estimates (50th low flow)
*21-Day Average-
Dissolved:
1.1E-04 - 3.8E+00
*28-Day Average:
1.2E-01 - 1.2E+04
NA
Near
(Scenario-
Specific)
WWM-PSC modeled
estimates (10th low flow)
*21-Day Average
Dissolved:
9.2E-03 - 1.8E+02
*28-Day Average:
1.1E-00 - 5.7 E+05
NA
Near
(Scenario-
Specific)
IIOAC modeled
(Deposition from air)
<1
<1
*<1
Near
(Scenario-
Specific)
Robsonet al. (2013)
(Deposition from air)
<1
<1
<1
Far
Biosolid Application
NA
NA
*30
Near
Modeled Estimates from
(EC/HC 2011)- 100 m
from facility
Raw Materials Handling
Near
5.0E-01 - 1.5E+01
3.6E+03 - 1.8E+05
NA
Compounding
1.0E-01 - 1.3E+00
3.3E+02 - 9.9E+03
NA
Page 190 of 723
-------
Modeled Estimates from
CEC/HC 201D - 5 km
from facility
Raw Materials Handling
Far
3.0E-01 - 1.0E+01
2.6E+03 - 7.7E+04
NA
Compounding
3.0E-02 - 9.0E-01
2.3E+02 - 7.0E+03
NA
Modeled Estimates from
cec 2008)
4.8E-5 - 3.7E+02
1.0E-3 - 4.0E+06
4.5E-1 - 9.1E+04
Near
Monitoring Data
All Extracted Data
High income:
2.5E-3 - 3.1E+00 (n= 3)
Non-high income:
6.0E-5 - 1.8E+00 (n=2)
High income:
7.5E-02 - 8.5E+04 (n=ll)
Non-high income:
5.0E-03 - 2.75E+04 (n=4)
High income:
1.4E+2- 1.3E+03 (n=l)
Non-high income:
5.0E-03 - 3.2E+04 (n=8)
Near
High income:
2.0E-7 - 6.7E-03 (n=6)
Non-high income:
9.5E-06 - 1.6E-03 (n=3)
High income:
2.0E-3 - 1.7E+03 (n=32)
Non-high income:
2.0E-03 - 1.0E+03 (n=13)
High income:
1.8E-1 - 1.0E+2 (n=3)
Non-high income:
4.0E-03 - 1.7E+03 (n=14)
Far
Final Extracted Dataset
(following statistical
analysis procedures)
High income:
2.5E-3 - 3.1E+00 (n=3)
*CT: 8.4E-01
*HE: 9.9E-01
High income:
5.0E-01 - 8.5E+04 (n=6)
*CT: 3.4E+03
*HE: 5.1E+03
High income:
1.4E+2- 1.3E+3 (n=l)
*CT: 1.0E+03
*HE: 1.3E+03
Near
High income:
2.0E-7 - 6.7E-03 (n=4)
*CT: 4.1E-04
*HE: 8.0E-04
High income:
2.2E-2- 1.7+03 (n=14)
*CT: 6.0E+00
*HE: 2.0E+01
High income:
1.8E-1 - 1.2E+01 (n=2)
*CT: 1.4E+00
*HE: 3.0E+00
Far
Asterisk (*) indicates values used in exposure estimates for risk estimation
NA = not available; CT = central tendency; HE = high-end
Page 191 of 723
-------
2.3.6 Sensitivity Analysis - Environmental Exposure
2.3.6.1_ Modeled Sediment
For estimated sediment concentrations from VVWM-PSC (Section 2.3.2.2.2), the default values, such as
suspended sediment concentration, fraction organic content, chlorophyll, and biomass content also
influence distribution. A targeted sensitivity analysis showed that Koc, half-life in sediment, fraction
organic content, and suspended solids concentration are parameters that tend to have more of an impact
on sediment concentrations. EPA considered variation of some of the more sensitive parameters, but
found that results using different inputs, showed similar magnitude and trends as the results presented.
This is likely because changes in of multiple parameters may have offset the impact of other parameters.
2.3.6.2 Monitoring Data (General)
Table 2-67 summarizes the sensitivity analysis associated with monitoring data. Potential variability in
the assumption that the central tendency estimate of the reported monitoring data represent the
geometric mean appear to have a limited impact on the estimate of the high-end (95th percentile) dose.
Increasing the geometric mean by 10% over the baseline value increased high-end dose by 4%, while
decreasing it by 10% decreased dose by 7%.
Table 2-67. Sensitivity Analysis of Central Tendency Estimate Assumptions in Monitoring Data
Est
Baseline GM
imated Dose in mg/kg/i
Baseline GM + 10%
lay
Baseline GM -10%
95th Percentile Dose
3.1E-04
3.2E-04
2.9E-04
% Change from Baseline
—
4%
-7%
GM = geometric mean
2.3.6.3 Fish Tissue
For fish tissue concentrations (Section 2.4.2), a wide range of BCF and BAF values are available in the
literature. Generally, BCF and BAF values are highly sensitive to variability in measured input values
(dissolved surface water concentration, lipid weight fish tissue concentration, and fraction lipid-content).
Small changes in these input values can result in large changes in associated BCF and BAF values.
2.3.6.4 Scenario Inputs (product amount, WWTR%)
As described in Section 2.2.15, EPA performed sensitivity analyses for three conditions of use at the per
site process volumes of 50,000 lbs/yr and 25,000 lbs/yr to examine the effect of process volume on the
resulting general population and environmental exposures. In addition, EPA chose to perform additional
sensitivity analyses by incorporating a higher onsite (direct release) wastewater removal when the
removal rates were unknown. For Scenario 1 (Repackaging of Import Containers), based on information
provided in Section 2.2.2, EPA applied 90% removal for releases to water. As mentioned in Section
2.3.2, when information regarding pretreatment for direct releases to surface was uncertain, EPA applied
a removal rate of 0%. In the sensitivity analysis presented here, a tiered approach was used to assess
these releases using both 0% removal and a higher removal rate.
Little information was found on the type or efficiency of onsite treatment used by direct discharging
facilities using HBCD. Due to its low water solubility (66 |ig/L), high log Kow (5.6) and physical state
(solid), HBCD is likely to partition to the organic phase, including organic particulates such as activated
sludge in biological wastewater treatment systems. At concentrations above its water solubility it is
expected to behave as a particulate in aqueous wastewater and be removed with other solids by gravity
settling during the wastewater clarification process. The efficiency of removal of HBCD may be
reflected in data for total suspended solids (TSS) removal.
Page 192 of 723
-------
HBCD processing and use activities described in this Risk Evaluation may be subject to the Organic
Chemicals and Plastics and Synthetic Fibers (OCPSF) Category Effluent Limitations Guidelines,
Pretreatment Standards, and New Source Performance Standards 40 CFR Parts 414 and 416 [FRL 3230-
5], The OCPSF limitations and standards establish effluent limitations guidelines and standards that
limit the discharge of pollutants into navigable waters and publicly owned treatment works (POTWs) by
existing and new sources in the Organic Chemicals, Plastics, and Synthetic Fibers (OCPSF) industrial
category. End-of -pipe biological treatment direct dischargers who are subject to subpart E of the
regulations must meet relevant discharge limits of priority pollutants. A facility may meet their limits by
virtue of the absence of a regulated pollutant in their process wastewater as confirmed by monitoring, or
the use of engineering controls or installation of end-of-pipe biological treatment. Where present and
properly maintained and operated, this type of treatment has been shown to remove chemicals with
similar tendency to sorb to sludge as that of HBCD (log Kow 5.6), examples include benz(a)anthracene
(log Kow 5.8), benz(a) pyrene (log Kow 6.1), and fluoranthene (log Kow 5.8). The EPA Development
Document for Effluent Limitations, Guidelines and Standards for Organic Chemicals, Plastics and
Synthetic Fibers Point Source Category ( 7) reported that the majority of the facilities in
the OCPSF category responding to the EPA 308 survey reported using the activated sludge treatment
process to treat their process wastewater. TSS removal in activated sludge treatment was reported by the
responding facilities with a of mean (67%), a median (81%), a minimum (-29%) and a maximum (99%)
for thirty nine observations.
HBCD may be released to wastewater incorporated into polystyrene particles. These particles may fall
into the range of "microplastics" <5 mm in diameter (Conlev et al. 2019). A number of studies have
demonstrated high removal of HBCD and microplastics in activated sludge treatment (Conlev et al.
2019). determined the microplastic (synthetic polymer materials <5mm in size) loads and removal
efficiencies of three activated sludge wastewater treatment plants (WWTPs) with different treatment
sizes, operations and service compositions discharging to Charleston Harbor, South Carolina, over the
course of a year. Microplastics concentrations (counts per L) varied within a factor of 2.5 in influent and
4.8 in effluent at each WWTP, and that neither concentrations nor removal efficiencies demonstrated a
seasonal trend. The largest wastewater treatment plant in the study, which also employed primary
clarification, had the highest MP removal efficiency of 97.6 ± 1.2%. The other two smaller facilities had
average removal efficiencies of 85.2 ± 6.0% and 85.5 ± 9.1%. Ruan (Ruan et al. 2019) investigated the
removal of microplastics and HBCD levels in microplastics at two Hong Kong wastewater treatment
plants. One plant employed primary treatment while the second plant utilized secondary treatment.
Greater than 90% removal of HBCD was observed in both plants. Approximately 60% and 87% removal
of microplastics occurred in the primary and secondary treatment systems, respectively.
Sun (S um et al.! ) conducted a comprehensive review of studies on the detection, occurrence and
removal of microplastics in WWTPs. The review included techniques used for collecting microplastics
from both wastewater and sewage sludge, and their pretreatment and characterization methods.
Microplastics removal in various stages of wastewater treatment and their retention in sewage sludge
were explored. Overall percent removals in secondary wastewater treatment from 7 studies conducted in
the U.S. and Europe were reported. Microplastics removal efficiencies ranged from 72 to 99% with a
mean value across all the studies of removal in secondary treatment of 92%. Carr (Carr et al. 2016)
conducted microplastics bench scale wastewater treatment simulations and studied effluent discharges
from seven tertiary and one secondary wastewater treatment plant in Southern California to determine
the fate of microplastics in these systems. The results of bench scale experiments with activated sludge
and raw wastewater, simulated high solids influent and gravity filtration suggested that the buoyancy of
Page 193 of 723
-------
microplastics facilitates removal by surface skimming, entrapment in influent suspended solids
facilitates removal by solids settling, and high retention of microplastics on typical gravity bed filter
materials leads to potential for high removal in secondary and tertiary wastewater treatment. Analysis of
influent and effluent samples for microplastic particles at both secondary and tertiary treatment plants
indicated removal >= 99%.
EPA considered these reported values and uncertainty in extrapolating from performance of the
treatment systems surveyed in the Effluent Guidelines document to those facilities using HBCD. EPA
also considered uncertainty associated with the use of TSS removal and microplastics removal as a
surrogate for HBCD removal. EPA selected 90% removal of HBCD in wastewater treatment for direct
dischargers. EPA is confident that some removal of HBCD will occur in onsite wastewater treatment.
Higher or lower removal of HBCD could occur based on the type of treatment employed and its
performance optimization.
EPA acknowledges the downward trend of environmental releases as the production volume of HBCD
has decreased over time. To account for this, EPA considered three separate estimates of releases for
conditions of use based on three different production volumes: 100,000, 50,000, and 25,000 kilograms
per year. EPA estimated surface water and sediment concentrations through the Point Source Calculator
for all combinations. EPA inferred that the days of release correlated with kg/site/day releases. For
example, as total releases decrease, the number of days of release also decrease. For this reason, any 1-
day surface water concentrations are approximately equal. Both the overall magnitude of the release and
the number of days of release influence estimated concentrations. When the overall magnitude of the
release is reduced by a factor of two or four, the corresponding environmental concentration is also
reduced by approximately a factor of two or four. When the number of days are reduced by factor of two
or four, the corresponding environmental concentration is reduced, however, the trend is not linear and
depends on the number of days of release. This is due to uncertainty in the timing of the release days and
the selected averaging periods (21-days for surface water and 28 days for sediment), 21-day average
water concentrations and 28-day average sediment concentrations are more sensitive to changes in
release estimates. EPA inferred that the release days occur intermittently rather than continuously
through the year. The timing of these releases, in addition to the number of release days, influence
potential exposure concentrations. EPA also varied other parameters in its surface water modeling that
have a large impact on estimated results. The selected flow values for mean-flow or low flow are highly
sensitive. EPA used a central tendency and a high-end estimate for each of these flow metrics. Estimated
sediment concentrations are highly sensitive to the sediment half-life used; hence, EPA used central
tendency and high-end estimates for sediment half-life in calculating sediment concentrations. Because
the percent removal of HBCD from different removal processes is likely variable, EPA also varied
percent removal expected based on three scenarios: on-site treatment (pre-treatment) [0%] and on-site
wastewater treatment plants [90%]. Some release estimates already account for treatment while others
do not. The efficiency of treatment across different industrial facilities and different wastewater
treatment plants will also vary.
Page 194 of 723
-------
Table 2-68. Summary of HBCD Surface Water Concentrations from Sensitivity Analysis: Varying
Production Volume and Waste Water Treatment Removal- Environmental Exposures
SCENARIO NAME
Production
Volume
(lbs / year)
% WW TP
Removal for
Direct Releases3
Surface Water 1-Day
Average Concentration
Range (ug/L)
Sediment
Acute:
50th %-ile
Chronic:
50th %-ile
11-d half-life:
50th %-ile
128-d half-life:
50th %-ile
Scenario 1. Import
and Re-packaging/
Processing:
Repackaging of
Import Containers
100,000
90%
3.7E-02 -
9.7E+00
3.0E-02 -
9.4E-01
3.4E+01 -
8.7E+02
7.7E+01 -
2.0E+03
50,000
90%
3.7E-02 -
9.4E+00
1.8E-02 -
5.0E-01
1.9E+01 -
5.4E+02
4.1E+01 -
1.2E+03
25,000
90%
3.7E-02 -
1.0E+01
8.8E-03 -
4.8E-01
8.5E+00 -
3.2E+02
1.9E+01 -
6.3E+02
Scenario 3. Processing:
Manufacturing of XPS
Foam using XPS
Masterbatch
100,000
0%
8.0E-03 -
2.9E+00
3.8E-04 -
1.4E-01
4.0E-01-
8.9E+01
8.9E-01 -
1.2E+02
50,000
0%
4.0E-03 -
1.5E+00
1.9E-04 -
7.1E-02
2.0E-01 -
4.5E+01
4.4E-01 -
6.0E+01
25,000
0%
2.0E-03 -
7.4E-01
3.8E-04 -
1.4E-01
1.0E-01 -
2.3E+01
2.2E-01 -
3.0E+01
Scenario 5.
Processing:
Manufacturing of EPS
Foam from Imported
EPS Resin Beads
100,000
0%
8.8E-01-
1.1E+02
2.9E-01 -
5.0E+00
3.3E+02-
5.5E+03
7.6E+02 -
1.2E+04
50,000
0%
4.4E-01 -
1.1E+02
1.5E-01 -
5.0E+00
1.7E+02 -
3.5E+03
3.8E+02 -
6.9E+03
25,000
0%
2.2E-01 -
1.1E+02
7.4E-02 -
5.0E+00
8.4E+01 -
3.2E+03
1.9E+02 -
4.9E+03
a There are no predicted direct releases for Scenario 1.
Page 195 of 723
-------
2.3.7 Assumptions and Key Sources of Uncertainty in Environmental Exposure
Assessment
Concentrations of HBCD in environmental and biological media are expected to vary. Close proximity
to facilities and other sources is likely to lead to elevated concentrations compared to locations which
are more remote. A combination of monitoring data from the U.S. and international sources were used in
this exposure assessment. In addition, monitoring data were collected in previous years when production
volume and associated releases of HBCD into the environment are expected to have been higher than
they are currently and expected to be in the future. When considering older monitoring data and
monitoring data from international sources, there are uncertainties associated with using these data
because it is unknown whether those sampling sites are representative of current sites within the U.S.
In modeling environmental concentrations of HBCD, EPA acknowledges the conservative nature of the
E-FAST model and the additional refinement provided by the PSC model. Water dilution models can be
used to determine the concentration of a chemical in the surface water after a source emits the chemical
into a water body. Since the E-FAST model default values encompass either a combination of upper
percentile and mean exposure parametric values, or all upper percentile parametric values, the resulting
model predictions represent high-end exposures estimates. A simple dilution model, such as EFAST,
provides exposure estimates that are derived from a simple mass balance approach, and does not account
for partitioning between compartments within a surface water body or degradation over time in different
media, parameters which are relevant to HBCD. For these reasons, EPA utilized a two-tier approach by
complementing the EFAST modeling with more refined estimates from the PSC model to further
describe environmental exposures.
When modeling using E-FAST, EPA assumed that primary treatment removal at POTWs occurred with
90% removal efficiency, however for direct discharges, EPA used 0% removal. EPA recognizes that this
is a conservative assumption that results in no removal of HBCD prior to release to surface water. This
assumption will give higher surface water and sediment concentrations compared to a removal
efficiency of 75 or 90% removal. This assumption reflects both the uncertainty of the type of wastewater
treatment that may be in use at a direct discharging facility and the HBCD removal efficiency in that
treatment. It is likely that under the COUs for HBCD, a facility's wastewater discharge is required to
meet NPDES discharge permit limits for total suspended solids, five-day biochemical oxygen demand
(BODs) and other wastewater treatment parameters. Treatment methods used to meet the limits (such as
activated sludge treatment) will likely also remove HBCD from wastewater to an uncertain, but non-
zero, extent due to the properties of HBCD.
EPA used a combination of chemical-specific parameters and generic default parameters when
estimating surface water, sediment, soil, and fish-tissue concentrations. For estimated soil concentrations
from biosolid application, specifically, EPA recognizes that different default parameters for tillage depth
and application rates are used by other U.S. agencies which may result in concentrations of a higher
magnitude. However, EPA used both central tendency and high-end values across model inputs to
characterize the variability within and across scenarios. EPA also used central tendency and high-end
model outputs. Comparison of model outputs with monitored values offers one way to ground-truth the
combination of model inputs and outputs used. EPA compared monitoring and modeled surface water,
sediment, soil, and fish-tissue concentration estimates. Estimates of fish-tissue concentrations are further
discussed in Section 2.4.3. In summary, EPA compared monitored and modeled fish tissue
concentrations using modeled 21-day average dissolved water concentrations and low-end BAF values
and found overlap and concordance between these values and fish-tissue monitoring data. When
Page 196 of 723
-------
modeling the HBCD concentrations in water and sediment, EPA did not consider the potential impact of
persistence and longer-term sinks in lake and estuary environments.
Recent and future estimated levels of HBCD in the area may be lower than past levels due to reported
reductions in releases over time. EPA assessed more recent releases. The predicted concentrations may
be lower than concentrations that consider more years of releases or releases associated with higher
production volumes.
2.4 Human Exposures
EPA considered four different receptor groups for the human exposure assessment: occupational,
general population, highly exposed, and consumers. The receptor groups were defined as:
• Occupational include individuals who work at a facility handling HBCD (workers) and
occupational non-users (ONU) who do not directly handle the chemical but perform work in an
area where the chemical is present.
• General population include individuals who are not expected to live close to point sources of
HBCD (far-field) and do not have a specific HBCD source within a living environment that has
been assessed by EPA in the consumer exposure assessment (i.e., home insulation, auto-
components, mouthing of recycled products). The general population experiences steady-state
chronic exposures resulting in risk from sustained background exposure in the environment due
to HBCD persistence.
• Highly exposed include individuals who are expected to live close by point sources of HBCD.
• Consumers include individuals who have articles containing HBCD in their homes or
automobiles.
A slightly different approach was used for each receptor group based on the exposure media/pathways
and available data. It is possible for an individual to fall into multiple receptor and potentially exposed
groups.
For all receptor groups, except general population, EPA developed scenario specific exposure estimates
based on condition of use (COU) release estimates described in Section 2.2. These exposures occur at or
near point sources (i.e., facilities that process, use, or dispose of HBCD or HBCD-containing materials)
or involve the use of articles containing HBCD. General population exposures estimates are non-
scenario specific in that they are based on media concentrations not related to a specific COU release
estimate (i.e., background or far from facility releases). HBCD exposures to the general population may
be variable as they are influenced by both sources into the environment, degradation and removal from
the environment. Estimates of general population exposures based on environmental monitoring and
biomonitoring data represent the conditions present at the time the data was collected. It is unknown
which combination of potential sources associated with conditions of use as described in this risk
assessment contribute to the monitoring data presented here. However, given the wide range of
exposures shown within and across the monitoring data, there is a plausible contribution from some of
the sources/conditions of use described within this document. Scenario-specific modeled releases for
individual exposure pathways (e.g., fish ingestion) were added to the aggregate background exposure
from all other pathways (i.e., all exposure pathways except fish ingestion). Exposures were not
aggregated within a particular exposure route across both biomonitored and modeled estimates.
Figure 2-2 shows the exposure pathways/media identified for each receptor group, and the assessment
approaches are further shown in Table 2-69.
Page 197 of 723
-------
Jit
m
iif
^ i®1:
Occupational
General Population
Highly Exposed
Consumers
At Facility
Far from Facility/Background
Near Facility
Near Consumer Products
Scenario Specific
Non-Scenario Specific
Scenario Specific
Scenario Specific
Inhalation
Dermal
Ingestion: Diet, Dust & Soil
Inhalation: Indoor & Outdoor Air
Dermal
Ingestion: Fish
Inhalation: Indoor Air
Inhalation: Indoor Air
Ingestion: Dust &
Mouthing Recycled Articles
Figure 2-2. Overview of receptor groups considered within the Risk Evaluation.
Table 2-69. Exposure Scenarios Descriptions for Receptor Groups
Scenario
Receptor
Group
Source
Pathway
Media
Approach
Approach Description
OES 1-13
Worker
HBCD
Inhalation
Dermal
Indoor
Air/Personal Air
Quantitative
Monitoring, Modeling,
Occupational Exposure
Limits
OES 1-13
ONU
HBCD
Inhalation
Dermal
Indoor
Air/Personal Air
Qualitative
Not Applicable
G1
General
Population
HBCD
Ingestion
Inhalation
Dermal
Diet, Dust, Soil,
Indoor Air,
Outdoor Air
Quantitative
Monitoring: Indirect
Estimation and Exposure
Reconstruction
HI
Highly
Exposed
HBCD emitted
from any point
source during its
lifecycle from
Scenarios described
in Section 2.2
Ingestion
Fish Tissue:
Emission into
water and uptake
into fish tissue
Quantitative
Modeling with PSC
combined with and Lipid
Normalized Upper
Trophic Level BAF
(monitoring). Monitoring
H2
Highly
Exposed
HBCD emitted
from any point
source during its
lifecycle from
Scenarios described
in Section 2.2
Inhalation
Air: Emission to
air and subsequent
inhalation of
particles
Quantitative
Modeling with IIOAC
CI
Consumers
XPS/EPS insulation
in residences
Inhalation
Indoor Air and
Dust: Emission
from insulation
into indoor air and
settled dust
Quantitative
Modeling with IECCU
C2
Consumers
HBCD contained in
automobile
components
Inhalation
Indoor Air and
Dust: Emission
into automobile
cabin air and
settled dust
Quantitative
Modeling with IECCU
Page 198 of 723
-------
Scenario
Receptor
Group
Source
Pathway
Media
Approach
Approach Description
C3
Consumers
Recycled consumer
articles that contain
HBCD
Ingestion
Articles:
Mouthing, direct
contact
Quantitative
Monitoring and Modeled
Qi
Highly
Exposed
EPS and XPS
insulation in
buildings during
use
Inhalation
Air: Emission
from building
interior to ambient
air surrounding
buildings
Qualitative
N/A
Q2
Highly
Exposed
Birds
HBCD sent to
landfill across the
lifecycle
Inhalation
Ingestion
Air, Soil, Water:
Comingled HBCD
containing
materials leach
into soil, disposed
food, and water
Qualitative
N/A
Occupational receptors are discussed first in Section 2.4.1. The section contains a detailed methodology
and approach for the enumeration of worker and ONUs, estimates of central and high-end inhalation and
dermal exposure for each of the thirteen conditions of use. EPA assessed exposure to male and female
workers including female workers of reproductive age of > 16 years to less than 50 years old, including
adolescents (16 to <21 years old). Adolescents are a small part of the total workforce (U.S. BLS. 2017).
Non-occupational receptors are discussed in Sections 2.4.2 (general population), Section 2.4.3 (highly
exposed), and Section 2.4.4 (consumer). Scenarios which were only qualitatively assessed are discussed
in Section 2.5.5. EPA assessed exposure to seven age groups, as appropriate: <1 year, l-<2 years, 2-<3
years, 3-<6 years, 6-
-------
recycling of these same articles throughout the United States suggest that there may be a continuing
sources for emission of HBCD extending into the future.
2.4.1 Occupational Exposures
EPA assessed workplace exposures pertaining to the following HBCD exposure scenarios:
• Repackaging of Import Containers
• Compounding of Polystyrene Resin to Produce XPS Masterbatch
• Processing of HBCD to Produce XPS Foam using XPS Masterbatch
• Processing of HBCD to Produce XPS Foam using HBCD powder
• Processing of HBCD to Produce EPS Foam from Imported EPS Resin Beads
• Processing of HBCD to Produce SIPs and Automobile Replacement Parts from XPS/EPS Foam
• Use: Installation of Automobile Replacement Parts
• Use: Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings,
and Other Structures
• Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures
• Recycling of EPS Foam and Reuse of XPS Foam
• Formulation of Flux/Solder Pastes
• Use of Flux/Solder Pastes
• Recycling of Electronics Waste (E-Waste) Containing HIPS
Components of the Occupational Exposure Assessment
The occupational exposure of each exposure scenario comprises the following components:
1. Number of Workers and Occupational Non-Users: An estimate of the number of workers and
occupational non-users (-workers, who do not directly handle the chemical but perform work in
an area where the chemical is present) potentially exposed to the chemical for the given exposure
scenario.
2. Inhalation Exposure: Central tendency and high-end estimates of inhalation exposure to
workers and occupational non-users. EPA assumes that all inhaled particulates are absorbed by
either the lung or intestine after ingestion as further discussed in Section 4.2.1.
3. Dermal Exposure: Estimates of dermal exposure to workers.
The process descriptions and facility estimates are included in Section 2.2 for each exposure scenario.
2.4.1.1 Occupational Exposures Approach and Methodology
Number of Workers and ONUs
Where available, EPA prefers to use CDR data to provide a basis to estimate the number of workers and
occupational non-users (ONUs). However, all companies that have historically reported HBCD
manufacturing and importation to CDR have ceased such operations. In lieu of current CDR data, EPA
used U.S. economic data to estimate the number of workers and ONUs using the following method:
• Identify the North American Industry Classification System (NAICS) codes for the industry
sectors associated with each exposure scenario.
• Estimate total employment by industry/occupation combination using the Bureau of Labor
Statistics' Occupational Employment Statistics data ( >16).
• Refine the occupational employment statistics estimates where they are not sufficiently
granular by using the U.S. Census' (2015) Statistics of U.S. Businesses (SUSB) data on total
employment by 6-digit NAICS.
• Estimate the number of potentially exposed employees per site (Census Bureau 2015).
Page 200 of 723
-------
• Estimate the number of potentially exposed employees within the exposure scenario, using
the number of sites estimated as described in Section 2.2.1.
EPA discussed the estimation of HBCD throughput and number of sites in Section 2.2.1.
EPA '.s General Approach to the Assessment of Inhalation Exposure
EPA provided occupational exposure results representative of central tendency conditions and high-end
conditions. A central tendency is assumed to be representative of occupational exposures in the center of
the distribution for a given exposure scenario. For Risk Evaluation, EPA may use the 50th percentile
(median), mean (arithmetic or geometric), mode, or midpoint values of a distribution as representative of
the central tendency scenario. EPA's preference is to provide the 50th percentile of the distribution.
However, if the full distribution is not known, EPA may assume that the mean, mode, or midpoint of the
distribution represents the central tendency depending on the statistics reported in the data source for the
distribution.
A high-end is assumed to be representative of occupational exposures that occur at probabilities above
the 90th percentile but below the exposure of the individual with the highest exposure ( i).
For Risk Evaluation, EPA plans to provide high-end results at the 95th percentile. If the 95th percentile is
not available, EPA may use a different percentile greater than or equal to the 90th percentile but less than
or equal to the 99.9th percentile, depending on the statistics available for the distribution. If the full
distribution is not known and the preferred statistics are not available, EPA may estimate a maximum or
bounding estimate in lieu of the high-end.
Exposures are calculated from datasets, comprised of data from one or more sources, depending on the
size of the dataset. For datasets with six or more data points, central tendency and high-end exposures
were estimated using the 50th percentile and 95th percentile. For datasets with three to five data points,
central tendency exposure was calculated using the 50th percentile and the maximum was presented as
the high-end exposure estimate. For datasets with two data points, the midpoint was presented as a
midpoint value and the higher of the two values was presented as a higher value. Finally, data sets with
only one data point presented the value as a what-if exposure. EPA did not have discrete data points for
the discussed monitoring data in this section. Only statistical summaries of the data sets were available,
and EPA did not combine or perform calculations with these reported statistics.
EPA follows the following hierarchy in selecting data and approaches for assessing inhalation
exposures:
1. Monitoring data:
a. Personal and directly applicable
b. Area and directly applicable
c. Personal and potentially applicable or similar
d. Area and potentially applicable or similar
2. Modeling approaches:
a. Surrogate monitoring data
b. Fundamental modeling approaches
c. Statistical regression modeling approaches
3. Occupational exposure limits:
Page 201 of 723
-------
a. Company-specific occupational exposure limits (OELs) (for site-specific exposure
assessments, e.g., there is only one processing site who provides to EPA their internal
OEL but does not provide monitoring data)
b. OSHA permissible exposure limit (PEL)
c. Voluntary limits (American Conference of Governmental Industrial Hygienists [ACGIH]
threshold limit value [TLV], NIOSH recommended exposure limit [REL], Occupational
Alliance for Risk Science [OARS] workplace environmental exposure level (WEEL)
[formerly by AIHA])
For occupational exposures, EPA used measured air concentrations, estimated air concentrations, or
occupational exposure limits to calculate exposure concentration metrics required for Risk Evaluation.
Specifically, EPA used these exposure concentration values to calculate acute exposure dose (AED) and
average daily dose (ADD). Additional explanation of the equations used to calculate AED and ADD,
and example calculations are located in Appendix E.3 and Appendix E.4, respectively. EPA then
multiplied the AED and ADD by the inhalation absorption factor of 100% (discussed in Section 3.2.2) to
estimate the acute absorbed dose (AAD) and chronic absorbed dose (CAD), respectively. The AED and
AAD are used to assess acute exposure risks. The ADD and CAD are used to assess chronic, non-cancer
risks. These calculations require additional parameter inputs, such as years of exposure, exposure
duration and frequency, and lifetime years.
For the final exposure result metrics, each of the input parameters (e.g., air concentrations, working
years, exposure frequency, lifetime years) may be a point estimate (i.e., a single descriptor or statistic,
such as central tendency or high-end) or a full distribution. EPA will consider three general approaches
for estimating the final exposure result metrics:
Deterministic calculations: EPA will use combinations of point estimates of each parameter to estimate a
central tendency and high-end for each final exposure metric result. EPA will document the method and
rationale for selecting parametric combinations to be representative of central tendency and high-end.
Probabilistic (stochastic) calculations: EPA will pursue Monte Carlo simulations using the full
distribution of each parameter to calculate a full distribution of the final exposure metric results and
selecting the 50th and 95th percentiles of this resulting distribution as the central tendency and high-end,
respectively.
Combination of deterministic and probabilistic calculations: EPA may have full distributions for some
parameters but point estimates of the remaining parameters. For example, EPA may pursue Monte Carlo
modeling to estimate exposure concentrations, but only have point estimates of working years of
exposure, exposure duration and frequency, and lifetime years. In this case, EPA will document the
approach and rationale for combining point estimates with distribution results for estimating central
tendency and high-end results.
EPA's determination of each of the input parameters for calculation of AED and ADD are explained in
Appendix E.3.
EPA quantitatively assessed exposure to male and female workers including female workers of
reproductive age of > 16 years to < 50 years old, which includes adolescents (16 to <21 years old). Male
adolescent workers are also potentially exposed to HBCD and their exposure dose (mg/kg-day) is in the
range assessed as their dose would be between estimates for average workers and female workers.
Page 202 of 723
-------
Adolescents (16 to < 21 years old) are a small part of the total workforce in the workplace (U.S. BLS.
2017).
EPA '.s Approach to the Assessment of HBCD Inhalation Exposure
EPA gathered and evaluated occupational exposure information in accordance with the process
described in the Application of Systematic Review in TSCA Risk Evaluations (U.S. EPA 2018b). The
results of EPA's systematic review include occupational monitoring data pertaining to the manufacture
and processing of HBCD in Europe. These data, which are presented in Appendix Appendix E, are
HBCD inhalation exposure concentration monitoring data pertaining to the manufacture and processing
of various grades of HBCD and include various types of data (e.g., personal breathing zone, area
monitoring, 8-hr TWA, etc.). The main source of these data is the European Union Risk Assessment
Report (EURAR) (NICNAS 2012b; ECHA 2008b). From these data, EPA selected particular data to
estimate worker inhalation exposure concentrations as discussed in Sections 2.4.1.2 through 2.4.1.13.
The overall quality confidence rating of all the data that EPA selected is high as determined by EPA's
systematic review. Table 2-70 contains a summary of EPA's approaches to the assessment of worker
inhalation exposure concentrations, and includes mention of the industrial processes and worker
activities that the selected worker monitoring data pertain to. The occupational monitoring data comprise
of HBCD concentrations in inhalable and respirable dust. EPA assessed worker exposure to inhalable
dust only and EPA's rationale for doing so is discussed in Section 4.2.1. A breathing rate of 1.25 m3/hr
was applied for all workers, representing elevated respiratory rate compared to at rest for workers
undergoing light activity (U.S. EPA 2011b). For each exposure scenario, EPA calculated acute and
chronic exposures from the estimated inhalation exposure concentrations. Equations and sample
calculations for acute and chronic exposures can be found in Appendix E.3 and Appendix E.4,
respectively.
In addition to the data mentioned above, the results of EPA's systematic review also include air
concentration data pertaining to the thermal cutting of XPS/EPS foam. As discussed in Sections 2.4.1.6,
2.4.1.7, and 2.4.1.9, XPS/EPS foam may be thermally cut with a hot wire during the processing of
HBCD to produce EPS foam from imported EPS resin beads, during the manufacture of SIPs and auto
parts from XPS/EPS foam and during the installation of XPS/EPS in buildings and other structures.
Zhang et al. (2012) reported the release of HBCD nanoparticles during the thermal cutting of XPS foam
and EPS foam in a laboratory glovebox. The HBCD that was released was mostly particles (99.9%) and
only a very small fraction was released as a vapor. The released particles were composed of HBCD and
other chemicals and included liquid particles and polystyrene foam fragments. The distribution of
HBCD concentration versus particle size of the released particles has a geometric mean of 237 and 150
nm for XPS and EPS, respectively, and geometric standard deviation of 2.2 and 1.9 for XPS and EPS,
respectively. The average concentration of XPS and EPS in the glovebox was 0.089 mg/m3 and 0.057
mg/m3, respectively. EPA did not incorporate these HBCD air concentration data into the estimates of
exposure concentrations of the relevant exposure scenarios because these data are measurements of
concentration in a laboratory glovebox and are not occupational monitoring data.
Table 2-70. Summary of Inhalation Exposure Assessment Approaches
Relevant Report
Section
Exposure Scenario
Approach to the Assessment of HBCD Potential Inhalation
Exposure Concentrations
Section 2.4.1.2
Repackaging of Import
Containers
EPA estimated the inhalation exposure concentrations to be equal to
surrogate HBCD worker inhalation exposure concentration
monitoring data. These surrogate data are worker monitoring data
that pertain to various worker activities during the manufacturing of
Page 203 of 723
-------
Relevant Report
Section
Exposure Scenario
Approach to the Assessment of HBCD Potential Inhalation
Exposure Concentrations
HBCD in Europe. The worker activities include packaging and
working in a warehouse.
Section 2.4.1.3
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
In the case of the exposure scenarios of the Compounding of
Polystyrene Resin to Produce XPS Masterbatch and the Processing
of HBCD to Produce XPS Foam Using HBCD Powder, EPA found
inhalation exposure concentration monitoring data pertaining to the
exposure scenario, but did not incorporate these data into the
estimates of inhalation exposure concentrations because these data
are not the preferred type. In the case of all of the three exposure
scenarios, EPA estimated inhalation exposure concentrations to be
equal to the assessed exposure concentrations reported in the
EURAR (ECHA 2008b) that Dcrtain to all nolvmer processing
operations involving standard grade HBCD. The bases of these
assessed exposure concentrations of the EURAR are HBCD
inhalation exposure concentration monitoring data that pertain to the
manufacture of EPS resin beads and are surrogate data for the three
exposure scenarios mentioned in the column to the left. These are
surrogate data because these data pertain to the manual addition of
HBCD to process equipment.
Section 2.4.1.5
Processing of HBCD to
Produce XPS Foam
Using HBCD Powder
Section 2.4.1.12
Formulation of
Flux/Solder Pastes
Section 2.4.1.4
Processing of HBCD to
Produce XPS Foam
using XPS Masterbatch
EPA estimated inhalation exposure concentrations to be equal to
inhalation exposure concentration monitoring data that pertain to this
exposure scenario. These monitoring data pertain specifically to the
secondary processing of XPS in Europe which EPA assumed
comprises cutting, sawing and/or machining of XPS foam.
Section 2.4.1.6
Processing of HBCD to
Produce EPS Foam
from Imported EPS
Resin Beads
EPA estimated the inhalation exposure concentrations of all these
exposure scenarios to be equal to surrogate HBCD worker inhalation
exposure concentration monitoring data. The surrogate data pertain
to the secondary processing of XPS as a part of the
manufacture of XPS foam at sites in Europe. EPA assumed the
secondary processing of XPS foam comprises cutting, sawing
and/or machining of XPS foam. EPA's single estimate of
inhalation exposure concentrations is applicable to all four scenarios.
Section 2.4.1.7
Processing of HBCD to
Produce SIPs and
Automobile
Replacement Parts from
XPS/EPS Foam
Section 2.4.1.9
Use: Installation of
XPS/EPS Foam
Insulation in
Residential, Public and
Commercial Buildings,
and Other Structures
Section 2.4.1.11
Recycling of EPS Foam
and Reuse of XPS foam
Section 2.4.1.10
Demolition and
Disposal of XPS/EPS
Foam Insulation
Products in Residential,
Public and Commercial
EPA estimated inhalation exposure concentrations to be equal to the
Occupational Safety and Health Administration (OSHA) permissible
exposure limit (PEL) for particulates not otherwise regulated
(PNOR) multiplied by the HBCD concentrations in XPS and EPS
foam.
Page 204 of 723
-------
Relevant Report
Section
Exposure Scenario
Approach to the Assessment of HBCD Potential Inhalation
Exposure Concentrations
Buildings, and Other
Structures
Section 2.4.1.8
Use: Installation of
Automobile
Replacement Parts
EPA does not expect these exposure scenarios to result in the
generation of dust, hence EPA does not estimate inhalation
exposures.
Section 2.4.1.13
Use of Flux/Solder
Pastes
Section
2.4.1.14
Recycling of
Electronics Waste (E-
Waste) Containing
HIPS
EPA estimated inhalation exposure concentrations to be equal to
inhalation exposure concentration monitoring data that pertain to this
exposure scenario. Specifically, these monitoring data pertain to the
recycling of e-waste in Europe.
EPA expects potential inhalation exposure of occupational non-users (ONUs) to HBCD, but EPA did
not quantify these exposures due to lack of adequate worker monitoring data and lack of relevant
mathematical models. ONUs are workers such as supervisors who work in or near areas where HBCD is
handled or processed, but whose work is not directly associated with HBCD. EPA expects that dust
containing HBCD that is generated during worker activities may be transported via indoor air or ambient
air currents to locations in which ONUs are present. The worker monitoring data identified through
EPA's systematic review process are presented in Appendix Appendix E, Inhalation Monitoring Data
Summary, and include personal and area monitoring data. Most of these data do not pertain to the
relevant ONUs for the following reasons: (1) the worker activities associated with the personal
monitoring data are not relevant to ONUs, and (2) the area monitoring data and the data for which the
type of sampling is not reported are either not relevant to the exposure scenarios or are not relevant to
ONUs. For example, in the case of the data pertaining to the Compounding of Polystyrene Resin to
Produce XPS Masterbatch Containing HBCD, which is 8-hr TWA area monitoring data, the sampling
location is the feed deck near typical operator positions. This data likely does not represent ONU
exposure because an ONU is unlikely to be present at the feed deck for an entire shift.
EPA assumes HBCD air concentrations that ONUs are potentially exposed to are lower than HBCD air
concentrations that workers are potentially exposed to because the dust is diluted as it is transported
through workspaces by indoor or ambient air currents. EPA also assumes the duration and frequency of
the ONUs' potential HBCD inhalation exposures to be lower than that of workers. The lower HBCD
potential inhalation exposure levels of ONUs would result in lower risk for ONUs as compared to
workers. Uncertainties related to EPA's assumptions related to ONU exposure levels are discussed in
Section 2.4.1.15.4.
General Dermal Exposures Approach and Methodology
EPA estimated high-end worker dermal potential dose rate in accordance with the EPA/OPPTDirect 2-
Hand Dermal Contact with Solids Model (U.S. EPA 2013a) in the case of the following exposure
scenarios: the repackaging of import containers, compounding of polystyrene to produce XPS
masterbatch, manufacturing of XPS foam using XPS masterbatch, manufacturing of XPS foam using
HBCD powder, and formulation of flux/solder pastes (these scenarios are discussed in Sections 2.4.1.2
through 2.4.1.5 and 2.4.1.12). This high-end potential dose rate is equal to 3,100 mg/day which is the
quantity of solids retained on a worker's skin during an event that results in the worker's contact with
Page 205 of 723
-------
the solids; the frequency of such events is assumed to be once per day ( 13a). The
EPA/OPPTDirect 2-Hand Dermal Contact with Solids Model does not include a central tendency value
of the potential dose rate although this model is based on data reported in Lansink (1996) and both the
high-end and central tendency values of these data are given in Lansink (1996). The central tendency
potential dose rate that is associated with the high-end potential dose rate of 3,100 mg/day is equal to
900 mg/day. The central tendency value of 900 mg is reported in Lansink et al. 1996 as cited in
Marquart et al. 2006. This central tendency value pertains to the manual loading of mixers with dusty
powder and is designated as the typical case exposure (Marquart et al. 2006)15.
EPA estimated high-end worker dermal potential dose rate in accordance with the EPA/OPPT Direct 2-
Hand Dermal Contact with Container Surfaces (Solids) Model ( 13 a) in the case of the use
of solder/flux pastes (this scenario is discussed in Section 2.2.13). This high-end potential dose rate is
equal to 1,110 mg/day which is the quantity of solids retained on a worker's skin during an event that
results in the worker's contact with the solids; the frequency of such events is assumed to be once per
day (U.S. EPA. 2013a). The EPA/OPPT Direct 2-Hand Dermal Contact with Container Surfaces
(Solids) Model does not include a central tendency value of the potential dose rate although this model is
based on data reported in Lansink (1996) and both the high-end and central tendency values of these
data are given in Lansink et al. The central tendency potential dose rate that is associated with the high-
end potential dose rate of 1,110 mg/day is equal to 450 mg/day. The central tendency value of 450 mg is
reported in Lansink (1996) as cited in Marquart et al. 2006. This central tendency value pertains to the
gathering of closed bags of powder and is designated as the typical case exposure (Marquart et al.
2006)16.
The two models that EPA used as mentioned above assume a single contact event per day and that the
amount of solid on the skin is not expected to be significantly reduced by wiping from the skin or
increased from repeated contact with the chemical {i.e., wiping excess solids from the skin does not
remove a significant fraction of the small layer of chemical adhering to the skin and additional contacts
with the chemical do not add a significant fraction to the layer). EPA calculated the potential dose for a
worker with no dermal protection by multiplying the quantity of solids on the skin by the weight fraction
of HBCD in the solids and the frequency of exposure events. EPA does not expect dermal exposure for
the remaining exposure scenarios because HBCD is entrained in the EPS and XPS foam (those in
Section 2.4.1.6 through 2.4.1.11).
In this Risk Evaluation, EPA provides comparison of the potential worker dermal dose rates calculated
by EPA and those estimated in the EURAR (ECHA. 2008b) and Australian Risk Assessment (NICNA.S
2012b). The EURAR and NICNAS both estimate potential dermal exposures using the Estimation and
Assessment of Substance Exposure (EASE) model. The EASE model was developed by the UK Health
and Safety Executive with the Health and Safety Laboratory. It predicts expected dermal exposures for a
wide range of substances and scenarios using situational information related to the chemical (Tickner et
15 The high-end value of 3,100 mg also pertains to manual loading of mixers with dusty powder. This value corresponds to
the value of 3,000 mg reported in Marquart et al. (Marquart et al 20061 as the reasonable worst case exposure pertaining to
loading of mixers and obtained from Lansink et al. (Lansink et al. .1.9961. EPA did not directly cite Lansink et al. (Lansink et
al. 1996') because, as stated in Marquart et al. (Marquart et at. 20061. this report has not been published in a scientific
journal.
16 The high-end value of 1,110 mg also pertains to the gathering of closed bags of powder. This value corresponds to the
value of 1,050 mg reported in Marquart et al. (Marquart et al. 20061 as the reasonable worst case exposure pertaining to the
gathering of closed bags of powder and obtained from Lansink et al. (Lansink et al. 1996"). EP A did not directly cite Lansink
et al. (Lansink et al. .1.9961 because, as stated in Marquart et al. (Marquart et al. 20061. this report has not been published in
a scientific journal.
Page 206 of 723
-------
al. 2Q05Y
For occupational exposures, EPA used the potential dermal dose rate estimated as described above to
calculate exposure concentration metrics required for risk assessment. Specifically, EPA used the
potential dermal dose rates and dermal absorption factor of 6.5% (discussed later in Section 3.2.2) to
estimate the AAD and CAD. The AAD calculation entails the multiplication of the dermal potential dose
rate by the dermal absorption factor, which is then divided by body weight. The CAD calculation is the
same, with the additional multiplication of exposure frequency and working years, followed by division
of the averaging time. The values used for body weight, exposure frequency, working years, and
averaging time are explained in Appendix E.3. The AAD is used to assess acute exposure risks. The
CAD is used to assess risks from chronic exposures.
Occupational non-users are workers who do not handle HBCD and thus, unlike workers, are not
potentially exposed to HBCD dermally as a result of handling HBCD. However, ONUs are potentially
exposed to HBCD dermally through contact with surfaces where HBCD dust has settled. EPA mentions
this type of potential ONU dermal exposure in the discussions of the relevant occupational exposure
scenarios, but EPA did not quantify these exposures due to lack of data, and EPA expects that dermal
exposures may be much less likely for this population. Potential ONU dermal exposure to settled dust is
unlikely in the case of the exposure scenarios that do not include worker dermal exposure because these
exposure scenarios pertain to material (EPS resin beads and XPS/EPS insulation) in which the HBCD is
entrained at low concentrations and worker or ONU contact with this material is unlikely to result in
dermal exposure.
A summary of approaches and EPA's overall confidence in the exposure estimates are provided in
Table 2-71.
Page 207 of 723
-------
Table 2-71. A Summary for Each of the 12 Occupational Exposure Scenarios (OESs)
[For many cases EPA was not able to estimate inhalation exposure for ONUs, but EPA expects these to be lower than inhalation exposure for workers; dermal exposure
not estimated for ONUs since they are not expected to be in direct contact with HBC'D.|
1
1
Inhalation Exposure
'Dermal Exposure
Occupational Exposure
1
1
Monitoring
Modeling
Overall
Confidence
1 Modelingc
Scenario (OES)
1
|Monitoring
1 Data
|
# Data
Points
Data
Quality
Rating
Worker
ONU
Worker
ONU
J Worker
ONU
Repackaging of Import Containers
1 S
10
H
~
X
X
X
M
1 ^
-
Compounding of Polystyrene Resin to Produce
XPS Masterbatch
16
H
~
X
X
X
M
-
Processing of HBCD to Produce XPS Foam using
XPS Masterbatch
1
1 /
1
9
H
~
X
X
X
M
1 s
-
Processing of HBCD to Produce XPS Foam using
HBCD Powder
1 ~
1
16
H
~
X
X
X
M
1 ~
-
Processing of HBCD to Produce EPS Foam Using
Imported EPS Resin Beads a
9
H
~
X
X
X
L to M j
-
Processing of HBCD to Produce SIPs and
1
¦
Automobile Replacement Parts from XPS/EPS
~
9
H
~
X
X
X
L to M
I
-
Foam a
|
Installation of Automobile Replacement Parts b
1 x
N/A
N/A
X
-
-
N/A |
-
Installation of XPS/EPS Foam Insulation in
Residential, Public and Commercial Buildings,
1 ~
9
H
~
X
X
X
L to M
1
-
and Other Structures a
1
Demolition and Disposal of XPS/EPS Foam
1
|
Insulation Products in Residential, Public and
1 *
N/A
N/A
X
~
X
L to M
¦
-
Commercial Buildings, and Other Structures a
1
Recycling of EPS Foam and Reuse of XPS foam a
|
9
H
~
X
X
X
L to M
1
-
Formulation of Flux/Solder Pastes
! ~
16
H
~
X
X
X
M
! ~
-
Use of Flux/Solder Pastes
i x
N/A
N/A
X
-
-
N/A
i ^
-
Recycling of Electronics Waste (E-Waste)
Containing HIPS
! y
24
H
~
X
X
X
M
! y
-
a EPA does not expect dermal exposure of workers to be a part of these exposure scenarios.
b The installation of automobile replacement parts is not expected to result in worker and ONU inhalation and dermal exposures.
0 The exposure scenarios preclude ONU dermal exposure because ONUs are not expected to handle HBCD.
Page 208 of 723
-------
Consideration of Engineering Controls and Personal Protective Equipment
OSHA requires and NIOSH recommends that employers utilize the hierarchy of controls to address
hazardous exposures in the workplace. The hierarchy of controls strategy outlines, in descending order
of priority, the use of elimination, substitution, engineering controls, administrative controls, and lastly
personal protective equipment (PPE). The hierarchy of controls prioritizes the most effective measures
first which is to eliminate or substitute the harmful chemical (e.g., use a different process, substitute with
a less hazardous material), thereby preventing or reducing exposure potential. Following elimination and
substitution, the hierarchy recommends engineering controls to isolate employees from the hazard,
followed by administrative controls, or changes in work practices to reduce exposure potential (e.g.,
source enclosure, local exhaust ventilation systems). Administrative controls are policies and procedures
instituted and overseen by the employer to reduce the potential for worker exposure to hazards, these
could include training employees on the hazards and how to avoid them, policies regarding scheduling
to reduce acute exposures, and housekeeping standards. As the last means of control, the use of personal
protective equipment (e.g., respirators, gloves) is recommended, when the other control measures cannot
reduce workplace exposure to an acceptable level. The National Institute for Occupational Safety and
Health (NIOSH) and the U.S. Department of Labor's Bureau of Labor Statistics (BLS) conducted a
voluntary survey of U.S. employers regarding the use of respiratory protective devices between August
2001 and January 2002 (NIOSH. 2003). For additional information, please also refer to
Memorandum NIOSH BLS Respirator Usage in Private Sector Firms, Docket # EPA-HQ-OPPT-2019-
0500 ( 20).
Respiratory Protection
OSHA's Respiratory Protection Standard (29 CFR Section 1910.134) requires employers in certain
industries to address workplace hazards by implementing engineering control measures and, if these are
not feasible, provide respirators that are applicable and suitable for the purpose intended. Respirator
selection provisions are provided in Section 1910.134(d) and require that appropriate respirators are
selected based on the respiratory hazard(s) to which the worker will be exposed and workplace and user
factors that affect respirator performance and reliability. Assigned protection factors (APFs) are
provided in Table 1 under Section 1910.134(d)(3)(i)(A) (see below in Table 2-72) and refer to the level
of respiratory protection that a respirator or class of respirators is expected to provide to employees
when the employer implements a continuing, effective respiratory protection program.
There are no OSHA or NIOSH exposure limits for the HBCD cluster: (CAS #s: 25637-99-4; 3194-55-6;
3194-57-8), however, HBCD is handled in a powdered form with mean particle size ranges from 20 to
150 |im. There is the potential for generation of airborne HBCD dust during different worker activities.
Employers should first consider elimination, substitution, engineering, and administrative controls to
reduce exposure potential and, if exposures still present workplace, employers are required to institute a
respiratory protection program and provide employees with NIOSH-certified respirators. Where other
hazardous agents could exist in addition to HBCD, consideration of combination cartridges would be
necessary. Table 2-72 can be used as a guide to show the protectiveness of each category of respirator;
EPA took this information into consideration as discussed in Section 4.2.2. Based on the APF, inhalation
exposures may be reduced by a factor of 5 to 10,000, when workers and occupational non-users are
using respiratory protection.
Page 209 of 723
-------
Table 2-72. Assigned Protection Factors for Respirators in OSHA Standard 29 CFR Section
1910.134
Type of Respirator
Quarter
Mask
Half Mask
Full
Facepiece
Helmet/
Hood
Loose-
fitting
Facepiece
1. Air-Purifying Respirator
5
10
50
-
-
2. Power Air-Purifying Respirator (PAPR)
-
50
1,000
25/1,000
25
3. Supplied-Air Respirator (SAR) or Airline Respirator
• Demand mode
-
10
50
-
-
• Continuous flow mode
-
50
1,000
25/1,000
25
• Pressure-demand or other positive-
pressure mode
-
50
1,000
-
-
4. Self-Contained Breathing Apparatus (SCBA)
• Demand mode
-
10
50
50
-
• Pressure-demand or other positive-
pressure mode (e.g., open/closed
circuit)
-
-
10,000
10,000
-
Source: 1910.13 4(d)(3 )(i)( A)
Dermal Protection
The Hand Protection section of OSHA's Personal Protective Equipment Standard (29 CFR Section
1910.138) requires employers to select and require workers to wear gloves to prevent exposure to
harmful substances. As with respirators, gloves are used to prevent employee exposures to hazards.
Employers base selection of gloves on the type of hazard encountered, conditions during use, tasks
performed and factors that affect performance and wear ability. Gloves, if proven impervious to the
hazardous chemical, and if worn on clean hands and replaced when contaminated or compromised, are
expected to provide employees with protection from hazardous substances. HBCD is a solid particulate
and would not be expected to permeate through gloves. Some examples of impervious gloves are nitrile,
butyl rubber, polyvinyl chloride, and polychloroprene.
EPA reviewed safety data sheets (SDSs) for HBCD powder, EPS resin beads containing HBCD, and
XPS and EPS foam containing HBCD. EPA did not find any SDSs for XPS masterbatch containing
HBCD.
The exposure scenarios in this Risk Evaluation in which workers may handle HBCD powder include
Repackaging of Import Containers, Compounding of Polystyrene Resin to Produce XPS Masterbatch,
Processing of HBCD to Produce XPS Foam, and Formulation of Flux/Solder Pastes. For HBCD powder,
an SDS from Great Lakes Chemical Corporation (Great Lakes Chemical 2003) recommended the use of
neoprene gloves and an SDS from Santa Cruz Biotechnology Company, Inc. (Santa Cruz Biotechnology
2009) recommended the use of gloves made of polychloroprene, nitrile rubber, butyl rubber, Viton, or
polyvinyl chloride.
Page 210 of 723
-------
The exposure scenarios in this Risk Evaluation in which workers may handle XPS or EPS foam
containing HBCD include: Processing to Produce XPS Foam using XPS Masterbatch, Processing of
HBCD to Produce XPS Foam, Processing to Produce EPS Foam from Imported EPS Resin Beads,
Processing to Produce SIPs and Automobile Replacement Parts from XPS/EPS Foam, Installation of
XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings, and Other Structures,
Demolition and Disposal of XPS/EPS Foam Insulation in Residential, Public and Commercial
Buildings, and Other Structures, and Recycling of EPS Foam. EPA reviewed seven SDSs for XPS and
EPS foam products containing HBCD. All the reviewed SDSs recommend suitable or appropriate gloves
and, in some cases, gloves to protect from mechanical injury. The SDSs do not recommend specific
glove materials (Dow Chemical Pacific 2018; DiversiFoam 2015; Insulfoam a Division of Carlisle
Construction 2015; Multi-Panels 10 * , ¦>1 I* t _0i», lite Plastics Co dba Fox 2008; A.C.H. Foam
Technologies 2007).
During Processing to Produce EPS Foam from Imported EPS Resin Beads, workers may handle EPS
resin beads containing HBCD. An SDS from BASF recommends the use of non-static gloves, such as
leather gloves, when handling EPS resin beads containing HBCD (BASF ). As indicated in Section
1.2.2, BASF has ceased the use of HBCD. EPA did not find additional glove material recommendations.
During Use of Flux/Solder Pastes, workers may handle flux/ solder paste formulations containing
HBCD. SDSs from Henkel and Kester recommend the use of nitrile rubber gloves (Henket 2016; Kester
2.015). The SDS from Kester also recommends the use of natural rubber gloves.
2.4.1.2 Repackaging of Import Containers
Imported HBCD is repackaged by unloading HBCD powder or granules from imported containers into
an intermediate storage vessel or directly into new containers. Workers and ONUs are potentially
exposed by inhalation to the HBCD dust that is generated during the transfer of HBCD. Also, there is a
potential for ONU dermal exposure through contact with surfaces where HBCD dust has settled.
Because of the larger particle size of the granules, inhalation exposure to dust during unloading of
granules is expected to be lower than that from unloading powders (NICNAS 2012b; ECHA 2008b).
Worker inhalation and dermal exposure during the unloading of imported EPS resin beads is not
expected due to the larger size of the beads and because HBCD is entrained within the polymeric matrix
of the EPS resin beads (NICNAS 201 :h; VQj \ 2008b).
Number of Potentially Exposed Workers and Occupational Non-Users
As discussed in Section 2.2.1, EPA developed release and exposure estimates for repackaging of import
containers at a single site. Of the five submitters to 2016 CDR, four submitters estimate that fewer than
10 workers are potentially exposed to HBCD, while the fifth submitter estimated that at least 10 but
fewer than 25 workers are potentially exposed to HBCD. However, the companies that previously
reported HBCD import volumes to 2016 CDR have stated to EPA that they permanently stopped the
activity in 2016 or 2017. Thus, in lieu of using this CDR data from companies that discontinued use of
HBCD, EPA estimated the number of workers potentially exposed using Bureau of Labor Statistics
(BLS) data.
Based on BLS data for NAICS code 493100, Warehousing and Storage, and related Standard
Occupational Classification (SOC) codes, there are on average an estimated three workers and one ONU
per site at warehousing and storage facilities. Based on these BLS data and one site for the repackaging
of import containers, EPA estimated that a total of three workers and one ONU are potentially exposed
during this exposure scenario.
Page 211 of 723
-------
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to surrogate HBCD
occupational inhalation exposure concentration monitoring data. These surrogate data are worker
monitoring data that pertain to various worker activities during the manufacturing of HBCD in Europe.
EPA also considered other HBCD occupational inhalation exposure concentration data as surrogate
monitoring but chose the data mentioned above as further discussed below.
HBCD occupational inhalation exposure monitoring data that EPA considered are shown in Table 2-73
below. EPA selected the data of Searl and Robertson (2005). which are noted as la in this table, as the
surrogate monitoring data from among all of the data in Table 2-73 because (a) the overall quality
confidence rating in these data is high as determined via EPA's systematic review, (b) the worker
activities that these data pertain to include packaging and working in a warehouse, (c) these data pertain
to standard grade HBCD, and (d) these data are 8-hr TWA personal breathing zone measurements.
EPA estimated central tendency and high-end exposure concentrations to be equal to the median value
of 0.89 mg/m3 and the 90th percentile value of 1.89 mg/m3 of the surrogate monitoring data, respectively.
EPA also considered worker monitoring data other than the data mentioned above as surrogate data.
Specifically, EPA considered data that pertain to worker activities that include addition of HBCD to
process equipment provided in Table 2-74.. These data are from Thomsen (2007). which are noted as la
and lb in this table, and the data from Searl and Robertson (2.005). which are noted as 2a-d in this table.
EPA did not select these data as surrogate data because the addition of HBCD to process equipment is
likely to involve handling of smaller quantities of HBCD as compared to the repackaging of HBCD.
The exposure frequency for this exposure scenario is a range of 29 to 250 days/year. As discussed in
Section 2.2.2, EPA estimated days of release at a repackaging site as a range from 29 to 300 days/year.
EPA expects this range of release days is also reflective of the operating days during which HBCD is
repackaged at an importation site and workers are potentially exposed to HBCD. However, EPA does
not expect that workers will be exposed greater than 250 day/year, accounting for a worker schedule of
five days per week and 50 weeks per year. EPA used the midpoint of this range of exposure frequency,
rounded up where the midpoint resulted in fractions of days, to calculate central tendency average daily
dose. EPA used the high-end of this range of exposure frequency to calculate high-end average daily
dose. Additionally, EPA estimated worker exposure over the full working day, or eight hours/day, as the
data used to estimate inhalation exposures are 8-hour time-weighted average (TWA) data.
Page 212 of 723
-------
Table 2-73. Inhalation Monitoring Data for Manufacturing of HBCD
Data Source/Study a
Exposure Scenario
Form of HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Sourcec
Overall
Confidence
Rating
Inhalation Monitoring Data Used to Estimate Worker Exposures Resulting from Repackaging
Searl and Robertson
(2005) - la
Manufacturing of
HBCD
Standard grade
HBCD
Personal
Breathing
Zone
Packaging,
compaction,
process
operations, and
working in the
warehouse
Mean: 1.23
Median: 0.89
90th percentile:
1.89
Max: 3 mg/m3
10
8-hr TWA
(ECHA
2008b)
(ECHA
2009b)
High
Other Inhalation Monitoring Data Pertaining to the Manufacturing of HBCD that EPA Considered as Surrogate Monitoring Data
Searl and Robertson
(2005)- lb
Manufacturing of
HBCD
Fine grade
HBCD
Personal
Breathing
Zone
Packaging,
compaction,
process
operations, and
working in the
warehouse
Mean: 23
90th percentile:
35
4
8-hr TWA
(ECHA
2008b)
High
Searl and Robertson
(2005) - lc
Manufacturing of
HBCD
HBCD of
unknown grade
NR
Packaging and
compaction of
powders
Respirable,
mean: 0.18
Inhalable,
Mean: 1.23
NR
NR
(ECHA
2009c)
High
Waindzioch (2000) -
la
Manufacturing of
HBCD
HBCD of
unknown grade
Area
Reactor
0.00028 -
0.0285
3
Short-term
(ECHA
2008b)
Unacceptable
Waindzioch (2000) -
lb
Manufacturing of
HBCD
HBCD of
unknown grade
Area
Filling Station
0.0094 - 0.097
2
Short-term
(ECHA
2008b)
High
Biesemeier (1996)
Manufacturing of
HBCD
HBCD of
unknown grade
NR
Bagging
HBCD product
4.0-4.5
NR
NR
(ECHA
2008b)
High
Velsicol (1978)
Manufacturing of
HBCD
HBCD of
unknown grade
Personal
Breathing
Zone
Transfer of the
HBCD in the
hammer-mill to
28 drums
1.9
1
300 minutes
(Velsicol
Chem
Core
1978)
High
Yietal. (2016)
Manufacturing of
HBCD
HBCD of
unknown grade
Personal
Breathing
Zone
NR
0.0102 -0.0283
14
NR
(Yi et al.
2016)
High
Page 213 of 723
-------
Data Source/Study a
Exposure Scenario
Form of HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Sourcec
Overall
Confidence
Rating
NR = Not Reported; N/A = Not Applicable
a - Where multiple datasets were available from one literature source, EPA distinguished data as la, lb, 2a, 2b, etc.
b - The statistical values were obtained from the referenced literature source and were not calculated by EPA.
c - The source where the respective data was extracted. All sources of the information are mentioned. In the case of multiple sources, information from the various
sources is presented as contained in these sources and EPA did not combine the information from the various sources.
Page 214 of 723
-------
Dermal Exposure Assessment
As described in Section 2.4.1.1. and assuming two-hand contact to solids containing 100% HBCD, EPA
calculated the potential dose for a worker to be 3,100 mg HBCD/day (high-end) and 900 mg HBCD/day
(central tendency) (U.S. EPA.: ).
The EURAR estimated dermal exposure during manufacturing of HBCD (importation and repackaging
was not included in the EURAR) using EASE model. The EURAR estimated an exposure to standard
grade HBCD powder of 1 mg/cm2-day. This translates into a dose of 1,070 mg/day, using EPA's two-
hand surface area of 1,070 cm2. The NICNAS report estimated dermal exposure during importation and
repackaging to standard grade HBCD powder of 0.1 to 1 mg/cm2-day using the EASE model. Using
EPA's two-hand surface area, this results in a dose of 107 to 1,070 mg/day.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed potential inhalation exposure concentrations presented
above. EPA considered the quality of the data, the assessment approach, and uncertainties in assessment
results to determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations based on inhalation exposure concentration monitoring
data that (a) are the preferred type of monitoring data (i.e., 8-hr TWA personal breathing zone data), and
(b) are surrogate data pertaining to various worker activities that include an activity that is relevant to
the assessed exposure scenario.
There is uncertainty about the extent to which the surrogate inhalation exposure concentration
monitoring data are valid surrogate data because of the following reasons. First, these concentrations are
based on worker monitoring data that pertain to various worker activities including activities that are not
relevant to the exposure scenario. Second, EPA is uncertain that the packaging process associated with
the worker monitoring data and the repackaging process in the U.S. are equivalent in terms of worker
exposure. There is also uncertainty in the estimated HBCD potential inhalation exposure concentrations
because these concentrations pertain to workers in Europe and the extent to which these concentrations
represent the distribution of inhalation exposure air concentrations pertaining to workers in the U.S. is
uncertain. Refer to Section 2.4.1.14 for additional discussion of uncertainty. Based on these strengths,
and uncertainties, EPA has medium confidence in the assessed occupational inhalation exposure air
concentrations.
2.4.1.3 Compounding of Polystyrene Resin to Produce XPS Masterbatch
Workers are expected to manually unload and transfer HBCD powder or granules into hoppers or other
equipment used to feed the HBCD into XPS masterbatch mixing equipment. This manual transfer may
result in worker inhalation exposure to HBCD dust and dermal exposure to solid HBCD. Additionally,
the generated dust from these transfer activities may result in ONU inhalation exposure to the HBCD
dust and ONU dermal exposure through contact with surfaces where HBCD dust has settled.
Workers may also be potentially exposed from occasional cleaning of process equipment and loading of
XPS masterbatch into packages, if these activities are manual.
Number of Potentially Exposed Workers and Occupational Non-Users
As discussed in Section 2.2.3, EPA developed exposure estimates for one site for this exposure scenario.
The two submissions in 2016 CDR that identify the industrial sector as "plastic material and resin
Page 215 of 723
-------
manufacturing" each estimate that at least 50 but fewer than 100 workers are potentially exposed to
HBCD. However, the companies that previously reported HBCD import volumes to CDR have stated to
EPA that they permanently stopped the activity in 2016 or 2017. Thus, in lieu of using this CDR data
from companies that discontinued use of HBCD, EPA estimated the number of workers potentially
exposed using Bureau of Labor Statistics (BLS) data.
Based on data from the Bureau of Labor Statistics (BLS) for NAICS code 325991, Custom
Compounding of Purchased Resins, and related Standard Occupational Classification (SOC) codes,
there are on average an estimated 20 workers and 7 ONUs per site at custom compounding facilities.
Based on these data and one modeled site for the production of XPS masterbatch, EPA estimated that a
total of 20 workers and 7 ONUs are potentially exposed during this exposure scenario.
Occupational Exposure Assessment
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to the assessed exposure
concentrations reported in the EURAR (ECH.A. 2008b) that pertain to all polymer processing operations
involving standard grade HBCD. These assessed exposure concentrations of the EURAR are based on
HBCD occupational inhalation exposure concentrations that pertain to the manufacture of EPS resin
beads. EPA considered HBCD occupational inhalation exposure concentration data that pertain to the
exposure scenario that is the subject of this section as well as data that pertain to other exposure
scenarios but chose the assessment approach mentioned above.
EPA found monitoring data that pertain to the exposure scenario that is the subject of this section and
the overall confidence rating of these data is high as determined via EPA's systematic review. These
data are the data of Searl and Robertson (2005). which are HBCD occupational inhalation exposure
concentration monitoring data pertaining to the compounding of polystyrene resin and production of
XPS masterbatch at sites in Europe and are presented in Table 2-74. and noted in this table as 3a-d. EPA
did not incorporate these data into the estimate of exposure concentrations because the grade of HBCD
associated with these data is not reported and the type of sample (personal breathing zone or area) is
reported for only half of these data.
Page 216 of 723
-------
Table 2-7^
. Summary of In
lalation Monitoring Data for Handling of HBCD
Literature
Study a
Exposure Scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or
Sampling Location
Exposure Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Sourcec
Overall
Confidence
Rating
Inhalation Monitoring Data Used to Estimate Worker Exposures (both in this Risk Evaluation and the EURAR)
Searl and
Robertson
(2005) - 2a
Manufacturing of
EPS Resin beads
Standard
grade
HBCD
Personal
Manual addition of HBCD
powder to reactor each
time a batch of EPS resin
was produced
Range: 2.89-21.5
Mean: 7.2
Median: 5.52
90th percentile: 10.5
12
Short-term (13 to
56 mins)
(NICNAS
2012b):
(ECHA
2008b)
High
Searl and
Robertson
(2005) -
2b
Manufacturing of
EPS Resin beads
Standard
grade
HBCD
Personal
Manual addition of HBCD
powder to reactor each
time a batch of EPS resin
was produced
Range: 0.12-3.36
Mean: 1
Median: 0.42
90th percentile: 1.11
(NICNAS 2012b): 1.3
(ECHA 2008b)
12
8-hr TWA - note
these are 8-hr
TWA values of
the data in the
above row
(NICNAS
2012b):
(ECHA
2008b)
High
Searl and
Robertson
(2005) -
2c
Manufacturing of
EPS Resin beads
Standard
grade
HBCD
Personal
Manual addition of HBCD
powder to reactor each
time a batch of EPS resin
was produced
Range: 0.07-14.7
Mean: 1.2
Median: 0.27
90th percentile: 1.10
18
8-hr TWA
(ECHA 2008b):
Full-Shift
(NICNAS
2012b)
(NICNAS
2012b):
(ECHA
2008b)
High
Searl and
Robertson
(2005) -
2d
Manufacturing of
EPS Resin beads
Standard
grade
HBCD
Personal
Weighing powder prior to
addition to reactor. HBCD
bags were weighed and
opened concurrently, or
weighed in advance, in
which case HBCD was
transferred from 25-kg
sacks using plastic scoop
(full-shift measurement).
Range: 4.35-12.1
Mean: 7.2
Median: 6.19
90th percentile: 10.5
(NICNAS 2012b): 10.6
(ECHA 2008b):
4
8-hr TWA
(ECHA 2008b):
Full Shift
(NICNAS
2012b)
(NICNAS
2012b):
(ECHA
2008b)
High
Other Inhalation Monitoring Data for Handling of HBCD
Searl and
Robertson
(2005) - 3a
Compounding of
Polystyrene resin to
produce XPS
Masterbatch
containing HBCD
HBCD of
unknown
grade
Area
Weighing and mixing
Max 7.5 (for 2 hours)
Mean: 1.89
Median: 0.83
90th percentile: 5.4
10
Short-term
(ECHA
2008b)
(ECHA
2009b)
High
Searl and
Robertson
(2005) - 3b
Compounding of
Polystyrene resin to
produce XPS
HBCD of
unknown
grade
Area
Weighing and mixing
Mean: 0.88
90th percentile: 1.36
10
8-hr TWA
(ECHA
2008b)
High
Page 217 of 723
-------
Literature
Study a
Exposure Scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or
Sampling Location
Exposure Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Sourcec
Overall
Confidence
Rating
Masterbatch
containing HBCD
Searl and
Robertson
(2005) - 3c
Compounding of
Polystyrene resin to
produce XPS
Masterbatch
containing HBCD
HBCD of
unknown
grade
NR
Extruder
Mean: 0.12
Median: 0.10
90th percentile: 0.16
4
5 hours
(ECHA
2008b)
(ECHA
2009b)
High
Searl and
Robertson
(2005) - 3d
Compounding of
Polystyrene resin to
produce XPS
Masterbatch
containing HBCD
HBCD of
unknown
grade
NR
Automated handling of
HBCD
Negligible
3
NR
(ECHA
2008b)
High
Abbott
(2001) - la
Manufacture of
XPS from HBCD
powder or granules
Standard
grade
HBCD
Area
At the feed deck near
typical operator positions
Range 0.24 - 1.6
Mean: 0.66
90th percentile: 1.45
(excluding 10 ND samples)
16(10
ND)
8-hr TWA
(ECHA
2008b)
High
Abbott
(2001) - lb
Manufacture of
XPS from HBCD
powder or granules
HBCD
powder and
granules
Personal
breathing
zone
Activities in the mixer
area, including operating a
closed automated process
excluding potential contact
with neat HBCD
Range: 0.0002-0.0009
Mean: 0.0005
Median: 0.0005
6
8-hr TWA
(ECHA
2008b)
(NICNAS
2012b)
High
Thomsen
(2007) - la
Manufacture of
XPS from HBCD
powder or granules
HBCD
powder and
granules
Personal
breathing
zone
Weighing and addition of
HBCD to the reactor and
subsequent washing,
centrifugation, sifting, and
transfer of product to a silo
container
Range: 0.001-0.15
Mean: 0.015
Median: 0.0027
24
8-hr TWA
Thomsen
(2007)
High
Thomsen
(2007) - lb
Manufacture of
XPS from HBCD
powder or granules
HBCD
granules
Mostly
area and
some
personal
breathing
zone
Feed deck near typical
operator positions
Range 0.005-0.9
Mean: 0.24
90th percentile: 0.47
(excluding 16 ND samples)
43 (16
ND)
60- 1435
minutes
(ECHA
2008b)
High
Page 218 of 723
-------
Literature
Study a
Exposure Scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or
Sampling Location
Exposure Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Sourcec
Overall
Confidence
Rating
Searl and
Robertson
(2005) - 4
Manufacture of
XPS from HBCD
powder or granules
HBCD
granules
Area
Logistics, extruding, and
laboratory
Mean: 0.00003
90th percentile: 0.00004
12
8-hr TWA
(ECHA
2008b)
High
Ransbotyn
(1999)
Manufacturing of
EPS Resin beads
Respirable
Dust
Inhalable
Dust
Personal
Addition of HBCDD to
reactor or the supervising
of the addition.
Respirable dust: <0.5
Total Inhalable dust: 2.0
Not specific to HBCD
5
Max 8-hr TWA
(ECHA
2008b)
High
NICNAS
(2012b)-
la
All industrial
polymer processing
sites d
Standard
grade
HBCD
Modelled
with
EASE
Addition of HBCD into
process operation
Typical: 2 to 5
"Worst-case": 5 to 50
N/A-
tliis is a
modelled
exposure
8-hr TWA
(NICNAS
2012b)
High
NICNAS
(2012b)-
lb
HBCD importation/
repackaging sites
and all industrial
polymer processing
sites d
HBCD
granules
Modelled
with
EASE
Repackaging with the use
of LEV (typical) and
without LEV (worst-case)
Typical: 0.2 to 0.5
"Worst-case": 0.5 to 5
N/A-
tliis is a
modelled
exposure
8-hr TWA
(NICNAS
2012b)
High
NR = Not Reported; N/A = Not Applicable
a - Where multiple datasets were available from one literature source, EPA distinguished data as la, lb, 2a, 2b, etc.
b - The statistical values were obtained from the referenced literature source and were not calculated by EPA.
c - All sources of the information are mentioned. In the case of multiple sources, information from the various sources is presented as contained in these sources and EPA did
not combine the information from the various sources.
d - Per NICNAS (2012b). this includes EPA's c\ do sure scenarios for Comooundine of Polvstvrene Resin to Product XPS Masterbatch. Processing HBCD to Produce XPS
Foam using XPS Masterbatch, Processing of HBCD to Produce XPS Foam using HBCD Powder, and Processing of HBCD to Produce EPS Foam from Imported EPS Resin
Beads.
Page 219 of 723
-------
The HBCD occupational inhalation exposure concentration data or modeling results from Thomsen
(2007), NICNAS (2012b). Abbott (2001) and Ransbotyn (1999). which are given in Table 2-74., pertain
to various processes other than the compounding of polystyrene resin and production of XPS
masterbatch. The overall confidence rating of all of these data is high as determined via EPA's
systematic review; however, EPA did not further consider these data as surrogate data because none of
these data are 8-hr TWA personal breathing zone data that are only associated with HBCD standard
grade powder.
The HBCD occupational inhalation exposure concentration data of Searl and Robertson (2005) that
pertain to the manufacture of EPS resin beads (provided in Table 2-74. and noted in this table as 2b-d)
are 8-hr TWA personal breathing zone data that are only associated with HBCD standard grade powder.
The overall confidence rating of all these data is high as determined via EPA's systematic review
process. EPA determined these data are surrogate data because these data pertain to the worker activity
of HBCD manual addition to process equipment which is the worker activity that is expected to result in
the largest exposure in the case of the exposure scenario that is the topic of this section. However, EPA
cannot incorporate all this surrogate data into estimates of exposure concentrations because the discrete
data points of the various datasets are not available, and EPA cannot calculate the 50th percentile and
95th percentile values of all the data to assess central tendency and high-end values. However, as
detailed in Appendix E.2, all of these data are the basis of the assessed "typical" and "reasonable worst-
case" HBCD occupational exposure concentrations that are reported in the EURAR and that pertain to
all polymer processing operations involving standard grade HBCD. Hence, EPA estimated HBCD
occupational exposure concentrations to be equal to these assessed exposure concentrations of the
EURAR. Specifically, EPA estimated high-end and central tendency exposure concentrations to be
equal to, respectively, the "reasonable worst-case" exposure concentration of 2.5 mg/m3 and the
"typical" exposure concentration that is equal to one half of the reasonable worst-case, or 1.25 mg/m3.
As discussed in Section 2.2.3, EPA estimated a range of release days of 10 to 60 days/year. EPA expects
this range of release days is also reflective of the operating days during which HBCD is processed at a
compounding site and workers are potentially exposed to HBCD. EPA used the midpoint of this range
of exposure frequency (rounded up) to calculate central tendency average daily dose and used the high-
end of this range of exposure frequency to calculate high-end average daily dose. Additionally, EPA
estimated worker exposures over the full working day, or eight hours/day, as the data used to estimate
inhalation exposures is 8-hour TWA data.
Dermal Exposure Assessment
As described in Section 2.4.1.1. EPA calculated dermal exposure assuming two-hand contact to solids
containing 100% HBCD (NICNAS 2012b; KemI 2009) because sites that produce HBCD flame-
retarded XPS masterbatch receive manufactured or imported HBCD in its pure form to be 3,100 mg
HBCD/day (high-end) and 900 mg HBCD/day (central tendency).
The EURAR estimated dermal exposure for the use of HBCD standard grade powder as an additive in
XPS masterbatch and XPS foam manufacturing. The EASE model estimated this exposure to be 0.1
mg/cm2-day two-hand surface area of 1,070 cm2. Using EPA's two-hand surface area of 1,070 cm2, this
results in a dose of 107 to 1,070 mg/day.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed inhalation exposure concentrations presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
Page 220 of 723
-------
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations based on inhalation exposure concentration monitoring
data that (a) are the preferred type of monitoring data (i.e., 8-hr TWA personal breathing zone data), (b)
are surrogate data pertaining to a worker activity that is certainly relevant to the assessed exposure
scenario, and (c) comprise multiple datasets.
The limitation of the assessment approach is the estimation of inhalation exposure concentrations based
on worker monitoring data that pertain to the worker activity that is expected to result in the largest
exposure but that do not pertain to other worker activities.
There is uncertainty in the estimated HBCD potential inhalation exposure concentrations because the
bases of these concentrations are data that pertain to workers in Europe and the extent to which these
concentrations represent the distribution of inhalation exposure air concentrations pertaining to workers
in the U.S. is uncertain. Refer to Section 2.4.1.14 for additional discussion of uncertainty. Based on
these strengths, limitation, and uncertainty, EPA has medium confidence in the assessed occupational
inhalation exposure air concentrations.
2.4.1,4 Processing of HBCD to Produce XPS Foam using XPS Masterbatch
Workers may be exposed to HBCD while manually unloading and transferring XPS masterbatch directly
into the extruder or into equipment used to feed the XPS masterbatch into the extruder. This manual
transfer may result in worker inhalation exposure to HBCD dust that was generated from abrasion of the
XPS masterbatch pellets or granules during transport (OECD 2009). Manual transfers may also result in
worker dermal exposure to solid HBCD. Additionally, the generated dust from these transfer activities
may result in ONU inhalation exposure to HBCD and ONU dermal exposure through contact with
surfaces where HBCD dust has settled.
Workers may also be potentially exposed from occasional cleaning of process equipment and cutting of
the foam (i.e., secondary processing) into slabs or other shapes (ECHA. 2009b).
Number of Potentially Exposed Workers and Occupational Non-Users
The 2016 CDR data identifies multiple submissions that claim industrial use in the "construction" and
"plastics product manufacturing" sectors (U.S. EPA. 2016c). These industrial sectors are broad and can
include a variety of sites, including sites that do not produce or install XPS and EPS foam, thus the
reported estimates of number of workers potentially exposed at these sites may not be applicable to this
exposure scenario.
EPA used workers and ONU estimates determined from an analysis of BLS data for the NAICS code
326140, Polystyrene Foam Product Manufacturing. These data indicate that there are, on average, 20
workers and 6 ONUs per site within NAICS code 326140. Based on these data and one modeled site for
the manufacturing of XPS foam from XPS masterbatch, EPA estimated that a total of 20 workers and 6
ONUs are potentially exposed in this exposure scenario.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to HBCD occupational
inhalation exposure concentration monitoring data pertaining to the exposure scenario discussed in this
section. The EURAR (ECHA. 2008b) includes HBCD occupational inhalation exposure monitoring data
pertaining to the manufacturing of XPS Foam at multiple sites in Europe using XPS masterbatch and
Page 221 of 723
-------
these data are presented in Table 2-75. As detailed in this table, these data pertain to various worker
activities or parts of the process for production of XPS foam from XPS masterbatch. These data were
obtained by sampling dust and analyzing the samples for HBCD (ECHA 2008b). Workers are
potentially exposed to HBCD contained in dust comprising airborne fragments of XPS foam (ECHA.
2.009b) or XPS masterbatch. Each of the data in Table 2-75 have an overall confidence rating of high as
determined via EPA's systematic review but EPA selected only the data in this table that pertain to the
Secondary Processing of XPS foam as the estimates of HBCD inhalation exposure concentrations
because EPA cannot calculate the mean and 95th percentile of all of the data given in this table. EPA
cannot calculate these statistical values because the individual data points of the various dataset of Table
2-75 are not reported in the EURAR.
Most of the samples associated with the Searl and Robertson (2005) datasets that are noted in the table
as (5a) and (5b) contained HBCD at levels below the detection limit. Specifically, HBCD was detected
in only three of the fourteen dust samples associated with the Searl and Robertson (2005) datasets noted
in the table as 5a and 5b. Nine of these fourteen samples were taken during the secondary processing of
XPS foam (the Searl and Robertson (2005) dataset (5a)), which EPA interprets to mean cutting, sawing,
and machining of XPS foam to manufacture shaped products (discussed Section 2.4.1.6) and the other
five samples were taken during XPS foam reclamation (the Searl and Robertson (2005) data set (5b)),
which is the shredding and reprocessing of process waste (ECHA. 2009b).
Although HBCD was not detected in most of the samples associated with the Secondary Processing of
XPA foam, EPA selected the data that pertains to this part of the process because these data include
larger values and a wider range of exposure concentrations as compared with the data that pertain to the
other parts of the process or worker activities. In conclusion, EPA estimated worker exposure to HBCD
during the production of XPS foam from masterbatch using the mean and high-end values of the data
that pertain to the secondary processing of XPS foam: 0.08 mg/m3 as a central tendency estimate of
exposure concentration and the 90th percentile value of 0.22 mg/m3 as the high-end estimate of exposure
concentration.
As discussed in Section 2.2.4, EPA estimated a range of release days of 1 to 16 days/year for air
releases. EPA expects this range of release days is reflective of the operating days during which HBCD
is processed at a converting site and workers are potentially exposed to HBCD. EPA used the midpoint
of the range of exposure frequency and rounded up when the midpoint resulted in fractions of days, to
calculate central tendency average daily dose. EPA used the high-end of this range of exposure
frequency to calculate high-end average daily dose. Additionally, EPA estimated worker exposure over
the full working day, or eight hours/day, as the data used to estimate inhalation exposures is 8-hour
TWA data.
Page 222 of 723
-------
Table 2-75. Summary of Inhalation Monitoring Data for the Manufacture of XPS Foam Using XPS Masterbatch Containing HBCD
Literature
Study a
Exposure
Scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or
Sampling Location
Exposure
Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Source
Overall Confidence
Rating
Inhalation Monitoring Data Used to Estimate Worker Exposure
Searl and
Robertson
(2005) - 5a
Processing of
HBCD to
Produce XPS
from XPS
Masterbatch
HBCD
in XPS
foam
NR
Secondary processing of
XPS foam
Mean: 0.08
90th percentile:
0.22 c
9
8-hr TWA
Orieinal source: (Searl and
Robertson 2005)
Reported in: (ECHA
2009b.2008b)
High
Other Inhalation Monitoring Data for Handling of XPS Foam
Searl and
Robertson
(2005) -
5b
Processing of
HBCD to
Produce XPS
from XPS
Masterbatch
HBCD
in XPS
foam
NR
Reclamation of XPS foam
- including shredding and
reprocessing of process
waste
Mean: 0.02
90th percentile:
0.02 c
5
8-hr TWA
Orieinal source: (Searl and
Robertson 2005)
Reported in: (ECHA
2009b.2008b)
High
Searl and
Robertson
(2005) -
5c
Processing of
HBCD to
Produce XPS
from XPS
Masterbatch
HBCD
in XPS
foam
NR
Other process control
operators
Mean: 0.03
90th percentile:
0.03 c
4
8-hr TWA
Orieinal source: (Searl and
Robertson 2005)
Reported in: (ECHA
2009b.2008b)
High
Searl and
Robertson
(2005) -
5d
Processing of
HBCD to
Produce XPS
from XPS
Masterbatch
XPS
Master-
batch
NR
Process operators
handling XPS
masterbatch
Mean: 0.03
90th percentile:
0.03 c
24
8-hr TWA
Orieinal source: (Searl and
Robertson 2005)
Reported in: (ECHA
2009b.2008b)
High
NR = Not Reported; N/A = Not Applicable
a - Where multiple datasets are reported in a single literature source, EPA distinguished the various datasets as la, lb, 2a, 2b, etc.
b - The statistical values were obtained from the referenced literature source and were not calculated by EPA.
c - The EURAR defines the secondary processing of EPS foam as the cutting, sawing, and machining of EPS foam and therefore EPA assumed the term "secondary processing
of XPS foam" to mean cutting, sawing and machining of XPS foam.
Page 223 of 723
-------
Dermal Exposure Assessment
As described in Section 2.4.1.1. EPA calculated dermal exposure assuming two-hand contact to solid
XPS masterbatch containing 70% HBCD (NICNAS 2012b; E( )08b). EPA used this weight
fraction because workers at sites that produce XPS foam from XPS masterbatch have the highest
potential dermal exposure concentration to HBCD during the unloading of XPS masterbatch. Using this
model and 70% HBCD, EPA calculated the potential dose for a worker to be 2,170 mg HBCD/day
(high-end) and 630 mg HBCD/day (central tendency). The EURAR and NICNAS report do not estimate
dermal exposures during this operation.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed inhalation exposure concentrations presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations to be equal to worker monitoring data pertaining to the
exposure scenario and the selection of a dataset of the worker monitoring data that includes the largest
range of values as the basis of the estimated exposure concentrations.
The limitations of the assessment approach are the following: (a) the estimation of inhalation exposure
concentrations to be equal to monitoring data pertaining to only a part of the process for the manufacture
of XPS foam using XPS masterbatch and not all of this process and (b) the worker monitoring data that
are the basis of the estimated inhalation exposure concentrations are not the preferred type because the
type of sampling (personal breathing zone or area monitoring) is not reported for this data.
There is uncertainty in the estimated HBCD potential inhalation exposure concentrations because most
of the worker monitoring data that are the basis of the estimated inhalation exposure concentrations are
non-detects (specifically, 3 or fewer of the total of 9 samples are non-detects). Also, these concentrations
pertain to workers in Europe and the extent to which these concentrations represent the distribution of
inhalation exposure air concentrations pertaining to workers in the U.S. is uncertain. Refer to Section
2.4.1.14 for additional discussion of uncertainty. Based on these strengths, limitations, and uncertainty,
EPA has medium confidence in the assessed occupational inhalation exposure air concentrations.
2.4.1.5 Processing of HBCD to Produce XPS Foam Using HBCD Powder
Workers are expected to manually unload and transfer HBCD powder directly into the extruder or into
equipment used to feed the powder into the extruder. This manual transfer may result in worker
inhalation exposure to HBCD dust and dermal exposure to solid HBCD. Additionally, the generated dust
from these transfer activities may result in ONU inhalation exposure to the HBCD dust and ONU dermal
exposure through contact with surfaces where HBCD dust has settled.
Workers may also be potentially exposed from occasional cleaning of process equipment and cutting of
the foam into slabs or other shapes, if these activities are manual. However, the unloading of HBCD
powder is expected to present the highest potential exposure to HBCD, as HBCD is at the highest
concentration during this activity.
Number of Potentially Exposed Workers and Occupational Non-Users
The 2016 CDR data identifies multiple submissions that claim industrial use in the "construction" and
"plastics product manufacturing" sectors (2016 CDR, U.S. EPA. 2016c). These industrial sectors are
Page 224 of 723
-------
broad and can include a variety of sites, including sites that do not produce or install XPS and EPS
foam, thus the reported estimates of number of workers potentially exposed at these sites may not be
applicable to this exposure scenario.
EPA used workers and ONU estimates determined from an analysis of BLS data for the NAICS code
326140, Polystyrene Foam Product Manufacturing. These data indicate that there are, on average, 20
workers and 6 ONUs per site within NAICS code 326140. Based on this data and one modeled site for
the manufacturing of XPS foam from HBCD powder, EPA estimated that a total of 20 workers and 6
ONUs are potentially exposed during this exposure scenario.
Inhalation Exposure Assessment
EPA estimated HBCD potential exposure concentrations to be equal to the assessed exposure
concentrations reported in the EURAR (ECH.A. 2008b) that pertain to all polymer processing operations
involving standard grade HBCD. These assessed exposure concentrations of the EURAR are based on
HBCD occupational inhalation exposure concentrations that pertain to the manufacture of EPS resin
beads. EPA considered HBCD occupational inhalation exposure concentration data that pertain to the
exposure scenario that is the subject of this section but chose that the assessment approach mentioned
above.
EPA identified other monitoring data pertaining to this exposure scenario with overall confidence
ratings of high as determined via EPA's systematic review. These data are given in Table 2-74. in
Section 2.4.1.3. Specifically, the data in this table referenced here are the data of Abbott (2001).
Thomsen (2007) and the data of Searl and Robertson (2005) that are noted in this table as (4). EPA
expects the handling of HBCD standard grade powder to result in the largest potential exposure
concentrations, and therefore EPA did not incorporate these data into the estimate of potential exposure
concentrations because these data are not 8-hr TWA personal breathing zone data that are only
associated with HBCD standard grade powder.
EPA expects the worker activity of manual addition of HBCD to process equipment to result in the
largest potential exposure concentration. Therefore, as in the case of the exposure scenario of
compounding of polystyrene resin to produce XPS masterbatch, EPA estimated HBCD inhalation
exposure concentrations to be equal to the assessed exposure concentrations reported in the EURAR
(ECH.A. 2008b) that pertain to all polymer processing operations involving standard grade HBCD.
Specifically, EPA estimated high-end and central tendency exposure concentrations to be equal to the
"reasonable worst-case" exposure concentration of the EURAR of 2.5 mg/m3 and the "typical" exposure
concentration of the EURAR of 1.25 mg/m3, respectively. Refer to Section 2.4.1.3 for the discussion of
these data of the EURAR.
Dermal Exposure Assessment
As described in Section 2.4.1.1. EPA calculated dermal exposure assuming two-hand contact to solid
containing 100% HBCD (NICNAS 2012b; ECH.A. 2008b). EPA used this weight fraction because
workers at sites that produce XPS foam from HBCD powder have the highest potential dermal exposure
concentration to HBCD during the unloading of HBCD powder. Using this model and 100% HBCD,
EPA calculated the potential dose for a worker to be 3,100 mg HBCD/day (high-end) and 900 mg
HBCD/day (central tendency).
The EURAR estimated dermal exposure for the use of HBCD standard grade powder as an additive in
XPS masterbatch and XPS foam manufacturing. The EASE model estimated this exposure to be 0.1
Page 225 of 723
-------
mg/cm2-day. Using EPA's two-hand surface area of 1,070 cm2, this results in a dose of 107 mg/day. The
NICNAS report uses EASE to model dermal exposure during the addition and weighing of HBCD into
processes. EASE estimated a dermal dose rate of 0.1 to 1 mg/cm2-day. Using EPA's two-hand surface
area of 1,070 cm2, this results in a dose of 107 to 1,070 mg/day. The EASE estimates provided in the
EURAR and NICNAS are lower than that estimated by EPA (3,100 mg/day) as the EPA/OPPTDirect 2-
HandDermal Contact with Solids Model predicts a higher quantity of solids on skin per day.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed inhalation exposure concentrations presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations based on inhalation exposure concentration monitoring
data that (a) are the preferred type of monitoring data (i.e., 8-hr TWA personal breathing zone data), (b)
are surrogate data pertaining to a worker activity that is certainly relevant to the assessed exposure
scenario, and (c) comprise multiple datasets.
The limitation of the assessment approach is the estimation of inhalation exposure concentrations based
on worker monitoring data that pertain to the worker activity that is expected to result in the largest
exposure but that do not pertain to other worker activities.
There is uncertainty in the estimated HBCD potential inhalation exposure concentrations because the
bases of these concentrations are data that pertain to workers in Europe and the extent to which these
concentrations represent the distribution of inhalation exposure air concentrations pertaining to workers
in the U.S. is uncertain. Refer to Section 2.4.1.14 for additional discussion of uncertainty. Based on
these strengths, limitation, and uncertainty, EPA has medium confidence in the assessed occupational
inhalation exposure air concentrations.
2,4.1,6 Processing of HBCD to Produce EPS Foam from Imported EPS Resin
Beads
EPS foam is produced from EPS resin beads by conditioning the beads and using molds to form blocks
of foam (further described in Section 2.2.6), and then followed by the secondary processing of the foam.
The secondary processing of EPS foam include the cutting, sawing and machining of EPS foam (ECHA,
2.008b). This is done to produce sheets or customer-required shapes (NICNAS. 2012b). and results in
cuttings and sawdust that are recycled within the plant (ECHA.. 2008b); the cuttings are granulated prior
to recycle (NICNAS. 2012b). Worker exposure to HBCD as a result of the conditioning of beads and the
formation of foam in molds is expected to be low based on the process description and because HBCD is
encapsulated in the EPS resin beads at a low concentration (<1 wt%) (NICNAS. 2012b). According to
HBCD importers in Australia, the cutting of XPS/EPS foam by manually sawing it or by using a hot
wire is unlikely to produce inhalable particles (NICNAS. 2012b). The EURAR (ECHA. 2008b) does not
include an assessment of occupational exposure pertaining to the manufacture of EPS foam from EPS
resin beads. According to the EURAR (ECHA. 2008b) worker exposure resulting from the cutting of
XPS/EPS foam that generates dust and from heating XPS/EPS with a hot wire is probably lower than
exposure resulting from handling of pure HBCD. EPA assessed potential worker and ONU exposure to
the dust that is generated during the secondary processing of EPS foam.
Page 226 of 723
-------
Number of Potentially Exposed Workers and Occupational Non-Users
The 2016 CDR data identifies multiple submissions that claim the industrial use in the "construction"
and "plastics product manufacturing" sectors ( ;). These industrial sectors are broad and
can include a variety of sites, including sites that do not product or install XPS and EPS foam, thus the
reported estimates of number of workers potentially exposed at these sites may not be applicable to this
exposure scenario.
EPA used workers and ONU estimates determined from an analysis of BLS data for the NAICS code
326140, Polystyrene Foam Product Manufacturing. These data indicate that there are, on average, 20
workers and 6 ONUs per site within NAICS code 326140. Based on these data and one modeled site for
the manufacturing of EPS foam from imported EPS resin beads, EPA estimated that a total of 20
workers and 6 ONUs are potentially exposed during this exposure scenario.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to surrogate HBCD
occupational inhalation exposure concentration monitoring data. The surrogate monitoring data pertain
to the secondary processing of XPS foam which is part of the process of the manufacture of XPS Foam
using XPS masterbatch.
The EURAR (ECHA. 2008b) includes worker inhalation exposure monitoring data pertaining to the
secondary processing of XPS foam but this process is not described. EPA assumed this process is
similar to the secondary processing of EPS foam because the manufacture of XPS foam includes the
trimming of XPS foam to desired shapes (ECHA. 2008b). Based on this, EPA determined the worker
inhalation exposure monitoring data pertaining to the secondary processing of XPS foam to be surrogate
data. This monitoring data is the data of Searl and Robertson (2005). which is presented in Table 2-74
and noted in this table as 5a. EPA estimated central tendency and high-end exposure concentrations to
be equal to the mean value of 0.08 mg/m3, and the 90th percentile value of 0.22 mg/m3of the surrogate
monitoring data, respectively. Refer to Section 2.4.1.4 for a discussion of this monitoring data. Cutting
of EPS foam can be done by machine using sawing or hot wire cutting or by handsaw (NICNAS.
2012b.) Hence, as discussed in Section 2.4.1.1, workers are possibly exposed to HBCD nanoparticles as
a result of cutting of EPS foam with a hot wire.
As discussed in Section 2.2.6, EPA estimated a range of release days of 16 to 140 days/year. EPA
expects this range of release days is also reflective of the operating days during which HBCD is
processed at a converting site and workers are potentially exposed to HBCD. EPA used the midpoint of
this range of exposure frequency, rounded up where the midpoint resulted in fractions of days, to
calculate central tendency average daily dose. EPA used the high-end of this range of exposure
frequency to calculate high-end average daily dose. Additionally, EPA estimated worker exposure over
the full working day, or eight hours/day, as the data used to estimate inhalation exposures is 8-hour
TWA data.
Dermal Exposure Assessment
EPA did not find data on potential levels of dermal exposure for workers engaged in activities related to
the production of EPS foam from EPS resin beads. The EURAR and Australian risk assessment did not
assess dermal exposures during this exposure scenario (NICNAS 2012b; ECHA. 2008b). HBCD is
entrained in the imported EPS resin beads and the potential dermal exposure from handling EPS and
XPS foams containing HBCD is low due to the small weight fraction of HBCD in the foam and because
HBCD is incorporated into the foam matrix, thus is not readily available for exposure (NICNAS 2012b;
Page 227 of 723
-------
ECHA 2008b). Due to the same considerations, dermal exposures to HBCD during this exposure
scenario are not expected.
Strengths, Limitations, and Confidence in Assessment Results
EPA has low to medium confidence in the assessed inhalation exposure concentrations presented above.
EPA considered the quality of the data, the assessment approach, and uncertainties in assessment results
to determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations to be equal to surrogate occupational inhalation
exposure concentration monitoring data.
The limitations of the assessment approach are the following: (a) the worker monitoring data that are the
basis of the estimated inhalation exposure concentrations are not the preferred type because the type of
sampling (personal breathing zone or area monitoring) is not reported for this data and (b) potential
worker exposure resulting from hot wire cutting of EPS foam is not estimated.
The uncertainty in the estimated HBCD potential inhalation exposure concentrations are as follows.
First, as discussed Section 2.4.1.4, most of the worker monitoring data that are the basis of the estimated
inhalation exposure concentrations are non-detects (specifically, three or fewer of the total of nine
samples are non-detects.) Second, there is uncertainty about the extent to which the surrogate inhalation
exposure concentration monitoring data are valid surrogate data because EPA is uncertain that the
secondary processing of XPS foam and the secondary processing of EPS foam are equivalent in terms of
worker exposure. Third, the extent to which the estimated occupational inhalation exposure
concentration data, which are data that pertain to workers in Europe, represent the distribution of
inhalation exposure air concentrations pertaining to workers in the U.S. is uncertain. Refer to Section
2.4.1.14 for additional discussion of uncertainty. Based on these strengths, limitation, and uncertainties,
EPA has low to medium confidence in the assessed occupational inhalation exposure air concentrations.
Page 228 of 723
-------
Literature
Study a
Exposure
Scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or
Sampling Location
Exposure
Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Source
Overall
Confidence
Rating
Inhalation Monitoring Data Used to Estimate Worker Exposure
Searl and
Robertson
(2005) - 5a
Manufacture of
XPS from XPS
Masterbatch
HBCD in
XPS foam
NR
Secondary processing
of XPS foam
Mean: 0.08
90th
percentile:
0.22 c
9
8-hr TWA
Original source:
(Searl and
Robertson 2005)
Reported in:
(ECHA 2008b):
(ECHA 2009b)
High
Other Inhalation Monitoring or Air Concentration Data for the Handling of XPS and EPS Foam
Searl and
Robertson
(2005) -
5b
Manufacture of
XPS from XPS
Masterbatch
HBCD in
XPS foam
NR
Reclamation of XPS
foam - including
shredding and
reprocessing of process
waste
Mean: 0.02
90th
percentile:
0.02 c
5
8-hr TWA
Original source:
(Searl and
Robertson 2005)
Reported in:
(ECHA 2008b):
(ECHA 2009b)
High
Searl and
Robertson
(2005) -
5c
Manufacture of
XPS from XPS
Masterbatch
Uncertain:
HBCD in
XPS foam
or XPS
Masterbatch
NR
Other process control
operators
Mean: 0.03
90th
percentile:
0.03 c
4
8-hr TWA
Original source:
(Searl and
Robertson 2005)
Reported in:
(ECHA 2008b):
(ECHA 2009b)
High
NR = Not Reported; N/A = Not Applicable
a - Where multiple datasets were available from one literature source, EPA distinguished data as la, lb, 2a, 2b, etc.
b - The statistical values were obtained from the referenced literature source and were not calculated by EPA.
Page 229 of 723
-------
2,4.1,7 Processing of HBCD to Produce SIPs and Automobile Replacement Parts
from XPS/EPS Foam
Workers are likely to manually unwrap and further handle the XPS and EPS foam boards during which
they will likely have dermal contact with the foam; however, HBCD is expected to be incorporated in
the foam matrix and not readily available for exposure (NICNAS 2012b). Attrition of the foam boards
during transportation to sites at which SIPs and automobile replacement parts are manufactured is
unlikely because of the large size of the boards and the limited opportunity for rubbing of boards against
each other. Therefore, worker inhalation exposure during unwrapping of the boards to dust resulting
from the attrition of the boards is unlikely (U.S. EPA. 2014a). To manufacture SIPs, the XPS and EPS
foam is cut into the desired size panel, either with saws or thermal wires (NICNAS 2012b; ECH.A.
2.008b). The panels are then adhered to steel, plastic, concrete, plasterboard, or other sheathing material
on either side, forming a sandwich, which is why these panels are also referred to as sandwich panels
(NICNAS 2012b). Once the SIPs are produced, they are shipped to construction sites for installation.
Cutting of the XPS and EPS foam results in particle generation that pose potential for worker and ONU
inhalation exposure.
Number of Potentially Exposed Workers and Occupational Non-Users
EPA estimated exposures for workers at two sites based on the methodology described in Section
2.4.1.1. The 2016 CDR data identify multiple submissions that claim the industrial use in the
"construction" and "plastics product manufacturing" sectors (U.S. EPA. 2016c). These industrial sectors
can include a variety of sites, including XPS and EPS foam sites and construction sites, thus the reported
estimates of number of workers potentially exposed at these sites may not be applicable to this exposure
scenario.
EPA used workers and ONU estimates determined from an analysis of BLS data for the NAICS code
326140, Polystyrene Foam Product Manufacturing. These data indicate that there are, on average, 20
workers and 6 ONUs per site within NAICS code 326140. Based on these data and one site for each of
the SIPs and automobile replacement part production, EPA estimated that a total of 39 workers and 11
ONUs are potentially exposed during this exposure scenario. EPA used unrounded figures for the
number of workers and ONUs per site to calculate these totals, resulting in the slight discrepancy.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to surrogate HBCD
occupational inhalation exposure concentration monitoring data. The surrogate monitoring data pertain
to the secondary processing of XPS foam which is part of the process of the manufacture of XPS Foam
using XPS masterbatch.
The EURAR (ECH.A. 2.008b) includes worker inhalation exposure monitoring data pertaining to the
secondary processing of XPS foam but this process is not described. As discussed in Section 2.4.1.6,
EPA assumed this process is similar to the secondary processing of EPS foam and hence comprises
cutting, sawing and/or machining of XPS foam. Based on this, EPA determined the worker inhalation
exposure monitoring data pertaining to the secondary processing of XPS foam to be surrogate data.
These monitoring data are the data of Searl and Robertson (2005). which is presented in Table 2-75 and
noted in this table as 5a. EPA estimated central tendency and high-end exposure concentrations to be
equal to the mean value of 0.08 mg/m3, and the 90th percentile value of 0.22 mg/m3of the surrogate
monitoring data, respectively. Refer to Section 2.4.1.4 for a discussion of this monitoring data. As
Page 230 of 723
-------
discussed in Section 2.4.1.1, workers are possibly exposed to HBCD nanoparticles as a result of cutting
of EPS foam with a hot wire.
As discussed in Section 2.2.7, EPA estimated a range of release days of 16 to 300 days/year. EPA
expects this range of release days is also reflective of the operating days during which HBCD is
processed at foam cutting sites and workers are potentially exposed to HBCD. However, EPA does not
expect that workers will be exposed greater than 250 day/year, accounting for a worker schedule of five
days per week and 50 weeks per year. Based on this, EPA estimated worker exposures over a range of
16 to 250 days/year. EPA used the midpoint of this range of exposure frequency, rounded up where the
midpoint resulted in fractions of days, to calculate central tendency average daily dose. EPA used the
high-end of this range of exposure frequency to calculate high-end average daily dose. Additionally,
EPA estimated worker exposure over the full working day, or eight hours/day, as the data used to
estimate inhalation exposures is 8-hour TWA data.
Dermal Exposure Assessment
EPA did not find data on potential levels of dermal exposure for workers engaged in activities related to
the manufacturing of SIPs and automobile replacement parts from XPS and EPS foam. The EURAR and
Australian risk assessment did not assess dermal exposures during this exposure scenario, with both
reports stating that these exposures are expected to be low because HBCD is incorporated into the foam
matrix, thus is not readily available for exposure flSflCNAS 2012b; EC |08b). The potential dermal
exposure from handling EPS and XPS foams containing HBCD is low due to the small weight fraction
of HBCD in the foam and because HBCD is incorporated into the foam matrix, thus is not readily
available for exposure (NICNAS 2012b; ECHA 2008b). Due to the same considerations, dermal
exposures to HBCD during this exposure scenario are not expected.
Strengths, Limitations, and Confidence in Assessment Results
EPA has low to medium confidence in the assessed inhalation exposure concentrations presented above.
EPA considered the quality of the data, the assessment approach, and uncertainties in assessment results
to determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations to be equal to surrogate occupational inhalation
exposure concentration monitoring data.
The limitations of the assessment approach are the following: (a) the worker monitoring data that are the
basis of the estimated inhalation exposure concentrations are not the preferred type because the type of
sampling (personal breathing zone or area monitoring) is not reported for this data and (b) potential
worker exposure resulting from hot wire cutting of EPS foam is not estimated.
The uncertainty in the estimated HBCD potential inhalation exposure concentrations are as follows.
First, as discussed in Section 2.4.1.4, most of the worker monitoring data that are the basis of the
estimated inhalation exposure concentrations are non-detects (specifically, three or fewer of the total of
nine samples are non-detects.) Second, there is uncertainty about the extent to which the surrogate
inhalation exposure concentration monitoring data are valid surrogate data because EPA is uncertain that
the secondary processing of XPS foam and the manufacture of SIPs and replacement auto parts from
XPS/EPS foam are equivalent in terms of worker exposure. Third, the extent to which the estimated
occupational inhalation exposure concentration data, which are data that pertain to workers in Europe,
Page 231 of 723
-------
represent the distribution of inhalation exposure air concentrations pertaining to workers in the U.S. is
uncertain. Refer to Section 2.4.1.14 for additional discussion of uncertainty. Based on these strengths,
limitation, and uncertainties, EPA has low to medium confidence in the assessed occupational inhalation
exposure air concentrations.
2.4.1.8 Use: Installation of Automobile Replacement Parts
EPA does not expect that workers at automobile repair sites further process the replacement parts
containing HBCD. Because the automobile replacement parts are received at repair shops as finished
articles containing XPS and EPS foam, in which HBCD is incorporated into the foam matrix, inhalation
and dermal exposures are not expected fNICNAS 2012b; ECHA 2008b).
2.4.1.9 Use: Installation of XPS/EPS Foam Insulation in Residential, Public and
Commercial Buildings, and Other Structures
Workers may saw or cut XPS/EPS foam boards at construction sites (ECHA. 2008b). The boards are
sawed with a bandsaw or manually fNICNAS 2012b) and cut with a knife or a hot wire fNICNAS
2012b; ECHA 2008b). The EURAR and NICNAS do not include an assessment of the occupational
exposure scenario. According to HBCD importers in Australia, the cutting of XPS/EPS foam boards by
manually sawing it or by using a hot wire is unlikely to produce inhalable particles but NICNAS does
not include any corroborating data fNICNAS. 2012b). According to the EURAR (ECHA. 2008b).
worker exposure resulting from the cutting of XPS/EPS foam boards that generates dust and from
heating XPS/EPS with a hot wire is probably lower than exposure resulting from handling of pure
HBCD. As discussed in Section 2.2.9, the amounts of XPS/EPS particles generated from sawing and
cutting XPS/EPS foam boards are reported in the EURAR but the particle sizes are not given. EPA
assessed potential worker exposure to the dust that is generated during the sawing or cutting of XPS/EPS
foam boards. ONUs may inhale this dust.
Number of Potentially Exposed Workers and Occupational Non-Users
As discussed in Section 2.2.9, EPA estimated the number of potential construction sites to be as few as
34 large construction sites (assumes HBCD use rate estimated for large-scale use) and as high as
2,696 residential construction sites (assumes HBCD use rate estimated for residential use) may install
insulation containing HBCD in a year.
EPA analyzed information from the Bureau of Labor Statistics for the NAICS code 238310, Dry wall
and Insulation Contractors, to determine an estimate of the number of workers and ONUs that may be
present at a construction site. These data indicate that there are, on average, 8 workers and 1 ONU per
contractor establishment within NAICS code 238310. Due to the low estimate of workers and ONUs per
establishment, EPA assumes that this estimate represents the size of one work crew and that one crew
would be present at job sites {i.e., construction sites) at a given time. Thus, EPA estimated 8 workers
and 1 ONU per job site. Furthermore, EPA assumes that different crews from separate contractor
establishments may install insulation containing HBCD and that these crews may install insulation
containing HBCD at more than one job site in a year, although there is the potential for variability.
Using these data for number of workers and ONUs and the lower value estimate of 34 construction sites,
a total of approximately 310 workers and 30 ONUs are potentially exposed. Using these data and the
upper value estimate of 2,696 residential construction sites, a total of approximately 25,000 workers and
2,400 ONUs are potentially exposed. EPA expects that this range accounts for both the scenario that job
crews may install insulation containing HBCD at multiple sites through a year and the scenario that a
job crew will only install insulation containing HBCD at one site in a year. These data are summarized
in Table 2-77. EPA used unrounded figures for the number of workers and ONUs per site to calculate
Page 232 of 723
-------
these totals, resulting in the slight discrepancy. EPA recognizes that smaller residential sites likely have
fewer workers than larger sites, thus this is likely an overestimate of the number of potentially exposed
people.
Table 2-77. U.S. Number of Establishments and Employees for Installation of XPS/EPS Foam
nsulation in Residential, Public and Commercial Buildings, and Other Structures
2016 NAICS
2016 NAICS Title
Number of Job Sites
Number of
Workers per Sitea
Number of ONUs
per Sitea
Lower value
(large commercial
sites)
Upper value
(residential sites)
238310
Drywall and
Insulation
Contractors
34
2,696
9
1
Lower value of total establishments and
number of potentially exposed workers
and ONUs = b
34
310
30
Upper value of total establishments and
number of potentially exposed workers
and ONUs = b
2,696
25,000
2,400
a - Rounded to the nearest whole number and two significant figures,
b - Unrounded figures were used for total worker and ONU calculations.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to surrogate HBCD
occupational inhalation exposure concentration monitoring data. The surrogate monitoring data pertain
to the secondary processing of XPS foam which is part of the process of the manufacture of XPS Foam
using XPS masterbatch.
The EURAR (ECFLA 2008b) includes worker inhalation exposure monitoring data pertaining to the
secondary processing of XPS foam but this process is not described. As discussed in Section 2.4.1.6,
EPA assumed this process is similar to the secondary processing of EPS foam and hence comprises
cutting, sawing and/or machining of XPS foam. Based on this, EPA determined the worker inhalation
exposure monitoring data pertaining to the secondary processing of XPS foam to be surrogate data.
These monitoring data are the data of Searl and Robertson (2005). which is presented in Table 2-74 and
noted in this table as 5a. EPA estimated central tendency and high-end exposure concentrations to be
equal to the mean value of 0.08 mg/m3, and the 90th percentile value of 0.22 mg/m3of the surrogate
monitoring data, respectively. Refer to Section 2.4.1.4 for a discussion of this monitoring data. As
discussed in Section 2.4.1.1, workers are possibly exposed to HBCD nanoparticles as a result of cutting
of EPS foam with a hot wire.
As discussed in Section 2.2.9, EPA estimated a range of release days of 1 to 3 days/year-site. However,
EPA expects that workers may install insulation containing HBCD at multiple sites in a year. EPA does
not expect that workers will be exposed greater than 250 day/year, accounting for a worker schedule of
five days per week and 50 weeks per year. Based on this, EPA expects the minimum number of
exposure days to be 1 day/per year and the maximum number of exposure days to be 250 days/year.
EPA used the midpoint of the range of 1 to 250 days/year of exposure frequency, rounded up to 126
days/year, to calculate central tendency average daily dose. EPA used the high-end of this range of
Page 233 of 723
-------
exposure frequency to calculate high-end average daily dose. Additionally, EPA estimated worker
exposure over the full working day, or eight hours/day, as the data used to estimate inhalation exposures
is 8-hour TWA data.
Dermal Exposure Assessment
The EURAR and Australian risk assessment did not assess dermal exposures during this exposure
scenario (NICNAS 2012b; ECHA 2008b). stating that these exposures are expected to be low. The
potential dermal exposure from handling XPS and EPS foams containing HBCD is low due to the small
weight fraction of HBCD in the foam and because HBCD is incorporated into the foam matrix, thus is
not readily available for exposure (NICNAS iO I l\r, { 1"U \ JOOBb). EPA does not expect dermal
exposures during this exposure scenario due to the same considerations.
Strengths, Limitations, and Confidence in Assessment Results
EPA has low to medium confidence in the assessed inhalation exposure concentrations presented above.
EPA considered the quality of the data, the assessment approach, and uncertainties in assessment results
to determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations to be equal to surrogate occupational inhalation
exposure concentration monitoring data.
The limitations of the assessment approach are the following: (a) the worker monitoring data that are the
basis of the estimated inhalation exposure concentrations are not the preferred type because the type of
sampling (personal breathing zone or area monitoring) is not reported for this data and (b) potential
worker exposure resulting from hot wire cutting of EPS foam is not estimated.
The uncertainty in the assessment results are as follows. First, as discussed in Section 2.4.1.4, most of
the worker monitoring data that are the basis of the estimated inhalation exposure concentrations are
non-detects (specifically, three or fewer of the total of nine samples are non-detects.) Second, there is
uncertainty about the extent to which the surrogate inhalation exposure concentration monitoring data
are valid surrogate data because EPA is uncertain that the secondary processing of XPS foam and the
sawing or cutting of XPS/EPS foam at construction sites are equivalent in terms of worker exposure
because the methods and frequencies of sawing or cutting of XPS/EPS foam boards at construction sites
and the ventilation rates at these sites may be different than the values of these parameters in the case of
industrial sites at which the secondary processing of XPS foam occurs. Refer to Section 2.4.1.14 for
additional discussion of uncertainty. Based on these strengths, limitation, and uncertainties, EPA has
low to medium confidence in the assessed occupational inhalation exposure air concentrations.
2,4,1,10 Demolition and Disposal of XPS/EPS Foam Insulation Products in
Residential, Public and Commercial Buildings, and Other Structures
EPA expects workers may break XPS and EPS foam insulation products during demolition, which may
generate dust that contains HBCD that workers and ONUs may inhale. The waste from demolition sites
will most likely be sent to construction & demolition landfills, incineration facilities, or recycled.
Insulation waste containing HBCD may be further broken down via shredders, or other equipment at
landfill and incineration facilities. Workers and ONUs at these facilities may be exposed to dust
containing HBCD. Occupational exposures during recycling is discussed in Section 2.4.1.11 Recycling
of EPS foam and Reuse of XPS foam.
Page 234 of 723
-------
Solid waste may be first sent to waste transfer facilities, where waste is consolidated onto larger trucks.
At many transfer stations, workers screen incoming waste located on conveyor systems, tipping floors,
or in waste pits to identify recyclables and wastes inappropriate for disposal (e.g., hazardous waste,
whole tires). Workers at transfer stations operate heavy machinery such as conveyor belts, push blades,
balers, and compactors, and may also clean the facility or perform equipment maintenance. Workers
may be exposed to poor air quality due to dust and odor, particularly in tipping areas over waste pits
(https://www.epa.gOv/sites/production/files/z S/documents/r02002.pdf). As reported for a
municipal landfill facility, waste may be dumped onto tipping floors for storage, then fed to a conveyor
system for sorting and eventual shredding of waste. The waste from these processes are either directly
loaded on trucks to be sent into the landfill or deposited in storage pits (Burkhart and Short 1995).
Heavy machinery operators may be exposed to particulates and other contaminates while in the cabs of
the machinery (https://www.cdc.gov/niosh/hhe/reports/pdfs/1996-010 ^ and
https://www.cdc.gov/niosh/hhe/reports/pdfs/1993-0696-2395.pdf). Mechanics servicing equipment may
be exposed to residues on machinery. In addition, workers may be exposed when removing dirty work
uniforms (https://www.cdc.gov/niosh/hhe/reports/pdfs/1996-0109-2616.pdf). EPA expects similar
processing of waste may occur at C&D landfills.
At Municipal Waste Combustors (MWCs), waste materials are not generally handled directly by
workers. Trucks may dump the waste directly into the pit, or waste may be tipped to the floor and later
pushed into the pit by a worker operating a front-end loader. A large grapple from an overhead crane is
used to grab waste from the pit and drop it into a hopper, where hydraulic rams feed the material
continuously into the combustion unit at a controlled rate.
Facilities that used the refuse-derived fuel (RDF) process may conduct on-site sorting, shredding, and
inspection of the waste prior to incineration to recover recyclables and remove hazardous waste or other
unwanted materials. Sorting is usually an automated process that uses mechanical separation methods,
such as trommel screens, disk screens, and magnetic separators. Once processed, the waste material may
be transferred to a storage pit, or it may be conveyed directly to the hopper for combustion. Tipping
floor operations may generate dust. Air from the enclosed tipping floor, however, is continuously drawn
into the combustion unit via one or more forced air fans to serve as the primary combustion air and
minimize odors. Dust and lint present in the air is typically captured in filters or other cleaning devices
in order to prevent the clogging of steam coils, which are used to heat the combustion air and help dry
higher-moisture inputs (Kj 2).
Number of Potentially Exposed Workers and Occupational Non-Users
EPA did not find information regarding the number of workers typically on a demolition site. To
estimate the number of workers potentially exposed per site, EPA assumed that demolition is
accomplished by workers who remove the insulation, as the insulation may be recycled or reused as
discussed in Section 2.2.11. To estimate the number of these workers, EPA assumed that this number of
workers is equivalent to the number of workers who install foam panels and utilized the same
methodology for estimating workers potentially exposed during the installation of insulation into
buildings, as described below and in Section 2.4.1.9.
As described in Section 2.4.1.9, EPA analyzed information from the Bureau of Labor Statistics (BLS)
for the NAICS code 238310, Dry wall and Insulation Contractors, to determine an estimate of the
number of workers and ONUs that may be present at a demolition site. These data indicate that there are,
on average, 8 workers and 1 ONU per contractor establishment within NAICS code 238310. Using these
data for number of workers and ONUs and the lower value estimate of 578 demolition sites, a total of
Page 235 of 723
-------
approximately 5,300 workers and 510 ONUs are potentially exposed. Using these data and the upper
value estimate of 45,832 residential demolition sites, a total of approximately 420,000 workers and
40,000 ONUs are potentially exposed.
For potential workers handling C&D waste, EPA reviewed data from the BLS for NAICS code 562212,
Solid Waste Landfill, and related Standard Occupational Classification (SOC) codes, there are on
average an estimated 3 workers and 2 ONUs per site at landfill facilities. An analysis using the BLS for
NAICS code 562219, Other Nonhazardous Waste provided the same estimate. Using BLS for NAICS
code 562213, Solid Waste Combustors and Incinerators, and related SOC codes, there are on average an
estimated 13 workers and 8 ONUs per incineration site. As stated in Section 2.2.10, EPA identified
estimates of 1,120 to 1,577 active C&D landfills and up to 107 waste-to-energy facilities in the U.S. It is
likely that some of these facilities may not receive insulation waste containing HBCD, depending on the
type of waste accepted at the facility and prevalence of XPS/EPS foam insulation containing HBCD in
nearby areas. An upper bound estimate would be 4,731 workers and 3,154 ONUs for solid waste
landfills and 1,391 workers and 856 ONUs for solid waste incinerators.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations in accordance with an estimation
method that is based on the Occupational Safety and Health Administration (OSHA) permissible
exposure limit (PEL) for particulates not otherwise regulated (PNOR) ( |). That is, EPA
estimated the HBCD potential inhalation exposure concentrations by multiplying the OSHA PEL for
PNOR, which is 15 mg/m3 for total dust, by the HBCD concentration in XPS and EPS foam, which are 2
wt% and 0.7 wt%, respectively (ECHA. 2008b). This modeling approach assumes that dust generated is
only from XPS/EPS foam and is proportional to the concentration of HBCD in the foam. Hence, EPA
calculated potential HBCD exposure concentrations ranging from 0.105 to 0.30 mg/m3. The OSHA PEL
for PNOR and EPA's estimate are 8-hour TWA values. The specific value of exposure concentration
using this method is dependent on the proportion of each type of foam, XPS and/or EPS, being broken
down.
EPA considered the use of the data discussed in Section 2.4.1.4, which is data for workers performing
secondary processing of XPS foam, which includes cutting, sawing, or machining of XPS foam. EPA
did not use these data as surrogate for this exposure scenario because, based on the process description,
EPA does not expect the use of the same tools for breaking down of foam in this exposure scenario as
those used for the secondary processing of XPS foam at an XPS foam manufacturing site, resulting in
different dust generation potential. Specifically, as discussed in the process description, this exposure
scenario involves manually breaking foam insulation, demolishing with equipment such as a wrecking
ball, or shredding of foam during waste processing. Based on the process description, the land disposal
for the most part does not involve the intentional breaking of waste although some processing steps such
as compaction and loading and unloading of waste may result in the breaking of articles. This approach
likely overestimates exposure experienced by workers at landfills as discussed below in strengths,
limitations, and confidence in assessment results.
As discussed in Section 2.2.10, EPA estimated a range of release days of 1 to 3 days/year-job site for
demolition sites. However, EPA expects that workers may demolish insulation containing HBCD at
multiple sites in a year. Landfill and incineration operations are expected to run year-round. EPA does
not expect that workers will be exposed greater than 250 day/year, accounting for a worker schedule of
five days per week and 50 weeks per year. Based on this, EPA expects the minimum number of
exposure days to be 1 day/per year and the maximum number of exposure days to be 250 days/year.
Page 236 of 723
-------
Workers may only perform demolition activities intermittently throughout a year. EPA believes the
upper estimate of 250 days/year is likely an overestimate for demolition workers but does not have any
data to estimate the exact number of working days. EPA used the midpoint of the range of 1 to 250
days/year of exposure frequency, rounded up to 126 days/year, to calculate central tendency average
daily dose. EPA used the high-end of this range of exposure frequency to calculate high-end average
daily dose. Additionally, EPA estimated worker exposure over the full working day, or eight hours/day,
as the data used to estimate inhalation exposures is 8-hour TWA limit.
Dermal Exposure Assessment
The EURAR and Australian risk assessment did not assess dermal exposures during this exposure
scenario (NICNAS JO I ";b; K'UA AQ8b). The potential dermal exposure from handling XPS and EPS
foams containing HBCD is low due to the small weight fraction of HBCD in the foam and because
HBCD is incorporated into the foam matrix, thus is not readily available for exposure (NICNAS 2012b;
ECHA 2008b). EPA does not expect dermal exposures to HBCD during this exposure scenario due to
the same considerations.
Strengths, Limitations, and Confidence in Assessment Results
EPA has low to medium confidence in the assessed inhalation exposure concentrations presented above.
EPA considered the uncertainties in assessment results to determine the level of confidence. EPA is
uncertain about the extent to which the OSHA PEL for PNOR is representative of occupational
inhalation exposure air concentrations during demolition of buildings and other structures and the
processing of waste. Inherent to EPA's approach is the assumption that XPS/EPS foam is the source of
all the dust that is generated and this assumption likely results in an overestimate of exposure
concentrations. In particular, EPA expects inhalation exposures for workers at landfill and incineration
facilities to be overestimated. Insulation waste is only a small contributor to waste received at C&D
landfills and site-specific processes may not involve the intentional breaking of waste. EPA estimates
concrete and wood products composed the largest proportion of waste materials at C&D landfills (U.S.
EPA 2018a). Based on these strengths, limitation, and uncertainty, EPA has low to medium confidence
in the assessed occupational inhalation exposure air concentrations.
2.4.1.11 Recycling of EPS Foam and Reuse of XPS Foam
EPS boards are recycled by grinding them and feeding the grinded material to the molding process
together with virgin EPS to form new boards (ECHA. 2008b). As discussed in Section 1.2.6, in the U.S.
the EPS produced by recycling EPS insulation boards is taken to polystyrene product manufacturers.
EPA assumes that the recycling of EPS insulation boards may include secondary processing of EPS,
which is a part of the process for manufacture of EPS from EPS resin beads and can be the cutting,
sawing and machining of the EPS foam (ECHA. 2008b). As discussed in Section 2.4.1.6, EPA believes
the secondary processing of EPS may result in potential worker and ONU inhalation exposure to the
dust that is generated during this process and hence EPA assessed worker potential exposure.
Number of Potentially Exposed Workers and Occupational Non-Users
EPA estimated exposures for workers at two recycling and reuse sites based on the information in
Section 2.2.11. As discussed above, EPS recycling is likely to be performed at sites with similar
operations to those described for EPS foam manufacturing in Section 2.2.6. Thus, EPA assumed the
same number of workers and ONUs as described in Section 2.4.1.6 (Processing of HBCD to Produce
EPS Foam from Imported EPS Resin Beads). For this estimate, EPA utilized worker and ONU estimates
determined from an analysis of BLS data for the NAICS code 326140, Polystyrene Foam Product
Manufacturing. These data indicate that there are, on average, 20 workers and 6 ONUs per site within
NAICS code 326140. Based on these data and two sites for the recycling of EPS foam and reuse of XPS,
Page 237 of 723
-------
EPA estimated that a total of 39 workers and 11 ONUs are potentially exposed during this life cycle
stage. EPA used unrounded figures for the number of workers and ONUs per site to calculate these
totals, resulting in the slight discrepancy.
EPA notes that the number of workers potentially exposed during reuse of XPS may differ from the
estimate above, if XPS is reused directly at construction sites and is not first processed {i.e., cut or
otherwise re-shaped) at industrial processing sites.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to surrogate HBCD
occupational inhalation exposure concentration monitoring data. The surrogate monitoring data pertain
to the secondary processing of XPS foam which is part of the process of the manufacture of XPS Foam.
The EURAR (ECHA. 2008b) includes worker inhalation exposure monitoring data pertaining to the
secondary processing of XPS foam but this process is not described. As discussed in Section 2.4.1.6,
EPA assumed this process is similar to the secondary processing of EPS foam and hence comprises
cutting, sawing and/or machining of XPS foam. Based on this, EPA determined the worker inhalation
exposure monitoring data pertaining to the secondary processing of XPS foam to be surrogate data.
These monitoring data are the data of Searl and Robertson (2005). which is presented in Table 2-75 and
noted in this table as 5a. EPA estimated central tendency and high-end exposure concentrations to be
equal to the mean value of 0.08 mg/m3, and the 90th percentile value of 0.22 mg/m3of the surrogate
monitoring data, respectively. Refer to Section 2.4.1.4 for a discussion of this monitoring data.
As discussed in Section 2.2.11, EPA estimated a range of release days of 1 to 140 days/year. EPA
expects this range of release days is also reflective of the operating days during which foam containing
HBCD is recycled at a converting site and workers are potentially exposed to HBCD. EPA used the
midpoint of this range of exposure frequency, rounded up where the midpoint resulted in fractions of
days, to calculate central tendency average daily dose. EPA used the high-end of this range of exposure
frequency to calculate high-end average daily dose. Additionally, EPA estimated worker exposure over
the full working day, or eight hours/day, as the data used to estimate inhalation exposures is 8-hour
TWA data.
Dermal Exposure Assessment
EPA did not find data on potential levels of dermal exposure for workers engaged in activities related to
the recycling of EPS foam. The EURAR and Australian risk assessment did not assess dermal exposures
during this exposure scenario, with both reports stating that these exposures are expected to be low
because HBCD is incorporated into the foam matrix, thus is not readily available for exposure (NICNAS
2012b; ECHA 2008b). The potential dermal exposure from handling EPS and XPS foams containing
HBCD is low due to the small weight fraction of HBCD in the foam and because HBCD is incorporated
into the foam matrix, thus is not readily available for exposure (NICNAS Am _b, I t'.H \ A'08b). EPA
does not expect dermal exposures to HBCD during this exposure scenario due to the same
considerations.
Strengths, Limitations, and Confidence in Assessment Results
EPA has low to medium confidence in the assessed inhalation exposure concentrations presented above.
EPA considered the quality of the data, the assessment approach, and uncertainties in assessment results
to determine the level of confidence.
Page 238 of 723
-------
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations to be equal to surrogate occupational inhalation
exposure concentration monitoring data.
The limitation of the assessment approach is that the worker monitoring data that are the basis of the
estimated inhalation exposure concentrations are not the preferred type because the type of sampling
(personal breathing zone or area monitoring) is not reported for this data.
The uncertainty in the estimated HBCD potential inhalation exposure concentrations are as follows.
First, as discussed Section 2.4.1.4, most of the worker monitoring data that are the basis of the estimated
inhalation exposure concentrations are non-detects (specifically, three or fewer of the total of nine
samples are non-detects.) Second, there is uncertainty about the extent to which the surrogate inhalation
exposure concentration monitoring data are valid surrogate data because EPA is uncertain that the
secondary processing of XPS foam and the recycling of EPS foam are equivalent in terms of worker
exposure. Third, the extent to which the estimated occupational inhalation exposure concentration data,
which are data that pertain to workers in Europe, represent the distribution of inhalation exposure air
concentrations pertaining to workers in the U.S. is uncertain. Refer to Section 2.4.1.14 for additional
discussion of uncertainty. Based on these strengths, limitation, and uncertainties, EPA has low to
medium confidence in the assessed occupational inhalation exposure air concentrations.
2.4.1.12 Formulation of Flux/Solder Pastes
EPA lacks information about the physical form and concentration of the HBCD received at the
flux/solder paste formulation site and assumed the HBCD is received as a solid either in pure form in
formulations containing nearly 100% HCBD. Workers at the formulation site will likely unload HCBD
into mixing equipment, where the HBCD is mixed with other ingredients and becomes suspended in the
solder flux component formulation. This HBCD transfer may result in worker inhalation exposure to
HBCD dust and dermal exposure to solid HBCD. Additionally, the generated dust from these transfer
activities may result in ONU inhalation exposure to the HBCD dust and ONU dermal exposure through
contact with surfaces where HBCD dust has settled.
Workers may also be potentially exposed from occasional cleaning of process equipment and loading of
formulations into containers to be shipped to China for final formulation of the flux/solder paste.
However, the unloading of HBCD powder is expected to present the highest potential exposure to
HBCD, as HBCD is at the highest concentration during this activity.
Number of Potentially Exposed Workers and Occupational Non-Users
As discussed in Section 2.2.13, EPA estimated exposures for workers at one solder flux component
formulation site.
The number of workers and ONUs potentially exposed during this exposure scenario was estimated
using BLS data for the NAICS code 325998, All Other Miscellaneous Chemical Product and
Preparation Manufacturing. These data are summarized in Table 2-78 below. Based on these data, EPA
estimated that a total of 14 workers and 5 ONUs are potentially exposed during this exposure scenario.
Page 239 of 723
-------
Table 2-78. U.S. Number of Establishments and Employees for Formulation of Solder Flux
Scenario
2016
NAICS
2016 NAICS Title
Number of
Establishments
Number of
Workers per
Site8
Number of
ONUs per
Site8
Formulation of
flux and solder
325998
All Other Miscellaneous
Chemical Product and
Preparation
Manufacturing
1
14
5
a Rounded to the nearest whole number.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations to be equal to the assessed exposure
concentrations reported in the EURAR (ECHA 2008b) that pertain to all polymer processing operations
involving standard grade HBCD. These assessed exposure concentrations of the EURAR are based on
HBCD occupational inhalation exposure concentrations that pertain to the manufacture of EPS resin
beads. EPA determined these data are surrogate data because these data pertain to the worker activity of
manual addition of HBCD to process equipment. EPA estimated high-end and central tendency exposure
concentrations to be equal to, respectively, the "reasonable worst-case" exposure concentration of the
EURAR which is equal to 2.5 mg/m3 and the "typical" exposure concentration of the EURAR which is
equal to one half of the reasonable worst-case, or 1.25 mg/m3. Refer to Section 2.4.1.3, Compounding of
Polystyrene Resin to Produce XPS Masterbatch, for a discussion of EPA's approach to the estimation of
these estimates of exposure concentration.
As discussed in Section 2.2.12, EPA estimated days of release at a formulation site as a range from 5 to
300 days/year. EPA expects this range of release days is also reflective of the operating days during
which HBCD is processed at a formulation site and workers are potentially exposed to HBCD.
However, EPA does not expect that workers will be exposed greater than 250 day/year, accounting for a
worker schedule of five days per week and 50 weeks per year. Based on this information, EPA estimated
worker exposures over the exposure frequency of 5 to 250 days/year. EPA used the midpoint of this
range of exposure frequency, rounded up where the midpoint resulted in fractions of days, to calculate
central tendency average daily dose. EPA used the high-end of this range of exposure frequency to
calculate high-end average daily dose. Additionally, EPA estimated worker exposure over the full
working day, or eight hours/day, as the data used to estimate inhalation exposures is 8-hour TWA data.
Dermal Exposure Assessment
As described in Section 2.4.1.1. EPA calculated dermal exposure assuming two-hand contact to solids
containing 100% HBCD. EPA used this weight fraction because workers have the highest potential
dermal exposure concentration to HBCD during the unloading of HBCD powder, prior to formulation.
EPA calculated the potential dose for a worker to be 3,100 mg HBCD/day (high-end) and 900 mg
HBCD/day (central tendency). The EURAR did not estimate dermal exposures during this exposure
scenario. The NICNAS report did use EASE to model dermal exposure during the addition and
weighing of HBCD into processes, which is covered in this exposure scenario. The NICNAS report
estimated a dermal dose rate of 0.1 to 1 mg/cm2-day. This results in a dose of 107 to 1,070 mg/day,
using EPA's two-hand surface area of 1,070 cm2 (NICNAS 2012b; ECHA 2008b).
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed inhalation exposure concentrations presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
Page 240 of 723
-------
determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high which is a strength of the assessment. The strength of the assessment approach is the
estimation of inhalation exposure concentrations based on surrogate HBCD inhalation exposure
concentration monitoring data that are the preferred type of monitoring data {i.e., 8-hr TWA personal
breathing zone data).
There is uncertainty about the physical form and concentration of the HBCD that is received at the
formulation site and hence there is uncertainty about the extent to which the surrogate data is relevant to
the exposure scenario. Also, there is uncertainty in the estimated HBCD potential inhalation exposure
concentrations because the bases of these concentrations are data that pertain to workers in Europe and
the extent to which these concentrations represent the distribution of inhalation exposure air
concentrations pertaining to workers in the U.S. is uncertain. Refer to Section 2.4.1.14 for additional
discussion of uncertainty. Based on these strengths, limitation, and uncertainty, EPA has medium
confidence in the assessed occupational inhalation exposure air concentrations.
2,4,1.13 Use of Flux/Solder Paste
The technical data sheets for the flux and solder products identified indicates that these formulations are
frequently supplied in small containers, such as syringes and 100-gram jugs (Indium Corporation.
2019b). Workers may be potentially exposed during unloading into application equipment.
Number of Potentially Exposed Workers and Occupational Non-Users
EPA estimated exposures for workers at 227 sites based on the information in Section 2.2.13. For this
estimate, EPA utilized workers and ONU estimates determined from an analysis of BLS data for the
NAICS code 334400, Semiconductor and Other Electronic Component Manufacturing. These data
indicate that there are, on average, 30 workers and 37 ONUs per site within NAICS code 334400. Based
on these data and 227 sites, EPA estimated that a total of 6,800 workers and 6,100 ONUs are potentially
exposed during this life cycle stage. EPA used unrounded figures for the number of workers and ONUs
per site to calculate these totals, resulting in the slight discrepancy.
Inhalation
During this exposure scenario HBCD is in paste form within the flux/solder paste and is not available
for particulate generation and exposure. Additionally, based on the process description, EPA does not
expect the use of flux/solder pastes to generate mists, other particulates, or vapors, due to the low
volatility of HBCD. The EURAR and NICNAS RAR indicate that HBCD begins to thermally degrade at
temperatures around 190 °C (NICNAS 2012b; ECHA 2008b). Typical soldering formulations start to
melt between 183-188 °C, with soldering temperatures set between 30°C to 50°C higher than the liquid
temperature of the alloy as a rule of thumb and expected to be set up to 300°C (Indium Corporation
2019a. b). EPA expects that the soldering process will destroy (via thermal degradation) the HBCD,
making it unavailable for exposure. Based on this description, EPA does not expect worker inhalation
exposure to HBCD during this exposure scenario.
Dermal
As described in Section 2.4.1.1. EPA used this model because the amount of dermal contact that workers
are potentially exposed to is likely smaller than that estimated in the other exposure scenarios. This
model uses a smaller quantity of solids on hands to estimate potential dose, based on worker contact
with container surfaces. EPA calculated dermal exposure assuming two-hand contact to solids
containing 1% HBCD. EPA calculated the potential dose for a worker to be 11.0 mg HBCD/day (high-
Page 241 of 723
-------
end) and 4.5 mg HBCD/day (central tendency). The EURAR and NICNAS did not estimate dermal
exposures during this exposure scenario (NICNAS 2012b; ECHA 2008b).
Strengths, Limitations, and Confidence in Assessment Results
EPA did not assess occupational inhalation exposures during this exposure scenario based on literature
and industry information indicating that the temperatures at which soldering occurs are likely to result in
the degradation of HBCD, as discussed above.
2.4X14 Recycling of Electronics Waste (E-Waste) Containing HIPS
HIPS is used in electronics such as household appliances, television sets, computers, phones, and other
electronic products (Morf et al. 2005). At the end of their life, electronics may be disposed of and
recycled at electronics recycling facilities. At electronics recycling facilities, workers may manually
disassemble electronics, sort electronic components, and operate equipment used to further process
electronic components, such as through crushing, grinding, and separation of materials (e.g., metal
scrap, plastics) (Rosenberg et al. 2011; Morf et al. 2005). These activities may generate dust that
contains HBCD that workers and ON Us may inhale (Rosenberg et al. ) or come into dermal contact
with once the dust settles on surfaces and workers or ONUs touch these surfaces (Zeng et al. 2016;
Rosenberg et al. 2011). EPA assessed potential worker exposure to the dust that is generated during
electronics recycling.
Number of Potentially Exposed Workers and Occupational Non-Users
EPA estimates that there are 745 electronics recycling sites currently in the US, including both certified
and uncertified sites, according to the two accredited certification programs e-Stewards and R2
Steward 2020; Sustainable Electronics Recycling 2020). BLS data for the NAICS code 562920,
Materials Recovery Facilities, indicate that there are, on average, two workers and two ONUs per site
within NAICS code 562920. However, Rosenberg et al. (2011) collected personal breathing zone
samples at four waste electrical and electronic equipment recycling sites, sampling between four and
seven workers per site. Therefore, the BLS data may underestimate the number of workers, so EPA
assessed seven workers per site. Because the BLS data indicate a similar number of workers and ONU at
these sites, EPA also assessed seven ONUs per site. This results in a total of 14 workers and ONUs per
site, which is similar to the average number of total employees per establishment for NAICS code
562920 (15 employees per site, of which workers and ONUs are a component).
Based on these data and 745 sites for electronics recycling, EPA estimated that a total of 5,215 workers
and 5,215 ONUs are potentially exposed in this exposure scenario.
Inhalation Exposure Assessment
EPA estimated HBCD potential inhalation exposure concentrations during electronics recycling using
inhalation monitoring data from Rosenberg et al. (2011). EPA selected the data of Rosenberg et al
(2011) because the overall confidence rating determined through EPA's systematic review of these data
is high and no other relevant HBCD monitoring were found for this exposure scenario.
Rosenberg et al. (2011) collected personal breathing zone samples for flame retardants, including
HBCD, at four waste electrical and electronic equipment (WEEE) recycling sites in Finland. Worker
activities at these sites included manual disassembly and sorting of WEEE, removal of hazardous and
valuable components, and operation of mechanical size reduction equipment, such as crushing and
grinding machinery (Rosenberg et al. 2011). These activities are consistent with the worker activities
EPA assumes to have exposure potential.
Page 242 of 723
-------
Rosenberg et al. (2011) took PBZ samples during two sampling events at each site, one before and one
after the implementation of improved engineering controls, including improved ventilation,
maintenance, and cleaning habits. A total of 45 PBZ samples were taken from 34 workers at the four
sites during one work shift, with 24 samples taken before engineering control improvements and 21
samples taken after engineering control improvements. The results of the sampling were summarized in
the supplemental file to Rosenberg et al ( ) by presenting the arithmetic mean, median, and range for
each site, before and after engineering controls were implemented. Individual sampling points were not
provided. Rosenberg et al (2011) does not provide individual sample times but indicates that the samples
were taken over a shift with sample times ranging from 191 to 408 minutes. EPA assumes that the
authors translated the individual sample results to a common time basis in order to calculate the
presented summary statistics, with the time basis most likely 8 hours based on the longest sampling time
of 408 minutes. Therefore, EPA assumes the data from Rosenberg et al (2011) are 8-hour TWA values.
EPA included the results from Rosenberg et al Q ) in Table 2-79 below. For this assessment, EPA
only used the data taken before the implementation of engineering controls to provide a conservative
assessment. As discussed in Section 2.4.1.1, EPA prefers the use of median over mean to estimate
central tendency. Therefore, to estimate central tendency exposure concentration, EPA took the average
of the median values presented for the four sites before the implementation of engineering controls,
resulting in a central tendency value of 13.9 ng/m3, which is 0.0000139 mg/m3. To estimate high-end
exposure concentration, EPA used the maximum of all the ranges presented for the four sites before the
implementation of engineering controls, resulting in a high-end value of 100 ng/m3, which is 0.0001
mg/m3. As discussed in Section 4.2.2, EPA determined that the MOE values for these inhalation
exposure concentrations were all above the benchmark values; therefore, EPA did not further refine the
central tendency and high-end inhalation exposure concentrations to account for the monitoring data
taken after the implementation of engineering controls.
EPA did not find data on the exposure frequency of this exposure scenario. EPA assessed the maximum
number of exposure days to be 250 days/year, based on a work schedule of five days per week and 50
weeks per year. EPA used this value to calculate high-end average daily dose. EPA used the midpoint of
the range of 1 to 250 days/year of exposure frequency, rounded up to 126 days/year, to calculate central
tendency average daily dose. Additionally, EPA estimated worker exposure over the full working day,
or eight hours/day, as the data used to estimate inhalation exposures are 8-hour TWA data.
Page 243 of 723
-------
Table 2-79. Inhalation Monitoring Da
a for HBCD at Electronics Recycling Sites
Data Source/Study a
Exposure Scenario
Form of HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(ng/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Source
Overall
Confidence
Rating
Inhalation Monitoring Data Used in this Assessment to Estimate Worker Exposures Resulting from Electronics Recycling
Rosenberg et al (2011)
- la
WEEE Recycling -
before engineering
controls
HBCD dust
Personal
Breathing
Zone
Site A -
Sorting,
disassembly,
process
controllers (for
mechanical
separations)
Range: 13 - 47
Mean: 29
Median: 27
ng/m3
6
8-hr TWA
(Rosenbers
et al. 2011)
High
Rosenberg et al (2011)
- lb
WEEE Recycling -
before engineering
controls
HBCD dust
Personal
Breathing
Zone
Site B -
Sorting,
disassembly,
process
controllers (for
mechanical
separations)
Range: 5.7 -
100
Mean: 38
Median: 15
ng/m3
7
8-hr TWA
(Rosenbers
et al. 2011)
High
Rosenberg et al (2011)
- lc
WEEE Recycling -
before engineering
controls
HBCD dust
Personal
Breathing
Zone
Site C -
Sorting,
disassembly,
process
controllers (for
mechanical
separations)
Range: 4.9 - 8.0
Mean: 6.5
Median: 6.4
ng/m3
5
8-hr TWA
(Rosenbers
et al. 2011)
High
Rosenberg et al (2011)
- Id
WEEE Recycling -
before engineering
controls
HBCD dust
Personal
Breathing
Zone
Site D -
Sorting,
disassembly,
process
controllers (for
mechanical
separations)
Range: 4.5 - 8.5
Mean: 6.6
Median: 7.2
ng/m3
6
8-hr TWA
(Rosenbers
et al. 2011)
High
Other Inhalation Monitoring Data Not Used in this Assessment
Rosenberg et al (2011)
-2a
WEEE Recycling -
after engineering
controls are
HBCD dust
Personal
Breathing
Zone
Site A -
Sorting,
disassembly,
process
Range: ND -
8.5
Mean: 2.9
6
8-hr TWA
(Rosenbers
et al. 2011)
High
Page 244 of 723
-------
Data Source/Study a
Exposure Scenario
Form of HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(ng/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Source
Overall
Confidence
Rating
implemented/
improved
controllers (for
mechanical
separations)
Median: 1.3
ng/m3
Rosenberg et al (2011)
-2b
WEEE Recycling -
after engineering
controls are
implemented/
improved
HBCD dust
Personal
Breathing
Zone
Site B -
Sorting,
disassembly,
process
controllers (for
mechanical
separations)
Range: 0.90 -
88
Mean: 23
Median: 3.5
ng/m3
5
8-hr TWA
(Rosenbers
et al. 2011)
High
Rosenberg et al (2011)
-2c
WEEE Recycling -
after engineering
controls are
implemented/
improved
HBCD dust
Personal
Breathing
Zone
Site C -
Sorting,
disassembly,
process
controllers (for
mechanical
separations)
Range: ND
Mean: ND
Median: ND
4
8-hr TWA
(Rosenbers
et al. 2011)
High
Rosenberg et al (2011)
-2d
WEEE Recycling -
after engineering
controls are
implemented/
improved
HBCD dust
Personal
Breathing
Zone
Site D -
Sorting,
disassembly,
process
controllers (for
mechanical
separations)
Range: ND
Mean: ND
Median: ND
6
8-hr TWA
(Rosenbers
et al. 2011)
High
ND = Non-detect for HBCD.
a - Where multiple datasets were available from one literature source, EPA distinguished data as la, lb, 2a, 2b, etc.
b - The statistical values were obtained from the referenced literature source and were not calculated by EPA.
Page 245 of 723
-------
Dermal Exposure Assessment
The potential dermal exposure from handling HIPS containing HBCD is low due to the small weight
fraction of HBCD in the HIPS and because HBCD is incorporated into the polymer matrix (EOTA
2009c). However, dust containing HBCD may be generated during electronics recycling. Workers and
ONUs may come into dermal contact with HBCD in these dusts when the dust settles on surfaces and
workers or ONUs touch these surfaces (Zeng et at. 2016; Rosenberg et al. 2011).
Therefore, EPA assessed worker dermal exposure to dusts containing HBCD at electronics recycling
sites. As described in Section 2.4.1.1, EPA used the EPA/OPPTDirect 2-HandDermal Contact with
Solids Model ( a) and Marquart et al. (2006) to estimate high-end and central tendency
worker dermal potential dose rate. These sources estimate potential dose rates of 3,100 mg/day (high-
end) and 900 mg/day (central tendency) for quantity of solids retained on a worker's skin. These
quantities do not pertain to dermal exposure to settled dust and are used here as conservative estimates.
To estimate the quantity of HBCD in solids, EPA used data on the concentration of HBCD in dust on the
floors at electronics recycling sites from Zeng et al. (2016). EPA used the data from Zeng et al. (2016)
because no other relevant data were found. The overall confidence rating determined through EPA's
systematic review of these data is medium.
Zeng et al. (2016) collected 48 samples of surface particulates from the floors of four major electronics
waste recycling sites in China. These samples were analyzed for multiple compounds, including HBCD.
Zeng et al. ( ) reported HBCD concentrations in these samples by providing the range and average
for the samples taken at each of the four sites. EPA used the maximum HBCD concentration from Zeng
et al. (2016). which is 57,000 ng HBCD/g dust equaling an HBCD fraction of 0.000057, to estimate
dermal exposures. EPA calculated the potential dose rate for a worker to be a high-end of 0.18 mg
HBCD/day (3,100 mg solids/day x 0.000057) and a central tendency of 0.051 mg HBCD/day (900 mg
solids/day x 0.000057). As discussed in Section 4.2.2.5, EPA determined that the MOE values for these
dermal exposure estimates were all above the benchmark values; therefore, EPA did not further refine
the dermal exposure estimates to account for the average HBCD concentrations presented in Zeng et al.
(2016) or a lower solids potential dose rate.
Strengths, Limitations, and Confidence in Assessment Results
EPA has medium confidence in the assessed inhalation exposure concentrations presented above. EPA
considered the quality of the data, the assessment approach, and uncertainties in assessment results to
determine the level of confidence.
The result of EPA's systematic review is inhalation exposure monitoring data with an overall confidence
rating of high, which is a strength of the assessment. The strength of the assessment approach is the use
of inhalation exposure monitoring data that (a) are the preferred type of monitoring data {i.e., 8-hr TWA
personal breathing zone data), (b) are directly applicable to this exposure scenario, and (c) comprise data
from multiple sites and workers.
The limitation of the assessment approach is the estimation of inhalation exposure concentrations based
on averaging the median values and using the maximum of the available monitoring data because
individual sampling points were not available. However, EPA did not refine this approach because all
calculated MOE values were above the benchmark. In addition, the assessment is limited because only
one relevant dataset from the literature was available.
There is uncertainty in the estimated HBCD potential inhalation exposure concentrations because these
concentrations pertain to workers in Europe and the extent to which these concentrations represent the
Page 246 of 723
-------
distribution of inhalation exposure air concentrations pertaining to workers in the U.S. is uncertain.
Refer to Section 2.4.1.14 for additional discussion of uncertainty. Based on these strengths, limitation,
and uncertainty, EPA has medium confidence in the assessed occupational inhalation exposure air
concentrations.
2,4.1,15 Assumptions and Key Sources of Uncertainties for Occupational Exposures
Uncertainty is "the lack of knowledge about specific variables, parameters, models, or other factors" and
can be described qualitatively or quantitatively. The following sections discuss uncertainties throughout
the assessed HBCD exposure scenario scenarios.
2.4.1.15.1 Number of Workers
There are a number of uncertainties surrounding the estimated number of workers potentially exposed to
HBCD, as outlined below.
First, BLS occupational employment statistics employment data for each industry/occupation
combination are only available at the 3-, 4-, or 5-digit NAICS level, rather than the full 6-digit NAICS
level. This lack of granularity could result in an overestimate of the number of exposed workers if some
6-digit NAICS are included in the less granular BLS estimates but are not, in reality, likely to use HBCD
for the assessed applications. EPA addressed this issue by refining the occupational employment
statistics estimates using total employment data from the U.S. Census SUSB. However, this approach
assumes that the distribution of occupation types (SOC codes) in each 6-digit NAICS is equal to the
distribution of occupation types at the parent 5-digit NAICS level. If the distribution of workers in
occupations with HBCD exposure differs from the overall distribution of workers in each NAICS, then
this approach may result in inaccuracy, resulting in either an overestimation or underestimation of the
number of potentially exposed workers.
Second, EPA's judgments about which industries (represented by NAICS codes) and occupations
(represented by SOC codes) are associated with the uses assessed in this assessment are based on EPA's
understanding of how HBCD is used in each industry. Designations of which industries and occupations
have potential exposures is nevertheless subjective, and some industries/occupations with few exposures
might erroneously be included, or some industries/occupations with exposures might erroneously be
excluded. This would result in inaccuracy and could either overestimate or underestimate the estimate of
exposed workers.
2.4.1.15.2 Estimation of Inhalation Exposure Concentration and Average
Daily Dose
For the most part, EPA estimated HBCD potential inhalation exposure concentrations to be equal to
surrogate HBCD inhalation exposure concentration monitoring data. There is uncertainty about the
extent to which these monitoring data are valid surrogate data. A reason for this is that there is
uncertainty about whether the process equipment and/or worker activities associated with the monitoring
data are comparable to the corresponding process equipment and/or worker activities pertaining to the
exposure scenarios that EPA assessed. Even if the process equipment and/or the worker activities were
comparable, there would still be uncertainty because for the most part EPA estimated HBCD potential
inhalation exposure concentrations to be equal to HBCD inhalation exposure concentration monitoring
data pertaining to workers at sites in Europe.
The extent to which HBCD inhalation exposure concentration monitoring data pertaining to workers at
sites in Europe represent the distribution of inhalation exposure air concentrations pertaining to workers
in the U.S. is uncertain because the determinants of HBCD occupational exposure in Europe and in the
Page 247 of 723
-------
U.S. may not be similar. These determinants include the engineering controls. The engineering controls
in Europe and in the U.S. may be different due to differing occupational exposure limits. For example,
the occupational exposure limit (OEL) for organic dust and mist in Sweden, which may be applicable to
HBCD, is 5 mg/m3 (EC 008b) but an OEL for HBCD is not established in the U.S. and the OSHA
PEL is 15 mg/m3.
EPA calculated average daily dose (ADD) for use in risk characterization assuming an exposure
frequency equal to the midpoint and high-end of the range of operating days per year, as discussed for
each exposure scenario. Use of the high-end exposure days assumes the workers are exposed every
working day, which may be an overestimate if workers do not conduct the worker activities that are
associated with the assessed exposure scenarios during each day of operation.
2.4.1.15.3 Modeling Dermal Exposures
To model dermal exposures, EPA used the EPA/OPPTDirect 2-HandDermal Contact with Solids
Model, EPA/OPPT Direct 2-Hand Dermal Contact with Container Surfaces (Solids) Model and the
typical exposure data reported in (Marquart. 2006) to estimate high-end and central tendency exposure
estimates. These estimates do not account for the potential exposure reduction due to glove use. In
addition, the potential dermal exposure estimates do not account for variations in the particle sizes of the
solid, amount being handled, or duration of worker activity performed. EPA modeled dermal exposures
using an upper-end estimate of 6.5% steady-state absorption (see Section 3.2.2). Absorption in
occupational settings may be lower than this value based on frequent hand washing or uneven
distribution across skin.
2.4.1.15.4 Occupational Non-User (ONU) Potential Inhalation Exposure
As discussed in Section 2.4.1.1, Occupational Exposures Approach and Methodology, EPA assumes
ONU potential HBCD inhalation exposure levels to be lower than those of workers. During the
construction and demolition of buildings, EPA believes that ONUs may work in close proximity to
workers and hence may be exposed to HBCD air concentrations similarly to workers. Furthermore, the
duration and frequency of the ONUs' work during the construction and demolition of buildings may
equal that of the workers at least for limited periods of time. That is, trade workers such electricians and
plumbers may work in close proximity to workers installing XPS/EPS insulation containing HBCD for
the duration of a particular construction project but that is not necessarily always the case. In conclusion,
there is uncertainty about whether the HBCD potential exposure level of ONUs in the case of
construction and demolition workers is lower than those of workers.
2.4.1.15.5 Firefighter Potential Occupational Exposure
Firefighters represent a subset of the worker population that could be exposed to HBCD from the
burning of building materials and other products. The exposure of HBCD could stem from multiple
conditions of use (e.g., building materials, automobile parts, and other plastics). For the HBCD Problem
Formulation, EPA did not include firefighters within the lifecycle diagram or conceptual model as an
assessment that EPA would include in this risk evaluation. However, EPA acknowledges that
firefighters may be exposed to HBCD and its thermal degradants via inhalation and dermal route.
EPA has identified limited information on firefighter exposure specific to HBCD. A review of literature
in general for firefighters did indicate firefighters may be exposed to flame retardants and combustion
by-products during firefighting, overhaul (searching for fire extending into building spaces), or through
contact with contaminated clothing and equipment and dust at firehouses (Minnesota Department of
Health. 2016). Firefighters generally wear SCBA, gloves, hoods, and coats as protective gear (Mayer et
at.. 2.019; Alexander and Baxter. 2016; Fent et at.. 2015). However, firefighters do not always wear
Page 248 of 723
-------
SCBA during exterior operations (deploying hoses, forcible entry) or during overhaul operations, and
even has been found to still be exposed to particulates and vapors to the neck and hands with PPE (Tent
et at.. 2015). Multiple sources indicate that flame retardants can accumulate on PPE over time (Mayer et
ai. 2019; Alexander and Baxter. 2016; Horn et at.. 2016; Fent et al.. 2015) and then transfer to the skin
of firefighters during activities such as turnout and cleaning. Alexander et al. (2016). Mayer et al.
(2019). and Horn et al. (2016) all sampled for and detected PBDEs, non-PBDE flame retardant and
organophosphate flame retardants on used firefighter PPE. Only Mayer et al. (2019) specifically
sampled for HBCD (on firefighter hoods), which was not detected in any hoods. This study was the only
information EPA identified that specifically looked at HBCD, EPA identified additional studies that
investigated other flame retardants including other brominated flame retardants (e.g., PBDEs, TBBPA).
Horn et al. (j ) measured area air concentrations of flame retardants including PBDEs and multiple
other flame retardants (not including HBCD) during a controlled active residential fire and overhaul,
detecting nine of the eighteen sampled flame retardants during the controlled fire (ranging from 1.2 to
2000 |ig/m3) and two flame retardants during overhaul (1.9 and 14 |ig/m3). Fent et al. (2018) measured
various chemical concentrations including hydrogen bromide in ten area samples that ranged from non-
detect to 19.8 mg/m3. NICNAS indicates that the combustion of brominated flame retardants can be a
source of hydrogen bromide (NICNAS. 2012b). EPA found one source that measured the concentration
of flame retardants in dust at fire stations (Shen et at.. 2018). Shen at al. (2018) did not sample for
HBCD, but did conclude that fire stations are contaminated with higher levels of flame retardants than
residences and other occupational settings.
Shaw et al. (20! 3) measured PBDEs and polybrominated dibenzo-p-dioxins and dibenzofurans
(PBDDFs) in the serum of 12 firefighters. This study notes that PBDDFs are produced during the
combustion of wastes containing brominated flame retardants (Shaw et al.. 2013). Shaw et al. ( )
found that levels of PBDEs in firefighters were higher than those detected in the general U.S.
population, concluding that the results are suggestive of significant occupational exposure to these
compounds during firefighting. Shaw et al. ( ) indicates that, while preliminary, the serum
concentrations of PBDDFs in firefighters suggest that occupational exposure to PBDDFs formed during
fires may be significant for firefighters.
In summary, EPA believes firefighters may be exposed to flame retardants, which may include HBCD.
However, EPA did not quantify these exposures as EPA lacks data specific to HBCD on these exposures
and exposures of other flame retardants are not easily translated to HBCD due to differences in chemical
properties, volumes, and uses. The potential exposures faced by firefighters is a source of uncertainty in
the occupational exposure assessment.
2.4.1.15.6 Summary of Occupational Exposures
For the risk characterization of occupational exposures, EPA used the 8-hour TWA exposure
concentrations (both central tendency and high-end values) that EPA selected for each exposure scenario
(refer to Sections 2.4.1.2 through 2.4.1.13 for rationale for these selections). Specifically, EPA used
these exposure concentration values to calculate acute exposure dose (AED) and acute daily dose
(ADD), which were then multiplied by the inhalation absorption factor of 100% (discussed in Section
3.2.2) to estimate the acute absorbed dose (AAD) and chronic absorbed dose (CAD), respectively.
Similarly, for dermal exposures, EPA used the potential dermal dose rates (refer to Sections 2.4.1.2
through 2.4.1.13 for rationale for EPA's determination of these values) to calculate AED and ADD,
them multiplied these values by a dermal absorption factor of 6.5% (discussed later in Section 3.2.2) to
estimate the AAD and CAD. Additional explanation of these equations and example calculations are
located in Appendix E.3 and Appendix E.4, respectively.
Page 249 of 723
-------
A summary of the 8-hour TWA or dermal dose rate, AAD, and CAD values used in this Risk Evaluation
is presented in Table 2-80 and Table 2-81 below. The ADD and CAD are used to characterize chronic,
non-cancer risks in Section 4.2.
Page 250 of 723
-------
Table 2-80. Acute and Chronic Inhalation Exposure Estimates, Worker Occupational Scenarios a
Eight-Hour TWA Exposures
Acute Absorbed Dose
Chronic Absorbed Dose
Occupational Scenario -
Inhalation Exposure
Chbcd, 8-hr TWA (mg/m3)
AADhbcd (mg/kg-day)
CADhbcd (mg/kg-day)
Characterization
High-End
Central
Tendency
High-End
Central Tendency
High-End
Central Tendency
Repackaging of import
containers
1.9E+00
8.9E-01
2.4E-01
1.1E-01
1.6E-01
4.27E-02
High-end: 90th percentile
Central Tendency: Median
Compounding of Polystyrene
Resin to Produce XPS
Masterbatch
2.5E+00
1.3E+00
3.1E-01
1.6E-01
5.1E-02
1.50E-02
High-end: Reasonable 'worst-
case' from EURAR
Central Tendency: Typical from
eurar'
Processing to Produce XPS
Foam Using XPS
Masterbatch
2.2E-01
8.0E-02
2.8E-02
1.0E-02
1.2E-03
2.47E-04
High-end: 90th percentile
Central Tendency: Mean
Processing of HBCD to
Produce XPS Foam
2.5E+00
1.3E+00
3.1E-01
1.6E-01
1.4E-02
3.85E-03
High-end: Reasonable 'worst-
case' from EURAR
Central Tendency: Typical from
eurar'
Processing to Produce EPS
Foam Using Imported EPS
Resin Beads
2.2E-01
8.0E-02
2.8E-02
1.0E-02
1.1E-02
2.14E-03
High-end: 90th percentile
Central Tendency: Mean
Processing to Produce SIPs
and Automobile Replacement
Parts from XPS/EPS Foam
2.2E-01
8.0E-02
2.8E-02
1.0E-02
1.9E-02
3.64E-03
High-end: 90th percentile
Central Tendency: Mean
Use: Installation of
Automobile Replacement
Parts b
--
--
--
--
--
--
Use: Installation of XPS/EPS
Foam Insulation in
Residential, Public and
Commercial Buildings, and
Other Structures
2.2E-01
8.0E-02
2.8E-02
1.0E-02
1.9E-02
3.45E-03
High-end: 90th percentile
Central Tendency: Mean
Demolition and Disposal of
XPS/EPS Foam Insulation in
Residential, Public and
Commercial Buildings, and
Other Structures
3.0E-01
1.1E-01
3.8E-02
1.3E-02
2.6E-02
4.53E-03
This is a range using the OSHA
PNOR PEL of 15 mg/m3 and
HBCD concentration of 0.7% in
EPS and 2% in XPS.
Page 251 of 723
-------
Occupational Scenario -
Inhalation Exposure
Eight-Hour TWA Exposures
Chbcd, 8-hr TWA (mg/m3)
Acute Absorbed Dose
AADhbcd (mg/kg-day)
Chronic Absorbed Dose
CADhbcd (mg/kg-day)
Characterization
High-End
Central
Tendency
High-End
Central Tendency
High-End
Central Tendency
Processing: Recycling of EPS
Foam
2.2E-01
8.0E-02
2.8E-02
1.0E-02
1.1E-02
1.95E-03
High-end: 90th percentile
Central Tendency: Mean
Formulation of Flux / Solder
Paste
2.5E+00
1.3E+00
3.1E-01
1.6E-01
2.1E-01
5.48E-02
High-end: Reasonable 'worst-
case' from EURAR
Central Tendency: Typical from
eur'ar'
Use of Flux / Solder Paste b
--
--
--
--
--
--
Recycling of Electronics
Waste (E-Waste) Containing
HIPS
1.0E-04
1.4E.05
1.3E-05
1.7E-06
8.6E-06
6.IE.07
High-end: High-end of range
Central Tendency: Average of
medians
a As discussed in Section 2.4.1.1 EPA expects potential inhalation exposure of an Occupational Non-User (ONU) in the case of some of the exposure scenarios but EPA
did not assess this exposure due to lack of data. EPA expects these exposures to be lower than the exposures of the corresponding workers. b EPA did not estimate
inhalation exposures for these exposure scenarios as EPA does not expect the generation of dust for these exposure scenarios.
able 2-81. Acute and Chronic Derma
Exposure Estimates, Worker Occupational Scenarios
Occupational Scenario - Dermal
Exposure
Potential Dose Rate
Dexp (mg/day)
Acute Absorbed Dose
AADhbcd (mg/kg-day)
Chronic Absorbed Dose
CADhbcd (mg/kg-day)a
Characterization
High-End
Central
Tendency
High-End
Central
Tendency
High-End
Central
Tendency
Repackaging of import containers
3.1E+03
9.0E+02
2.5E+00
7.3E-01
1.7E+00
2.8E-01
Chronic absorbed dose -
High-end: Maximum number
of exposure days
Central tendency: midpoint
of exposure days
Compounding of Polystyrene Resin to
Produce XPS Masterbatch
3.1E+03
9.0E+02
2.5E+00
7.3E-01
4.1E-01
7.0E-02
Chronic absorbed dose -
High-end: Maximum number
of exposure days
Central tendency: midpoint
of exposure days
Page 252 of 723
-------
Occupational Scenario - Dermal
Exposure
Potential Dose Rate
Dexp (mg/day)
Acute Absorbed Dose
AADhbcd (mg/kg-day)
Chronic Absorbed Dose
CADhbcd (mg/kg-day)a
Characterization
High-End
Central
Tendency
High-End
Central
Tendency
High-End
Central
Tendency
Processing to Produce XPS Foam Using
XPS Masterbatch
2.2E+03
6.3E+02
1.8E+00
5.1E-01
7.7E-02
1.3E-02
Chronic absorbed dose -
High-end: Maximum number
of exposure days
Central tendency: midpoint
of exposure days
Processing of HBCD to Produce XPS
Foam
3.1E+03
9.0E+02
2.5E+00
7.3E-01
1.1E-01
1.8E-02
Chronic absorbed dose -
High-end: Maximum number
of exposure days
Central tendency: midpoint
of exposure days
Processing to Produce EPS Foam Using
Imported EPS Resin Beads
-
-
--
--
--
--
Processing to Produce SIPs and
Automobile Replacement Parts from
XPS/EPS Foam
-
-
--
--
--
--
Use: Installation of Automobile
Replacement Parts
-
-
--
--
--
--
Use: Installation of XPS/EPS Foam
Insulation in Residential, Public and
Commercial Buildings, and Other
Structures
-
-
--
--
--
--
Demolition and Disposal of XPS/EPS
Foam Insulation in Residential, Public
and Commercial Buildings, and Other
Structures
-
-
--
--
--
--
Processing: Recycling of EPS Foam
-
-
--
--
--
--
Formulation of Flux / Solder Paste
3.1E+03
9.0E+02
2.5E+00
7.3E-01
1.7E+00
2.6E-01
Chronic absorbed dose -
High-end: Maximum number
of exposure days
Central tendency: midpoint
of exposure days
Use of Flux / Solder Paste
1.1E+01
4.5E+00
8.9E-03
3.7E-03
6.1E-03
1.0E-03
Chronic absorbed dose -
High-end: Maximum number
of exposure days
Central tendency: midpoint
of exposure days
Page 253 of 723
-------
Occupational Scenario - Dermal
Exposure
Potential Dose Rate
Dexp (mg/day)
Acute Absorbed Dose
AADhbcd (mg/kg-day)
Chronic Absorbed Dose
CADhbcd (mg/kg-day)a
Characterization
High-End
Central
Tendency
High-End
Central
Tendency
High-End
Central
Tendency
Recycling of Electronics Waste (E-Waste)
Containing HIPS
1.8E-01
5.IE.02
1.4E-04
4.2E-05
9.8E-05
1.5E-05
Chronic absorbed dose -
High-end: Maximum number
of exposure days
Central tendency: midpoint
of exposure days
Page 254 of 723
-------
2.4.2 General Population (Background) Exposures
2.4,2,J_ General Population Exposure Approach and Methodology
HBCD is used primarily as an additive flame retardant in a variety of materials. HBCD has been
detected in the indoor and outdoor environment and in human biomonitoring indicating that some
amount of exposure is occurring in some individuals, although exposures likely vary across the general
population. S qq Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on General Population, Environmental, and Consumer Exposure Assessment or a summary
of environmental and biomonitoring studies where HBCD has been detected.
The migration of additive flame retardants from indoor sources such as building materials, plastics, and
other articles (from in-service products/articles at the end of their life cycle (Section 1.2.8) as well as
historical releases (Section 1.2.9) resulting from HBCD's persistence in the environment) appears a
likely source of flame retardants found in indoor dust, suspended particles, and indoor air (Guo 2013;
Dodson et al. 2012; Weschler and Nazaroff 2010). However, the relative contribution of different
sources of HBCD in these matrices is not well characterized. For example, HBCD present in building
insulation, textiles, and recycled XPS and EPS materials are likely to have differing magnitudes of
emissions. The totality of background exposure includes steady-state environmental exposures ongoing
releases not associated with a particular COU, background/indirect exposures from minor use products
(e.g., textiles, electrical and electronic products, adhesives, and coatings) (Section 1.2.8), and releases
stemming from historical activities (Section 1.2.9) due to HBCD's persistence in the environment
Emission of HBCD is likely to occur through the following mechanisms: diffusion from sources and
gas-phase mass-transfer, abrasion of materials to form small particulates through routine use, and direct
transfer from articles to dust adhered to the article surface. Releases of flame retardants to the outdoor
environment may occur through direct releases to water and air as well as indirect releases from the
indoor environment. For a more detailed discussion about indoor SVOC exposure, fate and transport in
the indoor environment, please see th q Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD),
Supplemental Information on General Population, Environmental, and Consumer Exposure Assessment.
Exposure to general population from non-scenario specific uses was estimated for emissions to water
and air, as depicted in Figure 2-3.
Uses Summary of Summary of Media Exposure Media Estimation
Release Types or Pathways Scenarios Methods
Non-scenario
Specific
Emission to water:
• Diet
• Soil
• Air
Emission to air:
• Air
• Dust
• Dermal
AmbientAir, Indoor
Air, Indoor Dust,
Soil, Diet,
Breastmilk
General
Background
Exposure
All media
• Use monitoring
data collected at
sites away from
manufacturing
facilities
• Reverse dosimetry
Figure 2-3. Overview of General Population Exposure Assessment
Page 255 of 723
-------
Figure 2-4 depicts the two approaches used by EPA to estimate exposures, both of which consider
multiple pathways of exposure. First, EPA estimated exposure doses using an indirect estimation
method that entailed combining environmental monitoring data (i.e., HBCD concentration in dietary
sources, dust, soil, ambient air, indoor air, and dermal loading) with age specific exposure factors and
activity patterns. EPA also estimated exposure doses using an exposure reconstruction method that
entailed combining human biomonitoring data from various environmental matrices with assumptions
about lipid content and generalized one-compartment half-life in the body. There is general concordance
between the two approaches. No modeling data was used for the general population receptor group.
Approach 1:
Indirect Estimation
m
Diet
Dust
Soil
Air
ill,
Dermal loading
Environmental media
concentrations
+
Age-specific
exposure
factors
+
Activity
patterns
Approach 2:
Exposure Reconstruction
w
W
Adipose tissue
Blood/serum
Breastmilk
Hair
Placental/fetal
tissue
Feces
Biomonitoring
concentrations
+
Lipid content
assumptions
Generalized one-
compartment
model
Figure 2-4. Two Exposure Assessment Approaches used to Estimate General Population Exposure
to HBCD
For each exposure pathway, central tendency and high-end doses were estimated. EPA's Human
Exposure Guidelines defined central tendency exposures as "an estimate of individuals in the middle of
the distribution." It is anticipated that these estimates apply to most individuals in the U.S. high-end
exposure estimates are defined as "plausible estimate of individual exposure for those individuals at the
upper end of an exposure distribution, the intent of which is to convey an estimate of exposure in the
upper range of the distribution while avoiding estimates that are beyond the true distribution." It is
anticipated that these estimates apply to some individuals, particularly those who may live near facilities
with elevated concentrations.
2.4.2.2 Indirect Estimation Using Environmental Monitoring Data and Exposure
Factors
EPA considered the following exposure pathways for the general population using the indirect
estimation approach:
• Dietary
a. Grains
b. Vegetables
c. Fruit
d. Meat
e. Dairy
f. Fats
Page 256 of 723
-------
g. Seafood (all age groups except infants)
h. Breast milk (for infant age group only)
• Ingestion of dust and soil
• Inhalation of particles (indoor and outdoor)
• Dermal absorption of dust, soil, and/or materials
Equations: EPA describes the equations and inputs used to estimate the exposures in Sections 2.4.2.2.1
through 2.4.2.2.5. For each pathway within the seven different age groups (ranging from infants to
adults), EPA calculated a central and high-end average daily dose (ADD) and then summed the pathway
specific doses to estimate aggregate doses for each age group. In this method EPA generally used central
tendency monitoring data and exposure factor inputs to calculate the central tendency ADD and high-
end values to calculate the high-end ADD. The calculated doses are presented in Section 2.4.2.2.5 for
each pathway individually and for the aggregate of all pathways.
Exposure Factors: Body weights, intakes rates, and other exposure factors used in the equations were
derived from EPA and other agency sources, in particular many were obtained from the Exposure
Factors Handbook (U.S. EPA. ). More information on the exposure factors used in each individual
media exposure assessment is presented in Appendix G.
Monitoring Data: For the indirect exposure approach, EPA screened, evaluated, and extracted
monitoring data for food, air, dust, and soil data. All studies with available monitoring data and passing
evaluation scores were considered for determining environmental concentrations and overall trends. The
following criteria were applied to obtain a representative final dataset for each media of interest:
• Location Type: Data were classified as near facility (point source) or away from facility (non-
point source or background) as discussed in the Environmental Exposure section. Data classified
as near facility were excluded from the general population analysis.
• Country: Since only limited U.S. data was identified through systematic review, data from all
high-income countries as classified by the World Bank (June 2019) were included in the final
analysis (https://datahelpdesk.worldbank.org/knowledeebase/art.icles/906519-world-bank-
country-and-lending-groups). High-income countries were selected as surrogate countries based
on the assumption that these countries have manufacturing, processing, and use characteristics
that are most likely to resemble those in the United States. Refer to Appendix G for a list of
monitoring data availability by country and media type.
• Unit Fraction: Only data in accepted fractions were included (see Table 2-82). Concentrations
were converted to acceptable unit fractions if conversion factors were provided in the study,
including TOC to dry weight. For food, only wet weight unit fractions were used since no dry
weight to weight conversion factors were available.
• Source of Food: For food groups, data reported from market basket studies were included and
data from wild caught studies were excluded. Wild caught fish monitoring data were considered
in the highly-exposed assessment.
A description of the statistical approach to estimating the central and high-end concentrations can be
found in Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on
Page 257 of 723
-------
General Population, Environmental, and Consumer Exposure Assessment. In short, EPA estimated an
arithmetic mean and 90th percentile value for each dataset based on its distribution type (lognormal or
normal), and from these values calculated an overall central tendency (mean of means) and high-end
value (average of 90th percentile). The distribution type was determined from the type and combination
of statistical parameters available in the study (i.e., geometric mean, arithmetic mean, median, geometric
standard deviation, standard deviation, minimum, and/or maximum). Most combinations were assigned
a lognormal distribution type, unless mean estimates were outside the range of reported data. A normal
distribution type was assigned to datasets with only a mean and standard deviation or when the mean
and medians were the same. Datasets were excluded from the final analysis dataset when not enough
parameters were available to estimate a mean or 90th percentile (i.e., only a range of values or only a
minimum or maximum value was reported). Table 2-82 provides a summary of the number of studies
extracted and number of studies used in the final dataset, and the selected unit fraction.
Table 2-82. Summary of Monitoring Studies Identified and Used in the General Population
Exposure Assessment
Media
Number of Studies
Extracted
Number of Studies
in Final Dataset
Fraction
Fruits
4
la
Wet
Vegetables
5
2b
Wet
Grains
7
2b
Wet
Meats
20
3C
Wet
Dairy
14
3d
Wet
Fats
9
2e
Wet
Seafood
22
8f
Wet
Breast milk
33
17g
Lipid
Indoor Air
Residential
8
4h
Gas and/or particulate
Public and commercial
buildings (PCB)
7
5*
Gas and/or particulate
Vehicles
3
2>
Gas and/or particulate
Ambient Air
20
7k
Gas and/or particulate
Indoor dust
Residential
34
241
Dry
Public and commercial building
(PCB)
20
16m
Dry
Vehicles
6
5"
Dry
Soil
17
2°
Dry
Handwipe
2
P
n/a
a Fruits: (Barehi et al. 2016)
Page 258 of 723
-------
b Vegetables and Grains: (Barglii et al. 2016: (Fsa 2006)
0 Meat: (Schecter et al. 2012): Barglii et al. 2016: (Fsa 2006)
d Dairy: (Barglii et al. 2016: (Fernandes et al. 2016): (Fsa 2006)
e Fats: (Schecter et al. 2012): (Fsa 2006)
f Seafood: (Driffield et al. 2008): (Schecter et al. 2012): (Kakimoto et al. 2012): (Ortiz et al. 2011): (Nakagawa et al. 2010):
Barglii et al. 2016: (Fernandes et al. 2016): (Sonet al. 2015)
g Breastmilk: (de Wit et al. 2012): (Abdallah and Harrad 2011): (Eggesbo et al. 2011): (Fangstrom et al. 2008): (Glynn et
al. 2011): (Carignan et al. 2012): (Toms et al. 2012): (Roosens et al. 2010b):(Eliarrat et al. 2009): (Ryan and Rawn 2014):
(Darnerud et al. 2015): (Harrad and Abdallah 2015): (Ryan et al. 2006): (Antignac et al. 2016): (Lignell et al. 2003): (Tao
et al. 2017):(de Wit et al. 2012)
h Indoor air - residential:(de Wit et al. 2012): (Abdallah et al. 2008): (Saito et al. 2007): Newton et al. 2015)
1 Indoor air - PCB: (de Wit et al. 2012): (Abdallah et al. 2008): (Saito et al. 2007)
J Indoor air - vehicle: (de Wit et al. 2012): (Abdallah and Harrad 2010)
k Ambient air: (Hoh and Hites 2005): (Drage et al. 2016): (Abdallah et al. 2008): Newton etal. 2015: (Vorkamp et al. 2015):
(Shoeib et al. 2014): (KLIF2010)
1 Indoor air - residential: (Stapleton et al. 2008): (Abb et al. 2011): (Abdallah and Harrad 2009): (Roosens et al. 2009):
(Santillo et al. 2003): (Abdallah et al. 2008): (Abdallah et al. 2008): (D'Hollander et al. 2010):(Johnson et al. 2013):
(Sahlstrom et al. 2012): (Ali et al. 2012): (Shoeib et al. 2012): (de Wit et al. 2012): (Roosens et al. 2010a): (Abdallah et al.
2008): (Stapleton et al. 2014): (Fromme et al. 2014): (Sclireder and La Guardia 2014):(Dodson et al. 2012): Newton et al.
2015: (Sahlstrom et al. 2015): (Mizouclii et al. 2015): (Kuang et al. 2016): (Coelho et al. 2016)
m Indoor air - PCB: (Abdallah and Harrad 2009): (Santillo et al. 2001): (Abdallah et al. 2008): (Abdallah et al. 2008):
(D'Hollander et al. 2010): (de Wit et al. 2012): (Roosens et al. 2010a): (Takigami et al. 2009):( Newton et al. 2015):
(Leonards et al. 2001): (Al Bitar 2004): (Takigami et al. 2008): (Allgood et al. 2016): (Harrad et al. 2010): Newton et al.
2015: (Mizouclii et al. 2015)
n Indoor air - vehicle: (Abdallah and Harrad 2009): (Abdallah et al. 2008): (Harrad and Abdallah 2011): (Allen et al. 2013):
(de Wit et al. 2012)
° Soil: (Covaci et al. 2009: Newton et al. 2015)
p Handwipe: (Tav et al. 2018)
2.4.2.2.1 Diet — Ingestion
For general population exposure, EPA estimated dietary exposure from all food groups based on
monitoring data. The exposure dose associated with ingesting food is generally derived by multiplying
the concentration of chemical in food or breastmilk by the ingestion rate for that food and dividing by
body weight (U.S. EPA 1992). Within this overall framework, exposures could be estimated by
grouping all foods and liquids together and using a generic overall exposure factor, disaggregating
discrete food groups and using food group specific exposure factors, or estimating exposures for unique
food items. EPA used available monitoring data to estimate central tendency and high-end
concentrations of HBCD in specific food groups.
Equations
The equation used to calculate the chronic dose for each age group due to dietary exposure of fruits,
grains, vegetables, meat, dairy, fats, and seafood is presented in Equation 2-12 below.
Equation 2-12
FCxIRx ED
Where
ADD = Average daily dose used for chronic non-cancer risk
calculations due to ingestion of food group (mg/kg-day)
Page 259 of 723
-------
FC = HBCD concentration in food group (mg/g)
IR = Food group ingestion rate by age group (g/kg bw-day)
ED = Exposure duration
AT = Averaging time
The equation used to estimate exposure from ingestion of breastmilk is presented in Equation 2-13
below.
Equation 2-13
ADD = BMC X BMR X p
Where
ADD =
BMC =
BMR =
P =
Average daily dose used for chronic non-cancer risk calculations
due to ingestion of breastmilk (mg/kg-day)
Chemical concentration in breastmilk lipids (mg/g)
Breastmilk lipid ingestion rate (mL/kg-day)
Density of human breastmilk, 1.03 (g/mL)
Concentrations
Table 2-83 shows the central and high-end HBCD concentrations in the various food groups and
breastmilk. The central tendency concentrations were used in the central ADD estimate and the high-end
concentration was used in the high-end ADD estimate. Charts depicting concentrations in all extracted
studies are provided in the Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on General Population, Environmental, and Consumer Exposure Assessment.
For fruits, grains, vegetables, meat, dairy, fats, and seafood EPA used market basket monitoring studies
to identify concentrations of HBCD present in different food groups. Only one U.S. market basket study
was identified. In this study (Schecter et al. 2012) measured HBCD stereoisomers in a variety of
common lipid-rich U.S. foods purchased from supermarkets in Dallas, TX in 2010. Thirty-six individual
foods were sampled, generally consisting of fish (including bottom-feeders), poultry, pork, beef and
peanut butter. HBCD were measured in only 15 individual food samples (detection frequency of 42%).
Total HBCD in the individual food samples ranged from non-detect to 1.366 ng/g ww. The median and
mean of total HBCD for all the samples were 0.012 and 0.114 ng/g ww, respectively. The highest
concentration was detected in canned sardines, followed by smoked turkey. Concentrations in this U.S.
study were similar to, although slightly lower than, market basket surveys in other countries. For
example, total HBCD concentrations ranged from non-detect to 0.75 ng/g ww in a 2004 UK Total Diet
Study (FSA 2006). non-detect to 10.1 ng/g ww in a 2013 UK study (Fernandes et al. 2016). and non-
detect to 4.90 ng/g ww in a 2012-2014 Korean study (Barghi et al. 2.016). Fish exhibited the highest or
close to highest concentrations in all these studies, but the detection was not restricted to only fish or
meat. In (Barghi et al. 2016). HBCD was detected in all eight food categories (fish, shellfish, meat, eggs,
cereals, vegetables, fruits and dairy products), and was only not detected in yogurt and onions.
Numerous other studies also examined seafood. The highest total HBCD wet weight concentration
measured in seafood was 77.3 ng/g ww in a sample collected from a Japanese market in 2005
(Nakagawa et al. 2010).
Market basket seafood is different from wild-fish caught in a river. Market-basket monitoring studies
typically collect many samples and may pool similar types of foods together for chemical or statistical
analysis. The levels of HBCD present in market basket food groups are typically lower than levels
Page 260 of 723
-------
detected in wild animals and plants, especially if collected in industrial areas or near point sources. For
example, in aquatic species, total HBCD has been detected in blubber of harbor porpoises at up to
19,208 ng/g ww in the UK (Law et al. 2006) and in bivalves at up to 362,900 ng/g lw in the U.S. at a
WWTP outfall (La Guardia et al. 20 i 0). Fish ingestion from wild-caught fish is discussed in the highly
exposed population section.
Breast milk ingestion is an exposure pathway specific to infants. HBCD may be present in mothers in
the general population or in highly exposed mothers through subsistence fish exposure or via
occupational exposures. It is likely that that breastmilk concentrations are higher in women who
consume more fish. The highest concentrations were observed by (Eliarret et al. 2009). in which HBCD
was measured in milk samples collected from women in Spain, ranging from ND to 188 |ig/kg lw, with
an average of 47 |ig/kg hv and a median of 27 |ig/kg hv. Another study by (Eeeesb0 et al. 2011).
collected milk samples from 193 mothers as part of the Norwegian Human Milk Study. HBCD levels in
breast milk ranged from 0.1 to 31 |ig/kg lw, with an average of 1.1 |ig/kg lw. In the United States,
(Carignan et al. ^measured HBCD in the breast milk of 43 mothers. HBCD was detected in all
samples with concentrations ranging from 0.36 to 8.1 |ig/kg lw, with a geometric mean of 1.02 |ig/kg
lw.
Ingestion Rates
For fruits, grains, vegetables, meat, dairy, fats, and seafood EPA used mean and 95th percentile age-
specific ingestion rates to calculate the central and high-end doses, respectively, with the exception that
50th percentile and 90th percentile ingestion rates were used for fish/shellfish. The ingestion rates
(mg/kg-day) were obtained from the Exposure Factors Handbook (U.S. EPA. ); ( [);
(U.S. EPA. 20171); (U.S. EPA. 20181) for fruits, vegetables, grains, meats, dairy, fats, and breast milk.
For seafood ingestion rates, EPA used data from ( '14b) along with mean body weights for
each age group from Exposure Factors Handbook (L M _ 2011 h) to calculate a g/kg-day ingestion
rate. Although infants (birth to one year) may consume fish, fish ingestion was considered to be
negligible for this group because fish would only be consumed for a fraction of the first year of life
(starting at 4 to 6 months when solid food is first introduced), the percent of the infant population that
consumes fish is extremely small (only 2.6% of the population), and the mean ingestion rates for fish
(0.03 g/kg/day for the whole population or 1.3 g/kg/day for consumers only from Exposure Factors
Handbook Q ^ \ 201 lb)) is a small fraction of the total diet. Table 2-83 shows the HBCD
concentrations and the range of ingestion rate used in the dose calculation by food group. See Appendix
G for the specific values used for each age group.
Breastmilk lipid ingestion rates were obtained from the Exposure Factors Handbook ( )
and age weighted to calculate an ingestion rate for birth to <1 year old, then multiplied by the density of
human breastmilk (1.03 g/mL) to obtain an ingestion rate of g/kg-day. The calculated central tendency
ingestion rate was 4.2 g/kg-day and the high-end ingestion rate was 6.4 g/kg-day.
Table 2-83. Summary of Concentrations and Ingestion Rates Used in General Population Diet
Exposure Estimate
Food group
H BCD concentration (mg/g ww)
Range of ingestion rates (g/kg-day)
Central tendency
High-end
Central tendency
High-end
Fruits
2.6E-08
5.5E-08
1.4E+00-9.9E+00
4.3E+00-2.7E+01
Vegetables
1.6E-07
1.9E-07
2.5E+00-6.7E+00
6.0E+00-1.9E+01
Page 261 of 723
-------
Grains
8.2E-08
1.1E-07
2.0E+00-6.4E+00
4.3E+00-1.3E+01
Meats
1.1E-07
1.8E-07
1.7E+00-4.0E+00
3.8E+00-9.6E+00
Dairy
1.6E-07
2.4E-07
3.3E+00-4.9 E+01
9.9E+00-1.0E+02
Fats
1.7E-07
2.3E-07
1.1E+00-4.6E+00
2.0E+00-8.9E+00
Seafood
2.0E-06
4.1E-06
1.9E-02-6.3E-02
1.5E-02-4.1E-02
Breastmilk
4.4E-06
8.7E-06
4.2E+00
6.4E+00
Exposure Duration and Averaging Time
The years within an age group (e.g., 1 year for infants) was used for the exposure duration and averaging
time.
HBCD in Drinking Water
EPA considered ingestion of drinking water but did not quantify those concentrations in this Risk
Evaluation. The concentration of HBCD in surface water is generally low and monitored levels of
HBCD in drinking water are unavailable. Other assessments have included drinking water as a pathway
and noted that expected exposures are quite low. The following exposure pathways are possible:
1. Ingestion of finished water at the tap, expected HBCD levels are low.
2. Ingestion of surface water, including suspended sediment, during recreation in lakes and rivers.
HBCD levels are likely slightly more elevated than drinking water but intake rates and frequency
of exposure are lower.
2.4.2.2.2 Dust and Soil — Incidental Ingestion
The exposure dose associated with incidentally ingested dust and soil is generally derived by
multiplying the chemical concentration in dust or soil by the empirically derived ingestion rate of dust or
soil and dividing by body weight (U.S. EPA. 1992). The ingestion rate can be derived through tracer
methods which measure tracer chemicals present both in soil and dust and in the urine and feces of
humans and through biokinetic methods that use biomonitoring data and physiologically based
pharmacokinetic (PBPK) models to back-calculate ingestion rates. An activity-pattern based method
models hand-to-mouth and object-to-mouth contact to derive transfer rates of soil and dust to the mouth
to estimate ingestion rate (Mova and Phillips 2.014). Estimated ingestion rates based on the activity-
pattern method are informed by empirically and estimated variables (Ozkavnak et al. 2011) including:
• Hand and object to mouth frequency indoors and outdoors
• Dust loading
• Object: floor dust loading ratio
• Soil skin adherence rate
• Skin/soil surface contact rate
• Maximum dermal loading of soil loading on hands
• Surface to hand dust transfer efficiency
• Hand to mouth and object to mouth transfer efficiency
• Area of object mouthed and fraction of hand mouthed/event
• Bath and hand wash removal efficiency and frequency
Chemical concentrations in dust or soil are required for the tracer and biokinetic methods. Loadings of a
chemical in dust or soil are required for the activity-pattern method. The chemical concentration in dust
or soil is defined as the mass of chemical present per mass of dust or soil. The chemical loading in dust
Page 262 of 723
-------
is defined as the mass of chemical per surface area.
These variables are all related, but often only one of the three is reported in monitoring studies. If the
surface area units are the same for loadings, the chemical dust loading divided by the total dust loading
is equal to the chemical concentration. However, dust loadings of overall levels can also vary
substantially by building or within a building. If paired chemical dust loading and chemical
concentration data are available, an empirical relationship can be used to derive a relationship and
conversion equation.
When an activity pattern method is used an overall dust or soil factor (units surface area/time) that
incorporates variability from the bulleted list above can be used to estimate intake.
A wide range of studies have reported HBCD concentrations in dust in a variety of indoor environments.
No studies were identified that specified HBCD loadings in dust. Therefore, empirically-derived
ingestion rates based on the tracer and biokinetic approaches were used for this assessment.
Equations
EPA used Equation 2-14 to estimate HBCD doses from dust ingestion and Equation 2-15 to estimate
HBCD doses from soil ingestion.
Equation 2-14
Where
ADD =
DCxIRxFDxCFlxED
BWxAT
ADD = Average daily dose used for chronic non-cancer risk calculations due to dust
ingestion (mg/kg-day)
DC = Dust concentration (|ig/mg) (see explanation below)
IR = Dust ingestion rate (g/day)
CFi = Conversion factor for mg/|ig
FD = Fraction of time spent awake spent in indoor microenvironments
ED = Exposure duration
BW = Body weight (kg)
AT = Averaging time
Equation 2-15
DCxIRxFDxCFlxED
ADD =
BWxAT
Where
ADD = Average daily dose used for chronic non-cancer risk calculations due to soil
ingestion (mg/kg-day)
DC = Soil concentration (|ig/mg)
IR = Soil ingestion rate (g/day)
CFi = Conversion factor for mg/|ig
ED = Exposure duration
BW = Body weight (kg)
AT = Averaging time
Page 263 of 723
-------
This HBCD assessment uses Equation 2-14 and Equation 2-15, while future assessments may use
Equation 2-16 depending on data availability.
Equation 2-16
. _ _ DL xDF xTSxED
ADD =
BWxAT
Where
ADD =
Average daily dose used for chronic non-cancer risk calculations due to soil or
dust ingestion (mg/kg-day)
DL =
Dust or soil loading (|ig/cm2)
DF =
Dust or soil factor (cm2/ |ig * mg/hr)
TS =
Time spent in different microenvironments, total should equal time awake (hr/day)
ED =
Exposure duration (years)
BW =
Body weight (kg)
AT =
Averaging time (years)
Concentrations
Table 2-84 presents the dust and soil concentrations that were used in Equation 2-14 and Equation 2-15
to estimate exposures from dust and soil ingestion. For dust, the concentrations were classified based on
the sampling microenvironment: residential, public and commercial building, and automobile. For soil,
the background (away from facility) concentrations estimated in the environmental exposure assessment
were also used in the general population assessment. The central tendency concentrations were used in
the central ADD estimate and the high-end concentration was used in the high-end ADD estimate.
Charts depicting concentrations in all extracted studies are provided in the Risk Evaluation for Cyclic
Aliphatic Bromide Cluster (HBCD), Supplemental Information on General Population, Environmental,
and Consumer Exposure Assessment.
Indoor dust studies were prevalent for residential and public/commercial building settings, and lesser so
for vehicles. The final dataset includes five studies (all residential) conducted in the U.S. between 2006
and 2012. In the most recent U.S. study (Stapleton et al. 2014). total HBCD was detected in detected in
all samples collected from thirty North Carolina homes in 2012, with a geometric mean of 3.4 E-04
|ig/mg (range of 7.8 E-05 to 2.6 E-03 |ig/mg). These values are within the same order of magnitude as
the central and high-end estimated indoor residential dust values used in this assessment, as shown in
Table 2-84.
Studies measuring the concentration of HBCD in soil are limited, with most studies measuring samples
located near industrial facilities. As discussed in the Environmental Exposure section, no U.S. soil
studies were identified and therefore background soil concentrations were derived from only two small
studies conducted in Belgium and Sweden (Covaci et al. 2009) and (Newton et al. 2015).
Exposure Factors
Fraction of time spent awake in indoor microenvironments: For dust, the concentration in each
microenvironment was weighted based on the fraction of time spent in each microenvironment. The
time spent by children and adults in each of these microenvironments was estimated for three generic
activity-pattern profiles (stay at home, part-time school/home, and full-time school/home) informed by
EPA's Consolidated Human Activity Patterns Database (U.S. EPA. 2009b). The hours spent in each
microenvironment were used to derive a fraction of the day that an individual was exposed to the
Page 264 of 723
-------
selected HBCD concentrations in each microenvironment. The median fraction was used for the central
ADD estimate and the maximum fraction was used for the high-end ADD estimate. See Appendix G for
the fraction of time spent awake spent in indoor microenvironments for the three generic activity
profiles.
Ingestion Rates: The central tendency and high-end dust ingestion and soil ingestion rates from the
Chapter 5 Update of Exposure Factors Handbook (U.S. EPA 2011b) were used in the central and high-
end ADD calculations, respectively.
Table 2-84. Summary of Dust and Soil Inputs Used in Estimating Dust and Soil Ingestion Dose for
HBCD
Parameter
Central
Tendency
High-end
Dust Concentration Residence (|ig/mg)
1.5E-03
2.9E-03
Dust Concentration PCBs (|ig/mg)
1.5E-03
2.9E-03
Dust Concentration Vehicle (|ig/mg)
1.7E-02
3.2E-02
Soil Concentration Background (|ig/mg)
1.4E-06
3.0E-06
Range of Dust Ingestion Rates (varies by age group) (mg/day)
2.0E+01-5.0E+01
6.0E+01-1.0E+02
Range of Soil Ingestion Rates (varies by age group) (mg/day)
1.0E+01-4.0E+01
5.0E+01-9.0E+01
Exposure Duration and Averaging Time
The years within an age group (e.g., 1 year for infants) was used for the exposure duration and averaging
time.
2.4.2.2.3 Air — Inhalation
Equations
Equation 2-17 was used to estimate dose from ingestion of suspended particles in air is below. For
indoor air, the concentration of HBCD particulate can be derived directly from air monitoring data or
estimated from measured indoor dust monitoring or total indoor air (vapor and particulate)
concentrations. This assessment uses air monitoring data for both outdoor and indoor environments.
Equation 2-17
Where
ADD
AC
IF
IR
FD
ED
BW
AT
ADD =
ACxIRx IFxFD x ED
BWxAT
Average daily dose used for chronic non-cancer risk calculations due to
suspended particle ingestion (mg/kg- day)
Concentration of particulates in air (mg/m3) See explanation below
Fraction of inhaled particles that are ingested (1; unitless)
Inhalation rate (m3/day)
Fraction of day spent in microenvironment (unitless)
Exposure duration
Body weight (kg)
Averaging time
Page 265 of 723
-------
Concentrations
Table 2-85 presents the indoor and outdoor air concentrations that were used to estimate exposures from
inhalation using Equation 2-17. As with dust, the air concentrations were also classified based on the
sampling microenvironment: outdoor, residential, public and commercial building, and automobile. The
central tendency concentrations were used in the central ADD estimate and the high-end concentration
was used in the high-end ADD estimate. Charts depicting concentrations in all extracted studies are
provided in the RiskEvaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on General Population, Environmental, and Consumer Exposure Assessment.
Studies of HBCD in ambient air are limited. In the only U.S. study, Hoh and Hites 2005 measured
HBCD in five sites across five states and detected HBCD in 120 of 156 samples. Across all sites central
tendency concentrations ranged from approximately 1 to 5 E-06 |ig/m3, which is approximately an order
a magnitude lower than the estimated concentrations in Table 2-85. Indoor air data was extracted from
ten studies, but no U.S. data was identified.
The distribution of HBCD between gas-phase and particle phase in indoor air and the resulting particle
size distribution is an important consideration. Smaller particles are expected to be respirable while
larger particles are expected to be inhalable. The particle size distribution was not available for many
monitoring studies, although most studies did report whether the sample was particulate or vapor. Only
particulate values were considered for this pathway.
Exposure Factors
Fraction of time in a day spent in indoor microenvironments:
Similar to dust, the fraction of time spent by children and adults in each of the microenvironments was
estimated for three generic activity-pattern profiles informed by EPA's CHAD (U.S. EPA 2009b) stay at
home, part-time school or work, and full-time work or school. For air, the fraction is based on a 24-hr
day. The median fraction was used for the central ADD estimate and the maximum fraction was used for
the high-end ADD estimate. See Appendix G for the fraction of time spent in the microenvironments
over 24 hours for the three generic activity profiles.
Inhalation rates: The central tendency and high-end dust inhalation rates from the Chapter 5 Update of
Exposure Factors Handbook (U.S. EPA 2011b) were used in the central and high-end dose calculations,
respectively.
Table 2-85 Inputs for Estimation of HBCD Inhalation Dose
Parameter
Central Tendency
High-end
Air Concentration Outdoors (ng/m3)
1.96E-05
2.96E-05
Air Concentration Residence (ng/m3)
1.00E-04
2.26E-03
Air Concentration P&CBs (ng/m3)
9.10E-04
1.91E-03
Air Concentration Vehicle (|ig/m3)
2.44E-06
3.27E-06
Range of Inhalation Rates (varies by age group) (m3/day)
5.4E+00-1.6E+01
9.2E+00-2.5E+01
Exposure Duration and Averaging Time
The years within an age group (e.g., 1 year for infants) were used for the exposure duration and
averaging time.
Page 266 of 723
-------
2.4.2.2.4 Dermal
EPA used a fractional absorbed approach to estimate dermal exposures from contact with dust, soil, and
materials containing HBCD. Two different estimation methods were used. The first method was based
on empirical data where levels of HBCD present in dust on people's hands was sampled using hand-
wipes (direct estimation method). The second method was based on measured dust and soil
concentrations and age-specific dust-skin and soil-skin adherence factors (indirect estimation method).
After estimating the dermal loading, an absorption fraction of 6.5% was applied as discussed in Section
3.2.2.1.1. These methods are described in more detail below.
For the direct estimation approach, Equation 2-18 was used:
Equation 2-18
CJ
ADD _ Chw x FRabs x bW xCF X EF xED
AT
ADD =
Average daily dose used for chronic non-cancer risk calculations due to
skin contact with dust, soil, or materials (mg/kg-day)
Chw
Concentration in hand wipe (pg/cm2)
/ I\ =
Dermal absorption fraction (6.5%)
SA/BW =
Surface area of both hands/body weight ratio (cm2/kg)
CF =
Conversion factor (10E-9 mg/pg)
EF =
Exposure frequency (1 event/day)
ED =
Exposure duration (years in age group)
AT =
Averaging time, non-cancer (years in age group)
EPA used HBCD-specific hand wipe concentrations from Tav et al. 2018. In this study hand wipe
samples were collected from a Norwegian cohort of 61 adults between November 2013 and April 2014.
Participants were instructed not to wash their hands at least 60 minutes prior to sampling. Samples were
collected from both hands separately using sterile gauze pads immersed in isopropanol, and combined
into one sample prior to analysis by ultra-performance liquid chromatography-mass spectrometry
(UPLC-MS). The LOQ ranged from 20 to 45 ng/participant per isomer. The mass of total HBCD
(including a, (3, and y isomers) per participant was 49 to 8,900 ng with a median of 180 ng and a mean
of 680 ng (n = 60; detection frequency of 57 to 80% per isomer). After normalization for the surface
area of the participants hand, as estimated by an equation adopted from which
incorporates the weight and height of the participant, total HBCD ranged from 27 to 11,000 pg/cm2,
with a median of 150 pg/cm2 and a mean of 760 pg/cm2. The mean value of 0.76 ng/cm2 was selected for
the central ADD estimate and the maximum value of 11 ng/cm2 was selected for the high-end ADD
estimate. The study also collected settled dust samples from elevated levels in the living room of
participants. The authors noted that positive and significant correlations were found between settled dust
and hand wipes for gamma HBCD, which indicates that the levels of HBCD on the skin surface might
be a consequence of contact with elevated surface dust in the home.
One other hand wipe study was identified (Stapleton et al. 2014). In this study hand wipe samples were
collected in 2012 from 43 children, age 2 to 6, living in North Carolina. The gauze samples were
analyzed for individual HBCD isomers (alpha, beta, and gamma) using gas chromatography-mass
spectrometry (GC/MS). Total HBCD (sum of isomers) ranged from ND (<0.05) to 10.8 ng/participant,
Page 267 of 723
-------
with a geometric mean of 0.97 ng/participant (n = 43; detection frequency of 40 to 53% per isomer).
Stapleton et al. 2014 did not normalize the results based on surface area of the participants hands. Using
a value of 3.7 cm2, the mean surface area of hands for children ages 2 to <6 years ( ), the
maximum and geometric mean values would be 2.9 and 0.26 ng/cm2, respectively.
The levels detected in handwipes based on sampling of adults in Tav et al. 2018 were about three to four
times higher in magnitude than the levels detected in handwipes based on sampling of children in
Stapleton et al. . Some of the difference may be attributable to differences in activity patterns and
hand-to-mouth behaviors between adults and young children. As both studies used a relatively small and
potentially homogeneous group of participants, the concentrations from Tav et al. 2018 were selected for
all populations as a conservative estimate.
The surface area to body weight ratios used in Equation 2-18 are based on the 50th percentile values
reported in Refer to Appendix G for a summary of exposure factors used in the human
exposure assessment.
For the indirect estimation approach, Equation 2-19 was used:
Equation 2-19
SA
Conc.x FRnhc xAFx dw x CF1 x CF2 x EF x ED
ADD = — ^
AT
ADD =
Average daily dose used for chronic non-cancer risk calculations due to
skin contact with dust, soil, or materials (mg/kg-day)
Cone. =
Concentration in soil and dust (|ig/mg)
FRabs =
Dermal absorption fraction (6.5%)
AF =
Adherence factor (mg/cm2)
SA/BW =
Surface area of hands, face, and arms/bodyweight ratio (cm2/kg)
CF1 =
Conversion factor (0.001 g/mg)
CF2 =
Conversion factor (0.001 mg/|ig)
EF =
Exposure frequency (1 event/day)
ED =
Exposure duration (years in age group)
AT =
Averaging time, non-cancer (years in age group)
EPA used the same dust concentrations (weighted average based on microenvironment) and soil
concentrations calculated for the general population (refer to Section 2.4.2.2.2) to estimate dermal
exposure. The amount the dust and soil expected to adhere to the skin was accounted for through
adherence factors weighted based on the surface area of the body parts exposed to the dust and soil.
Exposure Factor Handbook ( ) provides recommended adherence factors by body part
for adults and children based on a limited number of observations for a limited set of activities that
primarily focus on soil. The recommended values for "activities with soil" were selected for the soil
pathways, and "residential, indoors" was selected for the dust pathway where available. The value was
not reported for adults, and the "activities with soil" was used in absence of dust-specific adherence
values. These values were weighted using equation 7-1 in U.S. EPA. 2 assuming that exposed body
parts are hands, lower legs (45% of total leg), and lower arms (50% of lower arms). For context, this
represents a short sleeve shirt and shorts scenario, or approximately 25% of the body. The surface area
to body weight ratios used in the equation are based on the 50th percentile values reported in Exposure
Page 268 of 723
-------
Factors Handbook (U.S. EPA 2011b). Refer to Appendix G for a summary of exposure factors used in
the human exposure assessment.
Table 2-86 provides the results for both methods. It should be noted that the direct handwipe method
results are an order of magnitude larger than the indirect adherence method results. The average of the
direct and indirect methods was used in the exposure assessment. For the indirect (adherence) method,
the dose from dust was approximately 10 to 200 times the dose from soil.
Table 2-86. Age Specific ADD for Dermal Exposure from Dust, Soil, and Materials
Dermal Central (mg/kg/day)
Dermal I
igh-End (mg/kg/day)
Direct
Indirect
Average
Direct
Indirect
Average
Infant (<1 year)
1.9E-05
8.4E-09
9.7E-06
1.3E-06
2.8E-09
6.7E-07
Young Toddler (l-<2 years)
1.9E-05
7.9E-09
9.4E-06
1.3E-06
2.7E-09
6.5E-07
Toddler (2-<3 years)
1.5E-05
7.2E-09
7.3E-06
1.0E-06
2.4E-09
5.0E-07
Small Child (3-<6 years)
1.4E-05
6.8E-09
7.1E-06
9.8E-07
2.3E-09
4.9E-07
Child (6-
-------
Table 2-88. General Population Central Tendency Source Contribution by Pathway and Age
Group (% Contribution to Total HBCD Exposure)
Age Group
DIET
DUST
SOIL
AIR
DERMAL
Infant (<1 year)
59.5%
37.3%
0.011%
1.5%
1.7%
Young Toddler (l-<2 years)
37.4%
58.3%
0.017%
2.0%
2.2%
Toddler (2-<3 years)
47.7%
46.5%
0.017%
3.0%
2.8%
Small Child (3-<6 years)
46.4%
46.6%
0.017%
3.4%
3.7%
Child (6-
-------
Central Tendency
High End
4.6%
0.013%
DUST
¦ SOIL
¦ AIR
DIET
¦ DERMAL
3.5%
2 3% 0.018%
Figure 2-5. Source Contribution by Pathway for Aggregate General Population Exposures
Estimated doses using the indirect estimation method from completed assessments were also examined
to compare against EPA's calculated doses. Dose estimates in completed assessments represent a wide
variety of countries (at least 18), populations, age groups, pathways, and exposure scenarios. The
extracted doses range from 1.8 E-13 to 2.75 mg/kg/day, with the highest dose attributed to intake from
the industrial use of HBCDD as textile back-coating agent (EINECS 2008). Other completed assessment
doses higher than the EPA calculated general population doses were reported for a local industry
specific scenario - emissions from EPS formulation (ECHA 2017b).
Occupational microenvironments represent settings where workers may be exposed to residual,
background levels of HBCD. These may include exposures due to formulated products and articles (e.g.,
textiles, electrical and electronic products, adhesives, and coatings). For estimating exposure from
occupational microenvironments, aggregate concentrations were estimated from various non-residential
microenvironments relevant to the general population i.e., mixed use, vehicle, commercial, public
buildings, and schools. These include available dust and air concentration data found in various school
rooms (classrooms, computer rooms, gymnasiums), government buildings, car cabins, car trunks,
airplanes, and waste electronics facilities) and represent a small subset of total aggregate general
population exposure (described in Section 2.4.2.1 with results presented in Section 2.4.2.2.5).
Concentrations were estimated as previously described for dust (Section 2.4.2.2.2) and indoor air
(Section 2.4.2.2.3), with high-end and central tendency doses for working age adolescents and adults
(age 16 - 70) derived only from data for public commercial buildings (PCBs) and automobiles.
Table 2-91 present exposure estimates for occupational microenvironments. Because occupational
microenvironments are represented by a subset of aggregate general population exposure, the table also
shows a relative comparison of those exposure estimates as a percentage of the total aggregate exposure
for each pathway and overall. Occupational microenvironments comprise the majority of aggregate dust
exposure but only a small minority of inhaled air exposures and less than a third of total aggregate
general population exposures. Exposures from formulated products and articles (e.g., textiles, electrical
and electronic products, adhesives, and coatings) comprise a non-quantifiable subset of the total
occupational microenvironment exposure since these aggregate exposures likely include other sources as
well, including releases stemming from historical activities (Section 1.2.9) due to HBCD's persistence.
2.4.2.2.6 Occupational Microenvironments
Page 271 of 723
-------
Table 2-91. Occupational Microenvironments Doses as a Percentage of Aggregate General
Population Exposure
Exposure
Level
PCB + Auto
Dust (mg/kg-
day)
PCB + Auto
Dust (% of
total dust)
PCB + Auto Air
(mg/kg-day)
PCB + Auto Air
(% of total air)
Occup.
Micro. Total
(mg/kg-day)
Occup. Micro.
(% of total aggregate)
High-End
4.7E-06
8.5E+01
5.2E-07
1.05%
5.3E-06
30.9%
Central
Tendency
8.3E-07
8.5E+01
1.6E-07
4.49%
9.9E-07
32.1%
2.4.2.3 Exposure Reconstruction Using Human Biomonitoring Data and Reverse
Dosimetry
EPA describes the approach used to estimate doses based on biomonitoring below. HBCD has been
quantified in human samples in blood serum in adults, cord serum, breast milk, and adipose tissue in
generally small, primarily European cohorts in a range of studies. An approach to estimate external
doses of HBCD based on biomonitoring data is reported in Aylward and Hays 2011. The approach uses
a simple one-compartment model with a 64 day half4ife of HBCD in the body (Gever et al. 2004)
coupled with an assumed percent lipid in the body, allowing ng/g lipid weight (lw) biomonitoring values
reported in various matrices to be converted to external exposure doses (mg/kg/day).
HBCD human biomonitoring data were previously extracted from peer-reviewed studies and curated to
produce one set of summary statistics per study. A total of 52 peer-reviewed studies, resulting in 64 data
sets with sampling years from 1973 to 2015, reported HBCD data in human adipose tissue, blood, breast
milk, feces, fetal tissue, hair, and placental tissue across the general population, occupational workers
and highly exposed populations. Table 2-92 provides the number of data sets for each population and
media type. Prior to any calculations of dose, the biomonitoring data were standardized to have the same
concentration units of ng/g lipid as follows:
1) For data reported as ng/g whole blood or ng/g serum, it was assumed that the lipid content in whole
blood and serum was 25%.
2) For data reported as ng/g hair, it was assumed that the lipid content in hair was 6%
3) For data reported as ng/L serum, the density of serum (1.024 g/mL as reported in Sniegoski and
Moody, 1979) was used to convert volume to mass.
Table 2-92. Human HBCD Biomonitoring Data Sets by Population, Type and Number
Population
Media Type
No. of Data Sets
General
Adipose Tissue
4
General
Blood / Serum
14
General
Breast Milk
34
General
Feces
1
General
Hair
1
General
Placental / Fetal Tissue
2
Highly Exposed
Blood
2
Highly Exposed
Breast Milk
4
Highly Exposed
Hair
1
Occupational
Breast Milk
1
Page 272 of 723
-------
For each set of human biomonitoring data, the estimated external dose of HBCD was estimated using
the approach in (Aylward and Hays 2011). This approach used a basic one-compartment, first-order
pharmacokinetic (PK) model to estimate chronic daily dose. The mass balance equation for change in
chemical mass in one compartment is:
AMc = (D ¦ BW ¦ At) - (k -Mc ¦ At)
where Mc is the mass of HBCD in the body [mg]
D is the chronic daily dose [mg/kg body weight/day]
BW is the body weight [kg body weight]
At is the change in time [days]
k is the first-order elimination rate constant [1/day]
The following equations can be substituted into the mass balance equation:
Mc
C = —
M lipid
Miipld = BW ¦ Fi
^1/2
where C is the mass of HBCD per mass of lipid in the body [mg/kg lipid]
Mupid is the mass of lipid in the body [kg lipid]
Fi is the fraction of body weight that is lipid [kg lipid/kg body weight]
ti/2 is the half-life of HBCD [days]
At steady state, this gives:
D = k C Ft
In this model, the assumptions are:
• Steady state conditions
• Elimination of HBCD from the body is due to a first-order degradation progress
• HBCD distributes equally in lipid throughout the body
• No difference in toxicokinetic parameters between different HBCD isomers
The parameter values used in (Aylward and Hays 2011). and subsequently used in the EPA calculations
were:
Page 273 of 723
-------
• Fraction of body weight that is lipid was assumed to be 25%
• Half-life of HBCD was previously estimated by (Gever et al. 2004) to be 64 days, with a range of
23 to 219 days. These values were calculated assuming:
o HBCD concentrations of 250-2400 ng/kg fat (mean of 700) in human breast milk from
non-occupationally exposed Swedish population based on values reported in (Barregard
2003).
o A daily intake rate of 142 ng/day by adult humans in Sweden based on a market basket
study, as reported in two studies of (Darnerud 2003) and (Lind et al. 2003).
o The fraction of dose absorbed from food was 100%.
Although the HBCD concentrations in breastmilk and the intake values used in the half-life calculations
are from abstracts or pre-published papers and could not be verified, the values are within similar
magnitudes as other published values for the Swedish population in literature.
Changes to either of the two parameters, fraction of body weight that is lipid (Fi) and HBCD half-life
(ti/2), would change the estimated dose.
The estimated doses across all population types ranged from 1.1E-09 to 1.5E-02 mg/kg/day.
2.4.2.4 Comparison of General Population Approaches
The figure below shows how these two approaches compare. The overall distribution based on the
biomonitoring data appears to be lognormal and the EPA estimated doses fall within the range of doses
derived from. This comparison provides confidence that EPA is within the correct order of magnitude to
estimate doses to the general population.
Page 274 of 723
-------
1e-1
1e-2
1e-3
|T 1e-4
~5>
|> 1e-5
0
o 1e-6
Q
1e-7
1e-8
1e-9 A,
Studies
Doses from Individual Human Biomonitoring Data
A General Population
O Highly Exposed Population
¦ Occupational Population
~ Placental and Fetal Tissues
Central Tendency Dose Range from Aggregate Exposure Pathways
High-End Dose Range from Aggregate Exposure Pathways
Figure 2-6. Comparison of HBCD Exposure via Environmental Monitoring/Exposure Factor and
Human Biomonitoring/Reverse Dosimetry Approaches
As described earlier in the section, it is unknown how scenario-specific estimates of exposure for highly
exposed populations compare to the doses estimated for the general population. It is also unknown how
temporal trends will ultimately impact biomonitoring studies. One recent study from Australia has
looked at biomonitoring of HBCD over time after their phase out. The authors note that while HBCD
levels are starting to decline, it may be some time before levels decline significantly due to the
persistence of HBCD in the body and ongoing sources of HBCD in the environment (Drage et al. 2015).
This approach is for total HBCD, not specific to the isomeric forms. While not specifically addressed in
this assessment, HBCD exists in three isomeric forms (alpha, beta, gamma). The different isomeric
forms have Koctanoi:Water values that differ by more than one log unit, whose biological half-lives vary
significantly (Szabo et al. 2011b; Szabo et al. 2011a. 2010). It is not known if the isomers have species
specific differences in toxicokinetics or toxicodynamics between animals and humans. Given these
uncertainties in the isomeric forms as well as in the pharmacokinetic data used in developing the
equivalent doses, there are uncertainties in the estimated external exposure doses based on
biomonitoring data. Biomonitoring studies in the literature are summarized in the Risk Evaluation for
Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on General Population,
Environmental, and Consumer Exposure Assessment (U.S. EPA 2019d). There is not a pharmacokinetic
model to fully describe the relationship between HBCD dose and lipid-adjusted HBCD concentrations in
humans, so therefore there is uncertainty associated with using a simpler approach to describe
toxicokinetics and toxicodynamics of HBCD.
Page 275 of 723
-------
2.4.2.5 General Population Subsistence Fisher Exposures
Aggregate exposures were also estimated for subsistence fishers. Subsistence fishers represent a PESS
group for HBCD due to their greatly increased exposure via fish ingestion (142.4 g/day compared to a
high-end of 22.2 g/day for the general population). Based on the increased ingestion rate (U.S. EPA
2000a) and various measured HBCD concentrations in fish both downstream (Near Field) and far away
(Far Field) from a releasing facility, EPA calculated aggregate general population exposure for
subsistence fishers. While EPA did not model subsistence fisher exposures due to releases associated
with a particular condition of use or OES, the use of measured HBCD concentrations in fish found
downstream of a nearby facility provide a reasonable estimate of HBCD exposure fish ingestion for a
highly exposed population.
EPA selected the best representative biomonitoring fish tissue concentrations from (Chen et al. 2011). In
this U.S. study, HBCD fish tissue concentrations were measured in 2006-2007 in three rivers, one
downstream of a nearby HBCD point source (Hyco River) and two others (Dan River, Roanoke River)
representing far-field fish tissue concentrations. The data from common carp was selected to use in the
Risk Evaluation because common carp represents an edible fish and generally contained the highest
HBCD concentrations. Table 2-93 presents the lipid-weight tissue concentrations as reported in (Chen et
al. 2011). and wet-weight concentrations converted from the lipid-weight concentrations using the
reported measured lipid content.
Table 2-93. Measured HBCD Concentrations From Various Species and Locations in (Chen et al.
2011)
River
Species
N
Mean Lipid
% (median in
parentheses)
Mean lipid weight concentrations (ng/g)
(median in parentheses)
Mean wet
weight conca
(ng/g)
a-
P"
v-
2HBCD
common
carp
7
9% (9%)
4270 (4700)
71 (54)
300 (250)
4640 (5010)
417.6
Hyco
channel
catfish
2
9% (9%)
3580 (3580)
60 (60)
46 (46)
3680 (3680)
331.2
(Near-
Field)
redhorse
sucker
2
6% (6%)
1340 (1340)
10(10)
53 (53)
1400 (1400)
84
gizzard
shad
2
9% (9%)
277 (277)
0.5 (0.5)
15 (15)
290 (290)
26.1
common
carp
7
13% (12%)
150 (73)
4.4 (3)
21 (21)
176 (100)
22.88
Dan
channel
catfish
9
12% (10%)
145 (111)
1.7(1)
5(3)
152(115)
18.24
(Far-
Field 1)
redhorse
sucker
3
10% (8%)
14(15)
<0.2 (0.2)
1.3(1)
16(16)
1.6
flathead
catfish
6
13% (11%)
667 (360)
17(6)
14 (6)
698 (370)
90.74
Roanoke
common
carp
7
11% (10%)
38 (32)
1.6(1)
14 (7.2)
54 (40)
5.94
channel
catfish
5
8% (7%)
58 (56)
0.7 (0.7)
1.9(1.8)
60 (59)
4.8
Page 276 of 723
-------
River
Species
N
Mean Lipid
% (median in
parentheses)
Mean lipid weight concentrations (ng/g)
(median in parentheses)
Mean wet
weight conca
(ng/g)
a-
P"
v-
2HBCD
(Far-
Field 2)
redhorse
sucker
8
8% (7%)
18(17)
0.3 (<0.2)
2.8 (2.6)
21 (20)
1.68
gizzard
shad
5
9% (8%)
10 (9)
0.4 (<0.2)
2.6 (2.4)
13 (12)
1.17
a Lipid weight concentrations were converted to wet weight concentrations using the reported mean lipid percentage.
These concentrations in ng/g were converted to mg/g and the dietary intake of fish for the subsistence
fisher was calculated using a fish ingestion rate of 142.2 g/day (U.S. EPA 2000a). Subsistence fishers
rely on fish for their protein intake, so the elevated fish ingestion exposures replaced the entirety of the
meat subset of diet. The subsistence fisher diet estimate was aggregated with other exposure pathways in
the same manner as was done for the general population (Section 2.4.2.2.5).
Central tendency exposure estimates for subsistence fishers for each exposure pathway and the
aggregated total are presented below in Table 2-94, with adult general population included for
comparison. The near-field subsistence fisher aggregate exposure is approximately 200-fold higher than
the adult general population. Based on reasonably available information, EPA is unable to determine
subsistence fisher exposure estimates specific to younger lifestages.
Table 2-94. Aggregate Central Tendency Exposure Comparison for Subsistence Fishers
Group
DIET
DUST
SOIL
AIR
DERMAL
ALL
Adult General Population
1.6E-06
9.7E-07
1.8E-10
1.7E-07
3.1E-07
3.1E-06
Subsistence Fisher NF
(Hyco)
7.4E-04
9.7E-07
1.8E-10
1.7E-07
3.1E-07
3.1E-06
Subsistence Fisher FF 1
(Dan)
4.2E-05
9.7E-07
1.8E-10
1.7E-07
3.1E-07
4.3E-05
Subsistence Fisher FF 2
(Roanoke)
1.2E-05
9.7E-07
1.8E-10
1.7E-07
3.1E-07
1.3E-05
All exposure values shown represent mg/kg.
2.4.3 Highly Exposed General Population Exposures
2.4.3.1_ Approach and Methodology
In this evaluation, highly-exposed general population include individuals who are expected to live close
to facility or residential sources, representing an example of Potentially Exposed or Susceptible
Subpopulations (PESS, see Section 2.4.8). EPA identified additional scenarios for the highly-exposed
general population, some of which were assessed quantitatively and some of which were assessed
qualitatively. This section contains discussion regarding two pathways that were assessed quantitatively:
1. emissions to water and subsequent ingestion of fish tissue (Scenario HI) and 2. emissions to air and
subsequent inhalation of particles (Scenario H2). Other scenarios considering exposure to EPS and XPS
insultation in buildings during use (Ql) and HBCD sent to landfill across the lifecycle (Q2) were
assessed qualitatively and discussed in Section 2.4.5.
Page 277 of 723
-------
Exposure from scenario-specific uses was estimated for emissions to water and air, as depicted in Figure
2-7. For quantitative analysis, exposure was modeled using the scenario-specific release estimates that
are summarized in Section 2.3.1. Modeled dust and indoor air concentrations, modeled outdoor air
concentrations, modeled water concentrations, and estimated soil, fish, and dietary concentrations will
be considered alongside available monitoring data.
Uses
Scenario-
specific
Summary of
Release Types
Emission to water:
• Surface water
• On-site WWT
• POTW
Emission to air:
• Fugitive
• Stack
• Incinerator
Summary of Media
or Pathways
Dietary (Fish)
(H1)
Air
(H2)
Exposure
Scenarios
Emission factor
Number of release
days
Treatment
fractional removal
Flow rates
Emission factor
Number of release
days
Media Estimation
Methods
Water
• Modeled: E-FAST
• Monitored
Fish
• Water & BAF
Air
Modeled: IIOAC
Figure 2-7. Overview of Exposure Assessment Method for Highly Exposed Scenarios
2.4.3.2 Near Facility Dietary (Fish) — Ingestion
EPA estimated highly exposed fish ingestion using modeled scenario-specific surface water
concentrations (point source) plus a lipid normalized upper trophic level fish BAF to convert the surface
water concentrations to fish tissue concentrations (Method 1). For comparison, EPA also estimated
possible dose ranges using all available fish-tissue monitoring data (Method 2), as well as all surface
water monitoring data plus lipid normalized upper trophic level fish BAF (Method 3). While the
modeled estimates apply to a smaller population who live near a facility and may ingest fish caught
within proximity to the river, the fish ingestion estimates based on monitoring data apply to whatever
conditions were present when those samples were taken.
Equations
The equation used to estimate exposure due to fish ingestion when monitored or modeled surface water
concentrations are available is presented in Equation 2-20 below. Exposure calculated from fish tissue
concentration directly uses the same basic equation, but the fish tissue concentration (|ig/kg) is
substituted for the surface water concentration and the BAF is removed.
Equation 2-20
Where
ADR
ADD
ADR or ADD
SWC x BAF xIRx CF1 x CF2 x ED
AT
Acute dose rate used for acute non-cancer risk
calculations due to fish ingestion (mg/kg-day)
Average daily dose used for chronic non-cancer risk
calculations due to fish ingestion (mg/kg-day)
Page 278 of 723
-------
swc =
Surface water (dissolved) concentration (|ig/L)
BAF =
Bioaccumulation factor (L/kg)
IR =
Age-specific fish ingestion rate (g/kg bw-day)
CFi =
Conversion factor for mg/|ig
CF2 =
Conversion factor for kg/g
ED =
Exposure duration (considers near facility residential
mobility) (year)
AT =
Averaging time (year)
Bioaccumulation factor
The surface water concentrations (measured or modeled) were converted to fish tissue concentrations
using a wet weight BAF of 46,488 L/kg determined from the upper trophic level lipid normalized BAF
in Wu et al. (2010V
The application of a BAF to measured fish tissue concentrations is not applicable.
Concentrations
Wet weight fish tissue concentrations, converted from surface water concentrations as appropriate, are
reported in for each method. The data is described below.
Table 2-95. Summary of HBCD Fish Concentration Data for Estimating Fish Ingestion Dose
Data Approach
Data Description
Surface Water
Concentration (ng/L)
Wet Weight Fish
Tissue Concentration
(mg/kg ww)
Method 1. Modeled
Surface Water
Concentration3
21-day average dissolved water concentrations
from PSC modeling using 10th and 50th
percentile mean flows
Overall range:
6.8E-02 - 3.4E+04
Overall range:
3.6E-03 - 1.6E+03
Method 2. Fish
Tissue Monitoring
Data (wild-caught)b
66 studies with 1774 samples collected from
over 27 countries
n/a
Overall range:
ND - 1.0E+01
CT range:
2.0E-06 - 4.9E+00
Method 3. Measured
Surface Water
Concentration3
14 studies with 600 samples collected from the
following countries: AQ, CA, CN, DK, GB,
JP, KR, PL, US, and ZA
Overall range:
ND - 3.1E+03
CT range:
4.3E-04 - 3.1E+03
Overall range:
ND - 1.4E+02
CT range:
2.0E-05 - 1.4E+02
a The measured and modeled surface water concentrations were converted to fish tissue concentrations using a low-end lipid normalized
upper trophic level fish BAF value of 46,488.
b If wet weight fish tissue concentrations were not available, lipid-weight fish tissue concentrations were calculated using a generic 5% lipid
content.
Method 1. Modeled Scenario Specific Surface Water Concentrations
Specifically, 21-day average dissolved surface water concentrations were obtained from modeling
performed using the Variable Volume Waterbody Model (VVWM) - Point Source Calculator (PSC)
(U.S. EPA 2019q). A summary of the condition of use scenarios modeled, including release estimates, is
described in the Environmental Exposure section (Section 2.3.1). A description of the modeling
approach is provided in Section 2.3.2.2.2. For fish ingestion, the modeling used harmonic mean surface
water flows which represent long-term average flow conditions. The 50th percentile flow was used to
estimate the central ADD and the 10th percentile flow used to estimate the high-end ADD. The 50th
percentile harmonic mean flow concentrations ranged from 6.8E-05 to 1.2 |ig/L and the 10 percentile
Page 279 of 723
-------
harmonic mean flow concentrations ranged from 1.9E-03 to 3.4 E+01 |ig/L. Results by specific
subscenario are presented in Appendix G.
Method 2. Measured Fish-Tissue Concentrations
Fish concentrations were reported in the literature on a lipid weight and wet weight basis. Species-
specific lipid content as reported by the individual studies, was not collected. Lipid content in fish
ranges from <1% to 15% (U.S. EPA. , ). To convert from lipid concentration to wet weight
concentration, Equation 2-21 is used.
Equation 2-21
Cone, ww = Cone, Iw x
% lipid
100%
Where
Cone, ww = Concentration on a wet weight basis, |ig/kg ww
Cone, Iw = Concentration on a lipid weight basis, |ig/kg lw
% lipid = Percentage of fish that is comprised of lipids
EPA used a generic default of 5% lipid content for any monitoring study that only reported fish-tissue
data in wet weight and did not provide enough detail on lipid-weight to estimate a lipid weight
concentration.
Charts depicting fish-tissue concentrations in all extracted studies are provided in the Risk Evaluation
for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on General Population,
Environmental, and Consumer Exposure Assessment.
As discussed above, Chen et al. (2011) provides the best representative U.S. study of fish biomonitoring
concentrations. In this study fish samples were collected from three rivers in Virginia and North
Carolina in 1999-2002 (n=189) and in 2006-2007 (n=183). The area in general is not industrial or
densely populated, but is in an area of textile production. Concentrations in fish varied significantly
between the rivers. The highest concentrations were from the Hyco River (maximum mean of 4640 ng/g
lw), which the authors hypothesized was because a textile-related facility was located approximately 10
km upstream from the sampling sites. The maximum mean concentrations in the Dan River fish and the
Roanoke River fish were lower at 698 and 60 ng/g lw, respectively. The authors hypothesize that levels
were higher in the Dan River watershed because the area has traditionally been home to more textiles
and furniture operations than the Roanoke watershed. A temporal analysis showed increase in
concentrations from 1992-2002 to 2006-2007, which may have been due to the emergence of HBCD
point sources in the mid-2000s in this local study. The use of HBCD in textiles is currently considered a
historical activity. More recent follow-up studies in this area are not available to investigate current
conditions and trends. The study results do indicate higher concentrations near point sources and lower
concentrations in diffuse source-derived areas. This is corroborated in the Chen et al. (2011) meta-
analysis of seventeen U.S. and international studies which showed HBCD concentrations were 1 to 2
orders of magnitude higher in freshwater fish sampled near point sources (38 to 6,660 ng/g lw) than in
freshwater fish sampled further away from sources (0.1 to 51.5 ng/g lw).
Method 3. Measured Surface Water Concentrations
Charts depicting surface water concentrations in all extracted studies are provided in the Risk Evaluation
Page 280 of 723
-------
for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on General Population,
Environmental, and Consumer Exposure Assessment (U.S. EPA 2019cT). As discussed in the
Environmental Exposure section, estimated central and high-end concentrations, respectively, were 0.84
and 0.99 |ig/L for near facility and 4.1E-04 and 8.0E-04 [j,g/L for away from facility (see Table 2-56).
Ingestion Rate
EPA used the same fish ingestion rates as for the general population assessment. Specifically, EPA used
data from (U.S. EPA 2014b) along with mean body weights for each age group from (U.S. EPA 2011b)
to calculate a g/kg-day ingestion rate. The high-end or ADR doses were calculated using the high-end
fish ingestion rate and the central or ADD doses were calculated using the central fish ingestion rate.
Exposure Duration and Averaging Time
An exposure duration and averaging time of 1 day was used for the acute ADR. For ADD calculation
using the modeled scenario-specific data, EPA assumed that children in the highly exposed group live
near a facility with elevated concentrations of HBCD for the entire duration of that life stage. EPA
assumed that adults in the highly exposed group live near a facility for a portion of their adult life,
depending on whether it was high-end or a central tendency estimate. The upper-end estimate for
residential mobility is 33 years and was selected for a high-end exposure duration (U.S. EPA 201 lb).
For a central tendency estimate for residential mobility, a value of 12 years was selected (U.S. EPA
2011b). For the other portion of their adult life, it was assumed that they were exposed to central
tendency fish-tissue concentration values based on monitoring data. Residential mobility was not
factored into the equation for the measured surface water and tissue methods because the values cannot
be attributed to a specific point source. For the averaging time, the ADD calculation used the years
within an age group.
Results
The central and high-end fish ingestion estimates from the scenario specific surface water modeling are
provided in Table 2-96 and Table 2-97 for the array of exposure scenarios and age groups.
Table 2-96. Highly Exposed Group: Range of High-End HBCD Fish Ingestion Dose by Scenario
and Age Group (mg/kg/day)
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
performed to include different source types and high-end and central tendency release estimates. For
scenarios with site-specific information, this information was used in the IIOAC model runs to
determine the meteorological station and population setting. When site-specific information was not
known, representative central tendency and high-end meteorological stations were used, along with
other default parameters (see Appendix G). For a given exposure scenario, a range of estimated air
concentrations was derived for each source type (fugitive, stack, incineration) at the fenceline and in the
community. Fenceline estimates were defined as air concentrations at 100-meter from the source while
community-averaged estimates were defined as average air concentrations within 100 to 1,000-meter
from the facility.
The range of modeled daily-averaged and annual-averaged results are presented in Table 2-98 for an
averaged indoor and outdoor air concentration by scenario and source type. Across all scenarios, the
average air concentration ranged from 1.5x 10"8 to 11.3 |j,g/m3 for fugitive sources, 4.70x 10"7 to 2.9
|j,g/m3 for stack sources, and 9.5><10"7 to 0.50 |j,g/m3 for incinerator sources. Ambient air concentrations
were modeled in IIOAC and averaged together with indoor air concentrations using the fraction of the
day spent outdoors, which was informed by the EPA's Consolidated Human Activity Patterns Database
(U.S. EPA 2009b). Indoor air concentrations were estimated using an indoor/outdoor ratio of 0.95 for
high-end estimates and 0.65 for central tendency estimates (U.S. EPA 2019r). When a choice was
available for central tendency or high-end average air concentrations, high-end fenceline results were
used for daily-averaged air concentrations and central tendency community-averaged results were used
for annual-averaged concentrations. For scenarios without site-specific information, IIOAC runs were
performed using both the representative central tendency and the high-end meteorological stations. In
these cases, the maximum high-end daily fenceline air concentration and the minimum mean
community-averaged annual air concentration are presented in Table 2-98.
Table 2-98. Overall Summary of HBCD Averaged Indoor and Outdoor Air Concentrations for 12
Emission Scenarios
Scenario Name
Fugitive Air
Concentration Range
(jig/m3)
24-Hour Average /
Yearly Average
Stack Air Concentration
Range (jug/m3)
24-Hour Average /
Yearly Average
Incineration Air
Concentration Range
(jig/m3)
24-Hour Average /
Yearly Average
1. Processing: Repackaging of
Import Containers
6.7E-02 - 5.9E+00 /
8.7E-04 - 4.4E-03
1.2E-02 - 8.5E-01 /
6.7E-04 - 3.4E-03
3.3E-04 - 3.2E-02 /
2.6E-04 - 1.3E-03
2. Processing: Compounding of
Polystyrene Resin to Produce
XPS Masterbatch
3.4E-03 - 2.6E-02 /
5.4E-06 - 6.4E-06
4.9E-04 - 3.8E-03 /
4.1E-06 - 4.9E-06
NA
3. Processing: Manufacturing
of XPS Foam using XPS
Masterbatch
1.3E-01 - 2.8E+00 /
5.1E-05 - 5.1E-05
1.9E-02 - 3.5E-01 /
3.9E-05 - 3.9E-05
NA
4. Processing: Manufacturing
of XPS Foam Using HBCD
Powder
1.6E-02 - 3.5E-01 /
6.4E-06 - 6.4E-06
2.3E-03 - 2.9E+00 /
4.9E-06 - 3.5E-04
6.8E-03 - 2.3E-01 /
1.8E-04 - 1.9E-04
5. Processing: Manufacturing
of EPS Foam from Imported
EPS Resin Beads
2.0E-01 - 1.1E+01 /
8.7E-04 - 4.4E-03
3.2E-02 - 1.6E+00 /
6.7E-04 - 3.4E-03
2.1E-02 - 5.0E-01 /
5.4E-03 - 1.0E-02
6. Processing: Manufacturing
of SIPs and Automobile
Replacement Parts from
XPS/EPS Foam
3.4E-03 - 5.1E-01 /
4.4E-05 - 2.0E-04
5.9E-04 - 7.2E-02 /
3.4E-05 - 1.5E-04
3.3E-03 - 3.1E-01 /
2.6E-03 - 6.5E-03
Page 284 of 723
-------
Scenario Name
Fugitive Air
Concentration Range
(jig/m3)
24-Hour Average /
Yearly Average
Stack Air Concentration
Range (jig/m3)
24-Hour Average /
Yearly Average
Incineration Air
Concentration Range
(jig/m3)
24-Hour Average /
Yearly Average
7. Use: Installation of
Automobile Replacement Parts
NA
NA
NA
8. Use: Installation of XPS/EPS
Foam Insulation in Residential,
Public and Commercial
Buildings, and other Structures
9.0E-04 - 8.9E-02 /
1.6E-08 - 5.8E-06
NA
1.4E-03 - 6.6E-02 /
9.5E-07 - 1.9E-04
9. Demolition and Disposal of
XPS/EPA Foam Insulation
Products in Residential, Public
and Commercial Buildings, and
Other Structures
8.0E-04 - 7.1E-01 /
1.5E-08 - 1.3E-05
NA
NA
10. Processing: Recycling of
EPS Foam and Reuse of XPS
Foam
1.4E-04 - 1.7E-01 /
6.1E-07 - 3.1E-06
2.2E-05 - 2.1E-02 /
4.7E-07 - 2.3E-06
1.5E-05 - 5.9E-03 /
3.8E-06 - 4.5E-06
11. Processing: Formulation of
Flux/Solder Pastes
2.9E-04 - 3.1E-02 /
6.6E-06 - 6.7E-06
1.9E-03 - 1.6E-01 /
7.5E-05 - 7.6E-05
NA
12. Use of Flux/Solder Pastes
NA
NA
5.8E-06 - 1.2E-03 /
4.5E-06 - 5.1E-06
Gray cells indicate no release data for this source.
The range of acute dose rate (ADR) and average daily dose (ADD) are presented in Table 2-99 and
Table 2-100, respectively, by scenario and age group. ADR and ADD were calculated using the average
air concentrations from Equation 2-22, and with a conservative assumption that 100% of inhaled
particles are ingested. The daily-averaged and annual-averaged air concentrations were used to calculate
the ADR and ADD, respectively. EPA used the Exposure Factors Handbook (U.S. EPA 2011b) to
inform age-specific body weights and inhalation rates. Specific exposure factors are provided in
Appendix G. Across all scenarios, COU 5 (Manufacturing of EPS Foam from Imported EPS Resin
Beads) resulted in the highest ADR values for all age groups, with infants having the maximum ADR.
Similarly, COU 5 also resulted in the highest ADD values for all age groups, with young toddlers having
the maximum ADD.
Equation 2-22
Where
ADD =
ADR =
AC =
InhR =
CFi =
ED =
BW =
AT =
. „ „ . „ ACxInhRxCFi xED
ADD or ADR = 1
BWxAT
Average daily dose (mg/kg-day)
Acute dose rate (mg/kg-day)
Average air concentration (|ig/m3), daily-averaged air concentration for ADR and
annual-averaged air concentration for ADD
Inhalation rate, in m3/hr for ADR and m3/day for ADD
Conversion factor from mg to |ig
Exposure duration, in days for ADR and years for ADD
Body weight (kg)
Averaging time, in days for ADR and years for ADD
Page 285 of 723
-------
Table 2-99. Highly Exposed Group: Range of HBCD Inhalation Dose by Scenario and Age Group, Acute Dose Rate (mg/kg/day)
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
Table 2-100. Range of HBCD Inhalation Dose by Scenario and Age Group, Average Daily Dose (mg/kg/day)
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
2,4.3.4 Aggregate Highly Exposed Population Exposure and Dose
Aggregate doses were calculated for the highly exposed population by summing the central tendency
general population dose pathways with the highly exposed dose pathway. Specifically, the aggregate
dose for Scenario HI is the sum of the highly exposed fish ingestion dose and all other central tendency
general population non-fish dose pathways. This calculation was not made for infants because infants
are not expected to ingest fish in their diet. For further discussion of risks from highly exposed fish
ingestion for infants and other lifestages, see Section 4.2.3.2. For Scenario H2, the aggregate dose is the
sum of the highly exposed inhalation dose and all other central tendency general population non-
inhalation dose pathways.
Table 2-101 and Table 2-102 show a summary of the ADR and ADR aggregate dose estimates for
Scenario HI, respectively. Table 2-103 and Table 2-104 show a summary of the ADR and ADR
aggregate dose estimates for Scenario H2, respectively.
Page 288 of 723
-------
Table 2-101. Range of HBCD Aggregate Exposure Acute Dose Rate (mg/kg/day) - Background and Modeled Fish Dose by Scenario
and Age
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
SCENARIO NAME
Infant
Young
Toddler
(l-<2 yrs)
Toddler
Small
Child
(3-<6 yrs)
Child
Teen
Adult
Adult
(<1 yr)
(2-<3 yrs)
(6-
-------
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
SCENARIO NAME
Infant
(<1 yr)
Young
Toddler
(l-<2 yrs)
Toddler
(2-<3 yrs)
Small
Child
(3-<6 yrs)
Child
(6-
-------
2.4.4 Consumer Exposures
2.4.4.1 Approach and Methodology
In this evaluation, consumers include individuals who have articles containing HBCD in their homes or
automobiles. Quantitative exposure estimates were developed for the consumer exposure scenarios as
described in Figure 2-9 based on the conditions of use within the scope of this Risk Evaluation.
Uses
Summary of Summary of Media Exposure
Release Types or Pathways Scenarios
Media Estimation
Methods
EPS/XPS Insulation in
Residences
(C1)
Mass transfer
from foam
(residence) or
replacement
parts
(automobile) to
particles
Indoor Air;
Settled and
Suspended Dust
Defined by:
• Consumer use
• Time spent in
location
• Emission rate
• Sorption rate
• Airflow rate
Air
• Modeled: IECCU
Dust
• Modeled: IECCU
Other
• Background (CT)
HBCD Contained in
Automobile Components
(C2)
Recycled Consumer
Articles
(C3)
Extraction by
saliva
Mouthing
Defined by:
• Concentration in
article
• Saliva extraction
efficiency
Mouthing
• Ingested Dose
Other
Background (CT)
Figure 2-9. Overview of Exposure Assessment Method for Consumer Exposure Scenarios
Scenario CI (emissions from XPS/EPS insulation installed in residential homes) corresponds to
condition of use #8 and #9 and Scenario C2 (emissions from HBCD-containing automobile components)
corresponds to condition of use #7. For these scenarios the presence and fate of HBCD in vapor phase,
settled dust, airborne particulate matter, and interior surfaces was investigated through a series of
simulations conducted for a "typical" residential building and a "typical" passenger vehicle by using
existing mass transfer models and simulation tools. Most parameters were either obtained from data in
the literature or estimated with empirical and QSAR models. All the simulations were conducted with
IECCU version 1.1 (U.S. EPA 2019p). The modeling results were compared with limited experimental
data. The predicted HBCD concentrations in settled dust in the living space were in line with the field
measurements. Additionally, the predicted temperature dependence of the HBCD emission rate is in
good agreement with the laboratory testing results reported by the Japanese researchers. Additional
details are provided in 5.4.2Appendix G. Doses over time were estimated using modeled concentrations
for the time spent in the simulated microenvironment and measured concentrations for time spent in
other environments (i.e., general population estimates).
2.4.4.2 XPS/EPS Insulation In Residences — Indoor Air and Settled Dust
Equation 2-23 and Equation 2-24 were used to estimate inhalation and dust ingestion, respectively, from
XPS/EPS insulation in residence. Total dose was calculated as a sum of the inhalation and incidental
ingestion routes.
Equation 2-23
ADR and ADD =
ACtotal x IR x CF x ED
BWxAT
Page 293 of 723
-------
ADR and ADD =
AC total
IR
CF
ED
BW
AT
Acute dose rate (ADR) or average daily dose (ADD) or due to inhalation
of vapor and airborne particulate matter (mg/kg-day)
Concentration in indoor air across all microenvironments accounting for
time spent in each microenvironment (|ig/m3)
Inhalation Rate (m3/day)
Conversion factor for mg/|ig (0.001)
Exposure duration (1 day for ADR or years in age group for ADD)
Body weight (kg)
Averaging time, non-cancer (1 day for ADR or years in age group for
ADD)
Where:
ACtotal = (ACmodeled x FDmodeled) + (ACother x (1 — FDother))
ACtotal = Concentration in indoor air across all microenvironments accounting for
time spent in each microenvironment (|ig/m3)
ACmodeled = Concentration in modeled indoor air of simulated microenvironment
(|ig/m3)
ACother = Concentration in air (ambient and indoor) of other microenvironments
(|ig/m3)
FD = Fraction of time spent in simulated residence over 24 hrs (unitless)
Equation 2-24
ADR and ADD =
DCtotal x IR x CF x ED
BWxAT
ADR and ADD =
DC
IR
CF
ED
BW
AT
Acute dose rate (ADR) or average daily dose (ADD) to inhalation of
vapor and airborne particulate matter (mg/kg-day)
Concentration in indoor dust across all microenvironments accounting for
time spent in each microenvironment (|ig/g)
Dust Ingestion Rate (g/day)
Conversion factor for mg/|ig (0.001)
Exposure duration (1 day for ADR or years in age group for ADD)
Body weight (kg)
Averaging time, non-cancer (1 day for ADR or years in age group for
ADD)
Where:
DCtotal = (DCmodeled x FDmodeled) + (DCother x (1 — FDother))
DCtotal --
DCmodeled =
DCother =
I'D =
Concentration in indoor dust across all microenvironments accounting for
time spent in each microenvironment (|ig/g)
Concentration in modeled indoor dust of simulated microenvironment
(|ig/m3)
Concentration in indoor dust of other microenvironments (|ig/m3)
Fraction of time spent in simulated residence while awake (unitless)
Page 294 of 723
-------
IECCU was used to model indoor air and dust concentrations in the living area of a three-zone generic
residential building, as described by Bevington et al. (Bevington et al. 2017). The HBCD source was
unfaced polystyrene insulation boards containing 0.5% HBCD, applied to both a vented attic and vented
crawlspace. Model inputs are described in further detail in Appendix G. The concentration of air and
dust in other microenvironments was assumed to be equivalent to the general population air and dust
estimates.
Age-specific ADR values used the highest modeled 24-hour average indoor air and dust concentration in
the simulated residence and the high-end general population estimates in the other microenvironment,
combined with a high-end intake. For age-specific ADD values, the long-term average indoor air and
dust concentration in the simulated residence and the central tendency general population estimates in
the other microenvironment were combined with central tendency intake. The concentrations were
weighted for the time spent in the simulated microenvironment (residence) and other microenvironments
(outdoors, vehicle, and/or commercial/public/government/child occupied facilities) over 24 hours (for
inhalation exposure) or while awake (for incidental ingestion exposure). The fractions of time spent
were derived from an analysis of CHAD activity pattern data for stay-at-home, part-time, and/or full-
time populations (U.S. EPA 2009b). The maximum fraction of time spent in the simulated environment
was used for the ADR (0.83 for air and 0.85 for dust) and the central fraction of time spent in the
simulated environment was used for the ADR (0.71 for air and 0.62 for dust). Age-specific inhalation
rates (mean and 95th percentiles), dust ingestion rates (mean and 95th percentiles), and bodyweights
(mean) for males and females were calculated from Exposure Factors Handbook (U.S. EPA 2011b) and
are provided in Appendix G.
The total dose estimates for Scenario CI are provided in Table 2-105, Table 2-106 for the ADD and
Table 2-107 for ADR. These tables also provide aggregate doses considering the addition of background
exposures from the diet, soil, and dermal pathways calculated for the general population.
Table 2-105. Age Specific ADR Associated with Residential Insulation Scenario CI
Age Group
TOTAL (Dust + Air) ADR
(mg/kg/day)
AGGREGATE ADR
(mg/kg/day)
Infant (<1 year)
2.3E-04
2.6E-04
Young Toddler (l-<2
years)
2.1E-04
2.9E-04
Toddler (2-<3 years)
1.8E-04
1.8E-04
Small Child (3-<6 years)
1.3E-04
1.4E-04
Child (6-
-------
Age Group
TOTAL (Dust + Air) ADD
AGGREGATE ADD
(mg/kg/day)
(mg/kg/day)
Small Child (3-<6 years)
2.7E-05
3.4E-05
Child (6-
-------
Table 2-107. Age Specific ADR Associated with HBCD in Automobile Component Scenario C2
Age Group
TOTAL (Dust + Air) ADR
(mg/kg/day)
AGGREGATE ADR
(mg/kg/day)
Infant (<1 year)
7.8E-04
8.0E-04
Young Toddler (l-<2
years)
6.8E-04
6.9E-04
Toddler (2-<3 years)
5.6E-04
5.7E-04
Small Child (3-<6 years)
4.2E-04
4.2E-04
Child (6-
-------
SA = Surface area of object in contact with the mouth (cm2)
T = Daily mouthing time (hr/day)
CF = Conversion factor for mg/|ig (0.001)
ED = Exposure duration (1 day for ADR and 1 year for ADD)
EF = Exposure frequency (1 day/year for ADR and 365 days/year for ADD)
BW = Body weight (kg)
AT = Averaging time (1 day for ADR and 1 year for ADD)
Where:
T = MDxCFxTA
T= Daily mouthing time (hr/day)
MD = Hourly mouthing duration (min/hr)
CF = Conversion factor for hr/min (0.0167)
TA = Time awake (hr/day)
EPA did not identify experimental data that measured migration of HBCD into saliva. Therefore, a
regression between concentration and migration rate into saliva for a variety of other chemicals found in
consumer products (U.S. EPA 2019r) was used to estimate the HBCD migration rate (MR; y=2E-
05xA0.9851). For surface area of objects in contact with the mouth (SA), the central tendency value (10
cm2) was used to calculate ADD while the high-end value (50 cm2) was used to calculate ADR (OECD
2019). Hourly mouthing durations are based time spent mouthing all non-pacifier items. The mean and
95th percentiles, used for the ADD and ADR respectively, are 7.1 and 13.1 min/hr for infants (0-
-------
The concentration of HBCD present in 97% of the consumer articles identified as likely to be mouthed
ranged from <1 ppm to 250 ppm. While HBCD can be present in many consumer articles, presence at
levels <250 ppm is not likely to impart flame retardancy and is likely due to mixing of recycled
feedstocks from many sources. Generally, as the concentration of HBCD increases, the potential for
imparting flame retardancy and the potential for exposure increases. Presence of HBCD at higher levels
(>250 ppm) may also be due to mixing of recycled feedstocks from many sources. For this analysis,
EPA used all data for products likely to be mouthed rather than identify a lower or upper cut-off based
on the potential for exposure and/or the potential for imparting flame retardancy. Based on these data,
the highest estimated ADR exposure was 1.86E-02 mg/kg/day for infants and 1.26E-02 mg/kg/day for 1
to 2 year olds. The highest estimated ADD exposure was 2.01E-03 for infants and 9.24E-04 for 1 to 2
year olds.
When the maximum mouthing specific ADD and ADR estimates are combined with the general
population background ADDs, the aggregate ADD and ADR, respectively, are 2.05E-03 and 1.86E-02
mg/kg/day for infants (0-<104 |ig/day (2.06><106 |ig ^ 100 days).
To estimate the effect of indoor emissions on ambient air, consider a 100-square mile, densely populated
urban area with a housing density of 1,000 units per square mile. In this example, the total source
strength is 2,06/ 109 |ig/day.
Total source strength = 2.06x 104 |ig/day x 100 mile2 x 1,000 units/mile2 = 2.06x 109 |ig/day
Next, the size of the air box that moves through the urban area over a 24-h period was calculated using
the mixing height, wind speed and travel distance, and diameter of the city area. The mixing height in
urban area is usually between 300 and 1,000 m. Consider the worse-case scenario with a mixing height
of only 150 m due to temperature inversion, which was the case during the London fog episode in 1952.
The worst-case scenario also occurs when there is little wind. In this calculation, a wind speed of 1 m/s
was used {i.e., the Beaufort number = 1 on a 0-to-12 scale). Thus, the distance of the air will travel over
a 24-h period is 1 m/s x 3,600 s/h x 24 h = 86,400 m. Furthermore, the diameter of the city area (100
mile2) is 18,200 m. From these values, the size of the air box moving through the city over a 24-h period
can be calculated:
Page 299 of 723
-------
Air box volume = 150 m x 86,400 m x 18,200 m = 2.35xlO11 m3
Dividing the total source strength by the air volume yields the HBCD concentration in the urban air
below the mixing height:
Possible Concentration = 2.06><109 jug 2.35xlOu m3 = 8.75xl0"3 |ig/m3
If other factors are considered such as other types of buildings which may have insulation and the
fraction of total buildings that have HBCD EPS or XPS insulation as opposed to other kinds of
insulation, there is additional variability that should be considered in the quantified air concentration. It
is noteworthy that this estimated air concentration is near the top-end of the range for extracted ambient
air monitoring data. In summary, emissions from HBCD insulation to ambient air represent a potential
ongoing source of exposure to the environment.
2.4.5.2 HBCD Sent to Landfill Across the Lifecycle
HBCD is not designated as a RCRA hazardous waste because it is not specifically listed as a known
hazardous waste and does not exhibit the characteristics of a hazardous waste (ignitability, corrosivity,
reactivity or toxicity) described under RCRA (40 CFR § 261). Because HBCD is not a RCRA hazardous
waste, HBCD wastes from across its lifecycle can be disposed of in hazardous waste, municipal, or
Construction and Demolition (C&D) landfills. Municipal and hazardous waste landfill design and
management controls such as coverings, liners, and leachate collection and treatment may partially or
fully mitigate migration of HBCD through landfills to groundwater. However, these features may be less
common for Construction and Demolition landfills which may be subject to less strict design
requirements and regulation.
The potential for landfilled HBCD to migrate through these landfills and reach receptors was
qualitatively assessed. There is a low potential for HBCD released to landfills to migrate through the
landfill to groundwater and reach receptors via groundwater ingestion or groundwater entering surface
water. HBCD is a solid and likely to be entrained in a solid matrix (XPS/EPS foam) when disposed of in
a landfill. HBCD's high soil organic carbon partition coefficient (>100,000) and low water solubility (66
ug/L) indicates it will preferentially partition to soil organic carbon and exhibit very slow movement
through soil to groundwater.
Very few studies to inform the potential for HBCD to migrate from landfills to the environment were
found. Available studies address the potential for HBCD to leach from substrates such as waste plastics
and XPS/EPS under laboratory conditions, and analysis of field collected landfill leachate for HBCD.
HBCD leaching through mixed waste and organic matter in lysimeters intended to represent conditions
in a municipal landfill were conducted by (Kajiwara et al. 2014). The waste contained approximately
13% by weight waste plastic. The plastic waste added to the lysimeter was determined to contain 390
ng/g total HBCD and the composite waste 4100 ng/g. The lysimeters were subjected to simulated
rainfall, and HBCD was detected in leachate from the beginning of the experiment before declining to
below the detection limit within 4 months. The study, which used waste materials as found (i.e., not
treated with additional HBCD), resulted in peak concentrations of 75 ng/L in leachate. ("Stubbings and
Harrad 2019) examined the leachability of HBCD from treated EPS and XPS foams. Concentrations in
pure water leachate from both foam types were in the high-|ig/L range, and in the low-mg/L range when
dissolved humic matter was included in the leaching fluid as an extractant. The concentrations measured
in this study were likely dominated by a fraction associated with small (<0.45 jam) foam particles
Page 300 of 723
-------
generated by abrasion. Preferential partitioning to these foam particle solids would be consistent with
the physicochemical properties of HBCD (low water solubility, high log Kow).
The results of these studies address the partitioning of HBCD from simulated landfill waste material or
XPS/EPS foam to leachate, but do not address mitigation of HBCD movement by landfill leachate
controls if present, or dilution/attenuation in soil. The relationship between HBCD loading in the study
and HBCD loading of U.S. landfills is unknown. Thus, applying these results to leaching of HBCD in
U.S. landfills to the environment would introduce significant uncertainty.
A limited number of HBCD landfill leachate monitoring studies were found. A study of leachates
collected from three landfills in South Africa found detectable concentrations of HBCD isomers in the
particulate phase but not in the dissolved phase (defined as passing through 0.45 |im filter paper) (Paso
et al. 2017; Olukunle and Okonkwo 2015; Gavitan-Garcia et al. 2017; Remberger et al. 2.004).
The available leachate monitoring data are limited to non-U.S. sites. The loading of HBCD to U.S.
landfills is unknown, as is whether the loadings result from current COUs or when they occurred.
Leaching from landfills may provide a pathway for exposure, but the resulting concentrations in
groundwater and surface water will be greatly attenuated by partitioning from water to solids and
retardation of particle transport.
The bioavailability of HBCD bound to solids in landfill leachate is also a source of uncertainty.
It is not currently possible to conduct a reliable quantitative assessment of HBCD exposure to human
and environmental receptors via landfill leachate. Insufficient information is reasonably available to
estimate HBCD landfill loadings on a per landfill basis that would be required to estimate leachate
concentrations. Generic characterization of the geology and hydrology associated with landfills
(underlying soil types, permeability, depth to groundwater, etc.) are lacking. Physical, chemical and
biological processes which may impact HBCD transport and transformation in a landfill are not well
understood.
2.4.5,3 Occupational Exposure Associated with the Condition of Use of Land
Disposal of Formulated Products and Articles
The Condition of Use of Land Disposal of Formulated Products and Articles is the land disposal of
articles that are a part of municipal solid waste (MSW.) The articles are specifically articles that are
associated with the minor uses of HBCD, which include textiles, electronics that contain HIPS, and
articles that contain adhesives and coatings.
Process Description:
Prior to disposal, solid waste may be first sent to waste transfer facilities where waste is compacted then
loaded onto larger vehicles for shipment to disposal sites such as landfills (https://www.epa.gov/sites/
production/files/2016-03/documents/r02002.pdf). At many transfer stations, workers screen incoming
waste located on conveyor systems, tipping floors, or in waste pits to identify recyclables and wastes
inappropriate for disposal (e.g., hazardous waste, whole tires). Workers at transfer stations operate heavy
machinery such as conveyor belts, push blades, balers, and compactors, and may also clean the facility
or perform equipment maintenance. Workers may be exposed to poor air quality due to dust and odor,
particularly in tipping areas over waste pits (https://www.epa.gov/sites/production/files/
03/documents/r02002.pdf).
Page 301 of 723
-------
Solid waste generally arrives at landfills in trucks and is dumped into a specific location in the landfill,
compacted with a compactor vehicle, and finally covered with soil from a nearby area
(https://www.cdc.gov/niosh/hhe/reports/pdfs/1993-0696-2395.pdf). Alternatively, the solid waste that is
received at a landfill is first shredded in a rotary hammer that pulverizes the waste and the shredded
waste is then dumped and spread on the landfill (https://www.cdc.gov/niosh/hhe/reports/pdfs/1991-
Q354-2532.pdf.) Workers at landfills operate heavy machinery such as compactors, loaders, and
bulldozers (https://www.cdc.gov/niosh/hhe/reports/pdfs/19 9-2616.pdO. This heavy machinery is
used to handle solid waste as well as soil used for daily cover
(https://www.cdc.gov/niosh/hhe/reports/pdfs/19' 9-2.616.pdf). Heavy machinery operators may be
exposed to particulates and other contaminates while in the cabs of the machinery
(https://www.cdc.gov/niosh/hhe/reports/pdfs/1996-0109-2616.pdf and
https://www.cdc.gov/niosh/hhe/reports/pdfs/1993-0696-2395.pdf). Mechanics servicing equipment may
be exposed to residues on machinery. In addition, workers may be exposed when removing dirty work
uniforms (https://www.cdc.gov/niosh/hhe/reports/pdfs/1996-0109-2616.pdf).
Number of Potentially Exposed Workers and Occupational Non-Users
EPA reviewed data from the BLS and related SOC codes. For workers handling waste at landfills, EPA
reviewed data for NAICS code 562212, Solid Waste Landfill, which indicates there are on average an
estimated 3 workers and 2 ONUs per site. For workers at waste transfer stations, EPA reviewed data for
NAICS code 562219, Other Nonhazardous Waste Treatment and Disposal, which provided the same
estimate for number of workers and ONUs. For 2019, the Waste Business Journal estimates 3,835 waste
transfer stations and 1,786 municipal solid waste landfills (Waste Business Journal. 2019). for a total of
7,198 landfills and waste transfer stations. Some of these facilities may not receive waste products and
articles containing HBCD, depending on the type of waste accepted at the facility. An upper bound
estimate would be 16,863workers and 11,242 ONUs for solid waste landfills and waste transfer stations.
Qualitative Inhalation Exposure Assessment
Occupational exposure to HBCD that results from the land disposal of MSW is comparable to
occupational exposure to HBCD that is the result of the processing or handling of articles that contain
HBCD because MSW disposal also involves potential worker exposure to HBCD as a result of the
processing or handling of articles. The relevant occupational exposure scenarios that involve exposure
that is the result of the processing or handling of articles that contain HBCD are as follows:
(1) Installation of XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings, and
Other Structures;
(2) Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures;
Based on the process description, the land disposal of MSW for the most part does not involve the
intentional breaking of articles although some processing steps such as compaction and loading and
unloading of MSW may result in the breaking of articles. The exception is the shredding of MSW. On
the other hand, all the above listed scenarios involve the intentional cutting or breaking of articles.
Accordingly, the rate of generation of inhalable dust resulting from the land disposal of an individual
article is likely less than the rate of generation of inhalable dust resulting from the processing of an
individual article in the case of any of the listed occupational exposure scenarios. Based on this, EPA
assumes the HBCD worker inhalation exposure concentration resulting from the disposal of MSW apart
from the shredding of MSW is lower than the HBCD worker inhalation exposure concentration
pertaining to the above listed exposure scenarios. The HBCD worker inhalation exposure concentration
resulting from the shredding of MSW may be greater than the HBCD worker inhalation exposure
Page 302 of 723
-------
concentration pertaining to the above listed exposure scenarios. HBCD inhalation exposure
concentration is dependent on not only the dust generation rate pertaining to an individual article but
also on (a) the concentration of HBCD in an article, (b) the number of articles processed per unit time,
(c) engineering controls, and (d) the rate of general mechanical or natural ventilation. These factors are
further discussed below in the discussion of uncertainties.
EPA assumes exposure duration and frequency are 8 hrs. per day and 250 days/year, respectively,
because EPA expects that articles that contain HBCD are randomly dispersed in MSW given that these
articles are used in buildings.
Discussion of Uncertainty in the Qualitative Assessment of Inhalation Exposure
1. The Concentration of HBCD in an Article:
EPA lacks data pertaining to the concentration of HBCD in textiles and in articles which contain
adhesives and coatings that contain HBCD. If these concentrations were greater than the
concentration of HBCD in XPS/EPS, and if dust were generated as a result of the land disposal
of these article, then the concentration of HBCD in this dust will exceed the concentration of
HBCD in dust that is generated as a result of cutting or breaking XPS/EPS during construction
and demolition. Furthermore, if the rate of inhalable dust generation during land disposal is
sufficiently high, then EPA's assumption that HBCD inhalation exposure concentration which
pertains to land disposal of articles is relatively low may not be valid.
2. The Number of Articles Processed per Unit Time:
The rate of total dust generation equals the rate of total dust generation from an individual article
on average multiplied by the number of articles containing HBCD that are processed per unit
time at a land disposal site. EPA lacks data pertaining to the number of articles containing
HBCD that are processed per unit time at a land disposal site. If this rate were sufficiently high,
then EPA's assumption that the HBCD inhalation exposure concentration which pertains to land
disposal of articles is relatively low may not be valid. The articles containing HBCD that are
processed per unit time at a land disposal site is likely low because of the following two factors:
first, the articles that contain HBCD likely comprise a small fraction of MSW and, second, for
the most part HBCD was used in XPS/EPS. With regard to the first factor, in 2017, 6.3% of the
municipal solid waste was from textiles, and less than 2% was from electronics (with 36% of this
recycled), (https://www.epa.gov/facts-and-figures-about-materials-waste-and-recvcling/guide-
facts-and-figures-report-about-materials#Materials and Products and (1, J \IS_-.QA'hi)- Even if
acute HBCD inhalation exposure concentration is relatively high because EPA's relevant
assumptions pertaining to inhalation exposure concentration are not valid, the average HBCD
inhalation exposure concentration may nonetheless be relatively low if the number of articles
were low.
3. Engineering Controls and General Mechanical or Natural Ventilation:
The land disposal of MSW is an open process (i.e., material is not processed in enclosed
equipment) although the shredding of MSW may be a partially enclosed process. Similarly, all
the occupational exposure scenarios listed above involve open processes. All the occupational
exposure scenarios discussed above including the Land Disposal of Formulated Products and
Articles involve outdoor and indoor workplaces and therefore all ventilation rates may be
comparable although EPA is uncertain of this.
4. HBCD Inhalation Exposure Concentration Associated with Shredding of MSW:
Page 303 of 723
-------
EPA lacks information about the prevalence of the shredding of MSW; if this step were note
prevalent, then the number of workers who would be potentially exposed would be low. The
average HBCD inhalation concentration may be low if the number of articles containing HBCD
that are shredded per unit time is low and/or if there are engineering controls to protect against
worker exposure. However, there may be acute worker exposure.
Qualitative Dermal Exposure Assessment
During the land disposal of MSW, workers may handle the articles that contain HBCD, but EPA does
not expect worker dermal exposure because the HBCD is entrained in the articles. Workers may be
exposed to settled dust that contains HBCD as a result of the shredding of MSW.
2.4.6 Sensitivity Analysis - Human Exposure
2.4.6.1 Sensitivity Analysis - Infant Exposures
For the highly exposed general population, EPA further considered infant exposures and reports
additional percentiles beyond the 95th percentile using different assumptions. In EPA's approach, the
selection of which upper percentile is assigned to the high-end monitoring data is generally more
sensitive than the selection of the geometric mean.
In this sensitivity analysis, EPA examined the effect of varying three assumptions related to the
stochastic modeling of HBCD aggregate dose for infants (<1 year) in the general population:
1. In the baseline stochastic analysis of HBCD doses modeled above, only the 95th percentile
estimate of modeled HBCD doses is reported as a high-end estimate. In this analysis, EPA also
reported the 96th, 97th, 98th, 99th, and 99.5th percentiles of estimated HBCD dose.
2. In the baseline (previous) analysis, environmental concentrations were assumed to follow
lognormal distributions, with the central tendency and high-end concentrations reported in
monitoring data used to define the shape of the lognormal distribution. Specifically, the central
tendency estimate from monitoring data was assumed to correspond to the median of the
lognormal distribution, while the high-end estimate from monitoring data was assumed to
correspond to either the 95th percentile (for soil and dust) or the 90th percentile (all other
environmental and biotic media). In this analysis, EPA varied this assumption by allowing all
high-end monitoring data values to represent the 90th, 95th, and 99th percentile of the underlying
lognormal distribution.
3. In the baseline analysis, the central tendency estimate from monitoring data was assumed to
correspond to the median of the lognormal distribution, which is equivalent to assuming that the
central tendency estimate was equal to the geometric mean of the underlying distribution. In this
analysis, EPA varied this assumption by 10% in either direction of the geometric mean to
evaluate the sensitivity of the output to the central tendency estimate.
The results of varying assumptions 1 and 2 in the sensitivity analysis are visualized in Figure 2-10. The
x-axis shows alternative percentiles that can be used to estimate the high-end dose, ranging from the 95th
to the 99.5th percentile of the output dose distribution. The y-axis displays the estimated dose in
mg/kg/day at each of these percentiles. The different curves each represent an alternative assumption
with respect to the shape of the underlying environmental distributions. Specifically, each series presents
an analysis based on assuming the reported high-end monitoring data value for environmental
concentrations represented either the 90th, 95th, or 99th percentile of the underlying lognormal
distribution; the baseline analysis is also pictured. Results for the 99%, and 99.5% percentile, and
Page 304 of 723
-------
maximum modeled dose for each assumption of the underlying distribution are provided in Table 2-110.
Results for the 99%, and 99.5%t percentile, and the maximum modeled dose for each assumption of the
underlying distribution are provided in Figure 2-10.
>¦>
ee
T3
clj
2
o
Q
0.01
Assumed Percentile for
High-End Monitoring
L001
0.0001
Baseline
95 96 97 98 99
Percentile of Aggregate Dose Estimates
100
Figure 2-10. Comparison of HBCD Dose for Infants in the General Population from Different
Sensitivity Analyses
Based on a Figure 2-10, it is possible to conclude:
1. High-end aggregate dose estimates are sensitive to the choice of percentile used to represent
high-end doses. Choosing the 99.5th percentile of the stochastic dose output instead of the 95th
percentile can increase estimated high-end dose by a factor of 3. This is consistent with the
theoretical expectation that dose estimates would be left skewed in their distribution with a long
tail to the right.
2. If it is assumed that the reported high-end value from monitoring data represents a higher end
percentile of the underlying distribution of environmental data (e.g., 99th percentile instead of
90th percentile), the estimated dose decreases. This is consistent with the theoretical expectation
that a longer tail will result in larger estimated dose.
3. The baseline analysis is very similar to the analysis in which the reported high-end value from
monitoring data represents the 90th percentile of the underlying distribution of environmental
data. This is because the baseline analysis assumes the reported monitoring high-end estimate
represents the 90th percentile for all distributions except soil and dust for which it was assumed
to represent the 95th percentile.
Page 305 of 723
-------
Table 2-110. Sensitivity Analysis of Upper End Monitoring Distribution Assumptions in
Monitoring Data
Assumed Percentile of
Monitoring Distribution
Upper End
Estimated Dose in mg/kg-day
99th Percentile
Estimated Dose
99.5th Percentile
Estimated Dose
Maximum
Modeled Dose
90th
8.7E-04
1.3E-03
3.6E-03
95th
3.5E-04
4.5E-04
1.0E-03
99th
1.7E-04
1.9E-04
3.2E-04
The results of varying assumption 3 in the sensitivity analysis are summarized in Table 2-111.
Table 2-111. Sensitivity Analysis of Central Tendency Estimate Assumptions in Monitoring Data
Estimated Dose Based on
Varying Monitoring Data
Central Tendency Assumption
Esl
timated Dose in mg/kg-day
Baseline GM
Baseline GM + 10%
Baseline GM -10%
95th Percentile Dose
3.1E-04
3.2E-04
2.9E-04
Maximum Estimated Dose
3.3E-03
3.5E-03
3.3E-04
% Change from Baseline
(95%tile)
—
4%
-7%
% Change from Baseline
(Maximum Dose)
—
6%
-0%
GM = geometric mean
The highest theoretical maximum aggregate exposure to infants is 3.59E-3 mg/kg-day where the
maximum modeled HBCD dose is combined with the lower (90th) assumed percentile of the underlying
distribution of environmental data. This value is similar to the maximum modeled HBCD dose from the
higher-end assumption (+10%) of the true central tendency value (3.45E-3 mg/kg-day).
2.4.6.2 Sensitivity Analysis - Variation in Production Volume
EPA considered releases using three production volumes acknowledging decreasing trends of releases.
EPA notes that chronic doses decrease by a factor of approximately two to four when releases are
similarly reduced by a factor of two to four. Acute doses are approximately the same because EPA
inferred that reduced release days when the magnitude of releases decreases. EPA also considered three
separate approaches to estimated fish doses and the overall magnitude and trends associated with all
three approaches are similar.
A sensitivity analysis examining varying production volume and waste water treatment removal was
conducted for human exposures, using a parallel approach as was described in Section 2.3.6 for
environmental exposures. The results are summarized in the table below. The estimated surface water
concentrations were used to derive fish ingestion doses as described previously in Section 2.4.3.2.
Page 306 of 723
-------
Table 2-112. Summary of Surface Water Concentrations from Sensitivity Analysis: Varying
HBCD Production Volume and Waste Water Treatment Removal -Human Exposures (Fish
Ingestion)
Scenario Name
Production
Volume
(lbs / year)
% WWTP
Removal for
Direct Releases
Surface Water 21-Day Average Dissolved
Concentration Range (jig/L)
Acute:
10th %-ile Flow
Chronic:
50th %-ile Flow
Scenario 1. Import
and Re-packaging/
Processing:
Repackaging of
Import Containers
100,000
90%
2.1E-01 - 1.4E+00
6.9E-03 - 2.0E-01
50,000
90%
1.2E-01 - 7.5E-01
4.1E-03 - 1.0E-01
25,000
90%
6.0E-02 - 7.1E-01
2.0E-03 - 1.0E-01
Scenario 3.
Processing:
Manufacturing of
XPS Foam using
XPS Masterbatch
100,000
0%
2.6E-03 - 9.2E-01
8.9E-05 - 3.2E-02
50,000
0%
1.3E-03 - 4.6E-01
4.4E-05 - 1.6E-02
25,000
0%
6.5E-04 - 2.3E-01
2.2E-05 - 8.2E-03
Scenario 5.
Processing:
Manufacturing of
EPS Foam from
Imported EPS
Resin Beads
100,000
0%
2.0E+00 - 3.4E+01
6.8E-02 - 1.2E+00
50,000
0%
4.4E-01 - 1.1E+02
1.7E-02 - 1.2E+00
25,000
0%
5.0E-01 - 3.6E+01
7.4E-02 - 5.0E+00
2.4.7 Assumptions and Key Sources of Uncertainty in the General Population, Highly
Exposed, and Consumer Exposure Assessment
Estimates of general population exposures based on environmental monitoring and biomonitoring data
represent the conditions present at the time the data was collected. It is unknown which combination of
potential sources associated with conditions of use as described in this Risk Evaluation contribute to the
monitoring data presented here. However, given the wide range of exposures shown within and across
the monitoring data, there is a plausible contribution from some of the sources/conditions of use
described within this document.
For the general population assessment, EPA used central tendency and high-end environmental
monitoring data informed by all studies for a given media that passed evaluation. EPA also compared
pathway specific estimates with completed assessments already reported in the literature. For example,
EPA's dietary assessment is of similar magnitude or higher than those reported for other countries (Lee
et al.. 2019; Cao et al.. 2018; Barghi et al.. 2016; Fromme et al.. 2015). EPA also used all extracted
biomonitoring data and estimated external doses based on assumptions of lipid-weight percentages first-
order elimination, constant lipid concentrations of HBCD throughout the body, and half-life. The half-
life is based on calculations using breastmilk concentrations and intakes for a group of women; however,
the intake was not based on paired measurements in the subjects of the breastmilk study but instead on a
general market basket study for the population as a whole. In addition, the dose reconstruction method
relies on an assumption of steady state and first order elimination. Longterm exposure to HBCD from
articles in the home may support an assumption of steady state, but elimination has not been adequately
characterized to firmly support a first-order assumption. Finally, the assumption that all lipid throughout
the body has the same concentration of HBCD cannot currently be verified using experimental data.
Page 307 of 723
-------
Because of these different assumptions, there is significant uncertainty in the dose reconstruction
calculations.
While there are approximately 400 monitoring studies across all media, there are limited studies within
the U.S. to characterize current and spatially diverse environmental levels. It is unknown whether the
currently available HBCD concentrations in environmental media outside of the U.S. are representative
of values in the U.S. While some media such as indoor dust and sediment have relatively more data,
other media such as human biota and surface water are less well characterized. A qualitative assessment
of the uncertainty, sensitivity, and variability associated with this approach is presented in Table 2-113
below.
Table 2-113. Qualitative Assessment of the Uncertainty and Variability Associated with General
'opulation Assessment
Variable Name
Data Source
Uncertainty
(L,H)
Variability
(L,H)
General Population Exposure Assessment (based on Environmental Monitoring)
Environmental Monitoring Data
Extracted and evaluated data (all)
plus key studies
L
H
Exposure Factors and Activity Patterns
Exposure Factors Handbook
L
L
General Population Exposure Assessment (based on Biomonitoring)
Biomonitoring Data
Extracted and evaluated data (all)
plus key studies
L
H
Half-life in the body
Select studies
H
H
Lipid weight in the matrix
Select studies
L
H
One-compartment approach
(Avlward and Havs 2011)
H
H
For the highly exposed group, EPA modeled three pathways: air, water to fish (fish ingestion), and
consumer articles to indoor air and dust. There are more input parameters used across these three
modeling approaches. EPA balanced a combination of central tendency and high-end inputs for these
modeled scenarios. Further, each scenario was split into many sub-scenarios to fully explore potential
variability. Modeled estimates were compared with monitoring data to ensure overlap and evaluate the
overall magnitude and trends. For example, fish ingestion doses were evaluated in three different ways
(see Section 2.4.3.2). A qualitative assessment of the uncertainty and variability associated with this
approach is presented in Table 2-114 below.
Page 308 of 723
-------
Table 2-114. Qualitative Assessment of the Uncertainty and Variability Associated with Highly
Exposed Population Assessment
Variable Name
Descriptor (data source)
Uncertainty
(L,H)
Variability
(L,H)
Environmental Exposure and Highly Exposed Groups Assessment (based on Exposure modeling)
Environmental Releases Category
Emission Factor
Range
(EU RAR, OECD ESD)
M
H
Days of Release
Range
(EU RAR, EU TGD, OECD ESD)
M
H
Production Volume
CDR volume threshold /Datamyne
H
L
Directly reported Releases
Reported values (TRI)
L
L
Environmental Fate Category
Physical-Chemical Properties:
KoC, Henry's Law Constant, etc
Point estimate
(measured values, modeled estimates)
L
L
BAF
Point estimate based on lower end of
range (measured studies)
L
H
Half-lives of HBCD in media
Range (measured studies)
L
H
Exposure Model Parameter Category
Water modeling defaults: river
flow, dimensions, characteristics
Range, CT and HE (PSC user guide)
L
H
Air modeling defaults:
meteorological data, indoor/outdoor
transfer,
Range, CT and HE (IIOAC user guide)
L
H
Consumer Article modeling
defaults: characterization of
emissions from articles,
characterization of residential and
auto environments)
Range, CT and HE (IECCU user guide
H
H
Exposure Factors and Activity
Patterns
Range, CT and HE (Exposure Factors
Handbook
L
L
L = low; M = moderate; H=high
EPA aggregated exposure across several pathways, in its general population assessment and found
general agreement between different approaches. EPA also substituted modeled estimates for scenario-
specific pathways for air, fish, and indoor air/dust for its assessment of highly exposed populations.
There was a wide range of release estimates reported within and across scenarios which results in
scenario-specific estimates that were lower than, of similar magnitude to, and higher than general
population estimates. When considering pathway specific estimates and aggregate exposures, there is
uncertainty associated with which pathways co-occur in a given population group. Further, there is
variability within a given exposure pathway. For the same exposure scenarios, central tendency
estimates are more likely to occur than high-end estimates. To address this, EPA used a stochastic
approach to simulate the effect of aggregated exposures. EPA used different combinations of exposures
sampling from the entire distribution for all pathways. This approach offers more clarity than static
sensitivity analyses based on combining assorted high-end and/or central tendency estimates of the
component distributions. For instance, combining the 95th percentile estimate of all component variables
Page 309 of 723
-------
in an exposure equation in a static sensitivity analysis may produce a conservative high-end estimate of
exposure that cannot be related to a specific percentile on the exposure distribution. Instead, EPA used a
stochastic analysis, and selected the 95th percentile to approximate a high-end exposure estimate. The
stochastic approach, however, is subject to uncertainty stemming from assumptions relating to the
component distributions. If the true component distributions differ in terms of shape and/or parameters
from the assumed distributions, the estimated exposure distribution may be potentially biased, especially
in the tails of the distribution.
Finally, EPA did not consider all possible exposure pathways, but rather focused on pathways that were
within the scope of its conceptual model. This may result in a potential underestimation of exposure in
some cases. Examples of exposure pathways that were not considered include incidental ingestion of
suspended sediment and surface water during recreational swimming and ingestion of non-fish seafood
such as aquatic invertebrates or marine mammals. However, EPA expects these exposures to be less
than those that were included in the aggregate assessment. As such, their impact will likely be minimal
and would be unlikely to influence the overall magnitude of the results.
2.4.8 Potentially Exposed or Susceptible Subpopulations
TSCA requires that a Risk Evaluation "determine whether a chemical substance presents an
unreasonable risk of injury to health or the environment, without consideration of cost or other non-risk
factors, including an unreasonable risk to a potentially exposed or susceptible subpopulation (PESS)
identified as relevant to the Risk Evaluation by the Administrator, under the conditions of use." TSCA
Section 3(12) states that "the term 'potentially exposed or susceptible subpopulation' means a group of
individuals within the general population identified by the Administrator who, due to either greater
susceptibility or greater exposure, may be at greater risk than the general population of adverse health
effects from exposure to a chemical substance or mixture, such as infants, children, pregnant women,
workers, or the elderly."
In developing the exposure assessment for HBCD, EPA analyzed the reasonably available information
to ascertain whether some human receptor groups may have greater exposure or susceptibility than the
general population to the hazard posed by HBCD. Exposures of HBCD would be expected to be higher
amongst groups living near industrial facilities {i.e., highly exposed general population), groups with
HBCD containing products in their homes, workers who use HBCD as part of typical processes, and
groups who have higher age and route specific intake rates compared to the general population.
EPA identified potentially exposed and susceptible subpopulations for further analysis during the
development and refinement of the life cycle, conceptual models, exposure scenarios, and analysis plan.
In Section 2.4, EPA addressed the potentially exposed or susceptible subpopulations identified as
relevant based on greater exposure. EPA addresses the subpopulations identified as relevant based on
greater susceptibility in Section 3.2.7.
Of the human receptors identified in the previous sections, EPA identifies the following as potentially
exposed or susceptible subpopulations due to their greater exposure and considered them in the Risk
Evaluation :
1. Workers and occupational non-users. EPA reviewed monitoring data found in published
literature including both personal exposure monitoring data (direct exposure) and area
monitoring data (indirect exposures) and identified data sources that contain measured
monitoring data and or/estimated data for the various conditions of use (including import and
processing of HBCD). Exposure estimates were developed for users (males and females workers
of reproductive age) exposed to HBCD as well as non-users or workers exposed to HBCD
Page 310 of 723
-------
indirectly by being in the same work area of the building (Table 2-80 and Table 2-81). Workers
exhibit higher breathing rates than the general population at rest, leading to elevated internal
dose even when exposed to similar air concentrations. Also, adolescents and female workers of
reproductive age (>16 to less than 50 years old) were also considered as a potentially exposed or
susceptible subpopulation as specified in Section 2.4.1.1. In Appendix E.8, EPA presents a
discussion and analysis of workers, including adolescents, by industry sector.
2. Consumers associated with consumer use/exposure. HBCD has been identified as being used in
products to which consumers may be exposed; however, only some individuals within the
general population may use these products. Therefore, those who do use these products are a
potentially exposed or susceptible subpopulation due to greater exposure. A description of the
exposure assessment for consumers is available in Section 2.4.4.
3. Subsistence fishers. Subsistence fishers ingest substantially more fish than the average member
of the general population and therefore experience much greater HBCD exposure via fish
ingestion. Aggregate exposure estimates for subsistence fishers are derived and described in
Section 2.4.2.5.
4. Highly exposed general population. Other groups of individuals within the general population
may be more highly exposed due to their proximity to conditions of use identified in Section 1.2
and Section 2.4.2.1 that result in releases to the environment and subsequent exposures (e.g.,
individuals who live or work near manufacturing, processing, distribution, use or disposal sites).
Section 2.4.3 provides an overview of types of receptors and exposure descriptors within the highly
exposed general population. EPA estimated age-specific exposures and doses for each overall exposure
group (Section 2.4.3.4) and acknowledges that individuals among the highly exposed general population
and other PESS categories overlap, as some individuals may belong to multiple receptor groups (as
described in Table 2-71). EPA also estimated ambient air concentrations for the highly exposed general
population, covering individuals of all lifestages living near facilities. Further characterization about
highly-exposed group and associated variability of exposure factors within the highly-exposed group is
discussed in the Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on General Population, Environmental, and Consumer Exposure Assessment.
In developing exposure scenarios, EPA considered age-specific differences (Section 2.4.2.1). For
HBCD, exposure scenarios that involve potentially exposed or susceptible subpopulations considered
age-specific behaviors, activity patterns, and exposure factors unique to those subpopulations. EPA used
the Exposure Factors Handbook (U.S. EPA. 201 lb) to inform body weights and intake rates for children
and adults also described in the Supplemental Information on General Population, Environmental, and
Consumer Exposure Assessment. Sections 2.4.2.2, 2.4.3.1, and 2.4.4.1 provide an overview of exposure
pathways considered for the different age groups.
There are some exposure scenarios where greater exposure from multiple sources may occur and
individuals who may have greater potential for exposure to HBCD. For example, as part of the Risk
Evaluation:
5. EPA used the CHAD database to inform how much time children spend in microenvironments
(Section 2.4.2.2.3) to determine children with elevated dust concentrations (Sections 2.4.4.2 and
2.4.4.3).
6. EPA considered breast milk concentration data and ingestion for breast-fed infants (< 1 year old)
in the exposure estimation (Section 2.4.2.3).
7. EPA used an activity-pattern based method to model hand-to-mouth and object-to-mouth contact
and to derive transfer rates of soil and dust to the mouth to estimate ingestion rate for children
Page 311 of 723
-------
and/or adults who ingest soil or sediment in environments where HBCD concentrations are
elevated (Sections 2.4.2.2.2 and 2.4.2.2.3), and for children who may mouth objects containing
HBCD (Section 2.4.4.4).
8. EPA completed an assessment of human dietary exposure from multiple sources for children or
adults who consume edible aquatic biota or terrestrial biota containing elevated levels of HBCD.
EPA considered available biomonitoring data in wildlife and dietary patterns across trophic
levels as part of its exposure assessment. These approaches were considered together to
determine HBCD concentrations in surface water, sediment, soil, and targeted wildlife biota. See
Section 2.4.2.2.1 for detailed information.
EPA also considered and analyzed the available data to ascertain whether some human receptor groups
may be exposed via pathways that may be distinct to a particular subpopulation or lifestage (e.g.,
children's crawling, mouthing or hand-to-mouth behaviors, see Appendix E) and whether some human
receptor groups may have higher exposure via identified pathways of exposure due to unique
characteristics (e.g., activities, duration or location of exposure) when compared with the general
population (U.S. EPA. 2006).
Page 312 of 723
-------
3 HAZARDS
3.1 Environmental Hazards
3.1.1 Approach and Methodology
During scoping and problem formulation, EPA reviewed potential environmental and health hazards
associated with HBCD. EPA identified the following sources of environmental hazard data: Technical
Review of HBCD ( 20]6e), Technical Review of Flame Retardant Alternatives for HBCD
(U.S. EPA. 2014d). National Industrial Chemicals Notification and Assessment Scheme (NICNAS)
Report on HBCD: Priority Existing Chemical Assessment (NICNAS 2012a). Environment Canada and
Health Canada Screening Assessment Report on HBCD (EC/HC 2011). European Union (EU)
Environmental Risk Assessment on HBCD (EINECS 2.008). EPA Risk-based Prioritization of HPV
Chemicals (U.S. EPA. 2008a). and S1DS Assessment of HBCD (OECD 2007). These sources describe
the hazards of HBCD to aquatic organisms including fish, aquatic invertebrates, aquatic plants and
sediment invertebrates exposed to relevant media under acute and chronic exposure conditions. These
publications report acute toxicity to aquatic invertebrates from HBCD, based on mortality and
immobilization as well as chronic toxicity to aquatic invertebrates (growth and reproduction) when
exposed to HBCD. Also, chronic toxicity was observed in benthic organisms based on reduced
survivability when exposed to HBCD. In addition, these assessments summarize the hazards of HBCD
to terrestrial organisms including soil invertebrates and avian species when exposed to relevant media
under acute and chronic exposure conditions.
Although the assessment documents mentioned above provide detailed information regarding the
environmental hazard of HBCD to aquatic and terrestrial organisms, they do not account for additional
and latest information published on HBCD. Therefore, EPA completed the review of environmental
hazard data/information sources during Risk Evaluation using the data quality review evaluation metrics
and the rating criteria described in the Application of Systematic Review in TSCA Risk Evaluations
document (U.S. EPA. 2018b). Studies that were considered "On Topic" were evaluated for acceptability.
The acceptable studies were rated as high, medium, or low for quality. The data quality evaluation
results are outlined in Supplemental File: Data Quality Evaluation of Environmental Hazard Studies
(U.S. EPA. 2019k). With the data, only studies rated as high, medium, or low for quality during data
evaluation were used during data integration. Any study rated as unacceptable was not used. Also, only
clearly adverse signs of toxicity (e.g., lethality, immobility, effects on growth and reproduction, organ
histopathology, abnormal behavior) were used to set toxicity effect levels such as lethal and effective
concentrations (i.e., LCso, EC so values) no-ob served-effect concentrations (NOECs) and lowest-
observed-effect concentrations (LOECs).
3.1.2 Hazard Identification
EPA identified 50 acceptable studies (i.e., rated high, medium or low) that contained aquatic toxicity
data (i.e., fish, aquatic invertebrates, algae) and terrestrial toxicity data (i.e., plants, earthworms, avian
species). Aquatic toxicity studies considered in this assessment are summarized in Table 3-1
This assessment evaluated not only studies that followed standard test guidelines (e.g., Office of
Chemical Safety and Pollution Prevention (OCSPP)), Organisation for Economic Co-operation and
Development [OECD]), but also non-standard toxicity tests that followed procedures that were
scientifically sound according to the Application of Systematic Review in TSCA. Risk Evaluations
document (U.S. EPA. 2018b). For this assessment, adverse signs of toxicity (e.g., lethality, immobility.
Page 313 of 723
-------
effects on growth and reproduction, organ histopathology, abnormal behavior) were used to set toxicity
thresholds.
Table 3-1. Environmental Hazard Characterization of HBCD to Aquatic and Terrestrial
Organisms
Test
Organism
Duration
Endpoint
Hazard
Value
Units
Effect Endpoint
Reference
Aquatic Organisms
Fish
Acute
96-hour LC50
>0.0025
mg/L
No effects on growth and
mortality
(Wildlife Intl 1997b)
(High)
96-hour
LOAEL
0.002
mg/L
Hatch delay
(Hu et al„ 2009a)
(High)
Chronic
88-day
NOEC
>0.0037
mg/L
No effects on growth,
reproduction or survival
(Drottar et al. 2001)
(High)
168-day
>0.02284
mg/kg
(diet)
disrupted thyroid
homeostasis
(Palace et al.. 2008;
Palace, et al., 2010)
(High)
42-day
>0.5
mg/L
No effects on growth and
mortality
(Zhana et al., 2008)
(High)
34-day
NOEC
>0.250
mg/L
No effects on growth and
mortality
(Foekema et al. 2014)
(High)
78-day
LOAEL
0.3
mg/kg
(lipid diet)
Thyroid effects
(Kuiuer et al.. 2007)
(High)
Sub-
chronic
17-day
50
mg/L
Abnormal growth,
malformation
(Hons et al., 2014)
(High)
Amphibians
Acute
8-day EC50
0.064
mg/L
Tail tip regression
(Schriks. 2006) (Hiah)
Invertebrates
(Pelagic)
Acute
48-hour EC50
>0.0032
mg/L
Immobilization
(Wildlife Intl 1998;
Wildlife Intl LTD 1997)
(High)
96-hour LC50
>0.8
mg/L
Mortality
(Shi et al. 2017a)
(High)
Chronic
21-day
NOEC
0.0031
mg/L
Reproduction; Growth
(weight and length)
(Wildlife Intl 1998)
(High)
21-day LOEC
0.0056
mg/L
21-day
MATC
0.0042
mg/L
Invertebrates
(Bentliic)
Chronic
28-day LOEC
>1,000
mg/kg dw
Mortality
(ACC 2003a. b)
(High)
28-day
NOEC
3.1a
mg/kg dw
Population
(Oetken et al. 2001)
(High)
28-day
NOEC
8.6 b
mg/kg dw
Population
28-day LOEC
28.7
mg/kg dw
Population
28-day
MATC
15.7
mg/kg dw
Population
Algae c
96-hour EC50
>0.0037
mg/L
Growth
(Wildlife Intl 1997b)
(High)
72-hour EC50
0.08
mg/L
Growth
(Walsh etal. 1987)
(High)
72-hour EC50
>0.041
mg/L
Growth
(Desiardins et al. 2004)
(High)
72-hour EC10
0.041
mg/L
Growth
72-hour EC50
0.052
mg/L
Growth
(Desiardins et al, 2005)
(High)
72-hour EC50
>0.01
mg/L
Growth
Terrestrial Organisms
Vegetation
Short-
term
96-hour
NOEC
>5,000
mg/kg dw
Emergence; Mortality;
Growth
(Wu et al. 2016c; Wu et
al. 2012; Porch et al.
02) (High)
Page 314 of 723
-------
Test
Organism
Duration
Endpoint
Hazard
Value
Units
Effect Endpoint
Reference
Chronic
21-day
NOEC
>5,000
mg/kg dw
Emergence; Mortality;
Growth
(Wu et al. 2016c; Wu et
al. 2012; Porch et al.
2002) (High)
Invertebrates
Chronic
56-day ECio
21.6
mg/kg soil
Effects not reported
Reproduction; Mortality
(Aufderheide et al.
2003) (High)
28-day
NOEC
128
mg/kg soil
56-dayNOEC
128
mg/kg soil
56-day LOEC
235
mg/kg soil
Reproduction
56-day LOEC
>4,190
mg/kg soil
Mortality
Avian Species
Chronic
22-day LOEC
0.001
mg/kg-d
Pipping success
(CruniD et al. 2010)
(High)
6-week
LOAEL
125
Hg/L
Hatching success
(MOE.T 2009)
(High)
15
mg/L
Offspring survival
2.1
mg/kg-d
5
mg/L
75-day
LOAEL
164.3
ng/g egg
WW
Corticosterone response in
males; Flying activities in
juvenile males; Predator
avoidance in juvenile
females
(Kobiliris 2010)
(High)
21-day
LOAEL
0.51-3.27
mg/kg-d
Delayed egg laying of
smaller eggs with thinner
eggshells
(Marteinson et al. 2012;
Fernie et al. 2011;
Marteinson et al. 2011;
Marteinson et al. 2010)
(High)
a Toxicity value reported by author.
b Toxicity value reported by author (normalized to organic carbon content in sediment).
c Because algae can cycle through several generations in hours to days, the data for algae was assessed together regardless of duration (/'. e., 48-hrs to 96 hrs).
Values in bold were used to derive Concentrations of Concern (COC) as described in Section 3.1.5 of this document. All values are listed individually with
study quality in [Data Quality Evaluation of Environmental Hazard Studies and Data Extraction for Environmental Hazard Studies. Docket: EPA-HQ-
OPPT-2019-05001 ¦
3.1.2.1 Aquatic Toxicity
Acute Fish Toxicity
Short-term effects of HBCD to fish were identified in six acceptable studies representing different
species including, rainbow trout (Oncorhynchus mykiss), zebrafish (Danio rerio) and Indian medaka
(Oryzias melastigma).
During an 96-hour acute toxicity study, rainbow trout (O. mykiss) were exposed to HBCD composed of
a, P, and y- diastereomers under flow-through conditions (Wildlife Intl 1997b). Rainbow trout (O.
mykiss) were exposed to six measured concentrations between 0 and 0.0025 mg/L. However no
mortalities or other effects were observed throughout the test. The results indicate that HBCD is not
acutely toxic to rainbow trout (O. mykiss) up to concentrations of >0.0025 mg/L. Other studies reported
adverse effects on the embryo toxicity (i.e., hatching, heart rate, development) of HBCD exposure.
However, most of these studies reported effects above HBCD's water solubility limit. In an embryo
toxicity study in zebrafish (D. rerio) conducted by Hu et al.. (2009a). delayed hatching was observed at
0.002 mg/L at 96-hours post fertilization, but not at the two highest exposure concentrations of 2.5 and
10 mg/L. Hatching success was not affected at any concentrations. In addition, no effects on survival or
malformation rates were observed in embryos exposed to concentrations up to 10 mg/L (highest
concentration tested). Other effects such as increase in heat shock protein at 0.01 mg/L and an increase
in malondialdehyde activity, used as a measure of lipid peroxidation, at 0.5 mg/L were observed. The
activity of superoxide dismutase was increased at 0.1 mg/L but decreased at 2.5 and 10 mg/L. The
Page 315 of 723
-------
author concluded that HBCD can cause oxidative stress and overexpression of Hsp70 in acute exposures
of zebrafish embryos.
Chronic Fish Toxicity
One acceptable study represents the chronic effects of HBCD to fish. In this study (Wildlife Intl
1997b). rainbow trout ((). mykiss) were exposed to HBCD at mean measured concentrations of
0.00025, 0.00047, 0.00083, 0.0018, and 0.0037 mg/L under flow-through conditions for 88 days.
Reagent grade acetone was used as a solvent control. The maximum nominal concentration was similar
to the measured water solubility of 0.0086 mg/L. No effects were found at the water solubility limit of
HBCD. The reported 88-day NOEC was >0.0037 mg/L. There were other studies that conducted sub-
chronic or chronic exposures of HBCD to fish and are summarized in the Risk Evaluation for Cyclic
Aliphatic Bromide Cluster (HBCD), Systematic Review Supplemental File: Data Extraction Tables of
Environmental Hazard Studies ( 319b). These other studies reported effects of chronic
exposure to HBCD including increased malformation rate, developmental abnormalities, oxidative
stress and apoptosis (Hone et at... 2014a). thyroid effects (Palace et al., 2008. ), oxidative damage
to lipids, proteins, and DNA and decreased antioxidant capacities in fish tissue (Zhang et al.. 2008).
Hong et al. ( ), examined the effects of HBCD on embryos of the marine medaka (Oryzias
melastigma). The embryos were exposed to HBCD for 17 days in an early life stage test. The
developmental abnormalities in medaka included yolk sac edema, pericardial edema, and spinal
curvature. Mechanistic findings in this study included increases in heart rate and sinus venosus-bulbus
arteriosus (SV-BA) distance, which are markers for cardiac development, induction of oxidative stress
and apoptosis, and suppression of nucleotide and protein synthesis. A maximum acceptable toxicant
concentration (MATC) of 0.03 mg/L was reported for this study. In contrast, Foekema et al. (2014)
observed no effects on mortality or development through metamorphosis (approximately 40 days post-
fertilization) in sole (Solea solea) embryos exposed in an early life stage test to concentrations of
HBCD up to 0.25 mg/L for 6 days, starting at 12 hours post-fertilization.
In other studies, thyroid effects were reported in juvenile rainbow trout (O. mykiss) following dietary
exposure to HBCD (Palace et al., 2008. 2010). Each of the diastereomers of HBCD (administered
separately via diet at concentrations of 5 ng/g of a-, P-, or y-HBCD for up to 56 days) disrupted thyroid
homeostasis, as indicated by lower free circulating T3 and T4 levels. No thyroid or other effects were
observed in European flounder (Platichthys flesus) following 78 days of diet or sediment exposure to
maximum concentrations of 3,000 |ig/g lipid in food and 8,000 |ig HBCD/g total organic carbon (TOC)
(Kuiper et al.. 2007).
Acute Invertebrate Toxicity
There are three acceptable studies that represent the acute toxicity of HBCD to aquatic invertebrates.
These studies include two water flea (Daphnia magna) studies and one copepod (Tigriopus japonicus)
study. The results of these acceptable studies show that HBCD is not acutely toxic to aquatic
invertebrates at the chemical's water solubility limit.
In one study (Wildlife Intl 1997). D. magna were exposed to mean measured concentrations of 0,
0.0018, 0.0021, 0.0023, 0.0024, and 0.0032 mg/L under flow-through conditions for 48 hours. No
effects were observed at the highest exposure concentration. In another study (Wildlife Intl 1998). The
water flea (D. magna) were exposed to mean measured concentrations of 0, 0.00087, 0.0016, 0.0031,
0.0056, and 0.011 mg/L for 21 days under flow-through conditions. No effects on mortality or
immobilization were observed at the highest exposure concentration after 48 hours of exposure to
HBCD. Both studies suggest that acute exposures to concentrations of HBCD below the HBCD water
Page 316 of 723
-------
solubility did not result in measured effects on mortality or immobilization behavior in D. magna.
Finally, the copepod (T. japonicus) were exposed to measured concentrations of 0, 0.008, 0.03, 0.08,
0.3, and 0.8 mg/L of HBCD for 96-hours (Shi et at ). Although the exposure concentrations were
tested above the water solubility limit, a solvent control (DMSO) was used. No effects were observed at
the highest exposure concentration.
Chronic Invertebrate Toxicity
There are four high-quality studies that represent the chronic toxicity of HBCD to aquatic invertebrates
representing freshwater and saltwater species in the water and sediment compartments. These studies
included one water flea (I). magna) study, two amphipod (Hyalella azteca) studies, one black worm
(Lumbriculus variegatus) study and one copepod (Tigriopus japonicus) study. There were effects on
growth and reproduction in the water flea (D. magna) after 21 days of exposure to HBCD. The
organisms were exposed to mean-measured concentration of 0, 0.00087, 0.0016, 0.0031, 0.0056, and
0.011 mg/L HBCD under flow-through conditions (Wildlife Intl 1998). An MATC of 0.042 mg/L was
calculated from a NOEC of 0.0056 mg/L and a LOEC of 0.031 mg/L. Also, there were effects on the
survival in the black worm (L. variegatus) after exposures to 0.05, 0.5, 50, and 500 mg/kg dry weight
(dwt) in sediment HBCD for 28-days (Oetken et al. 2.001). The effects are relevant at the population
level. In addition, HBCD induced developmental delay after 40 days of exposure to T. japonicus (Shi et
al. 2017a). Marine copepods were exposed to nominal concentrations of 0, 0.008, 0.03, 0.08, 0.3, 0.8
mg/L in water under static conditions. DMSO was used as a solvent. After 20 days of exposure, HBCD
caused growth delay to the copepod (T. japonicus nauplii). The LOEC for developmental delay (from
the nauplii to copepodid) were 0.03 and 0.008 mg/L for the Fo and Fi generations, respectively.
Similarly, the LOEC for developmental delay (nauplii to adults) were 0.3 and 0.03 mg/L for the Fo and
Fi generations, respectively, suggesting that the Fi generation was more sensitive to HBCD than the Fo
generation. For the water flea (H. azteca) no effects were observed at exposures of 31, 63, 125, 250,
500, and 1,000 mg/kg dw sediment (nominal concentrations) HBCD for 28 days in the presence of 2%
and 5% TOC (A.CC 2003a. b).
Other Acute and Chronic Effects
A wide range of effects of HBCD have been reported in fish (e.g., developmental toxicity, embryo
malformations, reduced hatching success, reduced growth, hepatic enzyme and biomarker effects,
thyroid effects, DNA damage to erythrocytes, and oxidative damage) and invertebrates (e.g.,
degenerative changes, morphological abnormalities, decreased hatching success, and altered enzyme
activity) in supporting studies that assessed endpoints beyond those evaluated in this assessment (Du et
al. 2015; Hong et al. 2015; Foekema et al. 2014; Hong et al. 2014; Zhang et al. 2014a; Wu et al. 2013;
Du et al. 2012a; Anselmo et al. 2011; Palace et al. 2010; Deng et al. 2009; Hu et al. 2009a; Smolarz and
Berger 2009; Aniagu et al. 2008; Palace et al. 2008; Zhang et al. 2008; Ronisz et al. 2004). Effects on
the thyroid in rainbow trout (O. mykiss) (reduced thyroid hormone (triiodothyronine, T3, and thyroxine,
T4)) (Palace et al. 2010; Palace et al. 2008; Kuiper et al. 2007; Lower and Moore 2007). are similar to
those observed in mammals. These studies were evaluated using metrics and the rating criteria described
in the Application of Systematic Review in TSCA. Risk Evaluations document ( 018b). These
studies were considered acceptable and are summarized in the Risk Evaluation for Cyclic Aliphatic
Bromide Cluster (HBCD), Systematic Review Supplemental File: Data Extraction Tables of
Environmental Hazard Studies document ( >).
Amphibians
For amphibians, one acceptable high quality study reported data on species of African clawed frog
(Xenopus laevis.)
Page 317 of 723
-------
Schriks et al. (2006) conducted a metamorphosis bioassay in the African clawed frog (X. laevis) to
detect thyroid hormone disruptive effects by HBCD. Pre-metamorphic X laevis tadpoles were pre-
treated with the goitrogen methimazole to inhibit thyroid synthesis prior to isolation of tadpole tips (6-8
mm). The tadpole tips were cultured ex vivo in dishes for 24 hours, and thereafter exposed to DMSO
(solvent control) and 6.4 mg/L of HBCD. On day 6, exposure of tail tips to 6.4 mg/L HBCD in
combination with 20 nM of T3, significantly (p < 0.05) potentiated tail tip regression with 35 ± 5%. All
lower HBCD exposures, including HBCD alone (0.64 mg/L), did not have any effects on tail tip
regression. At 6.4 mg/L HBCD alone or in combination with 20 nM T3 resulted in a very fast regression
of tail tips in the first two days of exposure. This was faster than tail tip regression in the 100 nM T3-
control, but after two days of exposure tail tip regression roughly stayed the same during the rest of the
experiment period. The study concluded that this fast regression was due to cytotoxic activity at this
concentration.
Aquatic Vegetation Toxicity
For aquatic plants hazard studies, algae are the common test species. Algae are cellular organisms which
will cycle through several generations in hours to days; therefore, the data for algae was assessed
together regardless of duration rather than being categorized as acute or chronic. There were five
acceptable studies for three species of algae (green algae and diatoms), including fresh and saltwater
species. Population changes were reported in the marine diatom (Skeletonema costatum) after 72 hours
exposure to HBCD (Walsh et al. 1987). The EC50 values were determined in four of the five test media
with different salinity for the marine diatom (S. costatum) and ranged from 0.009 to 0.012 mg/L. The
geometric mean EC50 was 0.010 mg/L. Also, in the same study, the green algae (Thalassiosira
pseudonana) were exposed to HBCD under the same conditions. The EC50 values were determined in all
six-test media and ranged from 0.050 to 0.370 mg/L. The reported EC50 value for T. pseudonana was
0.08 mg/L. No effects on population changes were reported at the solubility limit of HBCD for this
study.
A subsequent study by Desjardins et al. (2004) supports the acute toxicity of HBCD to the marine
diatom (S. costatum). In this study, the marine diatom (S. costatum) was exposed to a single test
concentration of HBCD, a negative control and a media control (no generator column) for 72 hours.
Measured test concentrations (as separate a, P and y isomers) were determined from samples of test
medium collected from the treatment and each control group at the beginning and end of the test. At test
initiation, an inoculum of the algal cells was added to each test chamber at a concentration of 77 000
cells/mL. Samples were collected from each replicate test chamber at 24-hour intervals to determine cell
densities and area under the curve (AUC) values. The arithmetic mean of total HBCD at test termination
was 0.041 mg/L and consisted of mean measured test concentrations for a-, P- and y-HBCD of 0.0305,
0.00886 and 0.161 mg/L respectively. The author reported that a concentration 0.041 mg/L resulted in
approximately 10% inhibition of growth in the marine diatom (S. costatum) after 72 hours exposure to
HBCD.
Desjardins et al. (2005) conducted another 72-hour study with S. costatum. This study consists of two
toxicity tests with HBCD using a co-solvent and performed at saturated solution. The biomass and the
growth rate were derived. For the co-solvent test, the primary stock solution was prepared in
dimethylformamide (DMF) at a nominal concentration of 0.10 mg/ml and diluted to secondary stock
solutions. Aliquots of the stock solutions were diluted with saltwater medium to prepare the nominal
concentrations of 0.00064, 0.0016, 0.004 and 0.010 mg/L. The analytical results performed at the
beginning of the test corresponded to 332, 131, 94 and 108% of the nominal concentration, respectively.
The solvent concentration in the solvent control an,d treatment groups was 0.1 ml/L. There was no
Page 318 of 723
-------
statistical difference between the control group and the test concentrations. The saturated solution test
was performed to study effects on algal growth of the mixed diastereomers of HBCD at the water
solubility limit. Only one test concentration was used. However, the concentrations used in the co-
solvent test and the concentration used in this assay combined, meets the requirement for an adequate
study. The authors mentioned that the test solution corresponded to the saturated solution of HBCD in
saltwater. The mean measured concentration of HBCD as a sum of the diastereomers was 0.0545 mg/L.
At the beginning of the test the following measured concentrations of the diastereomers were found: y =
0.00354 mg/L, P = 0.0152 mg/L and a = 0.0358 mg/L. The growth rate inhibition during the study was
17% compared to the column control after 24 hours, 29% after 48 hours and 51% after 72 hours. A non-
linear regression fitting to the cumulative normal distribution was used to calculate an ECso. The 72-hr
ECso for biomass and growth rate was calculated to be 0.027 and 0.052 mg/L, respectively.
There was one acceptable freshwater algal study conducted with HBCD. In this study, there were no
effects reported on abundances and population growth rate after 96-hour exposure to HBCD to
Selenastrum capricornutum (currently known as Raphidocelis subcapitata) (Wildlife Intl 1997b). This
freshwater green algae species was exposed to mean measured concentrations of 0.0013, 0.0022, 0.0033,
0.0042 and 0.0064 mg/L under static conditions for 96 hours. No dose response was found. Inhibition of
around 10% based on AUC after 96-hour was observed in the highest tested treatment. Averaging the
measured concentrations at the start and the end of the test for the highest exposed test group resulted in
a mean exposure concentration of 0.0037 mg/L.
3.1.2.2 Terrestrial Organisms
Toxicity to Soil Invertebrates
Three acceptable studies reported data on two species of earthworms. All three studies were rated as
high-quality. Aufderheide et al. 2003 conducted a 56-day study where earthworms (Eisenia fetida) were
exposed to HBCD to evaluate effects regarding reproduction and mortality. At 28-days a NOEC of
4,190 mg/kg dw soil was reported for mortality. The 56-day reproduction NOEC was 128 mg/kg dw
soil.
Another study examined the bioaccumulation potential of HBCD in earthworms (E. fetida and
Metaphire guillelmi) (Li et al. 2016). The tissue concentrations of a- and y-HBCDs were substantially
higher in E. fetida compared to those in M guillelmi, with the higher lipid and protein contents in E.
fetida as the primary reason for this difference. Other processes, such as uptake, depuration, metabolism
and isomerization, also differed between the two earthworm species and led to a difference in the
bioaccumulation of P-HBCD. The P- and y-HBCDs were bioisomerized to a-HBCD in the earthworms,
but to a greater extent in E. fetida.
Shi et al... (2017a) examined the effects of HBCD on the growth rate of the earthworm (E. fetida)
exposed to nominal concentrations of 0, 50, 100, 200, and 400 mg/kg dw and control (acetone). A
significant (p < 0.01) up-regulation of superoxide dismutase (SOD) expression level was observed in
earthworms exposed to HBCD at 400 mg/kg dw soil. The transcript level of Hsp70 gene was
significantly up-regulated (p < 0.01) when earthworms exposed to HBCD at 400 mg/kg (2.61-fold). A
LOAEL of 400 mg/kg dw soil was reported.
Toxicity to Avian Species
There are 11 studies that report data for exposure to HBCD for three avian species. These studies
include the domestic chicken (Gallus domesticus), Japanese quail (Coturnix japonica), and American
Page 319 of 723
-------
kestrel (Falco sparverius). The results of these high-rated studies show that HBCD is toxic all three
avian species with reported adverse effect on body weight, reproduction, development, behavior and
thyroid hormone regulation. In one study, short-term exposure to HBCD to the chicken (G. domesticus)
at nominal concentrations of 0, 0.006, 0.06, 0.6, 1.9, and 6.4 mg/L resulted in a significant up-regulation
of enzymes involved with metabolism of xenobiotic (Crump et al. 2.008). Also, significant down-
regulation of proteins associated with the thyroid hormone pathway and lipid regulation occurred in this
concentration range. A 36-hour LOAEL of 0.06 mg/L was reported. A follow-up to this study, (Crump
et al. 2010). reports the effects of HBCD on embryo toxicity, isomer-specific accumulation in liver and
cerebral cortex, and hepatic gene expression in the chicken (G. domesticus). HBCD was injected into the
air cell of chicken eggs prior to incubation. Measured concentrations of 220, 430, 1,500 (nominal),
4,980 and 50,000 (nominal) mg/L were used. Nominal concentrations were noted because the measured
concentrations for 300 and 10,000 ng/g could not be quantified. In addition, the authors reported the
effect concentrations based on the nominal stock concentrations. Eggs were observed for pipping
success for 22 days. Reduced pipping success was observed at 100 and 10,000 ng/g HBCD. Also,
isomeric composition of HBCD was significantly altered in hepatic tissue at 100 and 10,000 ng/g.
In another study (MOEJ 2009) adult mortality observed for the Japanese quail (C. japonica) increased at
1,000 mg/L. Also, dietary exposure of HBCD for C. japonica, resulted in reproductive toxicity (M1
2009). Quails were fed diets containing 0, 17.5, 33.4, 61.5 or 126.9 mg/kg-bw/day of HBCD (a mixture
of isomers: a, 27%; P, 30%; y, 43%) for six weeks. HBCD exposure resulted in a reduction in
hatchability at all concentrations examined. Statistically significant reduction in eggshell thickness (P<
0.05) was also observed at concentrations above 17.5 mg/kg/day. Also, HBCD exposure resulted in
decreased egg weights and production rate and an increase in cracked eggs at the two highest exposure
concentrations (61.5 mg/kg/day and 126.9 mg/kg/day). The effect on reproduction and development are
relevant for population effects.
Four acceptable studies reported data on the reproductive, development and behavior effect of HBCD in
American kestrels (/•'. sparverius), (Marteinson et al. 2012; Fertile et al. 2011; Marteinson et al. 2011;
Kobiliris 2010; Marteinson et al. 20 i 0).
Kobiliris (2010) reported a reduced "corticosterone response" (where "corticosterone response" was
defined as a stimulation of the adrenal cortex to produce and release corticosterone into the
bloodstream), reduced flying activities of juvenile males during hunting behavior trials, and delayed
response times of juvenile females during predator avoidance behavior trials in American kestrels (F.
sparverius) exposed in ovo to 164.13 ng/g wet weight (ww).
Fernie et al. (2011) examined the reproductive effects of HBCD on the American kestrels (/•'.
sparverius). HBCD dissolved in safflower oil was injected into the brains of dead cockerels daily. The
kestrels were fed a ration of cockerels equivalent to approximately 0.51 mg/kg-day. Dietary exposure
began three weeks prior to pairing and continued through courtship, egg laying, and incubation, until the
first chick hatched (approximately 75 days). Exposed kestrels laid eggs with average tissue
concentrations of 163.5 ng a- HBCD/g ww, 13.9 ng P-HBCD/g ww, and 2.6 ng y- HBCD/g ww.
Exposed birds displayed reduced time from pairing to egg laying and laid larger clutches of smaller eggs
(volume, mass). Exposed eggs lost more weight than control eggs during incubation, but egg shell
thickness was not affected. No effect on reproductive success was identified.
In a related study, Marteinson et al. (2010) found that accidental exposure of male in ovo American
kestrel (F. sparverius) to small concentrations of HBCD during exposure to pentaBDE technical
formulation (DE-71) may have contributed to synergistic/additive effects. HBCD levels in male
Page 320 of 723
-------
offspring of kestrels were measured at 3.27 ± 0.68 ng/g ww (low exposure), and 15.61 ± 2.63 ng/g ww
(high exposure) based on sibling eggs. HBCD levels were significantly negatively correlated with
reproductive success parameters of the male offspring: clutch size, fertility, copulation and behavior.
However, because PBDE levels were also significantly correlated to these parameters, the authors
determined that it was difficult to separate the influences of HBCD from those of PBDE.
In a subsequent study, Marteinson et al. ( ) exposed the American kestrel (/•'. sparverius) to dietary
HBCD at 0.51 mg/kg bw-day and found increased testes weight in unpaired males, an effect on testis
histology in unpaired males (increased number of seminiferous tubules containing elongated spermatids;
p = 0.052), marginally increased testosterone levels in breeding males (increased at the time the first egg
was laid; p = 0.054), and no significant effect on sperm counts. Plasma T4 levels were reduced in
breeding males throughout the study, which the authors suggest that the thyroid disruption may have
contributed to the observed increase in testes weight.
Toxicity to Terrestrial Mammals
The toxicity of HBCD to mammals is characterized in Section 3.2 of this document. In rodents, HBCD
isomers are biotransformed in the liver and are distributed in fat, liver, skeletal muscle and skin. Oral
toxicity studies in rodents show that HBCD exposure can affect thyroid function. HBCD exposure can
result in liver weight, steatosis, hypertrophy and inflammation. Reproductive toxicity in female rats
included decreases in pregnancy, number of litters lost at high exposure dose to F1 dams and decrease
primordial follicles. In male rats, no consistent effects were found relating reproductive effects to HBCD
exposures. HBCD exposure to rats resulted developmental effects including reduced offspring viability,
decreased pup body weight, altered development and skeletal system, and delayed eye opening.
Neurological effects as reported in experimental studies in rats resulted in neurodevelopmental
milestones, locomotor activity and executive function and neurological outcomes related to changes in
auditory sensitivity, dopamine system function, and brain weight. Immune system effects in rats exposed
to HBCD during development also resulted in changes to organ weights.
Toxicity to Terrestrial Vegetation
For terrestrial plants, three acceptable studies reported data on six species. All studies have a high-
quality rating. Phytotoxicity was reported in a 21-day exposure to HBCD to six species of plants (Porch
et al. 2.002). Mean measured test concentrations were 3 1.2, 97.7, 297.1, 764.6, 2,230 and 6,200 mg/kg
dry weight. In one study, three monocots (corn, onion and rye grass) and three dicots (cucumber,
soybean and tomato) were tested. For each species, a control group, and the five treatments were
maintained. Each group consisted of four replicates each containing 10 seeds. During the 21-day test,
weekly observations of seedling emergence and a qualitative assessment of the condition of each
seedling were made. Onions exposed to 276 mg/kg HBCD resulted in significant (p<0.05) reductions in
mortality. There were no signs of treatment-related phytotoxicity observed on seedlings of any species at
any test concentration. In another study, the accumulation and toxicity of a, P, and y-HBCDs in maize
were examined after exposure of 0, 0.002, 0.005, 0.01, 0.02, and 0.05 mg/L (Wu et al. 2012). In another
study, Wu et al. (2016c) investigated the accumulation of HBCD in maize. Young seedlings were
exposed to HBCD at concentrations of 0, 0.002, 0.005, 0.01, 0.02, and 0.05 mg/L. The uptake kinetics
showed that the HBCD concentration reached an apparent equilibrium within 96 hours, and the
accumulation was much higher in roots than in shoots. A reduction in maize root and shoot growth
resulted from an exposure to 0.002 mg HBCD/L.
3.1.3 HBCD Trophic Transfer in the Environment
EPA initially assessed the PBT characteristics of HBCD in accordance with the U.S. EPA TSCA Work
Plan Chemicals: Methods Document (U.S. EPA. 2012d). The potential of HBCD trophic transfer in both
Page 321 of 723
-------
aquatic and terrestrial ecosystems was evaluated in this Risk Evaluation by using the U.S. EPA Final
Water Quality Guidance for Great Lakes System ( ), U.S. EPA Wildlife Exposure Factors
Handbook (' \ s Jb) and European Chemicals Agency (ECHA) Guidance on Information
Requirements and Chemical Safety Assessment (Chapter R. 16: Environmental exposure estimate;
ECHA, 2016). Ingestion rates from the U.S. EPA Wildlife Exposure Factors Handbook (U.S. EPA.
1993b) are used to estimate the exposure of both aquatic and terrestrial predatorial organisms; the same
ingestion rates for aquatic organisms are also used in the U.S. EPA Final Water Quality Guidance for
Great Lakes System ( ). Different methodologies of predicting potential HBCD trophic
transfer were utilized because each method focuses on predators with different feeding habits; organisms
were chosen for each of the methods based on data availability and method-specific requirements.
EPA has assessed the available studies collected in accordance with The Application of Systematic
Review in TSCA Risk Evaluations ( ) relating to the bioaccumulation and
bioconcentration (BAF/BCF) of HBCD. To evaluate HBCD uptake via dietary and media exposure,
different approaches were used to incorporate various sources {i.e., environmental monitoring and
modeled surface water and sediment concentrations) and types of exposure media {i.e., uptake via diet or
environmental media). The calculations used to predict HBCD trophic transfer for both the aquatic
(mink and osprey) and terrestrial (American kestrel) predators are provided in Appendix H.2.
Estimations for HBCD trophic transfer as presented in Table 3-2 were calculated using exposure factors
from the U.S. EPA Wildlife Exposure Factors Handbook (U.S. EPA. 1993b) and HBCD biomonitoring
data.
HBCD bioaccumulation in both aquatic and terrestrial ecosystems has been demonstrated, as detailed
above in Section 2.1.2. Specifically, BAFs and BCFs up to 50,000,000 and 18,100 for HBCD have been
measured for freshwater fish (He et at.. 2013; Veith et a )), and HBCD has been ubiquitously
measured in taxa spanning all trophic levels in aquatic ecosystems (Wu et al. 2011). There is a greater
likelihood of the release of HBCD from the modeled exposure scenarios and respective conditions of
uses into surface water, thus higher trophic level organisms that reside in and primarily consume aquatic
prey have the greatest potential for exposure to and bioaccumulation of HBCD. Despite limited data
regarding HBCD exposure and hazard for terrestrial organisms, the presence of HBCD in the tissue,
eggs {e.g., peregrine falcons and chickens) and milk of {e.g., bobcats) higher biologically-organized
terrestrial organisms suggest the exposure of HBCD through trophic transfer and the likelihood of sex-
specific transfer of HBCD to offspring (Boyles et al. 2017; Tao et al. 2.016; Guerra et al. 2.012).
Mink {Neovison vison) was selected as a model aquatic predator because they primarily consume aquatic
prey, specifically higher trophic level fish. American kestrel {F. sparverius) was selected as a model
terrestrial avian predator because they primarily consume prey that inhabit terrestrial ecosystems.
American kestrel serve as a terrestrial predator avian counterpart to mammals (mink), where the dietary
exposures of HBCD from either only terrestrial and aquatic prey, respectively, can be compared. Mink
was selected to represent a higher trophic level mammal because a majority of their diet is composed of
fish and other aquatic prey. Specifically mink diet consists of 56, 26, 3, and 4% of trout, non-trout fish,
unidentified fish, and crustaceans ( ), respectively. This dietary composition is
comparable to the 90% of mink diet attributed to aquatic prey in trophic level 3 (U.S. EPA. 1995). The
components of American kestrel diet are not as well categorized as that of mink, however approximately
31% consists of small rodents ( 93b). The assessment does not assume that the remaining 10
and 69% of mink and American kestrel diet, respectively, has HBCD, and this is one of the uncertainties
that may underestimate high trophic level organism exposure to HBCD via diet. Despite HBCD being
found predominantly in aquatic media {e.g., sediment), HBCD trophic transfer may result in HBCD
Page 322 of 723
-------
source fluxes between aquatic and terrestrial ecosystems. Specifically, HBCD source movement from
aquatic to terrestrial ecosystems, via trophic transfer, is another area that was briefly explored by
estimating HBCD trophic transfer to a terrestrial mammal (e.g., mink) that primarily consumes aquatic
prey (e.g., trout) ( )). However, despite there being available data on general categories of
prey that mink, American kestrel, and osprey may consume, the dietary uptake of HBCD estimated for
these predators ultimately depends on the availability of both lab and field-acquired HBCD tissue
concentration data.
Given the higher likelihood that HBCD is present in the environment due to its persistent and
bioaccumulative characteristics, chronic exposures are of greater relevance to higher trophic level
organisms. The currently available data on HBCD toxicity to higher trophic level organisms are limited
to a few avian species that do not consume prey from aquatic ecosystems (i.e., Japanese quail and
American kestrel), where the greatest releases of HBCD are expected. Therefore allometric scaling of
the American kestrel reproductive LOEC (70,380 ng/d) was conducted to extrapolate toxicologically
equivalent doses of orally administered HBCD from adult female American kestrels to adult female
ospreys (Fernie et al. 2.011); the methodology is detailed in Appendix H.2. There is uncertainty as to
whether allometric scaling, derived from data on the results of American kestrel chronic exposure to
HBCD, will hold when extrapolating to doses in osprey. This uncertainty arises because of the absence
of quantitative information to characterize the toxicokinetic and toxicodynamic differences between two
avian species with very different lifestages and dietary preferences. No assessment factor was used in
addition to the allometric scaling of the adult female kestrel LOEC of 510 ng/g bw-d to the adult female
osprey LOEC of 40.8 ng/g bw-d (Fernie et al. 2011). or the consumption of 70,380 ng HBCD/d
(calculations available in Appendix H.2). The potential trophic transfer of HBCD from aquatic
ecosystems is more easily estimated than that from terrestrial ecosystems due to the greater amount of
both environmental and biomonitoring information and hazard data for aquatic ecosystems and
organisms, respectively.
The ECHA Guidance on Information Requirements and Chemical Safety Assessment (Chapter R. 16:
Environmental Exposure Estimate) (ECHA. 2.008a) was used to estimate HBCD uptake via fish- and
earthworm-consuming predators. Rainbow trout and earthworm bioconcentration factors (BCF) and
HBCD exposure concentrations in water and soil, respectively, were used to derive Corganism values, as
presented in Table 3-3 The BCF for rainbow trout was used to remain consistent with taxa used in Table
3-2, despite the availability of more conservative BCFs for other fish species (e.g., fathead minnows).
As compared to BAFs, BCFs can often underestimate HBCD uptake because only media exposure
concentrations are accounted for. BCFs are used per methodologies provided in the ECHA Guidance on
Information Requirements and Chemical Safety Assessment (ECHA. 2008a). The body burden of HBCD
in rainbow trout and earthworms, as presented in Table 3-3 does not represent the predicted
environmental concentration in food PECorai, predator for predators of rainbow trout or earthworms,
respectively. Total HBCD BMFs for rainbow trout and earthworms were unavailable, and isomer-
specific HBCD BMFs for rainbow trout were not used to derive PECorai, predator for predators of rainbow
trout because of uncertainties due to processes (i.e., bioisomerization, degradation) that would
significantly impact HBCD isomer uptake and depuration. There is additional uncertainty due to the use
of BCFs that are not normalized to the amount of lipids present in the samples of tissues used for the
referenced studies; there is additional uncertainty in using earthworms and rainbow trout as
representative organisms for their respective trophic levels using this analysis as lipid normalization
generally accounts for species differences (i.e., size, age, seasonal variations in diet, sex).
Page 323 of 723
-------
The above-mentioned methodologies used to estimate HBCD uptake via prey consumption and media
exposure only use available biomonitoring and hazard data. As compared to biomonitoring and
environmental monitoring data, which provides a snapshot of real time information on HBCD
concentrations found in wildlife and various media, these data cannot be specifically attributed to a
condition of use of HBCD that is evaluated in this risk assessment. As described below in Section 4.1, a
two-tiered approach was used to predict HBCD concentrations in various compartments (i.e., surface
water, pore water, sediment) as a result of HBCD releases expected from different model sub-exposure
scenarios of each condition of use. In Section 2.2, the terminology of "surface water" release is used to
describe release scenarios where HBCD is released into surface water without any treatment processes
being used, whereas "POTW" and "WWTP" both indicate the implementation of some type of
wastewater treatment process occurring before the effluent is released into the environment. In Section
3.1 (Environmental Hazards), and 4.1 (Environmental Risk), the terminology of "direct release" will be
used to describe the release of HBCD into surface water without the implementation of a wastewater
treatment process. "Surface water" concentrations reflect modeled surface water concentrations from
HBCD release scenarios regardless of wastewater treatment processes because ultimately all three
release scenarios (direct release, POTW, and WWTP) result in the release of HBCD surface water.
In addition to the HBCD concentrations predicted to be in each of the compartments using the Point
Source Calculator (PSC), HBCD physical chemical properties (i.e., Koc=100,000; logK0w=5.62; Water
solubility=66 |ig/L) were used as input parameters for the Kow (based) Aquatic BioAccumulation Model
(KABAM) version 1.0 ( 39c), which estimates the bioconcentration, bioaccumulation, and
biomagnification of HBCD in aquatic food webs. Specifically, mammal and avian uptake of HBCD
through diet and water intake were estimated and attributed to predicted surface water, pore water, and
sediment concentrations for modeled sub-exposure scenarios 3.3 (Processing: Manufacturing of XPS
Foam using XPS Masterbatch) and 5.7 (Processing: Manufacturing of EPS Foam from Imported EPS
Resin beads). As explained below in Section 4.1, a sensitivity analysis was conducted to evaluate
whether production volume and percent of HBCD removed from facility direct releases would impact
the predicted concentrations of HBCD in various media for three modeled sub-exposure scenarios (two
of which are selected for evaluation for trophic transfer) that have the highest releases of HBCD. The
two model sub-exposure scenarios (3.3 and 5.7) within exposure scenarios 3 and 5 were selected
because between the exposure scenarios that were targeted in the sensitivity analysis, these represent
three types of water treatment of releases from facilities (i.e., direct release, POTW, and WWTP) and
generally have the highest predicted surface water and sediment concentrations. KABAM predictions of
HBCD bioavailability through diet and water are used to categorize exposure and predict body burdens
and the contribution to body burden due to diet. Predicted bioaccumulation, bioconcentration and
biomagnification factors can also be predicted for representative organisms within each trophic level.
American kestrel and Sprague Dawley rats are used as proxy organisms for terrestrial avian and
mammalian wildlife organisms, respectively, that may be exposed to HBCD through trophic transfer and
various media exposure. Specifically, for this model, based on the assumption that the modeled
organisms have the same effect or response to the same effect concentration as those of the proxy
organisms, hazard data on the proxy organisms are also input parameters for KABAM. All KABAM
outputs (predicted body burdens, BAF, BCFs, etc.) are provided in Appendix H.3.
Methods used to estimate HBCD trophic transfer demonstrate HBCD uptake solely via prey ingestion do
not account for media exposure to HBCD, whereas the use of KABAM relates potential BAF, BCFs,
and other indications of trophic transfer to water releases of HBCD that can be tied to a specific COU.
Environmental monitoring data, as presented above, demonstrates the higher likelihood that aquatic
organisms are exposed to greater concentrations of HBCD than terrestrial organisms, especially near
Page 324 of 723
-------
facilities that process waste containing HBCD (Zhu et al. 2017). Furthermore, the data from both
monitoring and modeled predictions suggest that not only can HBCD undergo trophic transfer, but that
organisms that not only reside in aquatic ecosystems, but prey on aquatic organisms, will also be
exposed to HBCD. This suggests that terrestrial organisms living within close proximity to aquatic
ecosystems may be exposed to HBCD through their diet. Different diastereomer profiles may also
depend on diet preferences, where carnivorous fish may have higher ratios of a-HBCD than omnivorous
species (Hlouskova et al. 2013). Although not explicitly addressed in this Risk Evaluation, the potential
for HBCD trophic transfer may also depend on diastereomer-specific uptake, metabolism,
bioaccumulation and excretion; diastereoisomer-specific metabolism and biotransformation may account
for diastereoisomer-specific accumulation observed in higher trophic level organisms (Du et al. 2015).
Finally, HBCD excretion will also determine predator exposure to HBCD through prey consumption;
following an aqueous exposure to 1.8 |ig HBCD/L and a depuration period of 19 days, exposed rainbow
trout were able to eliminate 50% of their HBCD body burden (Drottar and Krueger 2000). The
approaches used below to estimate HBCD trophic transfer do not take excretion into consideration. The
equations used to derive HBCD ingestion in Table 3-2 and Table 3-3 are provided in Appendix H.2.
Table 3-2. Potential Trophic Transfer of HBCD in Aquatic and Terrestrial Ecosystems Using the
U.S. EPA Final Water Quality Guidance for Great Lakes System and U.S. EPA Wildlife Exposure
factors Handbook
Organism's Attribute
Assumption
Reference
Amount of
HBCD
Consumed
per Day
Amount of HBCD
Consumed per day
Normalized to Body
Weight
Deer mouse ingestion
rate (female)
0.45 g food/
g bw-d
Millar, 19791
Deer
Mouse
0.35 (via
fruit) + 200
(via
arthropods)
= 200.4 ng
HBCD/d
Deer Mouse 0.008
mg/kg BW-d
Deer mouse % diet of
fruit in summer
25%
Wolff et al.,
19851
Deer mouse body weight
(female)
24.5 g
Millar and
Innes, 19831
HBCD in fruits
(biomonitoring data: food
basket study in South
Korea)
0.127 ng
HB CD/kg
WW
Barghi (2016)
HBCD in grasshopper
(biomonitoring data: near
electronic-waste
dismantling facilities in
China)
32.4 ng
HBCD/g bw
Zhu (2017)
Deer mouse % diet of
arthropods in summer
56%
Wolff et al.,
19851
American kestrel
ingestion rate
(vertebrates-winter)
0.18 g/g bw-
d
Koplin et al.,
19801
American
kestrel
64.4 ng
HBCD/d
(via Deer
mouse)
American kestrel
0.0005 mg/kg BW-d
(via Deer mouse)
American kestrel % diet
of mammals
31.7%
Meyer and
Balgooyen,
19871
Page 325 of 723
-------
Organism's Attribute
Assumption
Reference
Amount of
HBCD
Consumed
per Day
Amount of HBCD
Consumed per day
Normalized to Body
Weight
American kestrel body
weight (female-winter)
138 g
Gessaman and
Haggas, 19871
Mink ingestion rate
0.16 g/g bw-
d
Bleavins &
Aulerich, 19811
Mink 700.7
ng HBCD/d
(via trout)
Mink 0.0004 mg/kg
BW-d (via trout)
Mink weight
1,734 g
Hornshaw et
al., 19831
Mink % diet of trout
56%
Alexander,
19771
HBCD in trout
4.51 ng
HBCD/g
Tomy (2004)
Osprey body weight
(female)
1,725 g
Pool, 19841
Osprey diet2:
2 - 3,370,200
ng HBCD/d
(listed below)
Osprey diet2:
lxlO"6 -2 mg/kg BW-d
(listed below)
Osprey % diet of fish
100%
Brown
and Amadon,
19681; Poole,
19891
HBCD in Rainbow trout
4.51 ng
HBCD/g
(Tomv et al.
2004)
1,479 ng
HBCD/d
0.001 mg/kg BW-d
HBCD in Northern
Snakehead
(biomonitoring data: food
basket study in South
China)
6.1 pg/g
(Mens et al.
2012)
2 ng HBCD/d
lxlO"6 mg/kg BW-d
HBCD in Brown trout
(biomonitoring data:
downstream of HBCD
manufacturing plant in
the UK)
6,758 ng
HBCD/g
(Allchin and
Morris 2003)
2,216,624 ng
HBCD/d
1.3 mg/kg BW-d
HBCD in Eel
(biomonitoring data:
downstream of HBCD
manufacturing plant in
the UK)
10,275 ng
HBCD/g
(Allchin and
Morris 2003)
3,370,200 ng
HBCD/d
2.0 mg/kg BW-d
1 Exposure factors, as indicated, were derived from the U.S. EPA Wildlife Exposure Factors Handbook. (U.S. EPA 1993b)
2HBCD tissue concentrations for osprey diet as categorized in the U.S. EPA Wildlife Exposure Factors Handbook. (U.S.
EPA 1993b) were not available for those listed suedes, however a ranee of higher troohic level fish soecies were used to
provide a range of potential HBCD uptake via osprey prey ingestion.
Page 326 of 723
-------
Table 3-3. Potential Trophic Transfer of HBCD in Aquatic and Terrestrial Ecosystems using the
ECHA Guidance on Information Requirements and Chemical Safety Assessment (Environmental
Exposure Assessment)
Organism's Attribute
Assumption
Reference
HBCD in Organism
Rainbow trout (whole body BCF)
8,974
Drottar and Kruger
(2000)
HBCD Rainbow trout concentration
(Cfish) =
16.2 mg/kg BW
HBCD exposure concentration to
Rainbow trout
1.8 ng/L
Drottar and Kruger
(2000)
Rainbow trout whole body lipid
percentage
0.083
Drottar and Kruger
(2000)
Lipid normalized HBCD Rainbow
trout concentration (Cfish) =60.1
mg/kg
Rainbow trout (whole body lipid
normalized BCF)
108,120.5
Drottar and Kruger
(2000)
Earthworm bioconcentration factor
(BCF)
4.5
Aufterheide (2003)
Earthworm concentration (Cearthworm)
= 18,855 mg/kg
HBCD exposure concentration to
earthworm
4,190 mg/kg
dry soil
Aufterheide (2003)
As presented in Table 3-2, it is likely for HBCD to undergo trophic transfer in both aquatic and
terrestrial ecosystems, however it is evident that aquatic organisms or predators of aquatic organisms are
more likely to be exposed to HBCD. In regards to the estimation of American kestrel dietary exposure to
HBCD, it is likely the results as presented in Table 3-2 underestimate American kestrel exposure to
HBCD because the primary source of HBCD is coming from the measurement of HBCD from one type
of fruit (watermelon), and grasshoppers; data limitations regarding the availability of more exposure
factors include characterizing the dietary composition of rodents and American kestrels, and measured
HBCD uptake and body burden data for prey organisms. In comparison to mink and osprey, where
100% of their diet can be attributed to higher trophic level fish and a greater availability of HBCD
uptake and body burden data for fish, estimations for American kestrel uptake of HBCD through prey
consumption is limited to less than a third of American kestrel diet (small mammal), as characterized by
the U.S. EPA Wildlife Exposure Factors Handbook. (U.S. EPA 1993b). Furthermore, approximately
20% of rodent uptake of HBCD is not encompassed in the presented estimations. Similarly, mink dietary
exposure to HBCD is also being underestimated because the above calculations only encompass
approximately 56% of mink diet; a majority of mink diet consists of upper trophic level fish, and it is
likely that mink are exposed to more than 700 ng HBCD/d if other higher trophic level fish were used
for the remaining half of mink diet.
EPA did not identify toxicity data for mink or any other higher trophic level terrestrial predator (other
than American kestrel) due to HBCD exposure, thereby making it difficult to determine a hazard
threshold for this terrestrial mammal. Osprey exposure to HBCD via estimations of higher trophic level
fish consumption is comparable to that estimated for mink, assuming 100% trout or higher trophic level
fish consumption. An allometrically-scaled Osprey LOEC of 40.8 ng/g BW-d (Fernie et al. 2011). or the
consumption of 70,380 ng HBCD/d (calculations in Appendix H.2), was significantly surpassed by 3
magnitudes when it was assumed that osprey consumed brown trout or eel (Allchin and Morris 2003).
which may result in reproductive toxicity (smaller clutches); should the reproductive toxicity estimate
for osprey apply, it is likely that their consumption of higher trophic level pelagic or benthic fish will
result in reproductive toxicity and population-level effects. As discussed above, from the presented
estimations, even if the entire American kestrel diet consisted of small mammal consumption, it is
unlikely that American kestrel would be exposed to HBCD concentrations that will result in
Page 327 of 723
-------
reproductive toxicity. Uncertainties due to the use of a representative species for a predator are outlined
below in Section 3.1.7, however, doing the same for prey species may also under- or overestimate
HBCD dietary exposure. Less than 50% of the diets of Deer mouse and American kestrel are being
accounted for and the lipid contents of the fruit (watermelon) and arthropods (grasshopper) used to
estimate the original dietary HBCD exposure of Deer mouse are lower than that of the higher trophic
level fish used to estimate osprey or mink dietary uptake. Data gaps make it difficult to ascertain
whether terrestrial predatory organisms such as American kestrel may be exposed to higher HBCD
concentrations through trophic transfer processes.
The estimated HBCD tissue concentrations and bioaccumulation of organisms in multiple trophic levels
(categorized by KABAM in Appendix G Sections G.3.1 and G.3.2) are based on either the 10th or 50th
percentile predictions for surface and pore water HBCD concentrations associated with exposure
scenario-related releases (3.3 and 5.7). In regard to the measured HBCD tissue concentration predictions
for higher trophic level fish (Table 3-2), these values are more comparable to the KABAM predictions
based on the 50th percentile surface and pore water concentrations. The higher concentrations of HBCD
in fish tissues represent sampling areas downstream of a manufacturing facility and are better
represented by the predicted KABAM values for fish tissue concentrations than the fish sampled in areas
not associated with industrial facilities. Although the 10th percentile KABAM predictions are all greater
than the measured fish tissue concentrations, the measured tissue concentrations are only an indication
of a background tissue concentration, and it is likely that releases from an industrial facility or use will
result in higher HBCD exposure and bioaccumulation. The predicted BAFs, are within the same
magnitude as measured BAFs (both lipid normalized) for upper trophic level fish (He et al. ; Wu et
at.. 201 lY
3.1.4 Weight of the Scientific Evidence
During data integration stage of systematic review, EPA analyzed, synthesized, and integrated the
environmental information for HBCD. This involved weighing scientific evidence for quality and
relevance, using a Weight of the Scientific Evidence (WOE) approach (U.S. EPA. 2018b).
During data evaluation of the relevant HBCD studies, a rating of high, medium, or low for quality based
on the TSCA criteria described in the Application of Systematic Review in TSCA Risk Evaluations was
applied (U.S. EPA. 2018b). While integrating environmental hazard data for HBCD, EPA gave more
weight and consideration to relevant data/information rated high or medium for quality. Only
data/information rated as high, medium, or low for quality was used for the environmental risk
assessment. Any information rated as unacceptable was not used to characterize the hazard of HBCD.
The factors for determining if environmental data/information were relevant, were based on whether the
source had biological, physical/chemical, and environmental relevance (U.S. EPA. 1998):
a. Biological relevance - correspondence among the taxa, life stages, and processes measured or
observed and the assessment endpoint.
b. Physical/chemical relevance - correspondence between the chemical or physical agent tested and
the chemical or physical agent constituting the stressor of concern.
c. Environmental relevance - correspondence between test conditions and conditions in the region
of concern. ( 998)
This WOE approach was used to assess the environmental hazard data of HBCD and develop
concentrations of concern (COCs) for the aquatic compartments (i.e., surface water, sediment) and
Page 328 of 723
-------
environmental concern levels for the terrestrial environment. Where high or medium quality studies
were available for a taxonomic group, low quality studies were not used to derive COCs or
environmental concern levels.
To assess aquatic toxicity from acute exposures, data for three taxonomic groups were available: algae,
aquatic invertebrates {i.e., surface water and sediment dwelling) and fish. For each taxonomic group,
data were available for these species as shown in Table 3-1. To characterize fish toxicity resulting from
acute exposures to HBCD, a 96-hour zebrafish (I). rerio) LOAEL of 0.002 mg/L based on delayed
embryo hatchability was used.
To assess aquatic toxicity from chronic exposures, data for two taxonomic groups were described in the
acceptable literature: fish, and two species for aquatic invertebrates {i.e., the water flea (I). magna), the
black worm {L. variegatus)). Therefore, the endpoints for fish and aquatic invertebrates including
surface water and sediment-dwelling organisms (MATC, NOEC, and an LOEC) were more biologically
relevant, because they measured a toxic effect. Of these values, the most sensitive species were a 21-day
MATC of 0.042 mg/L measuring reproduction in aquatic invertebrates {D. magna) and a 28-day MATC
of 15.7 mg/kg dw measuring worm survival in L. variegatus.
To assess the toxicity of HBCD to algae, data on four acceptable high-quality studies reported data on
three species of freshwater and marine vegetation {i.e., green algae and diatoms). The most sensitive
endpoint reported for marine diatom {Skeletonema costatum) was a 72-hour EC so of 0.010 mg/L from
Walsh et al. (1987). As previously stated, algae data were assessed together with acute and chronic
endpoints regardless of duration and not separated into acute and chronic, because durations normally
considered acute for other species {e.g., 48, 72 hours) can encompass several generations of algae. A
NOEC of 10 |ig/L was reported by Desjardins et al. (2004) for the same species. This study provides
support for the high toxicity of HBCD to algae that was reported in Walsh et al. (1987) by measuring the
exposure concentrations.
To assess terrestrial toxicity from chronic exposures, data for three taxonomic groups were described in
the acceptable literature: terrestrial plants, soil invertebrates and avian species. Therefore, the endpoints
for terrestrial plants, soil invertebrates and avian species (ECso, MATC, LOEC, NOEC, NOAEL,
LOAEL and LOEC) were more biologically relevant, because they measured a toxic effect. Of these
values, the most sensitive species were a 4-day maize {Zeal mays) measuring growth reduction and
reporting a LOAEL of 0.002 mg/L, a 14-day earthworm {Eisenia fetida) reporting a MATC of 200
mg/kg/day measuring reproduction effects and a 21-day LOAEL in American kestrel {F. sparverius)
measuring reproduction reporting a LOAEL of 3.27 ng/g ww.
3.1.5 Concentrations of Concern
The concentrations of concern (COCs) for aquatic species were calculated based on the environmental
hazard data for HBCD, using the weight of evidence approach described above and EPA methods (Suter
, I. r M:> \ .Q13c. 2012c). For HBCD, EPA derived an acute COC, a chronic COC, and an algal
COC. Algae was assessed separately and not incorporated into acute or chronic COCs, because
durations normally considered acute for other species {e.g., up to 96 hours) can encompass several
generations of algae.
After weighing the evidence and selecting the appropriate toxicity values from the integrated data to
calculate an acute and chronic COC, an assessment factor (AF) is applied according to EPA methods
(Suter 2016: U.S. EPA. 2013c. 2012c). The application of AFs provides a lower bound effect level that
would likely encompass more sensitive species not specifically represented by the available
Page 329 of 723
-------
experimental data. AFs also account for differences in inter- and intra-species variability, as well as
laboratory-to-field variability. These AFs are dependent on the availability of datasets that can be used
to characterize relative sensitivities across multiple species within a given taxa or species group.
However, they are often standardized in risk assessments conducted under TSCA, since the data
available for most industrial chemicals are limited. For fish and aquatic invertebrates (e.g., daphnia) the
acute COC values are typically divided by an AF of 5. The acute toxicity value is derived from an
embryo toxicity endpoint, and the AF used to derive the hazard threshold for acute exposure is 5. For
chronic COCs, an AF of 10 is used (U.S. EPA 2013c. 2012c). Environmental concentration levels were
derived for terrestrial organisms. An AF of 10 was also used to derive a COC for algae because the
effects measured (reproduction and growth) are generally considered to be associated with a longer time
frame or chronic exposure for this taxa.
Table 3-4. Concentrations of Concern (COCs) for Aquatic Toxicity
Environmental Toxicity
Effects
Hazard
Value
Assessment
Factor
Concentration
of Concern
(COC)
Reference
Score
Acute toxicity
Zebrafish (1). rerio)
96-hr LOAEL
Delay Hatching
2 (ig/L
5
0.4 (ig/L
(Hu et al.. 2009a)
High
Chronic toxicity
Water flea (I). magna)
21-d MATC (surface
water)
Reduced length
of surviving
young
4.2 (ig/L
10
0.417 (ig/L
(Drottar and
Krueeer 1998)
High
California blackworm
(L. variegatus)
28-day MATC (sediment)
Reduction in
worm number
15,700
|ig/kg dw
10
1,570 ng/kg/dw
(Oetken et al.
2001)
High
Algae
Marine diatom (S. costatum)
72-hr ECso
Growth Rate
10 ng/L
10
1 f-ig/L
(Walsh et al.
1987)
High
To calculate the acute COC of 0.4 |ig/L, the acute value from the zebrafish (IX rerio) 96-hour LOAEL
of 2 |ig/L was divided by an AF of 5.
In regards to calculating a chronic COC, the aquatic invertebrate (D. magna) 21-day MATC chronic
value of 4.2 |ig/L was divided by an AF of 10, per established EPA methods (U.S. EPA 2013c. 2012c).
resulting in a chronic COC of 0.4 |ig/L. Similarly, the algae 72-hr ECso of 10 |ig/L was divided by an
AF of 10, resulting in a COC of 1 |ig/L.
A chronic COC of 1,570 |ig/kg dw, based on the benthic organism, L. variegates, was derived from the
28-day MATC of 15,700 |ig/kg dw, which was divided by an AF of 10, per established EPA methods
(U.S. EPA 2013c. 2012c).
Table 3-5. Terrestrial Effect Concentrations (Hazard) used to Evaluate Toxicity to Terrestrial
Organisms
Environmental
Effects
Hazard
Reference
Score
Toxicity
Value
Maize
Growth (root and shoot)
2 f-ig/L
(Wu et al. 2016c)
High
Page 330 of 723
-------
4-d LOAEL
Earthworm
56-day MATC
Reproduction/ Mortality
173,000 ng/kg dwt
(Aufderheide et al.
20031
High
American kestrel
21-d LOAEL
Reproduction (clutch size, egg production
timing)
0.51 mg/kg bw
(Fertile et al. 2011)
High
Rat
2-generation NOAEL
Thyroid hormones response, Reproduction
10 mg/kg bw
(Etna et al, 2008)
High
Studies where terrestrial organisms were exposed to HBCD were evaluated and those with high data
evaluation scores (using either environmental hazard Systematic Review metrics) and relevant
environmental exposure pathways were used to assess risk to terrestrial organisms. The organisms
identified in the abovementioned studies in Table 3-5 were chosen to represent their respective taxa
classifications {i.e., vegetation, invertebrate, vertebrate). To evaluate HBCD hazard thresholds for
terrestrial wildlife, environmental hazard levels were used as reported by study authors because there
was not enough information available to derive assessment factors.
3.1.6 Summary of Environmental Hazard
HBCD presents a significant concern for adverse effects on the environment. This conclusion is based
on the observed potential for bioaccumulation, trophic transfer, altered reproductive behavior, as well as
toxicity due to both acute and chronic HBCD exposure. Bioconcentration factors (BCFs) and
biomagnification factors (BMFs) as high as 18,100 and 29.7, respectively, have been observed in fish
(Zhang et al. 2014b; Du et al. 2.012a; Law et al. 2006). BMF values of 26 (lipid-weight) and 1.6-3 have
also been observed in birds (Haukas et al. 2010b) and mammals (Shaw et al. ) respectively.
Observed toxicity values as low as 0.009 mg/L for a 72-hour ECso (reduced growth in the marine
diatom, S.costatum) fWalsh et al. 1987), and 0.0042 mg/L (MATC for reduced size (length) of surviving
young in 1). magna (Drottar and Krueger 1998)). indicate high aquatic toxicity due to acute and chronic
HBCD exposure.
Reduced chick survival in Japanese quail {C. japonica) fed a 15 ppm HBCD diet (2.1 mg/kg bw-day)
(MOEJ 2009) and altered reproductive behavior (reduced courtship and brood-rearing activity) and
reduced egg size in American kestrels (/•'. sparverius) fed 0.51 mg/kg bw-day (Marteinson et al. 2.012;
Fertile et al. .'01 t; Marteinson et al. 2.011; Marteinson et al. 2010) indicate high toxicity for terrestrial
organisms as well.
Assessment of HBCD aquatic toxicity is complicated by the low water solubility of the chemical and
differences in the solubility of the three main HBCD isomers, which makes testing difficult and
interpretation uncertain for studies conducted above the water solubility. Studies conducted at
concentrations above the water solubility of HBCD are essentially testing the effects at the maximum
HBCD concentration possible. In contrast with the studies cited above, other acute and chronic aquatic
toxicity studies conducted using methods, test species, and endpoints recommended by the EPA reported
no effects at saturation or near the limit of water solubility. However, water solubility is not considered a
limiting factor for hazard determination for aquatic species since there are studies showing adverse
effects at or below the water solubility of HBCD. In addition, the potential for HBCD to bioaccumulate,
bio-magnify, and persist in the environment, significantly increases concerns regarding HBCD exposure
for aquatic organisms.
Page 331 of 723
-------
A wide range of effects of HBCD have been reported in fish (e.g., developmental toxicity, embryo
malformations, reduced hatching success, reduced growth, hepatic enzyme and biomarker effects,
thyroid effects, DNA damage to erythrocytes, and oxidative damage) and invertebrates (e.g.,
degenerative changes, morphological abnormalities, decreased hatching success, and altered enzyme
activity) in supporting studies that assessed endpoints beyond those evaluated in this assessment (Du et
it .iM . il mgn it _*¦ m . ^ t«ekema et at 2014; Hong et ol 201 !; /.bang et at 2014a; Wu et al. 2013;
Du et jL A' L-si, Anselmo et al. 2011; Palace et al. 2010; Deng et al. 2009; Hu et al. 2009a; Smolarz and
Berger 2009; Aniagu et at 2008; Palace et al. 2008; Zhang et al. 2.008; Ronisz et al. 2004). Effects on
the thyroid in fish (reduced thyroid hormone (triiodothyronine, T3, and thyroxine, T4) in rainbow trout
(Palace et al. 2010; Palace et al. 2008; Kuiper et al. 2007; Lower and Moore 2007) are similar to those
observed in mammals. These studies were also evaluated using metrics and the rating criteria described
in the Application of Systematic Review in TSCA Risk Evaluations document ( 018b).
COCs derived for aquatic organisms are summarized in Table 3-2. EPA calculated the chronic COC for
HBCD based on two high quality studies at 4.2 ppb and 157 |ig/kg dw, based on an MATC for I),
magna and L. variegatus, respectively.
Also, the terrestrial effect concentrations derived for terrestrial organisms are summarized in Table 3-3.
EPA calculated the environmental concern levels for terrestrial receptors for HBCD based on three
acceptable studies at 2 |ig/L and 173 |ig/kg dw and 10 |ig//kg bw based on a LOAEL for maize, a
MATC for earthworms, and aNOAEL for rats, respectively.
As stated previously, algae were assessed separately from other aquatic organisms, because durations
normally considered acute for other species (e.g., 48, 72 hours) can encompass several generations of
algae. EPA calculated an algal COC for HBCD at 1 |ig/L, based on a geometric mean of a LOEC and
NOEC for growth in the marine diatom (S. cos latum) from Walsh et al. (1987). a study rated high for
quality.
3.1.7 Assumptions and Key Sources of Uncertainty for the Environmental Hazard
Assessment
After evaluating all available environmental hazard data on HBCD, EPA has high confidence in the
environmental hazard data used to assess the environmental hazard of HBCD and high confidence that
the data incorporates environmentally-protective acute and chronic concentrations of concern (as
described above). Despite the high confidence in the data used to assess the environmental hazard of
HBCD, there are sources of uncertainty regarding the extrapolation of available data and methods used
to select a representative species and taxa that are addressed below.
In characterizing the environmental hazard of HBCD, some uncertainty in the analysis of environmental
exposure is due to the inherent nature that the proportion of diastereomers in HBCD mixtures will differ
based on commercial and consumer products used, and the changes of such proportions that may occur
following environmental release. Similarly, the environmental hazard of HBCD will depend on the
exposure to varying proportions and concentrations of HBCD diastereomers; most studies reported
exposure and effects concentration in total HBCD, however studies that concentrated on
bioisomerization generally parsed out exposure based on individual diastereomer. The sole use of
HBCD diastereomer-specific partitioning and toxicity data may result in the underestimation of overall
HBCD environmental hazard because diastereomer proportions will continue to change in the
environment.
Page 332 of 723
-------
For evaluating the potential trophic transfer of HBCD in the environment, many assumptions and
uncertainties were taken into consideration due to the complexity of food web dynamics. In general,
there is an inherent uncertainty when using proxy organisms to represent all terrestrial and aquatic prey
and predators; the selection was based on data availability, thus making it difficult to represent more
than three levels of prey-predator relationships. Organism selection for this evaluation was exclusively
from the available exposure factors in the U.S. EPA Wildlife Exposure Factors Handbook (also
incorporated in the U.S. EPA Final Water Quality Guidance for Great Lakes System. Variations in diet
categories due to life stage, gender, and seasonal differences are not addressed in this evaluation because
the specificity and calculation of each exposure factor are based on the methodologies used in their
respective original references cited by the U.S. EPA Final Water Quality Guidance for Great Lakes
System and U.S. EPA Wildlife Exposure Factors Handbook. Further, the inability to account for
complete diets and the potential variations in diet may have resulted in the under- or overestimation of
HBCD uptake. Specifically, in regard to mink diet, HBCD uptake calculations using methodologies
from the U.S. EPA Final Water Quality Guidance for Great Lakes System and U.S. EPA Wildlife
Exposure Factors Handbook., and trout HBCD biomonitoring data could only account for 56% of mink
diet; an additional 26% and 18% of their diet was labeled "non-trout" fish, and miscellaneous items,
respectively. Like the other organisms used to calculate potential HBCD uptake via ingestion, large
portions of mink diet are unaccounted for due to a lack of reasonably available information on either the
diet composition, or HBCD body burden in prey organisms. Further underestimations of HBCD uptake
by terrestrial predators, as compared to aquatic predators in this assessment (i.e., calculated by
evaluating Kestrel ingestion of mice) may also be due to the use of fruit and grasshopper HBCD
biomonitoring data as the original source of HBCD for kestrel, as opposed to smaller mammals with a
higher body fat composition. The limited data regarding HBCD in terrestrial organisms contributes to
the uncertainty regarding HBCD trophic transfer in terrestrial food webs. Additionally, HBCD trophic
transfer was not quantified or evaluated for every level of biological organization because biomonitoring
data were available for many lower trophic level organisms. The uncertainties regarding the ingestion of
HBCD also do not take into consideration physiological processes that impact the absorption,
metabolism, distribution and elimination of HBCD, once ingested. The available literature regarding
how HBCD is absorbed, metabolized, distributed and eliminated are largely evaluations of the
bioisomerization of HBCD once ingested.
HBCD has physical-chemical properties that are within the model domains of KABAM (vl), which
allows for the prediction of potential trophic transfer of a chemical within a freshwater aquatic
ecosystem. KABAM (vl) provides an opportunity to model potential HBCD bioaccumulation,
bioconcentration, and trophic transfer due to predicted releases for individual sub-scenarios within a
specific COU (using PSC-VMWS), which thereby correspond with risk estimates calculated for pelagic
and benthic organisms. However, there are limitations involved with the extrapolation of the model
outputs, one of which being that the amount of HBCD predicted to undergo trophic transfer, is predicted
for trophic levels and not specific species. Further, the default model ecosystem for KABAM is a
freshwater pond that receives pesticides in both runoff and spray drift from an adjacent 10-ha treatment
field; HBCD is not a pesticide, thus the introduction of HBCD to the model freshwater pond may not be
representative of the exposure scenarios used to assess environmental risk.
The analysis focuses on HBCD uptake via prey ingestion as an indicator for potential HBCD trophic
transfer in aquatic and terrestrial food webs, and does not take into consideration the uncertainties
regarding the physiological processes that impact the absorption, metabolism, distribution and
elimination of HBCD, once ingested. Specifically, the available literature primarily focuses on HBCD
diastereomer-specific body burdens as a function of the potential bioisomerization of a-, P-, and y-
Page 333 of 723
-------
HBCD. However, as there is no consensus on the uptake, biotransformation, and elimination of HBCD
diastereomers once ingested, it is difficult to ascertain whether HBCD diastereomer-specific uptake and
exposure is a function of environmental concentrations and/or bioisomerization of HBCD once ingested.
Similar to polybrominated diphenyl ethers, where different congeners are found differentially in aquatic
or terrestrial organisms, potentially resulting in different dietary exposures, there is also speculation on
whether aquatic or terrestrial ecosystem conditions differentially result in diastereoisomer-specific
isomerization and degradation of HBCD (Potter et al. 2009).
As mentioned in Appendix C.2, a-HBCD bioaccumulate and biomagnifies to a greater extent than either
P- and y- diastereomers in aquatic food webs, despite y-HBCD being the isomer primarily found in
commercial mixtures. Furthermore, the bioisomerization of y-HBCD to a-HBCD in fish (Du et al..
2012a) and the higher water solubility of a-HBCD (as compared to the other diastereomers) suggest that
regardless of the percentages of diastereomers in commercial mixtures, once released into the
environment, there is a higher likelihood of organisms being exposed to a-HBCD. Diastereomer-specific
excretion will also influence whether higher trophic level predators will be exposed to HBCD via prey
ingestion. In rats that were orally exposed to all three HBCD diastereomers, HBCD diastereomer
excretion through both feces and urine was greater for P- and y- diastereomers, than a-HBCD (Hakk
2016). Species-specific differences in physiological processes will also greatly impact predator-specific
uptake of HBCD. Prey habitat and diet (e.g., types of organic matter) may also impact gut microbiome
composition and physiological ability to ingest, metabolize and store bioaccumulative chemicals, such
as HBCD. Due to the higher lipid and protein found in the earthworm, E.fetida, as compared toM
guillelmi, as well as differences in in HBCD uptake, depuration, metabolism and isomerization, the biota
soil accumulation factor for HBCD was higher in E. fetida. Furthermore, the bioisomerization of P- and
y-HBCD to a-HBCD was observed to a greater extent in E.fetida than inM guillelmi. In addition to
having a longer half-life than P- and y-HBCD, a-HBCD also bioaccumulated to a greater extent than the
other two diastereomers in earthworms exposed to soil samples individually containing HBCD
diastereomers (Li et al. 2016). In general, evaluating the trophic transfer of HBCD using any method
will not be able to account for all sources of physiological differences (i.e., age, gender, and seasonal
impacts on prey availability) that will ultimately affect HBCD exposure and bioavailability.
Finally, the AFs used to derive concentrations of concern do not take into account organisms being
exposed to HBCD via multiple pathways (i.e., media, dietary), or the other uncertainties discussed
above. Unfortunately, there is insufficient information available on the impact of organism sensitivity
resulting from either different or simultaneous exposure pathways to HBCD.
3.2 Human Health Hazards
3.2,1 Approach and Methodology
EPA used the approach described in Section 1.5 to evaluate, extract and integrate HBCD's human health
hazard and dose-response information.
Page 334 of 723
-------
Data
Summaries for
Adverse
Endpoints
(Supplemental
Human Health
Document)
Risk Characterization
Human Health Hazard Assessment
Data Evaluation
After full-text screening,
apply pre-determined data
quality evaluation criteria
to assess the confidence of
key and supporting studies
identified from previous
assessments as well as
new studies not
considered in the previous
assessments
• Uncertainty and variability
• Data quality
• PESS
• Alternative interpretations
Risk Characterization
Analysis
Determine the qualitative
and/or quantitative human
health risks and include, as
appropriate, a discussion of:
Data Integration
Integrate hazard information by considering quality (i.e.5
strengths, limitations), consistency, relevancy, coherence and
biological plausibility
Hazard ID
Confirm potential
hazards identified
during
scoping/problem
formulation and
identify new hazards
from new literature (if
applicable)
Dose-Response
Analysis
Benchmark dose
modeling for
endpoints with
adequate data;
Selection of PODs
Output of
Systematic
Review
Stage
WOE
Narrative by
Adverse
Endpoint
(Section 3.2.4)
Summary of
Results and
PODs
(Section 3.2.5)
Risk Estimates
and
Uncertainties
(Section 4.2)
Figure 3-1. EPA Approach to Hazard Identification, Data Integration, and Dose-Response
Analysis for HBCD
Specifically, EPA reviewed key and supporting information from previous hazard assessments as well as
the existing body of knowledge on HBCD's human health hazards. These data sources17 included the
TRI Technical Review of HBCD (U.S. EPA. 2016e). the TSCA Work Plan Problem Formulation and
Initial Assessment, (U.S. EPA 2015 a). Preliminary Materials for the IRIS Toxicological Review of
HBCD (U.S. EPA 2014f) as well as other publications (U.S. EPA 2016e. 2014d: NICNAS 2012a:
EC/HC 2011: EINECS 2008: U.S. EPA 2008a: OECD2007). Additional scientific support from the
Office of Research and Development subsequent to these publications also contributed to this human
health hazard assessment.
All non-cancer health hazards of HBCD previously identified in these reviews were described and
reviewed in this Risk Evaluation, including: acute toxicity, liver toxicity, thyroid effects, reproductive/
developmental toxicity, neurotoxicity, immunotoxicity, sensitization and irritation. EPA relied heavily
on the aforementioned existing reviews along with scientific support from the Office of Research and
Development in preparing this Risk Evaluation. Development of the HBCD hazard and dose-response
assessments considered EPA and National Research Council (NRC) risk assessment guidance.
The new literature was screened against inclusion criteria in the PECO statement and the relevant
studies (e.g., useful for dose-response)18 were further evaluated using the data quality criteria for human,
animal, and in vitro studies described in the Application of Systematic Review in TSCA Risk Evaluations
(U.S. EPA 2018b) (see Section 1.5). EPA skipped the PECO screening step of the key and supporting
studies and entered them directly into the data quality evaluation step based on their previously
identified relevance to the Risk Evaluation.
17 HBCD does not have an existing EPA IRIS Assessment.
18 Some of the studies that were excluded based on the PECO statement were considered later during the systematic review
process as needed. For example, EPA reviewed mode of action information to qualitatively support the health hazard
assessment.
Page 335 of 723
-------
EPA considered studies of low, medium, or high confidence for hazard identification and dose-response
analysis. Information from studies that were rated unacceptable were only discussed on a case-by-case
basis for hazard ID and weight-of-evidence assessment but were not considered for dose-response
analysis. EPA considered the specific reasons for the unacceptable scoring in determining whether
unacceptable studies could remain useful for hazard ID or weight-of-evidence.
EPA has not developed data quality criteria for all types of hazard information. This is the case for
toxicokinetics and many types of mechanistic data which EPA typically uses for qualitative support
when synthesizing evidence. As appropriate, EPA evaluated and summarized these data to determine
their utility for supporting the Risk Evaluation (e.g., ADME data).
Following the data quality evaluation, EPA integrated the toxicological information from each relevant
study. In the last step, the strengths and limitations of the data were evaluated for each endpoint and a
weight-of-the-scientific evidence narrative was developed. Data for each selected hazard endpoint was
modeled to determine the dose-response relationship (Appendix I). Finally, the results were
summarized, and the uncertainties were presented. The process is described in Figure 3-1.
The weight of scientific evidence (WOE) analysis included integrating information from toxicokinetics
and toxicodynamics in relation to the key hazard endpoints: acute toxicity, liver toxicity, thyroid effects,
reproductive/ developmental toxicity, neurotoxicity, immunotoxicity, sensitization and irritation. EPA
considered both data quality and relevance in selecting human health studies to move forward for dose-
response analysis in order to quantitatively assess each key hazard endpoint. EPA also considered
supportive data on mode of action (MOA) for these endpoints in evaluating the WOE for each
endpoints.
Dose-response analyses using benchmark dose modeling (BMD) was performed for each hazard
endpoint of concern where possible. In an effort to address some of the limitations of the
NOAEL/LOAEL approach, the BMD approach was developed as a more robust alternative that
considers all the data in the dose-response relationship ( ). A summary table which
includes all endpoints considered for this assessment, the no-observed- or lowest-observed-adverse-
effect levels (NOAEL and LOAEL) for non-cancer health endpoints by target organ/system, the
incidence for cancer endpoints, and the results of the data quality evaluation is provided in Risk
Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Systematic Review Supplemental File: Data
Quality Evaluation of Human Health Hazard Studies. (U.S. EPA. 2019n).
EPA considered points of departure (POD) from studies that were PECO relevant, scored acceptable in
the data quality evaluation, and contained adequate dose-response information. It is used as the starting
point for subsequent dose-response (or concentration-response) extrapolations and analyses. PODs can
be a no-observed-adverse-effect level (NOAEL), a lowest-observed-adverse-effect level (LOAEL) for
an observed incidence, or change in level of response, or the 95% lower confidence limit of the
benchmark dose (BMDL)19. PODs were adjusted as appropriate to conform to the specific exposure
scenarios evaluated.
The only available repeat-dose toxicity studies available on HBCD were conducted via the oral route of
exposure (except for a single 14-day inhalation study (Song et al.: )). These studies were evaluated
19 The benchmark dose (BMD) is a dose or concentration that produces a predetermined change in response range or rate
of an adverse effect (called the benchmark response or BMR) compared to baseline.
Page 336 of 723
-------
for dose-response assessment, and oral PODs were extrapolated for use via the inhalation route because
it is assumed that inhaled HBCD will be absorbed either through the lungs or via the GI tract following
incidental ingestion. Limited toxicological data are available by the dermal route and physiologically
based pharmacokinetic/ pharmacodynamic (PBPK/PD) models that would facilitate route-to-route
extrapolation have not been identified for HBCD. Therefore, oral PODs were also extrapolated for use
via the dermal route, with adjustments made for absorption. The PODs estimated based on effects in
adult animals were converted to Human Equivalent Doses (HEDs) employing a standard dosimetric
adjustment factor (DAF) consistent with EPA guidance ( 11c).
Section 3.2.5 describes the dose-response assessment guiding the selection of PODs for non-cancer
endpoints. The benchmark dose analysis is discussed in Appendix I, and th q Risk Evaluation for Cyclic
Aliphatic Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard. (U.S. EPA.
2.019e).
3.2,2 Toxicokinetics
3.2.2.1 ADME
3.2.2.1.1 Absorption
Absorption in the human gastrointestinal (GI) tract is expected given the detection of
hexabromocyclododecane (HBCD) in samples of human milk, maternal blood/cord blood, or fetal
tissue, and in food samples collected in several regions of the world (Rawn et al. 2014b; Rawn et al.
2.014a: NICNAS 2012a: EC/HC 20111
HBCD isomers were rapidly and extensively absorbed in the GI tracts of mice given single oral doses of
y-[14C]-HBCD (Szabo etal. 20101 a [14C] HBCD (Szabo et al. 2.01 la). or P-HBCD (Sanders et al.
2013) and rats given single oral doses of [14C]- y-HBCD (mixed with technical-grade HBCD containing
-75% y-HBCD) (Yu and Atallah 1980). For example, the rat study indicated nearly complete
absorption; after 72 hours, 72% of the administered radioactivity was detected in feces (as nonidentified
metabolites), 16% in urine, and 17% in tissues excluding the GI tract (Yu and Atallah 1980). In studies
of mice, absorption percentages between 85 and 90% were reported, based on tissue levels and
cumulative fecal and urinary excretion of radioactivity (Sanders et al. ":0L'»: Szabo et al. 201 la. 2010).
The dermal absorption of HBCD has also been investigated in a few studies. Various ex vivo and in
vitro skin models demonstrate that —30-50% of dermally exposed HBCD will partition into skin tissue
(Pawar et al. 2016: Abdallah et al. 2015). The absorption of HBCD is influenced by both the
composition of skin and the relative isomeric mixture of HBCD. HBCD is preferentially absorbed into
sebum compared to sweat, and absorption increases from y-HBCD < P-HBCD < a-HBCD. Substantially
less HBCD penetrates through skin for systemic absorption. One study (Roper et al. 2007) estimated less
than 0.1%) systemic absorption of HBCD dissolved in acetone, with 35% delivered into the skin and only
1.35%) remaining in the skin following washing and drying. Data from skin models suggests that 4.95 -
6.46%) of a-HBCD dissolved in acetone is absorbed, with other isomers permeating even less (Abdallah
el )•
For the purposes of this risk evaluation, an upper-end estimate of 100%> gastrointestinal absorption will
be used. It is assumed that any inhaled HBCD particles will be either absorbed through the lungs or
swallowed and absorbed through the GI tract, although GI absorption is expected to predominate
because the majority of particles are likely too large to reach the deep lung (further explained in Section
4.2.1). Based on available ex vivo and in vitro data, the highest-end estimate of 6.5%> dermal absorption
Page 337 of 723
-------
of HBCD is used as a conservative health-protective assumption. Comparison of this upper end estimate
for fraction absorbed with a calculation for flux/permeability produces very similar results (see
Appendix L for full discussion), justifying use of the fraction absorbed method for risk estimation. The
actual percentage of HBCD absorbed dermally is variable based on multiple factors including the
relative percentage of each isomer in the mixture, particle size, the bioavailability of HBCD when
entrained within foam or other particles, the presence of any potential organic solvent or non-aqueous
media for HBCD particles, and the relative ratio of sweat to sebum on skin.
3.2.2.1.2 Distribution
Numerous studies of HBCD concentrations in samples of human milk, blood, fatty tissues, or fetal
tissues have noted that a-HBCD is the predominant isomer detected, even though y-HBCD is the
predominant isomer in commercial HBCD products (Rawn et al. 2014b; Rawn et al. 2014a; NICNAS
2012a; EC/HC ^ ). These results indicate preferential tissue accumulation (especially in fat) of a-
HBCD, compared with y-HBCD or P-HBCD. In these studies, measurements of HBCD in maternal
serum and umbilical cord serum of pregnant women have demonstrated that HBCD can cross the
placenta and enter the fetal circulatory system.
In rats and mice, radioactivity from oral or intravenous (i.v.) administered [14CJ-HBCD distributes
widely in the body, with the highest levels in fat, liver, skeletal muscle, and skin (Sanders et al. 2013;
Szabo et al. JO I ib; S ubo et al. 2010; Yu and Atallah 1980). For example, 8 hours after administration
of a single oral dose of [14C]- y-HBCD (mixed with technical-grade HBCD) in female rats, radioactivity
was detected in the fat (20% of administered dose), muscle (14%), and liver (7%) with smaller amounts
(<1%) in the blood, heart, lung, gonads, uterus, spleen, kidney, and brain (Yu and Atallah 1980). A
similar relative distribution pattern was observed in male rats, except that the levels of radioactivity
(expressed as a percentage of administered dose) in fat and muscle of males were lower (about one-half
to three-quarters of the levels in females). Radioactivity in most tissues decreased over the course of 72
hours, but remained elevated in the fat. Nonpolar metabolites of HBCD accounted for all of the
radioactivity in fat; isomeric composition in the fat was not determined.
The three HBCD isomers exhibit differential accumulation in mice exposed by gavage (Sanders et al.
2013; Szabo et al. 201 I h, Szabo et al. 2010). At 1-3 hours after single radiolabeled doses of 3 mg/kg of
each isomer were given, concentrations of HBCD-derived radioactivity were highest in the liver,
followed by the adrenals, kidneys, and bladder (after exposure to y-HBCD); fat, kidneys, and lung (after
exposure to P-HBCD); or blood, kidney, and brain (after exposure to a HBCD). Tissue concentrations
were markedly higher after exposure to a-HBCD {e.g., peak of 47,628 ng/g liver) than after exposure to
the other isomers (peaks of 4,462 ng/g liver for P-HBCD and 2,309 ng/g liver for y-HBCD). Tissue
concentrations peaked 3-8 hours after exposure to either p or y-HBCD, and declined steadily thereafter.
In contrast, after exposure to a-HBCD, concentrations in the skin, muscle, and adipose tissue peaked
1-2 days later, indicating redistribution and accumulation of radioactivity in these tissues. Four days
after exposure to each isomer, concentrations were markedly decreased in all tissues; at that time, the
highest tissue concentrations were in the fat after exposure to P- and a HBCD (13,320 and 498 ng/g,
respectively), and in the adrenal glands after exposure to y-HBCD (492 ng/g) (Sanders et al. 2013;
Szabo et al. 201 lb; Szabo et al. 2010). The results indicate greater deposition of a-HBCD or its
metabolites in most tissues, especially fat, compared with y-HBCD and P-HBCD. Similar findings were
reported by (WIL Research 2001) based on data from fat tissue samples collected from rats exposed to
technical-grade HBCD for 90 days at a gavage dose of 1,000 mg/kg-day; P and y-HBCD tissue
concentrations were only 8-18% of the concentration of a-HBCD.
Page 338 of 723
-------
Sex-dependent differences in distribution were observed in rats exposed by gavage for 28 days to
commercial HBCD at doses from 0.3 to 200 mg/kg-day (van der Yen et al. 2006). Concentrations of
total HBCD were higher (on average 5-fold higher) in livers of female than male rats over the entire
dose range. Fat tissue from female rats contained HBCD concentrations approximately 4.5-fold higher
than those measured in male fat tissue (based on data from two rats/sex in the 10 mg/kg-day dose
group). Findings from the 90-day rat study by fWIL Research 2001) showed a smaller sex-dependent
difference in fat tissue concentrations. In rats exposed by gavage at a dose of 1,000 mg/kg-day, the mean
a-HBCD concentrations in fat tissues was 40% greater in female rats than males at exposure day 89; the
mean concentrations of P- and y-HBCD in fat tissues in males and females were similar. Based on same
collections on days 2, 6, 13, 20, 27, 55, 89, 104, and 118 of the study, the patterns of distribution into fat
tissues in males and females were similar.
3.2.2.1.3 Metabolism
Studies in laboratory animals and in vitro studies show that HBCD isomers can undergo
stereoisomerization, hydroxylation, and debromination, and that y-HBCD and P-HBCD are more rapidly
and extensively metabolized than a-HBCD. The results also indicate that cytochrome P450 (CYP450)
enzymes are involved in metabolism of HBCD, but the predominant metabolic pathways and terminal
excretory metabolites have not been fully characterized. Debrominated metabolites of HBCD have been
detected in human breast milk samples, suggesting that debromination steps inferred from metabolites
identified in laboratory animals are applicable to humans (Abdallah and Harrad 2011).
In vivo stereoisomerization of the y- to the a-isomer has been demonstrated in toxicity studies of rats,
and available data suggest that stereoisomerization is more important at higher doses. Dose-dependent
stereoisomerization was observed in rats repeatedly exposed to commercial HBCD (with composition
10% a, 9% P, and 81% y) by gavage (van der Yen et al. 2006; esearch 2001) or dietary
administration (van der Yen et al. 2009). In these studies, the ratios of the lipid-normalized
concentrations of y-isomer to the a-isomer (measured as parent compound using liquid
chromatography/mass spectrometry [LC/MS]) in liver differed from the ratios in the administered
material, and these ratios declined with increasing dose. For example, in adult rats exposed for 28 days
(van der Yen et al. 2006). the ratios of the y-isomer to the a-isomer (P-HBCD comprised <1.5% of the
total HBCD in tissues) in females ranged from 4.2 at the low dose (0.3 mg/kg-day) to 0.4 at the high
dose (200 mg/kg-day); in males, at the same doses, the ratios ranged from 2.3 at the low dose to 0.9 at
the high dose. These values were all lower than the ratio of 8.1 in the administered material. This dose-
dependent shift in the ratio of y:a isomers was also observed in 11-week-old offspring of rats exposed
before and during mating and during gestation and lactation (van, der Yen et al. 2009).
Analysis of excreta and tissues following oral administration of [14C]-HBCD to rats (Yu and Atallah
1980) showed extensive metabolism of y-HBCD. None of the radioactivity recovered in urine or feces
could be identified as parent y-HBCD following oral administration of [14C]-y-HBCD (mixed with
technical-grade HBCD containing -75% y-HBCD). Several polar metabolites of uncharacterized
structure were found in extracts of feces and urine; these metabolites constituted 88% of the cumulative
radioactivity excreted during the 72 hours after dosing (Yu and Atallah 1980).
Results of oral exposure studies in mice given the same dose of each isomer demonstrated more
extensive metabolism of P- and y-HBCD compared with a-HBCD (Sanders et al. 2013; Szabo et al.
201 la. 2010). For example, more radioactivity was excreted in the urine after oral dosing with P-HBCD
(-45% of administered dose over 4 days) than after the same dose of either a- or y HBCD (-20-28% of
administered dose). The urine contained only metabolites; none of the radioactivity in the urine was
associated with the parent isomers (Sanders et al. 2013; Szabo et al. 201 la. 2010). Extraction of feces
Page 339 of 723
-------
samples for thin layer chromatography analysis of radioactivity showed that a significant proportion of
fecal radioactivity was not extractable after exposure to a HBCD (64%) or y-HBCD (52%), while a
lower proportion was not extractable after exposure to P HBCD (30%). (Szafao et at 2010) hypothesized
that nonextractable radioactivity in feces represented remnants from reactive metabolites covalently
bound to proteins or lipids. Of the extractable radioactivity in feces, polar metabolites comprised the
largest percentage of extractable fecal radioactivity after dosing with y HBCD (85%); polar metabolites
comprised smaller percentages after dosing with a-HBCD (66%) or P-HBCD (39%). After exposure to
P- and y-HBCD, but not a HBCD, isomerization products were detected in feces. Total extractable fecal
radioactivity contained 4% P-HBCD and 7% a-HBCD after exposure to y-HBCD, and 16% y-HBCD
after exposure to P-HBCD. No isomerization of a-HBCD was evident in any of the matrices examined.
Data on the excretion of parent compound provide the strongest evidence for greater metabolism of P-
and y HBCD compared with a-HBCD: a larger percentage of extractable fecal radioactivity was
associated with parent compound after administration of a-HBCD (34%) than after dosing with P HBCD
(14%>) or y-HBCD (4%). Given that oral absorption of all three isomers was similar (85—90%), the
differences in excreted parent compound appear to reflect greater metabolism of the P- and y-isomers.
More rapid metabolism of P- and y-HBCD relative to a-HBCD was demonstrated in in vitro studies
using rat liver microsomes (Abdallah et al. 2014; Esslinger et al. 201 lb; et al. 2005). Following
incubation of the liver microsomes with NADPH and a 1:1:1 mixture of a-, P-, and y-HBCD, LC/MS
peaks for P- and y HBCD in the incubation fluid were greatly diminished after 90 minutes, whereas the
peak for a HBCD was essentially unchanged. In addition, degradation rates for enantiomeric isomers (+)
a and (-) a-HBCD were faster in rat liver microsomes than rates for (+) P-, (-) P-, or (-) y-HBCD
(Esslinger et al. ) ( ittah et al. 2014) calculated half-times of 17.14, 11.92, and 6.34 seconds
for in vitro rat liver microsomal metabolism of a-, y-, and P-HBCD, respectively.
Hydroxylation and debromination have been identified as metabolic pathways for HBCD isomers based
on partial characterization of metabolites in animal and in vitro studies. Analysis of adipose, liver,
muscle, and lung tissue extracts from rats exposed to 100 mg/kg-day commercial HBCD (enriched in the
y-isomer) for 28 days identified mono- and dibrominated metabolites of HBCD as well as
monohydroxylated derivatives of the dibrominated metabolites pentabromocyclo~,dodecene and
tetrabromocyclododecene (Brandsma et al. 2009). No sex dependent differences in metabolite profiles
were observed (Brandsma et al. 2.009). Hydroxylated metabolites of P- and y HBCD, along with other
unidentified metabolites, were also detected by LC/MS of incubation fluid after rat liver microsomes
were incubated with a mixture of a-, P-, and y-HBCD (1:1:1) and NADPH (Zegers et al. 2005).
Although specific enzymatic pathways for metabolism of HBCD have not yet been identified, results of
animal in vivo and in vitro studies are consistent with hydroxylation catalyzed by CYP450 enzymes, as
suggested by the observation that HBCD induced messenger ribonucleic acid (mRNA) levels for
CYP2B1/2 and CYP3A1/3 in livers of rats following 28 days of dietary exposure to commercial HBCD
(Canton et al. 2008; Germer et al. 2006). There are no data describing the potential contribution of gut-
mediated HBCD metabolism. However, it is likely that fecal metabolites are predominantly liver-
derived, as only radioactive metabolites (no parent compounds) were found in the bile of mice orally
exposed to a- or y-[ 14CJ-HBCD (Szabo et al. 201 la. 2010).
The available data are consistent with the proposed generalized metabolic pathways shown in Figure
3-2, in which debromination occurs via undetermined enzymes and hydroxylation occurs via CYP450
oxygenases (Brandsma et al. 2009). The generalized metabolic scheme in Figure 3-2 does account for
the in vivo and in vitro evidence that isomer-specific metabolic pathways may exist in laboratory
Page 340 of 723
-------
animals or data suggesting that HBCD metabolites may be conjugated prior to excretion. (Hakk et al.
2012) found evidence for different metabolic products of y-HBCD and a-HBCD using LC/MS analysis
of extractable and nonextractable HBCD metabolites in blood, fat, brain, bile, urine, and feces collected
in the toxicokinetic studies of mice exposed to radiolabeled y-HBCD (Szabo et al. 2010) and a-HBCD
(Szabo et al. 2011a). After a-HBCD exposure, two glutathione conjugates of a tri- or tetra-brominated,
unsaturated C6 hydrocarbon were identified in urine, and a monohydroxylated, hexabrominated
metabolite was identified in feces (Hakk et al. 2012). After y HBCD exposure, greater numbers of
metabolites were identified in urine and feces: (1) two carboxylic acid derivatives (indicative of ring
opening), a hydroxylated, pentabrominated derivative, and a putative methyl mercapturate of a
tetrabrominated derivative in urine; and (2) three debrominated and oxidized derivatives in feces (Hakk
et al. 2012). In rat liver microsomes tested in vitro, varied monohydroxylated HBCD products for each
of several tested enantiomeric substrates were detected: one from (+) a-HBCD; three from (-) a-HBCD;
two from (+) y-HBCD; and three from (-) y-HBCD (Esslinger et al. 201 lb).
HBCD -
OH
Monohydroxy
HBCD
,-OH
f
Dihydnoxy -
HBCD '
Br
A
z
Br
2
PBCDe
^OH
I
Br
Monohydroxy
PBCDe
OH
j
ir
V Dihydroxy
PBCDe
Br
2
TBCDe
OH
r
Monohydroxy
TBCDe
HBCD = hexabromocyclododecane; PBCDe = pentabromocyclododecene; TBCDe = tetrabromocyclododecene
Source: Adapted from (Brandsma et al. 2009).
Figure 3-2. Proposed Pathways for Metabolism of HBCD in Rats
3.2.2.1.4 Elimination
Elimination of radioactivity associated with administration of HBCD isomers is rapid, with most
eliminated over the first 24 hours post administration, after either oral or i.v. dosing in female mice
(Sanders et al. 2013; Szabo et al. 2011a. 2010) or oral administration in the rat (Yu and Atallah 1980).
Fecal and urinary excretion are the primary excretory pathways for absorbed HBCD, although the
detection of HBCD isomers in many studies of human breast milk samples indicates that breast milk fat
represents an additional elimination pathway.
Page 341 of 723
-------
The fecal:urine excretion ratios (based on samples collected over 48 hours postdosing) for absorbed
HBCD in mice exposed by gavage to 3 mg/kg were approximately 2.4 for a-[14C]-HBCD, 1.2 for P-
[14CJ-HBCD, and 2.1 for y [14C] HBCD (Sanders et al. 2013; Szabo et al. 201 la. 2010). Similar ratios
were seen after i.v. dosing at the same exposure level (Sanders et al. 2013; Szabo et al. 2011 a, 2010).
Together, urinary and fecal excretion 48 hours after dosing accounted for -70% of the administered
radioactivity (at 3 mg/kg) after exposure to the a isomer and -90% after exposure to the P- and y
isomers (Sanders et al. ; Szabo et al. 2.01 la. 2010). Excretion was essentially complete within 48
hours after either oral or i.v. dosing; studies evaluating elimination over longer time periods showed
little additional excretion after 48 hours (Szabo et al. 201 la. 2010).
The overall kinetics of urinary and fecal elimination in the rat is similar to mice, but sex-dependent
differences were suggested by data in rats. Forty-eight hours after dosing with [14C] y HBCD (mixed
with technical-grade HBCD containing -75% y-HBCD), fecal elimination accounted for 63% of
radioactivity in four female rats and 95% in two male rats (Yu and Atallah 1980). Over the same time
frame, urinary elimination accounted for 4.8 and 15.3% of radioactivity in female and male rats,
respectively.
In female mice administered a-[14C]-HBCD by gavage, a dose-dependent shift in fecal elimination was
observed (Szabo et al. 201 la). Fecal elimination accounted for about 48% of the administered radiolabel
at 3 mg/kg, but only about 32% following a 100 mg/kg dose (Szabo et al. 201 la). The mechanism for
the dose-dependent decrease in fecal excretion has not been identified; however, since radioactivity
derived from absorbed a-[14C]-HBCD is extensively excreted into feces, this outcome suggests a
possible capacity limitation in the secretion (e.g., biliary) mechanism. This dose-dependency was not
observed in similar studies of y-[14C]-HBCD in mice (Szabo et al. 2010). In mice given single doses of
P-[14C]-HBCD of 3, 30, or 100 mg/kg, the amount of administered radioactivity in 24-hour feces was
greater after 3 mg/kg (—50%) than after 100 mg/kg (—30%), but no dose-dependent difference was noted
in cumulative 96-hour feces (Sanders et al. 2013).
Biphasic elimination kinetics of radioactivity from blood and tissues of mice were observed following
oral administration of a-, P-, or y-[14C]-HBCD in corn oil vehicle (Sanders et al 2013; Szabo et al.
201 la. 2010). Tissue half-life values for the rapid phase in mice ranged from 0.1 to 0.4 days for a-
HBCD, from 0.02 to 0.2 days for P-HBCD, and from 0.3 to 1 day for y-HBCD. Terminal tissue half-life
values were longer for a HBCD (range, 0.5-17 days) than for y-HBCD (range, 0.8-5.2 days) or P-
HBCD (0.2-7 days). In particular, the terminal half-lives for fat tissue were 17 days for a-HBCD, 3.6
days for y-HBCD, and 2.5 days for P HBCD, indicating that, with repeated oral exposures, a-HBCD
would be expected to accumulate in fat to a greater extent than y HBCD or P-HBCD. Similar biphasic
excretory kinetics were observed in rats following single gavage doses of commercial HBCD with y-
[14CJ-HBCD (Yu and Atallah 1980). At the higher end of the range, (Gever et al. 2004) derived an
HBCD terminal elimination half-life of 64 days via estimation of human daily intake and body burden
(estimate for breast milk) as well as via estimation of half-life in adipose tissue of rats. Tissue excretory
kinetic data for humans are not available.
Breast milk lipid represents an additional elimination pathway for HBCD, and concentrations of HBCD
in human breast milk samples have been well studied; only a few reports are summarized here. Most
biomonitoring studies report total HBCD concentrations in breast milk around 1 ng/g. For example, the
following lipid-normalized median concentrations were reported: 0.9 ng/g lipid (range: 0.3-2.2 ng/g)
and 0.4 ng/g (range: 0.2-1.2 ng/g) for populations in the United States (Texas) in 2002 and 2004,
Page 342 of 723
-------
respectively (Ryan and Rawn 2014); 0.7 ng/g (range: 0.1-28.2 ng/g) in Ontario, Canada; 3.83 ng/g
(range 1-22 ng/g) in the United Kingdom (Abdallah and Harrad 2011); 0.6 ng/g (range: 0.6-5.7 ng/g) in
Belgium (Roosens et al. 2010b): and 0.86 ng/g (range: less than the limit of quantitation [LOQ] -31
ng/g) in Norway (Thomsen et al. 2010). (Ryan et al. 2006) reported that most of the HBCD detected in
breast milk from Texas women was the a-isomer, whereas in Japanese women, mean lipid-normalized
concentrations of a-, P-, and y-HBCD in breast milk were 1.5, <0.1, and 2.6 ng/g, respectively
(Kakimoto et al. 2008).
3.2.2.2 Description of Toxicokinetic Models
No physiologically based pharmacokinetic (PBPK) models are available for HBCD. An unpublished,
empirical two-compartment open kinetic model for orally-administered 14C-HBCD was developed from
data collected using Sprague-Dawley rats given single oral doses of commercial HBCD labeled with y-
[14CJ-HBCD (7-9 mg/kg) (Yu and Atallah 1980). The model did not explicitly describe the metabolism
of HBCD; however, the model did estimate an elimination constant. The elimination constant accounted
for metabolism of HBCD and excretion of metabolites into urine and feces. The central compartment of
the model comprised blood, muscle, liver, kidney, heart, spleen, lung, gonads, and uterus, and the
remaining compartment represented fatty tissues. The calculated concentrations of radioactivity in the
central and fat compartments were compared with respective observed concentrations in the blood and
fat. The pattern of predicted values of radiolabel in blood and fat generally reflected the pattern of
observed values in blood and fat. This kinetic model addressed the distribution of radioactivity only, and
did not explicitly describe metabolism.
(Aylward and Hays 2011) proposed the use of lipid-adjusted tissue concentrations of HBCD as an
internal dose metric that would reduce uncertainties associated with the inter- and intraspecies
extrapolation based on external dose. They derived a simple first-order elimination model to estimate the
steady-state lipid concentration of HBCD (in ng/g lipid) corresponding to a given daily HBCD intake (in
mg/kg-day) as follows:
D = CI x F1 x k
where D = chronic daily dose in mg/kg day, CI = lipid concentration (in mg/kg lipid), F1 = fraction of
body weight that is lipid (assumed to be 25%), and k = elimination rate calculated from the half-life
(HL, assumed to be 64 days in days) as k = In (2)/HL.
As noted by (Aylward and Hays 2011). uncertainty in the steady-state lipid concentration of HBCD
derived using this model comes from the assumed values for the half-life of HBCD (which is on the
higher end of estimates from several studies (see Section 3.2.2.1.4)) and the proportion of lipid in the
body. If used for purposes of interspecies extrapolation, uncertainty is also introduced by potential
toxicokinetics differences across species (e.g., differences in rates of metabolism of the different HBCD
isomers), and consideration of whether summed or isomer-specific doses should be used. If humans
clear individual isomers at a different rate than animals, and if the toxicity of individual isomers differs,
the internal summed dose could either over- or under predict the response. Finally, it should be noted
that a systematic examination of whether lipid-adjusted tissue concentrations better correlate with
response than other measures of dose (e.g., blood concentration, total concentration) has not been
conducted. Based on the absence of a robust, peer reviewed PBPK model and the uncertainties inherent
in the limited simple models, EPA relied on traditional route-to-route extrapolation, uncertainty factors,
and dosimetric adjustment factor in the derivation of HEDs.
Page 343 of 723
-------
3.2.3 Hazard Identification
The HBCD database includes six epidemiological studies that examined associations between HBCD
exposure and endpoints related to effects on the thyroid, nervous system, and male reproductive system.
The evaluation of HBCD epidemiology studies by each of the five aspects of study design - study
population characteristics and representativeness, exposure measures, outcome measures, confounding,
and analysis - is discussed below; a summary of the results from these studies and the data quality
evaluation of individual studies is provided in Risk Evaluation for Cyclic Aliphatic Bromide Cluster
(HBCD), Supplemental Information on Human Health Hazard. (U.S. EPA. 2Q19e) and Risk Evaluation
for Cyclic Aliphatic Bromide Cluster (HBCD), Systematic Review Supplemental File: Data Quality
Evaluation of Human Health Hazard Studies ( )19n). Overall, EPA determined that the
epidemiological database was insufficient for dose-response assessment.
Experimental animal studies of HBCD that underwent study evaluation consisted of studies designed to
examine repeat-dose oral toxicity and specialized studies of various non-cancer hazards. The majority of
the experimental animal studies were considered informative and useful for characterizing the health
hazards associated with exposure to HBCD, and results from these studies were extracted into evidence
tables in the Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information
on Human Health Hazard (!.' ^ I'P \ JO 19e) and [Risk Evaluation for Cyclic Aliphatic Bromide Cluster
(HBCD), Systematic Review Supplemental File: Data Extraction Tables for Human Health Hazard
Studies (U ,S. EPA. 2019gYl. Some limitations were noted for each study (see the Risk Evaluation for
Cyclic Aliphatic Bromide Cluster (HBCD), Systematic Review Supplemental File: Data Quality
Evaluation of Human Health Hazard Studies (U.S. EPA. 2.019n). Any study evaluation concerns that
may have meaningfully influenced the reliability or interpretation of the results were brought forward
into the synthesis of evidence for a given hazard. Two studies were considered for dose-response
assessment of all endpoints (Ema et al. 2008; WIL Research 2001). both of which scored a High in data
evaluation.
Animal studies of ingested HBCD reported effects on the thyroid, liver, development, reproduction,
nervous system, and immune system, in addition to limited studies demonstrating overt toxicity
following acute exposure and sensitization/irritation. The potential health effects of inhaled HBCD have
not been adequately investigated in humans or animals. There is not adequate available information to
assess the carcinogenic potential of HBCD.
3.2.3.1 Non-Cancer Hazards
Data evaluation results for all studies can be found in the [Risk Evaluation for Cyclic Aliphatic Bromide
Cluster (HBCD), Systematic Review Supplemental File: Data Quality Evaluation of Human Health
Hazard Studies (U.S. EPA. 2019m VI and data extraction results including author-reported PODs can be
found in the [Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Systematic Review
Supplemental File: Data Extraction Tables for Human Health Hazard Studies ( ,019 e)/.
For additional, more detailed information on toxicity information, weight of evidence, and mechanistic
data see Section 3.2.4 and [Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on Human Health Hazard ( !.019e)1.
3.2.3.1.1 Thyroid Effects
In humans, (Eeaesbg et al. 2011) reported elevated but non-statistically significant odds ratios for
increased thyroid stimulating hormone (TSH) in relation to increased HBCD levels in breast milk.
Confidence intervals (CIs) around point estimates were relatively wide and a clear dose-response was
not observed. Therefore, this study is considered as a no-effect finding. Similarly, other studies in
Page 344 of 723
-------
humans (Kicinskt et al. 2012; Raze et al. 2009; Johnson et al. ^ also did not observe any statistically
significant correlations with HBCD exposure and thyroid effects among populations of various
lifestages.
Although the human evidence was inconclusive, oral toxicity studies in rodents provide evidence that
HBCD exposure can result in dose-related perturbations of thyroid function. In studies of HBCD-
induced perturbation of serum thyroid hormone levels {i.e., TSH, T4, and T3), TSH was elevated in
three studies (Saegusa et al. 2009; Ema et al. 2008; [ ^search 2.001). two of which also reported
decreases in serum T4 (Ema et al. 2008;] ^search 2.001). Of the several studies that measured T3
(Saegusa et al. 2009; van der Yen et al. 2009; Ema et al. 2008; van der Yen et al. 2006; W il Research
2001). only one reported a treatment-related effect (Saegusa et al. 2009). with a statistically significant
reduction observed at the highest dose. Exposure to HBCD was also associated with histopathological
changes, including decreased thyroid follicle size (Ema et al. 2008; van der Yen et al. 2006). follicular
cell hypertrophy (Rasinger et al 2018; Saegusa et al. 2009; WIL Research 2001). colloid depletion
QVIJ Research. 1997). and increased thyroid weight (Saegusa et al. 2009; Ema et al. 2008; van der Yen
et al. 2006; ] ;search 2001). These changes were observed across multiple rat strains, sexes,
exposure durations, and study designs.
3.2.3.1.2 Liver Effects
There are no epidemiological studies that investigated the potential for an association between HBCD
exposure and liver outcomes; however, some evidence for liver toxicity was identified in several rodent
studies. The most consistently observed liver outcome was liver weight changes. Dose-related increases
were consistently observed across species, sexes, and age from multiple studies of various designs and
exposure durations (Maranghi et al 2013; Saegusa et al 2009; Ema et al. 2008; ] ;search 2001.
1997). Limited support for HBCD effects on the liver are provided by histopathological examination. A
subset of the rat studies (Saegusa et al. 2009; WIL Research 2001. 1997) and two mouse studies
(Rasinger et al. 2018; Maranghi et al. 2013) reported increased vacuolation (generally of minimal to
mild severity) in HBCD-exposed animals, but these responses were not dose-related. Other histological
findings were less frequently observed and included some additional evidence of fatty change (steatosis)
(Yanagisawa et al. 2014). hypertrophy (Yanagisawa et al. 2014; WIL Research 1997). and inflammation
(Maranghi et al. 2013). In a single-dose mouse study, only 49.5 |ig/kg of HBCD administered for 28
days in a fish-based diet also resulted in a statistically significant increase of lymphocytic infiltration,
and hyperaemic vessels (Rasinger et al. 2018). Of note, (Yanagisawa et al. 2014) scored Unacceptable in
data quality evaluation due to relying on an intermittent lx/week dosing schedule, however observations
from that study still contribute to hazard identification. Statistically or biologically significant elevations
in serum liver enzymes were not consistently associated with HBCD exposure in rats or mice
(Yanagisawa et al. 201 I; WTi Research 1997). although a dose-responsive (but non-statistically
significant) increase in alanine aminotransferase (ALT) was observed in female rats and statistically-
significant elevated gamma-glutamyl transferase (GGT) was observed in the high dose group of both
sexes (WIL Research 2001).
3.2.3.1.3 Reproductive Effects
Female reproductive effects
There are no epidemiological studies evaluating female reproductive outcomes. In animals, some
evidence for an association between HBCD exposure and female reproductive system effects comes
from findings of effects on fertility and pregnancy outcome as reported in a two-generation reproductive
toxicity study for HBCD in rats (Ema et al. 2008); signs of reproductive toxicity included dose-related
decreases in pregnancy incidence in F0 and F1 generations, and a statistically significant incidence of
total litter loss in multiple high-dose F1 dams. Decreased primordial follicles were also observed in the
Page 345 of 723
-------
F1 dams (this endpoint was not evaluated in FO females). In a single-dose study, only 49.5 |ig/kg of
HBCD administered to female mice for 28 days in a fish-based diet also resulted in histopathological
changes to the uterus, decreased oestradiol 17p, and an increased oestradiol 17p /testosterone ratio
(Rasinger et al. 2018).
Male reproductive effects
Two epidemiological studies investigated reproductive endpoints in male subjects from a birth cohort
and adult males seeking infertility treatments (Johnson et al. 2.013; Meiier et al. 2012); these studies
provide some evidence of a weak to moderate negative correlation between HBCD exposure and serum
testosterone or sex hormone binding globulin (SHBG) levels, but not other hormones.
In animal studies, no consistent effects on male reproductive organ weights, reproductive development,
hormone concentrations, or spermatogenic measures were associated with 28-day, 90-day, or
developmental exposure to HBCD (Saegusa et al. 2009; van der Yen et al. 2009; Ema et al. 2008; van
der Yen et al. 2006; WIL Research 2001).
3.2.3.1.4 Developmental Effects
There are no epidemiological studies evaluating developmental-specific outcomes. However, several
studies of rodents exposed during gestation and lactation provide some evidence of developmental
effects associated with HBCD, including reduced offspring viability (Ema et al. 2008). decreased pup
body weight (Maranghi et al. 2013; Saegusa et al. 2009; van der Yen et al 2009; Ema et al 2008).
altered development of the skeletal system, and delayed eye opening (Ema et al. 2008). Evidence of
adverse developmental effects is based on findings of reduced offspring survival and decreased pup
body weight. Reduced viability was observed in F2 pups of the two-generation study by (Ema et al.
2008); the decreases in viability were dose-related and observed on both post-natal day (PND) 4 and 21.
The fact that effects were seen only in F2 offspring is consistent with decreased viability manifesting
after multigenerational exposure, although that hypothesis cannot be established based on the current
developmental literature for HBCD (i.e., a single two-generation study). Effects on pup body weight
were demonstrated in several studies in rats using different strains and exposure durations (Saegusa et al.
2009; van der Yen et al. 2009; Ema et al. 2008). Other developmental effects, including changes in bone
development and delayed eye opening, were only reported in a single study and with a less clear dose-
response relationship (van der Yen et al. 2009; Ema et al. 2008).
3.2.3.1.5 Neurological Effects
Developmental exposure
In an epidemiological birth cohort study in the Netherlands (Roze et al. 2009). the associations between
maternal HBCD levels (week 35 of pregnancy) and multiple neuropsychological domains were
inconsistent across the measured domains. A second epidemiological study in adolescents in Belgium
(Kicihski et al. , ) did not observe associations between HBCD levels and six neurobehavioral
measures. In rodents, there is some evidence to support HBCD-mediated neurotoxicity following
developmental exposure. Early-life exposure in rodents affected several measures of neurotoxicity,
including neurodevelopmental milestones (Miller-Rhodes et al. 201 I; l-'tua et al. 2008). locomotor
activity and executive function (Miller-Rhodes et al. 2014; Ema et al. 2008; Eriksson et al. 2006). and
other neurological outcomes related to changes in auditory sensitivity, dopaminergic system function
(Lilienthal et al. 2009). and brain weight (van der Yen et al. 2009; Ema et al. 2008). (Eriksson, et al.
2.006) evaluated effects in young adult (3-month-old) mice that were administered a single dose of
HBCD on PND 10, which corresponds with a period of rapid growth and maturation for motor and
sensory neural networks in mice.
Page 346 of 723
-------
Adult exposure
There are no epidemiological studies evaluating nervous system effects following adult exposure. In
animals, four studies in rats or mice exposed only as adults found no changes in the nervous system
endpoints evaluated (i.e., striatal levels of dopamine, Functional Occupational Battery (FOB), locomotor
activity, brain weight, or gross brain pathology) (Genskow et al. 2.015; van der Yen et al. 2006; WIL
Research 2001. 1997). Notably, HBCD was cytotoxic to neuronal cell lines and reduced expression of
dopaminergic transporters in mice despite not affecting overall levels of striatal dopamine (Gemskew et
al.. 2015). Results on locomotor activity indicated that mice failed to habituate to the novel environment
of the testing arena, however this result was not confirmed in a longer duration study (Miller-Rhodes et
• raa et al. 2008).
3.2.3.1.6 Immune System Effects
There are no epidemiological studies evaluating immune system effects. In animals, there is some
evidence of HBCD-mediated immune system effects. The strongest evidence comes from alterations in
IgG antibodies, a functional measure of immune system response, in rats exposed to HBCD during
development (Hachtsuka et al. 2010; van der Yen et al. 2009; Etna et al. 2008). Changes were also
observed in other indicators of immunomodulation, including changes in immune organ weights
(thymus and spleen), changes in hematological parameters, and histopathology. Decreased thymus
weight (e.g., thymic atrophy), especially during development, have the potential to cause significant
chronic long-term immune effects. These observed changes were variable and inconsistent (including
directionality) however in both developing and adult animals (Hachisuka et al. 2010; van der Yen et al.
2009; Etna et al. 2008; van der Yen et al.. 2006). Recent mechanistic studies (Almughamsi and Whalen
2.016; Anisuzzaman and Whalen 2.016; Canbaz et al. 2016a; Koike et al. 2016) along with bioassays
from the EPA ToxCast Dashboard
(https://comptox.epa.gov/dashboard/dsstoxdb/results?search=hbcd#invitrodb) demonstrate changes in
cytokine secretion and cell surface marker expression from immune cells following HBCD exposure;
these changes were not always consistent however and could not be directly linked to any particular
toxicological outcome.
3.2.3.1.7 Overt Toxicity Following Acute/Short Term Exposure
Acute/short term studies in animals consist of either single or short-term exposures (14-days or less) at
high doses specifically designed for assessing the dose at which lethality occurs or for examining overt
toxicity. Several acute lethality studies in rodents and rabbits by the oral, dermal, and inhalation routes
with HBCD are available (GSK1 ! Momma etal. 1993; BASF 1990; ! fa, b, c; Lewis and
Pal anker 1978a). The acute lethality of HBCD is relatively low via the oral, dermal and inhalation
routes. Oral LDso values are equal to or greater than 680 mg/kg-bw, in rats and mice. Various neurotoxic
signs observed in oral studies included ptosis (upper eyelid drooping), apathy, trembling, and
hypoactivity. Additional effects included lacrimation (tears), diarrhea, and inflammation (
2015a). No lethality was observed in rabbits following acute dermal exposure to doses as high as 8.0
g/kg (Lewis and Palanker 1978a). Several inhalation studies have demonstrated no mortality in rats
following exposure to up to 200 mg/L (200,000 mg/m3) HBCD for 1 -4h (U.S. EPA. 2015a). with only
minor symptoms observed (such as eye squint, slight dyspnea, salivation, lacrimation, and nasal
discharge). A recent study confirmed that the HBCD LCso for 4-h inhalation exposure in rats is greater
than 5000 mg/m3 (Song et al. 2016). In that same study, HBCD also did not produce any adverse effects
(clinical signs or organ-specific pathology) up to 2000 mg/m3 administered 6h/day for 14 days.
3.2.3.1.8 Sensitization/Irritation
The available literature indicates that HBCD is not a dermal irritant in guinea pigs (Lewis and Palanker
1978b). Acute eye irritation studies in rabbits showed HBCD to be a mild transient ocular irritant (Lewis
Page 347 of 723
-------
and Patamker 1978b). (Gulf South Research Institute, 1988). One study (Momma etai 1993) found
HBCD to be a mild skin allergen in guinea pigs, however (Microbiological Associates 1996b) did not
observe any sensitization reaction at the same dose (5%) or neat in corn oil (-100%) (NRC 2000b). Two
mechanistic studies suggest that HBCD enhances the allergenic response to dust-mites (Canbaz et al.
2.016a; Canbaz et al. 2016b)l. and there is some evidence of HBCD stimulating the release of various
pro-inflammatory cytokines that may promote allergic responses (Almughamsi and Whalen 2016;
Anisuzzaman and Whalen 2016; Canbaz et al. 2016a; Koike et al. 2016).
3.2.3.2 Genotoxicity and Cancer Hazards
Genotoxicitv
A limited number of studies have investigated the genotoxicity of HBCD. The majority of these studies
were standard Ames tests for detecting mutagenic potential in the bacteria (Salmonella typhimurium)
These tests, which employ different strains of bacteria that have been developed with pre-existing
mutations, including S. typhimurium TA98, TA100, TA1535, TA1537, and TA1538, are referred to as
reversion assays (Maron and Ames 1983). Most of these assays conducted with HBCD yielded negative
results (Huntingdon Research. Center 1990; IBT Labs 1990; Litton Bionetics 1990; Pharm.akologisch.es
Institut 1990; SRI International 1990; Zeigeretal. 1987). Negative results were also obtained in (GSRI
(1978)). (IBT Labs. 1990)and (Huntingdon Research Center (1990)). however these studies scored
Unacceptable. Among the few assays performed to determine the genotoxicity of HBCD in eukaryotic
systems, one in yeast (Litton. Bionetics 1990) and one detecting chromosomal aberrations in human
peripheral lymphocytes in vitro (Microbiological Associates 1996a) were negative, even when tested at
cytotoxic concentrations. A single in vivo mouse micronucleus test following intraperitoneal (i.p.)
injections of HBCD (BASF 2000) was also negative, however the full study was unavailable for data
quality review.
Some positive results have been reported. S. typhimurium strain TA1535 was positive for reverse
mutations at the highest dose only using a liquid residue of HBCD in DM SO ( ,abs 1990). and
strain TA100 was positive also at the highest dose using an unidentified mixture characterized only as
HBCD bottoms in acetone (Ethyl Corporation 1990b). In this same study, TA1535 was positive at >100
Hg/plate without addition of an S9 microsomal fraction (Ethyl Corporation 1990b). The number of
revertants increased with dose. This was the only Ames study to report dissolving the test article in a
solvent other than DMSO (in this case, acetone). DMSO is a free-radical scavenger and can potentially
obscure genetic damage due to oxidative radicals. Both strains TA1535 and TA100 were designed to be
sensitive to detecting reversions by base substitution, a type of genetic lesion that can result from
oxidative DNA damage due to reactive oxygen species (ROS). However, there is only limited evidence
in the literature indicating that HBCD exposure may induce oxidative stress (An et al. 2013; Hu et al.
2009b).
In mammalian systems, a reverse mutation assay with Chinese hamster ovary (CHO) Sp5 and SPD8 cell
lines exposed to HBCD (Hetledav et al. 1999) yielded positive results. These two clones exhibit a partial
duplication of the hprt gene, causing lethality unless a reversion occurs, either via homologous
recombination (SPD8) or non-homologous recombination (Sp5). A statistically significant, dose-
dependent increase in reversion frequency was observed in both clones, although at higher doses, there
was a significant inhibition of cloning efficiency. In addition, a test of unscheduled DNA synthesis with
rat hepatocytes exposed to HBCD bottoms was positive (Ethyl Corporation 1990a) as well as comet
assays in human hepatocyte L02 and hepatoma HepG2 cells (An et al.. 2013; Huang et al. 2016) and
each study showed a dose-responsive increase in response. Interestingly a follow-up study by An et al.
(2016) found that pre-incubation of L02 cells with sub-mutagenic doses of HBCD promoted adaptive
responses that protect against genotoxic effects of subsequent high doses.
Page 348 of 723
-------
It is noteworthy that in these studies, the positive results were dose-dependent, observed at nontoxic
doses, and in two assays, specific for detecting mutations. However, the tests in bacteria and yeast along
with the single mammalian in vivo study (BASF 2000) were predominantly negative.
Page 349 of 723
-------
Carcinogenicity
The carcinogenic potential of HBCD was not evaluated in any epidemiological studies. The only
experimental animal study to examine cancer endpoints is an 18-month dietary study in mice that was
only available as an incomplete report (Kurokawa et al. 1984). That study concluded that HBCD was not
carcinogenic at dietary concentrations of 100, 1000, and 10,000 ppm.
Full details for all studies are provided in Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD),
Supplemental Information on Human Health Hazard ( e).
3.2.4 Weight of the Scientific Evidence
For more detailed discussion on weight of evidence and mode of action, see Risk Evaluation for Cyclic
Aliphatic Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard (U.S. EPA.
2019 e).
3.2.4.1 Non-Cancer Hazards
3.2.4.1.1 Thyroid Effects
The human database was considered too limited for drawing conclusions regarding the relationship
between HBCD exposure and thyroid effects. Several human epidemiological studies investigated the
association between HBCD exposure and alteration of thyroid hormones at various lifestages. (Eggesb.0
et al. ) reported an elevated but non-statistically significant odds ratio for increased TSH in relation
to increased HBCD levels in breast milk, but confidence intervals around point estimates were relatively
wide and a clear dose-response was not observed. Other studies also found no significant correlations
with HBCD exposure and thyroid effects. In general, these HBCD studies were limited by small sample
sizes (Kim and Oh 1 I, tohnson et al. 2013; Roze et al. 2009) or HBCD exposure quantification
methods (Kim and Oh 2014; Kicinski et al. 2 )
Animal toxicity studies provided evidence of thyroid perturbation associated with HBCD exposure,
including altered levels of thyroid hormones, histological changes, and increased thyroid weight, with
effects observed across multiple lifestages. Increased TSH is a sensitive early indicator of disruption of
the thyroid hormone economy, including decreased thyroid hormone synthesis or secretion, decreased
serum concentrations of T4, or decreased deiodination of T4 to T3 in peripheral tissues. A pattern of
increased TSH and decreased T4 that was observed in a two-generation reproductive study (Ema et al.
2008) is consistent with the multi-loop feedback system of the HPT-axis (Fisher and Nelson 2012). A
similar pattern of effect in TSH and T4 was reported by (WIL Research 2.001); however, this study
scored a low in data quality for thyroid outcomes despite scoring a high in data quality for other
endpoints due to inadequate reporting of thyroid hormone measurement methods, questionable control
data (unrealistically low TSH measurements), inconsistent data reporting across tables, and small sample
sizes. Although these two studies did not observe significant changes in T3, this finding is not surprising
given that T4 is the major thyroid hormone in the blood and most T3 is created by deiodination of T4 in
the peripheral tissues (Rosol et al. 2013). In addition to changes in serum hormone levels, evidence of
thyroid activation, including histopathological changes (Saegusa et al. 2009; Etna et al. 2008; van der
Yen et al. 2006; WIL Research 2001. 1997) and increased thyroid weight (Saegusa et al. 2009; Em.a et
al. 2008; van der Yen et al. 2006; esearch 2001). were observed in both sexes and across studies
of different exposure durations (subchronic, short-term, and one- and two-generations).
Regulation of thyroid hormones is complex and homeostasis is largely maintained via hypothalamic-
pituitary-thyroid (HPT) axis feedback mechanisms. Reductions in serum T3 or T4 triggers release of
TSH from the pituitary, which stimulates the thyroid gland to increase secretion of T3 and T4 stores
Page 350 of 723
-------
from the colloid (Fisher and Nelson 2012). Decreased T4 is expected to be the primary driver of HBCD-
mediated thyroid effects that triggers release of TSH. Reduced T4 alone can lead to adverse effects on
the developing nervous system even in the absence of changes to T3 and TSH levels (although a
statistically significant increase in TSH levels following HBCD exposure was observed in parallel in
(Etna et al. 2008)). Indeed, this is supported by mechanistic studies that indicate that observed decreases
in T4 may be largely driven by hepatic induction of enzymes that metabolize this hormone (Shelby et al.
2003; Van sell and Klaassen 2002; Kelly 2000). Furthermore, reduced T4 levels can also play a key role
in other downstream effects such as liver toxicity, developmental neurotoxicity, as well as other
developmental processes (Finken et al. 2013; Jutvez et al. 2013; Roman et al. 2013; Henrichs et al.
2010; Haddowetal. 1999). A few studies demonstrate that HBCD may induce these human health
hazards downstream of thyroid hormone dysregulation through direct activation of the DNA-binding
thyroid receptor. HBCD-mediated activation of the thyroid receptor has been shown to affect gene
expression, cell proliferation, and morphological development (Haroers et al. 2.006; Schriks et al. 2006).
Mechanistic Evidence
Available mechanistic data suggest that HBCD may interfere with normal thyroid hormone function.
Indirectly, HBCD may decrease circulating thyroid hormone levels by inducing liver xenobiotic
enzymes that are responsible for metabolizing thyroid hormones. Directly, HBCD may act via the
thyroid receptor and regulate thyroid-responsive genes. Other related, but less supported possible
mechanisms, include competition for thyroid hormone binding proteins and dysregulation of
deiodinases. See [Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental
Information on Human Health Hazard (I S I "P \ -I019e)1 for more details.
Relevance and sensitivity of thyroid hormone effects in rodents compared to humans
There is debate as to whether rodents are more sensitive than humans to thyroid hormone disruption.
A review on thyroid disruption by perchlorate by the National Academies of Science (NAS) (NRC
2005) concludes that while thyroid function and regulation are qualitatively similar in rats and humans,
differences in clearance rates and thyroid stimulation require careful consideration for interpreting
thyroid hormone or histopathology changes in quantitative risk assessment. This NAS assessment also
states that humans are less susceptible than rats to disruption of thyroid hormone based on these
differences. This review was targeted to the effects of perchlorate however, with all conclusions
caveated in that they apply specifically to perchlorate exposure and the formation of thyroid tumors,
which is not an expected outcome of HBCD exposure. The mode of action (MO A) for perchlorate
involves inhibition of sodium-iodide symporter (NlS)-mediated iodide uptake in the thyroid, and NAS
recommends use of this effect as the basis for the perchlorate point of departure (POD). There is no
evidence that HBCD modulates thyroid hormones through inhibition of iodide uptake.
Available mechanistic evidence suggests that HBCD is likely to function at least partially indirectly
through upregulation of the enzyme uridine diphosphate glucuronyl transferase (UGT) (Crump et al..
2010; Canton et al.. 2008; Cmrnp et al.. 2008; Palace et al.. 2008; van, der Yen et al.. 2006) resulting in
increased thyroid hormone metabolism and excretion (Kato et al. 2008; Klaassen and Hood 2001). This
mechanism would be expected to act on thyroid hormone levels directly, unlike the MOA for
perchlorate. Additionally, a review of the HPT axis across species published more recently than the
NAS review (Zoeller et al. 2007) states that there is minimal evidence linking biochemical and
metabolic differences in thyroid hormones (due primarily to reduce serum binding proteins in rodents) to
differences in sensitivity among rodents and humans except on a MOA-specific basis. The review
concludes that "total T4 in rodents is a valid measure of thyroid function if serum binding proteins are
not being affected by the treatment under study." While there is conflicting limited mechanistic evidence
Page 351 of 723
-------
investigating whether HBCD may affect transcription of the serum binding protein transthyretin (TTR)
(Crump et at.. 2008; Hamers et at.. 2006). the majority of mechanistic data supports an MOA involving
increased thyroid hormone clearance through induction of UGT.
A review by the National Institute of Environmental Health Sciences (N1EHS) (Choksi et at. 2003)
concludes that while the thyroid system is highly conserved between rodents and humans in general,
differences that need to be considered in extrapolating results from animal data include: "metabolic
turnover rates, basal TSH levels, sodium-iodide symporter sensitivities, windows of susceptibility, the
role of the thyroid system on reproductive tract development and function, and the magnitudes of
thyroid system changes that result in adverse health effects, among others. Additionally, thyroid
hormone glucuronidation by UGT is only a minor pathway in humans under euthyroid conditions,
although this can be modulated by upregulated T3 levels or xenobiotic exposure.
Biochemical and metabolic differences among adult rodents and humans may result in quantitative
differences in dose-response and downstream outcomes as a result of decreased serum hormones levels.
Thyroid hormone levels have much shorter half-lives in adult rats compared to humans, potentially due
to a lack of high-affinity T4 binding proteins (e.g., thyroxine-binding globulin, TBG), possibly making
T4 more susceptible to removal (Zoetter et at. 2007). Importantly, TBG is expressed in neonatal rodents
and only decreases following weaning. TBG increases during pregnancy in both rats and humans, while
only in mice does TBG decrease throughout pregnancy (Choksi et at. 2003). In general, there are
significantly fewer differences in thyroid hormone regulation between rodents and humans during
development. In humans, mild to moderate maternal thyroid insufficiency (i.e., low T4 levels) is
associated with higher risk for persistent cognitive and behavioral deficits in children (Finken et at.
2013; Julvez et at. 2013; Roman et at. 2013; Henrichs et at. 2010; Haddowetat. 1999). Similar effects
have been described in animal studies, with modest reductions in maternal T4 during gestation resulting
in behavioral alterations, learning deficits, and neuroanatomical changes in offspring (Gilbert et at.
2014; Gilbert et at. qhert. 2011; Liu et al. 2010; Auso et al. 2004). Therefore, developmental
effects of thyroid disruptors following gestational exposure are expected to be highly comparable
between rats and humans, with substantially increased susceptibility in developing individuals of both
species compared to adults. Additionally, because rats are more altricial than humans, thyroid
maturation (and thyroid hormone-associated growth and development) proceeds later in rats than
humans. Consequently, human offspring may be more susceptible in utero to many developmental
outcomes that were observed only postnatally in rats (e.g., mortality, reduced body weight). In contrast,
in some cases humans exposed only neonatally may have developed compensatory mechanisms that are
not yet fully formed in newborn rodents.
Overall the weight-of-evidence indicates that rodents are a relevant model for assessment of thyroid
disruption by HBCD. While there are some significant differences in the thyroid system between rodent
and human adults, gestational HBCD exposure is likely to result in qualitatively and quantitatively
similar developmental outcomes. Perturbations in thyroid hormones observed in animal studies
following HBCD exposure as well as effects observed in mechanistic studies [Risk Evaluation for Cyclic
Aliphatic Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard (U.S. EPA.
2019e)1. support EPA conducting dose-response analysis on this endpoint. In addition, the other hazards
associated with HBCD toxicity are likely downstream results of the dysregulation of thyroid hormones
and the HPT axis, key events in the associated adverse outcome pathway leading to multiple adverse
outcomes (Forhead and Fowden 2014; Gilbert and Zoetter 2010; Hutbert 2000). This hazard endpoint is
an upstream event of other adverse outcomes and was carried forward for dose-response analysis.
Page 352 of 723
-------
3.2.4.1.2 Liver Effects
No epidemiological studies are available to inform potential adverse effect of HBCD on liver. In
laboratory animals, there is evidence for liver toxicity. The most consistent hepatic change was
increased liver weight, which was observed in the majority of studies, in both sexes, in both rats and
mice, and following both adult and developmental exposures (Maranghi et at 2013; Saegusa et al. 2009;
van der Yen et al. 2009; Ema et al. 2008; van der Yen et al. 20*- V. \¥U Lesearch 2001. Although
the toxicological significance of increased liver weight is not clear, these data are supported by some
histological and mechanistic data. Vacuolation was observed in several rat studies (Saegusa et al. 2009;
search 2001. 1997) and one mouse study (Maranghi et al. 2013). The content of the
hepatocellular vacuoles was investigated by (WIL Research 2001) and characterized as lipid. Studies
reported evidence of inflammatory effects in the liver of mice following HBCD exposure through a
standard chow diet (Maranghi et al. 2013) and enhancement of hepatic fatty changes (steatosis) in mice
when HBCD was added to a high-fat diet (Yanagisawa et al. 2014). Statistically or biologically
significant elevations in serum liver enzymes were not associated with HBCD exposure in rats or mice
(Yanagisawa et al. u, ^ ^ 'search 2001. 1997).
Mechanistic Evidence
HBCD may dysregulate lipid metabolism and transport based on the presence of lipid vacuoles in
hepatocytes (WIL Research. 2001) along with observed increased triglycerides and elevated expressed
of lipid metabolism and transport genes (Yanagisawa et al. 2014). Mechanistic evidence also suggests a
potential role of HBCD in the induction of hepatic microsomal enzymes, a proposed key event in
initiating the perturbation of the HPT axis that leads to reduced T4 levels (see Thyroid section above).
Liver toxicity appears to be especially apparent following a high-fat diet, which may represent a
susceptibility factor for HBCD toxicity (Bernhard et al. 2016).
HBCD has been shown to induce the expression of several hepatic microsomal enzymes (Crump et al.
2010; Crump et al. 2008; Germer et al. 2006). which may result in interplay between liver and thyroid
hormone effects. HBCD may also impair lipid homeostasis. Several studies observed increased
vacuolation in hepatocytes (Maranghi et al. 2013; Saegusa et al. 2009; WIL Research 2001. 1997) and
the only study to evaluate vacuole contents indicated that they predominantly consisted of lipid (WIL
Research 2.001). Additionally, various gene expression studies lend supportive evidence for HBCD-
mediated disruption of genes involved in lipid metabolism and transport. HBCD-mediated alterations in
the regulation of lipid metabolism have also been observed in avian species and in vitro. The lack of
increased incidence of necrosis or apoptosis and/or serum enzymatic markers of hepatocellular damage
suggests that HBCD is not highly cytotoxic in liver. However, there is evidence to suggest the exposure
to HBCD can increase the production of reactive oxygen species (ROS). See [Risk Evaluation for Cyclic
Aliphatic Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard (U.S. EPA.
2019e)1 for more details.
Overall, liver toxicity following HBCD exposure is supported by observations in animal and
mechanistic studies. Additionally, liver toxicity may be exacerbated when HBCD exposure is combined
with a high-fat diet. Therefore, this hazard was carried forward for dose-response analysis.
3.2.4.1.3 Reproductive Effects
Female Reproductive Effects
The potential for HBCD to affect the female reproductive system has not been investigated in humans.
There is evidence for female reproductive hazard in animals, primarily based on effects observed in a
two-generation reproductive toxicity study (Ema et al. 2008). (Ema et al. 2008) reported dose-related
Page 353 of 723
-------
decreased incidence of pregnancy in the FO and F1 generations and a reduced pool of primordial
follicles in the F1 generation. The only other study that looked at a measure of pregnancy incidence was
a one-generation study (van der Yen et al. 2009) that reported no significant dose-response trend on
successful matings (i.e., the rate of matings that results in offspring). Because (van der Yen et al. 2009)
used a lower dose range than (Etna et al. 2008). the lack of effects on reproductive performance from
this study is only informative of an absence of effects at lower doses and does not contradict the
outcomes observed in (Ema et al. 2008) at higher doses. HBCD exposure did not affect other fertility
and pregnancy outcomes (e.g., gestational duration, number of implantation sites, litter size) (Saeeusa et
al. 2009; van der Yen et al. 2009; Em a et al. 2.008). Investigation of other female reproductive outcomes
provides little supportive evidence of reproductive toxicity. Statistically significant changes in sex
hormone levels were limited to increased follicle-stimulating hormone (FSH) as reported by (Ema et al.
2008) and increased testosterone as reported by (Maranehi et al. 2013); levels of other hormones showed
no dose-related changes. Evidence of changes in time to vaginal opening, a measure of reproductive
differentiation and development, were inconsistent across studies. No consistent effects were observed
on measures of reproductive organ weight.
Mechanistic Evidence
Human and rodent cell culture models provide some evidence to support the potential for HBCD to alter
the function of several reproductive hormones. Various studies suggest that HBCD may act as an
androgen receptor agonist (Christen et al. 2010) and a disruptor of FSH are mixed. In addition to
hormone receptor level effects, several studies indicate that HBCD may also perturb enzymes involved
in the synthesis and metabolism of reproductive hormones. See [Risk Evaluation for Cyclic Aliphatic
Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard (U.S. EPA. 2019e )] for
more details.
Evidence for female reproductive toxicity following HBCD exposure is supported by observations in
animal and mechanistic studies. Therefore, this hazard was carried forward for dose-response analysis.
Male Reproductive Effects
Both human and animal evidence for male reproductive effects were insufficient for drawing
conclusions regarding the relationship between HBCD exposure and male reproductive toxicity. Two
epidemiological studies (Johnson et al. 2013; Meiier et al. ^ ) provided limited evidence of male
reproductive effects (effects on serum testosterone and SHBG levels) associated with HBCD exposure
in humans, and animal studies revealed inconsistent effects in all measures of male reproductive
endpoints. Limited mechanistic data on male reproductive toxicity are available [Risk Evaluation for
Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard (U.S.
EPA. )].
Evidence for male reproductive toxicity following HBCD exposure in animal studies was limited and
inconsistent. Therefore, this hazard was not considered further for dose-response analysis.
3.2.4.1.4 Developmental Effects
Studies were not identified that looked at developmental-specific outcomes in humans. Epidemiological
studies pertaining to other organ-/system-specific hazards following developmental exposure are
discussed in Sections 3.2.3.1.1 (thyroid), 3.2.3.1.3 (male reproduction), and 3.2.3.1.5 (nervous system).
Animal toxicity studies provide evidence of a developmental hazard. These data suggest that early life
exposure to HBCD can affect various developmental outcomes, including reduced offspring viability
(Ema et al. 2008) and decrements in pup weight (Maranehi et al. 2013; Saegusa et al. 2009; van der Yen
Page 354 of 723
-------
et al. 2.009; Ema et al. 2.008). Ontogeny of developmental landmarks were either unaffected {i.e., incisor
eruption or pinna unfolding) or effected inconsistently (i.e., eye opening) (Ema et al. 2008). The support
for developmental toxicity is strongest in F2 animals, with effects seen in both sexes in the high-dose
group. This evidence is consistent with developmental thyroid hormone disruption.
Mechanistic data
HBCD exposure in zebrafish is associated with increased ROS generation and induction of apoptotic
cell pathways resulting in malformations and reduced viability CPu et al. 201 _ b, \ >cag et al. 2009; Hu et
al. 2009a) as well as effects on cardiac function (Wu et al. 2016a; Wu et al. 2013). Disruption of thyroid
hormones is strongly associated with downstream developmental effects including growth restriction,
skeletal development, and neurological abnormalities. Although there is limited mechanistic data overall
regarding HBCD-mediated effects on development, perturbations in thyroid hormones could lead to
developmental toxicity because of the role thyroid hormones play during development (Zoeller et al..
2007; Forhead a yden 2014; Gilbert and Zoeller 2010; Hulbert 2000). See [Risk Evaluation for
Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard (U.S.
EPA. 2019e)l for more details.
Evidence for developmental toxicity following HBCD exposure is supported by observations in animals
and mechanistic data on HBCD and thyroid hormone disruption. Therefore, this hazard was carried
forward for dose-response analysis.
3.2.4.1.5 Neurological Effects
Developmental Exposure
The two available epidemiological studies did not find consistent effects on the nervous system
following developmental exposure (Raze et al. 2009; Kicinski et al. 2012). Therefore, the available
human evidence ranges from equivocal to negative.
Some evidence of potential nervous system effects of HBCD comes from early-life exposure studies in
rodents. Perinatal HBCD exposure altered neurodevelopmental milestones (Miller-Rhodes et al. 2014;
Ema et al. 2008). elicited changes in locomotor activity and executive function that persisted into
adulthood (Miller-Rhodes et al. 20 u, * >na et al. 2008; Eriksson et al. 2006). and affected other
neurological endpoints related to changes in auditory sensitivity, dopamine system function (Lilienthal
et al. 2009). and brain weight (van der Yen et al. 2009; Ema et al. 2008). Across the database, nervous
system effects were observed in both sexes and across a wide range of doses and exposure durations
(ranging from acute to multigenerational). However, interpretation of these data was complicated by
study quality issues, including lack of blinding, poor health in the animals, pooling of data across
timepoints, and failure to measure potential confounders. Furthermore, there were considerable
inconsistencies in outcomes across studies that evaluated similar neurodevelopmental endpoints,
including development of sensorimotor reflexes, locomotor activity, learning ability in swim maze tests,
and brain weight.
Mechanistic Evidence
Thyroid hormones are known to play a key role in development of the vertebrate central nervous system,
and perinatal exposure to thyroid-disrupting chemicals has been shown to have lasting effects on
cognitive and behavioral outcomes (Gilbert et al. 2012; Howdeshell 2002; Koibuchi and Chin 2000).
HBCD specifically has been shown to interfere with thyroid hormone-mediated neurogenesis and
differentiation, calcium homeostasis, and neurotransmitter reuptake. Normal neurodevelopment is
dependent on tight regulation of all of these systems and perturbations are associated with persistent
changes in behavior and neurological function (Finken et al. 2013; Julvez et al. 2013; Roman et al. 2013;
Page 355 of 723
-------
Henrichs et al. 2010; Haddowetal. 1999). See [Risk Evaluation for Cyclic Aliphatic Bromide Cluster
(HBCD), Supplemental Information on Human Health Hazard (U.S. EPA. 2019e)l for more details.
Additionally, a recent publication (Rasimeer et al. 2018) identified genomic and proteomic changes in
the brain of female mice related to estradiol signaling, cell-cell junctions, endocytosis, and sirtuin
signaling. Modulation of these functions could result in dysregulated calcium homeostasis, oxidative
stress, and impaired cholesterol/fatty acid metabolism, all of which could result in detrimental
neurological effects.
Overall, there is evidence from animal studies to support potential nervous system effects associated
with HBCD exposure during development. However, although the data support a qualitative assessment
of developmental neurotoxicity, there are notable inconsistencies and/or limitations with the database.
Treatment-related effects were observed in all but one study that evaluated the effects of developmental
exposure on nervous system function, but there was no consistent pattern of effect across studies.
Furthermore, study quality issues {i.e., lack of blinding, health issues in the animals, pooling of data,
failure to measure potential confounders, wide variation in response, and questions regarding the
statistical methodology) were identified in several studies. In light of these uncertainties, selection of
data sets from the available developmental neurotoxicity studies for dose-response analysis was not
supported.
Adult Exposure
Neurotoxicity following HBCD exposure during adulthood was not supported by observations in animal
studies (Gemskow et al. 2015; van der Yen et al. 2006; ] ;search 2001. 1997). Adult male mice
exposed to 25 mg/kg-day for 30 days showed decreased striatal levels of dopamine transporter and
vesicular monoamine transporter 2, regulators of dopamine homeostasis and neurotransmission
(Gemskow et al. 2015). Similarly, an in vitro study found a dose-related reduction in dopamine and
gamma-aminobutyric acid uptake in rat synaptosomes and vesicles exposed to HBCD (Mariussen and
Fonnum 2003). Although prolonged deficits in reuptake mechanisms could result in excessive
stimulation of the post synaptic cell or deplete neurotransmitter stores in the presynaptic cell, (Gemskow
et al. 2015) did not find significant changes in tissue concentrations of dopamine or its metabolites in
adult mice exposed for 30 days. Therefore, this hazard was not carried forward for dose-response
analysis.
3.2.4.1.6 Immune System Effects
The potential immunotoxicity of HBCD has not been investigated in human populations. The effects of
HBCD on both functional and structural immune endpoints were evaluated in animal models. Of the
endpoints evaluated, measures of T cell-dependent antibody responses—functional immune endpoints
and therefore more sensitive and predictive indicators of potential immunotoxicity (Luster et al. 2005)—
were given more weight.
Developmental Exposure
In studies in rats, early-life HBCD exposure altered antibody responses to sheep red blood cells (SRBC)
(increased) (van der Yen et al. 2009) and keyhole limpet hemocyanin (KLH) (decreased) (Hachisuka et
al. 2010). Healthy immune function is maintained as a delicate balance between: (1) an immune
response adequate to provide protection from certain types of cancers and infectious diseases, and (2)
pathological loss of immune system control resulting in conditions such as autoimmunity,
hypersensitivity, and chronic inflammation. Unintended immunomodulation in either direction {i.e.,
immunosuppression or immunostimulation) may be considered adverse (WHO 2012). Therefore, the
difference in direction of effect in the only two measures of antibody response does not necessarily
minimize the validity of the findings in early lifestage animals. These antibody responses were not
Page 356 of 723
-------
adequately supported by consistent observational endpoints. Specifically, a statistically significant
decrease in thymus and spleen weight was observed in the F2 generation of (Etna et al. 2008) but not in
any other study.
Adult Exposure
HBCD did not cause changes in functional immune endpoints in adult rats or mice (Watanabe et al.
2010; van der Yen et al. 2006). The database does not provide a clear and consistent pattern of effect on
immune organ weights, hematology, or histopathology following adult exposure. Given the diversity of
study designs, exposure conditions, and analytical methods represented in this database, it is difficult to
identify the underlying reasons for the differences in observations across studies.
Mechanistic Evidence
Mechanistic data suggests that HBCD stimulates pro-inflammatory cytokines, however some of these
responses are not consistently observed. HBCD may stimulate an immune response by increasing the
activity of antigen-presenting cells (Koike et al.. 2016) and appears to alter human natural killer (NK)
cell function in vitro (Hinkson and Whalen 2010. 2009). See [Risk Evaluation for Cyclic Aliphatic
Bromide Cluster (HBCD), Supplemental Information on Human Health Hazard (U.S. EPA. 2019e)l for
more details.
Overall, while there is some evidence to support immune system effects following HBCD exposure (at
least for early-life exposure), the data are limited and inconsistent. Therefore, the WOE is inconclusive
and this hazard was not carried forward for dose-response analysis.
3.2.4.1.7 Overt Toxicity Following Acute/Short Term Exposures
Studies examining the toxicity of HBCD in humans following acute exposures have not been identified.
There is limited evidence from acute toxicity studies in both rodents and rabbits exposed to high levels
of HBCD for some minor and reversible neurological effects (e.g., ptosis (upper eyelid drooping),
apathy, trembling, and hypoactivity) via the oral route, and mortality via the oral, dermal, and inhalation
routes. Mortality or clinical signs of toxicity were not observed in rats following inhalation exposure to
2000 mg/m3 HBCD administered 6h/day for 14 days (Sons et al. 2016). While this study conflicts with
data from repeat-dose oral studies, the study is of too-short of a duration to examine any chronic effects.
Additionally, the study did not examine the critical effects of thyroid hormone regulation or any
reproductive/developmental outcomes.
Evidence for overt toxicity or mortality at toxicologically relevant doses is not supported by the
available data from high dose acute exposure studies. Additionally, since these shorter-term oral
exposure studies were either acute lethality studies or studies involving only single doses, they were not
considered amenable to quantitative analysis. Therefore, this hazard was not carried forward for dose-
response analysis.
3.2.4.1.8 Sensitization/Irritation
No studies have been identified examining the irritation or sensitization potential of HBCD in humans.
A few studies in animals have found evidence for sensitizing potential of HBCD (Canbaz et al. 2016a;
Momma et al. 1993) and HBCD stimulated release of pro-inflammatory cytokines, however, dermal
sensitization has not been consistently observed (NRC 2000b; Microbiological Associates 1996b).
Overall, there is insufficient evidence of irritation and inconsistent data regarding skin sensitization from
HBCD exposure. In addition, there is only qualitative information available on these hazards. Therefore,
they were not carried forward for dose-response analysis.
Page 357 of 723
-------
3,24.2 Genotoxicity/Carcinogenicity
Overall, given the limited data and mixed results between mammalian and non-mammalian systems,
there is indeterminate evidence to make a conclusion on the genotoxicity of HBCD.
The only experimental animal study to examine cancer endpoints concluded that HBCD was not
carcinogenic, however, this study was only available as an incomplete report (Kurokawa et al. 1984).
Therefore, according to the U.S. EPA Guidelines for Carcinogen Risk Assessment ( £005).
there is "inadequate information to assess the carcinogenic potential" of HBCD. Despite the limited
evidence, it is unlikely that the results of any potential additional studies would significantly alter the
conclusions about the hazard due to the mixed results and the negative incomplete report. As a result,
this hazard was not carried forward for dose-response analysis.
3,2,4,3 Summary of Human Health Hazards Used to Evaluate Acute and Chronic
Exposures
The EPA considered adverse effects for HBCD across organ systems. A comprehensive systematic
review table can be found [Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Systematic
Review Supplemental File: Data Quality Evaluation of Human Health Hazard Studies (U.S. EPA.
2019n)l. The full list of human health effects was screened to those that are relevant, sensitive, and
found in multiple studies. The HBCD human health hazard systematic review process screened 1,890
studies and obtained 53 studies that were relevant and applicable to the PECO statement. Only two of
these studies were unacceptable based on data evaluation criteria. The remaining database of 51 studies
included epidemiological studies that examined associations between HBCD exposure and endpoints
related to effects on the thyroid, nervous system, and female reproductive system as well as repeat-dose
experimental animal studies examining dose-responses for the endpoints of thyroid effects, liver effects,
male and female reproductive effects, developmental toxicity, neurotoxicity, and immunotoxicity. EPA
additionally considered data on toxicity following acute exposures, irritation, sensitization, genotoxicity,
and carcinogenicity. From these effects, the EPA selected endpoints supported by the weight of the
scientific evidence for non-cancer adverse outcomes that were amenable to quantitative analysis for
dose-response assessment as discussed in more detail below in Section 3.2.5 In the following sections,
the EPA identifies the appropriate toxicological studies to be used for acute and chronic exposure
scenarios.
3.2.5 Dose-Response Assessment
3.2.5.1 Selection of Studies for Non-Cancer Dose-Response Assessment
As discussed in Section 3.2.4, studies in humans were not adequate to support conclusions regarding the
relationship between HBCD exposure and effects on the thyroid, male reproduction, or nervous system,
and accordingly do not support dose-response analysis. In the absence of adequate human data, animal
toxicity studies were used for dose-response analysis.
The EPA evaluated data from studies described above (Section 3.2.3.1) to characterize the dose-
response relationships of HBCD and selected studies and endpoints to quantify risks for specific
exposure scenarios. One of the additional considerations was that the selected key studies had adequate
information to perform dose-response analysis for the selected Points of Departure (PODs). A POD can
be the 95% lower bound of the benchmark dose (BMDL) for an estimated incidence based on a
designated change in response level (BMR) or a NOAEL/LOAEL for an observed incidence or change
in the level of response.
Page 358 of 723
-------
Based on the WOE evaluation, four health effect domains were selected for non-cancer dose-response
analysis: (1) thyroid; (2) liver; (3) female reproductive; and (4) developmental. These hazards have been
carried forward for dose-response analysis. While there is also evidence to support nervous system
toxicity following exposure to HBCD during development (and this is a likely downstream outcome of
thyroid hormone deficiency), these data sets were insufficiently robust to support dose-response
analysis. Data sets for male reproductive effects, adult neurological effects, immune system effects,
genotoxicity, and cancer were also not carried forward for dose-response analysis. For a complete
discussion, see Section 3.2.4.
Studies that evaluated each of the four health effect domains were identified in Section 3.2.3, and are
considered in this section for dose-response analysis. In order to identify studies for dose-response
analysis, several attributes of the studies were reviewed. Preference was given to studies using designs
reasonably expected to detect a dose-related response. Chronic or subchronic studies are generally
preferred over studies of less-than-subchronic duration for deriving chronic and subchronic reference
values. Studies with a broad exposure range and multiple exposure levels are preferred to the extent that
they can provide information about the shape of the exposure-response relationship. Additionally,
studies that can reliably measure the magnitude and/or degree of severity of the effect are preferred.
Experimental animal studies considered for each hazard and effect were evaluated using systematic
review study quality considerations discussed in the Systematic Review Methods section. Only studies
that scored an acceptable rating in data evaluation were considered for use in dose-response assessment.
For HBCD, all evaluated repeated-dose studies that were considered acceptable received a medium or
high rating in data evaluation {Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD),
Systematic Review Supplemental File: Data Quality Evaluation of Human Health Hazard Studies (U.S.
EPA. 2019n). In addition to the data quality score, considerations for choosing from among these studies
included study duration, relevance of study design, and the strength of the toxicological response.
Details on these considerations for each endpoint are provided below. For all endpoints other than liver
toxicity, (Etna et al. 2008) was considered the best study for dose-response assessment. The study was
an OECD Guideline 2-generation reproductive toxicity study and scored a high in data evaluation. The
90-day repeat-dose oral study fWIL Research 2001) also scored a high and was additionally considered
for use in dose-response assessment only for the liver toxicity endpoint. See Section 3.2.5.2 for a more
detailed explanation of EPA's basis for selection of these studies and derivation of PODs for each
endpoint.
Given the different HBCD exposures scenarios considered (both acute and chronic), different endpoints
were used based on the expected exposure durations. For non-cancer effects and based on a WOE
analysis of toxicity studies from rats, risks for developmental effects including developmental disruption
of thyroid hormone homeostasis that may result from a single exposure were evaluated for both acute
(short-term) exposures and chronic (long-term, repeated/continuous) exposures, whereas risks for other
adverse effects (e.g., thyroid toxicity, liver toxicity, and female reproductive toxicity) were evaluated
only for chronic exposures to HBCD. Although developmental studies typically involve multiple
exposures, these studies are considered relevant for evaluating single exposures when the adverse effect
may plausibly result from a single exposure during a critical window of development (Davis et al. 2009;
Van Raaii et al. 2003b). This is consistent with EPA's Guidelines for Reproductive Toxicity Risk
Assessment ( M) and Guidelines for Developmental Toxicity Risk Assessment (U.S. EPA.
1991). which state that repeated exposure is not a necessary prerequisite for the manifestation of
developmental toxicity.
Page 359 of 723
-------
While there is uncertainty whether postnatal effects such as neonatal pup loss and decreased body
weight can result from single developmental exposures, there is increased risk following acute exposures
for HBCD, which is a persistent and bioaccumulative toxicological agent with a long half-life. Unlike
many other chemicals with short half-lives (on the order of hours or less), HBCD has a derived
elimination half-life as high as 64 days in humans (Gever et al. 2004). indicating that even a single
exposure may result in a retained body burden for an extended period of time. Consequently, in this Risk
Evaluation EPA concluded that single or acute exposures to HBCD could result in detrimental and
potentially irreversible effects on postnatal growth and viability, while acknowledging that risk for these
endpoints is dependent on the specific timing of exposure. There is strong evidence that HBCD can
reduce thyroid hormone levels in pregnant rats (Etna et al. 2008) and evidence from other thyroid
disruptors suggests that acute or short-term exposure can result in thyroid hormone effects (Paul et al.
2010; Hedge et al. 2009; Zhou et al. 2001). including in weanlings. These changes would presumably
result in downstream effects on developmental endpoints (Forhead and Fowden 20 i I; < lilbert and
Zoeller 2010; Hulbert 2000). Using the developmental endpoints as acute PODs is a health protective
approach as it takes the results from a chronic two-generation study, where exposures lasted throughout
pregnancy of the animal through weaning and sexual maturity. EPA also assumes that a single acute
exposure could lead to the same effects if that exposure occurs during a critical window within the
pregnancy term. Nonetheless, this approach has a biologically supported basis.
3.2.5.2 Derivation of Points of Departure and Uncertainty Factors
A set of dose-response models were applied to empirically model the dose-response relationship in the
range of the observed data. The models in EPA's Benchmark Dose Software (BMDS, version 2.6) were
applied. Consistent with EPA's Benchmark Dose Technical Guidance Document (U.S. EPA. 2012a). the
benchmark dose (BMD) and 95% lower confidence limit on the BMD (BMDL) were estimated using a
benchmark response (BMR) to represent a minimal, biologically significant level of change, when
possible. The BMR is represented by a specified amount of change, or relative deviation (RD), for
continuous data. The BMR for dichotomous data is represented by a specified incidence, or extra risk
(ER). In the absence of information regarding the level of change that was considered biologically
significant, a BMR of 1 standard deviation (SD) from the control mean for continuous data or a BMR of
10% ER for dichotomous data was used to estimate the BMD and BMDL, and to facilitate a consistent
basis of comparison across endpoints, studies, and assessments. Endpoint-specific BMRs are described
further below. Where modeling was feasible, the estimated BMDLs were used as points of departure
(PODs); the PODs are summarized in Table 3-9. Further details, including the modeling output and
graphical results for the model selected for each endpoint, can be found in Appendix I and [Risk
Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental Information on Human Health
Hazard (U.S. EPA. 2019e). Where dose-response modeling was not feasible, NOAELs or LOAELs were
also identified and are summarized.
Selecting the model to use for POD computation
The following approach is recommended for selecting the model(s) to use for computing the BMDL to
serve as the POD for a specific dataset according to EPA Benchmark Dose Guidance (U.S. EPA. 2012a).
This guidance was followed for HBCD BMD modeling analysis.
a) Assess goodness-of-fit, using a value of a = 0.1 to determine a critical value (or a = 0.05 or
a = 0.01) if there is reason to use a specific model(s) rather than fitting a suite of models.
b) Further reject models that apparently do not adequately describe the relevant low- dose
portion of the dose-response relationship, examining residuals and graphs of models and data.
Page 360 of 723
-------
c) As the remaining models have met the recommended default statistical criteria for adequacy
and visually fit the data, any of them theoretically could be used for determining the BMDL.
The remaining criteria for selecting the BMDL are necessarily somewhat arbitrary and are
suggested as defaults.
d) If the BMDL estimates from the remaining models are sufficiently close (given the needs of the
assessment), reflecting no particular influence of the individual models, then the model with the
lowest AIC may be used to calculate the BMDL for the POD. This criterion is intended to help
arrive at a single BMDL value in an objective, reproducible manner. If two or more models
share the lowest AIC, the simple average or geometric mean of the BMDLs with the lowest AIC
may be used. Note that this is not the same as "model averaging", which involves weighing a
fuller set of adequately fitting models. In addition, such an average has drawbacks, including
the fact that it is not a 95% lower bound (on the average BMD); it is just the average of the
particular BMDLs under consideration (i.e., the average loses the statistical properties of the
individual estimates).
e) If the BMDL estimates from the remaining models are not sufficiently close, some model
dependence of the estimate can be assumed. Expert statistical judgment may help at this point
to judge whether model uncertainty is too great to rely on some or all of the results. If the
range of results is judged to be reasonable, there is no clear remaining biological or statistical
basis on which to choose among them, and the lowest BMDL may be selected as a reasonable
conservative estimate. Additional analysis and discussion might include consideration of
additional models, the examination of the parameter values for the models used, or an
evaluation of the BMDs to determine if the same pattern exists as for the BMDLs. Discussion
of the decision procedure should always be provided.
f) In some cases, modeling attempts may not yield useful results. When this occurs and the most
biologically relevant effect is from a study considered adequate but not amenable to modeling,
the NOAEL (or LOAEL) could be used as the POD. The modeling issues that arose should be
discussed in the assessment, along with the impacts of any related data limitations on the results
from the alternate NOAEL/LOAEL approach.
3.2.5.2.1 PODs for Acute Exposure
Developmental Effects
Acute exposure in humans is defined for occupational settings as exposure over the course of a single 8-
hour work shift and for the general population as a single 24-hour day. Consistent with EPA's
Guidelines for Reproductive Toxicity Risk Assessment, as discussed in Section 3.2.5.1, developmental
toxicity is considered relevant for calculating risks associated with acute occupational or general
population exposure.
Reduced offspring viability is a sensitive endpoint that is considered a marker for developmental
toxicity. A single study reported reductions in postnatal offspring viability (Em a et al. 2008) and was
judged to support dose-response analysis of viability as a measure of developmental effects.
Reduced offspring body weight is a sensitive endpoint that is considered a marker for fetal growth
restriction. Decreased pup body weight was reported in four studies (Mammght et al. 2013; Saegusa et al.
2009; van der Yen et al. 2009; Em a et al. 2008). (Maranghi et al.. ) only used a single dose level.
Observed effects were not consistently dose-responsive in (van der Yen et al. 2009). Additionally, the
magnitude of decreased pup body weight reported by (Ema et al. 2008) was substantially greater than
(Saegusa et al. 2009). Finally, (Ema et al. 2008) examined a larger number of animals per group than
Page 361 of 723
-------
other studies and covered a broader dose range than all other studies except (Saegusa et al. 2009). For
the above reasons, (Ema et al. 2008) was selected for dose-response analysis of pup body weight as a
measure of developmental effects following acute exposures. Table 3-6 summarizes study design
features considered in evaluating the strength of each study that reported changes in pup weight for
purpose of dose-response analysis.
Delayed eye opening is a marker of disrupted developmental maturation. A consistent dose-responsive
increase in delayed opening (based on reductions in eye opening at PND 14) was observed in both male
and female F2 offspring (F1 data was inconsistent) in (Ema et al. 2008). Therefore, this study was
judged to also support dose-response analysis of delayed opening as a measure of developmental effects.
Table 3-6. Study Design Features of Developmental Toxicity Studies
Study reference
Route
Exposure
duration
Number of
dose groups3
Number of
animals/ group
Dose range
(mg/kg-d)
Data
Quality
(Ema et al. 2008)
Diet
Two-
generation
3
13-24 rat litters
10-1,570b
High
(1.0)
(van der Ven et al.
2009)
Diet
One-generation
7
>14 rats
0.1-100
High
(1.2)
(Saegusa et al.
2009)
Diet
Gestation and
lactation (-42
d)
3
10-14 rats/sexc
15-1,505
High
(1.2)
(Maranghi et al.
2013)
Diet
28 days
1
10-15 female mice
199
High
(1.3)
aExcludes the control group.
bDoses differed by sex and generation (see, for example, Table 1-4).
Tor PND 0 data, exact number of animals examined per dose group was unclear based on the published study.
In a study by (van Raaii et al. 2003a) a comparison between repeated and single dose studies across a
range of chemicals showed that the NOAELs and LOAELs for fetal body weight were 2-4 fold lower
than those for single-dose studies, thereby indicating that fetal body weight is more sensitive to repeated
exposures. Body weight reduction in pups is therefore generally most applicable to estimating risks for
chronic exposures (at least for short half-life chemicals). Nonetheless, there remains uncertainty
regarding the applicability of the limited dataset examined in (van Raaii et al. 2003a) to persistent
chemicals with long half-lives such as HBCD. It is uncertain whether the dose-duration relationships
identified in (van Raaii et al. 2003a) for fetal body weight are also applicable to postnatal effects
observed following HBCD exposure. While offspring loss was only observed in the F2 generation (Ema
et al. 2008). suggesting a multigenerational effect (possibly due to increasing bioaccumulation) over
repeated exposures, the data does not exclude the possibility of this effect occurring following acute
exposures during a critical window of development. As discussed in Section 3.2.3.1, evidence from
other thyroid disruptors suggests that acute or short-term exposure can result in thyroid hormone effects
(Paul et al. 2010; Hedge et al. 2009; Zhou et al. 2001). including in weanlings, and these hormonal
changes could result in downstream effects on developmental endpoints (Forhead and Fowden 2014;
Gilbert and Zoeller 2010; Hulbert 2000). Additionally, due to HBCD's long half-life a single exposure
results in a chronic internal dose. Therefore, in order to be health protective given the persistence of
HBCD in the body and the absence of any other usable PODs from other potential acute endpoints (such
as neurotoxicity) for considering acute exposure scenarios, EPA considered the developmental endpoints
Page 362 of 723
-------
of F2 offspring loss, reduced F2 pup body weight, and delayed eye opening as the basis for the dose-
response analysis for acute exposures to HBCD.
Offspring Loss
Increased offspring loss in the F2 generation from the (Etna et al. 2008) study was amenable to BMD
nested modeling, using individual animal data obtained from the study authors (personal
communication, (Makris 2016) with implantation size (number) use as a covariate. Two datasets were
modeled: offspring loss (indicating decreased offspring viability) from implantation through PND 4 and
offspring loss from PND 4 (post-culling) through PND 21. Maternal gestational doses (10, 100, and 995
mg/kg-day) were used to model offspring loss from the implantation through PND 4 dataset and
modeling for the PND 4 post-culling through PND 21 dataset was performed using the maternal
lactational doses (20, 179 and 1,724 mg/kg-day).
From a statistical standpoint, most reproductive and developmental studies with nested study designs
typically support a BMR of 5% extra risk (ER) (U.S. EPA. 2012a). A smaller BMR of 1% ER was used
in this case to address the severity of this endpoint {i.e., offspring loss), in accordance with EPA
Benchmark Dose Guidance ( 312a). which supports use of smaller BMRs for more severe or
"frank" effects. The use of a 1% ER is justified for mortality, because death is clearly not a reasonable
risk for any percentage of the population. For purposes of comparison, a POD based on the NOAEL is
presented in addition to the BMDLoi (see Section 3.2.5.3). The NCTR/Rai and Van Ryzin model was
used for offspring loss from implantation through PND 4 based on selection of the lowest BMDL (see
step 5 in BMD guidance), and the NLogistic model with intra-litter correlation but without the covariate
was used for PND 4 through PND 21 loss based on selection of the lowest AIC (see step 4 in BMD
guidance).
Pup body weight
Changes in F2 pup body weight as reported in the two-generation reproductive toxicity study by (Ema et
al. 2008) were amenable to BMD modeling. A BMR of 5% RD from control mean was applied in
modeling pup body weight changes under the assumption that it represents a minimal biologically
significant response. In adults, a 10% decrease in body weight in animals is generally recognized as a
biologically significant response associated with identifying a maximum tolerated dose; during
development, however, identification of a smaller (5%) decrease in body weight is consistent with the
assumptions that development represents a susceptible lifestage and that the developing animal is more
adversely affected by a decrease in body weight than the adult. In humans, reduced birth weight is
associated with numerous adverse health outcomes, including increased risk of infant mortality as well
as heart disease and type II diabetes in adults (Barker 2007; Reyes and Manalich 2005). The selection of
a 5% BMR is additionally supported by data from (Kavtock et al. 1995) which found that a BMR of 5%
RD for fetal weight reduction was statistically similar to several other BMR measurements as well as to
statistically-derived NOAEL values, however EPA acknowledges the uncertainty in extrapolating this
fetal data to postnatal effects. For these reasons, a BMR of 5% RD was selected for decreased pup
weight. The exponential (M4) model was used for male weanlings based on lowest BMDL (see step 5 in
BMD guidance) and the linear model was used for female weanlings based on lowest AIC (see step 4 in
BMD guidance).
Delayed eve opening
Delayed eye opening data in both male and female offspring of the F2 generation from the (Ema et al.
2008) study were amenable to BMD nested modeling, using individual animal data obtained from the
study authors (personal communication) (Jacobs 2019). Calculated F2 offspring doses (15, 139 and 1360
Page 363 of 723
-------
mg/kg-day) were used to model delayed eye opening. Modeling was performed using the most recent
version of BMDS (v 3.1.1). The only model included in the software for use with nested data is the
nested logistic model. The NCTR model from earlier BMDS versions is scheduled for addition to the
software at some point in the future. The Rai and Van Ryzin model from earlier BMDS versions is no
longer supported. Significant model fit was achieved with intra-litter correlation included, and the
selected covariate of number of implantations did not make a significant difference on model fit. The
male dataset reflected very high levels of uncertainty based on a BMD/BMDL ratio of 16-31, so the
resulting BMDL was not selected as a POD and instead the NOAEL was used. The female dataset
resulted in a BMD/BMDL ratio of only -2.6, and visual inspection of model fit along with review of
scaled residuals confirmed adequate model fit. A BMR of 5% extra risk was selected for similar reasons
stated above for pup body weight, because delayed eye opening is a sensitive marker of potentially
irreversible broader physiological and/or neuromuscular developmental outcomes. The nested logistic
model with intra-litter correlation but without the covariate was selected based on selection of the lowest
AIC (see step 4 in BMD guidance). See Appendix 1.1.4 for BMD modeling results of all developmental
endpoints.
Thyroid hormone effects
As discussed above, evidence from other thyroid disruptors suggests that acute or short-term exposure
can result in thyroid hormone effects (Paul et al. 2010; Hedge et at 2009; Zhou et al. 2001). and these
hormonal changes could result in downstream effects on developmental endpoints (Forhead and Fowden
2.014; Gilbert and Zoeller 2.010; Hulbert 2000). A recent study (O'Shaughnessv et al. 2019)
demonstrated that hypothyroidism over only a 5-day gestational/post-natal window is sufficient to cause
cortical heterotopia in rat offspring, a permanent brain malformation that is associated with epilepsy and
learning disabilities in humans. Therefore, thyroid hormone changes in dams from chronic studies were
considered as adverse for acute exposure scenarios in a developmental context. As described in Section
3.2.4.1.1, while there are some significant differences in the thyroid system between rodent and human
adults, gestational HBCD exposure is likely to result in quantitatively similar developmental outcomes.
Additionally, because rats are more altricial than humans, thyroid maturation (and thyroid hormone-
associated growth and development) proceeds later in rats than humans. Consequently, human offspring
may be more susceptible in utero to many developmental outcomes that were observed only postnatally
in rats (e.g., mortality, reduced body weight).
Changes in maternal serum thyroxine (T4) was selected as the endpoint representative of thyroid effects.
See the full discussion of study selection and BMD modeling considerations for this endpoint in Section
3.2.5.2.2 below. In short, T4 data sets from (Ema et al. 2008) were selected for dose-response analysis.
Only data from female rats was considered for acute exposure scenarios, since gestational effects are of
primary concern. A BMR of 10% RD from control means, supported by the literature on the effects of
thyroid insufficiency in pregnant females and their offspring, was applied in modeling the female T4
data. The exponential (M4) model was selected for derivation of all BMDLs for the thyroid endpoint
based on lowest BMDL for females (step 5 in BMD guidance). Further discussion is provided below in
Section 3.2.5.2.2. See Appendix 1.1.1 for all BMD modeling results on the T4 dataset.
3.2.5.2.2 PODs for Chronic Exposures
Chronic exposure was defined for occupational settings as exposure reflecting a 40-hour work week at 8
hrs/day. Chronic exposure to the general population represents exposure averaged over 24 hours/day,
365 days/year, for the number of years living near a facility (either 13 or 33 years). Non-cancer
endpoints selected as most relevant for calculating risks associated with chronic (repeated) occupational
exposures to HBCD included toxicity to the thyroid, liver, female reproductive, and developmental
effects.
Page 364 of 723
-------
Table 3-12 summarizes the hazard studies and health endpoints by target organ/system that the EPA
considered suitable for the Risk Evaluation of chronic exposure scenarios for HBCD. Key studies in
Table 3-12 are briefly described in Non-Cancer Hazards, Section 3.2.3.1, along with other toxicity and
epidemiological studies. BMD modeling was performed for these endpoints in a manner consistent with
EPA Benchmark Dose Technical Guidance. BMR was selected for each endpoint.
Thyroid hormone effects
Changes in serum thyroxine (T4) was selected as the endpoint representative of thyroid effects based on
the following: (1) changes in T4 were observed in multiple studies; (2) T4 is likely to be the primary
driver of HBCD-mediated thyroid effects; and (3) it is well established that perturbations in T4 are
associated with biologically significant health effects. Specifically, adequate levels of T4 are necessary
for normal growth and development, and altered thyroid homeostasis has the potential to affect
numerous organ systems, including neuronal, reproductive, hepatic, and immune systems (Forhead and
Fowden 2014; Gilbert and Zoeller 2010; Hulbert 2000). Additionally, reductions in maternal T4 during
pregnancy or the early postnatal period are strongly associated with adverse neurological outcomes in
offspring. In humans, mild to moderate maternal thyroid insufficiency is associated with higher risk for
persistent cognitive and behavioral deficits in children (see below).
Based on considerations of study design and magnitude of T4 response, T4 data sets from (Ema et al.
2008) were selected for dose-response analysis. The 2-generation study design used by (Ema et al. 2008)
involved a longer exposure duration and larger group size than (van der Yen et al. 2006). while
inadequate reporting of thyroid hormone measurement methods, small sample sizes, and questionable
control data reduced the confidence in the thyroid hormone results from (WIL Research 2001). Table
3-7 provides an overview of the study designs for those studies reporting T4 levels that were evaluated
for dose-response analysis of thyroid effects.
Table 3-7. Study Design Features of Studies that Examined T4 Levels
Study reference
Route
Exposure
duration
Number of
dose groups3
Number of
animals/
group
Dose range
(mg/kg-d)
Data
Quality
(Ema et al. 2008)
Diet
Two-
generation
3
8 rats/sex
10-1,363a
High
(1.0)
(WIL Research 2001)
Gavage
90 days
3
5-10 rats/sex
100-1,000
Low*
(3)
(van der Ven et al.
2006)
Gavage
28 days
7
4-5 rats/sex
0.3-200
High
(1.3)
aDoses differed by sex and generation
*This study received a calculated score of 1.3 but was manually downgraded to Low for thyroid outcomes.
Specifically, T4 data from F0 male and female rats and from F1 female rats in (Ema et al. 2008) were
used for quantitative analysis. Because the magnitude of response in F1 male rats was smaller than the
response in these generations (by one-third to one-half), T4 data from F1 male rats was not modeled.
Based on the data observed in both humans and animals demonstrating downstream health effects
associated with a reduction of 10% and above in maternal T4 levels (Gilbert et al. 2014; Gilbert et al.
2013; Gilbert 2011; Liu et al. 2010; Auso et al. 2004). a BMR of 10% RD from control mean was
Page 365 of 723
-------
determined to be a minimally biologically significant degree of change when performing BMD
modeling using female rat data. Additionally, one study (Gilbert et al. 2016) demonstrated that mild T4
(<20%) reduction in dams can become amplified in offspring (>45%), resulting in long-lasting
reductions in neurotrophin gene expression leading to learning deficits. The available thyroid literature
does not support identification of a biologically significant change in T4 levels in adult males as
decreases in T4, and more generally thyroid function, have not been conclusively linked to similarly
severe outcomes as in females. Nevertheless, males with depressed T4 values are part of the
subpopulation that experiences thyroid dysfunction and there is no evidence to suggest that they are
consistently more sensitive to T4 changes than females. Consistent with EPA's Benchmark Dose
Technical Guidance Document (U.S. EPA. 2012a). a BMR of one control SD change from the control
mean was applied in modeling T4 data from male rats in the absence of a biological basis for selecting a
BMR. Additionally, a BMR of 10% RD from control means, supported by the literature on the effects of
thyroid insufficiency in pregnant females and their offspring, was also applied in modeling the male T4
data. Under the assumption that differences in thyroid hormone response in male and female rats
exposed to HBCD are not sex-specific but rather a reflection of hormone variability, using a BMR of
10%) RD was also considered appropriate for this dataset. The exponential (M4) model was selected for
derivation of all BMDLs for the thyroid endpoint (based on lowest AIC for males [step 4 in BMD
guidance] and based on lowest BMDL for females [step 5 in BMD guidance]). See Appendix 1.1.1 for
all BMD modeling results on the T4 dataset.
Liver Effects
Although increased liver weight is not adverse on its own, it serves as an effective and sensitive
quantitative indicator for liver toxicity when associated with other toxicological indicators, especially
within a potentially susceptible population. Evidence suggests that HBCD exposure impairs hepatic lipid
homeostasis, potentially through the production of ROS (Section 3.2.4.1.2), however establishing a
dose-response and adverse threshold for these indicators is difficult. Increased liver weight was therefore
selected as the representative endpoint for dose-response analysis of liver effects based on being the
most consistently observed toxicological effect. Increased liver weight was reported in six studies in rats
(Saegusa et al. 2009; Em a et al. 2008; van der Yen et al. 2.006; WIL Research 2001. 1997) and mice
(Maranghi et al. 2013). Increased liver weight was also accompanied by increased hepatocellular
vacuolization in (Maranghi et al. 2013; Saegusa et al. 2009; WIL Research 2001. 1997). hypertrophy in
0 isearch 1997). and inflammation in (Maranghi et al. 2013).
(Em a et al. 2008) con si stently observed increased liver weights in rats across multiple generations (i.e.,
F0, Fl, and F2), lifestages (i.e., postnatal day [PND] 26 offspring and adults), and in both sexes,
particularly at the high dose. Elevated liver weight was also observed along with hepatocellular
vacuolization in both sexes of rats across all dose groups in a 90-day study by (WIL Research 2001).
This study also observed statistically-significant elevated serum gamma-glutamyl transferase (GGT) and
a dose-responsive (non-statistically significant) increase in alanine aminotransferase (ALT), indicators
of liver damage, at the highest dose. Both studies were selected for dose-response analysis because they
provided robust dose-related responses that were consistent across sex and generations (for (Em.a et al.
2008). unlike (Saegusa et al. 2009)) and following longer exposure durations than other studies. Table
3-8 provides an overview of the study designs for those studies reporting relative liver weight that were
evaluated for dose-response analysis.
Page 366 of 723
-------
Table 3-8. Study Design Features of Studies that Examined
Study reference
Route
Exposure
duration
Number of
dose groups3
Number of
animals/ group
Dose range
(mg/kg-d)
Data
Quality
(Ema et al. 2008)
Diet
Two-
generation
3
13-24 rats/sex
10-1,570a
High
(1.0)
(WIL Research
2001)
Gavage
90 days
3
10 rats/sex
100-1,000
High
(1.0)
(van der Ven et al.
2006)
Gavage
28 days
7
4-5 rats/sex
0.3-200
High
(1.3)
(WIL Research
1997)
Gavage
28 days
3
6 rats/sex
125-1,000
High
(1.3)
(Saesusa et al.
2009)
Diet
Gestation and
lactation (-42
d)
3
10 rats/sex
15-1,505
High
(1.2)
(Maranshi et al.
2013)
Diet
28 days
1
10-15 female
mice
199
High
(1.3)
aDoses differed by sex and generation
Jver Weight
Liver effects as reported in the (Ema et al. 2008) and (WIL Research 2001) studies were evaluated using
BMD modeling. Liver weight data from (Ema et al. 2008) were amenable to modeling. For weanling
(PND 26) datasets, the average exposures across gestation and lactation (F1 = 16.5, 168, and 1,570
mg/kg-day; F2 = 14.7, 139, and 1,360 mg/kg-day) were used for modeling because there was no
evidence to indicate whether this effect was the result of prenatal exposure, postnatal exposure, or a
combination of both. The exponential (M4) model was selected for derivation of all BMDLs for the liver
endpoint from (Ema et al. 2008) based on visual fit and lowest AIC (steps 3 and 4 in BMD guidance).
The linear model was additionally applied to data from F1 rat adults. A BMR of 10% RD from the
control mean was applied in modeling relative liver weight changes under the assumption that it
represents a minimal biologically significant change, with liver weight changes considered analogous to
the 10% change in body weight that has been used to identify a maximum tolerated dose. Data on liver
effects derived from (WIL Research 2001) could not be modeled because none of the models provided
adequate fit; therefore, LOAELs were chosen for the PODs derived from these data (step 6 in BMD
guidance).
Female Reproductive Effects
Pregnancy incidence and primordial follicle count were selected for dose-response analysis as endpoints
representative of female reproductive effects. These effects were reported in a two-generation
reproductive toxicity study by (Ema et al. 2008) that included three dose groups in addition to the
control. Pregnancy incidence was measured in two generations with exposure durations ranging from
approximately 13 weeks (F0) to continuous lifetime exposure (Fl); primordial follicle count was only
evaluated in the Fl generation. (Ema et al. 2008). the only study to evaluate effects on pregnancy
incidence and primordial follicle count, was selected for dose-response analysis of these measures of
female reproductive toxicity.
Page 367 of 723
-------
Primordial follicle count
Decreased primordial follicle count as reported in the two-generation reproductive toxicity study by
(Etna et al. 2008) was amenable to BMD modeling. Because primordial follicles are formed during
gestation, the average dose during this critical window was used for BMD modeling. While there is no
consensus regarding the degree of change considered to be adverse, a BMR of 10% RD from control
levels was applied in modeling this endpoint under the assumption that it represents a minimal
biologically significant effect based on what may be considered a reasonably detectable decrease in
follicle number (Heindet 1998). The exponential (M4) model was selected for derivation of all BMDLs
for decreased follicle count based on being the only model with adequate fit (step 1 in BMD guidance).
Pregnancy incidence
In the study by (Ema et al. 2008). the increased incidence of non-pregnancy (indicating reduced female
fertility index) in HBCD-exposed F0 or F1 rats alone was not statistically significant with either
pairwise test (as reported by authors) or Cochran-Armitage trend test (conducted by EPA). Dose-
response curves were shallow and never reached a high response percentage. Nevertheless, EPA
considered this change to be biologically relevant. To increase statistical power and obtain a more
precise estimate of the BMD and BMDL, consideration was given to combining F0 and F1 datasets.
Cochran-Mantel-Haenszel statistics on F0 and F1 data stratified by dose groups were not significant (p =
0.59, a = 0.05), indicating no statistical association between generation and response after adjusting for
dose. Equality of responses in F0 and F1 rats was also not rejected (p > 0.2, a = 0.05) by the Breslow-
Day test for homogeneity of the odds ratios, and their background response percentages were not
detectably different (Fisher's exact, p = 1.00). The results of these statistical tests indicated that F0 and
F1 datasets were compatible for combining.
Despite these tests indicating that the datasets were compatible for combining, EPA determined that the
F0 and F1 data were not truly independent related datasets. Due to HBCD's bioaccumulation over time,
the F1 generation experiences additional continuous exposure compared to F0 animals, and the
statistical tests may not account for this confounder. Therefore, the data for increased incidence of non-
pregnancy was not considered appropriate for combining, and without statistical significance on either
data set alone, the endpoint does not represent a confirmed adverse effect.
Developmental Effects
As described above, developmental effects may result from single as well as repeated exposures at a
developmentally critical period; therefore, decreased pup body weight and decreased viability (Ema et
al. 2008) were the endpoints selected as most relevant to calculating risks associated with developmental
toxicity following chronic as well as acute exposures. A smaller BMR of 1% ER was used in this case to
address the severity of this endpoint (i.e., offspring loss). A BMR of 5% RD from control mean was
applied in modeling pup body weight changes under the assumption that it represents a minimal
biologically significant response.
3.2.5.2.3 Human Equivalent Doses
Human equivalent doses (HEDs) for oral exposures were derived from the PODs according to the
hierarchy of approaches outlined in EPA guidance (' v \ M l:). The preferred approach is
physiologically-based pharmacokinetic (PBPK) modeling. Other approaches can include using
chemical-specific information in the absence of a complete PBPK model. As discussed in Section 3.2.2,
an appropriate toxicokinetic model for HBCD is not available. In the absence of either chemical-specific
models or data to inform the derivation of human equivalent oral exposures, body weight scaling to the
3/4 power (i.e., BW3'4) was applied to extrapolate toxicologically equivalent doses of orally administered
agents from adult laboratory animals to adult humans.
Page 368 of 723
-------
Consistent with EPA guidance (l_J5 ,H1.\ -.QJ.tfX the PODs estimated based on effects in adult animals
were converted to HEDs employing a standard dosimetric adjustment factor (DAF) derived as follows:
BWA mr
DAF =
Where
BWa = animal body weight
BWh = human body weight
Using BWa of 0.25 kg for rats and BWh of 80 kg for humans (U.S. EPA. 2005). the resulting DAF for
rats is 0.24. Applying this DAF to the PODadj identified for HBCD effects in adult rats, a PODhed was
derived as follows:
PODhed = Laboratory animal dose (mg/kg-day) x DAF
BW3/4 scaling was not employed for deriving HEDs for increased relative liver weight in pups, offspring
loss, or decreased pup weight as reported by (Ema et al. 2008) where doses were administered to early
postnatal animals. There is uncertainty as to whether allometric (e.g., BW3'4) scaling, derived from data
in adult animals, holds when extrapolating doses in neonatal animals. This uncertainty arises because of
the absence of quantitative information to characterize the toxicokinetic and toxicodynamic differences
between animals and humans in early lifestages (U.S. EPA. 2011c).
3.2.5.2.4 Uncertainty Factors
Four areas of uncertainty and variability were considered in benchmark MOE derivation, as summarized
below.
A UF for extrapolation from a LOAEL to NOAEL, UFl, of 1 was applied when the POD was based on a
BMDL, and the BMR was selected under the assumption that it represented a minimal biologically
significant response level. A UFl of 1 was applied to offspring loss where the POD was based on a
NOAEL, and a value of 10 was applied to relative liver weight data from (WIL Research 2001) because
the POD was based on a LOAEL.
A subchronic to chronic UF, UFs, was applied to account for the possibility that longer exposure may
induce effects at a lower dose when data are derived from less-than-lifetime exposures. (Ema et al.
2008) is a multigenerational study where the parental generation is exposed for approximately 15-18
weeks and the offspring are exposed for approximately 21-24 weeks. Given HBCD's propensity to
bioaccumulate it is also expected that internal exposure could increase with longer external exposure
durations. For thyroid hormone effects, a UFs of 10 was applied when effects were observed in parental
(F0) animals because exposure was subchronic in duration. UFs was reduced to 1 for PODs for thyroid
effects derived from F1 offspring, which have already experienced bioaccumulation across generations
following up to 42 weeks of chronic exposure. A UFs of 1 was also applied to liver weight and both
reproductive endpoints from (Ema et al. 2008). which incorporate data from the F1 generation, for the
same reasoning. A UFs of 3 was applied for liver effects from (WIL Research 2001). a subchronic 90-
day study. UFs was reduced from 10 to 3 for that endpoint because the feedback interaction between
liver metabolism and the HPT axis along with inconsistently observed histopathological or biochemical
changes in other studies (see Section 3.2.4.1.2) suggests that there may only be limited adversity with
Page 369 of 723
-------
increasing exposure. For pup weight and offspring loss, which are developmental endpoints, a UFs of 1
was applied because the developmental period is recognized as a susceptible lifestage where exposure
during certain time windows during development is more relevant to the induction of developmental
effects than lifetime exposure (U.S. EPA 1991).
With the exception of endpoints measured in neonatal animals, a UF for interspecies extrapolation, UFa,
of 3 (1012 = 3.16, rounded to 3) was applied to all PODs because BW34 scaling was used to extrapolate
oral doses from laboratory animals to humans. Although BW3 4 scaling addresses some aspects of cross-
species extrapolation of toxicokinetic and toxicodynamic processes, some residual uncertainty remains.
In the absence of chemical-specific data to quantify this uncertainty, EPA's guidance on BW3 4 scaling
(U.S. EPA 2011c) recommends the use of a UFa of 3. BW3 4 scaling was not used to derive HEDs for
relative liver weight in weanling rats, decreased pup weight, or offspring loss because of the absence of
information on whether allometric (i.e., body weight) scaling holds when extrapolating doses from early
postnatal animals to adult humans due to presumed toxicokinetic and/or toxicodynamic differences
between lifestages (U.S. EPA 2011c; Hattis et al. 2004). For these developmental endpoints, interspecies
extrapolation was based on administered dose, and an UFa of 10 was applied to account for the lack of
quantitative information to characterize toxicokinetic and toxicodynamic differences between animals
and humans at this lifestage.
An intraspecies UF, UFh, of 10 was applied to account for variability and uncertainty in toxicokinetic
and toxicodynamic susceptibility within the subgroups of the human population that are most sensitive
to the health hazards of HBCD (U.S. EPA 2002). In the case of HBCD, the PODs were derived from
studies that used an inbred rat strain and that is not considered sufficiently representative of the exposure
and dose-response of the most susceptible human subpopulations. In certain cases, the toxicokinetic
component of this factor may be replaced when a PBPK model is available that incorporates the best
available science on variability in toxicokinetic disposition in the human population (including sensitive
subgroups). For HBCD, the available information is insufficient to quantitatively estimate variability in
human susceptibility; therefore, the full value for the intraspecies UF was applied.
3.2.5.3 Points of Departure for Human Health Hazard Endpoints
Table 3-9 summarizes the oral PODs (and sequence of adjustments leading to the derivation of a human
equivalent POD or PODhed) by target organ/system. As described and justified in Section 3.2.5.2, all of
the PODs except for liver toxicity to be used for risk characterization were derived from the two-
generation reproductive toxicity study by (Ema et al. 2008). For liver toxicity, the POD selected for risk
characterization was obtained from (WIL Research 2001). a 90-day oral toxicity study conducted
according to OECD testing guidelines.
Table 3-9. Summary of BMB
Modeling Resuli
ts and Derivation of
TEDs for
HBCD
Endpoint
and
Reference
Species/
Sex
Model8
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
PODadj"
(mg/kg-d)
PODhedc
(mg/kg-d)
Thyroid
Decreased
T4
(Ema et al.
2008)
F0 rats
(Sprague-
Dawley)/
male, adults
Exponential
(M4)
10% RD
23.9
6.99
6.99
1.68
Page 370 of 723
-------
Endpoint
and
Reference
Species/
Sex
Model8
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
PODadj"
(mg/kg-d)
PODhedc
(mg/kg-d)
Decreased T4
(Ema et al.
2008)
F0 rats
(Sprague-
Dawley)/
male, adults
Exponential (M4)
1 SD
101
29.5
29.5
7.08
Decreased
maternal T4
(Ema et al.
2008)
FO rats
(Sprague-
Dawley)/
female, adults
Exponential
(M4)
10% RD
334
93.8
93.8
22.5
Decreased
maternal T4
(Ema et al.
2008)
F1 rats
(Sprague-
Dawley)/female,
adults
Exponential (M4)
10% RD
448
127
127
30.5
Liver'1
Relative liver
weight
(Ema et al.
2008)
F1 rats
(CRL)/male
weanlings, PND
26
Exponential (M4)
10% RD
163
109
109
109
Relative liver
weight
(Ema et al.
2008)
F1 rats
(CRL)/weanling
s, PND 26
Exponential (M4)
10% RD
165
115
115
115
Relative liver
weight
(Ema et al.
2008)
F1 rats (CRL)/,
male, adults
Linear
10% RD
680
573
573
138
Relative liver
weight
(Ema et al.
2008)
F1 rats (CRL)/,,
female, adults
Exponential (M4)
10% RD
569
184
184
44.2
Relative liver
weight
(Ema et al.
2008)
F2 rats (CRL)/,
weanlings
Exponential (M4)
10% RD
215
116
116
116
Relative liver
weight
(Ema et al.
2008)
F2 rats (CRL)/,
weanlings
Exponential (M4)
10% RD
286
166
166
166
Relative liver
weight and
hepatocellula
r
vacuolization
(WIL
Research
)01)
Rats (Sprague-
Dawley)/, male
adults
No model fit
LOAEL = 100 (17% RD liver
weight, 300% RD vacuolization)
100
24
Page 371 of 723
-------
Endpoint
and
Reference
Species/
Model8
BMR
BMD
BMDL
PODadj"
PODhedc
Sex
(mg/kg-d)
(mg/kg-d)
(mg/kg-d)
(mg/kg-d)
Relative liver
weight and
hepatocellula
r
vacuolization
(WIL
Research
Rats (Sprague-
Dawley)/,
female adults
No model fit
LOAEL = 100 (24% RD liver
weight, 200% RD vacuolization)
100
24
2001)
Reproductive
Primordial
follicles
(Ema et al.
2008)
F1 parental rat
(Sprague-
Dawley)/,
adults
Exponential
(M4)
10% RD
10.1
2.87
2.87
0.689
Developmental
Offspring loss
1% ER
109
54.5
54.5
54.5
from
F2 offspring rats
(CRL)/male and
female
5% ER
316
158
158
158
implantation
through PND
4
(Ema et al.
2008)
NCTR/Rai and
Van Ryzin
NOAEL = 100 (-2% ER)
100
100
Offspring
1% ER
16.9
9.03
9.03
9.03
loss from
F2 offspring
rats
(CRL)/male
and female
5% ER
88.1
47.1
47.1
47.1
PND 4 post-
culling
through
PND 21
NLogistic
NOAEL =19.6 (7% ER)
19.6
19.6
(Ema et al.
2008)
Decreased
F2 rats
(CRL)/male
weanlings
pup weight
(Ema et al.
2008)
Exponential
(M4)
5% RD
354
89.6
89.6
89.6
Decreased
pup weight
(Ema et al.
2008)
F2 rats (CRL)/
female
weanlings
Linear
5% RD
417
297
297
297
Delayed eye
opening, F2
rats, female
F2 rats (CRL)/
female
weanlings
NLogistic
5% ER
75.61
28.73
28.73
28.73
weanlings
(Ema et al.
2008)
10% ER
159.62
60.66
60.66
60.66
Delayed eye
opening, F2
rats, male
weanlings
F2 rats (CRL)/
male weanlings
NLogistic
NOAEL =
139 mg/kg (25.5% ER)
139
139
Page 372 of 723
-------
Endpoint
and
Reference
Species/
Sex
Model8
BMR
BMD
(mg/kg-d)
BMDL
(mg/kg-d)
PODadj"
(mg/kg-d)
PODhedc
(mg/kg-d)
(Ema et al.
2008)
a For modeling details, see Appendix I and [Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD),
Supplemental Information on Human Health Hazard (U.S. EPA 2019e)l.
bAll studies involved dietary administration. Therefore, no adjustments to estimate the average daily dose were
required, and BMDL, NOAEL, or LOAEL values were equivalent to the PODadj in all cases.
c PODhed values for endpoints measured in adult animals were calculated using BW3 4 scaling. PODhed values for
endpoints measured in neonatal animals were expressed as administered dose.
dRelative liver weiaht from both (Ema et al. 2008) and (WIL Research 2001) is expressed as a/100 a BW.
Note: Both (Ema et al. 2008) and (WIL Research 2001) scored a Hiah in data evaluation.
Table 3-10 and Table 3-11 are a continuation of Table 3-9, Table 3-10 summarizes the human
equivalent PODs and a breakdown of UFs for each relevant endpoint, leading to the derivation of
benchmark MOEs for the Risk Evaluation of acute exposure scenarios. Table 3-11 provides the same
information for the Risk Evaluation of chronic exposure scenarios.
Table 3-10. PODs and Benchmark MOEs for Effects Following Acute Exposure to HBCD
Endpoint and
reference
Developmental
exposure window
PODhed*
(mg/kg-d)
POD
type
UFl
UFs
UFa
UFh
Benchmark
MOE
Thyroid
Decreased
maternal T4, F0
rats, female
adults
(Ema et al. 2008)
Throughout
gestation and
lactation
22.5
BMDLio
1
1
3
10
30
Decreased
maternal T4, F1
rats, female
adults
(Ema et al. 2008)
Throughout gestation
and lactation
30.5
BMDLio
1
1
3
10
30
Developmental
F2 Offspring loss
(Ema et al.
2008)
Implantation -
PND 4
54.5
158
100
BMDLoi
BMDL05
NOAEL
1
1
10
10
100
F2 Offspring
9.03
BMDLoi
loss
PND 4 - PND 21
47.1
BMDLos
1
1
10
10
100
(Ema et al. 2008)
19.6
NOAEL
Decreased pup
weight, F2 rats,
male weanlings
(Ema et al. 2008)
GD 0 - PND 21
89.6
BMDLos
1
1
10
10
100
Page 373 of 723
-------
Endpoint and
reference
Developmental
exposure window
PODhed3
(mg/kg-d)
POD
type
UFl
UFS
UFa
UFh
Benchmark
MOE
Decreased pup
weight, F2 rats,
female weanlings
(Ema et al. 2008)
GD0-PND21
297
BMDL05
1
1
10
10
100
Delayed eye
opening, F2 rats,
female
weanlings
(Ema et al. 2008)
GD 0 - PND 21
28.73
60.66
BMDLos
BMDLio
1
1
10
10
100
Delayed eye
opening, F2 rats,
male weanlings
(Ema et al. 2008)
GD0-PND21
139
NOAEL
1
1
10
10
100
Table 3-11 POPs and Benchmark MOEs for Effects Following Chronic Exposure to HBCD
Endpoint and reference
PODhed3
(mg/kg-d)
POD type
UFl
UFs
UFa
UFh
Benchmark
MOE
Thyroid
Decreased T4, F0 rats, male
adults
(Ema et al. 2008)
1.68
BMDLio
1
10
3
10
300
Decreased T4, F0 rats, male
adults
(Ema et al. 2008)
7.08
BMDLisd
1
10
3
10
300
Decreased T4, F0 rats, female
adults
(Ema et al. 2008)
22.5
BMDLio
1
10
3
10
300
Decreased T4, F1 rats, female
adults
(Ema et al. 2008)
30.5
BMDLio
1
1
3
10
30
Liver
Relative liver weight, F1 rats,
male weanlings, PND 26
(Ema et al. 2008)
109
BMDLio
1
1
10
10
100
Relative liver weight, F1 rats,
female weanlings, PND 26
(Ema et al. 2008)
115
BMDLio
1
1
10
10
100
Relative liver weight, F1 rats,
male adults
(Ema et al. 2008)
138
BMDLio
1
1
3
10
30
Page 374 of 723
-------
Endpoint and reference
PODhed3
(mg/kg-d)
POD type
UFl
UFs
UFa
UFh
Benchmark
MOE
Relative liver weight, F1 rats,
female adults
(Ema et al. 2008)
44.2
BMDLio
1
1
3
10
30
Relative liver weight, F2 rats,
male weanlings
(Ema et al. 2008)
116
BMDLio
1
1
10
10
100
Relative liver weight, F2 rats,
female weanlings
(Ema et al. 2008)
166
BMDLio
1
1
10
10
100
Relative liver weight and
hepatocellular vacuolization,
rats, male adults
(WIL Research 2001)
24
LOAEL
10
3
3
10
1,000
Relative liver weight and
hepatocellular vacuolization,
rats, female adults
(WIL Research 2001)
24
LOAEL
10
3
3
10
1,000
Relative liver weight and
hepatocellular vacuolization,
rats, female adults
(WIL Research 2001)
24
LOAEL
10
3
3
10
1,000
Reproductive
Primordial follicles, F1
parental rat, female adults
(Ema et al. 2008)
0.689
BMDLio
1
1
3
10
30
Developmental
F2 Offspring loss
(Ema et al. 2008); Implantation
-PND4
54.5
158
100
BMDLoi
BMDL05
NOAEL
1
1
10
10
100
F2 Offspring loss
(Ema et al. 2008): PND 4 -
PND 21
9.03
47.1
19.6
BMDLoi
BMDL05
NOAEL
1
1
10
10
100
Decreased pup weight, F2
rats, male weanlings
(Ema et al. 2008): GD 0 - PND
21
89.6
BMDLos
1
1
10
10
100
Decreased pup weight, F2 rats,
female weanlings
(Ema et al. 2008): GD 0 - PND
21
297
BMDL05
1
1
10
10
100
Page 375 of 723
-------
Endpoint and reference
PODhed3
(mg/kg-d)
POD type
UFl
UFS
UFa
UFh
Benchmark
MOE
Delayed eye opening, F2 rats,
female weanlings
(Ema et al. 2008)
28.73
60.66
BMDLos
BMDLio
1
1
10
10
100
Delayed eye opening, F2 rats,
male weanlings
(Ema et al. 2008)
139
NOAEL
1
1
10
10
100
Table 3-12 lists the PODheds selected for use in risk estimation by target organ/system and exposure
category (i.e., acute vs. chronic). The two studies considered for derivation of PODs both received a
High in data quality evaluation and all derived BMDLs were considered similarly reasonable for use in
risk estimation. Therefore, EPA selected the lowest resulting POD among BMDL modeling results in
order to be health-protective.
Table 3-12. PODs Selected for Risk Estimation for Each Target Organ/System
Toxicity Endpoint
PODhed
(mg/kg-d)
Benchmark
MOE
Effects following acute exposure
Thyroid
Decreased maternal T4 (Ema et al. 2008)
22.5
30
Developmental
F2 generation offspring loss (Ema et al. 2008)
9.03
100
Decreased F2 generation dud weight (Ema et al.
2008)
89.6
100
Delayed F2 generation eve opening (Ema et al.
2008)
28.73
100
Effects following chronic exposure
Thyroid
Decreased T4 (Ema et al. 2008)
1.68
300
Liver
Increased relative liver weight and vacuolization
(WIL Research 2001)
24
1000
Female
Reproductive
Reduced Drimordial follicles (Ema et al. 2008)
0.689
30
Developmental
F2 generation offsnring loss (Ema et al. 2008)
9.03
100
Decreased F2 generation dud weight (Ema et al.
2008)
89.6
100
Delaved F2 generation eve onening (Ema et al.
2008)
28.73
100
Page 376 of 723
-------
3.2.6 Assumptions and Key Sources of Uncertainties for the Human Health Hazard
Assessment
3.2.6.1 Toxicokinetics
In vivo animal studies of the individual isomers have not been conducted. Therefore, it is not possible to
predict whether the toxicity of an environmental HBCD mixture would differ from the toxicity of
commercial mixtures {i.e., those tested in toxicity studies). It is known, however, that the three major
isomers have somewhat different physical/chemical properties (see Section 1.1) and differ
toxicokinetically. For example, the a-isomer accumulates to a greater extent in tissues, especially fat,
when compared to y- or P-HBCD; y- and P-HBCD are more rapidly and extensively metabolized than a-
HBCD (see Section 3.2.2.1.3). Mechanistic studies provide limited evidence of differences in biological
activity of the three. Thus, the composition of HBCD mixtures to which humans are exposed is likely to
differ from the commercial mixtures used in toxicity testing. Whether, and to what extent, the toxicity of
the environmental mixtures differs from the toxicity of the commercial mixtures used to derive the
PODs is not known based on the available health effects literature. Similarly, HBCD toxicokinetics
including absorption and bioaccumulation differ greatly among isomers and are greatly affected by the
relative fat content of tissues and surrounding media {e.g., water, air, diet, breastmilk). For both
consistency and health-protectiveness, these issues were accounted for by utilizing the upper range of
absorption estimates across available studies and including a 10X sub chronic-to-chronic UF based on
assumed increasing bioaccumulation over time. This adjustment was not included for developmental
endpoints or for effects observed following multi-generational exposure, which should already
encompass chronic bioaccumulation. EPA believes that the use of this 10X uncertainty factor is likely to
be protective of risk from bioaccumulation in human tissues, however there is insufficient available data
to confirm this presumption.
EPA utilized data exclusively from oral studies in developing PODs. While it is assumed that any
inhaled particulates will be either absorbed through the lung or swallowed and absorbed in the GI, there
could be potentially significant differences metabolic outcomes between these routes. Similarly, oral
data was extrapolated for evaluating dermal exposure. The absence of a usable PBPK model to
quantitatively account for differences between routes represents an important uncertainty when
considering the application of oral PODs to other exposure routes.
EPA assumed an upper-end dermal absorption estimate of 6.5% based on a steady-state value from in
vitro data following 24hr HBCD exposure as a thin, evenly distributed layer on skin. The actual
percentage of HBCD absorbed dermally is variable based on multiple factors including the relative
percentage of each isomer in the mixture and the relative ratio of sweat to sebum on skin. This value
likely overestimates average dermal absorption when accounting for other factors such as washing or
wiping skin clean and uneven distribution along the skin surface area. Additionally, the test data
involves HBCD dissolved in acetone, where HBCD is much more soluble than in water.
3.2.6.2 Human Health Endpoints
PODs were derived from two studies, (Ema et al. 2008) and (V search 2001). These studies were
selected because they both scored high in data evaluation, followed OECD guidance and Good
Laboratory Practice, and were of longer duration with effects observed more consistently than other
high-quality studies that we evaluated. PODs were derived from these studies using BMD modeling
when possible in order to obtain more precise values. BMD modeling results always contain some level
of uncertainty, and various factors such as model fit and BMR selection may have a large effect on the
final POD value.
Page 377 of 723
-------
Endpoints for Acute Exposures
EPA considered the two developmental toxicity endpoints to be applicable to acute exposures. There is
uncertainty surrounding this consideration because the precise critical exposure window is unknown and
it is unknown how well the two generational rodent study predicts acute effects in humans. EPA
determined that the sustained persistence of HBCD in human tissue suggests that a single exposure
could have sustained effects. Therefore, despite the uncertainties, neonatal mortality and body weight
reduction were considered relevant to acute exposures. Offspring loss represents the most severe
endpoint representing the developmental toxicity hazard and is also the lowest available POD relevant to
acute exposures, thus making EPA's approach health protective. EPA also considered maternal
decreases in T4 levels for acute exposure scenarios, because short-term changes in thyroid hormones
may result in irreversible developmental outcomes such as neurotoxicity and other effects. There are no
available studies examining acute developmental HBCD exposure, however there is evidence of acute
developmental neurotoxicity (Sections 3.2.3.1.5 and 3.2.4.1.5) and evidence from other thyroid
disruptors suggests that acute or short-term exposure can result in thyroid hormone effects (Paul et al.
2010; Hedge et al. 2009; Zhou et al. 2001). Therefore, it is assumed that decreased maternal T4 can
serve as a sensitive quantitative measure of potential developmental effects that cannot otherwise be
quantified.
Endpoints for Chronic Exposures
The available information on weight of evidence and HBCD mode of action suggests that most if not all
HBCD human health hazard endpoints are downstream of dysregulation of the hypothalamic-pituitary-
thyroid axis as indicated by decreased T4 levels. Therefore, in addition to representing the lowest
available POD, changes in T4 thyroid hormone levels were identified as the most important endpoint
relevant to chronic exposures. There is some uncertainty over the use of rodent thyroid hormone data for
quantitative human health risk assessment, however the complexity of the system makes it difficult to
determine whether rodents would in fact be more sensitive to the specific effects of HBCD. Direct
extrapolation of adult rodent thyroid hormone effects to adult humans and use of a 10% BMR is health-
protective and may potentially overestimate risk to human adults. However, developmental effects of
thyroid disruptors following gestational exposure are expected to be highly comparable between rats and
humans, with substantially increased susceptibility in developing individuals of both species compared
to adults. While there are some significant differences in the thyroid system between rodent and human
adults, gestational HBCD exposure is likely to result in quantitatively similar developmental outcomes.
Therefore, there is reduced concern about overestimation when considering thyroid hormone changes as
a biochemical marker of downstream developmental toxicity.
No BMD model provided adequate fit to the data from fWIL Research 2001) and therefore a LOAEL
value was used, introducing additional uncertainty in the form of a large cumulative uncertainty factor
and benchmark MOE. This is likely to overestimate risk for that endpoint due to the large default values
used for various uncertainty factors. Nonetheless, EPA believes that the selected PODs best represent
the hazards associated with HBCD for quantitative risk estimation. The liver POD from fWIL Research
2001) is still less protective than the thyroid effects POD from (Ema et al. 2008). so its inclusion does
not significantly impact the risk conclusions.
Additionally, EPA determined that there was evidence to support potential nervous system effects
following HBCD exposure, however limitations in the available data precluded use of any particular
study for dose-response analysis of the hazard. Nonetheless, other more sensitive endpoints such as
thyroid hormone changes are expected to be protective of neurotoxicity and any other qualitative health
effects. Overall, there is medium confidence in all endpoints applicable to chronic exposure, including
Page 378 of 723
-------
the most sensitive endpoint of thyroid effects. There is additionally some uncertainty in the evaluation of
inhalation hazards due to the lack of reasonably available subchronic or chronic inhalation studies. The
14-day study by (Song et at 2016) only performed gross pathological examination of organs and did not
closely examine respiratory-specific indications of toxicity such as measuring bronchoalveolar lavage
fluid (BALF).
3.2.7 Potentially Exposed or Susceptible Subpopulations
TSCA requires that a Risk Evaluation "determine whether a chemical substance presents an
unreasonable risk of injury to health or the environment, without consideration of cost or other non-risk
factors, including an unreasonable risk to a potentially exposed or susceptible subpopulation (PESS)
identified as relevant to the Risk Evaluation by the Administrator, under the conditions of use." TSCA
Section 3(12) states that "the term 'potentially exposed or susceptible subpopulation' means a group of
individuals within the general population identified by the Administrator who, due to either greater
susceptibility or greater exposure, may be at greater risk than the general population of adverse health
effects from exposure to a chemical substance or mixture, such as infants, children, pregnant women,
workers, or the elderly." In developing the hazard assessment, EPA evaluated available data to ascertain
whether some human subpopulations may have greater susceptibility than the general population to the
chemical's hazard(s). As discussed further, EPA identified the following susceptible groups: pregnant
women, women of reproductive age who may become pregnant, developing fetus and breastfed infants,
postnatal infants, infants, obese individuals or those on a high-fat diet, and individuals with pre-existing
health conditions or genetic predispositions.
Early lifestages are potentially susceptible to HBCD exposure. HBCD is widely detected in breast milk
and umbilical cord serum, indicating a strong potential for prenatal and lactational exposure (Fangstrom
et al. 2008; Kakimoto et al. 2008; Meiier et al. 2008; Fanestrom et al. 2005). Additionally, HBCD has
been detected in placenta and fetal liver tissue (Rawn. et al.. 2014a).
In animal studies, HBCD exposure resulted in thyroid alterations. Thyroid hormones play a critical role
in coordinating complex developmental processes, and perturbations of thyroid hormone levels in a
pregnant woman or neonate can have persistent adverse health effects for the child (Zoeller et al. 2007)).
including adverse neurological outcomes (Finken et al. , 1 uhez et al. 2013; Roman et al. 2013;
Henrichs et al. 2010; Haddowetal. 1999). During early gestation, the developing fetus relies solely on
thyroid hormones of maternal origin. As the fetus begins to produce thyroid hormones, there is less
reliance on maternal thyroid hormones; however, early development remains a sensitive life stage for
hormone deficits, largely due to minimal reserve capacity when compared to adults (Gilbert and Zoeller
2010). Effects on female reproduction parameters are an additional consideration for identifying
pregnant and lactating females as a susceptible subpopulation.
Some gender-specific differences in distribution, metabolism, and elimination of HBCD have been
noted in animals. A toxicokinetic study in rats administered a single oral dose of [14C]-HBCD found that
males had faster elimination rates and lower tissue concentrations when compared to females (Yu and
Atall 3). These data are consistent with observations that female rats had higher liver
concentrations of HBCD following repeated oral exposure for 28 days (van d et al. 2006) or
following gestational, lactational, and dietary exposure (van der Yen et al. 2009). Measures of
mechanistic endpoints provide limited evidence of gender-specific responses to HBCD. For example,
(Germer et al. 2006) reported significant induction of CYP3A1/3 mRNA and the associated proteins in
both sexes of rats exposed to HBCD for 28 days, but the effect was greater and occurred at lower doses
in females (doses of >3 mg/kg-day in females and >30 mg/kg-day in males). In another 28-day study,
female rats exposed to HBCD had, overall, a significantly higher number of up- or down-regulated
Page 379 of 723
-------
hepatic genes than males (Canton et al. 2.008); however, genes involved in phase I and II metabolism
were up-regulated predominantly in males. In vivo toxicity studies, however, do not show a clear pattern
of sex-specific toxicity associated with HBCD exposure (for non-reproductive/developmental
endpoints). It is therefore unclear whether either males or females are more biologically susceptible to
HBCD toxicity on non-reproductive/developmental endpoints.
HBCD is preferentially deposited in adipose tissue, especially the a-HBCD isomer (see Section 3.2.2).
The bioaccumulative nature of HBCD suggests that individuals who consume a high-fat diet may be at
increased risk for HBCD toxicity. Additionally, individuals with higher body fat content may also be at
greater susceptibility to HBCD. This is corroborated by multiple studies demonstrating increasing liver
toxicity in mice administered a high-fat diet (Semihard et al. 2016; Yanaeisawa et al. 2014). Specific
preexisting conditions that may result in increased liver fat content include obesity, metabolic disease,
hypercholesterolemia, non-alcoholic fatty liver disease, alcoholic liver disease, and Hepatitis B or C
viral infections. Higher body fat content will also lead to increasing body burden, leading to increased
toxicity over time as HBCD is distributed from fat to other tissues.
Humans with pre-existing health conditions or genetic predispositions related to any of the affected
health domains (e.g., thyroid, liver, reproductive, neurological, immune) would also be expected to be
especially susceptible to HBCD toxicity, perhaps at significantly lower doses than healthy populations.
Page 380 of 723
-------
4 RISK CHARACTERIZATION
4.1 Environmental Risk
The environmental risk characterization of HBCD was conducted to evaluate whether the potential
releases of HBCD into various media types will exceed the HBCD concentrations observed to result in
hazardous effects to aquatic and terrestrial organisms. In evaluating the environmental hazard of HBCD,
a weight of evidence approach was used to select hazard effect concentrations for the derivation of risk
quotients for both aquatic and terrestrial organisms. The selected hazard effect concentrations reflect
studies with high data quality evaluation scores (as determined by the Systematic Review Metrics for
Environmental Toxicological Studies), where measured discrete exposure concentrations resulted in
observed effects due to acute and chronic exposures. Algal hazard thresholds were separated from being
categorized as an acute or chronic exposure because the relatively shorter study durations used for algal
toxicity tests measure toxicological effects (e.g., growth, reproduction) that are typically associated with
chronic effects. Concentrations of concern (COCs) are summarized below in Table 4-1 for aquatic
organisms. Hazard thresholds are summarized below in Table 4-2 for terrestrial organisms. COCs or
toxic reference values (TRVs) were not derived for terrestrial organisms because the general limitations
of available HBCD data for terrestrial organisms results in an inability to derive appropriate assessment
factors that address uncertainties due to duration, field-to-lab extrapolations, and endpoint-specific
modes of actions and implications. Finally, the environmental hazard studies used to derive hazard
thresholds and COCs were based on high data quality, measured hazard effects concentrations below the
water solubility limits of HBCD, and data relevant to the exposure pathway of interest.
As described in Section 2.2, EPA assessed releases of HBCD to the environment based on the
production volume of HBCD, emission factors, and number days of release per year. In a few cases,
EPA used TRI release data in lieu of the production volume of HBCD and emission factors. A two-
tiered modeling approach was used to predict both surface water and sediment HBCD concentrations
using two models, E-FAST (surface water) and the PSC (surface water and sediment). Briefly, E-FAST
was used for all conditions of use where water releases were predicted to occur. If the E-FAST
predicted 7Q10 surface water concentrations were greater than the chronic or acute COCs, the PSC
model was then used to confirm whether the predicted surface water concentration exceeds the chronic
or acute COC. While both E-FAST and PSC consider dilution and variability in flow, the PSC model
can further estimate a time-varying surface water concentration, partitioning to suspended and settled
sediment, and degradation within compartments of the water column.
As explained in Section 2.3, EPA used Standard Industrial Classification (SIC) codes to determine
industry-specific dilution factors and stream flows to predict surface water and sediment HBCD
concentrations. In lieu of having site-specific release information for HBCD, EPA used SIC code
information to determine 10th and 50th percentile flow rates to crosswalk with specific COUs. Surface
water releases for each exposure scenario were utilized to estimate surface water concentration using
flow values from both the 10th and 50th percentile facility for the SIC code. The 10th percentile flow
values are approximately a factor of 10 lower than the 50th percentile flows for the SIC codes chosen
(lower flow volume will result in higher predicted concentrations of HBCD in the surface water and
sediment). The 10th and 50th percentile facilities were estimated in the Risk Evaluation to account for the
variability in receiving stream flows (all risk estimates are provided in Appendix J).
As described in Section 2.3, to assess the estimated release of HBCD via air deposition from specific
exposure scenarios, IIOAC was used to provide an estimated concentration of HBCD that could be in
soil via air deposition in both fenceline (less than 100 m from an industrial facility) and community
Page 381 of 723
-------
(100-1,000 m from an industrial facility) scenarios. Although the IIOAC was also applied to a generic
farm pond setting to calculate concentrations of HBCD in pond surface water and pond sediment, only
the soil concentrations resulting from air deposition were used. Estimated surface water and sediment
HBCD concentrations using the IIOAC were not used because as compared to E-FAST or the PSC,
IIOAC is a simpler model, providing a two-compartment (surface water and sediment) concentration of
HBCD with no accounting for media exchange of the chemical of interest or partitioning to other
suspended solids in the surface water.
In addition to modeling, environmental monitoring and biomonitoring data was reviewed, and screened
to assess wildlife exposure to HBCD. The key studies that were reviewed and used for the
environmental exposure assessment are summarized in Section 2.3.1. Environmental monitoring data
summarized below in Table 4-3 and Table 4-4 demonstrate that the predicted surface water and
sediment HBCD concentrations using both E-FAST and the PSC support measured HBCD
concentrations near industrial facilities in most modeled exposure scenarios, except for exposure
scenario 12 (Use of Flux or Solder Pastes). For exposure scenario 12, all predicted releases of HBCD
are below the concentrations of HBCD that have been measured in surface water and sediment near
industrial facilities, yet some surface water concentrations based on the 7Q10 50th percentile
predictions are greater than the measured surface water concentrations of HBCD found near general
populations (Vernier et at. 2014).
Incorporating both environmental monitoring and predicted environmental concentrations of HBCD
provides information that can be used to evaluate exposure scenarios within each COU. Environmental
monitoring data cannot provide HBCD release information that can be attributed to a specific exposure
scenario or exposure scenario-specific parameter, nor can it be used to determine HBCD releases from
a specific time period {i.e., historic or current releases). However, the incorporation of measured
environmental monitoring data does provide context for the persistence of HBCD in the environment,
despite recently observed reductions in HBCD production and use. Environmental monitoring data also
provides insight regarding how previous releases of HBCD may also contribute to the current
environmental exposures of HBCD.
Modeled HBCD surface water and sediment concentrations were obtained by using information that is
specific to an exposure scenario or that pertains to an industrial or commercial sector that is related to
an exposure scenario {e.g., polymer processing, use of spray polyurethane foam). Modeled HBCD
surface water and sediment concentrations however can only be attributed to the assessed releases in
the case of each exposure scenario. Although HBCD is expected to partition out of the water column
quickly, thereby reducing exposure for pelagic organisms, modeled HBCD surface water and sediment
concentrations also do not account for the bioavailability of HBCD to pelagic organisms due to the
presence of suspended solids {i.e., resuspension of sediment, presence of natural organic matter).
Therefore, predicted surface water concentrations used to characterize risk from surface water releases
associated with current conditions of uses may underestimate exposure to HBCD for pelagic
organisms.
To characterize environmental risk due to historical activities (as explained in Section 1.2.9), monitoring
approaches were used to evaluate exposure. Monitoring information likely encompasses HBCD releases
from both historical and ongoing conditions of use and it is difficult to ascertain what proportions may
be due to any specific release at a specific time period or geographical location. Risk estimates for
background exposure would therefore be expected to incorporate exposures from any and all potential
historical uses that may have resulted in releases to the environment, and inclusion of any historical
Page 382 of 723
-------
COUs in the Risk Evaluation does not result in additional environmental exposures beyond what was
previously assessed based on monitoring data. Chronic exposures are being evaluated for historical
activities of HBCD primarily because of the persistent, bioaccumulative and toxic (PBT) nature of
HBCD, and the lack of information regarding the explicit releases from historical uses and the
potentially resulting acute exposures. However, it is due to these unique PBT characteristics of HBCD,
that EPA acknowledges the likelihood that historical uses of HBCD may contribute to current HBCD
exposures.
As stated in Section 2.2.14, EPA performed a sensitivity analysis for three conditions of use using the
per site process volumes of 50,000 lbs/yr and 25,000 lbs/yr to examine the effect of process volume on
modelled environmental exposures. Due to HBCD declining use, EPA did not identify a current import
volume for HBCD, and conservatively used the CDR reporting threshold for small firms of 100,000
lbs/yr as explained in Section 1.2.3. If import is occurring at all, the current import volume could be
lower than the threshold volume of 100,000 lbs/yr. For select conditions of use, EPA assessed the most
recently identified import volume in 2017 of-50,000 lbs/yr (see Table 1-4) and to account for the
declining use of HBCD, EPA also considered 25,000 lbs/yr. The selected conditions of use
(Repackaging of Import Containers, Manufacturing of XPS foam from XPS masterbatch, and
Manufacturing of EPS foam from EPS resin) considered in the sensitivity analysis represent conditions
of use that are expected to result in high surface water and sediment concentrations.
KABAM (vl) predictions of HBCD bioavailability through diet and water are also used to categorize
exposure and predicts body burdens and the contribution to body burden due to both diet and media
exposure. Predicted bioaccumulation, bioconcentration and biomagnification factors can also be
predicted for representative organisms within each trophic level. American kestrel and Sprague Dawley
rats are used as proxy organisms for terrestrial avian and mammalian wildlife organisms, respectively,
that may be exposed to HBCD through trophic transfer and various media exposure. Specifically, for
this model, based on the assumption that the modeled organisms will experience the same hazardous
effect as those of the proxy organisms, hazard data on the proxy organisms are also input parameters for
KABAM. Both the predicted hazard effect concentrations and exposure to HBCD through diet and
media exposures are used to calculate risk estimates for mammal and avian species within multiple
trophic levels.
For the estimation of environmental risk to wildlife via trophic transfer (dietary exposure), the hazard
thresholds and environmental exposure data (e.g., media or tissue HBCD concentrations) selected were
based on studies that have been evaluated through the Systematic Review Process; the hazard effect
concentrations and environmental media or tissue concentrations of HBCD were evaluated as high
quality studies using Systematic Review Environmental Hazard or Exposure Metrics, respectively. As
noted previously, one of the constraints to characterizing dietary exposure and estimating HBCD trophic
transfer in aquatic and terrestrial ecosystems is having hazard data where the exposure regime and
methodology used to quantify chemical uptake used is compatible with the available monitoring data.
This evaluation of environmental risk resulting from trophic transfer is limited to the available data and
exposure factors, therefore risk quotients were only calculated for kestrel, osprey (via allometric scaling
from Kestrel reproductive toxicity data), rainbow trout and earthworms because dietary exposure to
HBCD was available (Fertile et al. 201 I; \tifderheide et al. 2003; Wildlife Intl 1997a)
There are many potential sources of uncertainty in all of the parameters involved in environmental
exposure estimates. As presented in Table 2-114, the greatest influence on exposure estimates given the
associated uncertainty and sensitivity (effect on the final values) stems from the selection of emission
Page 383 of 723
-------
factor and days of release. Production volume is highly uncertain but not very sensitive, while other
factors such as physical-chemical properties, BAF, HBCD half-lives, and exposure model parameters
were all estimated to contain low uncertainty. In order to account for these uncertainties and variability
among release estimates and exposure considerations including wastewater treatment, EPA provided
risk estimates based on a range of exposure sub-scenarios. EPA believes that these sub-scenarios
sufficiently capture the range of risk estimates for all reasonably expected environmental exposures,
with minimal remaining unaccounted for uncertainty. Therefore, EPA has high confidence in the range
of risk estimates for the highly-exposed aquatic and terrestrial organisms.
4.1,1 Environmental Risk Estimation
The environmental risk of HBCD is characterized by calculating risk quotients (RQs) (U.S. EPA.
1998a; Barnthouse et at.. 1982). The concentrations of concern (COCs) derived from hazard data are
used to calculate RQs for aquatic organisms. The hazard effects concentrations are used to calculate
RQs for terrestrial organisms (COC calculation methodologies, specified below, were not originally
meant for terrestrial organisms). The environmental concentration for each compartment {i.e.,
wastewater, surface water, sediment, soil) is based on measured and/or modeled concentrations of
HBCD.
4.1.1.1 Environmental Effect Levels of HBCD
The methods for calculating the environmental concentrations of concern (COCs) are based on
published EPA methods ( HO 13 a; 2012d). As described above, the selection of hazard effect
concentrations was based on a weight of the scientific evidence approach that takes into consideration:
data evaluation quality scores, relevancy of exposure and effect measured, and the availability of
supporting studies.
The environmental hazard evaluation that is summarized in Section 3.1 of this evaluation is based on
high quality studies. The algal hazard threshold was based on a 72-hr exposure to HBCD (Walsh et at..
1987) with measured observed hazardous effect {i.e., growth) on a marine algae species {Skeletonema
costatum), where the exposure concentration was below the water solubility of HBCD. As described in
Sections 2.1.2.6 and 2.1.2.7, the ubiquitous presence of HBCD in the tissues of marine organisms
indicates the exposure of HBCD to marine organisms, despite a lack of information regarding the
source of HBCD. The data availability for freshwater pelagic organisms exposed to HBCD was more
expansive, and as the industrial release of HBCD is more likely to occur in freshwater water bodies,
the acute and chronic COCs were based on hazard thresholds for freshwater organisms. The acute
hazard threshold is based on a 96-hr HBCD exposure to zebrafish embryos, where hatching delay
occurred when exposed to 2 (.ig/L (Hu et at. 2009a). resulting in an acute COC of 0.4 (.ig/L. The chronic
MATC of 4.2 (ig/L derived from a 21-d study using the aquatic invertebrate, D. magna, was used to
calculate the chronic COC of 0.417 (.ig/L (Drottar and Ktueger. 1998). The chronic COC to represent
benthic organisms (/.. variegatus) was also based on the same requirements mentioned above (Oetken
et at.. 2001). In regard to terrestrial organisms, the effect concentration levels as provided in Table 4-2
similarly represent three trophic levels, and the rationale for selecting these studies is based on high
data evaluation quality scores, and the pertinence of the tested exposure and effect measured. The
hazard effects concentrations cover a range of observed effects {i.e., growth, reproductive success,
oxidative stress), and the potential for organisms to be exposed to such concentrations was evaluated
by using both environmental monitoring {i.e., surface water, sediment, and soil) and modeled surface
water and sediment HBCD concentrations to calculate risk estimates.
4.1.1.2 Acute and Chronic Concentrations of Concern
The COC's for acute toxicity were determined by dividing the acute effect level {i.e., reduction of
Page 384 of 723
-------
Zebrafish embryo hatching) by an assessment factor of 5, and the algae (i.e., growth) and chronic (i.e.,
growth of Water Fleas and reproduction of California blackworms) COCs were calculated using an
assessment factor of 10. Further details on the calculations used to derive COCs are described above in
Section 3.1.5.
Table 4-1. Concentrations of Concern (COCs) Derived to Evaluate Toxicity to Aquatic Organisms
for HBCD
Environmental
Media
Organism
and
Endpoint
Hazard Effect
Concentration
Assessment
Factor
Effect
Concentration
of Concern
(COC)
Reference
Data
Evaluation
Score
Zebrafish
(Danio rerio)
96-hr LOAEL
2 (ig/L
5
Delayed
embryo
hatching
0.4 (ig/L
(Hu et al.
2009a)
High
Surface Water
Water flea
(D. magna)
21-d MATC
4.2 (ig/L
10
Reduced
length of
surviving
young
0.417 (ig/L
(Drottar
and
Krueeer
1998)
High
Marine algae
(S. cost a turn)
72-hr EC50
10 ng/L
10
Growth
Rate
1 f-ig/L
(Walsh et
al. 1987)
High
Sediment
California
blackwonn
(Lumbriculus
variegatus)
28-day MATC
15,700
|ig/kg dw
10
Reduction
in worm
number
1,570 ng/kg/dw
(Oetken et
al. 2001)
High
The methodology used to derive concentrations of concern as presented in Table 4-1 are described
above in Sections 3.1.6 and 3.1.7.
Table 4-2. Hazard Effect Concentrations used to Evaluate Toxicity to Terrestrial Organisms
Organism and
Endpoint
Hazard Effect
Concentration
Effect
Reference
Data
Evaluation
Score
Maize
4-d LOAEL
2 f-ig/L
Growth (root and shoot)
(Wu et al. 2016c)
High
Earthworm
14-d MATC
173,000 ng/kg
bw
Reproduction/mortality
(Aufderheide et al. 2003)
High
American kestrel
21-d LOAEL
0.51 mg/kgbw
Reproduction (clutch size,
egg production timing)
(Fernie et al.. 2011)
High
Rat
2-generation NOAEL
10 mg/kgbw
Thyroid hormones
response. Reproduction
(Ema et al. 2008)
High
Studies where terrestrial organisms were exposed to HBCD were evaluated and those with high data
evaluation scores (using environmental Systematic Review metrics) and relevant environmental
exposure pathways were used to assess risk to terrestrial organisms. The studies identified in Table 4-2
provide a summary of studies where chronic exposures to HBCD were conducted with terrestrial
Page 385 of 723
-------
organisms. The organisms identified in the abovementioned studies were chosen to represent their
respective taxa classifications {i.e., vegetation, invertebrate, vertebrate). Out of the four terrestrial
vegetation studies (all rated with high data evaluation scores), (Wu et at 2016c) exposed maize to
HBCD via water exposure; without information regarding biosolid application and exposure, this study
was the most relevant because the exposure is not diastereomer-specific and has a discrete effect
concentration resulting in significant reductions in root and shoot growth. Risk estimates were not
calculated for maize because it is unlikely that terrestrial plants will be exposed to HBCD through
precipitation (as done in the study). For soil organisms in the terrestrial environment the earthworm {E.
fetida) is the most biologically-relevant species for the terrestrial soil environment. The effects of HBCD
exposure to E. fetida has been summarized in the previous section reporting a MATC of 173,000 |ig/kg
bw (Aufderheide et al. 2003). In the 10 highly-evaluated studies, chronic exposures to HBCD resulted in
varying reproductive and developmental effects {e.g., reduced hatching time, smaller egg production,
and the presence of HBCD in eggs) in terrestrial avian species (Table 3-1). As described in Table 4-2,
rats exposed to HBCD resulted in a T4 response in male rats, which corresponds with downstream
reproductive and developmental effects at similar doses (Ema et al. 2008).
4.1.2 Calculation of Risk Quotient (RQ) Values for HBCD
Environmental risk was characterized by calculating risk quotients or RQs ( 3a;
Bamthouse et al.. 1982); the RQ is defined as:
RQ = Environmental Concentration / Effect Level
For aquatic organisms, the "effect level" is a derived COC based on a hazard effects concentration. For
terrestrial organisms, the "effect level" is the hazard effect concentration identified in Table 4-2. COC
calculation methodologies were not originally meant for terrestrial organisms and as mentioned above,
COCs or TRVs were not calculated for terrestrial organisms, where an assessment factor is data-derived
to compensate for varying sources of uncertainties associated with the interpretation and extrapolation of
a hazard threshold. An RQ equal to 1 indicates that the exposures are the same as the concentration that
causes effects. If the RQ is above 1, the exposure is greater than the effect concentration and risk is
indicated. If the RQ is below 1, the exposure is less than the effect concentration and risk is not
indicated.
4.1.3 Risk Estimation Approach
The concentrations of concern (COC) used to calculate risk quotients (RQ) for aquatic organisms were
derived from hazard values resulting from acute and chronic exposures to HBCD. RQs for terrestrial
organisms were derived from the raw hazard values resulting from acute and chronic exposures to
HBCD (no COCs were calculated).
Environmental risk for conditions of use releases was primarily characterized with modeled releases
resulting in estimated media-specific HBCD environmental concentrations, and environmental
monitoring information was used to characterize background exposure to HBCD that is not attributed to
exposure scenario-specific releases for the abovementioned conditions of use or historical uses.
However, in lieu of having exposure scenario-specific media releases, background monitoring data was
used to characterize environmental risk. The totality of background exposure includes steady-state
environmental exposures from ongoing releases that are not associated with a particular COU,
background/indirect exposures from minor use products {e.g., textiles, electrical and electronic products,
adhesives, and coatings) (Section 1.2.8), and releases stemming from historical activities (Section 1.2.9)
due to HBCD's persistence in the environment. Furthermore, background HBCD concentrations derived
from measured environmental monitoring data were not aggregated with modeled exposure-scenario-
Page 386 of 723
-------
specific releases for any media type (e.g., surface water, sediment or soil) to provide an overall total
exposure due to previously existing and potentially current releases because of the uncertainty involved
in discerning the proportions of HBCD that may have come from releases resulting from either historic
or current conditions of uses.
Environmental monitoring data (i.e., surface water, sediment and soil concentrations of HBCD) are also
evaluated below, in the context of the same hazard and COC values as those used for the modeled
surface water and sediment HBCD concentrations predicted by E-FAST and PSC, and the soil
concentrations predicted by IIOAC (via air deposition). The use and derivation of environmental media
HBCD concentrations attained from various sources of monitored data is discussed above in Sections
2.3.2. and 2.3.3. RQ calculations using environmental monitoring data are provided below in Table 4-3,
Table 4-4, and Table 4-6. RQ calculations using environmental modeling data are provided below in
Table 4-5 (surface water and sediment), and Appendix J (soil).
Briefly, environmental monitoring data sampled from the U.S. as well as other high-income countries
(rationale provided in Section 2.3.2) with enough data for the estimation of an arithmetic mean and 90th
percentile value were used to calculate risk estimates. As explained in Section 2.4.2, sampling location
characterization is not feasible because not all literature sources provide this information nor is it always
possible to categorize environmental monitoring data based on industrial sector. Therefore, the
monitoring data is categorized by qualifiers study authors used to indicate sampling proximity to a point
source or non-point source of HBCD.
Risk estimation approach for aquatic organisms
RQ calculations using predicted modeling data are provided below in Table 4-5 (surface water and
sediment) and Table Apx J-13 (soil). Surface water and sediment HBCD concentrations were not
predicted for the following conditions of uses: "Use: Installation of Automobile Replacement Parts",
"Processing: Formulation of Flux/Solder Pastes" , "Processing: Recycling of electronics waste
containing HIPS that contains HBCD" and "Use: Other Formulated Products and Articles (e.g.,
adhesives, coatings, textiles, and electronics)" because surface water releases were not predicted to
occur (as explained in Section 2.2). Surface water releases are likely to occur for all the other exposure
scenarios, and therefore have risk characterized for aquatic organisms. However, although water releases
are predicted to occur for the condition of use "Land disposal of textiles, electrical and electronic
products, adhesives, and coatings", via the potential leaching capacity of HBCD from these facilities
(not through the disposal process of these formulated products and articles) or runoff, there is very
limited information regarding this topic. In lieu of having media-specific release information for this
condition of use via leaching or surface runoff, background information (measured monitoring data) is
used as a proxy to characterize the risk for the "Land disposal of textiles, electrical and electronic
products, adhesives, and coatings".
Further explanations regarding model parameters used for the different scenario labels are provided in
Section 2.3.2. Briefly, E-FAST was used for all conditions of use where water releases were likely to
occur. If the EFAST predicted 7Q10 surface water concentrations (SWCs) were greater than the COCs,
the PSC model was then used to affirm whether the predicted SWC exceeds the COCs using different
parameters. EFAST considers dilution and variability in flow for days exceeded estimates. The PSC also
considers dilution but can further estimate a time-varying surface water concentration, partitioning to
suspended and settled sediment, and degradation within compartments of the water column within a
river segment. To derive risk estimates for pelagic species, the 1- and 21-d predicted surface water
concentrations were compared to the acute, algae, and chronic COCs. To derive risk estimates for
Page 387 of 723
-------
benthic species, the 28-d predicted sediment concentrations based on either the 11- or 128-d HBCD half-
lives were compared to the chronic COC for lumbriculus. Modeled surface water and sediment
concentrations were used to characterize risk for aquatic organisms for all exposure scenarios with
surface water releases except for "Land disposal of textiles, electrical and electronic products, adhesives,
and coatings" because as stated above, there is limited data regarding the HBCD leaching from
associated facilities (described in Section 2.4.5.2). Therefore measured background information (e.g.,
near industrial facilities and near general population) is used to characterize risk to aquatic organisms
due to this condition of use, while understanding that measured background information for specific
media types can be attributed to any releases that occur due to historic or current conditions of use.
The sensitivity analysis on how production volume and percentage of HBCD removal from the direct
release of HBCD into surface water was conducted to reflect declining production volumes and the
likelihood that the HBCD will partition to TSS (Appendix J. 1.2). The surface water and sediment
concentrations were predicted for three production volumes (100,000, 50,000 and 25,000 lbs/yr) due to
the declining use of HBCD and lack of information regarding the current import volume of HBCD to
account for the current processing and use associated with HBCD. Furthermore, the selected exposure
scenarios (Repackaging of Import Containers, Manufacturing of XPS foam from XPS masterbatch, and
Manufacturing of EPS foam from EPS resin) were considered in the sensitivity analysis using the three
production volumes because they were expected to result in high surface water and sediment
concentrations. The estimated emissions from the three exposure scenarios cover emission data from
process-specific industry data and OECD ESDs. The resulting risk estimates from the sensitivity
analysis regarding production volume will not be used for the risk conclusions because the lower
volumes of predicted HBCD production and use are not certain and instead provide support for the
current estimates based on a production volume of 100,000 lbs/yr; based on estimates using the 10th
percentile surface water and sediment HBCD concentrations, decreasing the production volume does not
reduce the number of exposure sub-scenarios with environmental risk.
Risk estimation approach for terrestrial organisms
EPA used IIOAC to estimate air deposition from facility releases, and calculated resulting soil
concentrations near the facilities. IIO AC uses pre-run results from a suite of AERMOD dispersion
scenarios at a variety of meteorological and land-use settings, as well as release emissions, to estimate
particle deposition at different distances from sources that release chemical substances to the air. To
derive risk for soil organisms, the predicted soil concentration from air deposition is compared to the
chronic COC for earthworms.
Soil concentrations (via air deposition) were not predicted for the following conditions of use: "Use:
Installation of Automobile Replacement Parts", and "Use: Other Formulated Products and Articles (e.g.,
adhesives, coatings, textiles, and electronics)", because air releases were not predicted to occur as
explained in Section 2.2). Air releases of HBCD are likely to occur for all the other exposure scenarios,
and therefore have risk characterized for terrestrial soil organisms.
Predicted soil concentrations of HBCD (via air deposition modeling) were used to characterize risk for
terrestrial soil organisms for all exposure scenarios with air releases except for the two listed above, as
well as "Recycling of electronics waste containing HIPS," and "Land disposal of textiles, electrical and
electronic products, adhesives, and coatings." Although air releases are predicted to occur for the
condition of use "Recycling of electronics waste containing HIPS," a semi-quantitative screening
approach (as explained below in Section 4.1.3.2.3) was used to compare industrial releases associated
with this exposure scenario to those of other exposure scenarios with air releases; the release days and
Page 388 of 723
-------
amount of HBCD released were factors considered to determine whether this exposure scenario will
likely have soil concentrations of HBCD that may exceed the chronic hazard threshold for earthworms.
In regards to "Land disposal of textiles, electrical and electronic products, adhesives, and coatings",
there is limited data regarding the release of HBCD into air from associated facilities. Therefore
measured background information (e.g., near industrial facilities and near general population) is used to
characterize risk to terrestrial soil organisms due to this condition of use, while understanding that
measured background information for specific media types can be attributed to any releases that occur
due to historic or current conditions of use.
4.1.3.1 Risk Estimation Based on HBCD Surface Water and Sediment
Concentrations using Environmental Monitoring Data and Modeling
Results
The COCs and hazard effect concentrations used to calculate RQs below are summarized above in
Section 4.1.2, with the respective toxicity data.
4.1.3.1.1 Risk Estimation Based on Surface Water and Sediment Monitoring
Data
Table 4-3. Calculated Risk Quotients based on HBCD Surface Water (jig/L) Concentrations as
Reported in Environmental Monitoring Studies
Site
Characteriza
-tion
Surface Water
Concentrations (|ig/L)
Acute RQ
(COC: 0.4 jig/L)
Algae RQ
(COC: 1 jig/L)
Chronic RQ
(COC: 0.417 jig/L)
Mean of
Mean
Avg of
90th
Percentile
Risk
estimate
using:
Mean of
Mean
Risk
estimate
using:
Avg of
90th
Percentile
Risk
estimate
using:
Mean of
Mean
Risk estimate
using: Avg of
90th
Percentile
Risk
estimate
using:
Mean of
Mean
Risk estimate
using: Avg of
90th
Percentile
Near
Industrial
Facility
(Point
Source)3
0.84
0.99
2.10
2.48
0.84
0.99c
2.02
2.38
Near General
Population
(Non-Point
Source)b
0.00041
0.0008
1.03E-
03
2.00E-03
4.10E-04
8.00E-04
9.83E-04
1.92E-03
Values in bold text and highlighted in red denote a risk (RQ>1) to the aquatic environment where the surface water concentration
(SWC) exceeds the concentration of concern (COC) for acute, chronic and algae environmental hazard. The algae RQ based on the
average 90th percentile SWC is bolded to indicate risk.
^References to characterize the mean of the mean and averaae of 90th percentile SWCs are listed here: (Ichihara et al. 2014; Kowalski
and Mazur 2014; Oh et al. 2014).
bReferences to characterize the mean of the mean and averaae of 90th percentile SWCs are listed here: (Law et al. 2006; Harrad et al.
2009; Ichihara et al. 2014; Venier et al. 2014)Values in bold text and highlighted in red denote a risk (RO>l) to the aauatic
environment where the surface water concentration (SWC) exceeds the concentration of concern (COC) for acute, chronic and algae
environmental hazard. The algae RQ based on the average 90th percentile SWC is bolded to indicate risk.
c The RQ of 0.99 is an indicator of risk to algae because although the surface water concentration of HBCD measured near industrial
facilities is an average of the high end (90th percentile) measured concentrations reported in these studies, there were measured
concentrations used that are above 0.99 ua/L (Ichihara et al. 2014; Kowalski and Mazur 2014; Oh et al. 2014). To be more conservative
of the wide ranges of HBCD measured concentrations in surface water, a RQ of 0.99 is still a likely indicator that algae near industrial
facilities will be exposed to HBCD at concentrations that may exceed the COC of 1 ng/L.
Page 389 of 723
-------
Table 4-4. Calculated Risk Quotients based on HBCD Sediment Concentrations (jig/kg) as
Reported in Environmental Monitoring Studies
Site Characterization
Sediment Concentrations (|Lig/kg)
Chronic RQ
(COC: 1,570 jig/kg)
Mean of Mean
Avg of 90th
Percentile
Mean of Mean
Avg of 90th
Percentile
Near Industrial Facility (Point
Source)3
3443
5073
2.193
3.231
Near General Population (Non-
Point Source )b
6.2
19.8
0.0039
0.0126
Values in bold text and highlighted in red denote a risk (RQ>1) to the aquatic environment where the sediment
concentration exceeds the concentration of concern (COC).
"¦References to characterize the mean of the mean and average of 90th percentile sediment concentrations are listed here:
(Sellstrom et al. 1998; Haukas et al. 2010b: La Guardia et al. 2010; Oh et al. 2014; Al-Odaini et al. 2015; Stiborova et al.
2017)
References to characterize the mean of the mean and average of 90th percentile sediment concentrations are listed here:
(Ramu et al. 2010; Klosterhaus et al. 2012; Yans et al. 2012; Harrad et al. 2009; Haukas et al. 2009; Kohler et al. 2008;
Minli et al. 2007; Morris et al. 2004; Rembereer et al. 2004; Jeons et al. 2014; Luisi et al. 2015; Lvons et al. 2015; Al-
Odaini et al. 2015; Anim et al. 2017)
Page 390 of 723
-------
4.1.3.1.2 Risk Estimation Based on Surface Water and Sediment Modeling Data
Table 4-5. Range of Risk Quotients for Modeled Surface Water and Sediment HBCD Concentrations for Each Condition of Use
Using a Production Volume of 100,000 lbs/yr (0% removal for direct release)
Risk quotients (RQs) for surface water are calculated using aquatic acute, algae and chronic COCs of 0.4,1.0 and 0.417 jig/L, respectively. RQs for sediment
are calculated using the sediment COC of 1,570 jug/kg. If the predicted surface water or sediment concentration was 0 or if the calculated RQ was< 0.005, the
RQ was rounded to 0. Values in bold text and highlighted in red denote exposure scenarios where at least half of the model sub-scenarios have risk (RQ>1) to
the pelagic or benthic environment where the surface water concentration (SWC) or sediment concentration, respectively, exceeds the concentration of
concern (COC) for environmental hazard.
Exposure Scenario
Surface Water
Sediment
Acute
Algae
Chronic
11-d half-life
128-d half-life
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
Section 2.2.2 - Repackaging
of Import Containers (1)
4.3-189
0.09-24.2
1.72-75.6
0.04-0.83
3.5-21.22
0.07-2.26
0.87-4.61
0.02-0.56
2.29-11.91
0.05-1.26
Section 2.2.3 - Compounding
of Polystyrene Resin to
Produce XPS Masterbatch (2)
3.48-34.75
0.09-2.08
1.39-31.3
0.04-0.83
0.19-4.22
0-0.1
0-0.77
0-0.02
0-1.86
0-0.04
Section 2.2.4 - Processing of
HBCD to produce XPS Foam
using XPS Masterbatch (3)
0.76-275
0.02-7.33
0.3-110
0.01-2.93
0.04-13.55
0-0.34
0.01-2.22
0-0.06
0.02-2.97
0-0.08
Section 2.2.5 - Processing of
HBCD to produce XPS Foam
using HBCD Powder (4)
0.91-107
0.02-2.85
0.02-2.85
0.01-1.14
0.05-5.25
0-0.13
0.01-0.87
0-0.02
0.02-1.16
0-0.03
Section 2.2.6 - Processing of
HBCD to produce EPS Foam
from Imported EPS Resin
Beads (5)
89.5-9,900
2.2-262.5
35.8-3,960
0.88-105
33.57-
563.55
0.71-12.01
8.73-
143.31
0.21-3.52
22.68-
361.78
0.48-7.77
Section 2.2.7 - Processing of
HBCD to produce SIPs and
Automobile Replacement
Parts from XPS/EPS Foam
(6)
0.97-
148.75
0.02-3.93
0.39-59.5
0.01-1.57
0.19-8.47
0-0.18
0-2.15
0-0.05
0-5.43
0-0.12
Page 391 of 723
-------
Risk quotients (RQs) for surface water are calculated using aquatic acute, algae and chronic COCs of 0.4,1.0 and 0.417 jig/L, respectively. RQs for sediment
are calculated using the sediment COC of 1,570 jug/kg. If the predicted surface water or sediment concentration was 0 or if the calculated RQ was< 0.005, the
RQ was rounded to 0. Values in bold text and highlighted in red denote exposure scenarios where at least half of the model sub-scenarios have risk (RQ>1) to
the pelagic or benthic environment where the surface water concentration (SWC) or sediment concentration, respectively, exceeds the concentration of
concern (COC) for environmental hazard.
Exposure Scenario
Surface Water
Sediment
Acute
Algae
Chronic
11-d half-life
128-d half-life
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
Section 2.2.9 - Installation of
XPS/EPS Foam Insulation in
Residential, Public, and
Commercial Buildings, and
Other Structures (8)
0.05-59.25
0.01-8.45
0.02-23.7
0-3.38
0-4.10
0-0.04
0.06-0.57
0.01-0.07
0.13-1.28
0.01-0.10
Section 2.2.10- Demolition
and Disposal of XPS/EPS
Foam Insulation Products in
Residential, Public and
Commercial Buildings, and
Other Structures (9)
0.05-59.25
0.01-8.45
0.02-23.7
0-3.38
0-4.10
0-0.04
0.01-0.10
0.002-0.02
0.001-0.01
0.0001-
0.0007
Section 2.2.11- Recycling of
EPS Foam and Reuse of XPS
Foam (10)
1.2-183.25
0.03-4.88
0.48-73.3
0.01-1.95
0.45-9.02
0.01-0.22
0.12-1.48
0-0.04
0.17-1.98
0-0.06
Section 2.2.13 - Use of
Flux/Solder Pastes (12)
0.58-1.19
0.02-0.15
0.23-0.47
0.01-0.06
0.03-0.06
0-0.01
0-0.01
0
0.01-0.02
0
Page 392 of 723
-------
4.1.3.1.3 Risk Estimation for the Recycling of Electronics Waste Containing
HIPS
To characterize the risk associated with the recycling of electronics waste containing HIPS that contain
HBCD, a screening level approach was used to compare the estimates for HBCD release and duration to
those used for other conditions of uses. To be specific, for aquatic environments, there is no information
available that suggests water releases will occur for this exposure scenario or related COU, therefore
surface water and sediment HBCD concentrations were not modeled for this exposure scenario.
However, similar to other exposure scenarios and historical uses that may release HBCD into the aquatic
environment, it is likely that the Recycling of Electronic Waste Containing HIPS may have also
contributed to measured background concentrations of HBCD discussed above in Section 4.1.3.1.1, and
therefore there is potential for there to be risk for aquatic organisms near industrial facilities that are
associated with the recycling of electronic waste containing HIPS.
4.1.3.2 Risk Estimation based on HBCD Soil Concentrations using Environmental
Monitoring and Modeling Data
4.1.3.2.1 Risk Estimation Based on Soil Monitoring Data
Table 4-6. Calculated Risk Quotients based on HBCD Soil Concentrations (jig/kg) as Reported in
Environmental Monitoring Studies
Data Source
HBCD Source
Site Characterization
Soil Concentrations
(Mg/kg)
Chronic RQ
(Hazard effect
concentration:
173,000 fig /kg)
Mean of
Mean
Avg of
90th
Percentile
Mean of
Mean
Avg of
90th
Percentile
Environmental
Monitoring
Air Deposition
Near Industrial Facility
(Point Source)3
1,016
1,254
5.87xlO"3
7.25xl0-3
Near General Population
(Non-Point Source)b
1.4
3.0
8.30xl0"6
1.74 xlO"5
There are no instances of risk estimates that denote a risk (RQ>1 indicating risk) to the terrestrial enviromnent where the
soil concentration exceeds the hazard effects concentration for earthworm reproduction (Aufderheide et al. 2003).
a References to characterize the mean of the mean and average of 90th percentile soil concentrations are listed
here:(Rembereer et al. 2004)
b References to characterize the mean of the mean and average of 90th percentile soil concentrations are listed here:
(Covad et al. 2009; Newton et al. 2015)
4.1.3.2.2 Risk Estimation Based on Soil Modeling Data
As presented in Appendix Table Apx J-13, there are no instances of risk quotients equal to or greater
than one (indicating risk) when using the highest IIOAC predictions for soil HBCD concentrations in
either the fenceline or community scenarios. The results suggest the unlikelihood that any of the
exposure scenarios alone will contribute sufficient HBCD to result in risk for terrestrial soil organisms.
The below table presents a summary of risk estimation for soil organisms based on both monitoring and
modeling data.
Page 393 of 723
-------
Table 4-7. Calculated Risk Quotients based on HBCD Soil Concentrations (jig/kg) as Reported in
Environmental Monitoring Studies and Calculated using Modeling Data
Data Source
HBCD Source
Site Characterization
Soil Concentrations (jig/kg)
Chronic RQ
(Hazard effect
concentration:
173,000 fig /kg)
Environmental
Monitoring
Air Deposition
Near Industrial
Facility (Point Source)
50th Percentile
1016
0.0059
90th Percentile
1254
0.0072
Near General
Population
(Background)
50th Percentile
1.44
0
90th Percentile
3.01
0
Model
Biosolid
Application
Agriculture (Point
Source)
* Based on LaGuardia
2010
Maximum
30.00
0.0002
Air Deposition
Near Industrial
Facility (Point Source)
Maximum
0.13
0
Combined
Air Deposition
Near Facility,
Biosolid
Application and
Background Levels
N/A
41.00
0.0002
There are no instances of risk estimates that denote a risk (RQ>1 indicating risk) to the terrestrial enviromnent where the
soil concentration exceeds the hazard effects concentration for earthworm reproduction (Aufderheide et al. 2003).
When the RQs are <0.0001, the RQ is rounded to 0.
4.1.3.2.3 Risk Estimation for the Recycling of Electronics Waste Containing
HIPS
Risk Estimates for Terrestrial Ecosystems based on Monitoring Data
HBCD releases from the recycling of electronics waste containing HIPs, has been identified as an
ongoing COU (associated exposure scenario Recycling of Electronic Waste Containing HIPS), and
environmental risk due to potential releases from electronics recycling sites is characterized below using
environmental monitoring data provided above in Section 4.1.3.2.1 (same analysis used to evaluate
HBCD background exposure). In regards to the use of environmental monitoring data to characterize
background HBCD concentrations (where releases from historic and current conditions of use have
likely contributed to), risk to terrestrial organisms due to chronic HBCD exposure is characterized by
soil concentrations (Table 4-6) measured near industrial facilities (point source exposure) or general
population (non-point source exposure).
Risk Estimates for Terrestrial Ecosystems based on Modeling Data
To characterize the risk associated with exposure scenario-specific media releases for each current
conditions of use to soil organisms, soil concentrations (via air deposition) were estimated using
methods outlined above in Section 2.3. To evaluate environmental risk to soil organisms due to the
Recycling of Electronic Waste Containing HIPS, a screening level approach was used to compare the
estimates for HBCD air release and duration to those used for other conditions of use. EPA estimated
central tendency and high-end air releases of HBCD from electronic recycling sites to be 0.024 and 0.38
kg/site-d, respectively, for a duration of 250 days. EPA compared the air release estimates for electronic
recycling sites to those that were previously used to quantify HBCD soil concentration (via air
Page 394 of 723
-------
deposition) for the other conditions of use. The daily release amounts of HBCD and number of release
days estimated for electronic recycling sites fall within the range as those used to characterize and
estimate soil HBCD concentrations from air deposition for other conditions of use. Specifically, in
comparison to exposure scenario 6.12, where the daily release of HBCD (3.8 kg/site-d) and number of
release days (300 days) are both higher than those predicted for electronic recycling sites, the resulting
soil HBCD concentration for exposure scenario 6.12 is 3.66E-03 |ig/kg for fenceline communities (near
industrial facilities). This exposure scenario's estimated soil concentration of HBCD does not surpass
the hazard threshold for soil organisms (173,000 |ig /kg), and therefore did not result in environmental
risk. Due to the unlikelihood that the lower release amounts and days for electronic recycling sites will
surpass those used for any of the current conditions of use, soil concentrations of HBCD due to air
deposition were not estimated using methods outlined above in Appendix F.1.2 for this condition of use.
There are no estimated HBCD soil concentrations resulting from modeled HBCD release via air
deposition that exceed the chronic COC for soil organisms (Appendix J.1.3.1) for any conditions of use,
including the Recycling of Electronic Waste Containing HIPS.
4.1.3.3 Risk Estimation based on Exposure via Trophic Transfer
To calculate RQs for the organisms in Table 4-8 and Table 4-9, hazard effect concentrations were
selected based on high data quality evaluation scores as well as the appropriateness of the endpoint in
regards to what is likely a chronic dietary exposure.
As summarized in Table 4-8, risk quotients (RQs) calculated for deer mouse, kestrel and osprey are
based on the amount of HBCD consumed per day normalized to body weight as calculated using the
values from Table 3-2 (exposure factors were from the U.S. EPA Final Water Quality Guidance for
Great Lakes System and U.S. EPA Wildlife Exposure Factors Handbook). The hazard values are
described in Section 3.1.2 and the Systematic Review Supplemental File: Data Extraction Tables of
Environmental Hazard Studies. The osprey hazard effect concentration is allometrically-scaled from the
kestrel 75-d LOAEL (reproductive toxicity).
As summarized in Table 4-9, RQs calculated for rainbow trout and earthworms are based on the amount
of HBCD consumed per day normalized to body weight as calculated using the values from Table 3-3
(exposure factors were from the ECHA Guidance on Information Requirements and Chemical Safety
Assessment (Environmental Exposure Assessment)). The hazard values are described in Section 3.1.2
and the Systematic Review Supplemental File: Data Extraction Tables of Environmental Hazard Studies.
Table 4-8. Calculated Risk Quotients based on Potential Trophic Transfer of HBCD in Aquatic
and Terrestrial Ecosystems Using the U.S. EPA Final Water Quality Guidance for Great Lakes
System and U.S. EPA Wildlife Exposure Factors Handbook
Organism
Amount of HBCD
consumed per day
normalized to body
weight (mg/kg bw)
Hazard Effect
Concentration
(mg/kg bw)
Reference for
Hazard Effect
Concentration
RQs
Kestrel
0.0005
0.51
(Fernie et al„ 2011)
0.001
Osprey
lxlO"6 - 2.0
0.51
(Fernie et al„ 2011)
2x10 6 - 3.92
Values in bold text and highlighted in red denote risk (RQ>1) where the dietary uptake of HBCD
exceeds the hazard threshold.
Page 395 of 723
-------
Table 4-9. Calculated Risk Quotients based on Potential Trophic Transfer of HBCD in Aquatic
and Terrestrial Ecosystems using the ECHA Guidance on Information Requirements and
Chemical Safety Assessment (Environmental Exposure Assessment)
Organism
Amount of HBCD
consumed per day
normalized to body weight
(mg/kg bw)
Hazard Effect
Concentration
(mg/kg bw)
Reference for
Hazard Effect
Concentration
RQs
Rainbow trout
16.2
0.0025
(Wildlife Intl
1997a)
6480
Earthworm
18855
173.00
Aufderheide et
al. (2003)
109
Values in bold text and highlighted in red denote risk (RQ>1) where the dietary uptake of
exceeds the hazard threshold.
HBCD
4.1.4 Environmental Risk Results
The risk of HBCD to aquatic and terrestrial ecosystems are summarized in Sections 4.1.4.1, and
Appendix J. Specifically, Table 4-3, Table 4-4 and Table 4-6 include risk quotients (RQ) based on
reported environmental monitoring data for HBCD concentrations in sampled surface water, sediment
and soil samples, respectively. Table 4-5 includes RQs based on predicted surface water and sediment
concentrations categorized by the different modeling scenarios for each exposure scenario (further
details provided in Section 2.3). The presented RQs are based on predicted surface water and sediment
concentrations using the 10th and 50th percentile flow. RQs based on predicted soil HBCD
concentrations via air deposition, are presented in Section J. 1.3.1; there were no RQs equal to or greater
than one. Table 4-5 includes RQs for each exposure scenario based on predictions of surface water and
sediment HBCD concentrations using the Variable Volume Waterbody Model (VVWM) - Point Source
Calculator (PSC). Table 4-8 and Table 4-9 depict RQs that are primarily based on environmental
monitoring data, where exposure is further characterized by diet-based exposure factors (U.S. EPA Final
Water Quality Guidance for Great Lakes System and U.S. EPA Wildlife Exposure Factors Handbook),
or laboratory-derived bioconcentration factors (BCFs).
Risk to the aquatic environment is characterized by evaluating both surface water and sediment
concentrations of HBCD, by using both environmental monitoring and predicted surface water and
sediment concentrations. Risk to the terrestrial environment was also characterized by using predicted
surface water and sediment concentrations as input values for KABAM (vl), in addition to soil HBCD
concentrations attained from environmental monitoring studies and predicted using the IIOAC air
deposition HBCD concentrations. Furthermore, to evaluate how HBCD trophic transfer would impact
predators in both aquatic and terrestrial environments, risk to both aquatic and terrestrial avian species
was derived for osprey, and kestrel, respectively, where the Kestrel reproductive hazard effect
concentration was allometrically-scaled for osprey. Due to the lack of hazard data regarding the
exposure of HBCD to higher trophic level aquatic organisms (despite large amounts of biomonitoring
data) and the greater likelihood that HBCD will be released into aquatic environments, osprey was
chosen as a representative species for an aquatic predator because the diet is easily characterized by fish
consumption. Measured data demonstrates that HBCD will bioconcentrate in Rainbow trout and
Earthworms (Table 3-3), therefore risk was also characterized to evaluate whether HBCD poses risk to
organisms where high HBCD bioconcentration has already been quantified.
Page 396 of 723
-------
The red shaded values in these tables denote when at least half of the calculated RQs are equal to or
greater than one, indicating that the modeled or measured surface water, sediment or soil concentration
of HBCD exceeds the COC or hazards effect concentration (terrestrial organisms only), resulting from
acute or chronic exposures.
4.1.4,1 Risk Characterization for Aquatic and Terrestrial Ecosystems based on
Environmental Monitoring Data
As seen in Table 4-3 and Table 4-4, acute, algae and chronic risk quotients based on measured
concentrations of HBCD near industrial facilities are equal to or greater than one, suggesting that
although these surface water and sediment HBCD concentrations are not indicative of a specific
exposure scenario or condition of use, there is concern regarding the potential additional release of
HBCD from industrial facilities. In addition, generally both the use of the mean of mean and average of
the 90th percentile measured surface water and sediment HBCD concentrations from environmental
monitoring studies yielded RQs equal to or greater than one. On the other hand, RQs based of measured
surface water and sediment concentrations of HBCD in sites unassociated with an industrial facility are
all at least one magnitude below one. HBCD may not be bioavailable for even benthic organisms
downstream of industrial facilities. HBCD is expected to have higher binding affinity for sediment and
organic matter and will partition out of the water column quickly. HBCD was undetectable in sediment
samples 60 km downstream of industrial facilities (Guerra et al. 2009). suggesting that it is unlikely for
aquatic organisms that do not inhabit areas within close proximity to industrial facilities to be at risk for
HBCD exposure.
Similarly, as depicted in Table 4-6, the calculated RQs based on measured HBCD soil concentrations
are all more than one magnitude below one, using the earthworm chronic hazard effect concentrations;
there is not predicted risk for soil-dwelling organisms either near industrial facilities or sites associated
with the general population. In regard to terrestrial vegetation, there were no hazard data available
regarding vegetation exposure to HBCD via soil. On the other hand, a reduction in root and shoot
growth was observed when maize was exposed to 2 |ig HBCD/L; there are no measured surface water
concentrations of HBCD that exceeds 2 |ig HBCD/L for either near sites categorized as being associated
with an industrial facility or near general populations.
As stated in Section 4.1.2, the goal of environmental risk characterization is to determine whether there
are risks to the aquatic or terrestrial environments from measured levels of HBCD found in surface
water, sediment or soil. The risk quotients (RQ) method ( a; Barnthouse et al.. 1982) was
used to determine whether the exposures of HBCD exceed either the concentrations of concern (COC)
or hazard effects concentrations for aquatic or terrestrial organisms, respectively. Regarding terrestrial
organisms, the risk is not as easily characterized because the available hazard and exposure data are not
completely compatible {i.e., the exposure media and corresponding units do not always match those
used in predictive models or reporting methods used to collect environmental monitoring or
biomonitoring data). Specifically, the terrestrial plants with data (regarding HBCD exposure) are all
agricultural crops and were exposed to HBCD using exposure solutions with dissolved HBCD; the most
relevant exposure pathway for HBCD to agricultural crops would be via the application of biosolids.
Therefore, a RQ cannot be calculated to determine whether the exposure concentration is above the
threshold where toxicological effects are observed due to biosolid application. Using the soil
environmental monitoring data as presented in Table 4-6 the risk of HBCD to soil invertebrates can be
evaluated by using the earthworm hazard effect concentration (56-d GMATC of 173,000 |ig/kg). There
are no RQs greater than one using the highest soil concentrations across the data sources presented,
suggesting that terrestrial invertebrates will not be exposed to HBCD concentrations that exceed the
exposure concentrations where toxicological effects were observed. As presented in Table 4-6, using
Page 397 of 723
-------
EPA methodology outlined in Section 2.3.3, a soil concentration of 41 |ig/kg was calculated, with
biosolid application contributing approximately 75% of the combined HBCD soil concentration, as
compared to air deposition or background levels. Using either environmental monitoring, modeled or a
combination of both types of data regarding soil HBCD resulted in RQs below one, further supporting
the unlikelihood that soil organisms such as earthworms will be exposed to concentrations of HBCD that
will exceed the threshold of hazard.
As a PBT chemical, considering the potential for chronic exposures to HBCD due to all sources {i.e.,
water releases, air deposition, biosolid application and background levels) is imperative because
evaluating any one release or exposure pathway for HBCD may underestimate HBCD exposure.
Specifically, evaluating air deposition alone may imply that there isn't risk to terrestrial organisms that
do not inhabit areas near industrial facilities (accounting for multiple conditions of uses). Measured soil
concentrations of HBCD associated with either industrial facilities or general populations, biosolid
application or background levels are greater than those predicted for specific exposure scenarios using
the IIOAC. For example, the highest predicted soil HBCD concentration (via air deposition) is 0.134
|ig/kg for the exposure scenario "Processing: Manufacturing EPS Foam from Imported EPS Resin
beads" from fugitive stacks, which is four and one magnitude less than the amounts of HBCD measured
near sites associated with industrial facilities and general population, respectively. This suggests that
there is risk to soil organisms even without the additional HBCD releases via specific exposure
scenarios and conditions of use and that predicted risk to soil organisms via air deposition is greatly
underestimated by the use of modeled releases alone.
Land disposal of other formulated products and articles (e.g., adhesives, coatings, textiles, and
electronics) (Exposure scenario characterized using solely monitoring data as a proxy)
As explained in Section 2.2, EPA did not assess a range of daily release rates based on data pertaining to
emission factors and number of days of release per year. There is uncertainty regarding the extent to
which these emission factors and number of days of release per year are applicable to the land disposal
of formulated products and articles such as adhesives, coatings, textiles and electronics that would occur
in the U.S. Releases of HBCD to the aquatic and terrestrial environment are likely to occur via this
exposure scenario. The above explanation regarding the use of background information to support the
predicted modeling of media-specific HBCD concentrations, apply to all conditions of use and historic
uses, however as explained above in Section 4.1.3, in lieu of having modeled surface water, sediment
and soil (via air deposition) HBCD concentrations for this exposure scenario, background concentrations
are used to characterize the risk to aquatic and terrestrial soil organisms. In short, RQs are greater than
one based on algae, acute and chronic hazard thresholds (COCs) near industrial facilities, using
measured surface water concentrations (Table 4-3). Likewise, using measured sediment concentrations
of HBCD near industrial facilities, RQs are greater than one based on the chronic hazard threshold for
benthic invertebrates (Table 4-4). Although background concentrations of HBCD encompass HBCD
releases from both historical and current conditions of use, there may be risk to aquatic organisms that
inhabit water bodies near facilities associated with the disposal of adhesives, coatings, textiles, and
electronics due to leaching and runoff. There are no measured soil concentrations due to air deposition
that result in risk to soil organisms, based on the chronic hazard threshold for earthworms.
4,1.4,2 Risk Characterization for Aquatic and Terrestrial Ecosystems based on
Modeled Surface Water and Sediment Concentrations
To evaluate the risk for organisms in aquatic ecosystems due to of HBCD exposure, modeled surface
water concentrations were compared to concentrations of concern (COC) based on acute, algae, and
chronic hazard effect concentrations, and modeled sediment concentrations were compared to a chronic
hazard effect concentration. The exposure scenarios are labeled with an exposure scenario number to
Page 398 of 723
-------
help orient the audience as the exposure scenarios do not necessarily follow the same order when
categorized by their condition of use (as seen in Section 4.5).
The risk quotients (RQ) for every exposure sub-scenario are available in Appendix J. 1.2, based on both
the 10th and 50th percentile surface water and pore water concentrations of HBCD. The risk
characterization for aquatic organisms is based on RQs derived from predicted surface water and
sediment concentrations for production volumes of 100,000 lbs/yr and 0% removal of HBCD from
directly released HBCD into surface water.
Additionally, a targeted sensitivity analysis was conducted to characterize how two additional
production volumes for three exposure scenarios (derived from the 10th and 50th percentile surface water
and sediment HBCD concentrations) may affect derived RQs for aquatic organisms. Using the predicted
surface water and pore water HBCD concentrations from the PSC, and proxy organism hazard data {i.e.,
rats and Japanese quail) as input parameters for KABAM (vl), RQs for multiple mammalian wildlife
species can be estimated (assuming that the effect concentrations are the same as those as the proxy
organism by scaling of body weight).
Repackaging of Import Containers (Exposure scenario 1)
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for
repackaging of import containers. This process can result in direct releases of HBCD into surface water,
or release through POTWs.
For each release medium, EPA assessed a range of daily release rates based on data pertaining to
emission factors and number of days of release per year. The emission factors were obtained from the
OECD ESD on Plastics Additives (OECD 2009). The number of days of release per year are estimated
values that are applicable to the basic chemical industry in general (ECB 2003). There is some
uncertainty regarding the extent to which these emission factors and number of days of release per year
are applicable to the repackaging of import containers that would occur in the U.S. Releases of HBCD to
the aquatic environment are due to the activity of repackaging of import containers.
Within this exposure scenario, there are eight exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the predicted 50th percentile surface water HBCD
concentrations, there are three acute and algae and two chronic risk estimates that are greater than one,
based on the acute, algae and chronic COCs of 0.4, 1.0 and 0.417 |ig/L, respectively. Based on the
predicted 10th percentile surface water HBCD concentrations, all eight exposure sub-scenarios have RQs
greater than one, based off of the acute, algae and chronic COCs.
In evaluating the 50th percentile predictions to calculate RQs for benthic organisms, there are two RQs
greater than one using the 128-d HBCD half-life; none of the RQs based on the 11-d HBCD half-life
resulted in RQs equal to or greater than one. Based on the predicted 10th percentile sediment HBCD
concentrations, all eight of the RQs based on the 128-d HBCD half-life had RQs greater than one. Based
on the predicted 10th percentile sediment HBCD concentrations, half of the RQs were greater than one,
using the 11-d HBCD half-life; the other four RQs were within approximately 10% of a RQ of 1,
demonstrating that the predicted releases based on the less conservative half-life of 11-d exceeds or are
very close to reaching the reproductive hazard threshold for L. variegains (Oetken et al. 2001).
Compounding of Polystyrene Resin to Produce XPS Masterbatch (Exposure scenario 2)
Page 399 of 723
-------
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for
compounding polystyrene resin to produce XPS Masterbatch. This process can result in direct releases
of HBCD into surface water, or release through POTWs and onsite wastewater treatment.
For each release medium, EPA assessed a range of daily release rates based on data pertaining to
emission factors and number of days of release per year. The emission factors pertain to sites in Europe
at which XPS Masterbatch was compounded (ECHA. 2008b). The data pertaining to the number of
release days per year are estimated values that are applicable to the polymer formulation industry in
general (ECB 2.003). There is some uncertainty regarding the extent to which these emission factors and
number of days of release per year are applicable to the compounding of XPS Masterbatch that would
occur in the U.S. Releases of HBCD to the aquatic environment are due to the activity of compounding
polystyrene resin to produce masterbatches of XPS.
Within this exposure scenario, there are twelve exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the predicted 50th percentile surface water concentrations of
HBCD, there are only two RQs greater than one when using the fish acute COC of 0.4 |ig/L (Hu et al.
2009a). Based on the predicted 10th percentile surface water concentrations of HBCD, eight out of
twelve exposure sub-scenarios have RQs greater than one when using the fish acute and algae COCs,
and four of those twelve have RQs greater than one when using the chronic COC based on water flea
reproductive hazard effect concentration (Drottar and Kmeeer 1998).
In regard to the predicted sediment HBCD concentrations, there are only two RQs that are equal to or
greater than one for predicted sediment concentrations, and both were calculated using the 10th
percentile prediction based on the longer 128-d HBCD half-life.
Manufacturing of XPS Foam using XPS Masterbatch (Exposure scenario 3)
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for
manufacturing of XPS foam using XPS Masterbatch. This process can result in direct releases of HBCD
into surface water, or release through POTWs and onsite wastewater treatment.
For each release medium, EPA assessed a range of daily release rates based on data pertaining to
emission factors and number of days of release per year. These emission factors and number of days of
release per year pertain to sites in Europe at which XPS Foam was manufactured (ECHA 2008b). There
is some uncertainty regarding the extent to which these emission factors and number of days of release
per year are applicable to the manufacture of XPS from Masterbatch that would occur in the U.S.
Releases of HBCD to the aquatic environment are due to the activity of manufacturing of XPS foam
using XPS Masterbatch.
Within this exposure scenario, there are 12 exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the predicted 50th percentile surface water concentrations,
there are no RQs greater than one for chronic hazard, but there are four and three RQs that exceed the
threshold for risk for the acute and algae COCs, respectively. Based on the predicted 10th percentile
surface water concentrations, there are 10, eight and five RQs that are greater than one, when using the
acute, algae and chronic COCs, respectively.
Based on the 50th percentile predictions for sediment HBCD concentrations, there were no instances of
risk estimates greater than one (indicating risk) using either the 11- or 128-d half-lives of HBCD. Based
Page 400 of 723
-------
on the predicted 10th percentile sediment HBCD concentrations, there are one and two RQs that exceed
the sediment COC, using the 11- or 128-d HBCD half-life, respectively.
Processing of HBCD to produce XPS Foam using HBCD Powder (Exposure scenario 4)
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for
manufacturing of XPS foam using XPS powder. This process can result in direct releases of HBCD into
surface water, or release through POTWs and onsite wastewater treatment.
For each release medium, EPA assessed a range of daily release rates based on TRI data and data
pertaining to emission factors and number of days of release per year. These emission factors in general
and number of days of release per year in the case of releases to water pertain to sites in Europe at which
XPS Foam was manufactured (EC 08b). In the case of releases to air, the data pertaining to the
number of release days are estimated values that are applicable to the industrial use of polymers in
general (ECB 2003). There is some uncertainty regarding the extent to which these emission factors and
number of days of release per year are applicable to the manufacture of XPS from HBCD that would
occur in the U.S. Releases of HBCD to the aquatic environment are due to the activity of manufacturing
XPS foam using HBCD powder.
Within this exposure scenario, there are six exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the predicted 50th percentile surface water HBCD
concentrations, there are no RQs greater than one using the chronic COC. However, based on the algae
and acute COC, there is one and two RQs greater than one, respectively. Based on the predicted 10th
percentile surface water HBCD concentrations, there is one RQ greater than one when using the chronic
COC, and four RQs that are greater than one when using either the acute or algae COC. Additionally, of
the six exposure sub-scenarios, the two RQs based on the acute COC that are less than one have surface
water concentrations that are within 10% of exceeding the acute COC, suggesting that all six acute RQs
either exceed or within the same magnitude of the zebrafish hazard effect concentration (Hu et al.
2009a).
In regard to sediment HBCD concentrations modeled using the PSC, there is only one risk estimate
greater than one, using the 10th percentile predictions based on the 128-d HBCD half-life, suggesting
that the water releases of HBCD from this exposure scenario will not result in sediment concentrations
of HBCD that will surpass the sediment COC.
Processing of HBCD to produce EPS Foam from Imported EPS Resin Beads (Exposure scenario 5)
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for
manufacturing of EPS foam imported EPS Resin Beads. This process can result in direct releases of
HBCD into surface water, or release through POTWs and onsite wastewater treatment.
For each release medium, EPA assessed a range of daily release rates based on data pertaining to
emission factors and number of days of release per year. The emission factors were obtained from the
OECD ESD on Plastics Additives (OECD 2009) or an EPA/OPPT screening-level model. The number
of days of release per year is an estimated value that is applicable to the industrial use of polymers in
general or is a value that pertains to the manufacture of EPS foam at a site in Australia (NICNAS.
2012b). There is some uncertainty regarding the extent to which these emission factors and number of
days of release per year are applicable to the manufacture of EPS foam that would occur in the U.S.
Furthermore, EPA's assessment of releases may be conservative based on a comparison of sources of
release and emission factors as assessed by EPA and as reported in EURAR and NICNAS (NICNAS
Page 401 of 723
-------
2012b; ECH.A. 2008b) for this exposure scenario. Releases of HBCD to the aquatic environment are due
to the activity of the processing of EPS foam from imported EPS resin beads.
Within this exposure scenario, there are 12 exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the 50th percentile surface water concentration predictions, a
majority of the RQs are greater than one, with 12, 11 and nine RQs greater than one when using the
acute, algae and chronic COCs, respectively. Based on the 10th percentile surface water concentration
predictions, all of the exposure sub-scenarios have RQs greater than one based on the acute, algae and
chronic COCs.
The 50th percentile sediment HBCD concentration predictions resulted in eight RQs greater than one
using either the 11- or 128-d HBCD half-lives. Based on the 10th percentile sediment HBCD
concentration predictions, all 12 RQs are greater than one using either the 11- or 128-d HBCD half-
lives.
Processing of HBCD to produce SIPs and Automobile Replacement Parts from XPS/EPS Foam
(Exposure scenario 6)
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for
manufacturing of structural insulated panels and automobile replacement parts from XPS/EPS foam.
This process can result in direct releases of HBCD into surface water, or release through POTWs and
onsite wastewater treatment.
EPA assessed a range of daily release rates based on data pertaining to particle generation from the
cutting or sawing of XPS/EPS foam reported in the EURAR (ECHA 2008b) and disposal of trimming
waste given in the Spray Polyurethane Foam Generic Scenario (U.S. EPA. 2018d). The data pertaining to
the number of release days are estimated values that are applicable to the polymer use industry in
general (ECB 2003). There is some uncertainty regarding the extent to which the emission factor data
reported in the EURAR and the data on the number of release days are applicable to these specific
exposure scenario activities that would occur in the U.S. Releases of HBCD to the aquatic environment
are due to the activity of manufacturing of structural insulated panels and automobile replacement parts
from XPS/EPS foam.
Within this exposure scenario, there are 12 exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the 50th percentile surface water concentration predictions,
there are no RQs that are greater than one, using the chronic COC, but there are two and one acute and
algae RQ, respectively, that exceed one. Based on the 10th percentile surface water concentration
predictions, there are eight, seven and four RQs that are greater than one, using the acute, algae and
chronic COCs.
Based on the 50th percentile sediment concentration predictions, there are no RQs greater than one, using
either the 11- or 128-d HBCD half-lives. On the other hand, based on the 10th percentile sediment
concentration predictions, there are two and four RQs that exceed the sediment COC using 11- or 128-d
HBCD half-lives, respectively.
Installation of XPS/EPS Foam Insulation in Residential, Public, and Commercial Buildings, and
Other Structures (Exposure scenario 8)
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for the
installation of XPS/EPS foam insulation in residential, public, and commercial buildings (and other
Page 402 of 723
-------
structures). This process can result in direct releases of HBCD into surface water, or release through
POTWs.
EPA assessed a range of daily release rates based on an estimated HBCD throughput at residential and
commercial buildings, emission data pertaining to particle generation from the cutting or sawing of
XPS/EPS foam reported in the EURAR (ECH.A. 2.008b) and disposal of trimming waste given in the
Spray Polyurethane Foam Generic Scenario ( Xd). The data pertaining to the number of
release days are estimated values given in the Spray Polyurethane Foam (SPF) Generic Scenario for
operating days at construction sites. There is some uncertainty regarding the extent to which the
emission factor data reported in the EURAR and installation days for SPF are applicable to this specific
exposure scenario that would occur in the U.S. Releases of HBCD to the aquatic environment are due to
the activity of installation of XPS/EPS foam insulation in residential, public and commercial buildings
(and other structures).
Within this exposure scenario, there are four exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the 50th percentile surface water concentration predictions,
there are no RQs that are greater than one, when using the chronic COC, but there is one RQ greater
than one, when using either the acute or algae COC. Based on the 10th percentile surface water
concentration predictions, there is one RQ greater than one, when using the chronic COC, and two RQs
greater than one when using either the acute or algae COC.
In regard to the predicted sediment HBCD concentrations, there are no RQs greater than one based off
of the 50th percentile sediment concentrations, using either the 11- or 128-d HBCD half-life, whereas
there is one RQ greater than one based off of the 10th percentile sediment concentration prediction when
using the 128-d half-life.
Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures (Exposure scenario 9)
Section 2.2 of this document describes how EPA estimated releases from sites at which structures
containing XPS/EPS foam insulation are demolished. This activity can result in releases of HBCD to
air, surface water, and/or POTWs.
EPA assessed a range of daily release rates based on particle generation factors pertaining to the manual
breaking of XPS/EPS foam boards and the cutting of XPS/EPS foam boards with a knife followed by
manual breaking.
Within this exposure scenario, there are four exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the 50th percentile surface water concentration predictions,
there are no RQs that are greater than one, when using the chronic COC, but there is one RQ greater
than one, when using either the acute or algae COC. Based on the 10th percentile surface water
concentration predictions, there is one RQ greater than one, when using the chronic COC, and two RQs
greater than one when using either the acute or algae COC.
In regard to the predicted sediment HBCD concentrations, there are no RQs greater than one based off
of either the 10th or 50th percentile sediment concentrations, using either the 11- or 128-d HBCD half-
life.
Recycling of EPS Foam and Reuse ofXPS Foam (Exposure scenario 10)
Page 403 of 723
-------
Section 2.2 of this document describes how EPA estimated releases from sites at which HBCD is
processed for the industrial recycling of EPS foam and the reuse of XPS foam. This process can result in
direct releases of HBCD into surface water, or release through POTWs and onsite wastewater treatment.
EPA assessed a range of daily release rates based on the emission data like the Manufacturing of EPS
foam from EPS resins as stated earlier in this section with the exclusion of releases from trimming
waste. There is some uncertainty regarding the extent to which the emission factor data and the data on
number of release days are applicable to this specific exposure scenario. Releases of HBCD to the
aquatic environment are due to the activity of the recycling of EPS foam and reuse of XPS foam.
Within this exposure scenario, there are 12 exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the 50th percentile surface water concentration predictions,
there are no RQs that are greater than one, when using the chronic COC, but there are four and two RQs
greater than one when using the acute or algae COC, respectively. Based on the 10th percentile surface
water concentration predictions, there are eight, six and two RQs greater than one, when using the acute,
algae and chronic COC, respectively.
Regarding the predicted sediment HBCD concentrations, there are no RQs greater than one based off the
50th percentile sediment concentrations, using either the 11- or 128-d HBCD half-life, whereas there are
two RQs greater than one based on the 10th percentile sediment concentration predictions, using either
the 11- or 128-d HBCD half-life.
Use of Solder/Flux Pastes (Exposure scenario 12)
Section 2.2 of this document describes how HBCD is processed at industrial sites specifically for the use
of solder or flux pastes. This process can result in the release of HBCD through POTWs and onsite
wastewater treatment.
EPA assessed a range of daily release rates based on estimated emissions from the use of solder paste
reported in the OECD ESD on Chemicals Used in the Electronics Industry ( D 2010a). The data
pertaining to the number of release days are estimated values that are applicable to the electronics
industry in general (ECB 2003). There is some uncertainty regarding the extent to which the emission
factor data reported in general for solder paste use in the ESD and the data on number of release days are
applicable to the current use of HBCD-containing flux/solder paste. Releases of HBCD to the aquatic
environment are due to the activity of the use of solder or flux pastes.
Within this exposure scenario, there are eight exposure sub-scenarios with predicted surface water and
sediment HBCD concentrations. Based on the 50th percentile surface water concentration predictions,
there are no RQs greater than one, whereas there are two RQs greater than one based on the 10th
percentile surface water concentration predictions using the acute COC.
All risk estimates are less than one when using the 10th or 50th percentile predictions for sediment
concentrations of HBCD, using either the 11- or 128-d HBCD half-life.
4,1.4,3 Risk Characterization for Aquatic and Terrestrial Ecosystems based on
Exposure via Potential Trophic Transfer of HBCD
As presented in Section 3.1.3, the trophic transfer potential of HBCD is evaluated for a representative
terrestrial and aquatic predator; the potential risk to terrestrial and aquatic organisms can be qualitatively
evaluated using this methodology. Table 4-8 resents the RQs based on the hazard value for American
kestrel ( lie et al. 2011) and measured biomonitoring data regarding the prey of American kestrel and
Page 404 of 723
-------
osprey. Specifically in regard to American kestrel, reproductive toxicity was observed in female kestrel
exposed to 0.51 mg/kg bw (Fertile et al. 2011). Table 3-2 suggests that American kestrel are exposed to
64.4 ng HBCD per day through the consumption of small mammals (i.e., mice), however mice only
comprise approximately a third of American kestrel diet; it is likely that these calculations underestimate
HBCD uptake through diet. On the other hand, because osprey diet is 100% characterized by the
consumption of fish, a variety of T2 and T3 fish biomonitoring data (rainbow trout, northern snakehead,
brown trout, eel) were used to examine whether osprey could consume enough fish to be exposed to
concentrations of HBCD that would exceed the allometrically-scaled kestrel reproductive hazard effect
concentration (Fertile et al. 2011). RQs for osprey were only greater than one when their diet comprised
of brown trout and eel that were sampled downstream of HBCD manufacturing plants, demonstrating
that prey type and availability, as well as their proximity to areas with higher concentrations of HBCD in
environmental media {i.e., industrial facilities) will be important variables to consider when
characterizing exposure.
RQs presented in Table 4-9suggest that in addition to the high likelihood for HBCD to bioconcentrate in
rainbow trout and earthworms, through both diet and media exposure, HBCD will exceed thresholds of
hazard, respectively. The PBT characteristics of HBCD also may result in changes in population-level
dynamics in aquatic and terrestrial ecosystems should organisms similar to rainbow trout and
earthworms be chronically exposed to HBCD.
Using predicted surface water and porewater HBCD concentrations, risk quotients for terrestrial
mammals were calculated using KABAM (vl); risk quotients for terrestrial birds could not be calculated
due to hazard data incapability with model outputs. Although the risk estimates for soil organisms {i.e.,
earthworms) are less than one, the potential for both dietary and environmental exposure to HBCD is
likely; exposure to HBCD is prolonged given the PBT characteristics of HBCD. Conflicting risk
estimates for earthworms using IIOAC modeled air deposition HBCD concentrations or laboratory
measured bioconcentration data suggest that the release of HBCD through exposure-specific scenarios
may not result in additional risk. However, current background soil concentrations of HBCD, without
the potential releases of HBCD from the various modeled exposure scenarios, already pose a risk to
earthworms, and potentially other terrestrial organisms, near industrial facilities or potentially receive
biosolid application from areas downstream of industrial facilities areas with high concentrations of
HBCD.
4.1.4.4 Targeted Sensitivity Analysis
Section 2.2.14 describes the context behind conducting a targeted sensitivity analysis based on
production volume. Briefly, due to the uncertainty with the imported volume and resulted estimates of
environmental releases and exposures to the general population and the environment, a targeted
sensitivity analysis on the impact of import volumes on environmental risk estimates is conducted in this
section. The exposure scenarios considered in the sensitivity analysis represent those that resulted in the
highest estimates of releases on a daily basis and include scenarios that rely on both industry data and
OECD ESDs. Specifically, those exposure scenarios are listed below with their respective discussions
on risk estimates for surface water and sediment concentrations of HBCD.
Originally as presented above in Section 4.1.4.2, all nine exposure scenarios with estimated water
releases containing HBCD were predicted to have production volumes up to 100,000 lbs/yr. The purpose
of the sensitivity analysis is to evaluate how the model input parameter of production volume may
impact the predicted surface water and sediment HBCD concentrations. In addition to deriving risk
quotients by using predicted surface water and sediment HBCD concentrations based on a production
volume of 100,000 lbs/yr, risk quotients were also derived using the production volumes of 50,000 and
Page 405 of 723
-------
25,000 lbs/yr for the three processing exposure scenarios: exposure scenario #1: Repackaging of import
containers, exposure scenario #2: Manufacturing of XPS Foam using XPS Masterbatch, and exposure
scenario #3: Manufacturing of EPS Foam from Imported EPS Resin Beads. Unlike exposure scenarios 3
and 5, exposure scenario 1 (Repackaging of Import Containers) does not have direct releases into
surface water, therefore resulting in less total exposure scenarios. As stated above, the same sources of
information regarding the range of daily release rates, emission factors, number of release days per year,
and surrounding uncertainties outlined for each exposure scenario apply to the same exposure scenarios
(1,3, and 5) below.
For two processing exposure scenarios (Manufacturing of XPS Foam using XPS Masterbatch, and
Manufacturing of EPS Foam from Imported EPS Resin Beads), risk quotients were also calculated based
on predicted surface and pore water HBCD concentrations for terrestrial avian and mammalian wildlife.
Specifically, two model sub-scenarios for these exposure scenarios (3.3 and 5.7) were selected because
despite both having predicted direct releases of HBCD into surface water, the water releases vary
greatly, with model sub-scenario 5.7 having greater HBCD surface water, pore water and sediment
concentrations than 3.3. These sub-scenarios were selected to provide a range in risk estimates that
reflect lower and higher water releases of HBCD. The purpose of using KABAM was to estimate HBCD
risk to terrestrial organisms that prey on aquatic wildlife.
4.1.4.4.1 Summary of Ranges of RQs: Production Volume
The below table provides a range of risk quotients (RQ) that were calculated using predicted surface
water or sediment concentrations for three exposure scenarios: Repackaging of Import Containers,
Processing of HBCD to produce XPS Foam using XPS Masterbatch, and Processing of HBCD to
produce EPS Foam from Imported EPS Resin Beads (Section 2.2.15). The sensitivity analysis evaluates
the impact of production volume on RQ values. However, amongst the exposure sub-scenarios, altering
the production volume only impacted the percentage of RQs to be equal to or greater than one, when
using surface water or sediment concentrations based off the 50th percentile predictions.
Page 406 of 723
-------
Table 4-10. Range of Risk Quotients for Modeled Surface Water and Sediment HBCD Concentrations for Three Conditions of Use
Scenarios Using a Production Volume of 100,000, 50,000, and 25,000 lbs/yr
The bolded and red highlighted values denote when half or more of the sub-scenario risk quotients (RQ) modeled for each exposure scenario are >1. If the
predicted surface water or sediment concentration was 0, or if the calculated RQ was< 0.005, the RQ was rounded to 0.
Exposure
Scenario
Production
Volume (lbs
/ year)
Surface Water
Sediment
Acute
Algae
Chronic
11-d half-life
128-d half-life
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
10th
percentile
50th
percentile
Repackaging of
Import
Containers (1)
100,000
4.3-189
0.09-24.2
1.72-75.6
0.04-9.68
3.5-21.22
0.07-2.26
0.87-4.61
0.02-0.56
2.29-
11.91
0.05-1.26
50,000
3.93-180.5
0.09-
23.38
1.57-72.2
0.04-9.35
I.99-
II.44
0.04-1.21
0.48-2.85
0.01-0.34
1.23-7.45
0.03-0.79
25,000
3.65-
192.25
0.09-10
1.46-76.9
0.04-10
0.97-10
0.02-1.16
0.22-1.68
0.01-0.21
0.57-3.66
0.01-0.4
Processing of
HBCD to
produce XPS
Foam using
XPS
Masterbatch
(3)
100,000
0.76-275
0.01-7.33
0.3-110
0.01-2.97
0.04-13.5
0.001-
0.33
0.01-2.22
0-0.06
0.02-2.97
0-0.08
50,000
0.38-138.5
0.01-3.7
0.15-55.4
0-1.48
0.02-6.81
0-0.17
0.01-1.12
0-0.03
0.01-1.5
0-0.04
25,000
0.19-69.25
0-1.85
0.08-27.7
0-0.74
0-3.41
0-0.08
0-0.56
0-0.1
0.01-0.75
0-0.02
Processing of
HBCD to
produce EPS
Foam from
Imported EPS
Resin Beads (5)
100,000
89.5-9,900
2.2-262.5
35.8-3,960
0.88-105
33.57-
563.55
0.70-12
8.73-
143.31
0.21-3.52
22.68-
361.78
0.48-7.77
50,000
44.75-
9850
1.10-
262.50
17.9-3940
0.44-105
16.76-
515.59
0.35-12
4.36-
89.17
0.11-2.23
11.34-
201.27
0.24-4.39
25,000
18.23-
9825
0.55-
262.5
7.29-3930
0.22-105
4.46-
491.61
0.18-12
1.15-
79.62
0.05-2.03
3-135.03
0.12-3.13
Page 407 of 723
-------
Repackaging of Import Containers (Exposure scenario 1)
Surface Water:
Whether the 10th or 50th percentile surface water concentrations of HBCD were used, there are risk
quotients (RQs) greater than one using the acute, algae and chronic COCs for all three production
volumes. Based on the 10th percentile surface water HBCD concentrations, at least half of the RQs are
greater than one, using the acute, algae and chronic COC, regardless of the production volume. Based on
the 50th percentile surface water HBCD concentrations, at least half of the RQs are greater than one,
using the acute COC for all three production volumes.
Sediment:
Similar to the predicted surface water HBCD concentrations, based on the 10th percentile sediment
concentrations of HBCD, the sediment chronic RQs are greater than one using both the 11- and 128-d
HBCD half-lives. In regard to the 50th percentile sediment concentrations of HBCD, there are only RQs
greater than one based on the 128-d HBCD half-life, using a production volume of 100,000 lbs/yr.
Processing of HBCD to produce XPS Foam using XPS Masterbatch (Exposure scenario 3)
Surface Water:
For all three production volumes, based on the 10th percentile surface water concentrations of HBCD
were used, there are risk quotients (RQs) greater than one using the acute, algae and chronic COCs for
all three production volumes. Specifically, over half of the derived RQs are greater than one using the
10th percentile surface water concentrations of HBCD, based on the acute and algae COCs for all three
production volumes; over half of the chronic RQs are greater than one using the production volume of
100,000 lbs/yr. Based on the 50th percentile surface water HBCD concentrations, there are RQs greater
than one for all three production volumes, using the acute COC, whereas RQs are only greater than one
for both 100,000 and 50,000 lbs/yr, when using the algae COC. There are no RQs greater than one based
on the 50th percentile surface water concentrations of HBCD, when using the chronic COC.
Sediment:
There are RQs greater than one, when using the 10th percentile sediment concentrations of HBCD and
either 11- or 128-d HBCD half-lives, for production volumes of 100,000 and 50,000 lbs/yr (there were
no RQs greater than one for the lowest production volume of 25,000 lbs/yr). Based on the 50th percentile
sediment concentrations of HBCD, there were no RQs greater than one using either HBCD half-lives for
any of the three production volumes.
Risk Estimates for Terrestrial Organisms (KABAM outputs):
For all three production volumes, based on the 50th percentile surface water and sediment
concentrations, there are no instances of risk estimates greater than one (indicating risk) for small and
large mink and small river otters. See Appendix 0. Estimates for terrestrial organisms were not modeled
using the 10th percentile surface water and sediment concentrations, but it is likely that there may be
risk estimates greater than one using more conservative media concentrations.
Processing of HBCD to produce EPS Foam from Imported EPS Resin Beads (Exposure scenario 5)
Page 408 of 723
-------
Surface Water:
Whether the 10th or 50th percentile surface water concentrations of HBCD were used, there are risk
quotients (RQs) greater than one using the acute, algae and chronic COCs for all three production
volumes. Additionally, regardless of the production volume or whether the 10th or 50th percentile surface
water concentrations of HBCD were used, more than half of the calculated RQs for this exposure
scenario have RQs greater than one.
Sediment:
Similarly, whether the 10th or 50th percentile sediment concentrations of HBCD were used, there are
RQs greater than one using either HBCD half-lives, for all three production volumes. Based on the
either half-lives, at least half of the calculated RQs are greater than one for all three production volumes,
using the 10th percentile sediment concentrations of HBCD. Based on the 50th percentile sediment
concentrations of HBCD, at least half of the calculated RQs are greater than one for all three production
volumes, using the 128-d HBCD half-life, whereas this is only the case for the production volume of
100,000 lbs/yr, using the 11-d HBCD half-life.
Risk Estimates for Terrestrial Organisms:
For all three production volumes, based on the 50th percentile surface water and sediment
concentrations, there are risk estimates greater than one for small and large mink, and small river otters
(nine out of 15 risk estimates). See Appendix J. 1.2.3.
Page 409 of 723
-------
4.2 Human Health Risk
4.2.1_ Risk Estimation Approach
The use scenarios, populations of interest and toxicological endpoints used for acute and chronic
exposures are presented in Table 4-11.
Table 4-11. Use Scenarios, Populations of Interest and Toxicological Endpoints Used for Acute
and Chronic Exposures
Population of Interest and
Exposure Scenario
Workers:
Acute- Adult workers (>21 vears old) and female workers of
reproductive age (>16 year to less than 50 years old) exposed to HBCD
for a single 8-hr exposure
Chronic- Adult workers (>21 vears old) and female workers of
reproductive age (>16 year to less than 50 years old) exposed to HBCD
for the entire 8-hr workday for 260 days per year for 40 working years
Occupational Non-User:
Acute or Chronic- Adult workers (>21 vears old) and female workers
of reproductive age (>16 year to less than 50 years old) exposed to
HBCD indirectly by being in the same work area of the building
General Population (Background Exposure):
Acute or Chronic - <1 vear, 1 to <2 vears, 2 to <3 vears, 3 to <6 vears,
6 to <11 years, 11 to <16 years, 16 to <70 years
Highly Exposed Population (Near-Facility, Consumers):
Acute or Chronic - <1 vear, 1 to <2 vears, 2 to <3 vears, 3 to <6 vears,
6 to <11 years, 11 to <16 years, 16 to <70 years
Health Effects,
Concentration and Time
Duration
Units for Non-Cancer Point of Departures (POD): mg/kg-day
Non-Cancer Health Effects: 2
Acute- Thvroid hormone effects and developmental effects
Chronic- Thvroid hormone effects, liver effects, reproductive effects,
and developmental effects
Uncertainty Factors (UF)
used in Non-Cancer Margin
of Exposure (MOE)
calculations
Benchmark MOEs: Vary by endpoint
Benchmark MOE3 = (UFs) x (UFa) x (UFh) x (UFl)
1 Adult workers (>21 years old) include both female and male workers.
2Female workers of reproductive age (>16 to less than 50 years old) are the population of interest for reproductive and
developmental effects becausel6 is the basic minimum ase for employment (httos://www.dol.sov/asencies/whd/fact-
sheets/43-child-labor-non-aericulture) and 50 is averase ase of menopause. For other health effects (e.s.. liver, kidnev. etc.).
female or male workers were assumed to be the population of interest. Estimation of the risk was calculated for each group
based on differences in bodv weieht as described in the Exposure Factors Handbook (U.S. EPA 201 lb).
3UFs=subchronic to chronic UF; UFA=interspecies UF; UFH=intraspecies UF; UFl=LOAEL to NOAEL UF.
4OSHA defines chronic workplace exposures to be 8 hr-workday for 240 days for 45 years. EPA typically uses 250 days and
calculated 50th (31 vears) and 95th percentile (40 vears) workins vears usins data from U.S. Census Bureau (Census Bureau
2016) see Appendix E.7.
Page 410 of 723
-------
The EPA uses a Margin of Exposure (MOE) approach to assessing non-cancer risk. The MOE is the
ratio of the point of departure (POD) dose divided by the human exposure dose. The MOE is compared
to the benchmark MOE. The MOE estimate was interpreted as human health risk if the MOE estimate
was less than the benchmark MOE (i.e., the total UF). On the other hand, the MOE estimate indicated
negligible concerns for adverse human health risks if the MOE estimate exceeded the benchmark MOE.
Acute or chronic MOEs (MOEacute or MOEchronic) were used in this assessment to estimate non- cancer
risks using Equation 4-1.
Equation 4-1. Equation to Calculate Non-Cancer Risks Following Acute or Chronic Exposures
Using Margin of Exposures
Non — cancer Hazard value (POD)
MOEacuteorchronlc = Human Exposure
Where:
MOE = Margin of exposure (unitless)
Hazard Value (POD) = HED (mg/kg)
Human Exposure = Exposure estimate (in mg/kg) from occupational exposure assessment
= Exposure estimate (in mg/kg) from general population and highly
exposed population exposure assessment
Acute Absorbed Doses (AADs) were used to calculate occupational non-cancer risks following acute
exposure and Chronic Absorbed Doses (CADs) were used for occupational non-cancer risks following
chronic exposure (see Section 2.4.1.1 for description). Acute Dose Rates (ADRs) were used to calculate
non-cancer risks to the general population following acute exposure (see Section 2.4.2 for description
and equations by media type).
EPA used margin of exposures (MOEs)20 to estimate risks from acute or chronic exposures for non-
cancer based on the following:
1. the lowest HEDs within each non-cancer health effects domain reported in the literature;
2. the endpoint/study-specific UFs applied to the HEDs per EPA RfD/RfC Guidance (U.S. EPA.
2002): and
3. the exposure estimates calculated for HBCD uses examined in this risk assessment (see Section 32
-Exposures).
MOEs allow for the presentation of a range of non-cancer risk estimates. The occupational exposure
scenarios (OES) considered both acute and chronic exposures. As discussed in Section 2.4.1.1,
inhalation exposures to occupational non-users (ONUs) were not quantified due to lack of adequate data
but are expected to be less than worker exposures. Different adverse endpoints were determined to be
appropriate based on the expected exposure durations. For non-cancer effects, risks for acute effects
(offspring loss) were evaluated for acute (short-term) exposures, whereas risks for thyroid effects were
evaluated for repeated (chronic) exposures to HBCD. EPA discusses other effects in Sections 3.2.3 and
3.2.4.
20 Margin of Exposure (MOE) = (Non-cancer hazard value, POD) (Human Exposure). Equation 4-1. The benchmark MOE
is used to interpret the MOEs and consists of the total UF as described in Section 3.2.5.3.
Page 411 of 723
-------
For general population (background) risks for only chronic exposure scenarios were considered because
they represent a steady-state, while for highly exposed populations (living near a facility) both acute and
chronic exposures were considered. Risks to the highly exposed population were associated with
specific COUs and exposure scenarios, while general population exposures represented baseline steady-
state exposures from persistent HBCD in environmental media. Different adverse endpoints were used
based on the expected exposure durations. For non-cancer effects, risks for developmental effects were
evaluated for acute (short-term) exposures, whereas risks for other adverse effects (toxicity to the
thyroid, liver, developmental effects, and the female reproductive system) were evaluated for repeated
(chronic) exposures to HBCD. For occupational exposure calculations, mg/kg values were used to
calculate MOEs for risk estimates following acute and chronic exposures.
The total UF for each non-cancer POD was the benchmark MOE used to interpret the MOE risk
estimates for each use scenario. The MOE estimate was interpreted as a potential human health risk if
the MOE estimate was less than the benchmark MOE {i.e., the total UF). On the other hand, the MOE
estimate indicated negligible concerns for adverse human health effects if the MOE estimate exceeded
the benchmark MOE. Typically, the larger the MOE, the more unlikely it is that a non-cancer adverse
effect would occur.
Risk estimates in the form of MOE values were calculated for all of the studies for each health effects
domain that EPA considered suitable for the Risk Evaluation of acute and chronic exposure scenarios.
The studies selected for dose-response assessment and derivation of PODs examined oral administration
of HBCD. These oral PODs are directly applicable to risks from oral exposures such as via soil, drinking
water, and diet. For inhalation exposure, EPA considered the quantification of incidental ingestion of
particulates that would result from exposure to HBCD dust in occupational, environmental, or
residential settings. It is assumed that any inhaled particulate would either be absorbed through the lungs
or swallowed and subsequently absorbed in the GI tract. Based on available toxicokinetic data, EPA
conservatively assumes 100% absorption through the lungs and GI tract, although the majority of HBCD
particles are likely to deposit in the upper respiratory tract and be ingested. EPA did not identify any
respiratory-specific hazards associated with HBCD exposure. Since all HBCD hazards evaluated
through dose-response analysis involve systemic toxicity, it is irrelevant for the purposes of this
assessment whether HBCD is absorbed through the lungs or GI tract. Therefore, EPA used total
inhalation exposure values (as opposed to only respirable) for risk estimation.
For dermal exposure, EPA performed route-to-route extrapolation from oral toxicity based on similar
principles to those described in the EPA Guidance Document Risk Assessment Guidance for Superfund
Volume I: Human Health Evaluation Manual (Part E, Supplemental Guidance for Dermal Risk
Assessment) ( 104). All risk calculations for dermal exposure incorporate an adjustment for
6.5% absorption, based on available toxicokinetic data (see Section 3.2.2).
4.2.1.1 Representative Points of Departure for Use in Risk Estimation
In order to more succinctly present the most important risk estimates, occupational risks were assessed
using a single endpoint representative of each health domain. EPA considers all of the endpoints
identified in Table 3-12 to be relevant to human health hazard from HBCD exposure. Therefore,
occupational risk estimates are presented for only those endpoints representing the most sensitive and
robust data within each health domain, with the presumption that evaluation of risks for these endpoints
also account for all other less sensitive yet relevant endpoints. These PODs are presented in Table 4-12.
For complete occupational MOE tables displaying risk estimates for all endpoints, see [Risk Evaluation
for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental File: Occupational Risk Calculator. (U.S.
19s) "I.
Page 412 of 723
-------
Table 4-12. Most Sensitive Endpoints From Each Health Domain Used for Risk Estimation
Toxicity Endpoint
PODhed
(mg/kg-d)
Benchmark
MOE
Effects applicable to acute exposure scenarios
Thyroid
Decreased maternal T4 (Ema et al. 2008)
22.5
30
Developmental
F2 generation offspring loss (Ema et al. 2008)
9.03
100
Effects following chronic exposure scenarios
Thyroid
Decreased T4 (Ema et al. 2008)
1.68
300
Liver
Increased relative liver weight and vacuolization
(WIL Research 2001)
24
1000
Female
Reproductive
Reduced primordial follicles (Ema et al. 2008)
0.689
30
Developmental
F2 generation offspring loss (Ema et al. 2008)
9.03
100
Risk estimates are shown for the representative POD of each health domain following acute or chronic
exposure, as shown below. As described above in Section 3.2.5.2.1, developmental toxicity outcomes
may result from a single acute exposure during a critical window of development. Given this, the most
relevant lifestage in the human population would be women of child-bearing age. However, due to
uncertainty in the mode of action for HBCD developmental toxicity (e.g., outcomes could be exclusively
due to effects on the exposed unborn fetus in utero or they could also result from permanent damage to
eggs) and the possibility of a bioaccumulative effect following a future acute exposure, risks for
developmental toxicity were characterized for all lifestages.
4.2.2 Risk Estimation for Workers
The tables and narratives below describe the conclusions of the risk estimation via inhalation or dermal
exposure for each use scenario following acute or chronic exposures. Risks were calculated for average
adult workers as well as for women of reproductive age. Results presented below are for average adult
workers. MOEs are approximately 10% lower for women of reproductive age compared to average adult
workers, and differences in risk conclusions are identified in the tables and risk characterization
narratives when applicable. For a complete list of all risk calculations, see [Risk Evaluation for Cyclic
Aliphatic Bromide Cluster (HBCD), Supplemental File: Occupational Risk Calculator (U.S. EPA
2019s)]. The risk estimates in the tables below and in the supplemental file are presented only for OES
associated with ongoing manufacturing or import. Risk estimation for recycling of electronics waste
containing HIPS is provided in Section 4.2.2.5.
EPA notes that OSHA requires employers apply the hierarchy of controls as discussed in Section 2.4.1.1
which first prioritizes elimination, substitution, engineering and administrative controls, and then if not
feasible to address the hazard, the implementation of a respiratory protection program. Adjusted MOEs
were not calculated based on glove protection because EPA does not expect any level of dermal
exposure to HBCD following proper use of gloves impervious to HBCD. As discussed in Section
2.4.1.1, impervious gloves, if worn on clean hands and replaced when contaminated or compromised,
are expected to provide employees with protection from HBCD. HBCD is a solid particulate and would
not be expected to permeate through gloves (unlike certain solvents). Some examples of impervious
Page 413 of 723
-------
gloves are nitrile, butyl rubber, polyvinyl chloride, and polychloroprene. EPA did not identify any
sufficient data or applicable model that can be used to adequately estimate inhalation exposure to
particulates for ONUs. EPA assumes that exposures to ONUs would be lower than those for workers
(Section 2.4.1.1).
PPE usage up to APF = 50 and including impervious gloves is assumed for all OES except installation
and demolition (& disposal) of XPS/EPS foam insulation, for which respirator use is not assumed. Table
4-13 presents this information below. Risk estimates are presented displaying the APFs expected to
mitigate risk for the exposure scenario (e.g., acute inhalation) in the sections below.
Table 4-13. Inhalation Exposure Data Summary and Respirator Use Determination
Occupational Exposure
Scenario
Inhalation
Exposure
Approach
Number
of Data
Points
Model Used
Approach for ONUs
Respirator
Use
Industrial or
Commercial
OES
Repackaging of Import
Containers
Monitoring
data
10 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial
Compounding of
Polystyrene Resin to
Produce XPS Masterbatch
Monitoring
data
16 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial
Processing of HBCD to
Produce XPS Foam using
XPS Masterbatch
Monitoring
data
9 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial
Processing of HBCD to
Produce XPS Foam using
HBCD Powder
Monitoring
data
16 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial
Processing of HBCD to
Produce EPS Foam Using
Imported EPS Resin
Beads
Monitoring
data
9 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial
Processing of HBCD to
Produce SIPs and
Automobile Replacement
Parts from XPS/EPS
Foam
Monitoring
data
9 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial
Page 414 of 723
-------
Occupational Exposure
Scenario
Inhalation
Exposure
Approach
Number
of Data
Points
Model Used
Approach for ONUs
Respirator
Use
Industrial or
Commercial
OES
Installation of Automobile
Replacement Parts
N/A-
inhalation
exposure
not
assessed
N/A-
inhalation
exposure
not
assessed
N/A-
inhalation
exposure not
assessed
N/A - inhalation exposure
not assessed
N/A-
inhalation
exposure
not
assessed
Commercial
Installation of XPS/EPS
Foam Insulation in
Residential, Public and
Commercial Buildings,
and Other Structures
Monitoring
data
9 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
Not
expected to
use
respirators
Commercial
Demolition and Disposal
of XPS/EPS Foam
Insulation Products in
Residential, Public and
Commercial Buildings,
and Other Structures
N/A-
modeling
only
N/A-
modeling
only
OSHA
PNOR PEL
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
Not
expected to
use
respirators
Commercial
Recycling of EPS Foam
and Reuse of XPS foam
Monitoring
data
9 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial/
Commercial
Formulation of
Flux/Solder Pastes
Monitoring
data
16 (8-hr
TWA)
N/A-
monitoring
data only
Exposures to ONUs are
assumed to be less than those
for workers. Risk estimates
for inhalation exposure to
ONUs were not quantified
May use
respirators
Industrial
Use of Flux/Solder Pastes
N/A-
inhalation
exposure
not
assessed
N/A-
inhalation
exposure
not
assessed
N/A-
inhalation
exposure not
assessed
N/A - inhalation exposure
not assessed
N/A-
inhalation
exposure
not
assessed
Industrial/
Commercial
EPA did not quantitatively assess occupational exposure associated with Land Disposal of Formulated
Products and Articles. As described in Section 2.4.5.3, EPA assumes a low concentration in municipal
waste disposed of at a landfill, and workers are not expected to be exposed to products or articles
containing HBCD on a regular basis. Therefore, acute and chronic risks to workers from this COU are
not expected in most circumstances. In worst-case scenarios where municipal waste is shredded,
exposures may be elevated and there may be a greater possibility of acute risks on days where HBCD-
containing products or articles are present. Therefore risks cannot be ruled out despite being of lower
likelihood. This uncertainty is further described in Section 4.3.2.3.
Page 415 of 723
-------
4,2,2,1 Occupational Risk Estimation for Non-Cancer Effects Following Acute
Inhalation Exposures
Risks to workers were estimated for non-cancer effects following acute inhalation exposures. Table 4-14
displays MOE values for all occupational scenarios and human health hazards associated with acute
exposure, including results assuming either respiratory protection of APF = 5 or APF =10. Risks were
not identified for any scenario assuming respiratory protection of APF = 5 or greater. Inhalation risks
were not estimated for the following exposure scenarios because worker inhalation exposures are not
expected: Installation of Automobile Replacement Parts and Use of Flux/Solder Paste.
Page 416 of 723
-------
Table 4-14. Risk Estimation for Non-Cancer Effects Following Acute Inhalation Exposures
Occupational Exposure Scenario -
Inhalation Exposure
[Benchmark MOE = 100]
[Benchmark MOE = 30]
PODhed (mg/kg) = 9.03
Developmental Toxicity
F2 offspring loss
(Ema et al. 2008)
PODhed (mg/kg) = 22.5
Thyroid Hormone Changes
Decreased maternal T4
(Ema et al. 2008)
High-End Exposure
Central Tendency Exposure
High-End Exposure
Central Tendency Exposure
No
Protection
APF
= 5
APF
= 10
No
Protection
APF
= 5
APF
= 10
No
Protection
APF
= 5
APF
= 10
No
Protection
APF
= 5
APF
= 10
Repackaging of Import Containers
38
191
382
81
406
812
95
476
952
202
1011
2022
Compounding of Polystyrene Resin to
Produce XPS Masterbatch
29
144
289
58
289
578
72
360
720
144
720
1440
Processing of HBCD to produce XPS Foam
Using XPS Masterbatch
328
1642
3284
903
4515
9030
818
4091
8182
2250
11250
22500
Processing of HBCD to produce XPS Foam
Using HBCD Powder
29
144
289
58
289
578
72
360
720
144
720
1440
Processing of HBCD to produce EPS Foam
Using Imported EPS Resin Beads
328
1642
3284
903
4515
9030
818
4091
8182
2250
11250
22500
Processing of HBCD to produce SIPs and
Automobile Replacement Parts from
XPS/EPS Foam
328
1642
3284
903
4515
9030
818
4091
8182
2250
11250
22500
Installation of Automobile Replacement Parts
--
--
--
--
--
--
--
--
--
--
--
--
Installation of XPS/EPS Foam Insulation in
Residential, Public and Commercial
Buildings, and Other Structures
328
1642a
3284a
903
4515a
9030a
818
4091a
8182a
2250
11250a
22500a
Demolition and disposal of XPS/EPS Foam
Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures
241
1204a
2408a
688
3440a
6880a
600
3000a
6000a
1714
8571a
17143a
Recycling of EPS Foam
328
1642
3284
903
4515
9030
818
4091
8182
2250
11250
22500
Formulation of Flux / Solder Paste
29
144
289
58
289
578
72
360
720
144
720
1440
Use of Flux / Solder Paste
--
--
--
--
--
--
--
--
--
--
--
--
- As discussed in Section 2.4.1.1 EPA expects potential inhalation exposure of an Occupational Non-User (ONU) in the case of some of the conditions of use but EPA did not
quantitatively assess these exposures due to lack of adequate data. EPA assumes that these exposures would be lower than the exposures of the corresponding workers.
- Bold/shaded text indicates MOE is less than the benchmark MOE. Non-bold text indicates the MOE is greater than the benchmark MOE. ~ indicates that exposures are not
expected during this exposure scenario.
a EPA is presenting MOEs for respiratory PPE up to APF = 10 as a what-if scenario, however EPA believes that workers in these OES are unlikely to wear respirators.
Page 417 of 723
-------
4,2,2.2 Occupational Risk Estimation for Non-Cancer Effects Following Chronic
Inhalation Exposures
Risks to workers were calculated for non-cancer effects following chronic inhalation exposures. Table
4-15 displays MOE values for all occupational scenarios and human health hazards associated with
chronic exposure, including results assuming either respiratory protection of APF =10 and APF = 50.
Risks were not identified for any scenario assuming respiratory protection of APF = 50 or greater.
Inhalation risks were not estimated for the following exposure scenarios because worker inhalation
exposures are not expected: Installation of Automobile Replacement Parts and Use ofFlux/Solder Paste.
Page 418 of 723
-------
Table 4-15. Risk Estimation for Non-Cancer Effects Following Chronic Inhalation Exposures
Occupational Exposure Scenario -
Inhalation Exposure
MARGIN OF EXPOSURE (MOE)
Benchmark MOE
= 300
Benchmark MOE
= 1000
Benchmark MOE = 30
Benchmark MOE
= 100
PODhed (mg/kg) = 1.68
Thyroid Effects
Decreased T4
(Ema et al. 2008)
PODhed (mg/kg) =
24
Liver Toxicity
Increased relative liver weight
and vacuolization
(WIL Research 2001)
PODhed (mg/kg) =
0.689
Female Reproductive
Toxicity
Reduced primordial
follicles
(Ema et al. 2008)
PODhed (mg/kg) =
9.03
Developmental Toxicity
F2 offspring loss
(Ema et al. 2008)
None
APF =
10
APF = 50
None
APF = 10
APF = 50
None
APF =
10
APF = 50
None
APF = 10
APF = 50
Repackaging of import containers (HE)
10
104
519
148
1483
7416
4
43
213
56
558
2790
Repackaging of import containers (CT)
39
394
1969
562
5624
28122
16
161
807
212
2116
10581
Compounding of Polystyrene Resin to
Produce XPS Masterbatch (HE)
33
327*
1635
467
4672
23360
13
134
671
176
1758
8789
Compounding of Polystyrene Resin to
Produce XPS Masterbatch (CT)
112
1121
5606
1602
16018
80091
46
460
2299
603
6027
30134
Processing of HBCD to produce XPS
Foam Using XPS Masterbatch (HE)
1394
13936
69682
19909
199091
995455
572
5716
28578
7491
74908
374540
Processing of HBCD to produce XPS
Foam Using XPS Masterbatch (CT)
6813
68133
340667
97333
973333
4866667
2794
27943
139714
36622
366217
1831083
Processing of HBCD to produce XPS
Foam Using HBCD Powder (HE)
123
1226
6132
1752
17520
87600
50
503
2515
659
6592
32960
Processing of HBCD to produce XPS
Foam Using HBCD Powder (CT)
436
4361
21803
6229
62293
311467
179
1788
8942
2344
23438
117189
Processing of HBCD to produce EPS
Foam Using Imported EPS Resin Beads
(HE)
159
1593
7964
2275
22753
113766
65
653
3266
856
8561
42805
Processing of HBCD to produce EPS
Foam Using Imported EPS Resin Beads
(CT)
786
7862
39308
11231
112308
561538
322
3224
16121
4226
42256
211279
Processing of HBCD to produce SIPs and
Automobile Replacement Parts from
XPS/EPS Foam (HE)
89
892
4460
1274
12742
63709
37
366
1829
856
8561
42805
Page 419 of 723
-------
Occupational Exposure Scenario -
Inhalation Exposure
MARGIN OF EXPOSURE (MOE)
Benchmark MOE
= 300
Benchmark MOE
= 1000
Benchmark MOE = 30
Benchmark MOE
= 100
PODhed (mg/kg) = 1.68
Thyroid Effects
Decreased T4
(Ema et al. 2008)
PODhed (mg/kg) =
24
Liver Toxicity
Increased relative liver weight
and vacuolization
(WIL Research 2001)
PODhed (mg/kg) =
0.689
Female Reproductive
Toxicity
Reduced primordial
follicles
(Ema et al. 2008)
PODhed (mg/kg) =
9.03
Developmental Toxicity
F2 offspring loss
(Ema et al. 2008)
None
APF =
10
APF = 50
None
APF = 10
APF = 50
None
APF =
10
APF = 50
None
APF = 10
APF = 50
Processing of HBCD to produce SIPs and
Automobile Replacement Parts from
XPS/EPS Foam (CT)
461
4611
23053
6586
65865
329323
189
1891
9454
4226
42256
211279
Installation of Automobile Replacement
Parts (HE)
--
--
--
--
--
--
--
--
--
--
--
--
Installation of Automobile Replacement
Parts (CT)
--
--
--
--
--
--
--
--
--
--
--
--
Installation of XPS/EPS Foam Insulation
in Residential, Public and Commercial
Buildings, and Other Structures (HE)
89
892 a
4460 a
1274
12742a
63709a
37
366 a
1829 a
479
4794 a
23971a
Installation of XPS/EPS Foam Insulation
in Residential, Public and Commercial
Buildings, and Other Structures (CT)
487
4867 a
24333a
6952
69524a
347619a
200
1996 a
9980 a
2616
26158a
130792a
Demolition and Disposal of XPS/EPS
Foam Insulation Products in Residential,
Public and Commercial Buildings, and
Other Structures (HE)
65
654 a
3270a
934
9344 a
46720 a
27
268 a
1341 a
352
3516 a
17578a
Demolition and Disposal of XPS/EPS
Foam Insulation Products in Residential,
Public and Commercial Buildings, and
Other Structures (CT)
371
3708 a
18540a
5297
52971a
264853 a
152
1521 a
7603 a
1993
19930a
99651a
Recycling of EPS Foam (HE)
159
1593
7964
2275
22753
113766
65
653
3266
856
8561
42805
Recycling of EPS Foam (CT)
864
8637
43183
12338
123380
616901
354
3542
17710
4642
46422
232109
Formulation of Flux / Solder Paste (HE)
8
78
392
112
1121
5606
3
32
161
42
422
2109
Page 420 of 723
-------
Occupational Exposure Scenario -
Inhalation Exposure
MARGIN OF EXPOSURE (MOE)
Benchmark MOE
= 300
Benchmark MOE
= 1000
Benchmark MOE = 30
Benchmark MOE
= 100
PODhed (mg/kg) = 1.68
Thyroid Effects
Decreased T4
(Ema et al. 2008)
PODhed (mg/kg) =
24
Liver Toxicity
Increased relative liver weight
and vacuolization
(WIL Research 2001)
PODhed (mg/kg) =
0.689
Female Reproductive
Toxicity
Reduced primordial
follicles
(Ema et al. 2008)
PODhed (mg/kg) =
9.03
Developmental Toxicity
F2 offspring loss
(Ema et al. 2008)
None
APF =
10
APF = 50
None
APF = 10
APF = 50
None
APF =
10
APF = 50
None
APF = 10
APF = 50
Formulation of Flux / Solder Paste (CT)
31
307*
1533
438
4380
21900
13
126
629
165
1648
8240
Use of Flux / Solder Paste (HE)
--
--
--
--
--
--
--
--
--
--
--
--
Use of Flux / Solder Paste (CT)
--
--
--
--
--
--
--
--
--
--
--
--
- As discussed in Section 2.4.1.1 EPA expects potential inhalation exposure of an Occupational Non-User (ONU) in the case of some of the conditions of use but EPA
did not assess these exposures due to lack of adequate reasonably available data. EPA assumes that these exposures would be lower than the exposures of the
corresponding workers.
- Bold/shaded text indicates MOE is less than the benchmark MOE. Non-bold/non-shaded text indicates the MOE is greater than the benchmark MOE. HE = High-End
exposure level; CT = Central Tendency exposure level; ~ indicates that exposures are not expected during this exposure scenario.
"None" refers to respiratory protection.
- * indicates that risks are identified for women of reproductive age only. See text below for details.
a EPA is presenting MOEs for respiratory PPE up to APF = 50 as a what-if scenario, however EPA believes that workers in these OES are unlikely to wear respirators.
Page 421 of 723
-------
4.2.2.3 Occupational Risk Estimation for Non-Cancer Effects Following Acute
Dermal Exposures
Risks to workers were calculated for non-cancer effects following acute dermal exposures, assuming
6.5% systemic absorption (see Section 3.2.2). Table 4-16 displays MOE values for all occupational
scenarios and human health hazards associated with acute dermal exposure. As mentioned above,
adjusted MOEs were not calculated based on glove protection because EPA does not expect any level of
dermal exposure to HBCD following proper use of impervious gloves.
Table 4-16. Risk Estimation for Non-Cancer Effects Following Acute Dermal Exposures
Occupational Exposure Scenario -
Dermal Exposure
[Benchmark MOE = 100]
[Benchmark MOE = 30]
PODhed (mg/kg) = 9.03
Developmental Toxicity
F2 offspring loss
(Ema et al. 2008)
PODhed (mg/kg) = 22.5
Thyroid Hormone
Changes
Decreased maternal T4
(Ema et al. 2008)
High-End
Central
Tendency
High-End
Central
Tendency
Repackaging of Import Containers
4
12
9
31
Compounding of Polystyrene Resin to Produce XPS
Masterbatch
4
12
9
31
Processing of HBCD to produce XPS Foam Using XPS
Masterbatch
5
18
13
44
Processing of HBCD to produce XPS Foam Using HBCD
Powder
4
12
9
31
Processing of HBCD to produce EPS Foam Using
Imported EPS Resin Beads
--
--
--
--
Processing of HBCD to produce SIPs and Automobile
Replacement Parts from XPS/EPS Foam
--
--
--
--
Installation of Automobile Replacement Parts
--
--
--
--
Installation of XPS/EPS Foam Insulation in Residential,
Public and Commercial Buildings, and Other Structures
--
--
--
--
Demolition of XPS/EPS Foam Insulation Products in
Residential, Public and Commercial Buildings, and
Other Structures
--
--
--
--
Recycling of EPS Foam
--
--
--
--
Formulation of Flux / Solder Paste
4
12
9
31
Use of Flux / Solder Paste
1010
2470
2517
6154
- As discussed in Section 2.4.1, there was no data to assess Occupational Non-User (ONU) exposures.
- Bold/shaded text indicates MOE is less than the benchmark MOE. Non-bold/non-shaded text indicates the MOE is
greater than the benclunark MOE. ~ Indicates that exposures are not expected during this exposure scenario.
Page 422 of 723
-------
4.2.2.4 Occupational Risk Estimation for Non-Cancer Effects Following Chronic
Dermal Exposures
Risks to workers were calculated for non-cancer effects following chronic dermal exposures, assuming
6.5% systemic absorption (see Section 3.2.2). Table 4-17 displays MOE values for all occupational
scenarios and human health hazards associated with chronic dermal exposure. As mentioned above,
adjusted MOEs were not calculated based on glove protection because EPA does not expect any level of
dermal exposure to HBCD following proper use of impervious gloves.
Table 4-17. Risk Estimate for Workers - Non-Cancer Effects Following Chronic Dermal Exposures
Occupational Scenario - Dermal Exposure
MARGIN OF EXPOSURE (MOE)
Benchmark MOE =
300
Benchmark MOE
IIHHI
Benchmark MOE
.ill
Benchmark MOE
inn
PODhed (mg/kg) =
1.68
Thyroid Effects
Decreased T4
(Ema et al. 2008)
PODhed (mg/kg) =
24
Liver Toxicity
Increased relative
liver weight and
vacuolization
(WIL Research
2001)
PODhed (mg/kg) =
0.689
Female
Reproductive
Toxicity
Reduced primordial
follicles
(Ema et al. 2008)
PODhed (mg/kg) =
9.03
Developmental
Toxicity
F2 offspring loss
(Ema et al. 2008)
Repackaging of Import Containers
1 (1.0)
2 (1.7)
14
25
0 (0.4)
(0.7)
5
9
Compounding of Polystyrene Resin to Produce
XPS Masterbatch
4
7
58
99
2
3
22
37
Processing of HBCD to produce XPS Foam
Using XPS Masterbatch
22
39
311
552
9
16
117
208
Processing of HBCD to produce XPS Foam
Using HBCD Powder
15
27
217
386
6
11
82
145
Processing of HBCD to produce EPS Foam
Using Imported EPS Resin Beads
--
--
--
--
--
--
--
--
Processing of HBCD to produce SIPs and
Automobile Replacement Parts from XPS/EPS
Foam
--
--
--
--
--
--
--
--
Installation of Automobile Replacement Parts
--
--
--
--
--
--
--
--
Installation of XPS/EPS Foam Insulation in
Residential, Public and Commercial Buildings,
and Other Structures
--
--
--
--
--
--
--
--
Demolition and Disposal of XPS/EPS Foam
Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures
--
--
--
--
--
--
--
--
Recycling of EPS Foam
--
--
--
--
--
--
--
--
Formulation of Flux / Solder Paste
1 (1.0)
2 (1.9)
14
27
0 (0.4)
1 (0.8)
5
10
Use of Flux / Solder Paste
274
540
3921
7718
113
222
1475
2904
- As discussed in Section 2.4.1, there was no adequate data available to quantitatively assess Occupational Non-User (ONU) exposures.
- Bold/shaded text indicates MOE is less than the benchmark MOE. Non-bold/non-shaded text indicates the MOE is greater than the
benchmark MOE.
- indicates that exposures are not expected during this exposure scenario.
- * indicates that risks are identified for women of reproductive age only. See text below for details.
Page 423 of 723
-------
4.2.2.5 Occupational Risk Estimation for the Recycling of Electronics Waste
Containing HIPS
HBCD from the recycling of electronics waste containing HIPS has been identified as an ongoing
exposure scenario and COU. Although HBCD is no longer used in electronics manufacturing, recycling
of old electronics waste containing HIPS with HBCD may result in acute and chronic exposures to
workers and ONUs. Occupational exposures for this OES are detailed in Section 2.4.1.14. As shown in
Table 4-18, risk estimates for this OES are well above the benchmark MOE and therefore risks are not
identified for either acute or chronic exposures from electronics waste recycling.
Table 4-18. Risk Estimates for Recycling of Electronics Waste Containing HIPS
Acute Exposures
Chronic Exposures
PODhed (mg/kg) = 9.03; Benchmark MOE = 100
Developmental Toxicity
F2 offspring loss
(Ema et al. 2008)
PODhed (mg/kg) = 1.68; Benchmark MOE = 300
Thyroid Effects
Decreased T4
(Ema et al. 2008)
High-End
Central Tendency
High-End
Central Tendency
722400
5197122
196224
2778904
Page 424 of 723
-------
4.2.3 Risk Estimation for General Population and Consumers
4,2,3,1 General Population Risk Estimation for Non-Cancer Effects - Background
Exposure
Risks were estimated for the general population, representing chronic, steady-state risks from sustained
background exposure in the environment due to HBCD persistence. In this assessment, general
population is considered to be individuals who are not expected to live close to point sources and are not
expected to have HBCD articles in their home. HBCD exposures to the general population are highly
variable and are influenced by both sources into the environment and degradation and removal from the
environment. Estimates of general population exposures based on environmental monitoring and
biomonitoring data represent the conditions present at the time the data was collected. It is unknown
which combination of potential sources associated with conditions of use as described in this risk
assessment contribute to the monitoring data presented here. However, given the wide range of
exposures shown within and across the monitoring data, there is a plausible contribution from some of
the sources/conditions of use described within this document. The totality of background exposure
includes steady-state environmental exposures ongoing releases not associated with a particular COU,
background/indirect exposures from minor use products (e.g., textiles, electrical and electronic products,
adhesives, and coatings) (Section 1.2.8), and releases stemming from historical activities (Section 1.2.9)
due to HBCD's persistence in the environment. To be health protective, general population risks for
background exposure were estimated based on the total aggregate exposure.
General population risk estimates account for steady-state background exposure in the environment
independent of any specific release. Therefore, only risks for chronic exposures are applicable. The
MOE tables below represent risks to aggregate steady-state HBCD exposure, combining dust, soil,
indoor air, diet, and dermal pathways. See Section 2.4.2 for a more detailed explanation of these
exposure pathways. Table 4-19 presents the MOEs for general population risks at both central tendency
(50th percentile) and high-end (95th percentile) exposure levels. General population risks from
background exposure are presented for the most sensitive and robust endpoints within each health
domain, as described in Section 4.2.1.1 and presented in Table 4-12.
Page 425 of 723
-------
Table 4-19. General Population Risk
stimation for Non-Cancer Effects - Background Exposure
Aggregate
Background
Exposure
Benchmark MOE
= 300
PODhed (mg/kg) =
1.68
Benchmark MOE
= 1000
PODhed (mg/kg)
= 24
Benchmark MOE
= 30
PODhed (mg/kg) =
0.689
Benchmark MOE
= 100
PODhed (mg/kg) =
9.03
Thyroid Effects
Decreased T4
(Ema et al. 2008)
Liver Toxicity
Increased relative liver
weight and vacuolization
(WIL Research 2001)
Female
Reproductive
Toxicity
Reduced primordial
follicles
(Ema et al. 2008)
Developmental
Toxicity
F2 offspring loss
(Ema et al. 2008)
AGE
GROUP
CT
HE
CT
HE
CT
HE
CT
HE
<1 year
42129
9959
601845
142270
17278
4084
226444
53529
l-<2 years
57455
15008
820789
214397
23563
6155
308822
80667
2-<3 years
92421
18315
1320304
261643
37904
7511
496764
98443
3-<6 years
124794
24118
1782765
344547
51180
9891
670765
129636
6-
-------
4.2.3.2 General Population Risk Estimation for Non-Cancer Effects - Subsistence
Fishers
Risks were also estimated for subsistence fishers based on aggregate exposure. Subsistence fishers
represent a PESS group for HBCD due to their greatly increased exposure via fish ingestion (142.4
g/day compared to a high-end of 22.2 g/day for the general population). Based on the increased
ingestion rate (U.S. EPA 2000a) and various measured HBCD concentrations in fish both downstream
(Near Field) and far away (Far Field) from a releasing facility, EPA estimated risks in a similar manner
as the general population (Section 4.2.3.1). See Section 2.4.2.5 for complete details on the exposure
assessment for subsistence fishers.
Table 4-20. General Population Risk Estimation for Non-Cancer Effects - Su
Aggregate
Exposure -
Subsistence
Fishers
Benchmark MOE
= 300
PODhed (mg/kg) =
1.68
Benchmark MOE
= 1000
PODhed (mg/kg)
= 24
Benchmark MOE
= 30
PODhed (mg/kg) =
0.689
Benchmark MOE
= 100
PODhed (mg/kg) =
9.03
Thyroid Effects
Decreased T4
(Ema et al. 2008)
Liver Toxicity
Increased relative liver
weight and vacuolization
(WIL Research 2001)
Female Reproductive
Toxicity
Reduced primordial
follicles
(Ema et al. 2008)
Developmental
Toxicity
F2 offspring loss
(Ema et al. 2008)
GROUP
CT
HE
CT
HE
CT
HE
CT
HE
Near Field
2252
2215
32168
31642
923
908
12103
11905
Far Field 1
38631
30060
551868
429435
15843
12328
207640
161575
Far Field 2
125980
65283
1799717
932610
51667
26774
677144
350895
jsistence Fishers
MOEs were several fold above the benchmark MOE for any health endpoint and therefore HBCD is not
expected to present risk to subsistence fishers living either nearby or distant from an HBCD point
source.
Page 427 of 723
-------
4,2.3,3 General Population Risk Estimation for Non-Cancer Effects - Highly
Exposed Populations
Risks were calculated for the highly exposed general population, a subset of Potentially Exposed or
Susceptible Subpopulations (PESS) living near a point source of HBCD release (e.g., for inhalation,
within 100 meters for high-end and within 1000 meters for central tendency). For simplicity, the tables
below present risks considering acute or chronic exposure via fish ingestion, inhalation, and additional
exposure pathways using the most sensitive POD for either acute or chronic exposure scenarios. MOEs
for all other hazards would be higher than the presented values. Exposure via fish ingestion is the
primary driver for any risks identified to the highly exposed general population, except for infants whom
are not anticipated to ingest fish in their diet. Infants would be uniquely exposed through breast milk,
with the received dose dependent on the body burden of the mother.
As discussed in Section 3.2.5.2.1, both reduced pup body weight and offspring loss were considered as
relevant hazard for evaluating risks following acute exposure. There is substantial uncertainty whether a
single exposure can produce a permanent adverse effect on postnatal mortality or body weight. EPA
determined that the sustained persistence of HBCD in human tissue suggests that a single exposure
could have sustained effects. EPA evaluated risks for offspring loss for all lifestages, including those
below reproductive age. While developmental effects would not be expected to present in younger
lifestages below reproductive age (i.e., they would be expected to affect the offspring of an exposed
individual), the bioaccumulation and persistence of HBCD in tissues suggests that initial exposure at an
earlier age could result in effects later in life. Additionally, it is unknown whether developmental effects
observed in gestationally exposed neonates could also present in older exposed children. Therefore,
despite the uncertainties, developmental outcomes were considered potentially applicable to acute
exposures at all lifestages, however developmental toxicity to teenagers and adults would be of highest
concern. Based on this health-protective approach, risk estimates are only provided for the most
sensitive endpoint of acute (offspring loss) and chronic (decreased T4 levels) exposure scenarios.
The MOE tables for fish ingestion and inhalation incorporate summed exposures from representative
fish ingestion or air inhalation modeled exposures and aggregate central tendency general population
biomonitoring-based exposures (representing background exposure). Background exposure estimates
were adjusted from the overall general population exposure values to remove the route of interest (e.g.,
fish ingestion or air inhalation) in order to avoid double-counting because exposure via a particular route
is likely geographically specific and risk estimates are only based on OES-specific exposures. Therefore,
exposures were only aggregated from different routes but not within routes. EPA evaluated exposures
for each exposure scenario assuming several differing release scenarios (see Table 2-54 and Table 2-55).
MOE tables in Section 4.2.3.3 present risks for two exposure sub-scenarios under each exposure
scenario, including both the scenario resulting in the highest exposure and a representative moderate
exposure level based on variability in estimated releases and wastewater treatment. The risk estimates in
the tables below are presented only for OES associated with ongoing manufacturing or import. Risks
were also estimated for Recycling of Electronics Waste Containing HIPS based on relative comparison
of release estimates and associated MOEs. These results are as presented in Section 4.2.3.3.1.
EPA is unable to model estimations of breast milk ingestion for <1 year old infants associated with an
exposure scenario, so exposures are based on monitoring data. Dietary risk estimation for highly
exposed infants was therefore based on high-end general population exposure values (applicable to
chronic exposures only). EPA additionally estimated risk for two scenarios from exposure to HBCD via
consumer articles. MOE tables for these scenarios incorporated the sum of cumulative dust and air
exposure and background general population exposure (with general population dust and air values
Page 428 of 723
-------
removed). Risk estimates are also provided for chronic exposure to HBCD via mouthing of plastic
articles containing HBCD.
EPA assessed risks to the highly exposed population following acute or chronic exposures
independently, however these do not necessarily represent independent populations. An individual living
near a facility would have both acute and chronic exposures to HBCD over time. Only short-term
residents or visitors would experience acute but not chronic exposures.
4.2.3.3.1 General Population Risk Estimation for Non-Cancer Effects Following
Acute Exposures - Highly Exposed Populations
Risks to the highly exposed population were calculated for non-cancer effects following acute exposures
based on fish ingestion and inhalation.
Risks via Fish Ingestion / Dietary Exposure
Risks were not estimated for the following exposure scenarios via dietary exposure because releases
were not identified, or associated exposures were not quantified:
OES #7, Installation of Automobile Replacement Parts
OES #11, Formulation of Flux / Solder Paste
A description of all subscenarios for OES resulting in fish ingestion exposure can be found in Table
2-54.
Highly Exposed Population
Infants
Infants <1 year old are not expected to ingest fish in their diet (U.S. EPA. ) (as discussed in
Section 2.4.2). Therefore, dietary risks to highly exposed infants were estimated based on high-end
general population exposure values, which incorporates breast milk in its dietary component as well as
high-end estimates of dust, dermal, air, and soil exposure. Infant risks are based on steady-state
exposures estimated via biomonitoring and are not associated with a particular exposure scenario.
Similar to the risk estimation for general population, the risk estimation for highly exposed infants is
therefore only relevant to chronic exposures. Therefore, risks were not estimated for highly exposed
infants following acute exposures.
Other lifestages
EPA estimated risks to the highly exposed general population following acute exposure via fish
ingestion. EPA selected high-end fish ingestion rates and 10th percentile stream flow rates for calculation
of ADR values in order to represent high-end acute exposures. ADR does not represent single-day
releases from a facility but instead high-end (as it is unlikely to be sustained every day) values for
ingestion of exposed fish. Fish concentrations were estimated based on 21-day average dissolved HBCD
in the water column and estimated BAF values. See Section 2.4.3.2 for a full description of the fish
ingestion exposure assessment.
Table 4-21 displays risk estimates for each condition of use and life stage following acute HBCD
exposure (as the sum of acute fish ingestion dose (ADR) and central tendency non-fish pathway dose)
based on the most sensitive relevant hazard endpoint of offspring loss. Scenario-specific discussions of
risk are below.
Page 429 of 723
-------
Table 4-21. Risk Estimation for Non-Cancer Effects Following Acute Exposure to Highly Exposed Population - Fish Ingestion
Developmental Toxicity - F2 Offspring Loss
PODhed (mg/kg) = 9.03; Benchmark MOE = 100
SCENARIO NAME
Age Group / Lifestage
1- <2
years
2- <3
years
3-<6
years
6 -<11
years
11-<16
years
16- <70
years
1.5 Repackaging of Import Containers (Moderate Exposure)
1678
2034
2223
2865
4749
2508
1.7 Repackaging of Import Containers (Highest Exposure)
336
407
445
573
949
500
2.11 Compounding of Polystyrene Resin to Produce XPS Masterbatch
(Moderate Exposure)
15033
18419
20261
26239
43649
23324
2.3 Compounding of Polystyrene Resin to Produce XPS Masterbatch (Highest
Exposure)
1763
2138
2337
3011
4992
2636
3.4 Manufacturing of XPS Foam using XPS Masterbatch (Moderate
Exposure)
7187
8751
9590
12383
20556
10907
3.3 Manufacturing of XPS Foam using XPS Masterbatch (Highest Exposure)
509
617
674
868
1439
759
4.2 Manufacturing of XPS Foam using HBCD Powder (Moderate Exposure)
14541
17810
19586
25360
42181
22530
4.1 Manufacturing of XPS Foam using HBCD Powder (Highest Exposure)
1308
1585
1732
2231
3699
1952
5.8 Manufacturing of EPS Foam using Imported EPS Resin beads (Moderate
Exposure)
139
168
184
237
392
207
5.7 Manufacturing of EPS Foam using Imported EPS Resin beads (Highest
Exposure)
14
17
18
24
39
21
6.4 Manufacturing of SIPs and Automobile Replacement Parts (Moderate
Exposure)
4234
5143
5629
7260
12043
6373
6.7 Manufacturing of SIPs and Automobile Replacement Parts (Highest
Exposure)
922
1117
1221
1573
2606
1375
8.1 Installation of Insulation in Buildings (Moderate Exposure)
16081
19721
21704
28119
46789
25026
8.3 Installation of Insulation in Buildings (Highest Exposure)
1687
2045
2235
2880
4775
2521
9.4 Demolition and Disposal of XPS/EPS Foam (Moderate Exposure)
2520
3057
3343
4309
7144
3775
9.3 Demolition and Disposal of XPS/EPS Foam (Highest Exposure)
254
307
336
432
717
378
10.3 Recycling of EPS Foam (Moderate Exposure)
7939
9672
10603
13695
22739
12073
10.7 Recycling of EPS Foam (Highest Exposure)
764
925
1011
1302
2158
1139
12.2 Use of Flux/Solder Paste (Moderate Exposure)
127338
171660
200726
273654
472637
290909
12.6 Use of Flux/Solder Paste (Highest Exposure)
80233
103797
118076
157180
266670
152982
- MOEs represent risk from aggregate exposure values from fish ingestion ADR and background general population (non-fish ingestion) exposure.
- Bold text indicates MOE is less than the benchmark MOE. Non-bold text indicates the MOE is greater than the benchmark MOE.
Page 430 of 723
-------
Estimated risks are above the benchmark MOE for the highly exposed general population for all
exposure scenarios except tor Manufacturing of EPS Foam from Imported EPS Resin Beads.
Manufacturing of EPS Foam from Imported EPS Resin Beads
The MOE is below the benchmark MOE for all lifestages from the highest exposure sub-scenario (5.7)
but not under the representative moderate exposure scenario (5.8). MOEs for sub-scenario 5.7 ranged
from 14 - 39, benchmark MOE = 100. Quantitative risk estimates are only provided for sub-scenarios
5.7 and 5.8 as representative exposure levels; however, EPA has determined that estimated risks are
below the benchmark MOE for at least the most sensitive lifestage (young toddlers) under 4 of the 12
evaluated sub-scenarios.
Risks via Inhalation
Risks were not assessed for the following exposure scenarios via dietary exposure because releases were
not identified or associated exposures were not quantified:
OES #7, Installation of Automobile Replacement Parts
A description of all subscenarios for OES resulting in outdoor air inhalation exposure can be found in
Table 2-55. Table 4-22 displays risk estimates for each occupational scenario and life stage following
acute HBCD exposure (as the sum of acute air inhalation dose (ADR) and central tendency non-air
pathways dose) based on the most sensitive hazard endpoint of offspring loss. Estimation of the risk is
above the benchmark MOE for the highly exposed population at all lifestages for all exposure scenarios
(including the highest exposure sub-scenarios) following acute exposures. Acute inhalation risks are
based on ADR exposures of average daily air concentrations at the fenceline of a facility, 100 meters
from the source. See Section 2.4.3.3 for a full description of the air inhalation exposure assessment.
Page 431 of 723
-------
Table 4-22. Risk Estimation for Non-Cancer Effects Following Acute Exposure to Highly Exposed Population - Inhalation
PODhed (mg/kg) = 9.03
Developmental Toxicity - F2 Offspring Loss; Benchmark MOE = 100
SCENARIO NAME
Age
Group / Lifestage
<1
years
1- <2
years
2- <3
years
3-<6
years
6 -<11
years
11-<16
years
16- <70
years
1.5 Repackaging of Import Containers (Moderate Exposure)
37630
40969
48241
64630
92743
129072
188277
1.3 Repackaging of Import Containers (Highest Exposure)
1307
1369
1551
2075
2950
3998
5844
2.5 Compounding of Polystyrene Resin to Produce XPS Masterbatch
(Moderate Exposure)
209835
280262
434299
587718
914715
1691903
2405171
2.3 Compounding of Polystyrene Resin to Produce XPS Masterbatch
(Highest Exposure)
128508
155035
206152
277322
410989
632620
915103
3.3 Manufacturing of XPS Foam using XPS Masterbatch (Moderate
Exposure)
20056
21431
24750
33138
47331
64982
94891
3.1 Manufacturing of XPS Foam using XPS Masterbatch (Highest
Exposure)
2743
2878
3264
4368
6212
8426
12316
4.7 Manufacturing of XPS Foam using HBCD Powder (Moderate Exposure)
39449
43033
50776
68031
97674
136136
198559
4.9 Manufacturing of XPS Foam using HBCD Powder (Highest Exposure)
2622
2751
3120
4175
5938
8053
11771
5.3 Manufacturing of EPS Foam using Imported EPS Resin beads
(Moderate Exposure)
4705
4948
5623
7525
10707
14541
21252
5.7 Manufacturing of EPS Foam using Imported EPS Resin beads
(Highest Exposure)
680
712
806
1078
1532
2075
3034
6.5 Manufacturing of SIPs and Automobile Replacement Parts
(Moderate Exposure)
154878
192899
267999
361101
542143
872628
1257281
6.3 Manufacturing of SIPs and Automobile Replacement Parts
(Highest Exposure)
14212
15094
17323
23189
33072
45215
66047
8.4 Installation of Insulation in Buildings (Moderate Exposure)
77282
87880
108591
145710
211652
305501
444324
8.2 Installation of Insulation in Buildings (Highest Exposure)
62609
70043
84954
113924
164690
234279
341143
9.1 Demolition and Disposal of XPS/EPS Foam (Moderate Exposure)
224448
305663
490103
664203
1046702
2044031
2889182
9.2 Demolition and Disposal of XPS/EPS Foam (Highest Exposure)
10310
10905
12465
16684
23771
32409
47352
10.7 Recycling of EPS Foam (Moderate Exposure)
140770
172342
233750
314673
469041
736216
1063132
10.3 Recycling of EPS Foam (Highest Exposure)
38255
41677
49110
65796
94433
131490
191797
11.1 Formulation of Flux/Solder (Moderate Exposure)
119229
142270
186480
250730
370065
561967
813854
11.3 Formulation of Flux/Solder (High Exposure)
39092
42627
50277
67361
96702
134743
196531
12.3 Use of Flux/Solder (Moderate Exposure)
222576
302353
482613
653925
1028773
1993909
2820626
12.1 Use of Flux/Solder (Highest Exposure)
221704
300817
479160
649188
1020530
1971119
2789416
- MOEs represent risk from aggregate exposure values from inhalation ADR and background general population (non-air) exposure.
- Bold text/red shading indicates MOE is less than the benchmark MOE. Non-bold/non-shaded text indicates the MOE is greater than the benchmark MOE.
Page 432 of 723
-------
Recycling of Electronics Waste Containing HIPS
EPA estimated central tendency and high-end air releases of HBCD from electronics recycling sites to
be 0.024 and 0.38 kg/site-d, respectively, for a duration of 250 days (Section 2.2.14). EPA compared the
air release estimates for this COU to those that were previously used to quantify HBCD air inhalation
exposure to the highly exposed general population from releases associated with current conditions of
use (Appendix F.1.2). The daily release amounts of HBCD are significantly less than would result in
risk based on assessed releases from current uses. For subscenario 5.7, which had the highest exposure
of any OES subscenario, daily release of HBCD was 14 kg/site/day using the higher value for emission
factor, resulting in an acute MOE of 680 (compared to a benchmark of 100, see Table 4-22). The high-
end air releases estimate from electronics recycling is only 0.38 kg/site/day, almost 40-fold less than that
of subscenario 5.7. Based on risk estimates above the benchmark for this and all other subscenarios from
acute exposure (and no instances of risk estimated for chronic air inhalation exposure), risks are not
expected for the highly exposed general population from recycling of electronics waste containing HIPS.
Water releases from this OES are not expected and therefore risks via fish ingestion are not relevant to
this exposure scenario.
Consumer Articles
Risks were also estimated for consumer articles. These use scenarios are specific to the highly exposed
general population and involve exposure to HBCD dust and indoor air. See Section 2.4.4 for more detail
on these exposure scenarios. Scenario CI corresponds to exposure scenario #8, Installation ofXPS/EPS
foam insulation in residential, public and commercial buildings, and other structures, and scenario A4
corresponds to exposure scenario #7, Installation of automobile replacement parts.
MOEs were calculated incorporating the summation of these exposures and background general
population non-dust, non-air exposures. Results are presented in Table 4-23.
Table 4-23. Risk Estimation for Non-Cancer Effects Following Acute Exposure to Highly Exposed
>opulations - Consumer Articles
PODhed (mg/kg) = 9.03
Developmental Toxicity - F2 Offspring Loss
Benchmark MOE = 100
SCENARIO NAME
Age Group / Lifestage
<1 year
1- <2
years
2- <3
years
3-<6
years
6 -<11
years
11-<16
years
16- <70
years
CI - XPS/EPS Insulation in
residences
35411
41456
49008
65906
103663
191193
285083
C2 - HBCD contained in
automobile components
11259
13163
15814
21297
35551
83611
128816
- MOEs represent risk from aggregate exposure values from combined dust and indoor air ADR along with background
general population (non-air/non-dust) exposure.
- Non bold/ non shaded text indicates the MOE is greater than the benchmark MOE.
Additionally, EPA estimated risks to the most sensitive lifestage of 0 to <1 year old infants based on
Mouthing of Plastic Articles Containing HBCD (see Table 2-109 for exposure values). For the highest
modeled acute exposure dose of 1.86E-02 mg/kg-day, when summed with central tendency aggregate
background exposure the total exposure is 1.86E-2 mg/kg-day, and MOEs are several fold above the
benchmark MOE (MOE = 485, benchmark MOE = 100).
Page 433 of 723
-------
4.2.3.3.2 General Population Risk Estimation for Non-Cancer Effects Following
Chronic Exposures - Highly Exposed Populations
Risks to the highly exposed population were calculated for non-cancer effects following chronic
exposures based on fish ingestion and inhalation. In addition to calculating risks for individual
lifestages, risks were calculated for an individual living near a facility across multiple lifestages. The
upper-end estimate of residential mobility of 33 years was selected for a high-end exposure duration
(U.S. EPA. 2 ). A central tendency value of 12 years was also selected (U.S. EPA. 201 lb), with risks
calculated both from birth through 12 years of age. Exposure is higher for younger lifestages, so
estimating risk for a resident starting from birth is protective of anyone for whom exposure began later.
The calculated MOEs based on integrated exposure across lifestages for these durations represent
estimations of the risk based on a weighted average of lifestage-specific exposures across the stated
period of time. As an example, for residency from birth to 12 years old, integrated fish ingestion
exposure is calculated as: (1/12 * [high-end aggregated general population infant exposure] + 1/12 * 1-2
year old exposure + 1/12 * 2-3 year old exposure + 3/12 * 3-6 year old exposure + 5/12 * 6-11 year old
exposure + 1/12 * 11-16 year old exposure). A similar weighted average was applied for 33-year
residency.
Risks via Fish Ingestion / Dietary Exposure
Risks were not estimated for the following exposure scenarios via dietary exposure because releases
were not identified or associated exposures were not quantified:
OES #7, Installation of Automobile Replacement Parts
OES #11, Formulation of Flux / Solder Paste
A description of all subscenarios for OES resulting in fish ingestion exposure can be found in Table
2-54.
Highly Exposed Population
Infants
Infants <1 year old are not expected to ingest fish in their diet (U.S. EPA. ) (as discussed in
Section 2.4.2). Therefore, dietary risks to highly exposed infants were estimated based on high-end
aggregate general population exposure values, which incorporates breast milk in its dietary component
as well as high-end estimates of dust, dermal, air, and soil exposure. Infant risks are based on steady-
state exposures estimated via biomonitoring and are not associated with a particular condition of use.
MOEs are several orders of magnitude above the benchmark MOE (MOE 9,959; Benchmark MOE =
300) based on 95th percentile aggregate exposures (Table 4-19).
EPA also modeled infant exposures up to and exceeding the 99.5th percentile and compared those with
available biomonitoring data (see Section 2.4.6.1). Estimation of the risk is above the benchmark MOE
even for the highest-end exposures (MOE = 468, benchmark MOE = 300), where the maximum
modeled HBCD dose is combined with the lower (90th) assumed percentile for the high-end of the
underlying distribution of environmental monitoring data. In this circumstance, the maximum estimated
dose is 3.59E-3 mg/kg-day (Table 2-110). This risk estimate should therefore be protective of the vast
majority of infants within the highly exposed general population.
Other lifestages
Table 4-24 provides risk estimates for each occupational scenario and life stage following acute HBCD
exposure (as the sum of chronic fish ingestion dose (ADD) and central tendency non-fish pathway dose)
based on the most sensitive hazard endpoint of thyroid effects. ADD values representing chronic
Page 434 of 723
-------
exposure utilized central tendency fish ingestion rates, which are expected to be more representative of
most populations over a sustained period. Fish concentrations were estimated based on 21-day average
dissolved HBCD in the water column and estimated BAF values, the same as for ADR estimates, except
ADD estimates used a 50th percentile stream flow rate which is expected to be more representative of
variance over a full year. Integrated exposure across lifestages incorporated the high-end (95th
percentile) aggregate exposure value for infants and high-end adult ADD. See Section 2.4.3.2 for a full
description of the fish ingestion exposure assessment. Scenario-specific discussions of risk are below.
Page 435 of 723
-------
Table 4-24. Risk Estimation for Non-Cancer Effects Following Chronic Exposure to Highly Exposed Population - Fish Ingestion
Thyroid Effects - Decreased T4 Levels
PODhed (mg/kg) = 1.68; Benchmark MOE = 300
SCENARIO NAME
Age Grou
p / Lifestage
1- <2
years
2- <3
years
3-<6
years
6 -<11
years
11-<18
years
16- <70
years
(CT
residency)
16- <70
years
(HE
residency)
Residency across
lifestages
Birth-12
Birth-33
1.5 Repackaging of Import Containers (Moderate Exposure)
13493
17342
20592
23615
42650
59420
23148
19574
22897
1.7 Repackaging of Import Containers (Highest Exposure)
3314
4070
4732
5210
9328
12945
4776
5109
5208
2.11 Compounding of Polystyrene Resin to Produce XPS
Masterbatch (Moderate Exposure)
42626
63202
81016
110413
207932
295491
160615
52906
93777
2.3 Compounding of Polystyrene Resin to Produce XPS
Masterbatch (Highest Exposure)
32594
45899
57123
72452
133941
188653
86982
42572
65066
3.4 Manufacturing of XPS Foam using XPS Masterbatch
(Moderate Exposure)
48741
74677
97669
140399
268363
384179
245238
58716
114345
3.3 Manufacturing of XPS Foam using XPS Masterbatch (Highest
Exposure)
15499
20103
23978
27744
50205
70010
27625
27982
26721
4.2 Manufacturing of XPS Foam using HBCD Powder (Moderate
Exposure)
52951
83034
110228
165280
319898
460850
346407
62521
130170
4.1 Manufacturing of XPS Foam using HBCD Powder (Highest
Exposure)
27971
38498
47322
58360
107161
150451
65798
37437
53551
5.8 Manufacturing of EPS Foam using Imported EPS Resin beads
(Moderate Exposure)
5376
6663
7778
8629
15476
21492
8008
8184
8577
5.7 Manufacturing of EPS Foam using Imported EPS Resin beads
(Highest Exposure)
587
712
824
898
1605
2225
811
920
904
6.4 Manufacturing of SIPs and Automobile Replacement Parts
(Moderate Exposure)
43862
65462
84244
115979
219015
311659
174116
54109
97727
6.7 Manufacturing of SIPs and Automobile Replacement Parts
(Highest Exposure)
23422
31533
38318
46085
84125
117784
49341
32129
43104
8.4 Installation of Insulation in Buildings (Moderate Exposure)
46588
70551
91604
129105
245391
350313
209658
56710
106800
8.3 Installation of Insulation in Buildings (Highest Exposure)
17074
22309
26704
31122
56408
78718
31393
24265
29811
9.4 Demolition and Disposal of XPS/EPS Foam (Moderate
Exposure)
22163
29660
35930
42931
78251
109483
45377
30613
40355
9.3 Demolition and Disposal of XPS/EPS Foam (Highest Exposure)
3388
4162
4840
5330
9545
13246
4889
5221
5327
10.3 Recycling of EPS Foam (Moderate Exposure)
34063
48323
60386
77322
143281
202036
94958
44152
68932
10.7 Recycling of EPS Foam (Highest Exposure)
20463
27165
32774
38826
70629
98730
40366
28533
36734
12.2 Use of Flux/Solder Paste (Moderate Exposure)
56195
89745
120590
187483
366992
531767
478246
65352
143436
12.6 Use of Flux/Solder Paste (Highest Exposure)
54800
86828
116056
177570
345834
499805
412935
64145
137608
- MOEs represent risk from aggregate exposure values from fish ingestion ADR and background general population (non-fish ingestion) exposure.
Page 436 of 723
-------
| - Non bold/ non shaded text indicates the MOE is greater than the benchmark MOE.
Page 437 of 723
-------
Risks via Inhalation
Risks were not assessed for the following exposure scenarios via dietary exposure because releases were
not identified or associated exposures were not quantified:
OES #7, Installation of Automobile Replacement Parts
A description of all subscenarios for OES resulting in outdoor air inhalation exposure can be found in
Table 2-55 and ADD ranges are provided in Table 2-104. Chronic inhalation risks are based on ADD
exposures of community average annual air concentrations, between 100 and 1000 meters from the
source. See Section 2.4.3.3 for a full description of the air inhalation exposure assessment. Estimated
chronic exposures (ADD + background) for all subscenarios and lifestages were below 1E-4 mg/kg,
which corresponds to an MOE of 16,800 for the most sensitive chronic endpoint of thyroid effects.
Therefore, the MOE is multiple orders of magnitude above the benchmark MOE for all lifestages of the
highly exposed population via inhalation following chronic exposures from any exposure scenario.
Consumer Articles
Risks were also calculated for consumer articles. These use scenarios are specific to the highly exposed
general population and involve exposure to HBCD dust and indoor air. See Section 2.4.4 for more detail
on these exposure scenarios. Scenario A3 corresponds to exposure scenarios #8, Installation ofXPSEPS
foam insulation in residential, public and commercial buildings, and other structures, and scenario A4
corresponds to exposure scenario #7, Installation of automobile replacement parts.
MOEs were calculated incorporating these exposures and background general population non-dust, non-
air exposures. Results are presented in Table 4-25.
Table 4-25. Risk Estimation for Non-Cancer Effects Following Chronic Exposure to Highly
Exposed Populations - Consumer Articles
PODhed (mg/kg) = 1.68
Thyroid Effects - Decreased T4 Levels
Benchmark MOE = 300
SCENARIO NAME
Age Group / Li
estage
<1 year
1- <2
years
2- <3
years
3-<6
years
6 -<11
years
11-<16
years
16-<70
years
CI - XPS/EPS Insulation in residences
22722
25427
38428
49422
76353
133286
187090
C2 - HBCD contained in automobile
components
52020
48691
56935
70657
103592
154935
209924
- MOEs represent risk from aggregate exposure values from combined dust and indoor air ADR along with background general
population (non-air/non-dust) exposure.
- Non bold/ non shaded text indicates the MOE is greater than the benchmark MOE (risk not identified).
Additionally, EPA estimated risks to the most sensitive lifestage of 0 to <1 year old infants based on
Mouthing of Plastic Articles Containing HBCD (see Table 2-109 for exposure values). For the highest
modeled acute exposure dose of 2.01E-03 mg/kg-day, when summed with central tendency aggregate
background exposure the total exposure is 2.05E-03 mg/kg-day, and the MOE is almost 3-fold above the
benchmark MOE (MOE = 819, benchmark MOE = 300).
4.2.3.4 Targeted Sensitivity Analysis
Section 2.2.15 describes the context behind conducting a targeted sensitivity analysis based on
production volume. Briefly, due to the uncertainty with the imported volume and resulting estimates of
Page 438 of 723
-------
environmental releases and exposures to the general population and the environment, a targeted
sensitivity analysis on the impact of import volumes on environmental risk estimates was conducted.
The exposure scenarios considered in the sensitivity analysis represent the exposure scenarios that
resulted in the highest estimates of releases on a daily basis and include scenarios that rely on both
industry data and OECD ESDs. Originally as presented above in Section 4.1.4.2, all nine exposure
scenarios with estimated water releases containing HBCD were predicted to have production volumes
up to 100,000 lbs/yr. The purpose of the sensitivity analysis is to evaluate how the model parameters of
production volume and percent of HBCD removed in in exposure scenarios with direct releases into
surface water may impact the predicted fish ingestion exposure values. In addition to the risk estimates
described throughout Section 4.2.3 based on a production volume of 100,000 lbs/yr, risk estimates were
also derived using the production volumes of 50,000 and 25,000 lbs/yr for the following three
processing exposure scenarios: exposure scenario #1: Repackaging of import containers, exposure
scenario #2: Manufacturing ofXPS Foam using XPS Master batch, and exposure scenario #3:
Manufacturing of EPS Foam from Imported EPS Resin Beads.
Estimation of the risk to highly exposed general population via fish ingestion was below the benchmark
MOE only for the higher sub-scenario of exposure scenario #5, Manufacturing of EPS Foam from
Imported EPS Resin Beads. The highest exposure sub-scenario for that exposure scenario, 5.7, assumed
direct discharge and 0% WWT removal. A sensitivity analysis based on estimated production volume
was performed only for that sub-scenario. Results are provided in Table Apx K-l.
Manufacturing of EPS Foam from Imported EPS Resin beads
Estimation of the risk is below the benchmark MOE for all lifestages only following acute exposure
from the highest exposure sub-scenario (5.7). Reduced PV has essentially no effect on acute exposures
and associated risk estimates.
4.3 Assumptions and Key Sources of Uncertainty for the Risk
Characterization
4.3.1 Assumptions and Key Sources of Uncertainties for the Environmental Risk
Characterization
In characterizing the environmental risk of HBCD, the same uncertainties mentioned above regarding
environmental hazard characterization also apply. Specifically, the uncertainty regarding the
diastereomer composition of HBCD will differ based on commercial and consumer products used, and
the changes of such proportions that may incur following environmental release.
For evaluating the potential trophic transfer of HBCD in the environment, many assumptions and
uncertainties were taken into consideration due to the complexity of food web dynamics. In general,
there is an inherent uncertainty when using proxy organisms to represent all terrestrial and aquatic prey
and predators; the selection was based on data availability, thus making it difficult to represent more
than three levels of prey-predator relationships. Organism selection for this evaluation was exclusively
from the available exposure factors in the U.S. EPA Wildlife Exposure Factors Handbook (also
incorporated in the U.S. EPA Final Water Quality Guidance for Great Lakes System). The
representative organisms used to evaluate trophic transfer in this Risk Evaluation also only represent a
small subset of prey-predator relationships and trophic levels within a food web and this evaluation does
not quantify how varying prey selection factors will ultimately affect the trophic transfer of HBCD.
Variations in diet categories due to life stage, gender, and seasonal differences are not addressed in this
evaluation because the specificity of each exposure factor differed based on the methodologies used in
Page 439 of 723
-------
their respective original references. Further, the inability to account for complete diets and the potential
variations in diet may have resulted in the under- or overestimation of HBCD uptake, metabolism and
elimination. Specifically, there is also an uncertainty regarding the impact of gut physiology on HBCD
uptake by prey and predator organisms used in this evaluation; as gut physiology and microbiology
becomes more complicated and diverse as the trophic level increases, there is an inherent likelihood that
the extent of HBCD uptake and depuration will be affected. Further underestimations of HBCD uptake
by terrestrial predators, as compared to aquatic predators in this assessment {i.e., calculated by
evaluating kestrel ingestion of mice) may also be due to the use of fruit and grasshopper HBCD
biomonitoring data as the original source of HBCD for kestrel, as opposed to smaller mammals with a
higher body fat composition. The limited data regarding HBCD in terrestrial organisms contributes to
the uncertainty regarding HBCD trophic transfer in terrestrial food webs. Underestimations of HBCD
uptake may have resulted from the inability to account for a majority of diet compositions for various
predators due to an overall lack of information on such species-specific preferences, and an inability to
account for varying sources of physiological differences amongst organisms. The evaluation of trophic
transfer may also overestimate uptake of HBCD from a specific prey type because HBCD metabolism
and elimination were not accounted for. Furthermore, the inability to quantify spatially- and temporally-
related trends regarding HBCD releases and exposure may explain why birds of prey have varying body
burdens of HBCD in urban and remote regions (Law et al. 2006; de Boer et al. 2004). Finally, exposure
to terrestrial organisms may be underestimated because exposure via the inhalation of suspended HBCD
particulate in the ambient air (from various release sources) was not characterized or aggregated with
exposure to soil or ingestion {i.e., diet). Only one available repeat-dose toxicity study was evaluated
where no adverse effects were observed up to 2000 mg/m3 administered 6h/day for 14 days, and the
reported LC50 for 4-h inhalation exposure in rats is greater than 5000 mg/m3 (Song et al. 2016). As seen
in Appendix F.3, the highest modeled air releases of HBCD are from exposure scenarios for
import/repackaging (2.18 x 10"2 mg/m3) and manufacturing of EPS foam from imported EPS resin beads
(2.28 x 10"2 mg/m3) for fenceline communities (100 m from the source), which are five magnitudes less
than the hazard thresholds observed for rats due to inhalation. Therefore, inhalation of HBCD may not
be the main driver for HBCD exposure for terrestrial organisms, even those that inhabit areas near
industrial facilities.
EPA assessed releases of HBCD to the environment or to disposal based on the production volume of
HBCD, emission factors, and number days of release per year. In a few cases, EPA used TRI release
data in lieu of the production volume of HBCD and emission factors. The emission factors were
obtained from the EURAR, OECD ESDs, an EPA GSs, or a scientific journal article and the number of
days of release per year were obtained from the EURAR, EU TGD, the NICNAS RAR, an OECD ESD,
or an EPA GS as discussed in detail in Section 2.2. These data do not specifically pertain to the sites that
are the subject of this Risk Evaluation. Therefore, in the case of each COU, EPA estimated a range of
emission factors and a range of number of days of release per year and calculated a range of daily
release rate from these estimated ranges to account for uncertainty about the values of the emission
factor and number of days of release. Also, in the case of some releases, there is uncertainty about
medium of release and therefore EPA assessed various media of release to account for this uncertainty.
The emission factors and numbers of days of release per year that are the basis of the assessment pertain
to HBCD processing or use that occur at sites that are not located in the U.S. or pertain to an industrial
or commercial sector that is related to a COU {e.g., polymer processing, use of spray polyurethane
foam). There is some uncertainty regarding the extent to which this data is applicable to processing or
use of HBCD in the U.S. To account for the uncertainties and variability among release estimates and
exposure considerations including wastewater treatment, EPA provided risk estimates based on a range
of exposure sub-scenarios. EPA believes this sufficiently captures the range of risk estimates for all
reasonably expected environmental exposures. In regard to the calculation of risk estimates using
Page 440 of 723
-------
predicted surface water or sediment concentrations of HBCD based on E-FAST or the PSC, all risk
estimates can be associated with a specific condition of use.
Water dilution models can be used to determine the concentration of a chemical in the surface water
after a source emits the chemical into a water body. Since the E-FAST model incorporates defaults that
encompass either a combination of upper percentile and mean exposure parametric values, or all upper
percentile parametric values, the resulting model predictions represent high-end exposures estimates.
Simple dilution models, such as EFAST provide exposure estimates that are derived from a simple mass
balance approach, and does not account for partitioning between compartments within a surface water
body or degradation over time in different media, parameters which are relevant to HBCD, therefore
EPA utilized a two-tier approach by complementing the EFAST modeling with more refined estimate
from the PSC model to further describe environmental exposures. However, these predicted surface
water and sediment concentrations will likely underestimate HBCD concentrations because they do not
take into consideration background HBCD concentrations (only what may be in these matrices due to
water releases containing HBCD from a specific condition of use).
Monitoring data on measured water, sediment, and soil concentrations of HBCD take into consideration
real time HBCD concentrations in these matrices, however they cannot be associated with a specific
release associated with historical or condition of use. Some monitoring studies will associate
measurements to a specific sector; however this categorization is still too broad for one to associate with
a historical or condition of use. Furthermore, although risk estimates can be condition of use- or sector-
specific, the sole use of surface water, sediment, and soil concentrations of HBCD will not account for
dietary-associated sources of HBCD and will underestimate the risk to both terrestrial and aquatic
organisms. Aggregation of HBCD exposure pathways was not conducted within an exposure pathway
(e.g., the summation of sediment HBCD exposure to benthic organisms from via both background
monitoring data and potential modeled current releases from an ongoing condition of use) or across
exposure pathways (e.g., diet, dermal, inhalation) to avoid double-counting because exposure via a
particular route is likely geographically specific and risk estimates are only based on specific exposures.
Measured background exposure concentrations are therefore potentially associated with releases from
both historical and current conditions of use, and may be used to semi-quantitatively evaluate exposure
should there be predicted releases. As discussed in Section 4.1.3, measured monitoring information
(background exposure) was used to characterize the risk to aquatic and terrestrial organisms due to
potential surface water and air releases of HBCD from the land disposal of other formulated products
and articles (e.g., adhesives, coatings, textiles, and electronics) via potential leaching and runoff of
HBCD, in lieu of having measured data. As stated previously, measured monitoring information can
encompass releases from all historical and current conditions of use, therefore the use of monitoring
information (background exposure) may overestimate contributions from this one exposure scenario.
For exposure scenarios where water and/or air releases were not predicted to occur, EPA does not expect
that commercial or consumer uses of products or articles containing HBCD will lead to releases to the
environment, however EPA cannot rule out this possibility. Any potential environmental exposure
resulting from ambient air releases of commercial/consumer products is expected to be captured as part
of the background assessment of environmental soil monitoring data near general population (non-point
source exposure) sites (Table 4-6).
Based on the HBCD releases resulting from exposure scenario 5.8 (EPS foam from imported EPS Resin;
input parameters further detailed in Section 2.2.6), using the 10th percentile predictions, an additional
sensitivity analysis was conducted to evaluate how a greater range of HBCD aerobic benthic half-lives
(i.e., 6-, 8-, 32-, 100-, 384-d) would affect PSC-predicted surface water and sediment HBCD
concentrations, in comparison to the analysis conducted in the Risk Evaluation using both the 11- and
Page 441 of 723
-------
128-d HBCD aerobic benthic half-lives. 11- and 128-d HBCD half-lives represent a selected range of
HBCD aerobic benthic half-lives and are based on the high data evaluation scores. Predicted surface
water HBCD concentrations are based on the HBCD half-life of 128-days, and predicted sediment
HBCD concentrations are based on both 11- and 128-d HBCD half-lives.
In regard to the 21-d average surface water and 28-d average sediment HBCD concentrations, there is an
average difference of 3% and 19.5%, respectively. The average difference between surface water and
sediment concentration ranges are 0.85-6.4, and 4.7-41.1%, respectively. In addition, the greatest
difference between predicted 21-d average surface water and 28-d average sediment HBCD
concentrations was when comparing the 11-, 32-, and 100-d HBCD aerobic benthic half-lives (average
difference in surface water and sediment HBCD concentrations of approximate 6%, and 18-41%,
respectively). When comparing the lowest and highest predicted HBCD surface water and sediment
concentrations using either the 6- or 384-d HBCD half-lives, there is a difference of 17.7 and 76.1% in
HBCD surface water concentration (19.9-24.2 |ig HBCD/L) and HBCD sediment concentration
(15,600-65,200 |ig/kg), respectively. Based on the average difference between the surface water and
sediment concentrations of HBCD due to the various half-lives presented in the sensitivity analysis,
selecting a different half-life only significantly impacts the sediment concentration (i.e., a longer half-
life results in higher HBCD sediment concentrations). The sensitivity analysis suggests that there is a
significant amount of uncertainty in regards to the exposure scenario-specific environmental risk based
on PSC-predicted sediment concentrations; given that HBCD is likely to partition to sediment, it is
likely that the current values underestimate sediment HBCD concentrations and resulting risk to benthic
organisms, especially since these calculations do not take into consideration background levels of
HBCD that pre-exist potential exposure scenario-specific releases. Finally, model-predicted media
concentrations do not take into consideration previously-released HBCD via historical or current
conditions of uses, and underestimate the overall HBCD exposure to aquatic or terrestrial organisms.
The degradation of plastic products in the environment has also resulted in concern regarding the uptake
of HBCD via exposure to microplastics. As discussed in Section 2.1.3, this evaluation does not quantify
exposure to microplastics, nor is it able to quantify potential of exposure to HBCD from microplastics
due to various factors impacting microplastic fate and transport, as well as those impacting the possible
desorption of HBCD from microplastics if ingested (i.e., physiological limitations to prey size, gut
physiology, microplastic physical-chemical properties). Microplastics, similar to other environmental
sorbents (i.e., natural organic matter, suspended solids) are able to act as both a source and sink of
hydrophobic contaminants, and introduce uncertainty to the Risk Evaluation of HBCD exposure and risk
to aquatic and terrestrial organisms due to the complexity involved in the characterization of both
microplastic and microplastic-associated contaminant bioavailability. The approach used here to
estimate release and exposure does not account for the fact that HBCD may be released in a polystyrene
matrix in the modeled exposure scenarios. The inability to quantify potential leaching of HBCD from
products containing HBCD may overestimate or underestimate HBCD environmental exposure in
various media.
Another uncertainty regarding the exposure and environmental risk of HBCD is the likelihood of sex-
specific transfer of HBCD to offspring. HBCD has been measured in peregrine falcon and chicken eggs
upwards of 15,000 and 5,800 ng HBCD/g lw (Tao et al. 2.016; Guerra et al. 2012). In addition, HBCD
has also been quantified in milk from both humans and dairy cows (10 and 5.3 ng HBCD/g lw,
respectively (Shi et al. 2017b; Glynn et al. 2011). The presence of HBCD in the eggs of both aquatic and
terrestrial birds, as well as the milk of terrestrial mammals, suggests that sex-specific transfer is an
elimination pathway of HBCD for female birds and mammals that are reproductively active and
resulting offspring are exposed to HBCD before and after birth. The Risk Evaluation does not take this
Page 442 of 723
-------
uncertainty into account, and it is likely that the current environmental Risk Evaluation underestimates
organism exposure to HBCD.
EPA assessed risks for many of the current uses of HBCD using an assumed annual production volume
of 100,000 lbs per site (Section 2.2.1). EPA considers this value to be an upper bound estimate for an
importer based on the 2016 CDR reporting estimates and small entity reporting requirements.
Considerably higher production volume (ranging as high as 10 to 50 million pounds) occurred in
previous years. Many of the previously manufactured products associated with this past production may
still be in use and therefore be contributing to current and future levels of release. There is insufficient
information available for EPA to quantify any additional level of current or future releases from the
COUs based on this past production. As previously stated, the 100,000 lbs/year-site value that EPA used
for the primary assessment may represent a conservative approach for current production, however, the
possibility of higher releases based on remnants of past and historical activities cannot be ruled out. As
stated above in Section 4.1, EPA performed a sensitivity analysis for three COUs (Repackaging of
Import Containers, Manufacturing of XPS foam from XPS masterbatch, and Manufacturing of EPS
foam from EPS resin) using the per site volumes of 50,000 lbs/yr and 25,000 lbs/yr to examine the effect
of process volume on modelled environmental exposures. Due to HBCD declining use, EPA did not
identify a current import volume for HBCD, and conservatively used the CDR reporting threshold for
small firms of 100,000 lbs/yr as explained in Section 1.2.3.
As previously discussed, historical activities are responsible for a subset of the total aggregate exposure
to the environment and general population. The specific percentage of these total exposures that stem
from historical activities cannot be determined and may differ both geographically and temporally.
4.3.1.1 Confidence in Risk Estimates
There are many sources of uncertainty confidence in the parameters used to estimate surface water and
sediment HBCD concentrations for each exposure scenario. As presented in Table 2-113, the uncertainty
and variability are summarized for each consideration regarding environmental releases, fate, and
exposure model parameters, including in some cases the multiple sources of information used for some
of these considerations {i.e., emission factors, days of release, physical-chemical properties). To account
for these sources of uncertainty and variability, EPA provided risk estimates that reflect a range of
considerations, resulting in multiple iterations of RQs (exposure sub-scenarios) for each exposure
scenario.
EPA believes that these sub-scenarios sufficiently capture the range of risk estimates for all reasonably
expected aquatic and terrestrial exposures, with minimal remaining unaccounted-for uncertainty. The
environmental monitoring and biomonitoring studies used to derive risk estimates are of high quality
and are used to evaluate background concentrations of HBCD and the potential for HBCD trophic
transfer in aquatic (osprey, mink and other fish-consuming predators) and terrestrial organisms.
Although the organisms covered in these analyses are limited and do not take into consideration
variabilities in diet preferences, similar to the selected hazard effect concentrations used to derive risk,
representative organisms were used; whether exposure is assessed using measured or predicted media or
tissue concentrations, there is generally greater risk calculated for aquatic organisms. Therefore, EPA
has high confidence in the range of risk estimates for both aquatic and terrestrial organisms.
Page 443 of 723
-------
4.3.2 Assumptions and Key Sources of Uncertainties for the Human Health Risk
Characterization
4.3.2,1 Physical-Chemical Properties and Toxicokinetics Considerations
HBCD toxicokinetics including absorption and bioaccumulation differ greatly among the three HBCD
isomers (a-, P-, and y-HBCD) and are greatly affected by the relative fat content of tissues and
surrounding media (e.g., water, air, diet, breastmilk). Reasonably available information on human health
hazard and exposure does not typically differentiate among the three isomers of HBCD, and it is
unknown whether a particular COU or exposure pathway may bias toward one isomer over another. In
the absence of reasonably available information, this Risk Evaluation only assessed HBCD as a variable
mixture and it cannot be determined whether how the risk estimates would compare to a more refined
isomer-specific assessment.
EPA estimated dermal risks assuming consistent 6.5% dermal absorption based on the highest-end
estimate from available ex vivo and in vitro data in order to be health-protective. The actual percentage
of HBCD absorbed dermally is variable based on multiple factors including the relative percentage of
each isomer in the mixture and the relative ratio of sweat to sebum on skin. Absorption in occupational
settings may be substantially lower than this value based on frequent hand washing or uneven
distribution across skin. The true percentage of any dermally delivered dose that would be systemically
absorbed is likely to vary between COUs and over time. A calculation of flux would account for the
effect of exposure duration on absorbed dose. However a quantitative comparison demonstrates that
fraction absorbed and permeability/flux methods result in approximately the same value when using
upper-bound estimates (Appendix L). For many COUs HBCD is expected to be entrenched within
granules or pellets for which absorption is not expected. This will significantly reduce the amount of
HBCD absorbed from within these materials. However, for most COUs the MOEs were more than an
order of magnitude below the benchmark MOE, so moderate refinements in dermal absorption are
unlikely to result in a different risk conclusion.
EPA did not evaluate potential risks to metabolites or degradants of HBCD. In vivo metabolism of
HBCD varies by stereoisomer (Section 3.2.2.1.3) and the expected distribution of resulting products
cannot be sufficiently quantified. Any toxicity from HBCD metabolites would likely be accounted for in
long-term animal studies on the parent compound. Environmental or industrial degradants (e.g., from
thermal cutting) are expected to be similarly diverse and there is insufficient information available for
accurately determining relative concentrations of any particular species given differing assumptions
about media of release, wastewater treatment, and fate. Uncertainty is compounded when considering
the limited availability of toxicological data on these potential degradants. It is unknown how much
additional risk can be attributed to these species.
Thermal cutting of XPS and EPS foam with a hot wire can result in the release of HBCD nanoparticles
(Section 2.4.1.1). In addition to potentially increased absorption, nanoparticles may have unique
toxicities independent of HBCD biochemistry. EPA cannot determine sufficient details on the nature of
these nanoparticles or what additional toxicity they may present. Additionally, EPA did not incorporate
HBCD nanoparticle air concentration data into the estimates of exposure concentrations of the relevant
exposure scenarios because these data are measurements of concentration in a laboratory glovebox and
are not worker monitoring data. The absence of quantitative risk estimates for nanoparticle toxicity
represents a potential underestimation of risk from exposure to HBCD from these uses.
Although some simplistic toxicokinetic models for HBCD exist (empirical two-compartment open
kinetic model; and a simple first-order elimination model to estimate the steady-state lipid
Page 444 of 723
-------
concentration); these models introduce significant uncertainties that reduce the value of their use.
Therefore, EPA was unable to model the potential effects of bioaccumulation in human tissues over
time. For both consistency and health-protectiveness, these issues were accounted for by utilizing the
upper range of absorption estimates across available studies and including a 10X subchronic-to-chronic
UF based on assumed increasing bioaccumulation over time. This adjustment was not included for
developmental endpoints or for effects observed following multi-generational exposure, which should
already encompass chronic bioaccumulation. EPA also conservatively evaluated risks to all receptors
from hazards only observed in the F2 population (i.e., only after 2 generations of bioaccumulation). EPA
believes that the use of this 10X uncertainty factor is likely to be protective of risk from
bioaccumulation in human tissues, however there is insufficient available data to confirm this
presumption.
4.3.2.2 Human Health Hazard Considerations
To derive the benchmark MOEs, the UF approach ( ; ) was applied to a
PODhed based on changes in thyroid hormone levels (T4) in male rats exposed to HBCD. UFs were
applied to the PODhed to account for extrapolating from an animal bioassay to human exposure, the
likely existence of a diverse population of varying susceptibilities, and subchronic to chronic duration
(chronic exposures only). For the most part, these extrapolations are carried out with default approaches
given the lack of data to inform individual steps. EPA presumes that in general these uncertainty factors
are health-protective and are unlikely to underestimate risk relative to more data-driven refinement of
uncertainty factors.
As discussed in Section 3.2.5.2.1, both reduced pup body weight and offspring loss were considered as
relevant developmental endpoints for evaluating risks following acute exposure. There is substantial
uncertainty whether a single exposure can produce a permanent adverse effect on postnatal mortality or
body weight. EPA determined that the sustained persistence of HBCD in human tissue suggests that a
single exposure could have sustained effects. Additionally, acute and short-term exposure has been
associated with thyroid hormone disruption, which would be expected to have downstream effects on
development. Therefore, despite the uncertainties, neonatal mortality and body weight reduction were
considered relevant to acute exposures. EPA also considered maternal decreases in T4 levels for acute
exposure scenarios, because short-term changes in thyroid hormones are likely upstream of those
developmental outcomes. Additionally, decreased maternal T4 can serve as a sensitive quantitative
measure of other potential developmental effects that cannot otherwise be quantified (such as
neurotoxicity). EPA evaluated general population risks for the most sensitive endpoint of offspring loss
for all lifestages, including those below reproductive age. While developmental effects would not be
expected to present in younger lifestages, the bioaccumulation and persistence of HBCD in tissues
suggests that initial exposure at an earlier age could result in effects later in life. Additionally, it is
unknown whether developmental effects on neonates could also present in young exposed children. This
is a health protective approach that will overestimate risks to the general population following acute
exposures, especially for those lifestages below reproductive age. There is substantially less uncertainty
for risk estimations of teenagers and adults.
For risks following chronic exposure, there is medium confidence in the risk estimates for most sensitive
endpoint of thyroid effects for all populations and lifestages. There is uncertainty over the use of rodent
thyroid hormone data for quantitative human health risk assessment, as the complexity of the system
makes it difficult to determine whether adult rodents would in fact be more sensitive to the specific
effects of HBCD. However, developmental effects of thyroid disruptors following gestational exposure
are expected to be highly comparable between rats and humans, with substantially increased
susceptibility in developing individuals of both species compared to adults. Direct extrapolation of
rodent thyroid hormone effects to humans is health-protective and may potentially overestimate risk to
Page 445 of 723
-------
human adults, but evidence supports its use as a sensitive quantitative endpoint upstream of various
detrimental developmental outcomes, including those which could not be quantified (e.g.,
developmental neurotoxicity).
4.3.2.3 Occupational Exposure Considerations and Confidence Statements
There is high confidence in the most sensitive human health endpoints for chronic and acute exposures,
and all endpoints are relevant to workers whom are all likely to be of reproductive age. Occupational
inhalation exposure estimates (see Section 2.4.1.15.4) were assigned Low-Medium to Medium
confidence (Table 2-71) based on inhalation monitoring data for all OES. Confidence is raised by the
evaluation of risk estimates using both central tendency and high-end exposure levels. Therefore,
estimated risks for occupational exposures are overall of medium confidence for OES with low-medium
exposure confidence and of medium-high confidence for all OES with medium confidence.
In the absence of data, the dermal exposures to workers for relevant COUs were estimated using a
dermal exposure model routinely used in the new chemicals program, "EPA/OPPT Direct 2-Hand
Dermal Contact With Solids." The dermal exposure levels were estimated using conservative
assumptions, however both high-end and central tendency dermal exposure was estimated. When
considering the variability in expected dermal absorption (see above), it is likely that dermal risk
estimates are overestimated for the majority of occupational scenarios. Given the various uncertainties,
the potential magnitude of overestimation cannot be determined. There is low-medium confidence in
occupational dermal risk estimates.
For the purposes of this evaluation, inhalation and dermal routes of exposure were not combined to
evaluate occupational risks to HBCD. Dermal and inhalation exposure were considered independently.
Combining exposure routes would entail too much uncertainty as to the actual internal dose at target
sites given the lack of a usable PBPK model and/or measured biomonitored doses. See Section 4.4.2 for
more discussion.
EPA expects potential inhalation exposure of occupational non-users (ONUs) to HBCD, but EPA did
not quantify these exposures due to lack of adequate worker monitoring data and lack of relevant
mathematical models as discussed in Section 2.4.1.1. EPA assumes HBCD air concentrations that ONUs
are potentially exposed to are lower than HBCD air concentrations that workers are potentially exposed
and also assumes the duration and frequency of the ONUs' potential HBCD inhalation exposures to be
lower than that of workers as discussed in Section 2.4.1.1. When risks are not identified for workers,
risks are unlikely for ONUs. However, during the construction (i.e., installation of XPS/EPS insulation)
and demolition of buildings, there is uncertainty about whether the HBCD potential exposure level of
ONUs in the case of construction and demolition workers is in fact lower than those of workers. EPA
believes that ONUs may work in close proximity to workers for these OES and hence may be exposed to
HBCD air concentrations similarly to workers. Furthermore, the duration and frequency of the ONUs'
work during the construction and demolition of buildings may equal that of the workers at least for
limited periods of time. Therefore, risks may be comparable between ONUs and workers for these OES.
EPA is unable to quantify risk estimates for occupational exposure associated with the COU Land
Disposal of Formulated Products and Articles. While exposures to HBCD from disposal of HBCD-
containing articles are expected to be less than that for other OES, in some circumstances municipal
solid waste may undergo shredding which can result in significant exposure to dust. Elevated exposures
under this scenario are unlikely for a sustained basis, however acute exposure is possible. EPA only
considers developmental endpoints as relevant to acute exposure, so presumably acute exposure would
only be of concern to a female who is pregnant concurrent with exposure. Therefore, risks from Land
Disposal of Formulated Products and Articles are unlikely but cannot be ruled out.
Page 446 of 723
-------
4.3.2.4 PPE Considerations
Non-cancer risk estimates (MOEs) for occupational exposure scenarios are presented in Section 4.2.2.
These tables also present the minimum respirator requirement needed to mitigate risk for all health
domains. The MOEs for these respirator scenarios assume workers are properly trained and fitted on
respirator use, and that they wear respirators for the entire duration of the work activity. The MOEs for
respirator scenarios following chronic exposure also assume that workers wear respirators for the entire
duration of the work activity throughout their career. Similar assumptions apply to the use of gloves and
their expected elimination of any dermal exposure.
EPA has considered these assumptions for each condition of use (Table 4-13). The majority of COUs and
exposure scenarios are likely to take place in an industrial setting with an effective and robust respiratory
protection program. However, for installation and demolition of EPS/XPS insulation products (OES #8
and #9), Based on expert judgment and evaluation of peer review comments, EPA believes that workers
in these scenarios are unlikely to wear respirators. Therefore, MOEs assuming respiratory PPE are
presented for these OES only as a what-if scenario, but risk estimates without respirators will be used for
risk determination. EPA believes that this approach reflects a reasonable application of PPE
considerations for each COU.
4.3.2.5 General Population/Consumer Exposure Considerations and Confidence
Statements
EPA evaluated risk to the general population for individual lifestages for both acute and chronic
exposure scenarios. For chronic exposure, EPA also evaluated risk for an individual living near a facility
throughout their lifetime using integrated exposure values across lifestages, representing a weighted
average across a lifetime.
Estimated risks to the highly exposed populations are driven by fish ingestion exposure. Therefore, these
estimated risks are highly dependent on the selected BAF value. EPA chose a BAF value at the low-end
of the reported range. This was done because the modeled dissolved surface water estimates are
generally larger than values reported in the literature. Pairing a higher BAF value with higher surface
water values could result in unreasonably high estimated fish-tissue concentrations. EPA compared the
range of reported fish-tissue concentrations from monitoring data and found the modeled fish tissue
concentrations (range of modeled dissolved surface water and low-end BAF) to be of a similar order of
magnitude. Therefore, while selection of a different BAF value would have a significant effect on fish
ingestion risk estimates, the values for BAF and resulting fish ingestion exposure are well-supported by
the data.
For estimating fish ingestion exposures to the highly exposed general population, EPA selected high-end
fish ingestion rates for calculation of ADR values in order to represent high-end acute exposures. ADD
values representing chronic exposure utilized central tendency fish ingestion rates, which are expected to
be more representative of the most populations over a sustained period. While these assumptions are
expected to protect the majority of populations, there is potential for higher risk among subpopulations
with consistently elevated fish consumption rates. Risk estimates for chronic exposure scenarios may
therefore underestimate risk to these subpopulations, however it is uncertain whether any of these
subpopulations with significantly elevated fish ingestion rates actually live nearby a HBCD facility. In
order to account for subpopulations with consistently elevated fish ingestion rates, EPA also evaluated
risks to subsistence fishers (Section 4.2.3.2), however reasonably available and reliable data on ingestion
rates were only available for adults (from ( 100a). The inability to confidently assess younger
lifestages underestimates risk to this PESS group.
Page 447 of 723
-------
Estimated days of release for a given OES are assumed to be evenly distributed throughout the year.
Additionally, days of release for certain sub-scenarios may be as low as a single day per year.
Toxicological data are not available comparing intermittent and continuous exposures for relative
chronic health outcomes, but the effects of these uncertainties are minimized due to the sustained
environmental persistence and elevated bioaccumulation of HBCD in tissues. Both acute and chronic
exposures via fish ingestion to the highly exposed general population are based on 21-day average
dissolved HBCD water concentration and a single BAF value. It is assumed that the average HBCD
concentration in fish to be consumed remains relatively constant and the more important variable is the
ingestion rate, however EPA also assumed 50th percentile flow rate for risk estimates based on chronic
exposures. Use of highest single-day water concentrations for acute exposure would provide a more
health-protective estimate, however this would introduce large uncertainties and incongruency between
the use of chronic BAF values and acute release/exposure scenarios. Similarly, risk estimates resulting
from chronic exposures based on 50th percentile flow rate may underestimate risks for certain water
bodies with consistently low flow rates, however there would be significant uncertainty whether a 10th
percentile flow rate could be valid over an entire yearly average.
EPA does not expect that commercial or consumer uses of products or articles containing HBCD will
lead to releases to the environment, however EPA cannot rule out this possibility. Any potential general
population exposure resulting from ambient air releases of commercial/consumer products is expected to
be captured as part of the aggregate background assessment (Section 4.2.3.1).
There are many potential sources of uncertainty in all of the parameters involved in general population
exposure estimates. As presented in Table 2-114, the greatest influence on highly-exposed exposure
estimates given the associated uncertainty and sensitivity (effect on the final values) stems from the
selection of emission factor and days of release. Production volume is highly uncertain but not very
sensitive, while other factors such as physical-chemical properties, BAF, HBCD half-lives, and exposure
model parameters were all estimated to contain low uncertainty. In order to account for these
uncertainties and variability among release estimates and exposure considerations including wastewater
treatment, EPA provided risk estimates based on a range of exposure sub-scenarios. EPA believes that
these sub-scenarios sufficiently capture the range of risk estimates for all reasonably expected general
population exposures, with minimal remaining unaccounted-for uncertainty. Consumer article modeling
defaults are believed to be highly uncertain and highly sensitive, however estimation of the risk for
consumer articles were orders of magnitude above the benchmark MOE. Therefore, EPA has high
confidence in the range of risk estimates for the highly exposed general population.
Overall, based on the considerations above there is medium confidence in fish ingestion risk estimates.
There is high confidence in risk estimates for inhalation exposure and low-medium confidence for
consumer articles. Confidence in risk estimates from acute exposure is lower for non-infant lifestages
below reproductive age because risk estimates from acute exposure are based on developmental
endpoints that are less likely to affect older children.
4,3,2,6 Considerations of Historical Production Volumes and Activities
EPA assessed risks for many of the current uses of HBCD using an assumed annual production volume
of 100,000 lbs per site (Section 2.2.1). EPA considers this value to be an upper bound estimate for an
importer based on the 2016 CDR reporting estimates and small entity reporting requirements.
Considerably higher production volume (ranging as high as 10 to 50 million pounds) occurred in
previous years. Many of the previously manufactured products associated with this past production may
still be in use and therefore be contributing to current and future levels of release. There is insufficient
information available for EPA to quantify any additional level of current or future releases from the
COUs based on this past production. As previously stated, the 100,000 lbs/year-site value that EPA used
Page 448 of 723
-------
for the primary assessment may represent a conservative approach for current production, however, the
possibility of higher releases based on remaining stockpiles cannot be ruled out.
As previously discussed throughout the Risk Evaluation, both legacy uses (described in Section 1.2.8)
and historical activities (described in Section 1.2.9) are responsible for a subset of the total aggregate
exposure to the environment and general population. The specific percentage of these total exposures
that stem from historical activities cannot be determined and may differ both geographically and
temporally.
4.4 Other Risk Related Considerations
4.4.1 Potentially Exposed or Susceptible Subpopulations
This Risk Evaluation included risk estimates for adult workers and female workers of reproductive age
in order to account for developmental endpoints and for various lifestages of the general population in
order to account for differential exposures. MOEs for female workers of reproductive age were 10%
lower than workers overall, however in most instances the risk conclusions were the same. EPA
indicated instances in which risk conclusions differed among average workers and female workers of
reproductive age in Section 4.2.2. When risk conclusions differ for average workers and women of
childbearing age, Table 4-27 presents the risk estimate for the more sensitive subpopulation of women
of childbearing age.
Risk estimates were calculated for the highly exposed general population (representing populations
living close to a facility with HBCD releases) using the most sensitive relevant POD for both the highest
exposure sub-scenario along with a representative moderate exposure scenario. Risk estimates for the
highly exposed general population incorporated aggregate background exposure levels in addition to
modeled COU-specific exposure pathways. EPA also estimated risks for all lifestages, including the
most susceptible lifestages of infants and young toddlers. For dietary risks to infants (who are not
expected to ingest fish), risks were estimated for the absolute worst-case scenario of aggregated
exposure (including breastmilk) based on biomonitoring data (Section 4.2.3.3.2). EPA additionally
evaluated risks to susceptible lifestages from ingestion of house dust or mouthing of plastic articles. An
individual can fall into multiple PESS categories. For example, an individual may be highly exposed
because they live near a facility and may also be biologically susceptible as a pregnant mother.
Alternatively, they may live near a facility and also and be a worker.
For estimating fish ingestion exposures to the highly exposed general population, EPA selected high-end
fish ingestion rates for calculation of ADR values in order to represent high-end acute exposures. ADD
values representing chronic exposure utilized central tendency fish ingestion rates, which are expected to
be more representative of the most populations over a sustained period. While these assumptions are
expected to protect the majority of populations, there is potential for higher risk among subpopulations
with consistently higher fish consumption rates. For some populations, such as Native American tribes,
fish consumption rates may differ from that of the general population, including the highly exposed
population. Fish consumption rates among multiple tribes have been investigated, and this information is
documented in EPA's Exposure Factors Handbook (U.S. EPA. ) and other publications (Burger
2002; Critic 1994). Because ingestion rates vary across tribes, use of a single value for fish consumption
rate may over or underestimate exposures. Infants, children and pregnant woman are also groups among
Native American tribes and these populations overlap with other potentially exposed or susceptible
subpopulations. For populations with higher rates of fish ingestion, this may result in elevated exposure.
Additionally, other activities unique to these communities (e.g., open burning, (Gochfeld and Burger
2011)) may lead to additional aggregate exposure pathways which have not been characterized in this
Page 449 of 723
-------
Risk Evaluation. While EPA was unable to provide risk estimates for tribal communities, EPA estimated
risk to subsistence fishers (Section 4.2.3.2), a subpopulation that is similarly highly exposed due to
increased fish consumption relative to the general population. While fish consumption for certain tribal
communities may exceed even that of subsistence fishers, EPA assumes that these risk estimates are
applicable to the majority of communities.
4.4.2 Aggregate and Sentinel Exposures
Section 2605(b)(4)(F)(ii) of TSCA requires EPA, as a part of the Risk Evaluation, to describe whether
aggregate or sentinel exposures under the conditions of use were considered and the basis for their
consideration. EPA has defined aggregate exposure as "the combined exposures to an individual from a
single chemical substance across multiple routes and across multiple pathways." A detailed description
of the aggregate exposure evaluation is presented in Section 2.4.2.2.5. The relative contribution of each
pathway to the aggregated background exposure is shown in Table 2-88 (central tendency) and Table
2-90 (high-end). As a result of the widespread occurrence of HBCD coupled with its persistence and
bioaccumulation, aggregate exposures to the general population including consumers were considered
for HBCD by evaluating multiple pathways, routes of exposure and age groups. For all general
population exposure routes, background aggregate exposures for all exposure routes were combined
with specific modeled exposures for the pathway of interest (i.e., fish ingestion, air inhalation,
dust/indoor air, mouthing). Aggregating general population exposures is appropriate because these
background exposures are based on monitoring data and account for the persistence of HBCD in
biological tissues. While there is significant uncertainty and potential for overestimation of dermal
exposure based on use of an upper-end absorption estimate, this is a very minor contribution to the
overall general population exposure and the additional dermal contribution is unlikely to overload
toxicokinetic processes. For workers however, dermal exposure estimates are significantly higher than
inhalation exposure and it would therefore be inappropriate to add a likely highly overestimated value to
the inhalation exposure estimates without the use of a PBPK model available for determining the effect
on internal dose estimates. Therefore, EPA chose not to employ simply additivity of exposure pathways
for workers because of the uncertainties present in the current exposure estimation procedures.
Conversely, not aggregating exposures may underestimate total exposure for a given individual.
Additionally, background general population exposures were not aggregated with occupational
exposures for risk estimation to workers because background general population exposures are orders of
magnitude less than occupational exposures and would only have a negligible effect on the overall risk
estimates.
EPA defines sentinel exposure as "the exposure to a single chemical substance that represents the
plausible upper bound of exposure relative to all other exposures within a broad category of similar or
related exposures." In this Risk Evaluation, EPA considered sentinel exposures by considering
exposures to populations who may have upper bound exposures due to their exposure factors (e.g.,
higher intake rates such as elevated fish consumption), who live in close proximity to point sources
associated with the conditions of use and spend time in environments with HBCD-containing building
materials or automobile replacement parts. EPA characterized high-end exposures in evaluating both
modeled and monitored exposures to various receptors. A description of the high-end exposure estimates
is provided in Section 2.4.1.1 for workers. For the general population, risk was characterized for the
most highly exposed lifestage (i.e., <1 year olds for dust/inhalation, 1 to <2 year olds for fish ingestion).
Page 450 of 723
-------
4.5 Risk Conclusions
4.5.1 Environmental Risk Conclusions
A summary of risk estimates is provided below for aquatic and terrestrial organisms. Risk estimates
presented in tables represent the most robust and sensitive values when accounting for all of the assessed
representative species.
4.5.1,1 Summary of Risk Estimates for Aquatic Organisms
As described in Section 3.1.5, the environmental hazard thresholds are based on environmental hazard
concentrations reported for both aquatic and terrestrial organisms; environmental risk estimates are
ratios that compare the hazard threshold to exposure. Risk estimates for aquatic organisms based on
environmental monitoring data (near industrial facilities and general population sites, as categorized by
study authors) are summarized for pelagic and benthic organisms above in Section 4.1.4.1 and below in
Table 4-26. The average of 90th percentile (high-end) and mean of means (central tendency) surface
water and sediment concentrations summarized in Table 4-3 and Table 4-4, respectively, are used to
calculate risk for aquatic organisms that inhabit ecosystems near industrial facilities (point source) or
general population (non-point source) sites. Specifically, the high-end and central tendency surface
water concentrations measured near industrial facilities or general population sites were compared to all
three pelagic COCs: algae (1 |ig/L), acute fish (0.4 |ig/L) and chronic water flea (0.42 |ig/L). The algae
COC is based on observed reductions in growth rate as a result of a 72-hour exposure to HBCD. The
acute COC is based on delayed zebrafish embryo hatching as a result of a 96-hour exposure to HBCD
and the chronic water flea COC is based on reduced growth in surviving young. For characterizing risk
to benthic organisms, high-end and central tendency sediment concentrations measured near industrial
facilities or general population sites were compared to the 28-d blackworm chronic COC (1,570 |ig/kg),
based on effects on reproduction and mortality after a 56-day exposure. Summarized below in Table
4-26, RQs were equal to or above 1 (denoting risk) for all three COCs for pelagic organisms (algae,
acute fish and chronic invertebrate COCs), and the one COC for benthic organisms (chronic invertebrate
COC) based on measured surface water and sediment concentrations near industrial facilities,
respectively. On the other hand, RQs were less than one for all aquatic organisms based on
environmental monitoring data attained near general population sites.
Table 4-26 also summarizes RQs for exposure scenarios that characterize specific COUs; Sections
4.1.3.1.3 and 4.1.3.2.3 characterize the screening approach used to characterize risk for the COU of
Recycling of electronics waste containing HIPS that contain HBCD. Exposure scenario-specific risk for
aquatic organisms is summarized above in Section 4.1.3.1.2, in Table 4-5, where both 10th (high-end)
and 50th percentile (central tendency) surface water and sediment concentrations are used to calculate
risk. Additionally, the environmental hazard endpoints used to derive COCs are summarized above in
Section 4.1.2, in Table 4-1. For characterizing risk to pelagic organisms, either 1- or 21-d average
surface water concentrations of HBCD were used. Specifically, the 1-d average surface water
concentrations were compared to both the acute fish (0.4 |ig/L) and algae (1 |ig/L) COCs, and the 21-d
average surface water concentrations were compared to the chronic water flea COC (0.42 |ig/L). For
characterizing risk to benthic organisms, the 28-d average sediment concentration was compared to the
28-d blackworm chronic COC (1,570 |ig/kg). The below discussion will focus on exposure scenarios
where there is at least one exposure sub-scenario with a RQ > 1 (denoting when media exposure
concentrations exceeds the hazard threshold), using either the 10th or 50th percentile predicted surface
water or sediment HBCD concentrations.
As explained above in Section 4.1.4.1, for the exposure scenario of land disposal of other formulated
products and articles (e.g., adhesives, coatings, textiles, and electronics), environmental monitoring data
Page 451 of 723
-------
was used exclusively as the proxy for characterizing risk to aquatic organisms (whereas modeled surface
water and sediment HBCD concentrations were used to characterize risk to aquatic organisms for the
other exposure scenarios with surface water releases).
Pelagic Organism Risk based on Predicted Surface Water Concentrations of HBCD
All exposure scenarios have at least one RQ > 1 using the acute, algae or chronic COC, based on
predicted HBCD surface water releases. The below listed exposure scenarios are categorized by the
whether or not there is at least one RQ > 1 based on the acute, algae and/or chronic COCs.
For the following exposure scenarios, there are risks for pelagic organisms relative to the acute, algae
and chronic COCs. All of the below listed exposure scenarios have RQs > 1 for at least half of the
exposure sub-scenarios for both the acute and algae COCs. The single asterisk depicts whether at least
half of the exposure sub-scenarios have RQs > 1 relative to all three COCs. The double asterisk depicts
when the RQs are > 1 based on measured monitoring data near industrial facilities (background
information).
• Repackaging of Import Containers (1)*
• Compounding of Polystyrene Resin to Produce XPS Masterbatch (2)
• Processing of HBCD to produce XPS Foam using XPS Masterbatch (3)
• Processing of HBCD to produce XPS Foam using HBCD Powder (4)
• Processing of HBCD to produce EPS Foam from Imported EPS Resin Beads (5)*
• Processing of HBCD to produce SIPs and Automobile Replacement Parts from XPS/EPS Foam (6)
• Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and Commercial
Buildings, and Other Structures (9)
• Recycling of EPS Foam and Reuse of XPS Foam (10)
• Land disposal of other formulated products and articles (e.g., adhesives, coatings, textiles, and
electronics)**
For the following exposure scenario, there are risks for pelagic organisms relative to the acute and algae
COCs. The asterisk depicts whether at least half of the exposure sub-scenarios have RQs > 1 relative to
both COCs.
• Installation of XPS/EPS Foam Insulation in Residential, Public, and Commercial Buildings, and Other
Structures (8)*
For the following exposure scenario, there is risk for pelagic organisms relative to the acute COC.
• Use of Flux/Solder Pastes (12)
For the following exposure scenario, it is unlikely that there is risk for pelagic organisms relative to any
of the COCs.
• Recycling of electronics waste containing HIPs (13)
Benthic Organism Risk based on Predicted Sediment Concentrations of HBCD
Most of the exposure scenarios have at least one RQ > 1 using the chronic COC, based on predicted
HBCD sediment HBCD concentrations using either the 11- or 128-d HBCD half-life. The below listed
exposure scenarios are categorized by the whether or not there is at least one RQ > 1 based on the
chronic COC.
For the following exposure scenarios, there are risks for benthic organisms relative to the chronic COC,
using both the 11- and 128-d HBCD half-life. The single asterisk depicts whether at least half of the
exposure sub-scenarios have RQs > 1 relative to the chronic COC, using both HBCD half-lives. The
Page 452 of 723
-------
double asterisk depicts when the RQs are > 1 based on measured monitoring data near industrial
facilities (background information).
• Repackaging of Import Containers (1)*
• Processing of HBCD to produce XPS Foam using XPS Masterbatch (3)
• Processing of HBCD to produce EPS Foam from Imported EPS Resin Beads (5)*
• Processing of HBCD to produce SIPs and Automobile Replacement Parts from XPS/EPS Foam (6)
• Recycling of EPS Foam and Reuse of XPS Foam (10)
• Land disposal of other formulated products and articles (e.g., adhesives, coatings, textiles, and
electronics)**
For the following exposure scenarios, there are risks for benthic organisms relative to the chronic COC,
using both 128-d HBCD half-life. The asterisk depicts whether at least half of the exposure sub-
scenarios have RQs > 1 relative to the chronic COC.
• Compounding of Polystyrene Resin to Produce XPS Masterbatch (2)
• Processing of HBCD to produce XPS Foam using HBCD Powder (4)
• Installation of XPS/EPS Foam Insulation in Residential, Public, and Commercial Buildings, and Other
Structures (8)*
For the following exposure scenario, it is unlikely that there is risk for benthic organisms relative to the
chronic COC, using either HBCD half-lives.
• Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential, Public and Commercial
Buildings, and Other Structures (9)
• Use of Flux/Solder Pastes (12)
• Recycling of electronics waste containing HIPs (13)
As summarized below in Table 4-26, the bolded and shaded in gray text indicate when at least half of
the modeled exposure subscenarios have RQ > 1.
4,5.1.2 Summary of Risk Estimates for Terrestrial Organisms
Risk estimates for terrestrial organisms based on environmental monitoring data (near industrial
facilities and general population sites, as categorized by study authors) are summarized for terrestrial
above in Section 4.1.4.2. The average of 90th percentile (high-end) and mean of means (central
tendency) soil concentrations summarized in Table 4-6 are used to calculate risk for soil organisms that
inhabit ecosystems near industrial facilities (point source) or general population (non-point source) sites.
Specifically, the high-end and central tendency soil concentrations measured near industrial facilities or
general population sites were compared to the chronic earthworm COC (173,000 |ig/kg). RQs are less
than one based on environmental monitoring data attained near both industrial facilities and general
population sites, and are therefore not presented below in Table 4-26.
Similarly, as presented in Appendix Table Apx J-13, all RQs are < 1 when using the highest IIOAC
predictions for soil HBCD concentrations, based on exposure scenario-specific releases, in either the
fenceline or community scenarios. Section 4.1.3.2.3 also describes the screening approach used to
characterize risk to soil organisms for the COU of Recycling of electronics waste containing HIPS that
contain HBCD. The results suggest that it is unlikely that any of the exposure scenarios alone will result
in soil concentrations of HBCD that will surpass the chronic COC. Due to there being an unlikelihood of
risk to soil organisms due to chronic HBCD from air deposition, these results are not provided below in
Table 4-26.
Page 453 of 723
-------
Table 4-26. Summary of Risk for Aquatic Organisms
Life Cycle
Stage
Categorya
Subcategory b
Exposure
Scenario c
Population d
Exposure
Route
Hazard Threshold
Risk Estimates e
High-End
Central
Tendency
Manufacture
Import
Import
Section 2.4.1.2 -
Repackaging of
Import Containers
(1)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
4.3-189
0.09-24.2
Algae (COC= 1 |ig
HBCD/L)
1.72-75.6
0.04-0.83
Chronic (COC=
0.417ng HBCD/L)
3.5-21.22
0.07-2.26
Sediment
11-d Half-Life (COC:
1,570 ng/kg)
0.88-4.61
0.02-0.56
128-d Half-Life (COC:
1,570 ng/kg)
2.29-11.91
0.05-1.26
Processing
Processing-
Incorporated into
formulation,
mixture or
reaction product
Flame retardants
used in custom
compounding of
resin (e.g.,
compounding in
XPS masterbatch)
and in solder paste
Section 2.4.1.3 -
Compounding of
Polystyrene Resin
to Produce XPS
Masterbatch (2)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
3.48-34.75
0.09-2.08
Algae (COC= 1 |ig
HBCD/L)
1.39-31.3
0.04-0.83
Chronic (COC=
0.417ng HBCD/L)
0.19-4.22
0-0.1
Sediment
11-d Half-Life (COC:
1,570 ng/kg)
0.03-0.77
0-0.02
128-d Half-Life (COC:
1,570 ng/kg)
0.08-1.86
0-0.04
Incorporated into
articles
Flame retardants
used in plastics
product
manufacturing
(manufacture of
Section 2.4.1.4 -
Processing of
HBCD to produce
XPS Foam using
XPS Masterbatch
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
0.76-275
0.02-7.33
Algae (COC= 1 |ig
HBCD/L)
0.3-110
0.01-2.93
Page 454 of 723
-------
Life Cycle
Stage
Categorya
Subcategory b
Exposure
Scenario c
Population d
Exposure
Route
Hazard Threshold
Risk Estimates e
High-End
Central
Tendency
XPS and EPS foam;
manufacture of
structural insulated
panels (SIPS) and
automobile
replacement parts
from XPS and EPS
foam)
(3)
Chronic (COC=
0.417|ig HBCD/L)
0.04-13.55
0-0.34
Sediment
11-d Half-Life (COC:
1,570 ng/kg)
0.01-2.22
0-0.06
128-d Half-Life (COC:
1,570 ng/kg)
0.03-2.97
0-0.08
Section 2.4.1.5 -
Processing of
HBCD to produce
XPS Foam using
HBCD Powder
(4)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
0.91-107
0.02-2.85
Algae (COC= 1 |ig
HBCD/L)
0.36-42.8
0.01-1.14
Chronic (COC=
0.417ng HBCD/L)
0.05-5.25
0-0.13
Sediment
11-d Half-Life (COC:
1,570 ng/kg)
0.01-0.87
0-0.02
128-d Half-Life (COC:
1,570 ng/kg)
0.02-1.16
0-0.03
Section 2.4.1.6 -
Processing of
HBCD to produce
EPS Foam from
Imported EPS
Resin Beads (5)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
89.5-9,900
2.2-262.5
Algae (COC= 1 ng
HBCD/L)
35.8-3,960
0.88-105
Clironic (COC=
0.417ng HBCD/L)
33.57-
563.55
0.71-12.01
Sediment
11-d Half-Life (COC:
1,570 ng/kg)
8.73-143.31
0.21-3.52
128-d Half-Life (COC:
1,570 ng/kg)
22.68-
361.78
0.48-7.77
Section 2.4.1.7-
Processing of
Aquatic
Organisms
Surface Water
Acute (COC= 0.4 j.ig
HBCD/L)
0.97-148.75
0.02-3.93
Page 455 of 723
-------
Life Cycle
Stage
Categorya
Subcategory b
Exposure
Scenario c
Population d
Exposure
Route
Hazard Threshold
Risk Estimates e
High-End
Central
Tendency
HBCD to produce
SIPs and
Automobile
Replacement Parts
from XPS/EPS
Foam (6)
Algae (COC= 1 |ig
HBCD/L)
0.39-59.5
0.01-1.57
Chronic (COC=
0.417ng HBCD/L)
0.19-8.47
0-0.18
Sediment
11-d Half-Life (COC:
1,570 |ig/kg)
0.05-2.15
0-0.05
128-d Half-Life (COC:
1,570 ng/kg)
0.12-5.44
0-0.12
Recycling
Recycling of XPS
and EPS foam,
resin panels
containing HBCD;
Recycling of
electronics waste
containing HIPS
that contain HBCD
Section 2.4.1.11 -
Recycling of EPS
Foam and Reuse
of XPS Foam (10)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
1.2-183.25
0.03-4.88
Algae (COC= 1 |ig
HBCD/L)
0.48-73.3
0.01-1.95
Chronic (COC=
0.417ug HBCD/L)
0.45-9.02
0.01-0.22
Sediment
11-d Half-Life (COC:
1,570 |ig/kg)
0.12-1.48
0-0.04
128-d Half-Life (COC:
1,570 |ig/kg)
0.17-1.98
0-0.06
Section 2.4.1.14-
Recycling of
electronics waste
containing HIPS
(13)
Aquatic
Organisms
As discussed in Section 4.1.3.1.3, HB CD is not expected to be released
into surface water from this exposure scenario, therefore it is unlikely
that there will be risk to aquatic organisms (both pelagic and benthic).
Distribution
Distribution
Distribution
Activities related to distribution (e.g., loading, unloading) are considered throughout the life cycle, rather than
using a single distribution scenario.
Commercial
/Consumer
Use
Building/
construction
materials
Plastic articles
(hard: construction
and building
materials covering
large surface areas
(e.g., XPS/EPS
foam insulation in
residential, public
and commercial
buildings, and other
structures) and
solder paste
Section 2.4.1.9 -
Installation of
XPS/EPS Foam
Insulation in
Residential,
Public, and
Commercial
Buildings, and
Other Structures
(8)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
0.05-59.25
0.01-8.45
Algae (COC= 1 |ig
HBCD/L)
0.02-23.7
0-3.38
Chronic (COC=
0.417ug HBCD/L)
0-0.41
0-0.04
Sediment
11-d Half-Life (COC:
1,570 |ig/kg)
0.06-0.57
0.01-0.07
128-d Half-Life (COC:
1,570 ng/kg)
0.13-1.28
0.01-0.1
Page 456 of 723
-------
Life Cycle
Stage
Categorya
Subcategory b
Exposure
Scenario c
Population d
Exposure
Route
Hazard Threshold
Risk Estimates e
High-End
Central
Tendency
Section2.4.1.13 -
Use of
Flux/Solder
Pastes (12)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
0.58-1.19
0.02-0.15
Algae (COC= 1 |ig
HBCD/L)
0.23-0.47
0.01-0.06
Chronic (COC=
0.417ng HBCD/L)
0.03-0.06
0-0.01
Sediment
11-d Half-Life (COC:
1,570 ng/kg)
0-0.01
0
128-d Half-Life (COC:
1,570 ng/kg)
0.01-0.02
0
Disposal
Disposal
Land disposal of
construction and
demolition waste
Section 2.4.1.10 -
Demolition and
disposal of
XPS/EPS Foam
Insulation
Products in
Residential,
Public and
Commercial
Buildings, and
Other Structures
(9)
Aquatic
Organisms
Surface Water
Acute (COC= 0,4|ig
HBCD/L)
0.05-59.25
0.01-8.45
Algae (COC= 1 |ig
HBCD/L)
0.02-23.7
0-3.38
Chronic (COC=
0.417ng HBCD/L)
0-4.10
0-0.04
Sediment
11-d Half-Life (COC:
1,570 ng/kg)
0.01-0.1
0.002-0.02
128-d Half-Life (COC:
1,570 ng/kg)
0.001-0.01
0.0001-
0.0007
Land disposal of
formulated products
(e.g., adhesives, and
coatings) and
articles (e.g. textiles,
electrical and
electronic products)
Near Industrial
Facilities
(Point Source
Background
Exposure)f
Aquatic
Organisms
Surface Water
Acute (COC= 0.4 j.ig
HBCD/L)
2.48
2.10
Algae (COC= 1 ng
HBCD/L)
0.99
0.84
Chronic (COC=
0.417ng HBCD/L)
2.38
2.02
Sediment
Clironic (COC: 1,570
lig/kg)
3.23
2.19
Near General
Population
(Non-Point
Source
Surface Water
Acute (COC= 0.4 j.ig
HBCD/L)
2.00E-03
1.03E-03
Algae (COC= 1 |ig
HBCD/L)
8.00E-04
4.10E-04
Page 457 of 723
-------
Life Cycle
Stage
Categorya
Subcategory b
Exposure
Scenario c
Population d
Exposure
Route
Hazard Threshold
Risk Estimates e
High-End
Central
Tendency
Background
Exposure)f
Chronic (COC=
0.417|ig HBCD/L)
1.92E-03
9.83E-04
Sediment
Chronic (COC: 1,570
lig/kg)
0.01
0.004
a These categories of conditions of use appear in the Life Cycle Diagram, reflect CDR codes, and broadly represent conditions of use of HBCD in industrial and/or
commercial settings.
b These subcategories reflect more specific uses of HBCD.
0 Exposure scenarios are numbered in parentheses. This numbering will be referred to throughout the document, including for exposure subscenarios (e.g., 3.1, 3.2, etc).
Only exposure scenarios with water releases are presented in this table.
d For terrestrial soil organisms, all soil concentrations attained either through measured background concentrations, or modeled for specific air releases attributed to an
exposure scenario are all less than one, and therefore risk is unlikely for terrestrial soil organisms.
e Risk quotient ranges are bolded if there is at least one risk quotient (RQ) are equal to or greater than one (exposure exceeds the hazard threshold). Risk quotients bolded
and highlighted in gray demonstrate when at least half of the RQs for an exposure scenario are equal to or greater than one (exposure exceeds the hazard threshold).
Risk based on modeled information attributed to a release from a specific exposure scenario: For aquatic organisms, exposure scenario-specific risk estimates based on
high-end and central tendency predictions for surface water and sediment concentrations are based on 10th and 50th percentile flow rates, respectively. For terrestrial
organisms, exposure scenario-specific risk estimates for fenceline and community sites did not result in risk estimates equal to or greater than one and are provided in
Appendix J. 1.3.1.
Risk based on measured background information that is not attributed to a release from a specific exposure scenario: For aquatic and terrestrial background exposure
where risk estimates are based on monitoring data, high-end and central tendency predictions for aquatic (i.e., surface water and sediment concentrations) and terrestrial
(i.e., soil) organisms are based on an average of 90th percentile and mean of mean measured enviromnental media concentrations, respectively. Terrestrial organism risk
resulting from background exposure is described in Section 4.1.3.2.3.
f Background information is used as a proxy to characterize the risk from the COU of Disposal of other formulated products and articles (e.g., adhesives, coatings, textiles,
and electronics) because water and air releases are predicted to occur, but in lieu of not having media-specific release information for this CPU.
Page 458 of 723
-------
4.5.2 Human Health Risk Conclusions
A summary of risk estimates is provided below for workers, the general population, and consumers.
Risk estimates presented in tables represent the most robust and sensitive values when accounting for all
of the assessed lifestages and PESS groups.
4.5.2,1 Summary of Risk Estimates for Workers
Table 4-27 summarizes the risk estimates for inhalation and dermal exposures for all occupational
exposure scenarios. Risk estimates that exceed the benchmark (i.e., MOEs less than the benchmark
MOE) are highlighted by bolding the number and shading the cell in gray. The occupational exposure
assessment and risk characterization are described in more detail in Sections 2.4.1 and 4.2.2,
respectively. Occupational non-users (ONUs) are expected to have lower exposure levels than workers
in most instances but exposures could not be quantified. Based on the particulate form of HBCD with
low volatility, ONUs are not expected to be exposed at comparable levels to workers (an exception is for
installation and demolition of XPS/EPS in insulation, see Section 4.3.2.3). Specific links to the relevant
exposure sections in the document are listed in Table 4-27 in the Occupational Exposure Scenario
column.
The risk summary below is based on the most sensitive and robust acute (offspring loss) and chronic
(thyroid hormone effects) endpoints. Thyroid hormone changes (both acute and chronic) are considered
the primary effect resulting from HBCD exposure, as they lead to all of the other observed downstream
endpoints. When risk conclusions differ for average workers and women of childbearing age, Table 4-27
presents the risk estimate for the more sensitive subpopulation of women of childbearing age.
Inhalation Exposure
For acute and chronic exposure scenarios via inhalation without PPE (i.e., no respirators) there are risks
for workers relative to the benchmarks for the following occupational exposure scenarios at both the
high-end and central tendency exposure level from acute and/or chronic exposure durations.
• Repackaging of import containers
• Compounding of polystyrene resin to produce XPS Masterbatch
• Formulation of flux/solder pastes
• Processing of HBCD to produce XPS foam using HBCD powder
For the following exposure scenarios, there are risks for workers relative to the benchmarks only at
high-end exposure level from acute and/or chronic exposure durations:
• Processing of HBCD to produce EPS Foam from imported EPS resin beads
• Processing of HBCD to produce SIPs and automobile replacement parts from XPS/EPS foam
• Recycling of EPS foam and reuse of XPS foam
• Installation of XPS/EPS foam insulation in residential, public, and commercial buildings, and
other structures
• Use of flux/solder pastes
• Demolition and Disposal of XPS/EPS foam insulation products in residential, public and
commercial buildings, and other structures
When respirators are worn (APF 5, 10, or 50), risks are mitigated to below the benchmarks for both
acute and chronic exposure durations at both exposure levels. Workers exposed through installation or
demolition of XPS/EPS foam in insulation are unlikely to wear respiratory protection. Therefore, when
considering assumed PPE usage, risk remains only for the following exposure scenarios:
Page 459 of 723
-------
• Installation of XPS/EPS foam insulation in residential, public, and commercial buildings, and
other structures
• Demolition and Disposal of XPS/EPS foam insulation products in residential, public and
commercial buildings, and other structures
The following exposure scenarios did not have risk for either acute (when applicable) or chronic
exposure scenarios at any exposure level:
• Processing of HBCD to produce XPS foam using XPS Masterbatch
• Occupational microenvironments
• Recycling of electronics waste containing HIPS
Dermal Exposure
For acute and chronic exposures via dermal contact without PPE (i.e., no gloves) there are risks for
workers relative to the benchmark for the following exposure scenarios at both high and central
tendency exposure levels:
• Repackaging of import containers
• Compounding of polystyrene resin to produce XPS Masterbatch
• Formulation of flux/solder pastes
• Processing of HBCD to produce XPS foam using XPS Masterbatch
• Processing of HBCD to produce XPS foam using HBCD powder
For the following exposure scenario, there are risks for workers relative to the benchmark following
chronic exposure at the high-end exposure level:
• Use of flux/solder paste
EPA does not expect any level of dermal exposure to HBCD following proper use of impervious gloves.
Therefore, risk estimates are not provided and risks are not identified for any exposure scenario when
impervious gloves are assumed to be worn and used appropriately.
Page 460 of 723
-------
Table 4-27. Occupational Risk Summary Table
Life Cycle
Stage/
Category
Subcategory
Occupational
Exposure Scenario
(#)
Population
Exposure
Route
Sub-
Scenario
Exposure
Level
Risk Estimates for No PPE
Risk Estimates with PPE
Acute Non-
Cancer
(benchmark
MOE = 100)
Chronic Non-
Cancer
(benclunark
MOE = 300)
Acute Non-
Cancer
(benclunark
MOE = 100)
Chronic Non-
Cancer
(benclunark
MOE = 300)
Manufacture -
Import
Import
Section 2.4.1.2 - Repackaging
of Import Containers (1)
Workers
Inhalation
High-End
38
10
191
(APF 5)
519
(APF 50)
Central
Tendency
81
39
406
(APF 5)
394
(APF 10)
Dermal
High-End
4
1
Exposure not expected with
impervious gloves
Central
Tendency
12
2
Processing -
Incorporated
into
formulation,
mixture or
reaction
product
Flame retardants used
in custom
compounding of resin
(e.g., compounding in
XPS masterbatch) and
in solder paste
Section 2.4.1.3 -
Compounding of Polystyrene
Resin to Produce XPS
Masterbatch (2)
Workers
Inhalation
High-End
29
33
144
(APF 5)
1635
(APF 50)
Central
Tendency
58
112
289
(APF 5)
560
(APF 5)
Dermal
High-End
4
4
Exposure not expected
with impervious gloves
Central
Tendency
12
7
Section 2.4.1.12 - Formulation
of Flux/Solder Pastes (11)
Workers
Inhalation
High-End
29
8
144
(APF 5)
392
(APF 50)
Central
Tendency
58
31
289
(APF 5)
1533
(APF 50)
Dermal
High-End
4
1
Exposure not expected
with impervious gloves
Central
Tendency
12
2
Processing -
Incorporated
into articles
Flame retardants used
in plastics product
manufacturing
(manufacture of XPS
and EPS foam;
Section 2.4.1.4 - Processing
of HBCD to produce XPS
Foam using XPS Masterbatch
(3)
Workers
Inhalation
High-End
328
1394
1642
(APF 5)
6970
(APF 5)
Central
Tendency
903
6813
4515
(APF 5)
34065
(APF 5)
Dermal
High-End
5
22
Exposure not expected with
impervious gloves
Central
Tendency
18
39
Page 461 of 723
-------
Life Cycle
Stage/
Category
Subcategory
Occupational
Exposure Scenario
(#)
Population
Exposure
Route
Sub-
Scenario
Exposure
Level
Risk Estimates for No PPE
Risk Estimates with PPE
Acute Non-
Cancer
(benchmark
MOE = 100)
Chronic Non-
Cancer
(benclunark
MOE = 300)
Acute Non-
Cancer
(benclunark
MOE = 100)
Chronic Non-
Cancer
(benclunark
MOE = 300)
manufacture of
structural insulated
panels (SIPS) and
automobile
replacement parts
from XPS and EPS
foam)
Section 2.2.5 - Processing of
HBCD to produce XPS Foam
using HBCD Powder (4)
Workers
Inhalation
High-End
29
123
144
(APF 5)
615
(APF 5)
Central
Tendency
58
436
289
(APF 5)
2180
(APF 5)
Dermal
High-End
4
15
Exposure not expected with
impervious gloves
Central
Tendency
12
27
Section 2.4.1.6 - Processing
of HBCD to produce EPS
Foam from Imported EPS
Resin Beads (5)
Workers
Inhalation
High-End
328
159
1642
(APF 5)
795
(APF 5)
Central
Tendency
903
786
4515
(APF 5)
3930
(APF 5)
Section 2.4.1.7 - Processing of
HBCD to produce SIPs and
Automobile Replacement Parts
from XPS/EPS Foam (6)
Workers
Inhalation
High-End
328
89
1642
(APF 5)
445
(APF 5)
Central
Tendency
903
461
4515
(APF 5)
2305
(APF 5)
Processing -
Recycling
Recycling of XPS and
EPS foam, resin,
panels containing
HBCD
Section 2.4.1.11- Recycling of
EPS Foam and Reuse of XPS
Foam (10)
Workers
Inhalation
High-End
328
159
1642
(APF 5)
795
(APF 5)
Central
Tendency
903
864
4515
(APF 5)
4320
(APF 5)
Recycling of
electronics waste
containing HIPS that
contain HBCD
Section 2.4.1.14 - Recycling of
electronics waste containing
HIPS
Workers
Inhalation
High-End
722400
196224
Not
calculated
Not
calculated
Central
Tendency
5197122
2778904
Distribution -
Distribution
Distribution
Activities related to distribution (e.g., loading, unloading) are considered throughout the life cycle, rather than using a single
distribution scenario
Commercial/
consumer use
-Building/
construction
materials
Plastic articles (hard:
construction and
building materials
covering large surface
areas (e.g., XPS/EPS
foam insulation in
Section 2.4.1.9 - Installation
of XPS/EPS Foam Insulation
in Residential, Public, and
Commercial Buildings, and
Other Structures (8)b
Workers
Inhalation
High-End
328
89
1642 a
(APF 5)
445 a
(APF 5)
Central
Tendency
903
487
4515 a
(APF 5)
2435 a
(APF 5)
Page 462 of 723
-------
Life Cycle
Stage/
Category
Subcategory
Occupational
Exposure Scenario
(#)
Population
Exposure
Route
Sub-
Scenario
Exposure
Level
Risk Estimates for No PPE
Risk Estimates with PPE
Acute Non-
Cancer
(benchmark
MOE = 100)
Chronic Non-
Cancer
(benclunark
MOE = 300)
Acute Non-
Cancer
(benclunark
MOE = 100)
Chronic Non-
Cancer
(benclunark
MOE = 300)
residential, public and
commercial buildings,
and other structures)
and solder paste
Section 2.4.1.13 - Use of
Flux/Solder Pastes (12)
Workers
Dermal
High-End
1010
274
Exposure not expected with
impervious gloves
Central
Tendency
2470
540
Commercial/
consumer use
- Other
Formulated products
(e.g., adhesives and
coatings) and articles
(e.g., textiles,
electrical and
electronic products)
Section 2.4.2.2.6 -
Occupational
Microenviromnents
Workers
Multiple
High-End
N/A°
>320,000
Not
calculated
Not
calculated
Disposal -
Disposal
Land disposal of
construction and
demolition waste
Section 2.4.1.10 - Demolition
and Disposal of XPS/EPS Foam
Insulation Products in
Residential, Public and
Commercial Buildings, and
Other Structures (9) b
Workers
Inhalation
High-End
241
65
1204 a
(APF 5)
654 a
(APF 10)
Central
Tendency
688
371
3440 a
(APF 5)
1855
(APF 5)
Land disposal of
formulated products
(e.g., adhesives and
coatings) and articles
(e.g., textiles,
electrical and
electronic products)
Section 2.4.2.2.6 -
Occupational
Microenviromnents
Workers
Multiple
High-End
N/A°
>320,000
Not
calculated
Not
calculated
a EPA is presenting MOEs for respiratory PPE as a what-if scenario, however EPA believes that workers in these OES are unlikely to wear respirators.
b ONUs may be exposed to HBCD air concentrations similarly to workers in this OES.
c Background general population exposures are only relevant to chronic hazards.
Page 463 of 723
-------
4.5.2.2 Summary of Risk Estimates for General Population and Consumers
Based on qualitative consideration of the physical-chemical and fate characteristics as well as low
concentrations in surface water and the absence of any monitored levels in drinking water, HBCD is not
expected to be present in drinking water. Therefore, risks were not identified for HBCD via drinking
water exposure.
Based on qualitative consideration of the low potential for HBCD sent to landfills (e.g., construction and
demolition landfills), HBCD is not expected to migrate through the landfill to groundwater and reach
receptors via groundwater ingestion or groundwater entering surface water. HBCD is a solid and likely
to be entrained in a solid matrix (XPS/EPS foam) when disposed of in a landfill. HBCD's high soil
organic carbon partition coefficient (>100,000) and low water solubility (66 |ig/L) indicates it will
preferentially partition to soil organic carbon and exhibit very slow movement through soil to
groundwater. Therefore, risks were not identified for general population from HBCD via landfill
leachate.
Table 4-28 summarizes the risk estimates for inhalation and dermal exposures for the highly exposed
general population (including consumers). Risk estimates that exceed the benchmark (i.e., MOEs less
than the benchmark MOE) are highlighted by bolding the number and shading the cell in gray. The
highly exposed general population exposure assessment and risk characterization are described in more
detail in Sections 2.4.3 and 4.2.2, respectively. Details on the exposure assessment for each highly
exposed general population scenario can be found in Section 2.4.3, and consumer scenarios are
described in Section 2.4.4.
The risk summary below is based on the most sensitive and robust acute (offspring loss) and chronic
(thyroid hormone effects) endpoints. Thyroid hormone changes (both acute and chronic) are considered
the primary effect resulting from HBCD exposure, as they are associated with all of the other observed
downstream endpoints.
Page 464 of 723
-------
Table 4-28. Highly Exposed General Population/Consumer Risk Summary Table
Life Cycle
Stage/
Category
Subcategory
Exposure Scenario
(#)
Population
Exposure
Route
Sub-
Scenario
Exposure
Level
Risk Estimates
Acute Non-
Cancer
(benchmark
MOE = 100)
Chronic
Non-Cancer
(benchmark
MOE = 300)
Manufacture -
Import
Import
Repackaging of Import
Containers (1)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
37630
>16800
Highest
1307
Fish
Ingestion
Moderate
1678
13493
Highest
338
3314
Processing -
Incorporated into
formulation,
mixture or
reaction product
Flame retardants used
in custom
compounding of resin
(e.g., compounding in
XPS masterbatch) and
in solder paste
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch (2)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
209835
>16800
Highest
128508
Fish
Ingestion
Moderate
15033
42626
Highest
1763
32594
Formulation of Flux/Solder
Pastes (11)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
119229
>16800
Highest
39092
Processing -
Incorporated into
articles
Flame retardants used
in plastics product
manufacturing
(manufacture of XPS
and EPS foam;
manufacture of
structural insulated
panels (SIPS) and
automobile
replacement parts
from XPS and EPS
foam)
Processing of HBCD to
produce XPS Foam
using XPS Masterbatch
(3)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
20056
>16800
Highest
2743
Fish
Ingestion
Moderate
7187
48741
Highest
509
15499
Processing of HBCD to
produce XPS Foam
using HBCD Powder (4)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
39449
>16800
Highest
2622
Fish
Ingestion
Moderate
14541
52951
Highest
1308
27971
Processing of HBCD to
produce EPS Foam from
Imported EPS Resin
Beads (5)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
4705
>16800
Highest
680
Fish
Ingestion
Moderate
139
5376
Highest
14
587
Processing of HBCD to
produce SIPs and
Automobile
Replacement Parts from
XPS/EPS Foam (6)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
154878
>16800
Highest
14212
Fish
Ingestion
Moderate
4234
43862
Highest
922
23422
Processing -
Recycling
Recycling of XPS and
EPS foam, resin,
panels containing
HBCD
Recycling of EPS Foam
and Reuse of XPS Foam
(10)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
140770
>16800
Highest
38255
Fish
Ingestion
Moderate
7939
34063
Highest
764
20463
Page 465 of 723
-------
Life Cycle
Stage/
Category
Subcategory
Exposure Scenario
(#)
Population
Exposure
Route
Sub-
Scenario
Exposure
Level
Risk Estimates
Acute Non-
Cancer
(benchmark
MOE = 100)
Chronic
Non-Cancer
(benchmark
MOE = 300)
Recycling of
electronics waste
containing HIPS that
contain HB CD
Recycling of electronics
waste containing HIPS
(13)
General
Population
(Highly
Exposed)
Air
Inhalation
Relative
Riskb
>680
>16800
Distribution -
Distribution
Distribution
Activities related to distribution (e.g., loading, unloading) are considered throughout the life
cycle, rather than using a single distribution scenario
Commercial/
consumer use -
Building/
construction
materials
Plastic articles (hard:
construction and
building materials
covering large surface
areas (e.g., XPS/EPS
foam insulation in
residential, public and
commercial buildings,
and other structures)
and solder paste
Installation of XPS/EPS
Foam Insulation in
Residential, Public, and
Commercial Buildings, and
Other Structures (8)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
77282
>16800
Highest
62609
Fish
Ingestion
Moderate
16081
46588
Highest
1687
17074
Consumers
Dust/
Indoor air
Single
Scenario
35411
22722
Use of Flux/Solder Pastes
(12)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
222576
>16800
Highest
221704
Fish
Ingestion
Moderate
127338
56195
Highest
80233
54800
Commercial/
consumer use -
Other
Automobile
replacement parts
Installation of Automobile
Replacement Parts (7)
Consumers
Dust/
indoor air
Single
Scenario
11259
52020
Plastic and other
articles
Mouthing of articles
containing HBCD
Consumers
Mouthing
Single
Scenario
944
2713
Formulated products
(e.g., adhesives and
coatings) and articles
(e.g., textiles,
electrical and
electronic products)
General Population
Background Exposure
General
Population
Multiple
Central
Tendency
N/A°
>42129
High-End
N/A°
>9959
Disposal -
Disposal
Other land disposal
(e.g., construction and
demolition waste)
Demolition and Disposal
of XPS/EPS Foam
Insulation Products in
Residential, Public and
Commercial Buildings, and
Other Structures (9)
General
Population
(Highly
Exposed)
Air
Inhalation
Moderate
224448
>16800
Highest
10310
Fish
Ingestion
Moderate
2520
22163
Highest
254
3388
a Background general population exposures are only relevant to chronic hazards.
b Exposure estimates were not formally calculated for this COU. Risk was estimated by comparing releases and potential MOEs
relative to worst-case sub-scenarios.
c Background general population exposures are only relevant to chronic hazards.
Page 466 of 723
-------
EPA also estimated risks to additional PESS groups based on aggregate exposures that could not be
directly tied to a particular exposure scenario or COU. Instead, risk estimates are based on aggregated
ambient or background exposure via all exposure routes and therefore only risks resulting from chronic
exposures were estimated. Table 4-29 presents a summary of risk estimates for these groups based on
the most sensitive endpoint of thyroid hormone effects. Risks were not identified for any of these PESS
groups even based on the most sensitive endpoint and exposure estimates.
Table 4-29. Risk Summary for Additional PESS Groups
Receptor
Chronic MOE
(benchmark MOE = 300)
Section Reference
Infants (<1 year old)
(Maximum Estimated Dose,
assumed 90%tile as high-end of monitoring data)
468
Section 4.2.3.3.2
Subsistence Fishers
(Near-Field, High-End [95%tile])
2215
Table 4-20
Page 467 of 723
-------
5 UNREASONABLE RISK DETERMINATION
5.1 Overview
In each Risk Evaluation under TSCA section 6(b), EPA determines whether a chemical substance
presents an unreasonable risk of injury to health or the environment, under the conditions of use. These
determinations do not consider costs or other non-risk factors. In making these determinations, EPA
considers relevant risk-related factors, including, but not limited to: the effects of the chemical substance
on health and human exposure to such substance under the conditions of use (including cancer and non-
cancer risks); the effects of the chemical substance on the environment and environmental exposure
under the conditions of use; the population exposed (including any potentially exposed or susceptible
subpopulations (PESS)); the severity of hazard (including the nature of the hazard, the irreversibility of
the hazard); and uncertainties. EPA also takes into consideration the Agency's confidence in the data
used in the risk estimate. This includes an evaluation of the strengths, limitations, and uncertainties
associated with the information used to inform the risk estimates and the risk characterization. This
approach is in keeping with the Agency's final rule, Procedures for Chemical Risk Evaluation Under the
Amended Toxic Substances Control Act ( 26).21
This section describes the final unreasonable risk determinations for the conditions of use in the scope of
the Risk Evaluation for the cyclic aliphatic bromide cluster chemicals. EPA evaluated two of the three
chemicals in the cluster: CASRN 25637-99-4 and CASRN 3194-55-6. In this final Risk Evaluation
document, the use of "HBCD" refers to either or both chemicals. No conditions of use were identified
for the third chemical, CASRN 3194-57-8. The final unreasonable risk determinations are based on the
risk estimates in the final Risk Evaluation, which may differ from the risk estimates in the draft Risk
Evaluation due to peer review and public comments. Therefore, the final unreasonable risk
determinations of some conditions of use may differ from those in the draft Risk Evaluation.
5,1.1 Human Health
EPA's Risk Evaluation identified non-cancer adverse effects from acute and chronic exposures to
HBCD. The health risk estimates from inhalation and dermal exposures for all conditions of use are in
Section 4.2 (Table 4-14 through Table 4-24).
EPA evaluated exposures to workers, ONUs, general population, and consumers, using reasonably
available monitoring and modeling data for inhalation, and dermal exposures, as applicable.
For the HBCD Risk Evaluation, EPA identified and evaluated as Potentially Exposed or Susceptible
Subpopulations: workers, occupational non-users (ONUs), subsistence fishers, females of reproductive
age, young children, and the highly exposed general population (and consumers) living near or with an
HBCD point source. (Section 4.4.1).
The description of the data used for human health hazard is in Section 3. Uncertainties in the analysis
are discussed in Section 4.3 and considered in the risk determination for each condition of use below,
including that EPA was unable to model the potential effects of bioaccumulation in human tissues over
time, EPA was unable to quantify ONU exposure due to lack of adequate data or relevant models, and
estimated fish ingestion exposure is highly dependent on the selected Bioaccumulation Factor (BAF)
value.
21 This risk determination is being issued under TSCA section 6(b) and the terms used, such as unreasonable risk, and the
considerations discussed are specific to TSCA. Other statutes have different authorities and mandates and may involve risk
considerations other than those discussed here.
Page 468 of 723
-------
5.1.1.1 Non-Cancer Risk Estimates
The risk estimates of non-cancer effects (MOEs) refers to adverse health effects associated with health
endpoints other than cancer, including to the body's organ systems, such as thyroid effects, liver effects,
and reproductive/developmental effects. The MOE is the point of departure (POD) (an approximation of
the no-observed adverse effect level (NOAEL) or benchmark dose level (BMDL)) for a specific health
endpoint divided by the exposure concentration for the specific scenario of concern. Section 3.2.5
presents the PODs for acute and chronic non-cancer effects for HBCD and Section 4.2 presents the
MOEs for acute and chronic non-cancer effects.
The MOEs are compared to a benchmark MOE. The benchmark MOE accounts for the total uncertainty
in a POD, including, as appropriate: (1) the variation in sensitivity among the members of the human
population (i.e., intrahuman/intraspecies variability); (2) the uncertainty in extrapolating animal data to
humans (i.e., interspecies variability); (3) the uncertainty in extrapolating from data obtained in a study
with less-than-lifetime exposure to lifetime exposure (i.e., extrapolating from subchronic to chronic
exposure); and (4) the uncertainty in extrapolating from a lowest observed adverse effect level (LOAEL)
rather than from a NOAEL. A lower benchmark MOE (e.g., 30) indicates greater certainty in the data
(because fewer of the default UFs relevant to a given POD as described above were applied). A higher
benchmark MOE (e.g., 1000) would indicate more uncertainty for specific endpoints and scenarios.
However, these are often not the only uncertainties in a Risk Evaluation. The benchmark MOE for the
most robust and sensitive acute non-cancer risks for HBCD is 100 (accounting for intraspecies and
interspecies variability). The benchmark MOE for the most robust and sensitive chronic non-cancer risks
for HBCD is 300 (accounting for interspecies and intraspecies variability as well as subchronic to
chronic extrapolation). Additional information regarding the benchmark MOE is in Section 3.2.6.
5.1.1.2 Cancer Risk Estimates
EPA did not evaluate cancer risk from exposure to HBCD. Overall, given the limited data and mixed
results between mammalian and non-mammalian systems, there is indeterminate evidence to make a
conclusion on the genotoxicity of HBCD. The only experimental animal study to examine cancer
endpoints concluded that HBCD was not carcinogenic, however, this study was only available as an
incomplete report (Kurokawa et al. 1984). Therefore, according to the U.S. EPA Guidelines for
Carcinogen Risk Assessment (U.S. EPA. 20051 there is "inadequate information to assess the
carcinogenic potential" of HBCD. As a result, this hazard was not carried forward for dose-response
analysis or risk estimation. (Section 3.2.4.2)
5.1.1.3 Determining Unreasonable Risk of Injury to Health
Calculated risk estimates (MOEs or cancer risk estimates) can provide a risk profile by presenting a
range of estimates for different health effects for different conditions of use. A calculated MOE that is
less than the benchmark MOE indicates likely risk to human health of non-cancer effects. A calculated
cancer risk estimate that is greater than the cancer benchmark indicates likely risk to human health of
cancer. Whether those risks are unreasonable will depend upon other risk-related factors, such as the
endpoint under consideration, the reversibility of effect, exposure-related considerations (e.g., duration,
magnitude, or frequency of exposure, or population exposed), and the confidence in the information
used to inform the hazard and exposure values. A calculated MOE greater than the benchmark MOE or a
calculated cancer risk estimate less than the benchmark, alone do not support a determination of
unreasonable risk, since EPA may consider other risk based factors when making an unreasonable risk
determination.
EPA may make an unreasonable risk determination when the risk affects the general population or a
PESS that was identified as relevant. For workers (who are one example of PESS), when making an
Page 469 of 723
-------
unreasonable risk determination, EPA also makes assumptions regarding workplace practices and
exposure controls, including engineering controls or use of personal protective equipment (PPE).
However, EPA does not assume that ONUs use PPE. For each condition of use of HBCD with an
identified risk for workers, EPA assumes, as a baseline, the use of a respirator with an APF of 5, 10, or
50. Similarly, EPA assumes the use of impervious gloves in industrial settings. However, EPA assumes
that for some conditions of use, the use of appropriate respirators is not a standard industry practice,
based on best professional judgment given the burden associated with the use of supplied-air respirators,
including the expense of the equipment and the necessity of fit-testing and training for proper use. Once
EPA has applied the appropriate PPE assumption for a particular condition of use in each unreasonable
risk determination, in those instances when EPA assumes PPE is used, EPA also assumes that the PPE is
used in a manner that achieves the stated APF or PF. EPA's decisions for unreasonable risk to workers
are based on high-end exposure estimates, in order to capture not only exposures for PESS but also to
account for the uncertainties related to whether or not workers are using PPE.
In the HBCD risk characterization, offspring loss was identified as the most robust and sensitive
endpoint for non-cancer adverse effect from acute exposures and thyroid effects were identified as the
most robust and sensitive endpoint for non-cancer adverse effects from chronic exposures for all
conditions of use. However, additional risks associated with other adverse effects (e.g. liver effects,
other reproductive/developmental effects) were also identified for acute and chronic exposures.
Determining unreasonable risk by using offspring loss and thyroid effects will also include the
unreasonable risk from other endpoints resulting from acute or chronic inhalation and dermal exposures.
When making a determination of unreasonable risk, the Agency has a higher degree of confidence where
uncertainty is low. Similarly, EPA has high confidence in the hazard and exposure characterizations
when, for example, the basis for the characterizations is measured or monitoring data or a robust model
and the hazards identified for risk estimation are relevant for conditions of use. Where EPA has made
assumptions in the scientific evaluation, whether or not those assumptions are protective is also a
consideration. Additionally, EPA considers the central tendency and high-end scenarios when
determining the unreasonable risk. High-end risk estimates (e.g., 95th percentile) are generally intended
to cover individuals or sub-populations with greater exposure (PESS) and central tendency risk
estimates are generally estimates of average or typical exposure.
EPA may make a determination of no unreasonable risk for conditions of use where the substance's
hazard and exposure potential, or where the risk-related factors described previously, lead the Agency to
determine that the risks are not unreasonable.
5,1.2 Environment
EPA's Risk Evaluation identified adverse effects resulting from acute and chronic exposures to HBCD
for both aquatic and terrestrial organisms for all conditions of use, as summarized in Section 3.1. The
environmental hazard threshold is calculated for both aquatic and terrestrial organisms. The hazard
threshold for aquatic organisms takes into account an assessment factor that represents uncertainties
explained in Section 3.1.5, therefore allowing a concentration of concern (COC) to be derived.
Limitations in data availability regarding HBCD toxicity to terrestrial organisms do not allow for an
assessment factor to be used to derive a COC, therefore the hazard threshold is based on reported hazard
effect concentrations reported by key studies summarized in Section 3.1.5. The description of the data
used for environmental exposure is in Section 2.3. The environmental concentration is determined based
on the levels of the chemical released to the environment (e.g., surface water, sediment, soil, biota)
under the conditions of use, based on the fate properties, release potential, and reasonably available
Page 470 of 723
-------
environmental monitoring data. Section 4.1. provides more detail regarding the risk quotient derivations
for HBCD.
EPA calculated a risk quotient (RQ) to compare environmental concentrations against a hazard
threshold. The environmental risk estimates from exposure to HBCD via water (e.g., surface water and
sediment) and air (e.g., soil) releases are characterized in Section 4.1 (Table 4-3 through Table 4-7).
Uncertainties in the analysis are discussed in Section 4.3 and considered in the risk determination for
each condition of use below, including the fact that despite HBCD being a PBT, exposure to HBCD
across and within media types were not aggregated to estimate risk (as explained in Section 4.1.3),
therefore environmental risk may be underestimated for aquatic and terrestrial organisms.
5,1,2.1 Determining Unreasonable Risk of Injury to the Environment
Calculated risk estimates (RQs) can provide a risk profile by presenting a range of estimates for different
environmental hazard effects for different conditions of use. A calculated RQ that is equal to or greater
than one indicates likely risk to environmental health (exposure exceeds the hazard threshold), whereas
a calculated RQ that is less than one indicates that there is unlikely to be risk to environmental health
(exposure is less than the hazard threshold). Consistent with EPA's human health evaluations, the RQ is
not treated as a bright line and other risk-based factors may be considered (e.g., confidence in the hazard
and exposure characterization, duration, magnitude, uncertainty) for purposes of making an
unreasonable risk determination.
EPA may make an unreasonable risk determination when the risk affects organisms that are identified as
being relevant. Based on the available hazard data for aquatic and terrestrial organisms, EPA based
environmental risk for conditions of use on predicted media-specific HBCD concentrations. Although
EPA acknowledges that due to the physical-chemical properties of HBCD that dietary exposure is likely,
HBCD release information cannot be directly used to extrapolate tissue concentrations of prey of either
aquatic or terrestrial organisms; monitoring data was primarily used for the trophic transfer estimation of
HBCD (Section 3.1.3), and that is used to evaluate the potential for HBCD to undergo trophic transfer
due to all activities and releases that likely contribute to HBCD background exposures. Due to the lack
of HBCD hazard information regarding terrestrial organism exposure, terrestrial organism risk resulting
from HBCD exposure is limited to that for soil organisms (e.g., earthworms), and EPA acknowledges
this uncertainty.
In the HBCD risk characterization, delayed hatching and reduced growth of offspring were identified as
the most robust and sensitive endpoints for pelagic organisms due to acute and chronic exposures of
HBCD, respectively. EPA evaluated algae risk separately from the categorization of an acute or chronic
exposure, and unreasonable risk of reduced algae growth was evaluated. The most robust and sensitive
endpoint identified for benthic organisms due to chronic HBCD exposure was reduced reproduction.
EPA also identified reduced reproduction and survival of soil organisms due to chronic exposure to
HBCD as being the most robust and sensitive endpoint.
When making a determination of unreasonable risk, the Agency has a higher degree of confidence where
uncertainty is low. Similarly, EPA has high confidence in the hazard and exposure characterizations
when, for example, the basis for the characterizations is measured or monitoring data or a robust model
and the hazards identified for risk estimation are relevant for conditions of use. Where EPA has made
assumptions in the scientific evaluation, whether or not those assumptions are protective is also a
consideration. Additionally, EPA considers the central tendency and high-end scenarios when
determining the unreasonable risk. High-end risk estimates (e.g., 90th percentile) are generally intended
Page 471 of 723
-------
to cover organisms or populations with greater exposure (those inhabiting ecosystems near industries)
and central tendency risk estimates are generally estimates of average or typical exposure.
EPA may make a determination of no unreasonable risk for conditions of use where the substance's
hazard and exposure potential, or where the risk-related factors described previously, lead the Agency to
determine that the risks are not unreasonable.
5.2 Detailed Unreasonable Risk Determinations by Condition of Use
Table 5-1. Categories and Subcategories of Conditions of Use Included in the Scope of the Risk
Evaluation
Life Cycle
Stage
Category a
Subcategory b
Unreasonable
Risk
Detailed Risk
Determination
Manufacture
Import
Import
yes
Section 5.2.1.1 and
Section 5.2.2.1
Processing
Incorporated into
formulation, mixture
or reaction product
Flame retardants used in custom
compounding of resin (e.g.,
compounding in XPS
masterbatch) and in solder paste
yes
Section 5.2.1.2 and
Section 5.2.2.2
Incorporated into
article
Flame retardants used in plastics
product manufacturing
(manufacture of XPS and EPS
foam; manufacture of structural
insulated panels (SIPS) and
automobile replacement parts
from XPS and EPS foam)
yes
Section 5.2.1.3 and
Section 5.2.2.3
Recycling
Recycling of XPS and EPS
foam, resin, panels containing
HBCD
yes
Section 5.2.1.4 and
Section 5.2.2.4
Recycling
Recycling of electronics waste
containing HIPs that contains
HBCD
no
Section 5.2.1.5 and
Section 5.2.2.5
Distribution
Distribution
Distribution
no
Section 5.2.1.6 and
Section 5.2.2.6
Commercial/
consumer Use
Building/construction
materials
Plastic articles (hard):
construction and building
materials covering large surface
areas (e.g., XPS/EPS foam
insulation in residential, public
and commercial buildings, and
other structures) and solder paste
yes
Section 5.2.1.7 and
Section 5.2.2.7
Other
Automobile replacement parts
no
Section 5.2.1.8
Plastic and other articles d
no
Section 5.2.1.8
Formulated products (e.g.,
adhesives and coatings) and
articles (e.g., electronics
products and textiles)
no
Section 5.2.1.10
Disposal
Disposal
Land disposal (e.g., EPS and
XPS foam insulation)
yes
Section 5.2.1.10 and
Section 5.2.2.8
Page 472 of 723
-------
Life Cycle
Stage
Category a
Subcategory b
Unreasonable
Risk
Detailed Risk
Determination
Land disposal of formulated
products (e.g., adhesives and
coatings) and articles (e.g.,
electronics products and textiles)
no
Section 5.2.1.11 and
Section 5.2.2.9
aThese categories of conditions of use appear in the Life Cycle Diagram, reflect CDR codes, and broadly represent conditions
of use of HBCD in industrial and/or commercial settings and of consumer uses.
b These subcategories reflect more specific uses of HBCD
** Although EPA has identified both industrial and commercial uses here for purposes of distinguishing scenarios in this
document, the Agency interprets the authority over "any manner or method of commercial use" under TSCA section 6(a)(5)
to reach both.
5.2.1 Human Health
In addition to COU-specific determinations below. EPA also evaluated risks to the general population
(Table 4-19) and other Potentially Exposed or Susceptible Subpopulations (PESS) (Table 4-29) based on
aggregate general exposure to HBCD not associated with any particular COU. The PESS groups include
subsistence fishers and newborns less than 1 year old. For each of these groups, EPA did not find
unreasonable risk.
While HBCD is released to landfills, EPA determined the evaluation does not support an unreasonable
risk determination to the general population via landfill (e.g., construction and demolition landfill)
leachate based on a qualitative assessment of HBCD's migration through the landfill to groundwater and
to receptors via groundwater ingestion or groundwater entering surface water. HBCD is a solid and
likely to be entrained in a solid matrix (XPS/EPS foam) when disposed of in a landfill. HBCD's high
soil organic carbon partition coefficient and low water solubility indicates it will preferentially partition
to soil organic carbon and exhibit very slow movement through soil to groundwater.
While HBCD is released to surface water, EPA determined during problem formulation that no further
analysis beyond what was presented in the problem formulation document would be done for the
drinking water exposure pathway in this Risk Evaluation. While this exposure pathway remains in the
scope of the risk evaluation, EPA found no further analysis was necessary. EPA determined that the
evaluation does not support an unreasonable risk determination to the general population via drinking
water based on a qualitative assessment of the physical chemical properties and fate of HBCD in the
environment as well as the absence of any detection of HBCD in monitored water samples.
5.2.1.1_ Manufacturing - Import - (Import)
Section 6(b)(4)(A) unreasonable risk determination for import of HBCD: Does not present an
unreasonable risk of injury to health (workers, ONUs, and highly exposed general population).
For workers, EPA found that there was no unreasonable risk of non-cancer effects from acute (offspring
loss) or chronic (thyroid effects) inhalation or dermal exposures at the central tendency or high-end,
when assuming use of PPE. For the highly exposed general population, EPA found that there was no
unreasonable risk of non-cancer effects from acute (offspring loss) or chronic (thyroid effects) inhalation
or fish ingestion at the moderate or highest sub-scenario exposure levels.
EPA's determination that the import of HBCD does not present an unreasonable risk is based on the
comparison of the risk estimates for non-cancer effects to the benchmarks (Table 4-14 through Table
4-17) and other considerations.
Page 473 of 723
-------
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• EPA assumes workers use PPE (respirators and gloves).
• For workers, when assuming use of respirators with APF of 5 and 10, the risk estimates of non-
cancer effects from acute and chronic inhalation exposures at the central tendency and high-end
are higher than the MOE and do not support an unreasonable risk determination (Table 4-27).
• EPA does not expect any level of dermal exposure to HBCD following proper use of impervious
gloves. Therefore, risk estimates are not provided and risks are not identified for any exposure
scenario when impervious gloves are assumed to be worn and used appropriately (Section
4.5.2.1).
• Exposures to ONUs are expected to be lower than those for workers. Risk estimates for
inhalation exposure to ONUs were not quantified (Table 4-13).
• For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and highest sub-scenario exposure
levels are above the MOE and do not support an unreasonable risk determination (
Page 474 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to health from the
import of HBCD.
5.2.1.2 Processing - Incorporated into Formulation, Mixture or Reaction Product
Maine Retardants used in Custom Compounding of Resin (e.gcompounding
in XPS masterbatch) and in Solder Paste
Section 6(b)(4)(A) unreasonable risk determination for processing of HBCD into a formulation: Does
not present an unreasonable risk of injury to health (workers, ONUs, and highly exposed general
population).
For workers, EPA found that there was no unreasonable risk of non-cancer effects from acute (offspring
loss) or chronic (thyroid effects) inhalation or dermal exposures at the central tendency or high-end,
when assuming use of PPE.
For the highly exposed general population, EPA found that there was no unreasonable risk of non-cancer
effects from acute (offspring loss) or chronic (thyroid effects) inhalation or fish ingestion at the
moderate or highest sub-scenario exposure levels.
EPA's determination that the processing of HBCD into a formulation, mixture or reaction product does
not present an unreasonable risk is based on the comparison of the risk estimates for non-cancer effects
to the benchmarks (Table 4-14 through Table 4-17) and other considerations.
Page 475 of 723
-------
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• EPA assumes workers use PPE (respirators and gloves).
• For workers, when assuming use of respirators with APF of 5 and 10, the risk estimates of non-
cancer effects from acute and chronic inhalation exposures at the central tendency and high-end
are higher than the MOE and do not support an unreasonable risk determination (Table 4-27).
• EPA does not expect any level of dermal exposure to HBCD following proper use of impervious
gloves. Therefore, risk estimates are not provided and risks are not identified for any exposure
scenario when impervious gloves are assumed to be worn and used appropriately (Section
4.5.2.1).
• Exposures to ONUs are expected to be lower than those for workers. Risk estimates for
inhalation exposure to ONUs were not quantified (Table 4-13).
• For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and high sub-scenario exposure
levels are above the MOE and do not support an unreasonable risk determination (
Page 476 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to health from
processing HBCD into a formulation.
5.2,1.3 Processing - Incorporation into an Article - Flame Retardants used in
Plastics Product Manufacturing (manufacture of XPS and EPS foam;
manufacture of structural insulation panels (SIPS) and automobile
replacement parts from XPS and EPS foam)
Section 6(b)(4)(A) unreasonable risk determination for Processing of HBCD into an article: Does not
present an unreasonable risk of injury to health (workers, ONUs, and highly exposed general
population).
For workers, EPA found that there was no unreasonable risk of non-cancer effects from acute (offspring
loss) or chronic (thyroid effects) inhalation or dermal exposures at the central tendency or high-end,
when assuming use of PPE.
For the highly exposed general population, EPA found that there was no unreasonable risk of non-cancer
effects from acute (offspring loss) or chronic (thyroid effects) inhalation or fish ingestion at the
moderate or highest sub-scenario exposure levels.
EPA's determination that the processing of HBCD into an article does not present an unreasonable risk
is based on the comparison of the risk estimates for non-cancer effects to the benchmarks (Table 4-14
through Table 4-17) and other considerations.
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• EPA assumes workers use PPE (respirators and gloves).
• For workers, when assuming use of respirators with APF of 5 and 10, the risk estimates of non-
cancer effects from acute and chronic inhalation exposures at the central tendency and high-end
are higher than the MOE and do not support an unreasonable risk determination (Table 4-27).
• EPA does not expect any level of dermal exposure to HBCD following proper use of impervious
gloves. Therefore, risk estimates are not provided and risks are not identified for any exposure
scenario when impervious gloves are assumed to be worn and used appropriately (Section
4.5.2.1).
• Exposures for ONUs are expected to be lower than for workers. Risk estimates for inhalation
exposure to ONUs were not quantified (Table 4-13).
• For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and high sub-scenario exposure
levels are above the MOE and do not support an unreasonable risk determination (
Page 477 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to health from
processing HBCD into an article.
5.2,1.4 Processing - Recycling - Recycling of XPS and EPS Foam, Resin, Panels
containing HBCD
Section 6(b)(4)(A) unreasonable risk determination for recycling of XPS and EPS foam, resin, panels
containing HBCD: Does not present an unreasonable risk of injury to health (workers, ONUs, and
highly exposed general population).
For workers, EPA found that there was no unreasonable risk of non-cancer effects from acute (offspring
loss) or chronic (thyroid effects) inhalation or dermal exposures at the central tendency or high-end,
when assuming use of PPE.
For the highly exposed general population, EPA found that there was no unreasonable risk of non-cancer
effects from acute (offspring loss) or chronic (thyroid effects) inhalation or fish ingestion at the
moderate or highest sub-scenario exposure levels.
EPA's determination that the recycling of XPS and EPS foam, resin, and panels containing HBCD does
not present an unreasonable risk is based on the comparison of the risk estimates for non-cancer effects
to the benchmarks (Table 4-14 through Table 4-17) and other considerations.
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• EPA assumes workers use PPE (respirators and gloves).
• For workers, when assuming use of respirators with APF of 5 and 10, the risk estimates of non-
cancer effects from acute and chronic inhalation exposures at the central tendency and high-end
are higher than the MOE and do not support an unreasonable risk determination (Table 4-27).
• EPA does not expect any level of dermal exposure to HBCD following proper use of impervious
gloves. Therefore, risk estimates are not provided and risks are not identified for any exposure
scenario when impervious gloves are assumed to be worn and used appropriately (Section
4.5.2.1).
• Exposures for ONUs are expected to be lower than for workers. Risk estimates for inhalation
exposure to ONUs were not quantified (Table 4-13).
• For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and highest sub-scenario exposure
levels are above the MOE and do not support an unreasonable risk determination (
Page 478 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to health from
recycling of XPS and EPS foam, resin, panels containing HBCD.
5.2.1.5 Processing - Recycling - Recycling of electronics waste containing HIPS that
contain HBCD
Section 6(b)(4)(A) unreasonable risk determination for recycling of electronics waste containing HIPS
that contain HBCD: Does not present an unreasonable risk of injury to health (workers, ONUs, and
highly exposed general population).
For workers, EPA found that there was no unreasonable risk of non-cancer effects from acute (offspring
loss) or chronic (thyroid effects) inhalation or dermal exposures at the central tendency or high-end,
when assuming use of PPE.
For the highly exposed general population, EPA found that there was no unreasonable risk of non-cancer
effects from acute (offspring loss) or chronic (thyroid effects) inhalation or fish ingestion at the highest
or moderate sub-scenario exposure levels.
EPA's determination that the recycling of electronics waste containing HIPS that contain HBCD does
not present an unreasonable risk is based on the comparison of the risk estimates for non-cancer effects
to the benchmarks (Table 4-14 through Table 4-17) and other considerations.
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• Risk estimates are well above the benchmark MOE for non-cancer effects from acute or chronic
exposures and do not support an unreasonable risk determination (Table 4-27).
• EPA assumes workers use PPE (respirators and gloves).
• For workers, when assuming use of respirators with APF of 5 and 10, the risk estimates of non-
cancer effects from acute and chronic inhalation exposures at the central tendency and high-end
are above the MOE (Table 4-27).
• EPA does not expect any level of dermal exposure to HBCD following proper use of impervious
gloves. Therefore, risk estimates are not provided and risks are not identified for any exposure
scenario when impervious gloves are assumed to be worn and used appropriately (Section
4.5.2.1).
• Exposures for ONUs are expected to be lower than for workers. Risk estimates for inhalation
exposure to ONUs were not quantified (Table 4-13).
• For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and highest sub-scenario exposure
levels are above the MOE and do not support an unreasonable risk determination (
Page 479 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to health from
recycling of electronics waste containing HIPS that contains HBCD.
5.2.1.6 Distribution in Commerce - Distribution - Distribution
Section 6(b)(4)(A) unreasonable risk determination for distribution in commerce of HBCD: Does not
present an unreasonable risk of injury to health (workers and ONUs).
For the purposes of the risk determination, distribution in commerce of HBCD is the transportation
associated with the moving of HBCD in commerce. Activities related to distribution (e.g., loading,
unloading) are considered throughout the life cycle, rather than using a single distribution scenario. EPA
assumes transportation of HBCD is conducted taking similar measures as the transportation of
hazardous materials.
5.2.1.7 Commercial/Consumer Use - Building/Construction Materials - Plastic
Articles (hard) Construction and Building Materials covering Large Surface
Areas (e.g., EPS/XPS foam insulation in residential, public and commercial
buildings, and other structures) and Solder Paste
Section 6(b)(4)(A) unreasonable risk determination for commercial/consumer use of
building/construction materials and solder paste: Presents an unreasonable risk of injury to health
(workers and ONUs). Does not present an unreasonable risk to health for the highly exposed general
population including consumers.
For workers and ONUs, EPA found that there was an unreasonable risk of non-cancer effects
from chronic (thyroid effects) inhalation or dermal exposures at the high-end, without assuming
use of PPE. For the highly exposed general population EPA found that there was no unreasonable risk
of non-cancer effects from acute (offspring loss) or chronic (thyroid effects) air inhalation or fish
ingestion at the highest or moderate sub-scenario exposure levels. In addition, for consumers, EPA
found that there was no unreasonable risk of non-cancer effects from acute (offspring loss) or chronic
(thyroid effects) exposure to dust and indoor air from installation of XPS/EPS foam insulation.
EPA's determination that commercial/consumer use of HBCD in building/construction materials by
workers presents an unreasonable risk to health is based on the comparison of the risk estimates for non-
cancer effects to the benchmarks (Table 4-14 through Table 4-17) and other considerations.
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• Exposure to HBCD for ONUs is expected to be similar to that of workers; risk estimates for
inhalation exposure to ONUs were not quantified (Table 4-13).
• Workers installing XPS/EPS foam insulation and for ONUs working in residential, public, and
commercial buildings, and other structures are unlikely to wear respiratory protection (Table
4-27).
• For workers installing XPS/EPS foam insulation and for ONUs working in residential, public,
and commercial buildings, and other structures, when assuming no use of respirators, the risk
Page 480 of 723
-------
estimates to workers of non-cancer effects from chronic inhalation exposures at the high-end
support an unreasonable risk determination (Table 4-27).
For workers installing XPS/EPS foam insulation and for ONUs working in residential, public
and commercial buildings, and other structures, when assuming no use of respirators, the risk
estimates to workers of non-cancer effects from chronic inhalation exposures at the central
tendency do not support an unreasonable risk determination (Table 4-27).
For installation workers and ONUs, when assuming no use of respirators, the risk estimates to
workers of non-cancer effects from acute inhalation exposures at the high-end and central
tendency do not support an unreasonable risk determination (Table 4-27).
For workers installing XPS/EPS foam insulation and for ONUs working in residential, public,
and commercial buildings, and other structures, exposure is not expected when wearing
impervious gloves (Table 4-27).
For consumers, the risk estimates of non-cancer effects for acute and chronic exposures to dust
and indoor air are above the MOE and do not support an unreasonable risk determination (
Page 481 of 723
-------
Table 4-28).
For workers using solder paste, exposure is not expected when wearing impervious gloves (Table
4-27).
Consumers are not expected to be exposed to HBCD from use of solder paste (Section 2.2.13).
For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and high sub-scenario exposure
levels do not support an unreasonable risk determination (
Page 482 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is unreasonable risk of injury to health (for workers
and ONUs) from use of building/construction materials and solder paste.
5.2,1,8 Commercial/Consumer Use - Other - Automobile Replacement Parts and
Plastic and Other Articles
Section 6(b)(4)(A) unreasonable risk determination for commercial/consumer use of automobile
replacement parts and use of plastic and other articles: Does not present an unreasonable risk of injury to
health of (workers, ONUs, general population, consumers).
For consumers, EPA found that there was no unreasonable risk of non-cancer effects from acute
(offspring loss) or chronic (thyroid effects) exposure to dust and indoor air from installation of
automobile replacement parts. For consumers (1-2 year olds), EPA found that there was no unreasonable
risk of non-cancer effects from acute (offspring loss) or chronic (thyroid effects) exposure from plastic
and other articles.
EPA's determination that the commercial/consumer use in automobile replacement parts and plastic and
other articles does not present an unreasonable risk is based on the comparison of the risk estimates for
non-cancer effects to the benchmarks (Table 4-14 through Table 4-17) and other considerations.
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• For consumers, the risk estimates of non-cancer effects from acute and chronic exposure to dust
and indoor air from installation of automobile replacement parts do not support an unreasonable
risk determination (
Page 483 of 723
-------
Table 4-28).
Workers and ONUs are not expected to be exposed to HBCD from commercial use (installation)
of automobile replacement parts (Section 2.2.8).
Use of plastic and other articles is a consumer scenario and workers and ONUs are not expected
to be exposed to HBCD from the condition of use (Section 2.4.4).
For consumers (1 to 2-year olds), the risk estimates of non-cancer effects from acute and chronic
exposure from plastic and other articles and do not support an unreasonable risk determination (
Page 484 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to health
(consumers) from automobile replacement parts and mouthing of plastic and other articles.
5.2.1.9 Commercial/Consumer Use - Other - Formulated Products and Articles
Section 6(b)(4)(A) unreasonable risk determination for commercial/consumer use of formulated
products and articles: Does not present an unreasonable risk of injury to health of consumers.
For workers and ONUs, when assuming the use of PPE, EPA found that there was no unreasonable risk
of non-cancer effects from chronic (thyroid effects) exposure to dust and indoor air from formulated
products and articles. For the general population and consumers, EPA found that there was no
unreasonable risk of non-cancer effects from acute (offspring loss) or chronic (thyroid effects) exposure
to dust and indoor air from formulated products and articles.
EPA's determination that the commercial/consumer use of formulated products and articles does not
present an unreasonable risk is based on risks of exposure to background levels. As explained in Section
5.1, EPA considered the health effects of HBCD, the exposures for the condition of use, and the
uncertainties in the analysis, including uncertainties related to the exposures for ONUs.
The key factors in the determination for this COU are:
• For workers and ONUs, the risk estimates of non-cancer effects from exposure to formulated
products and articles do not support an unreasonable risk determination (Table 4-27).
• For the general population and consumers, the risk estimates of non-cancer effects from chronic
exposure to formulated products and articles do not support an unreasonable risk determination (
Page 485 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to health from
exposure to formulated products and articles.
5.2.1,10 Disposal - Other Land Disposal (e.g. construction and demolition waste) -
Demolition and Disposal of XPS/EPS Foam Insulation Products in
Residential, Public and Commercial Buildings and Other Structures
Section 6(b)(4)(A) unreasonable risk determination for disposal of building/construction materials:
Presents an unreasonable risk of injury to health (workers and ONUs); does not present an
unreasonable risk of injury to health to the highly exposed general population.
For workers and ONUs, EPA found that there was an unreasonable risk of non-cancer effects
from chronic (thyroid effects) inhalation at the high-end, without assuming use of PPE. For the
highly exposed general population, EPA found that there was no unreasonable risk of non-cancer effects
from acute (offspring loss) or chronic (thyroid effects) inhalation or fish ingestion at the highest or
moderate sub-scenario exposure levels.
EPA's determination that demolition and disposal of XPS/EPS foam insulation products presents an
unreasonable risk to health is based on the comparison of the risk estimates for non-cancer effects to the
benchmarks (Table 4-14 through Table 4-17) and other considerations.
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• Exposure to HBCD for ONUs is expected to be similar to that of workers; risk estimates for
inhalation exposure to ONUs were not quantified (Table 4-13).
• Workers exposed to HBCD when demolishing XPS/EPS foam insulation and for ONUs working
in residential, public and commercial buildings are unlikely to wear respiratory protection (Table
4-27).
• For workers exposed to HBCD when demolishing XPS/EPS foam insulation and for ONUs
working in residential, public and commercial buildings, when assuming no use of respirators,
the risk estimates of non-cancer effects to workers from chronic inhalation exposures at the high-
end support an unreasonable risk determination (Table 4-27).
• For workers exposed to HBCD when demolishing XPS/EPS foam insulation and for ONUs
working in residential, public and commercial buildings, when assuming no use of respirators,
the risk estimates of non-cancer effects from chronic inhalation exposures at the central tendency
do not support an unreasonable risk determination (Table 4-27).
• For demolition workers and ONUs, when assuming no use of respirators in residential, public
and commercial buildings, the risk estimates to workers of non-cancer effects from acute
inhalation exposures at the high-end and central tendency do not support an unreasonable risk
determination (Table 4-27).
Page 486 of 723
-------
• For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and highest sub-scenario exposure
levels do not support an unreasonable risk determination (
Page 487 of 723
-------
• Table 4-28).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is an unreasonable risk of injury to health (workers
and ONUs) from demolition of EPS/XPS foam insulation in residential, public and commercial
buildings, and other structures.
5,2.1.11 Disposal -Disposal of Formulated Products and Articles
Section 6(b)(4)(A) unreasonable risk determination for disposal of formulated products and articles:
Presents no unreasonable risk of injury to health (workers and ONUs); does not present an unreasonable
risk of injury to health to the highly exposed the general population.
For workers and ONUs, EPA found that there was no unreasonable risk of non-cancer effects from
chronic (thyroid effects) inhalation at the high-end, without assuming use of PPE. For the highly
exposed general population, EPA found that there was no unreasonable risk of non-cancer effects from
acute (offspring loss) or chronic (thyroid effects) inhalation or fish ingestion at the highest or moderate
sub-scenario exposure levels.
EPA's determination that disposal of formulated products and articles presents an unreasonable risk to
health is based on the comparison of the risk estimates for non-cancer effects to the benchmark MOE
(Table 4-14 through Table 4-17) and other considerations.
As explained in Section 5.1, EPA considered the health effects of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis, including uncertainties related to the exposures for
ONUs. The key factors in the determination for this COU are:
• Exposure to HBCD for ONUs is expected to be similar to that of workers; risk estimates for
inhalation exposure to ONUs were not quantified (Table 4-13).
• Workers exposed to HBCD when demolishing XPS/EPS foam insulation and for ONUs working
in residential, public and commercial buildings are unlikely to wear respiratory protection (Table
4-27).
• For workers exposed to HBCD when demolishing XPS/EPS foam insulation and for ONUs
working in residential, public and commercial buildings, when assuming no use of respirators,
the risk estimates of non-cancer effects to workers from chronic inhalation exposures at the high-
end support an unreasonable risk determination (Table 4-27).
• For workers exposed to HBCD when demolishing XPS/EPS foam insulation and for ONUs
working in residential, public and commercial buildings, when assuming no use of respirators,
the risk estimates of non-cancer effects from chronic inhalation exposures at the central tendency
do not support an unreasonable risk determination (Table 4-27).
• For demolition workers and ONUs, when assuming no use of respirators in residential, public
and commercial buildings, the risk estimates to workers of non-cancer effects from acute
inhalation exposures at the high-end and central tendency do not support an unreasonable risk
determination (Table 4-27).
Page 488 of 723
-------
• For the highly exposed general population, the risk estimates of non-cancer effects from acute
and chronic air inhalation and fish ingestion at the moderate and highest sub-scenario exposure
levels do not support an unreasonable risk determination (
Page 489 of 723
-------
• Table 4-28;).
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is an unreasonable risk of injury to health (workers
and ONUs) from disposal of formulated products and articles.
5,2.2 Environment
The unreasonable risk determinations in this section are based on the risk of adverse effects for aquatic
and terrestrial organisms. Risk estimates are presented at Table 4-26 for both the 10th (high-end) and 50th
percentile (central tendency) of estimated HBCD concentrations in surface water, sediment, and soil. For
aquatic organisms, the hazard endpoint identified for acute exposure is the delay of zebrafish embryo
hatching. The hazard endpoint identified for algae is reduction of growth. For chronic exposures, the
endpoints identified are growth effects for water flea surviving young and reduced reproduction of
California blackworms (Section 4.1.1.2). EPA also evaluated risks to terrestrial species from chronic
exposure to HBCD in soil for earthworms.
In addition to evaluating risk of intended, known, and reasonably foreseen uses of HBCD, EPA also
derived risk estimates based on monitoring data of HBCD in the environment that reflect releases of
HBCD from those uses and historical releases from discontinued uses that are not intended, known, or
reasonably foreseen to occur. RQs are equal to or above 1 (denoting risk) for all three hazard thresholds
for pelagic organisms (algae, acute fish and chronic invertebrate COCs), and the one hazard threshold
for benthic organisms (chronic blackworm invertebrate COC) based on measured monitoring surface
water and sediment concentrations near industrial facilities, respectively. On the other hand, RQs were
less than one for all aquatic organisms based on environmental monitoring data attained near general
population sites. RQs were also less than one based on the hazard threshold for earthworms near
industrial facilities and general population sites.
In regard to water releases, it is unlikely that three exposure scenarios (Installation of Automobile
Replacement Parts, Formulation of Flux/Solder Pastes and Recycling of Electronics Waste Containing
HIPS) will result in risk for aquatic organisms (pelagic and benthic) because EPA does not expect these
scenarios to result in the release of HBCD into surface water or sediment. Similarly, in regard to air
releases, it is unlikely that one exposure scenario (Installation of Automobile Replacement Parts) will
result in risk for soil organisms because EPA does not expect these scenarios to result in the presence of
HBCD in soil due to air deposition. However, although these exposure scenarios are not expected to
have water and/or air releases of HBCD, it is possible that for a specific COU corresponding to these
exposure scenarios, that there are other exposure scenarios characterizing a COU may have water or air
releases of HBCD. Despite the unlikelihood of environmental risk due to either having media-specific
releases that are less than the hazard value (RQ < 1) or the unlikely release of HBCD into specific
medias, since modeled HBCD exposures were not aggregated with measured background concentrations
of HBCD, current exposure scenario-related RQs may underestimate exposure. For the risk
determination EPA assumes background levels of HBCD add an indeterminate level of risk to each
COU but it is not aggregated quantitatively with the modeled HBCD media-specific concentrations.
5.2,2.1 Manufacturing - Import - (Import)
Section 6(b)(4)(A) unreasonable risk determination for import of HBCD: Presents an unreasonable
risk of injury to the environment (aquatic organisms); does not present an unreasonable risk to
terrestrial organisms.
Page 490 of 723
-------
For aquatic organisms, EPA found that there was an unreasonable risk of adverse effects from acute and
chronic exposures at the high-end and central tendency of concentrations in surface water and sediment.
There is also unreasonable risk of adverse effects to algae at the high-end concentration in surface water.
EPA's determination that the import of HBCD presents an unreasonable risk is based on the comparison
of the risk estimates for adverse effects to the benchmarks (Table 4-26) and other considerations. As
explained in Section 5.1, EPA considered the hazard of HBCD, the exposures for the condition of use,
and the uncertainties in the analysis. The key factors in the determination for this COU are:
• For aquatic organisms, the risk estimates of adverse effects from acute and chronic exposures in
surface water at the central tendency and high-end support an unreasonable risk determination.
The risk estimates of adverse effects for algae due to high-end surface water concentrations also
support an unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects from exposure in sediment at the
central tendency and high-end support an unreasonable risk determination.
• For terrestrial organisms, the risk estimates of adverse effects from chronic exposure in soil do
not support an unreasonable risk determination.
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is unreasonable risk of injury to the environment
from the import of HBCD.
5.2.2.2 Processing - Incorporated into Formulation, Mixture or Reaction Product
Maine retardants used in Custom Compounding of Resin (e.gcompounding
in XPS masterbatch) and in Solder Paste
Section 6(b)(4)(A) unreasonable risk determination for processing of HBCD into a formulation:
Presents an unreasonable risk of injury to the environment (aquatic organisms); does not present
an unreasonable risk to terrestrial organisms.
For aquatic organisms, EPA found that there was an unreasonable risk of adverse effects from acute and
chronic exposures at the high-end of concentrations in surface water and sediment. There is also
unreasonable risk of adverse effects to algae at the high-end concentration in surface water.
EPA's determination that the processing of HBCD into formulation presents an unreasonable risk is
based on the comparison of the risk estimates for adverse effects to the benchmarks (Table 4-26) and
other considerations. As explained in Section 5.1, EPA considered the hazard of HBCD, the exposures
for the condition of use, and the uncertainties in the analysis. The key factors in the determination for
this COU are:
• For aquatic organisms, the risk estimates of adverse effects from acute and chronic exposures in
surface water at the high-end support an unreasonable risk determination. The risk estimates of
adverse effects for algae due to high-end surface water concentrations also support an
unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects from exposure in sediment at the
high-end support an unreasonable risk determination.
Page 491 of 723
-------
• For terrestrial organisms, the risk estimates of adverse effects from chronic exposure in soil do
not support an unreasonable risk determination.
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is unreasonable risk of injury to the environment
from processing HBCD into formulation.
5.2.2.3 Processing - Incorporation into an Article - Flame Retardants used in
Plastics Product Manufacturing (manufacture of XPS and EPS foam;
manufacture of structural insulation panels (SIPS) and automobile
replacement parts from XPS and EPS foam)
Section 6(b)(4)(A) unreasonable risk determination for processing of HBCD into an article: Presents an
unreasonable risk of injury to the environment (aquatic organisms); does not present an unreasonable
risk to terrestrial organisms.
For aquatic organisms, EPA found that there was an unreasonable risk of adverse effects from acute and
chronic exposures at the central tendency and high-end of concentrations in surface water and sediment.
There is also unreasonable risk of adverse effects to algae at the central tendency and high-end
concentrations in surface water.
EPA's determination that the processing of HBCD into articles presents an unreasonable risk is based on
the comparison of the risk estimates for adverse effects to the benchmarks (Table 4-26) and other
considerations. As explained in Section 5.1, EPA considered the hazard of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis. The key factors in the determination for this COU
are:
• For aquatic organisms, the risk estimates of adverse effects from acute and chronic exposures in
surface water at the central tendency and high-end support an unreasonable risk determination.
There is also unreasonable risk of adverse effects to algae at the central tendency and high-end
concentration in surface water.
• For aquatic organisms, the risk estimates of adverse effects from exposure in sediment at the
central tendency and high-end support an unreasonable risk determination.
• For terrestrial organisms, the risk estimates of adverse effects from chronic exposure in soil do
not support an unreasonable risk determination.
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is unreasonable risk of injury to the environment
from processing HBCD into an article.
5.2.2.4 Processing - Recycling - Recycling of XPS and EPS Foam, Resin, Panels
Containing HBCD
Section 6(b)(4)(A) unreasonable risk determination for recycling of XPS and EPS form, resin, and
panels containing HBCD: Presents an unreasonable risk of injury to the environment (aquatic
organisms); does not present an unreasonable risk to terrestrial organisms.
For aquatic organisms, EPA found that there was an unreasonable risk of adverse effects from acute and
chronic exposures at the central tendency and high-end of concentrations in surface water and sediment.
Page 492 of 723
-------
There is also unreasonable risk of adverse effects to algae at the central tendency and high-end
concentrations in surface water.
EPA's determination that the recycling of XPS and EPS foam, resin, panels containing HBCD presents
an unreasonable risk is based on the comparison of the risk estimates for adverse effects to the
benchmarks (Table 4-26) and other considerations. As explained in Section 5.1, EPA considered the
hazard of HBCD, the exposures for the condition of use, and the uncertainties in the analysis. The key
factors in the determination for this COU are:
• For aquatic organisms, the risk estimates of adverse effects from acute and chronic exposures in
surface water at the central tendency and high-end support an unreasonable risk determination.
The risk estimates of adverse effects for algae due to central tendency and high-end surface
water concentrations also support an unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects from chronic exposure in sediment at
the central tendency and high-end support an unreasonable risk determination.
• For terrestrial organisms, the risk estimates of adverse effects from chronic exposure in soil do
not support an unreasonable risk determination.
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is unreasonable risk of injury to the environment
from recycling of XPS and EPS foam, resin, panels containing HBCD.
5.2.2.5 Processing - Recycling - Recycling of Electronics Waste Containing HIPS
that Contain HBCD
Section 6(b)(4)(A) unreasonable risk determination for recycling of electronics waste containing HIPS
that contain HBCD: Does not present an unreasonable risk of injury to the environment (aquatic and
terrestrial organisms).
For aquatic and terrestrial organisms, EPA found that there was no unreasonable risk of adverse effects
from exposures.
EPA's determination that recycling of electronics waste containing HIPS that contain HBCD does not
present an unreasonable risk is based on EPA's expectation that HBCD is not released from this
exposure scenario into surface water; therefore it is unlikely that there will be risk to aquatic organisms
(both pelagic and benthic). It is unlikely that air releases of HBCD from the recycling of electronics
waste containing HIPS that contain HBCD will result in risk to soil organisms.
In summary, EPA determined that there is no unreasonable risk of injury to the environment from the
recycling of electronics waste containing HIPS that contain HBCD.
5.2.2.6 Distribution in Commerce - Distribution - Distribution
Section 6(b)(4)(A) unreasonable risk determination for distribution in commerce of HBCD: Does not
present an unreasonable risk of injury to the environment (aquatic and terrestrial organisms).
For the purposes of the risk determination, distribution in commerce of HBCD is the transportation
associated with the moving of HBCD in commerce. Activities related to distribution (e.g., loading,
unloading) are considered throughout the life cycle, rather than using a single distribution scenario. EPA
Page 493 of 723
-------
assumes transportation of HBCD is conducted taking similar measures as the transportation of
hazardous materials.
5,2.2,7 Commercial/Consumer Use - Building/Construction Materials - Plastic
Articles (hard) Construction and Building Materials Covering Large Surface
Areas (e.g., EPS/XPS foam insulation in residential, public and commercial
buildings, and other structures) and Solder Paste
Section 6(b)(4)(A) unreasonable risk determination for use of construction and building materials and
solder paste containing HBCD: Presents an unreasonable risk of injury to the environment (aquatic
organisms); does not present an unreasonable risk to terrestrial organisms.
For aquatic organisms, EPA found that there was an unreasonable risk of adverse effects from acute
exposures at the central tendency and high-end of concentrations in surface water and chronic exposures
at the high-end of concentrations in sediment. There is also unreasonable risk of adverse effects to algae
at the central tendency and high-end concentrations in surface water.
EPA's determination that the use of construction and building materials and solder paste containing
HBCD presents an unreasonable risk is based on the comparison of the risk estimates for adverse effects
to the benchmarks (Table 4-26) and other considerations. As explained in Section 5.1, EPA considered
the hazard of HBCD, the exposures for the condition of use, and the uncertainties in the analysis. The
key factors in the determination for this COU are:
• For aquatic organisms, the risk estimates of adverse effects from acute exposures in surface
water at central tendency and high-end support an unreasonable risk determination. The risk
estimates of adverse effects for algae due to central tendency and high-end surface water
concentrations also support an unreasonable risk determination. The risk estimates of adverse
effects from chronic exposures in sediment at high-end support an unreasonable risk
determination.
• For aquatic organisms, the risk estimates of adverse effects from chronic exposure in surface
water do not support an unreasonable risk determination.
• For terrestrial organisms, the risk estimates of adverse effects from chronic exposure in soil do
not support an unreasonable risk determination.
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is unreasonable risk of injury to the environment
from commercial/consumer use of construction and building materials and solder paste.
5,2.2,8 Disposal - Other Land Disposal (e.g. construction and demolition waste)
Demolition and Disposal of XPS/EPS Foam Insulation Products in
Residential, Public and Commercial Buildings and Other Structures
Section 6(b)(4)(A) unreasonable risk determination for disposal of HBCD: Presents an unreasonable
risk of injury to the environment (aquatic organisms); does not present an unreasonable risk to
terrestrial organisms.
Page 494 of 723
-------
For aquatic organisms, EPA found that there was an unreasonable risk of adverse effects from acute and
chronic exposures at the central tendency and high-end of concentrations in surface water. There is also
unreasonable risk of adverse effects to algae at the central tendency and high-end concentrations in
surface water.
EPA's determination that the disposal of HBCD presents an unreasonable risk is based on the
comparison of the risk estimates for adverse effects to the benchmarks (Table 4-26) and other
considerations. As explained in Section 5.1, EPA considered the hazard of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis. The key factors in the determination for this COU
are:
• For aquatic organisms, the risk estimates of adverse effects from acute exposures in surface
water at the central tendency and high-end support an unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects from chronic exposure in surface
water at the high-end support an unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects for algae in surface water at the
central tendency and high-end support an unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects from exposure in sediment do not
support an unreasonable risk determination.
• For terrestrial organisms, the risk estimates of adverse effects from chronic exposure in soil do
not support an unreasonable risk determination.
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is unreasonable risk of injury to the environment
from disposal.
5,2,2,9 Disposal - Land Disposal of Formulated Products and Articles
Section 6(b)(4)(A) unreasonable risk determination for disposal of HBCD: Does not present an
unreasonable risk of injury to the environment (aquatic organisms); does not present an unreasonable
risk to terrestrial organisms.
For aquatic organisms, EPA found that there was no unreasonable risk of adverse effects from acute and
chronic exposures at the central tendency and high-end of concentrations in surface water. There is also
no unreasonable risk of adverse effects to algae at the central tendency and high-end concentrations in
surface water.
EPA's determination that the disposal of HBCD does not present an unreasonable risk is based on the
comparison of the risk estimates for adverse effects to the benchmarks (Table 4-26) and other
considerations. As explained in Section 5.1, EPA considered the hazard of HBCD, the exposures for the
condition of use, and the uncertainties in the analysis. The key factors in the determination for this COU
are:
• For aquatic organisms, the risk estimates of adverse effects from acute exposures in surface
water at the central tendency and high-end do not support an unreasonable risk determination.
Page 495 of 723
-------
• For aquatic organisms, the risk estimates of adverse effects from chronic exposure in surface
water at the high-end do not support an unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects for algae in surface water at the
central tendency and high-end do not support an unreasonable risk determination.
• For aquatic organisms, the risk estimates of adverse effects from exposure in sediment do not
support an unreasonable risk determination.
• For terrestrial organisms, the risk estimates of adverse effects from chronic exposure in soil do
not support an unreasonable risk determination.
In summary, the risk estimates, the health effects of HBCD, the exposures, and consideration of
uncertainties support EPA's determination that there is no unreasonable risk of injury to the environment
from disposal.
5.3 Changes to the Unreasonable Risk Determination from Draft Risk Evaluation to Final
Risk Evaluation
In response to peer review and public comments on the draft Risk Evaluation , EPA conducted
additional assessments, including estimation of environmental risk at the 10th percentile concentrations
of concern in surface water, sediment, and soil. For the human health assessment, EPA assumed that
workers and ONUs are unlikely to use respirator protection for installation or demolition of XPS/EPS
foam insulation and therefore did not apply in the risk determination the assumption that PPE is used for
these two uses. EPA also added assessments and unreasonable risk determinations for select PESS
groups not associated with releases from a particular COU. Ultimately EPA made determinations of
unreasonable risk for six of the 10 conditions of use.
5.4 Unreasonable Risk Determination Conclusion
5.4.1 No Unreasonable Risk Determinations
TSCA section 6(b)(4) requires EPA to conduct Risk Evaluations to determine whether chemical
substances present unreasonable risk under their conditions of use. In conducting Risk Evaluations,
"EPA will determine whether the chemical substance presents an unreasonable risk of injury to health or
the environment under each condition of use within the scope of the Risk Evaluation ..." 40 CFR
702.47. Under EPA's implementing regulations, "[a] determination by EPA that the chemical substance,
under one or more of the conditions of use within the scope of the risk evaluation, does not present an
unreasonable risk of injury to health or the environment will be issued by order and considered to be a
final Agency action, effective on the date of issuance of the order." 40 CFR 702.49(d).
EPA has determined that the following conditions of use of HBCD do not present an unreasonable risk
of injury to health or the environment:
• Processing; recycling of electronics waste containing HIPS that contains HBCD (Section 5.2.1.1,
Section 5.2.2)
• Distribution (Section 5.2.1.6, Section 493)
• Consumer/Commercial Use of replacement automobile parts (Section 5.2.1.8)
• Consumer/Commercial Use of plastics and other articles (Section 5.2.1.8)
Page 496 of 723
-------
• Consumer/Commercial Use of formulated products and articles (Section 5.2.1.9)
• Disposal of formulated products and articles (Section 5.2.1.11, Section 5.2.2.9)
This subsection of the final Risk Evaluation therefore constitutes the order required under TSCA section
6(i)(l), and the "no unreasonable risk" determinations in this subsection are considered to be final
agency action effective on the date of issuance of this order. All assumptions that went into reaching the
determinations of no unreasonable risk for these conditions of use, including any considerations
excluded for these conditions of use, are incorporated into this order.
The support for each determination of "no unreasonable risk" is set forth in Section 5.2 of the final Risk
Evaluation, "Detailed Unreasonable Risk Determinations by Condition of Use." This subsection also
constitutes the statement of basis and purpose required by TSCA section 26(f).
5,4.2 Unreasonable Risk Determinations
EPA has determined that the following conditions of use of HBCD present an unreasonable risk of
injury to the environment and two conditions of use (Commercial/consumer use of construction/building
materials and solder paste, and Disposal) also present an unreasonable risk of injury to health:
• Manufacturing (Import) (Section 5.2.1.1, Section 5.2.2, Section 4, Section 3, and 2.)
• Processing of HBCD: incorporation into a formulation, mixture, or reaction products (Section
5.2.1.1, Section 5.2.2, Section 4, Section 3, and 2.)
• Processing of HBCD: incorporation into an article (Section 5.2.1.1, Section 5.2.2, Section 4,
Section 3, and 2.)
• Recycling of XPS/EPS foam, resin, panels containing HBCD (Section 5.2.1.1, Section 5.2.2,
Section 4, Section 3, and 2.)
• Commercial/consumer use of HBCD in construction/building materials and solder paste (Section
5.2.1.1, Section 5.2.2, Section 4, Section 3, and 2.)
• Disposal of HBCD in construction and demolition waste (Section 5.2.1.1, Section 5.2.2, Section
4, Section 3, and 2.)
EPA will initiate TSCA section 6(a) risk management actions on these conditions of use as required
under TSCA section 6(c)(1). Pursuant to TSCA section 6(i)(2), the "unreasonable risk" determinations
for these conditions of use are not considered final agency action.
Page 497 of 723
-------
REFERENCES
\ 1 [{ k *xiin Technologies. (2007). Material safety data sheet (MSDS): foam-control EPS and foam-
control EPS with perform guard. Westminster, CO: ACH Foam Technologies.
http://www.octaform.com/wp-content/uploads/EPSMSDS.pdf.
Abb. M; Sta nz. W. (2011). Analysis of brominated flame retardants in house dust.
Chemosphere 85: 1657-1663. http://dx.doLo /i.chemosphere.2011.06.022.
Abbot. W. (2001). Summary of workplace and exposure monitoring data for Hexabromocyclododecane.
Abbot, W.
Abdallah. MA; Harrad. S. (2009). Personal exposure to HBCDs and its degradation products via
ingestion of indoor dust. Environ Int 35: 870-876.
http://dx.doi.on /int.2009.03.002.
Abdallah. MA; Harrad. S. (2011). Tetrabromobisphenol-A, hexabromocyclododecane and its
degradation products in UK human milk: relationship to external exposure. Environ Int 37: 443-
448. http://dx.doLo /i.envint.2010.11.008.
Abdallah. MA; Harrad. S; Covaci. A. (2008). Hexabromocyclododecanes and tetrabromobisphenol-A in
indoor air and dust in Birmingham, U.K: implications for human exposure. Environ Sci Technol
42: 6855-6861. http://dx.doi.org/10.1021/esS a,
Abdallah. MA; Pawar. G; Harrad. S. (2015). Evaluation of 3D-human skin equivalents for assessment of
human dermal absorption of some brominated flame retardants. Environ Int 84: 64-70.
http://dx.doLorg/10.1016/j.envint.-°L 0 L
Abdallah. MA; Sharkey. M; Berresheim. H; Harrad. S. (2018). Hexabromocyclododecane in polystyrene
packaging: A downside of recycling? Chemosphere 199: 612-616.
http://dx.doi.on .chemosphere.2018.02.084.
Abdallah. MA; Uchea. C; Chipman. IK; Harrad. S. (2014). Enantioselective biotransformation of
hexabromocyclododecane by in vitro rat and trout hepatic sub-cellular fractions. Environ Sci
Technol 48: 2732-2740. hji|;> u\.doi.org/10,102 i/es404644s.
Abdallah. M mrad. S. (2010). Modification and Calibration of a Passive Air Sampler for
Monitoring Vapor and Particulate Phase Brominated Flame Retardants in Indoor Air:
Application to Car Interiors. Environ Sci Technol 44: 3059-3065.
http://dx.doi.org Kl l ^ t n
ACC. (2003a). Hexabromocyclododecane (HBCD): A Prolonged Sediment Toxicity Test with Hyalella
azteca Using Spiked Sediment with 2% Total Organic Carbon. In Wildlife International Ltd (pp.
150). (OTS: NA; 8EHQ Num: FYI-03-01472; DCN: 84040000010; TSCATS RefID: NA; CIS:
FYI-03-01472). Easton, MD.
ACC. (2003b). Hexabromocyclododecane (HBCD): A Prolonged Sediment Toxicity Test with Hyalella
azteca Using Spiked Sediment with 5% Total Organic Carbon. (OTS: NA; 8EHQ Num: FYI-03-
01472; DCN: 84040000010; TSCATS RefID: NA; CIS: FYI-03-01472).
ACC/North American Flame Retardant Alliance (2019): Public Comment EPA-HQ-OPPT-2019-0238-
0020.
Airlite Plastics Co dba Fos (2008). Material safety data sheet (MSDS): fox blocks insulating
concrete forms (ICF). Omaha, NE: Airlite Plastics Co dba Fox Blocks.
https://formingsolutionsicf.corn/wp-content/uploads/2Ql5/0 N«\-Blocks-tPM> U-nendi\ \ I
MSDS.pdf.
itar. F. (2004). Hazardous chemicals in Belgian house dust: Report on chemical content in house
dust samples collected in Belgian homes and offices. Brussels, Belgium: Greenpeace Belgium.
http://www.greenpeace.org/eii-unit/Global/eu-iinit/reports-briefin.es/2007/hazardous-chemicals-
in-belgian-2.pdf.
Page 498 of 723
-------
Albemarle. (2005). Hexabromocyclododecane IUCLIP Data Set (CAS No. 25637-99-4): Submitted to
U.S. EPA's High Production Volume (HPV) Chemical Program.
http://www.epa.gov/chemrtk/pubs/siimmaries/cvclodod/cl3459iT.pdf
Albemarle. (2017). [Personal communication between Wesley Ware, Albemarle Corporation, and Sue
Slotnick, EPA, regarding Hexabromocyclododecane (HBCD)]. EPA-HQ-OPPT-2016-0735-0113
in regulations.gov.
Alexander. BM; Baxter. CS. (2016). Flame-retardant contamination of firefighter personal protective
clothing - A potential health risk for firefighters. J Occup Environ Hyg 13: D148-D155.
http://dx.doi.org/10.1080/15459624.; 316
All.' , \C \ in l':n Eede. N: Gooses J ilarrad. S: Neels. H: 'T Mannetie. A: Coaklev. J:
Douwes /aci. A. (2012). Occurrence of alternative flame retardants in indoor dust from
New Zealand: indoor sources and human exposure assessment. Chemosphere 88: 1276-1282.
http://dx.doi.or j. chemosphere.2012.03.100.
Allchin. CR: Morris. S. (2003). Hexabromocyclododecane (HBCD) diastereoisomers and brominated
diphenyl ether congener (BDE) residues in edible fish from the rivers Skerne and Tees, UK.
Organohalogen Compd 61: 41-44.
Allen. JG: Stapleton. HM; Vallarino. J; Mcneelv. E; Mcclean. MP; Harrad. SI; Ran )en.gler.
(2013). Exposure to flame retardant chemicals on commercial airplanes. Environ Health 12:
17. http://dx.doi.org/10.1186/1476-0'
Allgood. JM; limah. T; Mcclaskev. CM; La Guardia. MI: Hammel. SC; Zeineddine. MM; Tang. IW;
Runnerstroio \k\ »''gunseitaii < * \ (2016). Potential human exposure to halogenated flame-
retardants in elevated surface dust and floor dust in an academic environment. Environ Res 153:
55-62. http://dx.doi.org/10. "1016/i.envres .01 .*_! j_0j0
Alliance of Automobile Manufacturers. (2018). Cover letter and attachment from Automobile Alliance
11-19-18 [Personal Communication],
Almughamsi. H; Whalen. MM. (2016). Hexabromocyclododecane and tetrabromobisphenol A alter
secretion of interferon gamma (IFN-y) from human immune cells. Arch Toxicol 90: 1695-1707.
http://dx.doi.org/10.1007/s00204-015-1586-6.
ni. NA; Shim. WJ; Han. GM; Jang. M; Hong. SH. (2015). Enrichment of
hexabromocyclododecanes in coastal sediments near aquaculture areas and a wastewater
treatment plant in a semi-enclosed bay in South Korea. Sci Total Environ 505: 290-298.
An. J; Guo. P; Strang. > hnuv Y; Zhai^ \ ^ _ (2016). The "adaptive
responses" of low concentrations of HBCP in L02 cells and the underlying molecular
mechanisms. Chemosphere 145: 68-76. http://dx.doi.ou jjj| ^i.chemosphere j s
An. J; Zou. W; Chen. C; Zhon\ I \ uk Q Want H i (2013). The cytological effects of HBCPs on
human hepatocyte L02 and the potential molecular mechanism. J Environ Sci Health A Tox
Hazard Subst Environ Eng 48: 1333-1342. http://dx.doi.org/10.1080/10934529.20i3.781875.
Aniagu. SO; Williams. TP; Allen V Katsiadaki j_i ttipman. IK. (2008). Global genomic methylation
levels in the liver and gonads of the three-spine stickleback (Gasterosteus aculeatus) after
exposure to hexabromocyclododecane and 17-beta oestradiol. Environ Int 34: 310-317.
http://dx.doi.or i.envint.2007.03.009.
Anint \i* il1 1 1! * oonetilleke. A; Muellei if- .\\i »ko. GA. (2017). Pistribution of PBPEs,
HBCPs and PCBs in the Brisbane River estuary sediment. Mar Pollut Bull 120: 165-173.
http://dx.doi.or i.marpolbul.2017.05.002.
Anisuzzaman. S; Whalen. MM. (2016). Tetrabromobisphenol A and hexabromocyclododecane alter
secretion of IL-ip from human immune cells. J Immunotoxicol 13: 403-416.
.org/10.3109/1547691X.2015.1111960.
Anon. (2015). ICL-IP closes HBCP flame retardant production line. Additives for Polymers 2015: 7-8.
Page 499 of 723
-------
Anselmo. HMR; Koerting. < _* Vvito. S; van den Berg. JHJ; Dubbeldam. M; Kwadiik. C; Murk \J
(2011). Early life developmental effects of marine persistent organic pollutants on the sea urchin
Psammechinus miliaris. Ecotoxicol Environ Saf 74: 2182-2192.
http://dx.doi.oii 10 101» i.ecoenv.201 I 0 ' 037.
Antignac. IP; Main. KM; Virtanen. HE; Boquien. €¥; Marcharnt P. \ ciiisseau. A; Guiffart> 1 I • i.marenvres.201 ¦> 0- * 011
A.ylv> Hays. SM. (2011). Biomonitoring-based risk assessment for hexabromocyclododecane
(HBCD) [Review], Int J Hyg Environ Health 214: 179-187.
http://dx.doi.org K1 i.iiheh.201 1 0..002.
Bakii Connor. I A; Rowland. SI; Hendriks. AJ; Thompson. RC. (2016). Relative importance of
microplastics as a pathway for the transfer of hydrophobic organic chemicals to marine life.
Environ Pollut 219: 56-65.
Barghi. M; Shin. ES; Son. MH; Choi. SD; Pvo. H; Chang. YS. (2016). Hexabromocyclododecane
(HBCD) in the Korean food basket and estimation of dietary exposure. Environ Pollut 213: 268-
277. http://dx.doi.o>v 1 ^1 * /i.envpol.201-> 0J.026.
Barb (2007). The origins of the developmental origins theory. J Intern Med 261: 412-417.
http://dx.doi.on 5.2007.01809.x
Batxegard. (2003). Abstract P1.1 -15 [Abstract], In Annual Conference of the International Society of
Exposure Analysis.
BASF. (1990). Report on the study of the acute oral toxicity of hexabromocyclododecane in the mouse
with cover letter dated 031290 [TSCA Submission], (86900000383). Wyandotte, MI: BASF
Corporation.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuery=OTS0522946
BASF. (2000). Cytogenetic study in vivo with of hexabromocyclododecane in the mouse micronucleus
test after two intraperitoneal administrations. (Project No. 26MO100/004018). Ludwigshafen,
Germany: BASF Aktiengesellschaft.
(2015). Safety data sheet (SDS): styropek BF-222 M. Florham Park, NJ: BASF Corporation.
http://www2.basf.us/DocToolWeb/DisplavDoc?oname=30246572&appid=Plastics MSDS Sear
ch&rend=pdf.
(2017). Personal communication between Jodi Visco, BASF, and Sue Slotnick, EPA, regarding
Hexabromocyclododecane (HBCD).
Page 500 of 723
-------
Bemat. J. (2015). Thyroid hormones in brain development and function. In LJ De Groot; G Chrousos; K
Dungan; KR Feingold; A Grossman; JM Hershman; C Koch; M Korbonits; R McLachlan; M
New; J Purnell; RRebar; F Singer; A Vinik (Eds.). South Dartmouth, MA: MDText.com, Inc.
Bernhard. A; Berntssen. MH; Lundebve. AK; Reryneberg Alvheim cher Myrmel. L; Fisere. E;
Torstensen. BE; Kristiansen. K; Madsen. 1 , Uuittelid. T; Rasinger. ID. (2016). Marine fatty
acids aggravate hepatotoxicity of a-HBCD in juvenile female BALB/c mice. Food Chem Toxicol
97: 411-423. http://dx.doi.org/lO 101 iJOJ ;J0 002.
Besseling. E; Wegn cema. EM; van den Heuvel-Greve. MI; Koelmans. AA. (2013). Effects of
microplastic on fitness and PCB bioaccumulation by the lugworm Arenicola marina (L.).
Environ Sci Technol 47: 593-600. http://dx.doij l/es302763x
Bevingtc mo. Z; Hong. T; Hubbard. H; Wong. E; Sleasman. K; Hetfield. C. (2017). A Modeling
Approach for Quantifying Exposures from Emissions of Spray Polyurethane Foam Insulation in
Indoor Environments. In J Sebroski; M Mason (Eds.), (pp. 199-227). West Conshohocken, PA:
ASTM International. https://doi.org/1 7STP15892.0150045.
Biesemeier. J A. (1996). Followup response letter from Great Lakes Chemical Corporation to Keml.
Dated 19 December 1996. Biesemeier, JA
Boyles. E; Tan. H; Wu. Y; Nielsen. CK; Shen. L; Reinei ^ s Chen O (2017). Halogenated flame
retardants in bobcats from the midwestern United States. Environ Pollut 221: 191-198.
http://dx.doi.on Jt * i.envpol01 o_]j u63.
Brandsma. SH; Leonards. P; Leslie. HA; de Boer. I. (2015). Tracing organophosphorus and brominated
flame retardants and plasticizers in an estuarine food web. Sci Total Environ 505: 22-31.
http://dx.doi.org/10 101o/j.seitotenv2011 08.072.
Brandsma. SH; Van der V< sr. I; Leonards. PE. (2009). Identification of hydroxylated
metabolites of hexabromocyclododecane in wildlife and 28-days exposed Wistar rats. Environ
Sci Technol 43: 6058-6063. http://dx.doi.org/10. i 021 /es900879k.
Burger. J. (2002). Daily consumption of wild fish and game: Exposures of high-end recreationists. Int J
Environ Health Res 12: 343-354. http://dx.doi.org/10.1080/0960312021000Q56393.
Burkhart. IE; Short. S. (1995). Health Hazard Evaluation Report HETA 91-0354-2532, South Dade
Disposal Site, Goulds, Florida. (NTIS/02988621 3). Burkhart, JE; Short, S.
Cahill. TM; Cousins. I; Macks (2003). Development and application of a generalized
physiologically based pharmacokinetic model for multiple environmental contaminants. Environ
Toxicol Chem 22: 26-34. http://dx.doi.org/10.1002/etc.5620220104.
Campine. (2017). Personal communication between Hilde Goovaertz, Campine, and Sue Slotnick, EPA,
regarding Hexabromocyclododecane (HBCD).
Canb re. MC; Logiantara. A; van Ree. R; van Rijt. LS. (2016a). Indoor pollutant
hexabromocyclododecane enhances house dust mite-induced activation of human monocyte-
derived dendritic cells. J Immunotoxicol 13: 1-7.
http://dx.doi.org/lO 1080/1547691X.1C4 * l 200224.
Canb iantara. A; Haroers. T; van Ree. R; van Rijt. LS. (2016b). Indoor Pollutant
Hexabromocyclododecane Has a Modest Immunomodulatory Effect on House Dust Mite
Induced Allergic Asthma in Mice. Environ Sci Technol 50: 405-411.
http://dx.doi.on tcs.est.5b05348.
Canton. Ki' i'-ninenbury \ \ Uoogenboom. RL; Piersma. AH; van dot \ »'n. > 1', v an den Berg. M;
Heneweer. M. (2008). Subacute effects of hexabromocyclododecane (HBCD) on hepatic gene
expression profiles in rats. Toxicol Appl Pharmacol 231: 267-272.
http://dx.doi.org/10 101»'/j.taap.2008.04 01 ?•.
C, _ \ _I ^ \ V I- tun. K; Wang. C; Baninla. V (2018). An overview of
hexabromocyclododecane (HBCDs) in environmental media with focus on their potential risk
Page 501 of 723
-------
and management in China [Review], Environ Pollut 236: 283-295.
http://dx.doi.on, 10 10[-* i.envpol.20l^ h1
Carignan. CC; Abdallah. MA; Wu. N; Heiger-Bemavs. W; Mcclean. MP; Harrad. S: Webs
(2012). Predictors of tetrabromobisphenol-A (TBBP-A) and hexabromocyclododecanes (HBCD)
in milk from Boston mothers. Environ Sci Technol 46: 12146-12153.
http://dx.doi.on ;s302638d.
C u. J; Tesoro. AG. (2016). Transport and fate of microplastic particles in wastewater
treatment. Water Res 91: 174-182.
Census Bureau. (2012). Code Lists and Crosswalks - Census 2012 Detailed Industry Code List
[Database], Retrieved from http://www.census.gov/people/io/methodology/
Census Bureau. (2015). Statistics of U.S. Businesses (SUSB).
https://www.census.gov/data/tables/2015/econ/susb/2Q15-susb-annual.html.
Census Bureau. (2016). Survey of Income and Program Participation - SIPP Introduction and History
[Database], Retrieved from http://www.census.gov/programs-surveys/sipp/about/sipp-
introduction-history. html
Chen. D; La Guardia. Ml; Luellen. PR; Harvey. E; Mainor. TM; Hale. RC. (2011). Po temporal and
geographical patterns of HBCP and PBPE flame retardants in U.S. fish reflect evolving
industrial usage? Environ Sci Technol 45: 8254-8261. http://dx.doi.or> 10 Iv'J t es2P I H hv.
Choksi. NY: Jahnke. GD; St Hilaire. C; Shelby. M. (2003). Role of Thyroid Hormones in Human and
Laboratory Animal Reproductive Health [Review], Birth Pefects Res B Pev Reprod Toxicol 68:
479-491. http://dx.doi.org, 10 J 002/bi'q^
Christen. V; Crettaz. P; Oberli-Schrammli. A; Fent. K. (2010). Some flame retardants and the
antimicrobials triclosan and triclocarban enhance the androgenic activity in vitro. Chemosphere
81: 1245-1252. http://dx.doi.Org/10.1016/i.chemosphere.2010.09.031.
Coetho. SD; Sousa. \{\ tsobe. T; Kim. JW; Kunisu^ i K ogueira ^tiuibe. S. (2016). Brominated.
chlorinated and phosphate organic contaminants in house dust from Portugal. Sci Total Environ
569-570: 442-449. http://dx.doi.oi^ K> 10 KVi.scitotenv.20.1 ^ l «
Conley. K: Clum. A: Deep .ne. H: Beckingham. B. (2019). Wastewater treatment plants as a
source of microplastics to an urban estuary: Removal efficiencies and loading per capita over one
year. Water Research: X 3: 100030.
Covad. A: Roosens j.MULAL Waegeneers. N: Van Overmeire. I: Neels. U, Sevens. L. (2009).
Brominated flame retardants in Belgian home-produced eggs: levels and contamination sources.
Sci Total Environ 407: 4387-4396.
CP SC. (2001). CPSC staff exposure and risk assessment of flame retardant chemicals in residential
upholstered furniture. Bethesda, MD. http://dx.doi.org/10 L'« !40'RG.2.1.3291.6646.
Critfc. (1994). A fish consumption survey of the Umatilla, Nez Perce, Yakama, and Warm Springs
Tribes of the Columbia River Basin. (Technical Report 93-4). Portland, OR.
http://www.deq.idaho.gov/media/895853-fish-consumption-survey-1994.pdf.
Crump. D; Chiu. S: Egloff. C; Kennedy. SW. (2008). Effects of hexabromocyclododecane and
polybrominated diphenyl ethers on mRNA expression in chicken (Gallus domesticus)
hepatocytes. Toxicol Sci 106: 479-487. http://dx.doi.org/lQ. 1093/toxsci/kfn 196.
Crump. D; Egloff. C; Chiu. S: Letcher. RJ; Chu. S: Kennedy. SW. (2010). Pipping success, isomer-
specific accumulation, and hepatic mRNA expression in chicken embryos exposed to HBCD.
Toxicol Sci 115: 492-500. http://dx.doi.org/10.1093/toxsci/kfq068.
D'Hollander. W: Roose^ * v. t»\ aci. t ^rnelis. C: Reynders. H: Campenhout. KV: Voopi * <\
Bervoets. L. (2010). Brominated flame retardants and perfluorinated compounds in indoor dust
from homes and offices in Flanders, Belgium. Chemosphere 81: 478-487.
ix.doi.on .chemosphere.2010.07.043.
Page 502 of 723
-------
Darnerud. P; Lignell. S; Aune. M; Isaksson. M; Cantillana. T; Red ntm. A. (2015). Time trends
of polybrominated diphenylether (PBDE) congeners in serum of Swedish mothers and
comparisons to breast milk data. Environ Res 138: 352-360.
http://dx.doi.oii .envres.2015.02.031.
Damemd. PO. (2003). Toxic effects of brominated flame retardants in man and in wildlife [Review],
Environ Int 29: 841-853. http://dx.doior; 50160-4120(03)00107-7.
Paso. AP; Rohwer. ER; Koot. :onkwo. JO. (2017). Preliminary screening of polybrominated
diphenyl ethers (PBDEs), hexabromocyclododecane (HBCDD) and tetrabromobisphenol A
(TBBPA) flame retardants in landfill leachate. Environ Monit Assess 189: 418.
http://dx.doi.ore/10.1007/sl0C: z.
Davis. A; Gift. IS; Woodall. GM; Narotskv. MG: Fourman. GL. (2009). The role of developmental
toxicity studies in acute exposure assessments: analysis of single-day vs. multiple-day exposure
regimens. Regul Toxicol Pharmacol 54: 134-142. http://dx.doi.or6 .yrtph.2009.03.006.
Davis. JW; Gonsior. SI; Marty. GT. (2003a). Evaluation of aerobic and anaerobic transformation of
hexabromocyclododecane in aquatic sediment systems. (Dow Study ID 021081). Arlington, VA:
American Chemistry Council.
Davis. JW; Gonsior. SI; M (2003b). Evaluation of aerobic and anaerobic transformation of
hexabromocyclododecane in soil. (Dow Study ID 021082). Arlington, VA: American Chemistry
Council.
Davis. JW; Gonsior. S: Marty. G: Ariano. J. (2005). The transformation of hexabromocyclododecane in
aerobic and anaerobic soils and aquatic sediments. Water Res 39: 1075-1084.
http://dx.doi.on .watres.2004.11.024.
Davis. JW; Gonsior. SI; Markhatt derich. U; Hunziker. RW; Ariano. JM. (2006).
Biodegradation and product identification of [14C]hexabromocyclododecane in wastewater
sludge and freshwater aquatic sediment. Environ Sci Technol 40: 5395-5401.
http://dx.doi.on ;s060009m.
De Bt«'i j Ulohin. C; Zegers. B; Boon. IP: Brandsma. SH; Morris. S: Kruiit. AW; Van Per Veen. I:
Van Hesselingen. JM; Haftka. JJ.H. (2002). HBCD and TBBP-A in sewage sludge, sediments
and biota, including interlaboratory study. In RIVO - the Netherlands Institute for Fisheries
Research (pp. 40). (RIVO report number CO33/02). Ymuiden and Yerseke, Netherlands: De
Boer, J; Allchin, C; Zegers, B; Boon, JP; Brandsma, SH; Morris, S; Kruijt, AW; Van Der Veen,
I; Van Hesselingen, JM; Haftka, JJH.
https://yosemite.epa. gov/oppts/epatscat8.nsf/by+Service/8298B It 18 DO I VOQ8525 6EEB004A.
5440/$Fiie/8903 0000021.pdf.
de Boer. J; Lest: jnards. PE; Bersuder. P; Morris. S; Allchin. CR. (2004). Screening and time
trend study of decabromodiphenylether and hexabromocyclododecane in birds (pp. 4). de Boer,
J; Leslie, HA; Leonards, PE; Bersuder, P; Morris, S; Allchin, CR.
http://dtsc.ca.gov/bfr2013/abstract download/2004/upload/Individual%20Papers/BFR2004%20
ract%20025%20deBoer.pdf.
de W )rklund. J A; Thuresson. K. (2012). Tri-decabrominated diphenyl ethers and
hexabromocyclododecane in indoor air and dust from Stockholm microenvironments 2: indoor
sources and human exposure. Environ Int 39: 141-147.
http://dx.doi.on iOJiii V i.envint.,^ I I 1
Deng. J; Yu. L; Liu. C; Yu. K: Shi. X; Yeuni n. PK: Wu. RS; Zhou. B. (2009).
Hexabromocyclododecane-induced developmental toxicity and apoptosis in zebrafish embryos.
Aquat Toxicol 93: 29-36. http://dx.doi.oi i.aqiiatox.2009.03.001.
Desiardins. D; Macgregor. J; Krueger. H. (2004). Final report: hexabromocyclododecane (HBCD): a 72-
hour toxicity test with the marine diatom (Skeletomema costatum). (Project Number: 439A-125).
Easton, MD: Desjardins, D; Macgregor, J; Krueger, H.
Page 503 of 723
-------
Desiardins. D; MacGregor ;er. HO. (2005). Final report. Chapter 1, Hexabromocyclododecane
(HBCD): A 72-hour toxicity test with the marine diatom (Skeletonema costatum) using a co-
solvent. Easton, MD: Wildlife International Ltd.
DiversiFoam. P. (2015). Safety data sheet: certifoam. Rockford, MN: DiversiFoam Products.
https://www. diversifoam. com/pdf/CertiF oam -SDS .pdf.
Dodds. WK; Oakes. RM. (2004). A technique for establishing reference nutrient concentrations across
watersheds affected by humans. Limnology and Oceanography: Methods 2: 333-341.
http://dx.doi.on om.2004.2.333.
Dodson. RE; Perovich. LI; Covaci. A; Van den Eede. N: lomas. AC: Dirtu. AC: Bnx iel. RA.
(2012). After the PBDE phase-out: a broad suite of flame retardants in repeat house dust samples
from California. Environ Sci Technol 46: 13056-13066. http://dx.doi.org/ _ ?303879n.
Dow Chemical. (2017). [Notes from phone call between Shawn Hunter, Dow Chemicals, and Sue
Slotnick, EPA, regarding Hexabromocyclododecane (HBCD)].
Dow Chemical Pacific. L. (2018). Material safety data sheet (MSDS): Styrofoam RTM-GV-NC-X
Extruded. Wan Chai, Hong Kong: Dow Chemical Pacific Limited.
https://webcache.googleusercontent.com/search?q=cache:ncdz8MbJuJUJ:https://www. dow.com/
en-U'S/ShowPDF.ashx%3Fid%3D090003e8808177d3+&cd=l&hl=en&ct=clnk&gl=us.
Drage. D: Mueller. IF; Birch. G; Eaglesham. G; Hearn. LK; Harrad. S. (2015). Historical trends of
PBDEs and HBCDs in sediment cores from Sydney estuary, Australia. Sci Total Environ 512-
513: 177-184. http://dx.doi.orK 10 10 L- i.scitotenv.-01> 0LO-14.
Drage. PS: Newton. S: de Wit. CA; Harrad. S. (2016). Concentrations of legacy and emerging flame
retardants in air and soil on a transect in the UK West Midlands. Chemosphere 148: 195-203.
http://dx.doi.org/10.1016/i.chemosphere..01 !0l_014.
Driffield. M: Harmer. N: Brad des. AR; Rose. M; Mortimer. D; Dicks. P. (2008).
Determination of brominated flame retardants in food by LC-MS/MS: diastereoisomer-specific
hexabromocyclododecane and tetrabromobisphenol A. Food Addit Contam Part A Chem Anal
Control Expo Risk Assess 25: 895-903. http://dx.doi.org/10.1080/026520307Q1882999.
Drottar. ICR; Krueger. HO. (1998). Hexabromocyclododecane (HBCD): a flow-through life-cycle
toxicity test with the cladoceran (Daphnia magna): Final report, with cover letter dated 5/18/1998
(pp. 80). (Wildlife International Ltd. Project Number: 439A-I08). Easton, MD: Wildlife
International Ltd.)
https://vosemite.epa. gov/oppts/epatscat8.nsf/by+Service/5A64B9( 8085256930004C50
68/$File/84980000i
Drottar. KR; Krueger. HO. (2000). Hexabromocyclododecane (HBCD): A flow-through
bioconcentration test with the rainbow trout (Oncorhynchus mykiss). Easton, MD: Wildlife
International Ltd.
Drottar. KR; Macgregc ager. HQ. (2001). Hexabromocyclododecane (HBCD): An early life-
stage toxicity test with the rainbow trout (Onchorhynchus mykiss). (PROJECT NUMBER:
439A-112; 8EHQ-01201-15037). Easton, MD: Wildlife International Ltd.
Du. M; Fang. C; Qiu. L; Dong. S; Zhang. X; Yan. C. (2015). Diastereoisomer-specific effects of
hexabromocyclododecanes on hepatic aryl hydrocarbon receptors and cytochrome P450s in
zebrafish (Danio rerio). Chemosphere 132: 24-31.
http://dx.doi.on 10 101 o i.chemosphere.,01 .s 02.049.
Du. M; Lin. L; Yan. C; Wang. C; Zhang. X. (2013). Enantiomer-specific bioaccumulation and
depuration of hexabromocyclododecanes in zebrafish (Danio Rerio). J Hazard Mater 248-249:
167-171. http://dx. doi. org/10.1016/i.i hazm at.2012.12.046.
Du. M; Lin. L; Yan. C; Zhang. X. (2012a). Diastereoisomer- and enantiomer-specific accumulation,
depuration, and bioisomerization of hexabromocyclododecanes in zebrafish (Danio rerio).
Environ Sci Technol 46: 11040-11046 fiHj.; »k.doi.org/"iO. 102 l/es302166p.
Page 504 of 723
-------
Du. M; Zhang. D; Yan. C; Zhang. X. (2012b). Developmental toxicity evaluation of three
hexabromocyclododecane diastereoisomers on zebrafish embryos. Aquat Toxicol 112-113: 1-10.
http://dx.doi.on .10_10Jo i.aquatox.2UIu I_0jj
e-Steward. (2020). List Recyclers [Website], http://e-stewards.ore/data/list-recyclers/.
EC/HC (Environment Canada and Health Canada). (2011). Screening Assessment Report on
Hexabromocyclododecane (pp. 1-125). Ottawa, Canada: EC/HC (Environment Canada and
Health Canada), http://www.ee.gc.ca/ese-ees/default.asp?lang=En&n=7882C148-l#a4.
ECB. (2003). Technical guidance document on risk assessment: Part II. (EUR 20418 EN/2).
Luxembourg: Office for Official Publications of the European Communities.
http://ihcp.irc.ec.europa.eu/our activities/public-
health/risk assessment of Biocides/doc/tgd/tgdpart2 2ed.pdf.
ECB. (2008). Risk assessment: Hexabromocyclododecane. Cas-No.: 25637-99-4. Luxembourg:
European Inventory of Existing Commercial Chemical Substances, Office for Official
Publications of the European Communities, https://echa.europa. eu/docum.emts/10162/661 bffl 7-
dcOa i I 9758-40bdd6198f82
ECHA. (2008a). Guidance on information requirements and chemicals safety assessment: Supporting
reference guidelines, chapters R12-R20. Helsinki, Finland.
http://echa.eiiropa.eii/web/giiest/giiidance-documents/giiidance-on-information-reqiiirements-
and-chemical-safetv-assessment.
ECHA.. (2008b). Risk assessment: hexabromocyclododecane. Helsinki, Finland.
https://echa.eitropa.eii/docum.en.ts, {«I /661bffl7-dc0ii 4 I > -9758-40bdd6198f82.
ECHA.. (2009a). Background document for hexabromocyclododecane and all major diastereoisomers
identified. Helsinki, Finland, https://echa.europa.eu/documents /hbedd en.pdf.
ECHA.. (2009b). Data on manufacture, import, export, uses and releases of HBCDD as well as
information on potential alternatives to its use. In IOM Consulting, supported by BRE, PFA and
Entec (pp. 108). (ECHA/2008/2). Helsinki, Finland.
https://echa.eiiropa.eu/documents, tech rep hbedd en.pdf.
ECHA.. (2009c). Prioritisation and Annex XIV background information: hexbromocyclododecane.
Helsinki, Finland, https://echa.europa.eu/documents ddec00-863a-4cff-abd2-
6d4b39a
ECHA.. (2017a). Adsorption/desorption: hexabromocyclododecane. Helsinki, Finland. Retrieved from
https://echa.europa.eu/registration-dossierA/regi stered-dossier/15003/5/5/2#
ECHA. (2017b). Chemical safety report: Hexabromocyclododecane and all major diastereoisomers
identified. Helsinki. Finland, https://echa.europa.eu/document? hn 2/7959599c-20e^ '
aldf-94fcdd23434e.
ECHA. (2017c). Chemical safety report: Hexabromocyclododecane and all major diastereoisomers
identified, Part 2. Helsinki, Finland, https://ech.a.europa.eu/documents kC/abj it « t. »0-
4d75-b253-( 3dd076.
Eggesb0. M; Thomsen. C; J'01'gensen. J\ , lecher > < 'dland. 10; Longnecker. MP. (2011).
Associations between brominated flame retardants in human milk and thyroid-stimulating
hormone (TSH) in neonates. Environ Res 111: 737-743.
http://dx.doi.on .envres.2011.05.004.
EINECS. (2008). Risk assessment: Hexabromocyclododecane. Cas-No.: 25637-99-4. Luxembourg:
European Inventory of Existing Commercial Chemical Substances, Office for Official
Publications of the European Communities, https://echa.europa. eu/documents/10162/661 bffl 7-
dcOa ?758-40bdd6198f82.
Eliarrat. E: de la Cal. A; Raldua. D; Duran. C: Barcelo. D. (2005). Brominated flame retardants in
Alburnus alburnus from Cinca River Basin (Spain). Environ Pollut 133: 501-508.
http://dx.doi.on .envpol.2.004.06.017
Page 505 of 723
-------
Eliarrat. E; de la Ca 1a. D; Duran. C; Barcelo. D. (2004). Occurrence and bioavailability of
polybrominated diphenyl ethers and hexabromocyclododecane in sediment and fish from the
Cinca River, a tributary of the Ebro River (Spain). Environ Sci Technol 38: 2603-2608.
http://dx.doi.org/10.1021/esQ301424
Eliarrat. E; Guerra. P; Martinez. E; Farre. M; Alvarez. JG: Lopez-Teiion. xelo. JX (2009).
Hexabromocyclododecane in human breast milk: levels and enantiomeric patterns. Environ Sci
Technol 43: 1940-1946. http://dx.doi.ore s802919e.
Ema. M; Fuiii. S: Hirata-Koizumi. M; Matsumoto. M. (2008). Two-generation reproductive toxicity
study of the flame retardant hexabromocyclododecane in rats. Reprod Toxicol 25: 335-351.
http://dx.doi.on .reprotox.2007.12.004.
Energy Recovery. C. (2018). 2018 Directory of waste-to-energy facilities. Energy Recovery Council.
http://energvrecovervcoiincil.ore/wp-content/iiploads/2019/10/ERC-2018-directorv.pdf.
Engter. RE. (2012). The complex interaction between marine debris and toxic chemicals in the ocean
[Review], Environ Sci Technol 46: 12302-12315. h\\\> .iv.doi.org-10 102 I '05302 i(^
EPS Industry Alliance. (2017). Preliminary information on manufacturing, processing, distribution, use,
and disposal: Cyclic aliphatic bromide cluster (HBCD). OCSPP. Public comment. (EPA-HQ-
OPPT-2016-073 5-0026).
Eriksson. P; Fischer. C; Wallin. M; Jakobssc iksson. A. (2006). Impaired behaviour, learning
and memory, in adult mice neonatally exposed to hexabromocyclododecane (HBCDD). Environ
Toxicol Pharmacol 21: 317-322. http://dx.doi.or10 JOI o 1 .etap.200> j 0 00 L
Esslinger. S; Becker. R; June. C; Schroter-Kermani. C; Bremser. W; Nehls. I. (201 la). Temporal trend
(1988-2008) of hexabromocyclododecane enantiomers in herring gull eggs from the German
coastal region. Chemosphere 83: 161-167. http://dx.doi.org 10 101» i.chemosphere.2010 _Q 0-V_.
Esslinger. S: Becker. R; Maul, k; Nehls. I. (2011b). Hexabromocyclododecane enantiomers:
microsomal degradation and patterns of hydroxylated metabolites. Environ Sci Technol 45:
3938-3944. http://dx.doi.org/10.1021 /es 1039584.
Esslinger. S; Becker. R; Muller-Belecke. A; Bremser. W; Jung. C; Nehls. I. (2010). HBCD stereoisomer
pattern in mirror carps following dietary exposure to pure gamma-HBCD enantiomers. J Agric
Food Chem 58: 9705-9710. http://dx.doi.org h' hLl lUOl I
Ethyl Corporation. (1990a). Genetic toxicology rat hepatocyte primary culture/DNA repair test on
hexabromocyclododecane with cover letter dated 030890. (TSCATS/405817. OTS0522234. Doc
I.D. 86900000163). Baton Rouge, LA.
https://ntri.ntis. gov/NT ihboard/searchResults.xhtml?searchQuerv=OTS0522234.
Ethyl Corporation. (1990b). Genetic toxicology salmonella/microsomal assay on
hexabromocyclododecane with cover letter dated 030890 [TSCA Submission],
(TSCATS/405818. OTS0522235. Doc I.D. 86900000164). Baton Rouge, LA.
https://ntri.ntis. gov/NT ihboard/searchResults.xhtml?searchQuerv=OTS0522235.
Fa. S: Pogrmic-Maikic. K; Samardziia. D; Hrubik. J; Glisic _ >vacevic. R; Andric. N. (2015).
HBCDD-induced sustained reduction in mitochondrial membrane potential, ATP and
steroidogenesis in peripubertal rat Leydig cells. Toxicol Appl Pharmacol 282: 20-29.
http://dx.doi.org i.taap.201 \ J 1.001.
Fa. S; Samardziia. D; Odzic. L; Pogrmic-Maikic. K; Kaisarevic. S; Kovacevic. R; Andric. N. (2014).
Hexabromocyclododecane facilitates FSH activation of ERK1/2 and AKT through epidermal
growth factor receptor in rat granulosa cells. Arch Toxicol 88: 345-354.
http://dx.doi.on 10 100 ' s0020 \ 0U I I 2.
Fangstrom. B; Athanasiadou. M Vihanassiadis. I; Bign^u \ n vndiean. P; Weihe. P; Bergman. A.
(2005). Polybrominated diphenyl ethers and traditional organochlorine pollutants in fulmars
(Fulmarus glacialis) from the Faroe Islands. Chemosphere 60: 836-843.
http://dx.doi.org h« to i.chemosphere.200^ 0J_065.
Page 506 of 723
-------
Fangstrom. B; Athanassiadis. I; Odsic )ren. K; Bergman. A. (2008). Temporal trends of
polybrominated diphenyl ethers and hexabromocyclododecane in milk from Stockholm mothers,
1980-2004. Mol Nutr Food Res 52: 187-193. http://dx.doi.< nfr.200700182.
Fent. KW: Evans. DE; Babik. K; Strilev. C; Bertke. S; Kerber. S; Smith. D; Horn. GP. (2018). Airborne
contaminants during controlled residential fires. J Occup Environ Hyg 15(5): 399-412.
http://dx.doi.Org/https://doi.org/l 0.1080/15459624.^
Fent. KW; Horn. GP; DeCrane. S. (2015). Firefighters' Perspective on Flame Retardants.
https://www.sfpe.org/page/FPE 2015 04 4
Fern an d( >rtim.er. D; Rose. M; Smith. F; Panton. S; Garcia-Lopez. M. (2016). Bromine content
and brominated flame retardants in food and animal feed from the UK. Chemosphere 150: 472-
478. http://dx.doi.on h>_lvM_*/i.chemosphere .*-_«L_>M.
Fernie. KJ; Marteinson. SC; BinI j *M; Ritchie U I etcher. RJ. (2011). Reproductive changes in
American kestrels (Falco sparverius) in relation to exposure to technical
hexabromocyclododecane flame retardant. Environ Toxicol Chem 30: 2570-2575.
http://dx.doi.on 02/etc.652.
Finizio. A; Mackav. D; Bidleman. T; Hamer. T. (1997). Octanol-air partition coefficient as a predictor
of partitioning of semi-volatile organic chemicals to aerosols. Atmos Environ 31: 2289-2296.
Finken. MI; van Eiisden. M; Loomans. EM; Vriikotte. TG; Rotteveet. J. (2013). Maternal
hypothyroxinemia in early pregnancy predicts reduced performance in reaction time tests in 5- to
6-year-old offspring. J Clin Endocrinol Metab 98: 1417-1426. http://dx.doi.org/10.1210/ic.20 i 2-
3389.
Fisher. DA; Nelson. JC. (2012). Application of TSH and free thyroxine measurements to thyroid
diagnosis: Laboratory support of diagnosis and management. Fisher, DA; Nelson, JC.
http://www.questdiagnostics.com/testcenter/testguide. action?dc=^ pTSH.
Flame Control Coatings. (2018). Notes of telephone conversation with Paul Pisarksi of Flame Control
Coatings and Sue Slotnick of EPA, 6-29-18. https://www.regutations. gov/document?D=EP A-
HO-OPI
Foekema, EM; Lopez Parron, M; Mergia, MT; Carolus, ER; Yd Berg, JH;
(2014). Internal effect concentrations of organic substances for early life development of
egg-exposed fish. Ecotoxicol Environ Saf 101: 14-22.
http://dx.doLorg/l 0.1016/i.ecoem .<' i «_C 006.
Forhead. >wden. AL. (2014). Thyroid hormones in fetal growth and prepartum maturation
[Review], J Endocrinol 221: R87-R103. http://dx.doi.oi^ !<> ! ^ kI { 0025.
Frascb HU t^tson. GS; Bunge. AL; Chen. C; Cherrie. JW; Kastii^ T ' v'N.004.
ime. H; Hilger. B; Kopp. E; Miser ok. M; Volkel. W. (2014). Polybrominated diphenyl ethers
(PBDEs), hexabromocyclododecane (HBCD) and "novel" brominated flame
retardants in house dust in Germany. Environ Int 64: 61-68.
http://dx.doi.on 10 [Oh i.envint.-0\ » l KM.'.
FSA. (2006). Brominated chemicals: UK dietary estimates.
https://web.archive.Org/web/20120403220652/https://www.food.gov.uk/multimedia/pdfs/fsisl00
[•
Page 507 of 723
-------
Gavilan-Garcia. I; Elvira. SS; Arturo. GG; Erik. BM; Gonzalez-Gonzalez. LA. (2017). Brominated
Flame Retardants (BFRs) analysis in leachates and sludge from a landfill and wastewater plant in
the metropolitan area of Mexico City. J Environ Anal Toxicol 7: 2161-0525.
Genskow. KR; Bradner. JM; Hossain. MM; Richardson. JR.; Caudle. WM. (2015). Selective damage to
dopaminergic transporters following exposure to the brominated flame retardant, HBCDD.
Neurotoxicol Teratol 52: 162-169. http://dx.doi.ore/10 I0t6/i.ntt.201 > 06.003.
Gerecke. AC; Giger. W; Hartmann. PC; Heeb. NY; Kohler. HP; Schmid. P; Zennegg. M; Kohler. M.
(2006). Anaerobic degradation of brominated flame retardants in sewage sludge. Chemosphere
64: 311-317. http://dx.doi.org/ chemosphere.2005.12.016.
Germer. S; Piersma. AH; van der Yen. L; Kamyschnikow. A; Fen- \_ ^chmitz. HI; SchrenL (2006).
Subacute effects of the brominated flame retardants hexabromocyclododecane and
tetrabromobisphenol A on hepatic cytochrome P450 levels in rats. Toxicology 218: 229-236.
http://dx.doi.on h1 h'l * i.tox.200' [0 0i,s
Geyer. HI; Schramm. K. -W; Darnemd. PO; Aune. A I. t 'oht. EA; Fried. KW; Henkelmann. El; Lenoir.
D; Schmid. P; McDonald. TA. (2004). Terminal elimination half-lives of the brominated flame
retardants TBBPA, HBCD, and lower brominated PBDEs in humans. Organohalogen Compd 66:
3820-3825.
Gilbert. ME. (201 1). Impact of low-level thyroid hormone disruption induced by propylthiouracil on
brain development and function. Toxicol Sci 124: 432-445.
http://dx.doi.org/10.1093/toxsci/kri ll J j.
Gilbert. ME; Hedge. JM; Valentin-Blasiw I Mount. BC; Kannan. K; Tietge 1 Veller. RT; Crofton.
KM; Jarrett. JM; Fisher. JW. (2013). An animal model of marginal iodine deficiency during
development: The thyroid axis and neurodevelopmental outcome. Toxicol Sci 132: 177-195.
http://dx.doi.on 93/toxsci/kfs335.
Gilbert. ME; Ramos. RL; McCloskev. DP; Goodman. JH. (2014). Subcortical band heterotopia in rat
offspring following maternal hypothyroxinaemia: Structural and functional characteristics. J
Neuroendocrinol 26: 528-541. http://dx.doi.or\ hMj 11 mc.L-' U2-
Gilbert. ME; Rovet. J; Chen. Z; Koibuchi. N. (2012). Developmental thyroid hormone disruption:
prevalence, environmental contaminants and neurodevelopmental consequences.
Neurotoxicology 33: 842-852. http://dx.doi.c 5/i.neuro.2011.11.005.
Gilbert. ME; Zoeller. RT. (2010). Thyroid hormones—impact on the developing brain: Possible
mechanisms of neurotoxicity. In GJ Harry; HA Tilson (Eds.), Target Organ Toxicology Series,
vol 28 (3rd ed., pp. 79-111). New York, NY: Informa Healthcare.
Gilbert. ME; Sanchez-Huerta. K; Wood. C. (2016). Mild thyroid hormone insufficiency during
development compromises activity-dependent neuroplasticity in the hippocampus of adult male
rats. Endocrinology 157: 774-787. http://dx.doi.org/ 0/en.2015-1643.
Glynn. A; Lignell. S; Darnemd. PQ; Aune. M; Haildin Ankarbon j Mn'gdafd i\ ^ loegciut <
Bensrvd. I. (2011). Regional differences in levels of chlorinated and brominated pollutants in
mother's milk from primiparous women in Sweden. Environ Int 37: 71-79.
http://dx.doi.on lOlo i.envint.^Mo o ^03.
Gochfeld. M; Burger. J. (2011). Disproportionate exposures in environmental justice and other
populations: the importance of outliers. Am J Public Health 101 Suppl 1: S53-S63.
http://dx.doi.org/10.2105/AJPH.2011.300121.
Great Lakes Chemi (2003). Material safety data sheet (MSDS): great lakes CD-75P. West
Lafayette, Indiana.
Greene. MA. (2002). Mouthing times for children from the observational study. Bethesda, MD: U.S.
Consumer Product Safety Commission.
GSRI. (1978). Mutagenicity test of GLS-S6-41A (not published). (TSCATS/443581. OTS0000947.
EPA/OTS Doc #FYI-OTS-0794-0947).
Page 508 of 723
-------
GSRI (1994). Initial submission: Letter from Ethyl Corp to USE PA re technical and toxicity data on
brominated flame retardants including hexabromocyclododecane, * w/attchmts, dated 5/23/88
[TSCA Submission], (FYIOTS07940947). Baton Rouge, LA: Ethyl Corporation.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuerv=OTSOOOQ947.
Guerra. P; Alaee. M; Jimenez. B; Pacepavicius. G; Marvin. C: Macini xelo. D;
Champoux. I , Ivmie. K. (2012). Emerging and historical brominated flame retardants in
peregrine falcon (Falco peregrinus) eggs from Canada and Spain. Environ Int 40: 179-186.
. ore/10.1016/i. en vint.2011.07.014.
Guerra. P; De La Cal \ Vlarsh. G; Elian6/NEJM19990819» 10801.
Hakk. H. (2016). Comparative Metabolism Studies of Hexabromocyclododecane (HBCD)
Diastereomers in Male Rats Following a Single Oral Dose. Environ Sci Technol 50: 89-96.
http://dx.doi.ore/10.1021/acs.est.5b04510.
Hakk. H; Szabo. DT; Huwe. J; Diliberto. J; Bimbaum. LS. (2012). Novel and distinct metabolites
identified following a single oral dose of a- or y-hexabromocyclododecane in mice. Environ Sci
Technol 46: 13494-13503. http://dx.doi.ore/ 303209e.
Hamers. T; Kamstra. JH; Sonneveld. E; Murk. AJ: Kester. MH; Andersson. PL; Leeler. J; Brouwer. A.
(2006). In vitro profiling of the endocrine-disrupting potency of brominated flame retardants.
Toxicol Sci 92: 157-173. http://dx.doi.ore/10.1093/toxsci/kfi 187.
Harrad. S; Abdallah. MA. (2011). Brominated flame retardants in dust from UK cars—within-vehicle
spatial variability, evidence for degradation and exposure implications. Chemosphere 82: 1240-
1245. http://dx.doi.ore/1 u j 0 i o [ chemosphere.2010 UjJJS.
Harrad. S; Abdallah. MA. (2015). Concentrations of polybrominated diphenyl ethers,
hexabromocyclododecanes and tetrabromobisphenol-A in breast milk from United Kingdom
women do not decrease over twelve months of lactation. Environ Sci Technol 49: 13899-13903.
http://dx.doi.on 21/acs.est.5b00539.
Harrad. S; Goosey. E; Desborouel idallah. MA; Roosens. L; Covaci. A. (2010). Dust from U.K.
primary school classrooms and daycare centers: the significance of dust as a pathway of
exposure of young U.K. children to brominated flame retardants and polychlorinated biphenyls.
Environ Sci Technol 44: 4198-4202. http://dx.doi.ore/10.1021 /es 1 ».
Page 509 of 723
-------
Harrad. S; Abdallah. MA; Rose. NL; Turner. SD; Davidson. TA. (2009). Current-use brominated flame
retardants in water, sediment, and fish from English lakes. Environ Sci Technol 43: 9077-9083.
http://dx.doi.org/10.1021/es9Q2185u.
Hattis. D; Goble. R; Russ. A; Chu. M: Ericson. J. (2004). Age-related differences in susceptibility to
carcinogenesis: A quantitative analysis of empirical animal bioassay data. Environ Health
Perspect 112: 1152-1158. http://dx.doi.ore/10.1289/ehp.6871.
Haukas. M; Hvlland. K; Bergi ¦Jygard. T; Mariussen. E. (2009). Spatial diastereomer patterns of
hexabromocyclododecane (HBCD) in a Norwegian fjord. Sci Total Environ 407: 5907-5913.
http://dx.doi.Org/10.1016/i.scitotenv.2009.08.02.4.
Haukas. M; Hvlland. K; Nygard. T; Ben Mariussen. E. (2010a). Diastereomer-specific
bioaccumulation of hexabromocyclododecane (HBCD) in a coastal food web, Western Norway.
Sci Total Environ 408: 5910-5916. http://dx.doi.oiT 10 101» i.scitotenv.20 i0.08.026.
Haukas. M; Runs. A; Hvlland. K; Berge. ariussen. E. (2010b). Bioavailability of
hexabromocyclododecane to the polychaete Hediste diversicolor: exposure through sediment and
food from a contaminated fjord. Environ Toxicol Chem 29: 1709-1715.
http: //dx. doi. or g/10.1002/etc. 201.
He. MI; Luo. XI; Yu. LH; Wu. IP; Chen. SI; Mai. BX. (2013). Diasteroisomer and enantiomer-specific
profiles of hexabromocyclododecane and tetrabromobisphenol A in an aquatic environment in a
highly industrialized area, South China: vertical profile, phase partition, and bioaccumulation.
Environ Pollut 179: 105-110. http://dx.doi.Org/10.1016/i.envpol.2
Hedge. JM; Devito. MI; Crofton. KM. (2009). In vivo acute exposure to polychlorinated biphenyls:
effects on free and total thyroxine in rats. Int J Toxicol 28: 382-391.
http://dx.doi.or >9344631.
Heindel. II. (1998). Oocyte quantitation and ovarian histology. In An evaluation and interpretation of
reproductive endpoints for human health risk assessment. Washington, DC: ILSI Press.
Helledav. T; Tuominen. KL; Bergman. A; Jenssen. D (1999). Brominated flame retardants induce
intragenic recombination in mammalian cells. Mutat Res 439: 137-147.
http://dx.doi. org/10.1016/S13 83-5718(98)00186-7.
Henkel. C. (2016). Safety data sheet (SDS): LF721 medium bulk. Bien Hoa City, Vietnam: Henkel
Corporation.
http://mvmsds.henkel.com/mymsds/Search.do7BIJS ARE A=OOQ6&DOCTYPE=MSDS&COUN
M.TNK=1391002.
Henrichs. J: Bongers-Schokking. II; Schenk. II; Ghassabian. Schmidt. HG; Visser. TJ; Hooijkaas. H;
de Muinck Keizer-Schrama. SM; Hofman. A; Jaddoe. VV: Visser. W; Steegers. EA; Verhulst.
\ I. deRiikc VH. Ficmeier. H. (2010). Maternal thyroid function during early pregnancy and
cognitive functioning in early childhood: The generation R study. J Clin Endocrinol Metab 95:
4227-4234. http://dx.doi.org/) 0 IJ 10/ic.^ < M * M 1 :
Hinkson. NC: Whalen. MM. (2009). Hexabromocyclododecane decreases the lytic function and ATP
levels of human natural killer cells. J Appl Toxicol 29: 656-661.
Hinkson. NC: Whalen. MM. (2010). Hexabromocyclododecane decreases tumor-cell-binding capacity
and cell-surface protein expression of human natural killer cells. J Appl Toxicol 30: 302-309.
http://dx.doi.org/10.1002/iat.1495.
Hlouskova. V; Lankova. D; Kalachova. K; Hradkova. P; Poustka. J; Haislov; lkrabova. J. (2013).
Occurrence of brominated flame retardants and perfluoroalkyl substances in fish from the Czech
aquatic ecosystem. Sci Total Environ 461-462: 88-98.
http://dx.doi.oi>. 10 101 > i.scitotenv.,vM ;_pj.
-------
Hone. H; Li. D; Shen. R; Wane. X; S (2014). Mechanisms of hexabromocyclododecanes induced
developmental toxicity in marine medaka (Oryzias melastigma) embryos. Aquat Toxicol 152:
173-185. http://dx.doi.ore/10/1016/i.aquatox.201 KM_0 k».
Hone. H; Shen. R: Liu. W: Li. D; Huang. L; Shi (2015). Developmental toxicity of three
hexabromocyclododecane diastereoisomers in embryos of the marine medaka Oryzias
melastigma. Mar Pollut Bull 101: 110-118. http://dx.doi.on, i(,l* j.marpolbul.20t. s s
Horn. GP; Kerber. S: Fent. KW: Fernhall. B: Smith. PL. (2016). Interim Report: Cardiovascular &
Chemical Exposure Risks in Modern Firefighting. (EMW-2013-FP-00766). Urbana-Champaign,
IL: Illinois Fire Service Institute, University of Illinois at Urbana-Champaign.
https://www.fsi.illinois.edit/documents/research/CardioChemRisksModei iterimReport201
f
Howdeshell. KL. (2002). A model of the development of the brain as a construct of the thyroid system
[Review], Environ Health Perspect 110 Suppl 3: 337-348.
http://dx.doi.ore/lO t_'89/ehi.» l ' *
HSDB. (2008). H ex abromocyclododecane (HBCD) [Database], Bethesda, MD: National Library of
Medicine. Retrieved from http://toxnet.nlm.nih.eov/cei-bin/sis/search/f?./temp/~XQzOHv: 1
Hu. J; Liane ;en. M: Wane. X. (2009a). Assessing the toxicity of TBBPA and HBCD by zebrafish
embryo toxicity assay and biomarker analysis. Environ Toxicol 24: 334-342.
http://dx.doi.ore/10.1002/tox.20436.
Hu. X: Hu. D; Xu. Y. (2009b). Effects of tetrabrominated diphenyl ether and hexabromocyclododecanes
in single and complex exposure to hepatoma HepG2 cells. Environ Toxicol Pharmacol 27: 327-
337. http://dx.doi.ore/V' !l'l '/i.etap.20CK 1 I <« 14
Huane. L: Fantke. P; Ernstoff. A; JoUiet. OA. (2017). Quantitative Property-Property Relationship for
the Internal Diffusion Coefficients of Organic Compounds in Solid Materials. Indoor Air 27:
1128-1140.
(2016). American housing survey, components of inventory change: 2011-2013.
https://www.hiidiiser.eov/portal/datasets/cinch/cinchl3/cint1i t L11 i l4£-
Hulbert. A.J. (2000). Thyroid hormones and their effects: a new perspective [Review], Biol Rev Camb
Philos Soc 75: 519-631.
Huntingdon Research Center. (1990). Ames metabolic activation test to assess the potential mutagenic
effect of und no. 49 with cover letter dated 031290 [TSCA Submission], (TSCATS/406642.
OTS0522948. 86900000385). Wyandotte, MI: BASF Corporation.
https://ntrl.ntis. eov/NT ihboard/searchResults.xhtml?seareh.Query=OTS0522.948.
Ill I Ijibs, (1990). Mutagenicity of two lots of FM-100 lot 53 and residue of lot 3322 in the absence and
presence of metabolic activation with test data and cover letter [TSCA Submission], In Epa/Ots
(pp. #86-900000267). (TSCATS/407259. OTS0523259. Doc I.D. 86900000267). West
Lafayette, IN: Great Lakes Chemical Corporation.
https://ntrl.ntis. eov/NT ihboard/searchResults.xhtml?seareh.Query=OTS0523259.
Ichihara. M; Yamamoto. A; Takakura. K; Kakutani. N: Sudo. M. (2014). Distribution and pollutant load
of hexabromocyclododecane (HBCD) in sewage treatment plants and water from Japanese
Rivers. Chemosphere 110: 78-84. http://dx.doi.ore/tO 101o/i.chemosphere.201 4 I
Ityina. I', Hunziker. RW. (2010). Scenarios of temporal and spatial evolution of
hexabromocyclododecane in the North Sea. Environ Sci Technol 44: 4622-4628.
http://dx.doi.on ;s9034599.
Indium. C. (2018a). Follow-up questions about Indium's use of HBCD-Email exchange with Indium
Corporation and Sue Slotnick in September 2018 [Personal Communication],
Indium. C. (2018b). Use of HBCD in manufacture of solder paste by Indium Corporation-Phone call
with Sue Slotnick [Personal Communication],
Page 511 of 723
-------
Indium Corporation. (2019a). Application Notes: Soldering 101- A Basic Overview. Indium
Corporation, https://www.indium.com/technical-documents/application-n.otes/.
Indium Corporation. (2019b). Product Data Sheets: TACFlux 483. Indium Corporation.
https://www.indium.com/technical-documents/product-data-sheets/.
I.N.EOS Styrenics. (2017). Analysis of alternatives: HBCDD use in EPS for building applications.
Helsinki, Finland: European Chemicals Agency.
https://echa.europa.eu/documents, I PL'. ilxit-'i I>.->0-45a5-97e2-lc9c7597a692.
Insulfoam a Division of Carlisle Construction. M. (2015). Safety data sheet (SDS): 085 insulfoam
molded EPS products. Puyallup, WA: Insulfoam a Division of Carlisle Construction Materials.
https://www.insulfoam.com/wp-content/uploads/2014/04/MSDS Aug-2015.pdf.
IRDC. (1978a). Acute inhalation toxicity study in rats with hexabromocyclododecane with attachments
and cover letter dated 042478 [TSCA Submission], (EPA/OTS Doc #88-7800137;
8EHQ04780137). Chicago, IL: Velsicol Chemical Corporation.
https://ntrl.ntis.gov/NT^ 'l;ishboard/searchResults.xhtml?searchQuerv=OTS020Q488.
IRDC. (1978b). Acute toxicity studies in rabbits and rats with hexabromocyclododecane with
attachments [TSCA Submission], (EPA/OTS Doc #88-7800065; 8EHQ02780065). Chicago, IL:
Velsicol Chemical Corporation.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuerv=OTS0200Q51.
IRDC. (1978c). Acute toxicity studies in rabbits and rats with residue of hexabromocyclododecane with
attachments and cover letter dated 030178 [TSCA Submission], (88-7800088; 8EHQ03780088).
Chicago, IL: Velsicol Chemical Corporation.
https://ntrl.ntis. gov/NT ;hboard/searehResults.xhtml?seareh.Query=OTS0200466.
Jacobs. K. (2019). [Personal communication with Keith Jacobs, EPA containing individual animal data
on physical development for Ema et al, 2008] [Personal Communication],
Jang. M; Shim. VV J. Han. GM; Rani. M; Sons. YK; Hone. SH. (2016). Styrofoam Debris as a Source of
Hazardous Additives for Marine Organisms. Environ Sci Technol 50: 4951-4960.
http://dx.doi.on _ ics.est.5b05485
Jang. M; Shi in. GM; Rani. M; Song. YK; Hong. SH. (2017). Widespread detection of a
brominated flame retardant, hexabromocyclododecane, in expanded polystyrene marine debris
and microplastics from South Korea and the Asia-Pacific coastal region. Environ Pollut 231:
785-794. http://dx.dQM s ^M6/i.envpol.20l 1 08.066
Jeong. GH; Hwang. NR.; Hwang. EH; Lee. BC; Yoon. I. (2014). Hexabromocyclododecanes in crucian
carp and sediment from the major rivers in Korea. Sci Total Environ 470-471: 1471-1478.
http://dx.doi.org 10 101 > • t.scitotenv.201 * 10 0.'<8.
Johnson. PI; Stapleton. HM; Mukheri-''' Hauser. R; Meekei J3"* (2013). Associations between
brominated flame retardants in house dust and hormone levels in men. Sci Total Environ 445-
446: 177-184. http://dx.doi.org 10 tOlo i.scitotenv.-^M. 1.
Julvez. J; Alvarez-Pedrerol. M. ar; Rebagliato. M; Murcia. M; Foms. J; Garcia-Esteban. R; Lertxundi.
N; Espada. M; Tardon. A; Riano Gal an. I; Sunyer. I. (2013). Thyroxine levels during pregnancy
in healthy women and early child neurodevelopment. Epidemiology 24: 150-157.
http://dx.doi.org/l 0.1097/EDE.0b013e318276ccd3.
Kaiiwara. N; Hi rata. O; Takigami. H; Noma. Y; Tachifuii. A; Matsufuii. Y. (2014). Leaching of
brominated flame retardants from mixed wastes in lysimeters under conditions simulating
landfills in developing countries. Chemosphere 116: 46-53.
http://dx.doi.off h' h'l * i.chemosphere.JO 14 01 025.
Kakimoto. K; Akutsu. K; Konishi maka. Y. (2008). Time trend of hexabromocyclododecane in the
breast milk of Japanese women. Chemosphere 71: 1110-1114.
.on U'J ° L* *. chemosphere.. ^«' J p {*
Page 512 of 723
-------
Kakirooto. K; Nagavoshi. H; Yoshida J, \Uitsu. K; Konisl^ > I ^ba. A; Havakawa. K. (2012).
Detection of Dechlorane Plus and brominated flame retardants in marketed fish in Japan.
Chemosphere 89: 416-419. http://dx.doi.org/ • \ chemosphere.2^m o < o 2.
Kataoka. T: Kuribara. I; Katagiri. k; Wada. T; Masunaga. S. (2012). Emission Rates of
Hexabromocyclododecane (HBCD) from the Surface of Building Materials. Japan: Chemicals
Evaluation and Research Institute Japan, Yokohama National University, Graduate School of
Environment and Information Sciences.
Kato. Y; Ikushiro. S: Emi. Y; Tamaki. S; Suzuki. H; Sakaki. T; Yamada. S: Deeawa. M. (2008). Hepatic
UDP-glucuronosyltransferases responsible for glucuronidation of thyroxine in humans. Drug
Metab Dispos 36: 51-55. http://dx.doi.ors imd. 107.018184.
Kavlock. . BC: Faustman. EM; Kimmel. CA. (1995). Dose-response assessments for
developmental toxicity .4. Benchmark doses for fetal weight changes. Toxicol Sci 26: 211-222.
http://dx.doi.on 06/faat. 1995.1092.
Kelly. GS. (2000). Peripheral metabolism of thyroid hormones: a review [Review], Altern Med Rev 5:
306-333.
Kerol. (2009). Proposal for harmonised classification and labelling: Substance name:
Hexabromocyclododecane. Helsinki, Finland: European Chemicals Agency.
http s: //eefa a. europa. eu/docum ents, '8284634d-8fa9-43ad-acaa-f449dc28ff8c.
Kester. (2018). Email exchange between Kester and EPA, August 2018 re use of HBCD [Personal
Communication].
Kester. I. (2015). Safety data sheet (SDS): EP256HA solder paste. Itasca, IL: Kester Inc.
http ://www.techni-tool. com/site/MSDS/488 SQ2566.pdf.
Kicinski. M; Viaene. MK; Den Hond. E; Schoeters. G: Covaci. den. V: Bruckers. L;
Croes. K; Sioen. I; Baevens. W; Van Larebeke. N; Nawrot TS. (2012). Neurobehavioral
function and low-level exposure to brominated flame retardants in adolescents: A cross-sectional
study. Environ Health 11: 86. http://dx.doi.on [0 i [S* H o 069X-11-86.
Kim. Ill; Oh. JE. (2014). Tetrabromobisphenol A and hexabromocyclododecane flame retardants in
infant-mother paired serum samples, and their relationships with thyroid hormones and
environmental factors. Environ Pollut 184: 193-200.
http://dx.doi.on .envpol.2.013.08.034.
Kissel. JC. (201 1). The mismeasure of dermal absorption. J Expo Sci Environ Epidemiol 21: 302-309.
http://dx.doi.on 38/ies.2010.22.
Kitto. JBStSC. (1992). Steam: Its Generation and Use. In JBSSC Kitto (Ed.), (40th ed.). Barberton,
Ohio: The Babcock & Wilcox Company.
Klaassen. CD; Hood. AM. (2001). Effects of microsomal enzyme inducers on thyroid follicular cell
proliferation and thyroid hormone metabolism [Review], Toxicol Pathol 29: 34-40.
http://dx.doi.on 80/019262301301418838.
KLIF. (2010). New organic pollutants in air, 2007. Brominated flame retardants and polyfluorinated
substances. In Norwegian Institute for Air Research (NILU) (pp. 61). (SPFO-report 1077/2010,
TA-2689/2010). Norway: Climate and Pollution Agency.
http://www.miliodirektoratet.no/old/klif/publikasioner/2689/ta2689.pdf.
Klosterhaus. SL; Stapleton. HM; La Guardia. Ml; Greij (2012). Brominated and chlorinated flame
retardants in San Francisco Bay sediments and wildlife. Environ Int 47: 56-65.
http://dx.doi.on, 10 101*' i.envint.^1 - 06.005.
Kobiliris. D. (2010). Influence of embryonic exposure to hexabromocyclododecane (HBCD) on the
corticosterone response and "fight or flight" behaviors if captive American kestrels (pp. 1-58).
Montreal, Canada: Kobiliris, D.
id=0&dvs=1488987858161-447.
Page 513 of 723
-------
Koelmar dr. A; Burton. GA; Janssen. CR. (2016). Microplastic as a Vector for Chemicals in
the Aquatic Environment: Critical Review and Model-Supported Reinterpretation of Empirical
Studies [Review], Environ Sci Technol 50: 3315-3326. http://dx.doi.ore/10.1021/acs.est.5b06069
Kohler. M; Zennegg. M; Bogdal. C; Gerecke. AC; Schmid. P; Heeb. NY: Sturm. M; Vonmont. H;
Kohler. HP; Giger. W. (2008). Temporal trends, congener patterns, and sources of octa-, nona-,
and decabromodiphenyl ethers (PBDE) and hexabromocyclododecanes (HBCD) in Swiss lake
sediments. Environ Sci Technol 42: 6378-6384. http://dx.doi.org/10.1021/es702586r.
Koibuchi. N: Chin. MW. (2000). Thyroid hormone action and brain development. Trends Endocrinol
Metab 11: 123-128. http://dx.doij 6/S1043-2760(00)00238-1.
Koike. E; Yanagisawa. R; Takano. H. (2016). Brominated flame retardants, hexabromocyclododecane
and tetrabromobisphenol A, affect proinflammatory protein expression in human bronchial
epithelial cells via disruption of intracellular signaling. Toxicol In Vitro 32: 212-219.
http://dx.doi.on h1 h'l * i.tiv.t J ot *
Kowalsk azur. M. (2014). The Simultaneous Determination of Six Flame Retardants in Water
Samples Using SPE Pre-concentration and UHPLC-UV Method. Water Air Soil Pollut 225:
1866. http://dx.doi.org/10 n m 014 IVS6-4
Kuang. J; M arrad. S. (2016). Concentrations of "legacy" and novel brominated flame retardants
in matched samples of UK kitchen and living room/bedroom dust. Chemosphere 149: 224-230.
http://dx.doi.on .cheMosphere.2016.01.092.
Kuiper. RV; Canton. RF; Leonards. PE; Jenssen. BM; Dubbeldam. M; Wester. PW; van den Berg. M;
Vos. <0 \ ethaak M >. (2007). Long-term exposure of European flounder (Platichthys flesus) to
the flame-retardants tetrabromobisphenol A (TBBPA) and hexabromocyclododecane (HBCD).
Ecotoxicol Environ Saf 67: 349-360. http://dx.doi.on 10 l0L«/i.ecoenv.200» _ j 00 j
Kurokawa. Y; Inoue. T; LJchida. Y; Momma. J. (1984). Carcinogenesis test of flame retarder
hexabromocyclododecane in mice (Summary, unpublished, translated into English). Department
of Toxicology, National Public Health Research Institute, Biological Safety Test and Research
Centre.
La Guardia. MI; Hale. RC; Harvey. E; Chen. D. (2010). Flame-retardants and other organohalogens
detected in sewage sludge by electron capture negative ion mass spectrometry. Environ Sci
Technol 44: 4658-4664. http://dx.doi.org/10.1021 /es9039264.
La Guardia. MI; Hale. RC; Harvey. E; Mainor. TM; Ciparis. S. (2012). In situ accumulation of HBCD,
PBDEs, and several alternative flame-retardants in the bivalve (Corbicula fluminea) and
gastropod (Elimia proxima). Environ Sci Technol 46: 5798-5805.
http://dx.doi.org/10.1021/es3004238.
Lambert. S; Sinclai jxatt. A. (2014). Occurrence, Degradation, and Effect of Polymer-Based
Materials in the Environment. Rev Environ Contam Toxicol 227: 1-53.
http://dx.doi.on fO J00 * ;?78J r 01 5. ' >J
Lansink. CJM; Breelen. MSC; Marqiniu j, van Hem, men. J J. (1996). Skin exposure to calcium
carbonate in the paint industry. Preliminary modelling of skin exposure levels to powders based
on field data. (V96.064). Rijswijk, The Netherlands: TNO Nutrition and Food Research Institute.
LANXESS. (2017a). Personal communication between Richard Henrich, LANXESS Solutions US, Inc.
(formerly Chemtura Corporation), and Timothy Lehman, EPA, regarding
Hexabromocyclododecane (HBCD).
Laveis j* Bont! _ _ Hutton. la. (2014). Plastic ingestion by Flesh-footed Shearwaters (Puffinus
carneipes): Implications for fledgling body condition and the accumulation of plastic-derived
chemicals. Environ Pollut 187: 124-129. http://dx.doi.org/10 [Q lo/i.envpol.2Q I.' 12 0-0
Law. K; HaHdorson. T; Danell. R; Stern. >;wurtz. S; Alaee. M; Marvin t\ W hutle. M; Tomy. G
(2006). Bioaccumulation and trophic transfer of some brominated flame retardants in a Lake
Page 514 of 723
-------
Winnipeg (Canada) food web. Environ Toxicol Chem 25: 2177-2186.
http://dx.doi.Org/10.1897/05-500R.l
Law. K; Palace. VP; Halldorson. T: Danell R; Wautier. K I . ans. >3, Utee. M; Marvin. C: Toms
(2006). Dietary accumulation of hexabromocyclododecane diastereoisomers in juvenile rainbow
trout (Oncorhynchus mykiss). I: Bioaccumulation parameters and evidence of bioisomerization.
Environ Toxicol Chem 25: 1757. http://dx.doi.org/10.1897/Q5-
Law. R. obin J.; Allchin. C. olin R ; de Boer, J. acob: Covan drian; Herzk*' t * < »rte; Lepom
Morri. S. teven; Tronczynski. J. acek: de Wit. C. vnthia A. (2006). Levels and trends of
brominated flame retardants in the European environment. Chemosphere 64.
http://dx.doi.on .chemosphere.2005.12.007.
Lee. tang. WH; Chen. HL. (2019). Dietary exposure and risk assessment of exposure to
hexabromocyclododecanes in a Taiwan population. Environ Pollut 249: 728-734.
http://dx.doi.on .envpol.2019.03.040.
Leisewitz. A; Kruse. H; Schramm. E. (2001). Substituting environmentally relevant flame retardants:
Assessment fundamentals. Berlin, Germany: Umweltbundesamt.
https://www.iimweltbimdesamt.de/en/piiblikationen/siibstitiiting-environmentallY-relevant-flame
Leonards. PEG; Santillo. D; Brigden. K; Veen. I; Van. Hesselinger er. J; Johnston. R. (2001).
Brominated flame retardants in office dust samples. Proceedings of the Second International
Workshop on Brominated Flame Retardants, 14-16 May 2001 (pp. 1-4). Stockholm, Sweden:
Leonards, PEG; Santillo, D; Brigden, K; Veen, I; Van Hesselingen, J; De Boer, J; Johnston, P.
http://edepot.wur.a
Letcher. RJ; Gebbink. WA; Sonne. C; Bom. EW; Mckinnev. MA; Dietz. R (2009). Bioaccumulation
and biotransformation of brominated and chlorinated contaminants and their metabolites in
ringed seals (Pusa hispida) and polar bears (Ursus maritimus) from East Greenland. Environ Int
35: 1118-1124. http://dx.doi.o /i.envint.2009.07.006.
Lewis. AC; Patanlq (1978a). A dermal LD50 study in albino rabbits and an inhalation LC50 study
in albino rats. Test material GLS-S6-41A [unpublished], (Experiment Reference No. 78385-2).
Fairfield, NJ: Consumer Product Testing.
Lewis. AC; Palanla (1978b). A primary dermal irritation study, a dermal corrosion study, and an
ocular irritation study in albino rabbits and an oral LD50 study in albino rats: Test material GLS-
S6-41A. (78385-1). Consumer Product Testing.
L lo. T; Sun. H; Zhang. Y; Yang. J. (2016). Diastereomer- and enantiomer-specific accumulation,
depuration, bioisomerization, and metabolism of hexabromocyclododecanes (HBCDs) in two
ecologically different species of earthworms. Sci Total Environ 542: 427-434.
http://dx.doi.on J^ *_• i.scitotenv.lV1 .s j Oj iV.
Lignell. S; Damerud. PO; Aune. M; Tomkvist. A. (2003). Persistent organic pollutants (POP) in
breastmilk from primiparae women in Uppsala County, Sweden, 2002-2003. In Report to the
Swedish Environmental Protection Agency (pp. 10). Lignell, S; Darnerud, PO; Aune, M;
Tomkvist, A. https://www.diva-portal .org/smash/get/diva2:65786!
Lilienthal. H; van der Ven. LT; Piersma. AH; Vos. JG. (2009). Effects of the brominated flame retardant
hexabromocyclododecane (HBCD) on dopamine-dependent behavior and brainstem auditory
evoked potentials in a one-generation reproduction study in Wistar rats. Toxicol Lett 185: 63-72.
http://dx.doi.org/10.1016/i.toxlet.200- * J ^_2.
Lind. Y; Darner xima. S; Aune. M; Becker. W; Bierselius. R; Cnattingius. S; Glynn. A. (2003).
Polybrominated diphenyl ethers in breast milk from Uppsala County, Sweden. Environ Res 93:
186-194.
Lindberg. P; Sellstrom. LI; Haggberg. L; de Wit. CA. (2004). Higher brominated diphenyl ethers and
hexabromocyclododecane found in eggs of peregrine falcons (Falco peregrinus) breeding in
Sweden. Environ Sci Technol 38: 93-96. http://dx.doi.org/10.1021/es034614q.
Page 515 of 723
-------
Litton Bionetics. (1990). Mutagenicity evaluation of 421-32B (final report) with test data and cover
letter [TSCA Submission], (TSCATS/407257. OTS0523257. 86900000265). West Lafayette, IN:
Great Lakes Chemical Corporation.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS0523257.
Lm H i eng. W; Slun \\t X. i\io. Y; Wang. S: Fan. C; Wane. H; Zhang. H. (2010). The effect of
maternal subclinical hypothyroidism during pregnancy on brain development in rat offspring.
Thyroid 20: 909-915. http://dx.doi.org/10.1089/thv.2009.0Q36.
Lohmann. R. (2017). Microplastics are not important for the cycling and bioaccumulation of organic
pollutants in the oceans-but should microplastics be considered POPs themselves? Integr
Environ Assess Manag 13: 460-465. http://dx.doi.org/10.1002/ieai
Lower. N: Moore. A. (2007). The impact of a brominated flame retardant on smoltifi cation and olfactory
function in Atlantic salmon (Salmo salar L.) smolts. Mar Behav Physiol 40: 267-284.
http://dx.doi.on 80/102362.40701592.104.
Luigi useppe. M; CI audio. R. (2015). Emerging and priority contaminants with endocrine active
potentials in sediments and fish from the River Po (Italy). Environ Sci Pollut Res Int 22: 14050-
14066. http://dx.doi.orj -015-4388-8.
Lyons. BP; Barber. J.L; Ruroney. HS: Bolam. TP; Bersuder. P; Law. RJ; Mason. C; Smith. A.J; Morris.
S; Devlin. Ml; Al-Enezi. M; Massoud. h lan. AS; Al-Sarawi. HA. (2015). Baseline
survey of marine sediments collected from the State of Kuwait: PAHs, PCBs, brominated flame
retardants and metal contamination. Mar Pollut Bull 100: 629-636.
http://dx.doi.off h' hU * i.marpolbul.2015.0^ < U I
Luster. MI; Johnson. VI; Yucesov neonova. PP. (2005). Biomarkers to assess potential
developmental immunotoxicity in children. Toxicol Appl Pharmacol 206: 229-236.
http://dx.doi.on .taap.2005.02.010.
MacGregor. J A; Nixon. WB. (2004). Determination of water solubility of hexabromocyclododecane
(HBCD) using a generator column method. (Project No. 439C-138). Easton, MD: Wildlife
International Ltd.
Mac! (1991). Multimedia environmental models: the fugacity approach. Chelsea, MI: Lewis
Publishers.
Makris. S. (2016). [Personal communication with Sue Makris, EPA containing individual animal data
for Ema et al, 2008] [Personal Communication],
Managaki. S; Miyake. Y; Yokoyama. Y; Hondo. H; Masunaga. S; Nakai. S; Kobayashi. T; Kameya. T;
Kimura. A; Nakarai \ (» % {S.-5
Maron. DM; Ames. BN. (1983). Revised methods for salmonella mutagenicity test. Mutat Res Environ
Mutagen Relat Subj 113: 173-215. http://dx.doi.org, 10 I Ok- O.i H • s l j3)90010 ^
Marquart. H; Warrc itinen. J; Van Hemmen. JJ. (2006). Default values for assessment of
potential dermal exposure of the hands to industrial chemicals in the scope of regulatory risk
assessments. Ann Occup Hyg 50: 469-489. http://dx.doi.org/10.1093/annhyg/melQ12.
Page 516 of 723
-------
Marvin. CH; Tomy. GT; Alaee. M; Macinnis. G. (2006). Distribution of hexabromocyclododecane in
Detroit River suspended sediments. Chemosphere 64: 268-275.
http://dx.doi.org/10/1016/i .chemosphere.200>_ 12 _ 0 ] ]
Marteinson. SC: Bird. DM: Letcher. RJ; Sullivan.. KM: Ritchie. I J: Fernie. KJ. (2012). Dietary exposure
to technical hexabromocyclododecane (HBCD) alters courtship, incubation and parental
behaviors in American kestrels (Falco sparverius). Chemosphere 89: 1077-1083.
http://dx.doi. org/10.10 "16/i. ch em osph ere. 2012.05.073.
Marteinson. SC: Bird. DM: Shun _i]_ I etcher. RJ: Ritchie j* i-cmie. KJ. (2010). Multi-generational
effects of polybrominated diphenylethers exposure: embryonic exposure of male American
kestrels (Falco sparverius) to DE-71 alters reproductive success and behaviors. Environ Toxicol
Chem 29: 1740-1747. http://dx.doi.org/10.1002/etc.200.
Marteinson. SC: Kimmins. S: Letcher. RJ: Palac M: Ritchie. I J: Fernie. KJ. (2011). Diet
exposure to technical hexabromocyclododecane (HBCD) affects testes and circulating
testosterone and thyroxine levels in American kestrels (Falco sparverius). Environ Res 111:
1116-1123. http://dx.doi.o: /i.envres.2011.08.006.
Marvin. CH: Tomy. GT \imitage. JM, Ai run \ Mccarty. L: Covad. A; Palace. V. (2011).
Hexabromocyclododecane: current understanding of chemistry, environmental fate and
toxicology and implications for global management. Environ Sci Technol 45: 8613-8623.
http://dx.doi.on _ ;s201548c.
Maye> _ V. ^ i_. V\ Bcrtke. S: Horn. GP: Smith. PL: Kerber. S: La Gaurdia. Ml. (2019). Firefighter
hood contamination: Efficiency of laundering to remove PAHs and FRs. J Occup Environ Hyg
16:2: 129-140. http://dx.doi.Org/https://doi.org/l0.1080/1545962.4.2.018.1540877
Metier. L: Martiin. A; Melessen. I: Brouwer. A; Weiss. J: de Jong. FH: Sauei (2012). Influence of
prenatal organohalogen levels on infant male sexual development: sex hormone levels, testes
volume and penile length. Hum Reprod 27: 867-872. http://dx.doi.org ?3/humrep/der426.
Meiier. L: Weiss. J: Van Velzen. M: Brouwer. A: Bergman uer. PI. (2008). Serum concentrations
of neutral and phenolic organohalogens in pregnant women and some of their infants in The
Netherlands. Environ Sci Technol 42: 3428-3433. http://dx.doi.oo ;s702446p.
Meng .\ \ \ufflg. N: Duan tChen. L: Zeng. EY. (2012). Hexabromocyclododecane in consumer
fish from South China: implications for human exposure via dietary intake. Environ Toxicol
Chem 31: 1424-1430. http://dx.doi.org/10.1002/etc. 1826.
Microbiological Associates. (1996a). Hexabromocyclododecane (HBCD): Chromosome aberrations in
human peripheral blood lymphocytes with cover letter dated 12/12/1996 [TSCA Submission],
(TSCATS/453439. OTS0573552. Doc I.D. 86970000358). Arlington, VA: Chemical
Manufacturers Association.
https://ntrl.ntis. gov/.NTRL/dashboard/searchResults.xhtml?searchQuery=€)TS05
Microbiological Associates. (1996b). Hexabromocyclododecane (HBCD): Maximization test in guinea
pigs with cover letter dated 12/12/1996 [TSCA Submission], (86970000356). Arlington, VA:
Chemical Manufacturers Association.
https://ntrl.ntis. gov/.NTRL/dashboard/searchResults.xhtml?searchQuery=€)TS0573 550.
Miller-Rhodes. P: Popescu. M: Goeke. C: Tirabassi. T: Johnson. L: Markowski. VP. (2014). Prenatal
exposure to the brominated flame retardant hexabromocyclododecane (HBCD) impairs measures
of sustained attention and increases age-related morbidity in the Long-Evans rat. Neurotoxicol
Teratol 45: 34-43. http://dx.doi.or i.ntt.2014.06.009.
Minh. NH: Isobe. T: Ueno. D: Matsumoto. K: Mine. M: Kaiiw cahashi. S: Tanabe. S. (2007).
Spatial distribution and vertical profile of polybrominated diphenyl ethers and
hexabromocyclododecanes in sediment core from Tokyo Bay, Japan. Environ Pollut 148: 409-
417. http://dx.d0i.0rg/l /i.envpol.2006.12.011.
Page 517 of 723
-------
Minnesota Department of Health. (2016). Flame Retardants and Firefighter Exposure and Health. St.
Paul, MN: Minnesota Department of Health, Environmental Surveillance and Assessment
Section.
Morris. S; Allchin. €R; Ze'gers. BN: Haftka. ' ' ' )n. IP; Belpaire. C; Leonards. PE; Van Leeuwen. SP;
De Boer. J. (2004). Distribution and fate of HBCD and TBBPA brominated flame retardants in
North Sea estuaries and aquatic food webs. Environ Sci Technol 38: 5497-5504.
http://dx.doi.oo ;s049640i.
Mizouchi. S: Ichiba. M; Takigarni. H; Kaiiw arouku. T; Miyaiima. T; Kodaroa. H; Someva.
T; Ueno. D. (2015). Exposure assessment of organophosphorus and organobromine flame
retardants via indoor dust from elementary schools and domestic houses. Chemosphere 123: 17-
25. http://dx.doi.orp 10 j01 >. i.chemosphere.Iv" l i J l ^28.
MOEJ. (2009). 6-Week Administration Study of 1,2,5,6,9,10-Hexabromocyclododecane for avian
reproduction toxicity under long-day conditions using Japanese Quail. Tokyo, Japan: Ministry of
the Environment, Japan.
Momma. J; Kaniwa. M; Sekiguchi. H; Ohno. K; Kawasaki. Y; Tsuda. M; Nakamura. A; Kurokawa. Y.
(1993). [Dermatological evaluation of a flame retardant, hexabromocyclododecane (HBCD) on
guinea pig by using the primary irritation, sensitization, phototoxicity and photosensitization of
skin], Eisei Shikenjo Hokokul8-24.
Morf. LS; Tremp. J; Gloor. R; Huber. Y; Stengele. M; Zennegg. M. (2005). Brominated flame retardants
in waste electrical and electronic equipment: substance flows in a recycling plant. Environ Sci
Technol 39: 8691-8699. http://dx.doi.ore/10.1021/esO51170k.
Morose. G. (2006). An overview of alternatives to tetrabomobisphenol A (TBBPA) and
hexabromocyclododecane (HBCD). In Lowell Center for Sustainable Production, University of
Massachusetts, Lowell (pp. 1-32). Lowell, MA.: Prepared for the Jennifer Altman Foundation.
http://www.chemicalspolicv.org/downloads/AternativestoTBBPAandHB
Mov; illips. L. (2014). A review of soil and dust ingestion studies for children [Review], J Expo
Sci Environ Epidemiol 24: 545-554. http://dx.doi.org/10.1038/ies.
Multi-Panels. C. (2015). Material safety data sheet (MSDS): XPS boards. La J oil a, California: Multi-
Panels Corporation. http://www.miilti-panels.com/MSDSXPS-Boards.pdf.
Nakagawa. R; Murata. S; Ashizuk3 _\ '-.hintani. Y; Horn T; Tsutsumi J_ (2010).
Hexabromocyclododecane determination in seafood samples collected from Japanese coastal
areas. Chemosphere 81: 445-452. http://dx.doi.org/" >/i.chemosphere.2.010.08.015.
Newton. S; Seilstrom. II; de ¥ (2015). Emerging Flame Retardants, PBDEs, and HBCDDs in
Indoor and Outdoor Media in Stockholm, Sweden. Environ Sci Technol 49: 2912-2920.
http://dx.doi.on ;s505946e.
NICNAS. (2012a). Hexabromocyclododecane: Priority existing chemical assessment report no. 34.
Australia.
http://www.nicnas.gov.au/Publications/CAR/PEC/PEC34/I 11lij_»^''port June 20J2, JVJ)I j>!f.
NICNAS. (2012b). Priority existing chemical assessments: Hyexabromocyclododecane (HBCD)-
PEC34. https://www.nicnas.gov.au/chemical-information/pec-
assessm.en.ts?result 34' suit page=H.
NIOSH. (2014a). Evaluation of Occupational Exposures at an Electronic Scrap Recycling Facility.
(2012-0100-3217). Washington, DC: U.S. Department of Health and Human Services.
https://www.cdc.gov/niosh/hhe/reports/pdfs/
NIOSH. (2014b). A Pilot Assessment of Occupational Health Hazards in the U.S. Electronic Scrap
Recycling Industry (pp. 26). Washington, DC: U.S. Department of Health and Human Services.
¦cdc.gov/niosh/hhe/reports/pdfs/e-scrap survey
Page 518 of 723
-------
NIQSBL (2003). Respirator Usage in Private Sector Firms. Washington D.C.: United States Department
of Labor, Bureau of Labor Statistics and National Institute for Occupational Safety and Health.
https://www.cdc.gov/niosh/docs/respsurv/.
NRC. (2000a). H ex abromocyclododecane. In Toxicological risk of selected flame-retardant chemicals.
Washington, DC: National Academy Press.
http://www.nap.edii/openbook.php7record id=9841&paee=53.
NRC. (2000b). Toxicological risks of selected flame-retardant chemicals. Washington, DC: National
Academy Press, http://dx.doi.ore/10.17226/9841.
NRC. (2005). Health implications of perchlorate ingestion. Washington, DC: National Academies Press.
http://dx.doi. or e/fattp s: //dot. or e/10.17226/11202.
(2013). ODE isipan. Turkey: ODE Yalitim Sanayi ve Ticaret AS.
http://www.ode.com.tr/assets/upload/kalite belgeleri/ode-isipan-(en).pdf.
OECD. (2007). S1DS initial assessment profile: H ex ab rom ocy cl ododeca ne (HBCDD) [OECD S1DS],
Paris, France, http://webn.et.oecd.ore/HPV/LJI/handler.axd?id=ea58acl l-e090-41 31-
200a ;6c.
OECD. (2009). Emission Scenario Document on Plastic Additives. In Series on Emission Scenario
Documents No 3. Paris: OECD Environmental Health and Safety Publications.
http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=env/im/mono(2004)8/r
ev 1 & docl an guage=en.
OECD. (2010a). Emission Scenario Document on Chemicals Used in the Electronics Industry. In Series
on Emission Scenario Documents No 25. Paris: OECD Environmental Health and Safety
Publications.
http://www.oecd.org/officialdocurnents/publicdisplavdocurnentpdf/?cote=env/irn/mono(2010)37
& docl an guage=en.
OECD. (2010b). Emission Scenario Document on Formulation of Radiation Curable Coatings, Inks and
Adhesives. Paris: OECD Environmental Health and Safety Publications, http://www.oecd-
ilibrarv.org/docserver/download/9' I Jj 'J pdf?expires=14970J I 7 14&id=id&accname=guest&c
hecksurr )39B8643ABA020.
OECD. (2019). Estimating mouthing exposure in children-compi 1 ation of case studies. (Series on
Testing And Assessment No. 306).
http://www.oecd.org/officialdocuments/publicdisplavdocumentpdf/?cote=ENV-JM-
Oetken. M; Ludwichowskt. K: Nagel. R (2001). Validation of the preliminary EU-concept of assessing
the impact of chemicals to organisms in sediment by using selected substances. (UBA-FB 299 67
411). Dresden, Germany: Dresden University of Technology, Institute of Hydrobiology.
Oh. IK: Kotani. K; Managaki. S: Masunaga. S. (2014). Levels and distribution of
hexabromocyclododecane and its lower brominated derivative in Japanese riverine environment.
Chemosphere 109: 157-163. http://dx.doi.org 10 k«l i.chemosphere.2011 01 0 4.
Olukunle. 01: Okonkw (2015). Concentration of novel brominated flame retardants and HBCD in
leachates and sediments from selected municipal solid waste landfill sites in Gauteng Province,
South Africa. Waste Manag 43: 300-306. http://dx.d0i.0rg/l 0.1016/i .wasman.2015.07.009.
terra. P; Diaz-Ferrero. J: Eliarrat. E: Barcelo. D. (2011). Diastereoisomer- and enantiomer-
specific determination of hexabromocyclododecane in fish oil for food and feed. Chemosphere
82: 739-744. http://dx.doi.org/10 [01 ¦ 1 chemosphere.2010 j0 0_N8.
O'Shaughnessv. KL: Thomas. SE; Spring. SR; Ford. JL; Ford. RL; Gilbert. ME. (2019). A transient
window of hypothyroidism alters neural progenitor cells and results in abnormal brain
development. Sci Rep 9: 4662. http://dx.doi.org/i 0.1038/s41598-019-40249-7.
Page 519 of 723
-------
Ott. W; Klepeis. N: Switzer. P. (2008). Air change rates of motor vehicles and in-vehicle pollutant
concentrations from secondhand smoke. J Expo Sci Environ Epidemiol 18: 312-325.
http://dx.doi.on 38/si.ies.7500601.
Ozkavnak. H; Xue. J; Zartaria n. G; Smith. L. (2011). Modeled estimates of soil and dust
ingestion rates for children. Risk Anal 31: 592-608. http://dx.doi.ore/'
692.4.20
Palace. V; Park. B; Pleskach. K; Gemmill. B; Tomy. G. (2010). Altered thyroxine metabolism in
rainbow trout (Oncorhynchus mykiss) exposed to hexabromocyclododecane (HBCD).
Chemosphere 80: 165-169. http://dx.doi.ore/h1 u* i chemosphere. 2010 0' 0 [-•.
Palace. VP: Pleskach. K; Halldorson. T; Danell. R; Wautier. K: Evans aee. M: Marvin. C: Tomy.
GT. (2008). Biotransformation enzymes and thyroid axis disruption in juvenile rainbow trout
(Oncorhynchus mykiss) exposed to hexabromocyclododecane diastereoisomers. Environ Sci
Technol 42: 1967-1972. hm,. tk.doi.org/10.102 S/es702565h.
Paul. KB: Hedge. JM, Dcvito. Ml: Crofton. KM. (2010). Short-term exposure to triclosan decreases
thyroxine in vivo via upregulation of hepatic catabolism in Young Long-Evans rats. Toxicol Sci
113: 367-379. http://dx.doi.ors 93/toxsci/kfp271.
Pawa bdallah. MA; de Saa. EV: Harrad. S. (2016). Dermal bioaccessibility of flame retardants
from indoor dust and the influence of topically applied cosmetics. J Expo Sci Environ Epidemiol
27: 100-105. http://dx.doi.ore/ 8/ies.2015.84.
Pharmakoloeisches Institut. (1990). Ames test with hexabromides with cover letter dated 03 1290 [TSCA.
Submission], (TSCATS/406636. OTS0522942. Doc. I.D. 86900000379). Washington, DC: U.S.
Environmental Protection Agency.
Peled. M: Sch aria. R: Sondack (1995). Thermal rearrangement of hexabromo-cyclododecane
(HBCD). In JRDesmurs; B Gerard; MJ Godlstein (Eds.), (pp. 92-99). New York, NY: Elsevier.
http://dx.doi.ore K(»S)80<'L
Pom; alta. P: Roscioli. C: Bettinetti. R: Guzzella. L. (2014). Concentrations and trophic
interactions of novel brominated flame retardants, HBCD, and PBDEs in zooplankton and fish
from Lake Maggiore (Northern Italy). Sci Total Environ 481: 401-408.
http://dx.doi.on h1 t.scitotenv.2014.02.063.
Potte ardia. MJ: Harvey. EP: Hale. RC. (2009). Polybrominated diphenyl ether
flame retardants in Chesapeake Bay region, USA, peregrine falcon (Falco peregrinus) eggs:
urban/rural trends. Environ Toxicol Chem 28: 973-981. http://dx.doi.ore/10.1897/08-350.1.
Porch. JR: Kendall TZ: Krueeer. HQ. (2002). Hexabromocyclododecane (HBCD): A toxicity test to
determine the effects of the test substance on seedling emergence of six species of plants. (126
pp.). Easton, MD: Porch, JR; Kendall, TZ; Krueger, HO.
Rarou. K: Isobe I', I'jkahashi. S: Kim. EY; Min U\ . We. SU: Tanabe. S. (2010). Spatial distribution of
polybrominated diphenyl ethers and hexabromocyclododecanes in sediments from coastal waters
of Korea. Chemosphere 79: 713-719. http://dx.doi.ore/' ?/i.chem.osphere.2010.02.048.
Rani. M; Shim. WJ: Jane. M; Han. GM; Hone. SH. (2017). Releasing of hexabromocyclododecanes
from expanded polystyrenes in seawater -field and laboratory experiments. Chemosphere 185:
798-805. http: //dx. doi. or e/10.1016/i. ch em osph. ere .2.017.07.042
Ransbotyn. G. (1999). Use of HBCD in Flame-Retarded EPS Grades. Letter to KEMI from APME,
dated 30 April. 1999.
Ransbotyn. G (2000). Input to Ell risk assessment on HBCDD from users in EPS and HIPS. Report
prepared by the EPS and PS Comittees of the APME. Dated 18 February (pp. 25). Brussels,
Belgium: Ransbotyn, G.
Rasineer. JD; Carroll. TS: Maranehi. F; Tassinari. R; Moracci. G: Altieri. I: Mantovani. idebye.
AK; Hoestrand. C. (2018). Low dose exposure to HBCD, CB-153 or TCDD induces
histopathological and hormonal effects and changes in brain protein and gene expression in
Page 520 of 723
-------
juvenile female BALB/c mice. Reprod Toxicol 80: 105-116.
http://dx.doi.on . reprotox. 2.018.06.010.
Rawn. DF; Gaertner. DW; Weber. D; Curran. IH; Cooke. GM; Goodyer. CG. (2014a).
Hexabromocyclododecane concentrations in Canadian human fetal liver and placental tissues.
Sci Total Environ 468-469: 622-629. http://dx.doi.o>g ^i/i.scitotenv.2013.VS I
Rawn. DF; Ryan. II; Sadler. AR; Sun. WF; Webei i * 1 > J005.09718.X.
Robson. M; Melym.uk. L; Bradley. L; Treen. B; Backus. S. (2013). Wet deposition of brominated flame
retardants to the Great Lakes basin - Status and trends. Environ Pollut 182: 299-306.
http://dx.doi.oii 10 k.J: i.envpol.2010' OlS.
Rochman. CM; Hoi irobe. T; Teh. SI. (2013). Ingested plastic transfers hazardous chemicals to
fish and induces hepatic stress. Sci Rep 3: 3263. http://dx.doi.org/ j/srep03263
Rochman. CM j_ cwlson. RL; Eriksen. M; Allen. H; Cook. AM; Teh. SI. (2014). Polybrominated
diphenyl ethers (PBDEs) in fish tissue may be an indicator of plastic contamination in marine
habitats. Sci Total Environ 476-477: 622-633. http://dx.doi.org 10 101 • i.scitotenv.201 4 01 0^8
Roman. GC; Ghassabian. A; Bongers-Schokking. II; Jaddoe. VW; Hofman. A; deRiike. YB; Verhulst.
FC; Tiemeier. H. (2013). Association of gestational maternal hypothyroxinemia and increased
autism risk. Ann Neurol 74: 733-742. http://dx.doi.org/10.1002/ana.23976.
Ronisz. D; Finne. EF; Karlsson. H; Forlin. L. (2004). Effects of the brominated flame retardants
hexabromocyclododecane (HBCDD), and tetrabromobisphenol A (TBBPA), on hepatic enzymes
and other biomarkers in juvenile rainbow trout and feral eelpout. Aquat Toxicol 69: 229-245.
Roosens. L; Abdallah. Ml Harrad. S; Neels. H; Covaci. A. (2009). Exposure to
hexabromocyclododecanes (HBCDs) via dust ingestion, but not diet, correlates with
concentrations in human serum: preliminary results. Environ Health Perspect 117: 1707-1712.
http://dx.doi.org/10.1289/ehp.0900869.
Roosens. L; Cometis. C; D'Hollander. W; Bervoets. L; Reyndeis H, \ ;m 1
-------
Rosenberg. C; Hameila. M; Tornaeus. J; Sakkinen. K; Puttonen. K; Korpi ilunen. M; Linnainmaa.
M; Hesso. A. (201 1). Exposure to flame retardants in electronics recycling sites. Ann Occup Hyg
55: 658-665. http://dx.doi.ore/10.1093/aimhve/mer033.
Rosol. TJ: DeLellis. RA: Harvey. PW: Sutcliffe. C. (2013). Endocrine system. In W Haschek; C
Rousseaux; M Wallig (Eds.), (3rd ed., pp. 2391-2492). Waltham, MA: Academic Press.
http://dx.doi.ore/lO t0lo/B97^ o l_ I! - - o
-------
SCCH. (2018a). Amendments to Annexes to the Stockholm Convention. Available online at
http://chm.pops.int/Countries/StatusofRatifications/Amendmentstoannexes/tabid/3486/Default.as
E,
SCCH. (2018b). Import and export of POPs. Available online at
http://chm.pops.int/Implementation/ImportandexportofPOPs/Overview/tabid/4455/Default.aspx
SCCH. (2018c). Register of Specific Exemptions. Available online at
http://chro.pGps.int/Iropleroentation/ExeroptiQns/RegistefQrSpeciftcExeroptions/tabii mi
lt.aspx
Schecter. A; Szabo. DT; Mille :nt. TL: Malik-Bass. N: Petersen. M; Paepke. O: Colacino. JA:
Hynan. LS: Harris. TR; Malta, S: Birnbaum. LS. (2012). Hexabromocyclododecane (HBCD)
Stereoisomers in U.S. Food from Dallas, Texas. Environ Health Perspect 120: 1260-1264.
http://dx.doi.Org/10.1289/e 4993.
Schluroroer. M; Maures agner. S: Berrang. A; Siebert. T; Knappich. F. (2017). Recycling of flame
retarded waste polystyrene foams (EPS and XPS) to PS granules free of
hexabromocyclododecane (HBCDD). Schlummer, M; Maurer, A; Wagner, S; Berrang, A;
Siebert, T; Knappich, F.
http://www.svnbratechnology.eom/media/l 1693/fraunhoferwebsite creasolv-processing-of-
hbcd-containing-polvstvrene-from-construction-eps-l.pdf.
Schreder. ED: La Guardia. Ml. (2014). Flame retardant transfers from U.S. households (dust and
laundry wastewater) to the aquatic environment. Environ Sci Technol 48: 11575-11583.
http://dx.doi.on ;s502227h.
Schriks. M; Vrabie. CM: Gutleb \C Kiassen i, Kiciiens. IM: Murk. AJ. (2006). T-screen to quantify
functional potentiating, antagonistic and thyroid hormone-like activities of poly halogenated
aromatic hydrocarbons (PHAHs). Toxicol In Vitro 20: 490-498.
http://dx.doi.on .tiv.2005.09.001.
Schriks. \ I \ mavashe. E: Furlou. JI \k Murk. AJ. (2006). Disruption of thyroid hormone-mediated
Xenopus laevis tadpole tail tip regression by hexabromocyclododecane (HBCD) and
2,2',3,3',4,4',5,5',6-nona brominated diphenyl ether (BDE206). Chemosphere 65: 1904-1908.
http://dx.doi.on .chemosphere.2006.07.077
Searl. A; Robertson. A. (2005). Workplace exposure to hexabromocyclododecane (HBCD) in the
European Union.Report for the European Brominated Flame Retardant Industry Panel. Searl, A;
Robertson, A.
Sellstrom. LI; Kierkkeg 'it. C; Jansson. B. (1998). Polybrominated diphenyl ethers and
hexabromocyclododecane in sediment and fish from a Swedish River. Environ Toxicol Chem
17: 1065-1072. http://dx.doi.o 8( 1998)017< 1065:PDEAHI>2.3,CO:2.
Shaw. SD; Berger. ML: Weiis. L: Covaci. A. (2012). Tissue-specific accumulation of polybrominated
diphenyl ethers (PBDEs) including Deca-BDE and hexabromocyclododecanes (HBCDs) in
harbor seals from the northwest Atlantic. Environ Int 44: 1-6.
http://dx.doi.Org/10.1016/i .envint. - 1 L-i) s < «0!
Shaw. SD: Berg irris. JH; Yun. SH: Wu. O; Liao. C; Blum. A; Stefani. A; Kannan. K. (2013).
Persistent organic pollutants including polychlorinated and polybrominated dibenzo-p-dioxins
and dibenzofurans in firefighters from Northern California. Chemosphere 91: 1386-1394.
http://dx.doi. org/http ://dx. doi. or g/ _ ;hemosphere.2012.12.070
Shelby. MK; Cherrington. NJ; Vansell. NR; Klaassen. CD. (2003). Tissue mRNA expression of the rat
UDP-glucuronosyltransferase gene family. Drug Metab Dispos 31: 326-333.
http://dx.doi.on iro.d.31.3.326.
Shen. B; Whitehead. TP; Gill. R; Dhaliwal. J; Brown. FR; Petreas. M; Patton. S; Hammond. SK. (2018).
Organophosphate flame retardants in dust collected from United States fire Stations. Environ Int
112: 41-48. http://dx.doi.Org/https://doi.oii h' l
-------
S Ju. W; Shen. R; Li. D; Hone. H. (2017a). Accumulation and developmental toxicity of
hexabromocyclododecanes (HBCDs) on the marine copepod Tigriopus japonicus. Chemosphere
167: 155-162. http://dx.doi.or i.chemosphere.2016.09.160.
Shi. Z; Zhang. L; Zhao \ Vn. Z: Zhou. X; Li. I; Wu. Y. (2017b). Dietary exposure assessment of
Chinese population to tetrabromobisphenol-A, hexabromocyclododecane and decabrominated
diphenyl ether: Results of the 5th Chinese total diet study. Environ Pollut 229: 539-547.
http://dx.doi. ore/ "10.10 "16/i. envpol.2017.06.093.
Shoeib. M; Ahrens. L; J an tun en. L; Burner. T. om. (2014). Concentrations in air of organobromine,
organochlorine and organophosphate flame retardants in Toronto, Canada. Atmos Environ 99:
140-147. http://dx.doi.ore/10 1016/i.atM.osenv.201 I 0'\040.
Shoeib. M; Hamer. T; Webster. GM; Sverkc eng. Y. (2012). Legacy and current-use flame
retardants in house dust from Vancouver, Canada. Environ Pollut 169: 175-182.
http://dx.doi. ore/10.1016/i. envpol.2012.01.043.
Shoeib. M; Hamer. T; Wilford. BH; Jones. KC; Zhu. J. (2005). Perfluorinated sulfonamides in indoor
and outdoor air and indoor dust: occurrence, partitioning, and human exposure. Environ Sci
Technol 39: 6599-6606. http://dx.doi.ors :s048340y.
Smith. K; Liu. €H; El-Hiti. GA; Kan^ * ' 5 C lament. SG; Checquei Howarth. OW;
Hursthouse. MB; Coles. SI. (2005). An extensive study of bromination of cis,trans,trans-1,5,9-
cyclododecatriene: product structures and conformations. Org Biomol Chem 3: 1880-1892.
Smolarz. K; Berger. A. (2009). Long-term toxicity of hexabromocyclododecane (HBCDD) to the
benthic clam Macoma balthica (L.) from the Baltic Sea. Aquat Toxicol 95: 239-247.
http://dx.doi.on .aqiiatox.2009.09.010.
Son. MB; Kim. J: Shin. ES; Seo. SH; Chang. YS. (2015). Diastereoisomer- and species-specific
distribution of hexabromocyclododecane (HBCD) in fish and marine invertebrates. J Hazard
Mater 300: 114-120. http://dx.doi.org/10.1016/i jfa.azTO.at.2015.06.023.
Song. N: Li. L; Li. H; Ai. W; Xie. W; Yu. W; Liu. W; Wane. C: Shen. G; Zhou. L; Wei. C; Li. D: Chen.
H. (2016). Single and 14-day repeated dose inhalation toxicity studies of
hexabromocyclododecane in rats. Food Chem Toxicol 91: 73-81.
fattp://dx.doi.orj i.fct.2016.02.020.
SRI International. (1990). In vitro microbiological mutagenicity studies of four Ciba-Geigy Corporation
compounds (final report) with test data and cover letter [TSCA Submission], (TSCATS/407254.
OTS0523254. Doc I.D. 86900000262). West Lafayette, IN: Great Lakes Chemical Corporation.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuery=OTS0523254.
Stapleton. H; Allen. I; Kelly. S: Konstantinc Llosterhaus. S; Watkins. D; Mcclean. M; Webster. T.
(2008). Alternate and new brominated flame retardants detected in U.S. house dust. Environ Sci
Technol 42: 6910-6916. http://dx.doi.ors ;s801070p.
Stapleton. HM; Misenheimer. J; Hoffman. K; Webster. TF. (2014). Flame retardant associations
between children's handwipes and house dust. Chemosphere 116: 54-60.
http://dx.doi.on 1^ 10J_* i.chemosphere.. h »i>
Stiborova. H; Kolar. M; Vrkoslavova < 1'iHkrabova. J; Haisloui j. Demnerova. U 4 hlik. Q. (2017).
Linking toxicity profiles to pollutants in sludge and sediments. J Hazard Mater 321: 672-680.
http://dx.doi.ore/10.1016/i jhazmat.201 o 01051.
Stubbing trrad. S. (2019). Laboratory studies on leaching of HBCDD from building insulation
foams. Emerging Contaminants 5: 36-44.
Sun. J; Dai. X; Wane. O: van. Loosdrecht. MC: Ni. BJ. (2019). Microplastics in wastewater treatment
plants: Detection, occurrence and removal. Water Res 152: 21-37.
Sun. YX: Luo. XI: Mo. L; He. MI; Zhane. O; Chen. SI: Zou. FS; Mai. BX. (2012).
Hexabromocyclododecane in terrestrial passerine birds from e-waste, urban and rural locations in
Page 524 of 723
-------
the Pearl River Delta, South China: levels, biomagnification, diastereoisomer- and enantiomer-
specific accumulation. Environ Pollut 171: 191-198.
http://dx.doi.on .envpol.2012.07.026.
Sustainable Electronics Recycling. I. (2020). Find a Recycler [Website],
https://siistain.ableelectron.ics.ore/recYclers.
Suter (2016). Weight of evidence in ecological assessment. (EPA 100R16001). Washington, DC:
U.S. Environmental Protection Agency.
https://cfpub.epa.eov/si/si public record report. cfm?dirEntryId=!
Szabo. DT; Diliberto. 11; Hakk. H; Huwe. IK; Birnbaum. LS. (2010). Toxicokinetics of the flame
retardant hexabromocyclododecane gamma: effect of dose, timing, route, repeated exposure, and
metabolism. Toxicol Sci 117: 282-293. http://dx.doi.ori 93/toxsci/kfql83.
Szabo. DT; Diliberto. II: Hakk. H; Huwe. IK; Birnbaum. LS. (201 la). Toxicokinetics of the flame
retardant hexabromocyclododecane alpha: effect of dose, timing, route, repeated exposure, and
metabolism. Toxicol Sci 121: 234-244. http://dx.doi.ori 93/toxsci/kfr059.
Szabo. DT; Diliberto. II; Huwe. IK; Birnbaum. LS. (201 lb). Differences in tissue distribution of HBCD
alpha and gamma between adult and developing mice. Toxicol Sci 123: 256-263.
http://dx.doi.ore h« i^93/toxsci/fatto I
Takieami. H; Suzuki •„ ' Krai \ . hliikawa. Y; Sun ami. M; Sakai. S. (2009). Flame retardants in indoor
dust and air of a hotel in Japan. Environ Int 35: 688-693.
http://dx.doi.on .envint.2008.12.007.
Takieami. H; Suzuki •„ ' Krai \ _ 'v-ikai. S. (2008). Transfer of brominated flame retardants from
components into dust inside television cabinets. Chemosphere 73: 161-169.
http://dx.doi.ore/l 0.1016/j .chemosphere.2008.06.032.
Takieami. H; Watanabe. M; Kaiiwara. N. (2014). Destruction behavior of hexabromocyclododecanes
during incineration of solid waste containing expanded and extruded polystyrene insulation
foams. Chemosphere 116: 24-33. http://dx.doi.ore/10 101 o/i.chemosphere.2.01 -1 01 082..
Tang J J » » \ > > G. (2014). Levels of flame retardants HBCD, TBBPA and TBC in surface
soils from an industrialized region of East China. Environ Sci Process Impacts 16: 1015-1021.
http://dx.doi.on ;3em00656e.
Tan I Vniu-EhxaM \Mallah. M; Ash worth. DC; Douglas. P; Toledano. MB; Harrad. S. (2017).
Emerging and legacy flame retardants in UK human milk and food suggest slow response to
restrictions on use of PBDEs and HBCDD. Environ Int 105: 95-104.
http://dx.doi.ore KMo i.envint.;0i 0^010
Tao. F; Matsukami. H; Suzuki U i ue. NM; Viet. HI. l'
-------
Prudente. M; Boonyatumanond. R; Zakaria. MP; Akkhavong. K; Oga lirai. H; Iwasa. S;
Mizukawa. K; Hagino. Y; Imamura. A; Safaa. M; Takada. H. (2009). Transport and release of
chemicals from plastics to the environment and to wildlife [Review], Philos Trans R Soc Lond B
Biol Sci 364: 2027-2045. http://dx.doi.org/10.1098/rstb.2008.Q284
Thomsen. C; Molander. P; Daae. HL; Janak. K; Froshaug. M; Liane. VH; Thornd. S; Beche
Dvbi (2007). Occupational exposure to hexabromocyclododecane at an industrial plant.
Environ Sci Technol 41: 5210-5216. http://dx.doi.org/10.1021/es0702622.
Thomsen. C; Stigum. H; Freshaug. M; Broadwell. SL; Becher. G; Eggesb.0. M. (2010). Determinants of
brominated flame retardants in breast milk from a large scale Norwegian study. Environ Int 36:
68-74. hi 11'; Mv.doi.org/10 tOk'/i.envint.200'- \0 002.
Tian. S; Sebroski. J; Ecoff. S. (2017). ASTM STP 1589 — Developing Consensus Standards for
Measuring Chemical Emissions from Spray Polyurethane Foam (SPF) Insulation Predicting
TCPP Emissions and Airborne Concentrations from Spray Polyurethane Foam Using USEPA i-
SVOC software: Parameter Estimation and Result Interpretation. West Conshohocken, PA:
ASTM International.
Tickner. 1 } j utr. J; Creelv. KS; Cherrie. JW; Pryde . DE; Kingston. J. (2005). The Development of the
EASE Model. Ann Occup Hyg 49: 105-110.
Tomko. G; Mcdonald. KM. (2013). Environmental fate of hexabromocyclododecane from a new
Canadian electronic recycling facility. J Environ Manage 114: 324-327.
http://dx.doi.oii 10 [01 > i.ienvman.2011 j0 0J4.
Toms. LM; Guerra. P; Elian a trcelo. D; Harden. FA; Hob son. P; Siodin. A; Ryan. E; Mueller JF
(2012). Brominated flame retardants in the Australian population: 1993-2009. Chemosphere 89:
398-403. http://dx.doi.Org/l0.1016/j.chemosphere.2012.05.053.
Tomy. GT; Budakowski. W; Halldorson. T; Whittle. DM; Keir. Ml; Marvin. C; Macinnis. G; Alaee. M.
(2004). Biomagnification of alpha- and gamma-hexabromocyclododecane isomers in a Lake
Ontario food web. Environ Sci Technol 38: 2298-2303. http://dx.doi.org/ l/es034968h.
Tomy. GT; Pteskach. K; Oswald. T; Halldorson. T; Helm. PA; Macinnis. G; Marvin. CH. (2008).
Enantioselective bioaccumulation of hexabromocyclododecane and congener-specific
accumulation of brominated diphenyl ethers in an eastern Canadian Arctic marine food web.
Environ Sci Technol 42: 3634-3639. http://dx.doi.org/10.1021/es703083z.
Townsend. T; Ingwersen. W; Niblick. B; Jain. P; Watty. I. (2019). CDDPath: A method for quantifying
the loss and recovery of construction and demolition debris in the United States. Waste Manag
84: 302-309. http://dx.doi.org/ 10 KM j wasmati -VIS I I 0 IN
;y (2014). Employee Tenure News Release. Available online at
http://www.bts.gov/news.retease/archives/teniire 0918: m
U.S. BLS. (2016). May 2016 Occupational Employment and Wage Estimates: National Industry-
Specific Estimates [Website], http://www.bts.gov/oes/tabtes.htm.
(1987). Development Document for Effluent Limitations Guidelines and Standards for the
Organic Chemicals, Plastics and Synthetic Fibers; Point Source Category, Volume I.
(440187009A). http://nepis.epa. gov/exe/ZyPURL.cgi?Dockey=Q0Q01BGD.txt.
(1991). Guidelines for developmental toxicity risk assessment (pp. 1-71). (EPA/600/FR-
91/001). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=23162.
(1992). Guidelines for exposure assessment. Federal Register 57(104):22888-22938 [EPA
Report], (EPA/600/Z-92/001). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisptay.cfm.7dei 53.
1, !:_} iA (1993a). Determination of rates of reaction in the gas-phase in the troposphere. 5. Rate of
indirect photoreaction: Evaluation of the atmospheric oxidation computer program of Syracuse
Research Corporation for estimating the second-order rate constant for the reaction of an organic
Page 526 of 723
-------
chemical with hydroxyl radicals. (EPA-744-R-93-001). Office of Pollution Prevention and
Toxics.
L EJL (1993b). Wildlife exposures factors handbook. Appendix: literature review database, vol 2
[EPA Report], (EPA/600/R-93/187b). Washington, D.C.
https://ntrl.ntis. eov/NT ;hb oard/ search Results.xhtm 1 ? search Query=PB94177789.
(1994). Methods for derivation of inhalation reference concentrations and application of
inhalation dosimetry [EPA Report], (EPA/600/8-90/066F). Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Research and Development, Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office.
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=71993^ 829&CFTOKEN=2
5006J I 7.
(1995). Final Water Quality Guidance for the Great Lakes System. 60: 15366-15422.
(1996). Guidelines for reproductive toxicity risk assessment (pp. 1-143). (EPA/630/R-
96/009). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.gov/sites/production/files/201 l_-1 I u^cuments/guidelines ret-n^ voricitv.pdf.
EPA. (1998a). Guidelines for ecological risk assessment [EPA Report], (EPA/630/R-95/002F).
Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.gov/risk/guidelines-ecological-risk-assessm.ent.
L EJL (1998b). Characterization of building-related construction and demolition debris in the
United States. Franklin Associates, https://www.epa.gov/smm/characterization-building-related-
construction-and-demoliti on-debris-united-states.
(2000a). Methodology for deriving ambient water quality criteria for the protection of human
health (2000). (EPA/822/B-00/004). Washington, DC: U.S. Environmental Protection Agency,
Office of Water, http://www.epa.gov/waterscience/criteria/humanhealth/method/complete.pdf.
(2000b). Science policy council handbook: Risk characterization [EPA Report], (EPA/100/B-
00/002). Washington, D.C.: U.S. Environmental Protection Agency, Science Policy Council.
https://www.epa.gov/risk/risk-characterization-handbook.
EPA. (2002). A review of the reference dose and reference concentration processes (pp. 1-192).
(EPA/630/P-02/002F). Washington, DC: U.S. Environmental Protection Agency, Risk
Assessment Forum, http://www.epa.gov/osa/review-reference-dose-and-reference-concentration-
processes.
EPA. (2003a). Methodology for deriving ambient water quality criteria for the protection of human
health (2000), technical support document. Volume 2: Development of national bioaccumulation
factors [EPA Report], (EPA-822-R-03-030). Washington, DC.
http://www.epa.gov/waterscience/criteria/humanhealth/method/tsdvol2.pdf.
U.S. EPA. (2003b). Estimating 2003 building-related construction and demolition materials amounts.
https://www.epa.gOv/smm/estimating-2003-building-related-construction-an.d-demolition-
m. ateri als-am cunts.
L S J iA (2004). Risk Assessment Guidance for Superfund (RAGS), Volume I: Human health
evaluation manual, (part E: Supplemental guidance for dermal risk assessment): Final.
(EPA/540/R/99/005). Washington, DC.
http://www.epa.gov/oswer/riskassessment/ragse/in.dex.htm.
(2005). Guidelines for carcinogen risk assessment [EPA Report] (pp. 1-166). (EPA/630/P-
03/001F). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
http://www2.epa.gov/osa/giiidelines-carcinogen.-risk-assessment.
(2006). A framework for assessing health risk of environmental exposures to children (pp. 1-
145). (EPA/600/R-05/093F). Washington, DC: U.S. Environmental Protection Agency, Office of
Research and Development, National Center for Environmental Assessment.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm.7dei 563.
Page 527 of 723
-------
U.S. EPA. (2008a). Initial risk-based prioritization of high production volume chemicals:
Chemical/category: Hexabromocyclododecane (HBCD), CAS 3194-55-6 1,2,5,6,9,10
hexabromocyclododecane, CAS 25637-99-4 hexabromocyclododecane. Washington, DC.
U.S. EPA. (2008b). Lifecycle construction resource guide. Pollution Prevention Program Office, Office
of Policy and Management, EPA Region 4.
https://nepis.epa.gov/Exe/ZyPURL.cgi7DockeY Mill .TXT.
U.S. EPA. (2009a). Air quality criteria for particulate matter.
U.S. EPA. (2009b). Consolidated Human Activity Database [Website], http://www.epa.gov/chadnet 1 /.
U.S. EPA. (2009c). User's guide and technical documentation: KABAM version 1.0 (Kow (based)
Aquatic BioAccumulation Model). https://www.epa.gov/sites/production/files/2015-
07/documents/kab; isers guide.pdf.
U.S. EPA. (2010). Hexabromocyclododecane (HBCD) action plan summary [EPA Report], Washington,
D C. https://www.epa.gov/assessing-and-managing-chemicals-under-
tsca/h exabrom ocvcl ododecan e-hb cd-acti on -pi an.
U.S. EPA. (201 la). EPI Suite results for CASRN 3194-55-6. Download EPI Suite TM v4.0. Available
online at
U.S. EPA. (201 lb). Exposure Factors Handbook: 201 1 Edition. (EPA/600/R-09/052F). Washington,
DC: U.S. EPA. http://www.epa.gov/ncea/efh.
U.S. EPA. (201 lc). Recommended use of body weight 3/4 as the default method in derivation of the oral
reference dose (pp. 1-50). (EPA/100/R11/0001). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum, Office of the Science Advisor.
https://www.epa.gov/risk/recommended-use-bodv-weight-34-default-method-derivation-oral-
reference-dose.
U.S. EPA. (2012a). Benchmark dose technical guidance. (EPA/100/R-12/001). Washington, DC: U.S.
Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.gov/risk/benchmark-dose4echnical-guidance.
U.S. EPA. (2012b). Estimation Programs Interface Suite™ for Microsoft® Windows, v 4.1 1 [Computer
Program], Washington, DC. Retrieved from https://www.epa.gov/tsca-screening-tools/epi-
suitetm-estim ation-program -interface
U.S. EPA. (2012c). Sustainable futures P2 framework manual [EPA Report], (EPA-748-B 12-001).
Washington DC. http://www.epa.gov/sustainable-futures/sustainable-futures-p2-framework-
manual.
U.S. EPA. (2012d). TSCA work plan chemicals: Methods document (pp. 1-28). Washington, DC: U.S.
Environmental Protection Agency, Office of Pollution Prevention and Toxics.
https://www.epa.gov/sites/production/files/201 ! -
03/documents/workplan methods document web final.pdf.
U.S. EPA. (2013a). ChemSTEER User Guide - Chemical Screening Tool for Exposures and
Environmental Releases. Environmental Protection Agency.
https://www.epa.gov/sites/production/files/2Q15-Q5/documents/user guide.pdf.
U.S. EPA. (2013b). Chem STEER User Guide: Chemical Screening T ool for Exposures and
Environmental Release. U.S. Environmental Protection Agency (EPA).
U.S. EPA. (2013c). Interpretive assistance document for assessment of discrete organic chemicals.
Sustainable futures summary assessment [EPA Report], Washington, DC.
http://www.epa.gov/sites/production/files/2Q15-Q5/documents/05-iad discretes iune2013.pdf.
U.S. EPA. (2014a). Draft Emission Scenario Document (ESD) on the Use of Additives in Plastic
Compounding. Chantilly, VA: Eastern Research Group.
U.S. EPA. (2014b). Estimated Fish Consumption Rates for the U.S. Population and Selected
Subpopulations (NHANES 2003-2010) (pp. 110). (EPA-820-R-14-002). Washington, DC.
https://www.epa.gov/sites/production/files/2Q15-Ql/documents/fish-consumption-rates-
Page 528 of 723
-------
2 f.https://hero.epa.gov/hero/index.cfm?action=search.view&reference id=4565445U.S.
EPA. (2014d). Flame retardant alternatives for hexabromocyclododecane (HBCD) [EPA
Report], (EPA/740/R-14/001). Washington, D.C. http://www2.epa.gov/saferchoice/part.nership-
evaluate-flame-retardant-alternatives-hbcd-publications.
(2014c). Exposure and Fate Assessment Screening Tool Version 2014 (E-FAST 2014).
Washington, DC: Office of Pollution Prevention and Toxics, https://www.epa.gov/tsca-
screening-tools/e-fast-exposure-and-fate-assessment-screening-tool-version--01 !
EPA. (2014d). Flame retardant alternatives for hexabromocyclododecane (HBCD) [EPA Report],
(EPA/740/R-14/001). Washington, D.C. http://www2.epa.gov/saferchoice/partnersfaip~evatuate~
fl am e~retardant~al tern ative s~hb cd-publi cati on s
L (2014e). Framework for human health risk assessment to inform decision making. Final
[EPA Report], (EPA/100/R-14/001). Washington, DC: U.S. Environmental Protection, Risk
Assessment Forum, fattps://www.epa.gov/risk/framework-human~health~risk~assessm.ent~inform~
decision-making.
1. !:_} L.lv (2014f). Preliminary materials for the Integrated Risk Information System (IRIS)
Toxicological review of hexabromocyclododecane (HBCD). In National Center for
Environmental Assessment, Office of Research and Development (pp. 84). (EPA/63 0/R-13/23 5).
Washington, DC: U.S. Envrionmental Protection Agency.
https://ofmpub.epa. gov/eims/eimscomm. getfile?p download id=521738.
EPA. (2015a). TSCA Work Plan chemical problem formulation and initial assessment. Cyclic
aliphatic bromide cluster flame retardants. (EPA 743-D1-5001). Washington, DC.
https://www.epa.gOv/sites/prodiiction/files/2015-09/dociim.en.ts/hbcd problem formulation.pdf.
L S J iA (2015b). Standard Operating Procedure for Using the NAFTA Guidance to Calculate
Representative Half-life Values and Characterizing Pesticide Degradation.
https://www.epa.gov/sites/production/files/2015-
08/documents/ftt sop using nafta guidance version2.pdf
L .1'(2016a). Implementation Study of the Electronics Recycling Standards: R2 and e-Stewards®
02/documents/u s epa fact sheet implementation study l.pdf.
(2016b). Instructions for reporting 2016 TSCA chemical data reporting. Washington, DC:
Office of Pollution Prevention and Toxics, https://www.epa.gov/chemical-data-
reporting/instructions-reporting-2016-tsca-chemical-data-reporting.
(2016c). Non-confidential 2016 Chemical Data Reporting (CDR) Database.
(2016d). Public database 2016 chemical data reporting (May 2017 release). Washington, DC:
US Environmental Protection Agency, Office of Pollution Prevention and Toxics. Retrieved
from https://www.epa.gov/chemical-data-reporting
(2016e). Technical review of hexabromocyclododecane (HBCD) CAS registry numbers
3194-55-6 and 25637-99-4. Washington, DC: U. S. Environmental Protection Agency.
http s: //www, re gul ati on s. gov/docum e sPA~H.Q~TR.I-2015-0607-0028.
(2017a). 1,4-dioxane (CASRN: 123-91-1) bibliography: Supplemental file for the TSCA
Scope Document [EPA Report], https://www.epa.gOv/sites/prodiiction/files/2
06/docum ents/14di oxan e comp bib.pdf.
L S EPA. (2017b). Cyclic aliphatic bromides cluster (HBCD) (CASRN: 25637-99- 4; 3194-55-6; 3194-
57-8) bibliography: Supplemental file for the TSCA Scope Document [EPA Report],
https://www.epa.gOv/sites/production./files/2. /docum.en.ts/hbcd comp bib.pdf.
(2017c). Re: Phase Down of HBCD by Dow Chemical, EMAIL from Christine Lukas to Sue
Slotnick. https://www.regulations.gov/document?D=EPA-HQ-OP 06.
Page 529 of 723
-------
(2017d). Scope of the Risk Evaluation for cyclic aliphatic bromides cluster [EPA Report],
(EPA-740-R1-7002). https://www.epa.gov/sites/production/files/2017-
06/documents/hbcd scope 06-:2f^11.
U.S. (2017e). Scope of the Risk Evaluation for Pigment Violet 29 (Anthra[2,1,9-def: 6,5,10-
d'e'f]diisoquinoline-l,3,8,10(2H,9H)-tetrone): CASRN: 81-33-4 [EPA Report], (740-R1-7011).
U.S. EPA, Office of Chemical Safety and Pollution Prevention, Office of Pollution Prevention
and Toxics, https://www.epa.gov/sites/prodiiction/files/ '6/documents/pv29 scope 06-22-
(2017f). Strategy for conducting literature searches for cyclic aliphatic bromine cluster
(HBCD): Supplemental document to the TSCA Scope Document. CASRN: 25637-99-4; 3194-
55-6; 3194-57-8 [EPA Report], https://www.epa.gov/sites/prodiiction/files/2017-
(2017g). Toxics Release Inventory (TRI) basic plus data file, Hexabromocyclododecane
(CAS # 25637-99-4), reporting year 2017. Retrieved from https://www.epa.gov/toxics-release-
inventorv-tri-program/tri-basic-plus-data-files-calendar-vears-1987-2017
EPA. (2017h). Toxics Release Inventory (TRI), reporting year 2015. https://www.epa.gov/toxics-
release-inventorv-tri-program/tri-data-and-tools.
1 c. 1 r \ (2017i). Update for chapter 5 of the Exposure Factors Handbook: Soil and Dust Ingestion
[EPA Report], (EPA/600R-17/384F). Washington, DC: National Center for Environmental
Assessment, Office of Research and Development.
h ttp s: //cfpub. epa. gov/n cea/efp/recordi spl ay. cfm ? dei d=3 3 7521.
(2017j). Use and market profile for hexabromocyclododecane (HBCD). Draft.
(2018a). Advancing sustainable materials management: 2015 fact sheet.
07/documents/2 nm msw factsheet 072.42018 fnl 508 002.pdf.
a * \ (2018b). Application of systematic review in TSCA Risk Evaluation s. (740-P1-8001).
Washington, DC: U.S. Environmental Protection Agency, Office of Chemical Safety and
06/documents/final application of sr in ts* df.
(2018c). Application of systematic review in TSCA Risk Evaluation s: DRAFT Version 1.0.
(740P18001). Washington, D.C.: U.S. Environmental Protection Agency, Office of Chemical
Safety and Pollution Prevention.
(2018d). Draft Generic Scenario on the Application of Spray Polyurethane Foam (SPF).
Chantilly, VA: Eastern Research Group.
L EPA (2018e). ECOTOX user guide: ECOTOXicology database system. Version 5.0.
(2018f). Problem formulation of the risk evaluation for C.I. Pigment Violet 29 (Anthra[2,l,9-
def:6,5,10-d'e'f]diisoquinolinel,3,8,10(2H,9H)-tetrone). CASRN: 81-33-4 [EPA Report], (EPA
Document# 740-R1-7021). United States Environmental Protection Agency , Office of Chemical
Safety and Pollution Prevention, Office of Pollution Prevention and Toxics.
https://www.epa.gov/assessin.g-and-m.an.aging-chem.icals-iinder-tsca/pigment-violet-2.9-
anthra219-def6510-defdiisoquinoline-0
L (2018g). Problem formulation of the Risk Evaluation for cyclic aliphatic bromides cluster
(HBCD). (EPA-740-R1-7012). Washington, DC: Office of Chemical Safety and Pollution
Prevention, United States Environmental Protection Agency.
https://www.epa.gov/sites/prodiiction/files/2018-06/doeuments/h.bcd problem formulation 05-
31-18.pdf.
EPA. (2018h). Strategy for assessing data quality in TSCA Risk Evaluation s. Washington DC:
U.S. Environmental Protection Agency, Office of Pollution Prevention and Toxics.
06/docum ents/hb cd
epa.gov/sites/prodiiction/files/2018.
Pollution Prevention, https
?production/files/2018
Page 530 of 723
-------
U.S. EPA. (2018i). Update for chapter 9 of the Exposure Factors Handbook: Intake of Fruits and
Vegetables [EPA Report], (EPA/600/R-18/098F). Washington, DC: National Center for
Environmental Assessment, Office of Research and Development.
U.S. EPA. (2018j). Update for chapter 11 of the Exposure Factors Handbook: Intake Of Meats, Dairy
Products, And Fats [EPA Report], (EPA/600/R-17/485F). Washington, DC: National Center for
Environmental Assessment, Office of Research and Development.
https://cfpub.epa. eov/ncea/efp/recordisplav.cfm?deid=340600.
U.S. EPA. (2018k). Sustainable management of construction and demolition materials [Website],
https://www.epa.eov/smm/sitstainable-managem.ent-constmction-and-demolition-materials.
U.S. EPA. (2019a). Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental File:
Occupational Exposure and Environmental Releases Calculations.
http s: //www, re gut ati on s. gov/docket?D= Q-OPPT-2019-023 7
U.S. EPA. (2019b). Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Systematic Review
Supplemental File: Data Extraction Tables of Environmental Hazard Studies. .
.regul ati on s. gov/docket?D=
U.S. EPA. (2019c). Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Systematic Review
Supplemental File: Updates to the Data Quality Criteria for Epidemiological Studies. .
http s: //www, re gul ati on s. gov/docket?D= Q-OPPT-2019-023 7.
U.S. EPA. (2019d). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Supplemental
Information on General Population, Environmental, and Consumer Exposure Assessment. .
http s: //www, re gul ati on s. gov/docket?D=. Q-OPPT-2019-02.3 7.
U.S. EPA. (2019e). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Supplemental
Information on Human Health Hazard.
http s: //www, re gul ati on s. gov/docket?D= Q-OPPT-2019-023 ?.
U.S. EPA. (2019f). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Extraction of General Population and Environmental Exposure Studies
. regul ati on s. gov/doc ket?D=
U.S. EPA. (2019g). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Extraction Tables for Human Health Hazard Studies. .
http s: //www, regul ati on s. gov/doc ket?D=. Q-OPPT-2019-0237.
U.S. EPA. (2019h). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Extraction Tables of Environmental Fate and Transport Studies.
http s: //www, re gul ati on s. gov/docket?D= Q-OPPT-2019-02.3 7.
U.S. EPA. (2019i). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Quality Evaluation for Occupational Exposure and Release Data for
Common Sources. https://www.regulations.gov/docket?D=EPA-H.Q-OPPT-2.019-0237.
U.S. EPA. (2019j). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Quality Evaluation for Occupational Exposure and Release Data.
.regul ati on s. gov/docket?D=
U.S. EPA. (2019k). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Quality Evaluation of Ecological Hazard Studies.
http s: //www, re gul ati on s. gov/docket?D= Q-OPPT-2019-023 7.
U.S. EPA. (20191). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Quality Evaluation of Environmental Fate and Transport Studies. .
http s: //www, re gul ati on s. gov/docket?D= Q-OPPT-2019-023 7.
U.S. EPA. (2019m). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic
Review Supplemental File: Data Quality Evaluation of General Population and Environmental
Exposure Studies. https://www.regulation.s.gov/docket?D=EPA-H.Q-OPPT-2019-0237.
Page 531 of 723
-------
(2019n). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Quality Evaluation of Human Health Hazard Studies.
http s: //www, re eut ati on s. gov/docl IP A-HQ-OPPT-2019-023 7.
U.S. EPA. (2019o). Durable Goods: Product-Specific Data, https://www.epa.gov/facts-and-figures-
about-materials-waste-and-recvcling/durable-goods-product-specific-data#Electronics.
(2019p). IECCU 1.1 User's Guide. In Simulation Program for Estimating Chemical
Emissions from Sources and Related Changes to Indoor Environmental Concentrations in
Buildings with Conditioned and Unconditioned Zones (IECCU). (EPA Contract # EP-W-12-
010). Washington, DC: U.S. Environmental Protection Agency, Office of Pollution Prevention
and Toxics.
L (2019q). Point Source Calculator: A Model for Estimating Chemical Concentration in Water
Bodies. Washington, DC: U.S. Environmental Protection Agency, Office of Chemical Safety and
Pollution Prevention.
(2019r). User's Guide: Integrated Indoor-Outdoor Air Calculator (IIOAC). Washington, DC:
U.S. EPA.
EPA. (2019s). Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Supplemental File:
Occupational Risk Calculator, http s: //www, regut ati on s. gov/docket?D=EP A-H < < • ^"PT-.>«If
0237.
L LEPA (2019t). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Quality Evaluation of Physical-Chemical Properties Studies.
https://www\ regulations.gov/docket?D=EPA-HQ-QPPT-2019-0237.
(2019t). Risk Evaluation for Cyclic Aliphatic Bromides Cluster (HBCD), Systematic Review
Supplemental File: Data Quality Evaluation of Physical-Chemical Properties Studies.
(2020) Memorandum: NIOSH/BLS Respirator Usage in Private Sector Firms [Personal
Communication], https://www\regulations.gov/document?D=EPA~H.Q~QPPT~2019-0500-0029.
LINEP. (2010a). Hexabromocyclododecane draft risk profile (pp. 1-39). Europe.
http://chm.pops.int/Convention/POPsReviewCommittee/hrPOPRCMeetings/POPRC5/POPRC5F
ollowupcommunications/HBCDInvitationforcommentsondraftRP/tabid/742/language/en-
nS/Default.aspx.
UNEP. (2010b). Report of the Persi stent Organic Pollutants Review Committee on the Work of Its Sixth
Meeting. Addendum. Risk Profile on Hexabromocyclododecane (pp. 44). Chatelaine,
Switzerland.
http://chm.pops.int/TheCon.vention/POPsReviewCom.mittee/Meetings/POPRC6/POPRC6Report
andDecisions/ta tult.aspx.
UNEP. (2011). Report of the Persistent Organic Pollutants Review Committee on the Work of Its
Seventh Meeting. Addendum. Risk management evaluation on hexabromocyclododecane. In
Stockholm Convention on Persistent Organic Pollutants. (UNEP/POPS/POPRC.7/19/Add. 1).
Geneva, Switzerland.
http://chm.pops.int/Convention/POPsReviewCommittee/POPRCMeetings/POPRC7/POPRC7Re
portandDecisions/tabid/2472/Default.aspx
van. Beusekom. OC; Eliarrat. E; Barcelo. D; Koelmans. AA. (2006). Dynamic modeling of food-chain
accumulation of brominated flame retardants in fish from the Ebro River Basin, Spain. Environ
Toxicol Chem 25: 2553-2560. http://dx.doi.< 7/05-409R.1.
van, der Yen. LT; Verhoef. A; van, de Kuil * W . I ^onards. PE; Visser. TJ; Hauliers. T; Herlin. M;
Hakansson. H; Olausson. H; Piersma. AH; Vos. JG. (2006). A 28-day oral dose toxicity study
enhanced to detect endocrine effects of hexabromocyclododecane in Wistar rats. Toxicol Sci 94:
281-292. http://dx.doi.org/ 3/toxsci/kfll 13.
Page 532 of 723
-------
van c . LTM; van de Knit. T; Leonards. PEG: Slob. W; Lilienthal. H; Litems. S; Herlin. M;
Hakansson. H; Canton. in. den Berg. M; Visser. TJ; van. Loveren. H; Vos. JG: Piersma.
AH. (2009). Endocrine effects of hexabromocyclododecane (HBCD) in a one-generation
reproduction study in Wistar rats. Toxicol Lett 185: 51-62.
http://dx.doi.on .toxtet.2008.12.003.
van. Raaii. MTM; Jan.se iersma. AH. (2003a). The relevance of developmental toxicity
endpoints for acute limit setting. (601900004). Bilthoven: RIVM.
Van Raaii. MTM; Janssen. PAH; Piersma. AH. (2003b). The relevance of developmental toxicity
endpoints for acute limits settings (pp. 1-88). (RIVM Report 601900004). Nederlands:
Nederlands National Institute for Public Health and the Environment.
http://www2.epa.eov/sites/prodiiction/files/2014-04/documents/mtg35b.pdf.
Vansell. NR.; Klaassen. CD. (2002). Effect of microsomal enzyme inducers on the biliary excretion of
triiodothyronine (T(3)) and its metabolites. Toxicol Sci 65: 184-191.
http://dx.doi.on 93/toxsci/65.2.184.
Veith. GD; DeFiv 5 H ; Bergsvut _H\ _ (1979). Measuring and estimating the bioconcentration factor of
chemicals in fish. J Fish Res Board Can 36: 1040-1048. http://dx.doi.org/10 I 139/f? M i •
Vel si col Chem. Corp. (1978). Industrial hygiene survey, Velsicol Chemical Corporation, El Dorado, Ark
Plant, Fire Master 680 Unit and semi-works summary with attachments and cover letter dated
071978 [TSCA Submission], (EPA/OTS Doc #88-7800228). Chicago, IL.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuerv=€)TS0200544.
Venier. M; Dov omanak. K; Backus. S; Hites. R. (2014). Flame retardants and legacy chemicals
in Great Lakes' water. Environ Sci Technol 48: 9563-9572. http://dx.doi.< l/es501509r.
Venkatesan. AK; Hal den. RtJ. (2014). Brominated flame retardants in U.S. biosolids from the EPA
national sewage sludge survey and chemical persistence in outdoor soil mesocosms. Water Res
55: 133-142. http://dx.doi.org/10 tOj j watres.201 i 02.021.
Voita. S; Becanova. J; Melym.uk. L; Komprdova. K; Kohoutek. J; Kukucka. P; Klanova. J. (2017).
Screening for halogenated flame retardants in European consumer products, building materials
and wastes. Chemosphere 168: 457-466. http://dx.doi.Org/10.1016/i.chemosphere.2016.l 1.032.
Vorkamp. K; Bossi. R; Rii 3v. H; Sonne. C; Dietz. R (2015). Novel brominated flame
retardants and dechlorane plus in Greenland air and biota. Environ Pollut 196: 284-291.
http://dx.doLorg/10.1016/i.envpol iO I j_0 00'__.
Waindzioch. A. (2000). Technical report of workplace air-measurements with respect to inhalable dust
and hexabromocyclododecane (HBCD) at broomchemie B.V., Terneuzen (pp. 7). (100/00443-
00). Taunusstein, Germany: Waindzioch, A.
Walsh. GE; Yoder. MI; Mclaughlin. LL; Lores. EM. (1987). Responses of marine unicellular algae to
brominated organic compounds in six growth media. Ecotoxicol Environ Saf 14: 215-222.
Waste Business Journal. (2019). Directory of waste processing & disposal sites 2019 (US edition)
[Website], https://www.wasteinfo.com/diratlas.htm.
Watanabe. W; Shimizu. T; Sawamura. R; Hi no. A; Konno. K; Hirose. A; Kurokawa. M. (2010). Effects
of tetrabromobisphenol A, a brominated flame retardant, on the immune response to respiratory
syncytial virus infection in mice. Int Immunopharmacol 10: 393-397.
http://dx.doi.on tO 10t.in.tim.p.200 i J ^ 14
Weil. ED; Levchik. SV. (2009). Flame Retardants for Plastics and Textiles. Munich, Germany: Carl
Hanser Verlag GmbH & Co.KG. http://dx.doi.org/10.3139/978344643Q655.ftn.
Weschler. CJ; Nazaroff. WW. (2010). SVOC partitioning between the gas phase and settled dust
indoors. Atmos Environ 44: 3609-3620. http://dx.doi.oi r\.atmosenv.2.010.06.029.
WHO. (2012). Guidance for immunotoxicity risk assessment for chemicals. (Harmonization Project
Document No. 10). Geneva, Switzerland.
http: //www, in ch em.org/ docum ents/h arm proi /h arm, proi /h. arm pr If.
Page 533 of 723
-------
search. (1997). Twenty-eight day repeated dose oral toxicity study of HBCD in rats, with cover
letter dated 3/18/1997 [TSCA Submission], (TSCATS/445005. OTS0558957. Doc I.D.
#86970000747). Washington, DC: Chemical Manufacturers Association.
search. (2001). 90-Day oral (gavage) toxicity study of HBCD in rats. (WIL-186012).
Washington, DC: Chemical Manufacturers Association.
Wildlife Intl. LTD. (1997a). Letter from Chem MFCS Assoc to USEPA regarding: toxicological
investigation of hexabromocyclododecane (HBCD) with attachments, dated 06/27/1997 [TSCA
Submission], (EPA/OTS Doc #FYI-OTS-1097-1306).
https://ntrl.ntis. gov/NT ihboard/searchResults/titleDetail/OTSOOOl 306. xhtml.
Wildlife Intl. LTD. (1996). Hexabromocyclododecane (HBCD): Closed bottle test with cover letter
dated 12/12/1996. (EPA/OTS; Doc #86970000357). Arlington, VA: Chemical Manufactures
Association.
Wildlife Intl. LTD. (1997a). Hexabromocyclododecane (HBCD): Determination of n-octanol/water
partition coefficient with cover letter dated 06/27/1997 [TSCA Submission], (EPA/OTS Doc
#86970000802). Washington, DC: Chemical Manufacturers Association.
https://ntrl.ntis.gov/NT^ '>,.shboard/search.Results/titleDetail/OTS0573665.xhtml.
Wildlife Intl (1997b). Letter from Chem Mfgs Assoc to USEPA Regarding: Toxicological
Investigation of Hexabromocyclododecane (HBCD) with Attachments, Dated 06/27/1997
[TSCA Submission] (pp. 328). (OTS0001306; EPA/OTS; Doc #FYI-OTS-1097-1306).
Wildlife Intl LTD. (1997c). Hexabromocyclododecane (HBCD): Determination of the vapor pressure
using a spinning rotor gauge with cover letter dated 08/15/1997 [TSCA Submission], (EPA/OTS
Doc #86970000839). Chemical Manufacturers Association.
https://ntrl.ntis. gov/.NTRL/dashboard/searchResults.xhtml?searchOuerv=OTSQ573 702.
Wildlife Intl. LTD. (1998). Initial submission: Hexabromocyclododecane (HBCD) - a flow-through life-
cycle toxicity test with the cladoceran (Daphnia magna), final report, with cover letter dated
5/18/1998 (pp. EPA/OTS; Doc #FYI-OTS-0698-1333).
Wildlife Intl. LTD. (2000). Letter from Amer Chem Cncl submitting flow-through bioconcentration test
w/rainbow trout and end-user survey-phase 1 study of brominated flame retardant, w/attchmts
and dated 8/28/00 [TSCA Submission], (EPA/OTS Doc #FYI-OTS-1000-1392). Arlington, VA:
American Chemistry Council.
Wildlife Intl LTD. (1997). Hexabromocyclododecane (HBCD): A 48-Hour Flow-Through Acute
Toxicity Test With The Cladoceran (Daphnia Magna) With Cover Letter Dated 06/20/1997. In
Technical Report TSCATS 452984 (pp. (EPA/OTS 1097-1300)). (TSCATS/452984).
Wu. IP; Guan \ ]' •' hang. V i uo. XI; Zhi. H; Chen. SJ; Mai. BX. (2010). Trophodynamics of
hexabromocyclododecanes and several other non-PBDE brominated flame retardants in a
freshwater food web. Environ Sci Technol 44: 5490-5495. http://dx.doi.org/10.1021 /es 101300t.
Wu. IP; Gue ing. Y; Luo. XI; Zhi. H; Chen. SJ: Mai. BX. (2011). Several current-use, non-
PBDE brominated flame retardants are highly bioaccumulative: evidence from field determined
bioaccumulation factors. Environ Int 37: 210-215.
http://dx.doi.org ^^m • j.envint._h«*•« <06.
Wu. M; Wu. D; Wang. C; Guc 3. Z. (2016a). Hexabromocyclododecane exposure induces
cardiac hypertrophy and arrhythmia by inhibiting miR-1 expression via up-regulation of the
homeobox gene Nkx2.5. J Hazard Mater 302: 304-313.
http://dx.doi.on h1 h'l • Lih.azmat.2i'I ^ i < < 004
Wu. M; Zuo J It H Uuatn I Chen. M; Wang. C (2013). Effects of low-level
hexabromocyclododecane (HBCD) exposure on cardiac development in zebrafish embryos.
Ecotoxicology 22: 1200-1207. http://dx.doi.org/
Page 534 of 723
-------
Wu. MH: Han. T: Xu. G: Zang. C; Li. YJ; Sun. R; Xu. BT: Sun \ < hen, fang. L. (2016bY
Occurrence of Hexabromocyclododecane in soil and road dust from mixed-land-use areas of
Shanghai, China, and its implications for human exposure. Sci Total Environ 559: 282-290.
http://dx.doi.oii K' Km i.scitotenv.ios <_0 ! *<
Wit. T; Huang. H; Zhang. S. (2016c). Accumulation and phytotoxicity of technical
hexabromocyclododecane in maize. J Environ Sci 42: 97-104.
http://dx.doi.on. 110 i? ties.2015.06.018.
Wu. T; Wang. S: Huang. H; Zhang. S. (2012). Diastereomer-specific uptake, translocation, and toxicity
of hexabromocyclododecane diastereoisomers to maize. J Agric Food Chem 60: 8528-8534.
http://dx.doi.on f302682p.
XPSA. (2017a). [Letter from John Ferraro, XPSA, to Sue Slotnick, EPA, regarding
Hexabromocyclododecane (HBCD)].
XPSA. (2017b). Preliminary information on manufacturing, processing, distribution, use, and disposal:
Cyclic aliphatic bromide cluster (HBCD). OCSPP. Public comment. (EPA-HQ-OPPT-2016-
0735-0017).
Yamashita. R; Takada. H; Fukuwaka. MA; Watanuki. Y. (2011). Physical and chemical effects of
ingested plastic debris on short-tailed shearwaters, Puffinus tenuirostris, in the North Pacific
Ocean. MarPollutBull 62: 2845-2849. http://dx.doi.oa > j.marpolbul.2011 10 008
Yanagisawa. R; Koike. E; Win-Shwe. imoto. M; Takano. H. (2014). Impaired lipid and
glucose homeostasis in hexabromocyclododecane-exposed mice fed a high-fat diet. Environ
Health Perspect 122: 277-283. http://dx.d0i.0rg/l 0.1289/ehp. 1307421.
Yang. R; Wei. H; Guo. J; Li. A. (2012). Emerging brominated flame retardants in the sediment of the
Great Lakes. Environ Sci Technol 46: 3119-3126. http://dx.doi.oa 1 KL * as20 ! I jjjr-.
Yi. S: Liu. JG; Jin. J; Zhu. J. (2016). Assessment of the occupational and environmental risks of
hexabromocyclododecane (HBCD) in China. Chemosphere 150: 431-437.
http://dx.doi.oa 10 KM - • 1.chemosphere. JO 1 01 0 I
i. Y.H. (1980). Pharmacokinetics of HBCD in rats [unpublished], Rosemont, IL: Vesicol
Chemical Corporation.
Yu. L; Luo. X; Zheng. X; Zen^ \ „ i hen. D; Wu. J; Mai. H (2013). Occurrence and biomagnification of
organohalogen pollutants in two terrestrial predatory food chains. Chemosphere 93: 506-511.
http://dx.d01.0rg/l0.1016/j.chemosphere.201 Ou 023.
Zege its. A; Van Bommel. R; Minkenberg. C; Hamers. T; Kamstra. JH; Pierce. GJ; Boon. IP.
(2005). Levels of hexabromocyclododecane in harbor porpoises and common dolphins from
western European seas, with evidence for stereoisomer-specific biotransformation by
cytochrome p450. Environ Sci Technol 39: 2095-2100. http://dx.doi.org/10.1021/es049209t.
Zeiger. E; Andersoi aworth. S; Lawlor. T; Mortelmans. K; Speck. W". (1987). Salmonella
mutagenicity tests: III. Results from the testing of 255 chemicals. Environ Mutagen 9: 1-109.
http://dx.doi.oa 02/em.2860090602..
Zeng. Y.H; Tarn K I_ uo. XI; Zhe»^ _\K 'ai. BX. (2016). Organohalogen pollutants in
surface particulates from workshop floors of four major e-waste recycling sites in China and
implications for emission lists. Sci Total Environ 569a€"570: 982-989.
http://dx.doi.oa tO i0 K• i.scitotenv.l'^l' 3.053.
Zhan erecke. AC; Wang. I. (2012). Co-release of hexabromocyclododecane (HBCD)
and Nano- and microparticles from thermal cutting of polystyrene foams. Environ Sci Technol
46: 10990-10996. http://dx.doi.org/10.1021/es3Q2559v.
Zhang. X; Yang I 'bang. \ \ I t10. T; Song. S; Wat^ 1 (2008). Induction of hepatic enzymes
and oxidative stress in Chinese rare minnow (Gobiocypris rarus) exposed to waterborne
hexabromocyclododecane (HBCDD). Aquat Toxicol 86: 4-11.
http://dx.doi.org i.aquatox.200 0 _002.
Page 535 of 723
-------
Zhang. Y; Sun. H: Liu. F; Dai. \ , » Mn. X: Ruan. \. 'liao. L; Gan. Z. (2013).
Hexabromocyclododecanes in limnic and marine organisms and terrestrial plants from Tianjin,
China: diastereomer- and enantiomer-specific profiles, biomagnification, and human exposure.
Chemosphere 93: 1561-1568. http://dx.doi.Org/10.1016/i.chemosphere.2013.08.004.
Zhan un. H; Ruan. Y. (2014a). Enantiomer-specific accumulation, depuration, metabolization and
isomerization of hexabromocyclododecane (HBCD) diastereomers in mirror carp from water. J
Hazard Mater 264: 8-15. http://dx.doi.oo jhazmat.2013.10.062.
Zhang. Y; Sun. H; Zhu. H; Ruan. Y; Liu. F; Liu. X. (2014b). Accumulation of
hexabromocyclododecane diastereomers and enantiomers in two microalgae, Spirulina subsalsa
and Scenedesmus obliquus. Ecotoxicol Environ Saf 104: 136-142.
http://dx.doi.org/10/1016/i. ecoenv .2014.02.027.
Zhou. T; Ross. DG: Devito. Ml: Crofton. KM. (2001). Effects of short-term in vivo exposure to
polybrominated diphenyl ethers on thyroid hormones and hepatic enzyme activities in weanling
rats. Toxicol Sci 61: 76-82.
Zhu. C: Wang. P; Li. Y; Chen. Z; Li. H; Ssebugere. P; Zhang. O, Jiang. G. (2017). Trophic transfer of
hexabromocyclododecane in the terrestrial and aquatic food webs from an e-waste dismantling
region in East China. Environ Sci Process Impacts 19: 154-160.
http://dx.doi.org/10.1039/c6em00617e.
Ziccardi. LM; Edgington. A; Hentz. K; Kulacki. KJ; Ki scoll. S. (2016). Microplastics as vectors
for bioaccumulation of hydrophobic organic chemicals in the marine environment: A state-of-
the-science review [Review], Environ Toxicol Chem 35: 1667-1676.
http://dx.doi.on 02/etc.3461
Zoellci IvJ. I >V (2007). General background on the hypothalamic-pituitary-thyroid
(HPT) axis [Review], CritRev Toxicol 37: 11-53.
http://dx.doi.on 80/10408440601123446.
Page 536 of 723
-------
APPENDICES
Appendix A REGULATORY HISTORY
A.l Federal Laws and Regulations
Table Apx A-l. Federa
Laws and Regulations
Statutes/Regulations
Description of Authority/Regulation
Description of Regulation
Toxic Substances
Control Act (TSCA) -
Section 5(a)
Once EPA determines that a use of a chemical
substance is a significant new use under TSCA
section 5(a), persons are required to submit a
significant new use notice (SNUN) to EPA at
least 90 days before they manufacture (including
import) or process the chemical substance for
that use.
In September 2015, EPA
promulgated a SNUR to
designate manufacture or
processing of HBCD for
use as a flame retardant in
consumer textiles (apart
from use in motor
vehicles) as a significant
new use. Manufacturers
(which includes importers)
and processors are required
to notify EPA 90 days
before commencing the
activity (80 FR 57293,
September 23, 2015).
TSCA - Section 6(b)
EPA is directed to identify and begin Risk
Evaluations on 10 chemical substances drawn
from the 2014 update of the TSCA Work Plan
for Chemical Assessments.
Cyclic Aliphatic Bromide
Cluster (HBCD) is on the
initial list of chemicals to
be evaluated for
unreasonable risk under
TSCA (81 FR 91927,
December 19, 2016).
TSCA - Section 8(a)
The TSCA section 8(a) CDR Rule requires
manufacturers (including importers) to give EPA
basic exposure-related information on the types,
quantities and uses of chemical substances
produced domestically and imported into the
United States.
HBCD manufacturing
(including importing),
processing, and use
information is reported
under the CDR rule (76 FR
50816, August 16, 2011)
TSCA - Section 8(b)
EPA must compile, keep current and publish a
list (the TSCA Inventory) of each chemical
substance manufactured, processed or imported
into the United States.
HBCD (CASRN 25637-
99-4 and CASRN 3194-
55-6) was on the initial
TSCA Inventory and
therefore was not subject
to EPA's new chemicals
review process (60 FR
16309; March 29, 1995).
Page 537 of 723
-------
Emergency Planning
and Community Right-
to-Know Act (EPCRA)
- Section 313
Requires annual reporting from facilities in
specific industry sectors that employ 10 or more
full-time equivalent employees and that
manufacture, process or otherwise use a TRI-
listed chemical in quantities above threshold
levels.
EPA listed HBCD on the
TRI under 81 FR 85440
effective November 28,
2016. The first TRI
reporting deadline for
HBCD is July 1,2018.
US EPA Policy on
Evaluating Risk to
Children (1995)
It is EPA's policy to consider the risks to infants
and children consistently and explicitly as a part
of risk assessments generated during its decision
making process, including the setting of
standards to protect public health and the
environment. To the degree permitted by
available data in each case, the Agency will
develop a separate assessment of risks to infants
and children.
HBCD Final Risk
Evaluation assessed risks
to infants and children.
Executive Order 13045
- Protection of Children
from Environmental
Health Risks and
Safety Risks (1997)
Executive Order (EO) 13045 pertains to
environmental health or safety risk that EPA has
reason to believe may disproportionately affect
children. EO 13045 states that each federal
agency "(a) shall make it a high priority to
identify and assess environmental health risks
and safety risks that may disproportionately
affect children; and (b) shall ensure that its
policies, programs, activities, and standards
address disproportionate risks."
HBCD Final Risk
Evaluation assessed
environmental health risks
and safety risks that may
disproportionately affect
children and complied with
EO 13045 (62 FR 19885;
April 23, 1997).
A.2 State Laws and Regulations
Table Apx A-2. State Laws and Regulations
Stale Actions
Description of Action
Classification of HBCD
as Chemical of Concern
to Children; law requiring
reporting by
manufacturers
Maine classifies HBCD as a chemical of high concern (Maine 38 M.R.S.A.
Section 1693-A(1))
Maine requires manufacturers or distributers to report the use of deca BDE
and/or hexabromocylododecane, when intentionally added to certain
children's products which are sold in the State of Maine. The first reporting
deadline was August 31, 2017. (Rule Chapter 889)
http://www.maine.gov/dep/safechem/
Minnesota classifies HBCD as a chemical of high concern (Toxic Free Kids
Act Minn. Stat. 2010 116.9401-116.9407)
Oregon's Toxic-Free Kids Act requires manufacturers of children's
products sold in Oregon to report products containing HBCD or other high
priority chemicals of concern for children's health if found at or above
specific levels in those products. Ultimately, manufacturers are to remove
these chemicals from certain products or seek a waiver. Products that fall
Page 538 of 723
-------
under this law are those that are marketed to or intended for children. The
first deadline for providing notice was January 2018.
Washington requires manufacturers of children's products sold in
Washington to report if their product contains certain chemicals of high
concern to children, including HBCD. The law also bans from manufacture
or sale, in the state, children's products or residential upholstered furniture
containing >1,000 ppm of five flame retardants, including HBCD (Wash.
Admin. Code Section 173-334-130)
Other
In California, HBCD is listed as an initial informational candidate under
California's Safer Consumer Products regulations, on the state's
Proposition 65 list (Cal. Code Regs, tit. 22, Section 69502.3, subd. (a))
California lists HBCD as a designated priority chemical for biomonitoring.
However, California has not yet started biomonitoring HBCD. (California
SB 1379)
The Oregon Department of Environmental Quality lists HBCD as a priority
persistent pollutant and publishes use, exposure pathways and release data
for HBCD (Oregon SB 737)
In Massachusetts, HBCD will be reportable under the Toxics Use
Reduction Act beginning in reporting year 2018. (300 CMR 41.00)
Page 539 of 723
-------
A.3 International Laws and Regulations
("on ill rv/Org:iniz:i1 ion
Requirements nnd Restrictions
Canada
In October 2016, the Regulations Amending the Prohibition of Certain
Toxic Substances Regulations, 2012 (the Amendments) were published in
the Canada Gazette, Part II: Vol. 150, No. 20 - October 5, 2016 and will
come into force in December 2016. The Amendments include controls on
HBCD that prohibit HBCD and certain products containing the substance.
Time-limited exemptions for certain uses are included to allow industry to
phase-out their use of HBCD (Government of Canada).
European Union
HBCD is listed as a substance of very high concern (SVHC) and it is also
listed under Annex XIV (Authorisation list) of European Union's
Registration, Evaluation, Authorisation and Restriction of Chemicals
(REACH). After August 21, 2015, only persons with approved authorization
applications mav continue to use the chemical (European Chemicals
Agency).
The Waste Electrical and Electronic Equipment (WEEE) directive in the
European Union requires the separation of plastics containing brominated
flame retardants prior to recvcling (European Commission WEEE).
Japan
HBCD is subject to mandatory reporting requirements in Japan under the
Chemical Substances Control Law (CSCL); specifically, Japan requires type
III monitoring for all substances that may interfere with the survival and/or
growth of flora and fauna (Ministry of Economy. Trade and Industry Japan).
United Nations
Stockholm Convention
on Persistent Organic
Pollutants (POPs)
In May 2013, HBCD was added to the United Nations Stockholm
Convention list of Persistent Organic Pollutants (POPs) with specific
exemptions for production and use in EPS or XPS in buildings. As required
by the convention, Parties that use these exemptions must register with the
secretariat and the exemptions, unless extended in accordance with the
obligations of the Convention, expire five years from after the date of entry
into force of the Convention with respect to the particular chemical (SCCH
2.018b).
Page 540 of 723
-------
Appendix B LIST OF SUPPLEMENTAL DOCUMENTS
Associated Systematic Review Data Evaluation Documents - Provides additional detail and
information on individual study evaluations including criteria and scoring results, and Associated
Systematic Review Data Extraction Documents - Provides data extracted from acceptable studies
following evaluation of individual studies.
1. Supplemental File:Supplemental Information on General Population, Environmental, and
Consumer Exposures (U.S. EPA. 2019d).
2. Supplemental File Supplemental Information on Human Health Hazard. ( ).
3. Supplemental File: Occupational Exposure and Environmental Release Calculations
Spreadsheet. (U.S. EPA. 2019a)
4. Supplemental File: Occupational Risk Calculator ( SO 19s).
5. Systematic Review Supplemental File: Updates to Data Quality Criteria for Epidemiological
Studies. (U.S. EPA. 2019c)
6. Systematic Review Supplemental File: Data Extraction Tables for Environmental Fate and
Transport Studies. (U.S. EPA. 2019h)
7. Systematic Review Supplemental File: Data Quality Evaluation of Environmental Release and
Occupational Exposure Data (U.S. EPA. )
8. Systematic Review Supplemental File: Data Quality Evaluation for Environmental Release and
Occupational Exposure - Common Sources (U.S. EPA. 20191)
9. Systematic Review Supplemental File: Data Quality Evaluation for General Population,
Environmental, and Consumer Exposure ( i)
10. Systematic Review Supplemental File: Data Quality Evaluation of Environmental Hazard
Studies (U.S. EPA. 2019k)
11. Systematic Review Supplemental File: Data Quality Evaluation of Human Health Hazard
Studies - Animal, In Vitro, and Epidemiological Studies ( |)
12. Systematic Review Supplemental File: Data Quality Evaluation of Environmental Fate and
Transport Studies (U.S. EPA. 20191)
13. Systematic Review Supplemental File: Data Extraction for General Population, Environmental,
and Consumer Exposure ( E019D
14. Systematic Review Supplemental File: Data Extraction Tables for Human Health Hazard Studies
(U.S. EPA. 2019a)
Page 541 of 723
-------
15. Systematic Review Supplemental File: Data Extraction of Environmental Hazard Studies (U.S.
EPA. 2019fr)
16. Systematic Review Supplemental File: Data Quality Evaluation of Physical-Chemical Properties
Studies ( 2019ft
Page 542 of 723
-------
Appendix € FATE AND TRANSPORT
C.J Biodegradation
A closed bottle screening-level test for ready biodegradability (OECD Guideline 301D, EPA OTS
796.3200) was performed using an initial HBCD concentration of 7.7 mg/L and an activated domestic
sludge inoculum (Wildlife Intl 1996 as cited in ECHA. 2008b; Albemarle. 2005). No biodegradation was
observed (0% of the theoretical oxygen demand) over the test period of 28 days under the stringent
guideline conditions of this test.
Degradation of HBCD during simulation tests with viable microbes, based on OECD 307 and 308, was
approximately 61% in anaerobic freshwater sediment, 44% in aerobic freshwater sediment, and 10% in
aerobic soil after 112-113 days (Davis et al. 2006; ECB. 2008). The results from this study correspond
to estimated HBCD half-lives of 92 days in anaerobic freshwater sediment, 128, 92, and 72 days for a-
, y-, and P-HBCD, respectively in aerobic freshwater sediment, and >120 days in aerobic soil. An
initial total 14C-HBCD concentration of 3.0-4.7 mg/kg dry weight in the sediment and soil systems
was used, allowing for quantification of individual isomers, metabolite identification, and mass
balance evaluation (Davis et al 2006; NICNAS 2.012a). Although very high spiking rates can be toxic
to microorganisms in biodegradation studies and lead to unrealistically long estimated half-lives, the
results of this study did not suggest toxicity to microorganisms. Tests with viable microbes
demonstrated increased HBCD degradation compared to the biologically inhibited control studies. In
combination, these studies suggest that HBCD will degrade slowly in the environment, although faster
in sediment than in soil, faster under anaerobic conditions than aerobic conditions, faster with
microbial action than without microbial action, and at different rates for individual HBCD
diastereomers (slower for a-HBCD than for the y- and P- stereoisomers. The same researchers
previously conducted a water-sediment simulation test for commercial HBCD based on OECD
guideline 308 using nominal HBCD concentrations of 0.034-0.089 mg/kg dry weight (Davis et al.
2003a. 2005; Albemarle. 2005; ECB. 2008). Aerobic and anaerobic microcosms were pre-incubated at
20 °C for 49 days and at 23 °C for 43-44 days, respectively. HBCD was then added to 14-37 g dry
weight freshwater sediment samples in 250 ml serum bottles (water:sediment ratio of 1.6-2.9) and the
microcosms were sealed and incubated in the dark at 20 °C for up to 119 days. For the aerobic
microcosms, the headspace oxygen concentration was kept above 10—15%. This study evaluated only
Y-HBCD and did not address interconversion of HBCD isomers or a- and P-HBCD degradation.
Disappearance half- lives of HBCD with sediment collected from Schuylkill River and Neshaminy
creek were 11 and 32 days in viable aerobic sediments, respectively (compared to 190 and 30 days in
abiotic aerobic controls, respectively), and 1.5 and 1.1 days in viable anaerobic sediments, respectively
(compared to 10 and 9.9 days in abiotic anaerobic controls). Data from these tests suggest that
anaerobic degradation is faster than aerobic degradation of HBCD in viable and abiotic sediments and
that degradation is faster in viable conditions than abiotic conditions. While these findings are
consistent with Davis et al. (2006). the actual degradation rates in this study are much faster. However,
results from this study do not provide a reliable indication of HBCD persistence. A mass balance could
not be established because only y-HBCD was used to quantify HBCD concentrations, 14C-radiolabeled
HBCD was not used, and degradation products were not identified; therefore, apparent disappearance
of HBCD in this study may not reflect biodegradation. In addition, there were concerns that
contaminated sediment may have been used, HBCD extraction was incomplete (HBCD recovery
varied from 33 to 125 %), and an interfering peak was observed in the LC/MS chromatograms
corresponding to y-HBCD (NICNAS 2012a; EC D8b).
Page 543 of 723
-------
Similarly, a soil simulation test was conducted based on OECD guideline 307 for commercial HBCD
using 50 g dry weight sandy loam soil samples added to 250 ml serum bottles (Davis et ai. 2003b;
Davis et al. 2005; Albemarle. 2005; ECHA. 2008b). The moisture content was 20% by weight. Aerobic
and anaerobic microcosms were pre-incubated at 20 °C for 35 days and at 23 °C for 43 days,
respectively. Activated sludge was added to the soil at 5 mg/g, and HBCD was added to the soil to
achieve a nominal concentration of 0.025 mg/kg dry weight. The microcosms were then incubated in the
dark at 20 °C for up to 120 days. The disappearance half- lives were 63 days in viable aerobic soil
(compared to >120 days in abiotic aerobic controls) and 6.9 days in viable anaerobic soil (compared to
82 days in abiotic anaerobic controls). As in the sediment studies, HBCD degradation in soil occurred
faster under anaerobic conditions compared to aerobic conditions, and faster in viable conditions than
abiotic conditions. The disappearance half-lives in soil were slower than those in sediment.
Biological processes were suggested to be responsible for the increased degradation of HBCD in this
study using viable conditions, relative to abiotic conditions; however, degradation was not adequately
demonstrated in soil because no degradation products were detected and only y-HBCD was used to
quantify HBCD concentrations, making it impossible to calculate a mass balance. HBCD recoveries on
day 0 of the experiment were well below (0.011-0.018 mg/kg dry weight) the nominal test
concentrations (0.025 mg/kg dry weight), suggesting rapid adsorption of HBCD to soil and poor
extraction methods (NICNAS 2012a; ECHA. 2008b).
In studies using 0.025-0.089 mg/kg HBCD (Davis et al. 2005). the estimated half-life values were
shorter than studies using 3.0-4.7 mg/kg HBCD (Davis et al. 2006) by approximately one order of
magnitude for aerobic, viable sediment (11-32 days compared to 72-128 days) and anaerobic viable
sediment (1.1-1.5 days compared to 92 days). The viable aerobic soil half-life using lower
concentrations of HBCD (Davis et al. 2005) was less than half of the half-life based on the higher
HBCD concentration (63 days compared to >120 days) (Davis et al. 2006). Both Davis et al. ((Davis et
al. 2006; Davis et al. 2005) studies suggest that HBCD degrades faster in sediment than in soil, faster
under anaerobic conditions than aerobic conditions, and faster with microbial action than without
microbial action. HBCD is poorly soluble, and it was suggested that at higher concentrations of HBCD,
degradation is limited by mass transfer of HBCD into microbes. However, results from the Davis et al.
(2005) study likely overestimate the rate of HBCD biodegradation, for the reasons noted above
(primarily, failure to use 14C-radiolabeled HBCD, quantify isomers other than y-HBCD, identify
degradation products, or establish a mass balance, but also procedural problems with contamination of
sediment, incomplete HBCD extraction, and occurrence of an interfering peak in the LC/MS
chromatograms corresponding to y-HBCD).
Furthermore, the rapid biodegradation rates from Davis et al. (2005) are not consistent with
environmental observations. HBCD has been detected over large areas and in remote locations in
environmental monitoring studies. Dated sediment core samples indicate slow environmental
degradation rates (NUCHAS 2012a; Marvin et al. 20 i *, K M \ .^08b; Davis et al. 2005). For example,
HBCD was found at concentrations ranging from 112 to 70,085 |ig/kg dry weight in sediment samples
collected at locations near a production site in Aycliffe, United Kingdom 2 years after the facility was
closed down (ECHA 2008b). Monitoring data do not provide a complete, quantitative determination of
persistence because HBCD emission sources, rates, and quantities are typically unknown, and all
environmental compartments are not considered. However, the monitoring data do provide evidence in
support of environmental persistence.
Rapid HBCD biodegradation has been demonstrated under laboratory conditions not representative of
typical environmental conditions. A study designed to elucidate HBCD degradation mechanisms and
Page 544 of 723
-------
optimize biodegradation capability reported an HBCD degradation half-life of only 0.66 days in
anaerobic digested sewage sludge amended with yeast and starch at 37 °C. In this test, a-HBCD had
lower susceptibility to degradation than P- or y-HBCD (Gerecke et al. 2006). The authors noted that
these results are specific to the anaerobic conditions established by the experiment, and that the
degradation rate constants are expected to vary based on redox conditions of each specific anaerobic
environment.
C.2 Bioconcentration/Bioaccumulation
HBCD has been shown in numerous studies to bioaccumulate and biomagnify in aquatic and terrestrial
food chains.
Bioisomerization
In general, a-HBCD bioaccumulates in organisms and biomagnifies through food webs to a greater
extent than the P- and y- diastereomers. Uncertainty remains as to the balance of diastereomer
accumulation in various species and the extent to which bioisomerization and biotransformation rates for
each isomer affect bioaccumulation potential. Some authors (e.g., (Law et al. 2.006)) have proposed that
y-HBCD isomerizes to a-HBCD under physiological conditions, rather than uptake being
diastereoisomer-specific. To test this theory, Esslinger et al. (Esslinger et al. 20 i 0) exposed mirror carp
(Cyprinus carpio morpha noblis) to only y-HBCD and found no evidence of bioisomerization. In
contrast, when Du et al. (Du et al. 2012a) exposed zebrafish (Danio rerio) to only y-HBCD, they found
detectable levels of a-HBCD in fish tissue, suggesting that bioisomerization occurred. Marvin et al.
(Marvin et al. 2011) hypothesized that differences in accumulation could also be due in part to a
combination of differences in solubility, bioavailability, and uptake and depuration kinetics.
(Zhang et al 2014b) calculated diastereomer-specific BCFs in algae and cyanobacteria ranging from 174
to 469. For the cyanobacteria (Spirulina snbsa/sa), the BCF for a-HBCD (350) was higher than the
BCFs for P-HBCD (270) and y-HBCD (174). However, for the tested alga (Scenedesmus obliquus), the
BCF for P-HBCD (469) was higher than that for the other isomers (390 - 407).
Bioconcentration
BCFs for HBCD in fish in the peer-reviewed literature range as high as 18,100, as shown in Appendix
C.2 (Zhang et al. 2014a; Wildlife Intl 2000; Veith et al. 1979). Drottar and Krueger (2000) provided
strong evidence that HBCD bioaccumulates in a bioconcentration test that was conducted according to
guidelines OECD Test Guideline (TG) 305 and Office of Prevention, Pesticides and Toxic Substances
(OPPTS) 850.1730. In this study, BCFs of 13,085 and 8,974 were reported in rainbow trout (O. mykiss)
exposed to 0.18 and 1.8 |ig/L, respectively. Concentrations of HBCD in tissue reached steady-state at
day 14 for fish exposed to 1.8 |ig/L and, during the subsequent depuration stage, a 50% reduction of
HBCD from edible and non-edible tissue and whole fish was reported on days 19 and 20 post-exposure.
In fish exposed to 0.18 |ig/L, an apparent steady-state was reached on day 21, but on day 35, the tissue
concentration of HBCD in fish increased noticeably; thus, steady-state was not achieved according to
study authors, and BCF values (for the exposure concentration of 0.18 |ig/L) were calculated based on
day 35 tissue concentrations. A kinetic BCF value 14039 for the 0.18 |ig/L exposure concentration was
calculated to address the possibility that steady state was not reached (ECH.A. 2008b). Clearance of 50%
HBCD from tissue of 0.18 |ig/L exposed fish occurred 30-35 days post-exposure.
Veith et al. (1979) further supports a conclusion that HBCD bioaccumulates in a study conducted prior
to the establishment of standardized testing guidelines for bioconcentration studies. The study reported a
BCF of 18,100 following exposure of fathead minnow to 6.2 |ig/L; the BCF was identified as a steady-
Page 545 of 723
-------
state BCF, but the report does not indicate time when steady- state was reached. A depuration phase was
not included in this study. Zhang et al. (2014a) calculated BCFs for each diastereomer in mirror carp and
found strong evidence that a-HBCD (BCF of 5,570-11,500) is much more bioaccumulative than P- and
y-HBCD (BCF of 187-642); BCF values that were normalized to lipid content were much higher
(30,700-45,200 for a-HBCD, 1,030-1,900 for P-HBCD, and 950-1,730 for y-HBCD) than non-
normalized BCFs.
Bioaccumulation
BAFs, which capture accumulation of HBCD from diet as well as water and sediment, were calculated
for freshwater food webs in industrialized areas of Southern China in two separate field studies. He et al.
(He et al. 2013) calculated log BAFs of 4.8-7.7 (corresponding to BAFs of 63,000-50,000,000) for
HBCD isomers in carp, tilapia, and catfish, and found higher BAFs for a-HBCD than P- and y-HBCD.
In a pond near an e-waste recycling site, Wu et al. (Wu et al. 2011) calculated log BAFs of 2.85-5.98 for
EHBCD (corresponding to BAFs of 700-950,000) in a freshwater food web. Log BAFs for each
diastereomer in this study were comparable to one another (see Appendix C.2). La Guardia et al. (La
Guardia et al. 2012) calculated log BAFs in bivalves and gastropods collected downstream of a textile
manufacturing outfall; these ranged from 4.2 to 5.3 for a- and P-HBCD (BAFs of 16,000-200,000), and
from 3.2 to 4.8 for y-HBCD (BAFs of 1,600-63,000).
Biota Sediment Accumulation
BSAFs calculated in studies of invertebrates and fish are generally lower than reported BCFs and BAFs.
Haukas et al. (2010b) reported BSAFs <0.006 calculated from lipid-normalized concentrations of HBCD
in ragworms and HBCD concentrations normalized to total organic content in sediment, indicating very
low bioavailability of HBCD from sediments. Ragworm tissue concentrations were all less than the limit
of detection. The pattern of diastereomers in sediments was found to generally resemble the composition
of technical HBCD (i.e., predominantly y-HBCD). This study also found that in ragworms exposed to
HBCD through a diet of contaminated mussels (containing diastereomer contributions of 48% a-HBCD,
7% P-HBCD, and 45% y-HBCD), the tissue concentration of a-HBCD was greater than that of P-HBCD
or y-HBCD, suggesting selective bioaccumulation of the a-diastereomer.
Log BSAFs calculated in bivalves and gastropods collected downstream of a textile manufacturing
outfall ranged from 0 to 0.9 (for a- and P-HBCD) and from -1.5 to 0 (for y-HBCD) (La Guardia et al.
2012). These correspond to BSAFs of 1-8 for a- and P-HBCD and 0.03-1 for y-HBCD. BSAFs in
benthivorous barbell (Barbus graellsii) and pelagic bleak (Alburnus alburnus) were calculated based on
measured concentrations of HBCD reported in Eljarrat et al. (2005; 2004) as cited in (van Beusekom et
al. 2006) and ranged from 0.1 to 1.44 and from 0.14 to 1.23, respectively (van Beusekom et al. 2006).
Biomagnification of HBCD was demonstrated by Law et al. (2006). who reported BMFs of 9.2 (a-
HBCD), 4.3 (P-HBCD), and 7.2 (y-HBCD). Uptake of HBCD into muscle from the diet of rainbow trout
was exponential for a-HBCD with a doubling time of 8.2 days, exponential for P-HBCD with a doubling
time of 17.1 days, and linear for y-HBCD with a rate constant of 0.006 per day. Depuration was rapid
during the first 14 days and slower for the remainder of the experiment for a-HBCD (overall depuration
rate was not determined). Depuration rates of 0.44x 10"2 and 0.48x 10"2 per day were found for P-HBCD
and y-HBCD, respectively. Steady- state was not reached for any of the diastereomers within the 52-day
exposure period.
Biomagnification
Additional studies are available that support the conclusion that HBCD has the potential to biomagnify.
Studies of zebrafish by Du et al. (Du et al. 2013; Du et al. 2012a) reported diastereo- and enantiomer-
Page 546 of 723
-------
specific biomagnification. When BMFs were calculated for diastereomers without accounting for
specific enantiomers, after 42 days of exposure and a 21-day depuration period, a-HBCD was shown to
biomagnify to a greater extent than P- and y-HBCD (maximum BMFs of 29.71, 11.63, and 7.76,
respectively). Enantiomer-specific BMFs calculated in zebrafish by Du et al. (2013) followed a similar
diastereomer pattern, although the BMF values were much lower than those from Du et al. ( ).
Additionally, the results of Du et al. (20! 3) suggest that the (+) enantiomers of P- and y-HBCD are
selectively magnified compared to their (-) enantiomers. This pattern did not hold true for a-HBCD.
Letcher et al. (2009) found evidence of biomagnification of HBCD from the ringed seal to the polar bear
in an East Greenland food web, reporting a BMF of 1.7. BMFs for a-HBCD in a harbor seal food web
varied according to prey fish species, but ranged from 0.54 to 3.0 (Shaw et al. 2012). Shaw et al. (2012)
calculated higher BMFs from prey fish to the livers of adult male harbor seals than to the blubber of
those seals.
BMFs for a-HBCD in gulls and common eiders in a coastal marine food web in Norway provide
evidence of biomagnification, ranging from 3.1 to 1,285 when calculated on a wet weight basis and from
2.8 to 26 when calculated on a lipid-weight basis (Haukas et al. 2.010a). In terrestrial food webs in
China, both Sun et al. ( ) and Yu et al. (2013) found evidence of biomagnification (see Appendix
C.2), with BMFs up to 30 in passerine birds and up to 16 in owls. Yu et al. (2013) found more (-) a-
HBCD in predator species than (+) a-HBCD, but other studies do not agree, suggesting that enantiomer
biomagnification may be species-specific.
Trophic Transfer/Trophic Magnification
Tomy et al.(2008) describes the extent of trophic transfer (transfer and accumulation of HBCD between
trophic levels) by calculating TMFs of 2.1 and 0.5 for a- and y-HBCD, respectively, based on the Arctic
marine food web. Samples of blubber were taken and analyzed from the beluga whale (Delphinapterus
leucas), narwhal (Monodon monoceros), and walrus (Odobenus rosmarus), while whole organisms were
analyzed for arctic cod (Boreogadus saida), shrimp (Pandalus borealis and Hymenodora glacialis),
clams (Mya truncate and Serripes groenlandica), deepwater redfish (Sebastes mentella), and mixed
zooplankton to determine HBCD concentrations in the tissue of animals of different trophic levels in
order to establish whether HBCD biomagnifies between trophic levels.
Brandsma et al. (2015) studied trophic magnification of HBCD through benthic and pelagic food webs
in the Western Scheldt estuary, The Netherlands, and found similar results: a-HBCD concentrations
increased and y-HBCD concentrations decreased with an increase in trophic level (TMFs of 2.2 and 0.3,
respectively). In a freshwater food web studied near an e-waste recycling site in South China, Wu et al.
Q ) calculated enantiomer-specific TMFs for a-HBCD of 2.18-2.2, and found evidence that as
HBCD migrates up through the food web, a-HBCD increases and y-HBCD decreases, while P-HBCD
comprises a very low proportion of £HBCD. This pattern, also demonstrated by data in Haukas et al.
(2010a). becomes more prominent at upper trophic levels. In marine and freshwater food webs, Zhang et
al- (2013) calculated TMFs greater than 1 for a-HBCD and EHBCD.
In summary, while HBCD has been shown in numerous studies to bioaccumulate and biomagnify in
aquatic and terrestrial food chains, diastereomer- and enantiomer-specific mechanisms of accumulation
are still unclear.
Page 547 of 723
-------
C.3 Calculation of Lipid Normalized Bioaccumulation Factors for
HBCD
The lipid normalized bioaccumulation factors were calculated for:
• He et al. (2013) using mean concentration for total HBCDs in field collected Nile tilapia and Plecostomus
expressed as lipid weight and total HBCD concentrations in the dissolved phase in water.
The lipid normalized BAF calculations are presented below where:
BAF = Cb/Cwd
Cb = chemical concentration in the organism (g/kg LW)
Cwd = freely dissolved chemical concentration in the water (g/L)
Sample
Mean concentration total
IIUCDs
Coin ersion
li.\l
Nile tilapia
92 ng/g lw
Cb = 9.2e-5 g/kg
2.32E6
Plecostomus
361 ng/g lw
Cb = 0.000361 g/kg
9.09E6
Mud carp
58.3 ng/g lw
Cb = 5.83e-5 g/kg
1.47E6
Water,
dissolved phase
39.7 pg/L
Cwd = 3.97e-l 1 g/L
n/a
Underlying data:
Table 1
Concentrations ofTBBFA arid HBCDs in sedi|ment. sediment cores, water, and fish in the Dongjiang River, South China.
Sediment
Sediment cores (ng/g dw)
Water
Fish (ng/g lw)
(ng/g dw)
Core 1
Core 2
Dissolved phase
Particulate phase
Mud carp
Nile tilapia
Plecostomus
N -42
N = 19
N = 19
JV ™ 5 (pg/L)
N ~ 5 (ng/g dw)
N = 9
N - 15
N- 10
Lipid I
IBblW
s-HBU)
P-hbld
.-HBCD
15.2 (nd—82.3}
034 (nd—3.3)
0,33 (nd—2.2)
5.7 (0.08-28.9)
6.9 (0.07—31,6)
91.6 (7.9-4501
lbr0^-S„)
U" U l-22l
62 '0 8-42 8*
e>5'05-r3 31
2.9(0.2-14)
as 10.2-2.1)
0.3 {0.1-0-8}
4.2 \ 0.3-^0.5)
5Ji 1.2-22.6)
1750(1110-2830)
lb Or 5-27.6)
4 : I nd-7 1)
2b^ ind-54.6)
Jl) 7 19 5—82.4)
1.3 (nda-1.6)b
3.9 (nd-3.9)
0.9 (fid—0.9)
6.8 (fid—11..3)
8.0 (nd-113)
2.7(1.2-4.7)
352 (6.5-66)
47.3(17.5-114)
3.6 (nd-6.7)
12,6 (nd-33.4)
58.3(175-154)
<7 1 1-M01
1B I rd-51)
nd—3b 1)
"" b i id—19 5)
6 ' (nd-11 4)
(i~ nd- '5)1 j
3 2(1 b-i 71
21 2 ind—5J 4}
35jfnd—823>
b 2(nd—9 2 J
6 5 ind-10 4}
361 1 nd-832^
a Non detected.
b Mean (range).
Page 548 of 723
-------
Wu et al. ("2010') using mean concentration for total HBCDs in field collected mud carp and Northern snakehead
expressed as lipid weight and total HBCD concentrations in the dissolved phase in water.
Sample
Mean concentration total
IIUCDs
Coin ersion
li \l
Mud Carp
868 ng/g lw
Cb = 0.000868 g/kg
1.45E7
Northern
Snakehead
187 ng/g lw
Cb = 0.000187 g/kg
3.12E6
Water,
dissolved phase
0.06 ng/L
Cwd = 6e-l 1 g/L
n/a
Underlying data:
TABU 1,
i If ilCls aid Other lan-Plll BroutiMl
«l Bane R«t»i
rliitS m tie
Auntie Specie
* Wl
lisiihel
1
a
1
1
i
m
iilMSlt i«,|| .
let wt) Ireii * E-wasti Itefeliiii Site, Sunt! Ciiitii
China*
mirth ein
BjystMjsiuil
a - «, IJj
prawn
Bxid carp
iiucidiiiani
iiMkeltwd
watw snake
watei
s^tti IHfP flf
B - ? 13]
n - 12, [8]
» = 18,17]
a = 6
a — 1
»-6. n
il = l,*[3]
lipid l°rl
0 5a 1 0 11!
2 39 - 0 V
zei_ 0 4i
?t~3 - C 71
1 49 > *) 91
1 0B . 0 15
i, HBCD
7 73 i 1 33
237 r SO 9
640 ^22S
102 ; 32 6
163 * S2 4
434^ 315
'1 UC • J u1
61,4 _ 1G 2
, HBCD
C 24 ^ 0 24
10 2 _= 2 74
24 R 9 48
G 42 ji 2 39
2 k2 * 7fi
8 76 i-1 22
lull
23 5 _ 1,07
-HBCD
5Vi1 25
ite 12« 3
„ fit y
21 ' -r 5 81
Iff ! 9 Gp
R4 J - 42 B
0 01 0 (AI
84,3 - 4 22
rnecDs"
13 3 1 2 61
39E i S4 5
8G8 280
120 _ 44?
187 „ 32 7
r-67^ 364
(i i% t n o1
169 12 1
BTBPE
07 1 j 36 4
J4.7 t S 04
518 _ 277
D23 _ JIB
1 71 ^ 1 11
9 22 ^ CJ 22
0 02 ^ r 01
4554 - 60S
DBDPE
bdl
84 ^ - &4 3
5J8- 171
M0 — 14 K
Lidl
hdl
lirtl
1796 r 770
HBB
2m - 511
197 - 55 4
2451 - 778
68C •
11H - 474
30U9 230?
it 52 r 04
8872 1ft ,.3
PBEB
U,: L 2 ^3
6 31 2 52
25 fi _ 1 , 1
3'.I? i 2 10
17 E* 5 0?
4 14 i 4 14
CUP ti no
132 i 6,12
PBT
3 60 » U JO
1 5^ ' 0 64
? 24 - 1 51
' 59 - C 45
1.20 Jr 0 57
".0* Hi 3
4 cu -cm
20,6 - 2.B9
BDE 4?"
4270 + S2e
4M0 - 1230
2091U - T74U
r8M) r1480
259fil> * 4K20
51B7u - &340
10.7 ±0,14
44130 t ~>17
k'um^j rif jrvhvidusl samplfs crliea^il fiji-res m ref 30 g
-------
Appendix D RELEASES TO THE ENVIRONMENT
D.l 2017 TRI Releases Not Used in this Assessment
Table Apx D-l. presents 2017 TRI data that was not used in this assessment. These HBCD release data
were reported by Flame Control Coatings, LLC for one site that previously used HBCD as a component
in flame retarded coatings. These TRI releases were not used in the assessment because Flame Control
Coatings, LLC has indicated that they have ceased use of HBCD and the use of coatings is not an
exposure scenario in this final Risk Evaluation, as discussed in Section 1.2.4 of this final Risk
Evaluation.
Table Apx D-l. 2017 TRI Data Not Used in this Assessment
Site Identity
Reported NAICS Code -
Meaning
Function Inferred
from
Communication
with Company
Annual HBCD Release per
Site (kg/site-year)
Flame Control
Coatings,
LLC, Niagara
NY
325510 - Paint and Coating
Manufacturing
Flame retardant in
architectural
coatings
Fugitive aira: 0.612
Stack air b: 5.505
a These fugitive air releases were reported under Section 5.1 of the TRI Form R, which correspond to on-site fugitive or non-
point air emissions.
b These stack air releases were reported under Section 5.2 of the TRI Form R, which correspond to on-site stack or point air
emissions.
Page 550 of 723
-------
D.2 Evaluation of Environmental Release Data Sources
EPA has reviewed acceptable sources for HBCD release data according to the data quality evaluation
criteria found in The Application of Systematic Review in TSCA Risk Evaluations (I. Sb).
Table Apx D-2 summarizes the results of this evaluation. The data quality evaluation indicated the
release sources included are of medium to high confidence and are used to characterize releases of
HBCD.
Page 551 of 723
-------
Table Apx D-2. Summary of Release Data and Systematic Review Results
Release Data from Source
Data Identifier
from Data
Extraction
and
Evaluation
(DEE)
Overall
Confidence
Rating from
DEE
Rationale for
Row
Exposure scenario
Identifier
Release
Source
Inclusion /
Exclusion
1
Compounding of
Polystyrene Resin to
Produce XPS Masterbatch
Site 1
Water: 0.12 kg HBCD/yr
Air: 2.6 kg HBCD/yr
(ECHA
2008b)
3970747
High
Included - EPA
calculated
emission factors
2
Compounding of
Polystyrene Resin to
Produce XPS Masterbatch
Site 2
Water: 0.27 kg HBCD/yr
Air: 1.2 kg HBCD/yr
(ECHA
2008b)
3970747
High
from these data
and used them to
estimate releases
3
Compounding of
Polystyrene Resin to
Produce XPS Masterbatch
Site 3
Water: 37 kg HBCD/yr
Air: 3.3 kg HBCD/yr
(ECHA
2008b)
3970747
High
in the
corresponding
exposure
scenario
4
Manufacturing of XPS Foam
using XPS Masterbatch
Site 1
Water: 2.2 kg HBCD/yr
Air: 0.31 kg HBCD/yr
(ECHA
2008b)
3970747
High
5
Manufacturing of XPS Foam
using XPS Masterbatch
Site 2
Water: 0 kg HBCD/yr
Air: 18 kg HBCD/yr
(ECHA
2008b)
3970747
High
Included - EPA
calculated
6
Manufacturing of XPS Foam
using XPS Masterbatch
Site 3
Water: 1.3 kg HBCD/yr
Air: 14 kg HBCD/yr
(ECHA
2008b)
3970747
High
emission factors
from these data
7
Manufacturing of XPS Foam
using XPS Masterbatch
Site 4
Water: 4.2 kg HBCD/yr
Air: 9.3 kg HBCD/yr
(ECHA
2008b)
3970747
High
and used them to
estimate releases
Calculated Site
in the
8
Manufacturing of XPS Foam
using XPS Masterbatch
Estimate - reported by
EURAR as worst-case
emission factor
derived from site-
specific data
Water: 7.9 kg HBCD/yr
Air: 17.4 kg HBCD/yr
(ECHA
2008b)
3970747
High
corresponding
exposure
scenario
9
Manufacturing of XPS Foam
using HBCD Powder
Site 1
Water: 4.4 kg HBCD/yr
Air: 1.5 kg HBCD/yr
(ECHA
2008b)
3970747
High
10
Manufacturing of XPS Foam
using HBCD Powder
Site 2
Water: 1.2 kg HBCD/yr
Air: 1.4 kg HBCD/yr
(ECHA
2008b)
3970747
High
Included - EPA
11
Manufacturing of XPS Foam
using HBCD Powder
Site 3
Water: 0.055 kg HBCD/yr
Air: 3.7 kg HBCD/yr
(ECHA
2008b)
3970747
High
calculated
emission factors
12
Manufacturing of XPS Foam
using HBCD Powder
Site 4
Water: 3.7 kg HBCD/yr
Air: 1.5 kg HBCD/yr
(ECHA
2008b)
3970747
High
from these data
and used them to
13
Manufacturing of XPS Foam
using HBCD Powder
Site 5
Water: 0.0024 kg HBCD/yr
Air: 1.1 kg HBCD/yr
(ECHA
2008b)
3970747
High
estimate releases
in the
14
Manufacturing of XPS Foam
using HBCD Powder
Site 6
Water: 0 kg HBCD/yr
Air: 0.73 kg HBCD/yr
(ECHA
2008b)
3970747
High
corresponding
exposure
15
Manufacturing of XPS Foam
using HBCD Powder
Site 7
Water: 6 kg HBCD/yr
Air: 0.54 kg HBCD/yr
(ECHA
2008b)
3970747
High
scenario
16
Manufacturing of XPS Foam
using HBCD Powder
Site 8
Water: 0.0029 kg HBCD/yr
Air: 0.7 kg HBCD/yr
(ECHA
2008b)
3970747
High
Page 552 of 723
-------
Release Data from Source
Data Identifier
from Data
Extraction
and
Evaluation
(DEE)
Overall
Confidence
Rating from
DEE
Rationale for
Row
Exposure scenario
Identifier
Release
Source
Inclusion /
Exclusion
17
Manufacturing of XPS Foam
using HBCD Powder
Site 9
Water: 0.0019 kg HBCD/yr
Air: 0.15 kg HBCD/yr
(ECHA
2008b)
3970747
High
18
Manufacturing of XPS Foam
using HBCD Powder
Site 10
Water: 0 kg HBCD/yr
Air: 0.4 kg HBCD/yr
(ECHA
2008b)
3970747
High
19
Manufacturing of XPS Foam
using HBCD Powder
Site 11
Water: 0 kg HBCD/yr
Air: 1.8 kg HBCD/yr
(ECHA
2008b)
3970747
High
20
Manufacturing of XPS Foam
using HBCD Powder
Site 12
Water: 0 kg HBCD/yr
Air: 1.8 kg HBCD/yr
(ECHA
2008b)
3970747
High
21
Manufacturing of XPS Foam
using HBCD Powder
Site 13
Water: 0.11 kg HBCD/yr
Air: 1.2 kg HBCD/yr
(ECHA
2008b)
3970747
High
22
Manufacturing of XPS Foam
using HBCD Powder
Site 14
Water: 15 kg HBCD/yr
Air: 1.5 kg HBCD/yr
(ECHA
2008b)
3970747
High
23
Manufacturing of XPS Foam
using HBCD Powder
Site 15
Water: 0.00004 kg HBCD/yr
Air: 0.59 kg HBCD/yr
(ECHA
2008b)
3970747
High
24
Manufacturing of XPS Foam
using HBCD Powder
Site 16
Water: 0.0004 kg HBCD/yr
Air: 0.91 kg HBCD/yr
(ECHA
2008b)
3970747
High
25
Manufacturing of XPS Foam
using HBCD Powder
Site 17
Water: 0.021 kg HBCD/yr
Air: 3.8 kg HBCD/yr
(ECHA
2008b)
3970747
High
26
Manufacturing of XPS Foam
using HBCD Powder
Site 18
Water: 2.5 kg HBCD/yr
Air: 0.23 kg HBCD/yr
(ECHA
2008b)
3970747
High
27
Manufacturing of XPS Foam
using XPS Masterbatch;
Manufacturing of XPS Foam
using HBCD Powder;
Manufacturing of EPS Foam
from Imported EPS Resin
Beads
Dow Chemical
Company, Pevely MO
Stack air: 1.81 kg HBCD/yr
Off-site transfer for Incineration/thermal
treatment: 30.8 kg HBCD/yr
Off-site M64, off-site transfer for disposal to
other landfills: 123 kg HBCD/yr
(U.S. EPA
2017g)
2017 TRI
Medium
Included - per the
company,
operations with
HBCD have
ceased. Data is
used as surrogate
for unidentified
site
28
Manufacturing of XPS Foam
using XPS Masterbatch;
Manufacturing of XPS Foam
using HBCD Powder;
Manufacturing of EPS Foam
from Imported EPS Resin
Beads
Dow Chemical
Company, Dalton GA
Stack air: 21.3 kg HBCD/yr
Off-site M64, off-site transfer for disposal to
other landfills: 109 kg HBCD/yr
Off-site M56, off-site transfer for Energy
Recovery: 23.1 kg HBCD/yr
(U.S. EPA
2017g)
2017 TRI
Medium
Included - per the
company,
operations with
HBCD have
ceased. Data is
used as surrogate
for unidentified
site
29
Manufacturing of SIPs and
Automobile Replacement
Parts from XPS/EPS Foam;
Installation of XPS/EPS
XPS Boards
5 g XPS particles/metric ton XPS sawed
(ECHA
2008b)
3970747
High
Included -
emission factors
were used in the
corresponding
Page 553 of 723
-------
Row
Exposure scenario
Release Data from Source
Source
Data Identifier
from Data
Extraction
and
Evaluation
(DEE)
Overall
Confidence
Rating from
DEE
Rationale for
Inclusion /
Exclusion
Identifier
Release
Foam Insulation in
Residential, Public and
Commercial Buildings, and
Other Structures
exposure
scenarios
30
Manufacturing of SIPs and
Automobile Replacement
Parts from XPS/EPS Foam;
Installation of XPS/EPS
Foam Insulation in
Residential, Public and
Commercial Buildings, and
Other Structures
EPS Boards
445 g EPS particles/metric ton EPS sawed
(ECHA
2008b)
3970747
High
31
Manufacturing of SIPs and
Automobile Replacement
Parts from XPS/EPS Foam;
Installation of XPS/EPS
Foam Insulation in
Residential, Public and
Commercial Buildings, and
Other Structures
EPS Boards
100 g EPS particles/metric ton EPS cut
(ECHA
2008b)
3970747
High
32
Demolition and Disposal of
XPS/EPS Foam Insulation
Products in Residential,
Public and Commercial
Buildings, and Other
Structures
Manual breaking of
EPS boards
90 g EPS particles/metric ton EPS broken
(ECHA
2008b)
3970747
High
Included -
emission factor
was used in
corresponding
exposure
scenario
33
Demolition and Disposal of
XPS/EPS Foam Insulation
Products in Residential,
Public and Commercial
Buildings, and Other
Structures
Manual breaking of
XPS boards
0 g XPS particles/metric ton XPS broken
(ECHA
2008b)
3970747
High
Included -
emission factor
was used in
corresponding
exposure
scenario
34
Formulation of Coatings
Flame Control
Coatings LLC,
Niagara NY
Fugitive air: 0.612 kg HBCD/yr
Stack air: 5.505 kg HBCD/yr
(U.S. EPA
2017g)
2017 TRI
Medium
Excluded - this
data is presented
in Appendix D. 1,
but this is not an
exposure
scenario
35
Formulation of Solder/Flux
Pastes
Indium Corporation of
America, Clinton, NY
Fugitive air: 0.454 kg HBCD/yr
Stack air: 6.350 kg HBCD/yr
Waste broker for disposal: 0.454 kg HBCD/yr
(U.S. EPA
2017g)
2017 TRI
Medium
Included - loss
quantity was
used in the
Page 554 of 723
-------
Row
Exposure scenario
Release Data from Source
Source
Data Identifier
from Data
Extraction
and
Evaluation
(DEE)
Overall
Confidence
Rating from
DEE
Rationale for
Inclusion /
Exclusion
Identifier
Release
Treatment via solidification/stabilization: 6.350
kg HBCD/yr
corresponding
exposure
scenario
Page 555 of 723
-------
Appendix E OCCUPATIONAL EXPOSURES
E.l Inhalation Monitoring Data Summary
This appendix contains a summary of the available data that EPA compiled from literature sources.
EPA compiled HBCD inhalation monitoring data that was available in literature into three tables based
on the associated worker activities:
• TableApx E-l contains inhalation monitoring data related to the handling of HBCD in various
forms, including fine grade powder, standard grade powder, and granules.
• Table Apx E-2 contains inhalation monitoring data related to the handling and processing of XPS
and EPS foam containing HBCD.
Page 556 of 723
-------
Table Apx E-l. Inhalation Monitoring Data for Handling of HBCD
Literature
Study a
Exposure
scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or Sampling
Location
Exposure
Concentration
(mg/m3) b
Number of
Samples
Sample Time
/ Type of
Measurement0
Source
Overall
Confidence
Rating
Searl and
Robertson
(2005) - la
Manufacturing
of HBCD
Standard
grade HBCD
Personal
Breathing
Zone
Packaging, compaction, process
operations, and working in the
warehouse
Mean: 1.23
Median: 0.89
90th
percentile:
1.89
Max: 3 mg/m3
10
8-hr TWA
(ECHA
2008b)
(ECHA
2009b)
High
Searl and
Robertson
(2005)- lb
Manufacturing
of HBCD
Fine grade
HBCD
Personal
Breathing
Zone
Packaging, compaction, process
operations, and working in the
warehouse
Mean: 23
90th
percentile: 35
4
8-hr TWA
(ECHA
2008b)
High
Searl and
Robertson
(2005) - lc
Manufacturing
of HBCD
HBCD of
unknown
grade
NR
Packaging and compaction of
powders
Respirable,
mean: 0.18
Inhalable,
Mean: 1.23
NR
NR
(ECHA
2009c)
High
Waindzioch
(2000) - la
Manufacturing
of HBCD
HBCD of
unknown
grade
Area
Reactor
0.00028 -
0.0285
3
Short-term
(ECHA
2008b)
Unacceptable
Waindzioch
(2000)- lb
Manufacturing
of HBCD
HBCD of
unknown
grade
Area
Filling Station
0.0094 - 0.097
2
Short-term
(ECHA
2008b)
High
Biesemeier
(1996)
Manufacturing
of HBCD
HBCD of
unknown
grade
NR
Bagging HBCD product
4.0-4.5
NR
NR
(ECHA
2008b)
High
Velsicol
(1978)
Manufacturing
of HBCD
HBCD of
unknown
grade
Personal
Breathing
Zone
Transfer of the HBCD in the
hammer-mill to 28 drums
1.9
1
300 minutes
(Velsicol
Chem Coro
1978)
High
Yi et al.
(2016)
Manufacturing
of HBCD
HBCD of
unknown
grade
Personal
Breathing
Zone
NR
0.0102-
0.0283
14
NR
(Yi et al.
2016)
High
Searl and
Robertson
(2005) - 2a
Manufacturing
of EPS Resin
beads
Standard
grade HBCD
Personal
Manual addition of HBCD
powder to reactor each time a
batch of EPS resin was produced
Range: 2.89-
21.5
Mean: 7.2
Median: 5.52
12
Short-term
(13 to 56
mins)
(NICNAS
2012b);
(ECHA
2008b)
High
Page 557 of 723
-------
Literature
Study a
Exposure
scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or Sampling
Location
Exposure
Concentration
(mg/m3) b
Number of
Samples
Sample Time
/ Type of
Measurement0
Source
Overall
Confidence
Rating
90th
percentile:
10.5
Searl and
Robertson
(2005) - 2b
Manufacturing
of EPS Resin
beads
Standard
grade HBCD
Personal
Manual addition of HBCD
powder to reactor each time a
batch of EPS resin was produced
Range: 0.12-
3.36
Mean: 1
Median: 0.42
90th
percentile:
1.11 (NICNAS
2012b); 1.3
(ECHA 2008b)
12
8-hr TWA -
note these are
8-hr TWA
values of the
data in the
above row
(NICNAS
2012b):
(ECHA
2008b)
High
Searl and
Robertson
(2005) - 2c
Manufacturing
of EPS Resin
beads
Standard
grade HBCD
Personal
Manual addition of HBCD
powder to reactor each time a
batch of EPS resin was produced
Range: 0.07-
14.7
Mean: 1.2
Median: 0.27
90th
percentile:
1.10
18
8-hr TWA
(ECHA
2008b): 275
to 504 mins
(NICNAS
2012b)
(NICNAS
2012b):
(ECHA
2008b)
High
Searl and
Robertson
(2005) - 2d
Manufacturing
of EPS Resin
beads
Standard
grade HBCD
Personal
Weighing powder prior to
addition to reactor. HBCD bags
were weighed and opened
concurrently, or weighed in
advance, in which case HBCD
was transferred from 25-kg sacks
using plastic scoop (full-shift
measurement).
Range: 4.35-
12.1
Mean: 7.2
Median: 6.19
90th
percentile:
10.5 (NICNAS
2012b): 10.5 &
10.6 (ECHA
2008b)
4
8-hr TWA
(ECHA
2008b): 124
to 350 mins
(NICNAS
2012b)
(NICNAS
2012b):
(ECHA
2008b)
High
Searl and
Robertson
(2005) - 3a
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
Area
Weighing and mixing
Max 7.5 (for 2
hours)
Mean: 1.89
Median: 0.83
90th
percentile: 5.4
10
Short-term
(ECHA
2008b),
(ECHA
2009b)
High
Page 558 of 723
-------
Literature
Study a
Exposure
scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or Sampling
Location
Exposure
Concentration
(mg/m3) b
Number of
Samples
Sample Time
/ Type of
Measurement0
Source
Overall
Confidence
Rating
Searl and
Robertson
(2005) - 3b
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
Area
Weighing and mixing
Mean: 0.88
90th
percentile:
1.36
10
8-hr TWA
(ECHA
2008b)
High
Searl and
Robertson
(2005) - 3c
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
NR
Extruder
Mean: 0.12
Median: 0.10
90th
percentile:
0.16
4
5 hours
(ECHA
2008b),
(ECHA
2009b)
High
Searl and
Robertson
(2005) - 3d
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
NR
Automated handling of HBCD
Negligible
3
NR
(ECHA
2008b)
High
Abbott
(2001) - la
Manufacture
of XPS from
HBCD powder
or granules
Standard
grade HBCD
Area
At the feed deck near typical
operator positions
Range 0.24 -
1.6
Mean: 0.66
90th
percentile:
1.45
(excluding 10
ND samples)
16 (10 ND)
8-hr TWA
(ECHA
2008b)
High
Abbott
(2001) - lb
Manufacture
of XPS from
HBCD powder
or granules
HBCD
granules
Mostly area
and some
personal
breathing
zone
Feed deck near typical operator
positions
Range 0.005-
0.9
Mean: 0.24
90th
percentile:
0.47
43 (16 ND)
60 - 1435
minutes
(ECHA
2008b)
High
Page 559 of 723
-------
Literature
Study a
Exposure
scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or Sampling
Location
Exposure
Concentration
(mg/m3) b
Number of
Samples
Sample Time
/ Type of
Measurement0
Source
Overall
Confidence
Rating
(excluding 16
ND samples)
Thomsen
(2007) - la
Manufacture
of XPS from
HBCD powder
or granules
HBCD
powder and
granules
Personal
breathing
zone
Activities in the mixer area,
including operating a closed
automated process excluding
potential contact with neat HBCD
Range: 0.0002-
0.0009
Mean: 0.0005
Median:
0.0005
6
8-hr TWA
(ECHA
2008b)
(NICNAS
2012b)
High
Thomsen
(2007) - lb
Manufacture
of XPS from
HBCD powder
or granules
HBCD
powder and
granules
Personal
breathing
zone
Weighing and addition of HBCD
to the reactor and subsequent
washing, centrifugation, sifting,
and transfer of product to a silo
container
Range: 0.001-
0.15
Mean: 0.015
Median:
0.0027
24
8-hr TWA
(ECHA
2008b)
(NICNAS
2012b)
High
Searl and
Robertson
(2005) - 4
Manufacture
of XPS from
HBCD powder
or granules
HBCD
granules
Area
Logistics, extruding, and
laboratory
Mean: 0.00003
90th
percentile:
0.00004
12
8-hr TWA
(ECHA
2008b)
High
Ransbotyn
(2000)
Manufacturing
of EPS Resin
beads
Respirable
Dust
Inhalable
Dust
Personal
Addition of HBCDD to reactor or
the supervising of the addition.
Respirable
dust: <0.5
Total Inhalable
dust: 2.0
Not specific to
HBCD
5
Max 8-hr
TWA
(ECHA
2008b)
High
NICNAS
(2012b)-
la
All industrial
polymer
processing
sites
Standard
grade HBCD
Modelled
with EASE
Addition of HBCD into process
operation
Typical: 2 to 5
Worst-case: 5
to 50
N/A - this is
a modelled
exposure
8-hr TWA
(NICNAS
2012b)
High
NICNAS
(2012b)-
lb
HBCD
importation /
repackaging
sites and all
industrial
polymer
processing
sites
HBCD
granules
Modelled
with EASE
Repackaging with the use of LEV
(typical) and without LEV (worst-
case)
Typical: 0.2 to
0.5
Worst-case:
0.5 to 5
N/A - this is
a modelled
exposure
8-hr TWA
(NICNAS
2012b)
High
NR = Not Reported; N/A = Not Applicable
Page 560 of 723
-------
a - Where multiple datasets were available from one literature source, EPA distinguished data as la, lb, 2a, 2b, etc.
b - Statistics were calculated by the cited source and are presented here as they were presented in the source.
c - Where information is presented in multiple sources all sources are listed. Information was not combined from these sources but was presented in all sources
independently.
Table Apx
1-2. Inhalation Monitoring Data
7or Handling of XPS and EPS Foam Containing HBCD
Literature
Study a
Exposure
scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or
Sampling Location
Exposure
Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Sourcec
Overall
Confidence
Rating
Searl and
Robertson
(2005) - 5a
Manufacture of
XPS fromXPS
Masterbatch
XPS foam
NR
Secondary processing of
XPS foam - including
cutting, sawing, and
machining to manufacture
shaped products
Mean: 0.08
90th percentile:
0.22 d
9
8-hr TWA
Original
source: Searl
and Robertson
(2005)
Reported in:
(ECHA
2008b);
(ECHA
2009b)
High
Searl and
Robertson
(2005) - 5b
Manufacture of
XPS fromXPS
Masterbatch
XPS foam
NR
Reclamation of XPS foam
- including shredding and
reprocessing of process
waste
Mean: 0.02
90th percentile:
0.02 d
5
8-hr TWA
Original
source: Searl
and Robertson
(2005)
Reported in:
(ECHA
2008b);
(ECHA
2009b)
High
Searl and
Robertson
(2005) - 5c
Manufacture of
XPS fromXPS
Masterbatch
XPS foam
NR
Other process control
operators
Mean: 0.03
90th percentile:
0.03 d
4
8-hr TWA
Original
source: Searl
and Robertson
(2005)
Reported in:
(ECHA
2008b);
(ECHA
2009b)
High
Page 561 of 723
-------
Literature
Study a
Exposure
scenario
Form of
HBCD
Handled
Type of
Sample
Worker Activity or
Sampling Location
Exposure
Concentration
(mg/m3) b
Number
of
Samples
Sample Time /
Type of
Measurement
Sourcec
Overall
Confidence
Rating
Searl and
Robertson
(2005) - 5d
Manufacture of
XPS from XPS
Masterbatch
XPS foam
NR
Process operators
handling XPS masterbatch
Mean: 0.03
90th percentile:
0.03 d
24
8-hr TWA
Original
source: Searl
and Robertson
(2005)
Reported in:
(ECHA
2008b);
(ECHA
2009b)
High
Zhang et al.
(2012) - la
Thermal cutting
of XPS boards
XPS foam
NR
Thermal cutting of XPS
boards in a closed
glovebox
Mean: 0.089
NR
NR
(Zhane et al.
2012)
High
Zhang et al.
(2012)- lb
Thermal cutting
of EPS boards
EPS foam
NR
Thermal cutting of EPS
boards in a closed
glovebox
Mean: 0.057
NR
NR
(Zhane et al.
2012)
High
NR = Not Reported
a - Where multiple datasets were available from one literature source, EPA distinguished data as la, lb, 2a, 2b, etc.
b - Statistics were calculated by the cited source and are presented here as they were presented in the source.
c - Where information is presented in multiple sources all sources are listed. Information was not combined from these sources but was presented in all sources
independently.
d - These exposure values were all originally reported in the same study, Searl and Robertson (2005). and discussed in the EURAR (ECHA 2008b) and an ECHA report
(ECHA 2009b). The dataset includes 42 total samples, taken at three XPS manufacturing sites in the EU. The EURAR reports that the first two rows, consisting of 14 total
data points, include all non-detects, except for three samples, indicating that the exposure potential during these activities is low, despite the fact that the exposure
concentrations in Searl and Robertson (2005) - 5a are the highest of the surveyed activities.
Page 562 of 723
-------
E.2 Summary of Other Assessment Approaches
EPA identified three HBCD risk assessments from other countries. These include:
• European Union (EU) - Risk Assessment, Hexabromocyclododecane (EC 38 b)
• Australian Government Department of Health and Ageing, National Industrial Chemicals
Notification and Assessment Scheme (NICNAS) - Priority Existing Chemical Assessment
Report No. 34, Hexabromocyclododecane (NICNAS 20.1.2b)
• Environmental Canada (EC), Health Canada - Screening Assessment Report on
Hexabromocyclododecane (EC/HC 20.1.0
• Note that this RAR only includes release assessments during raw materials handling and
compounding and does not assess occupational exposures.
EPA compiled the assessment approaches from the above three sources for each exposure scenario
assessed in this assessment below. TableApx E-3 and TableApx E-4 specifically list the inhalation
exposure assessment methodology in the EU and NICNAS RARs, respectively, lists methodology for
oral and dermal exposure, as well as environmental release assessment methodology.
Page 563 of 723
-------
TableApx E-3. Summary of HBCD Occupational Inhalation Exposure Assessment Results and the Associated Assessment Basis and
Assessment Approach that are Reported in EU (2008)
Assessment Parameter
Chemical Process: Manufacture of HBCD
HBCD standard grade
powder
RWC:
1.9 mg/m3
The basis is the worker exposure
monitoring data for the manufacture of
HBCD that are reported in Searl and
Robertson (2005) - lao f Table Adx E-l
of this report.
The rationale is that this is the only
worker exposure monitoring data for
The RWC exposure concentration was assessed to be
equal to the 90th percentile of the concentration
measurements referenced under Basis.
Typical exposure concentration: refer to footnote (1).
Typical:
0.95 mg/m3
Exposure
Concentration
HBCD granules
RWC:
0.19 mg/m3
HBCD manufacturing that is specifically
associated with the HBCD standard grade
powder product.
This data were also used as the basis for
the assessment of exposure
concentrations in the case of the HBCD
granules product.
The typical exposure concentration was assumed to
be equal to 10 percent of the RWC exposure
concentration that was assessed in the case of the
HBCD standard grade powder product. The rationale
for this assumption is that 10 percent of particles in
the HBCD granules product were assumed to have a
size of less than 100 |im. which is the assumed
maximum particle size for HBCD standard grade
powder.
Typical exposure concentration: refer to footnote (1).
Chemical Process: Compounding of Polystyrene Resin to Produce XPS Masterbatch; Manufacture of XPS from HBCD powder, granules, or XPS masterbatch; and
Manufacture of EPS resin beads
Exposure
Concentration
HBCD standard grade
powder
RWC:
2.5 mg/m3
The basis is the worker exposure
monitoring data for the manufacture of
EPS resin that are reported in Searl and
Robertson (2005) - 2a-d of Table Adx
E-l of this report.
The rationale is that this data is based on a
greater number of samples.
The RWC exposure concentration was assessed by
accounting for both addition and weighing as follows:
1. Addition of HBCD - the 90th percentile
value, 1.3 mg/m3 (Searl and Robertson
(2005) - 2b). was used.
2. Weighing of HBCD - the 90th percentile
value, 10.5 mg/m3 (Searl and Robertson
(2005)- 2d), was used. This task is 10-15
percent of the long-term working time due to
task rotation and therefore, only a fraction of
this concentration was assessed (~10 percent
or 1.1 mg/m3).
Typical:
1.2 5 mg/m3
Page 564 of 723
-------
Assessment Parameter
The RWC concentration used in this exposure
assessment is the sum of 1.3 mg/m3 and 1.1 mg/m3,
which is approximately equal to 2.5 mg/m3.
Typical exposure concentration: refer to footnote (1)
HBCD granules
RWC:
0.22 mg/m3
Typical:
0.11 mg/m3
The basis is the monitoring data for the
manufacture of XPS from HBCD
granules that are reported in Abbott
(2001) - lb of Table AoxE-l of this
report.
The approach is not explained beyond that the data
referenced under Basis is more representative than
other similar data (i.e.. Thomsen (2007) - la-bof
Table Apx E-l) and that more emphasis on personal
sampling was given in selecting an assessed value.
Typical exposure concentration: refer to footnote (1).
master batch
RWC:
0.22 mg/m3
Typical:
0.11 mg/m3
The basis is the monitoring data for the
manufacture XPS from master batch that
are reported in Searl and Robertson
(2005) - 3a-dof Table Adx E-l of this
report.
The RWC exposure concentration was assessed to be
equal to the 90th percentile of the concentration
measurements referenced under Basis.
Typical exposure concentration: refer to footnote (1).
Source: (ECHA 2008b) European Chemicals Agency. Risk Assessment forHexabromocyclododecane: Final Report. May 2008.
RWC - Reasonable Worst Case
1 Typical concentration was assessed to be equal to one half of the assessed RWC concentration. The rationale for this approach is that measured data indicates that the
median value is approximately half the RWC.
Page 565 of 723
-------
TableApx E-4. Summary of HBCD Occupational Exposure Assessment Results and the Associated Assessment Basis and Approach
that are Reported in NICNAS (2012)
Assessment Parameter
Chemical Process: Compounding of Polystyrene Resin to Produce XPS Masterbatch, Manufacture of XPS from HBCD powder or granules. Manufacture of XPS from
XPS Master Batch, and Manufacture of EPS Resin
HBCD standard grade
RWC:
1.1 mg/m3 (addition)
10.5 mg/m3
(weighing)
The basis is the worker exposure
monitoring data for the manufacture of
EPS resin that are reported in Searl and
Robertson (2005) - 2b (for addition) and
Searl and Robertson (2005) - 2d (for
weighing) of Table Apx E-l of this
The RWC exposure concentration was assessed to be
equal to the 90th percentile of the concentration
measurements referenced under Basis.
Exposure
Concentration
powder
Typical:
0.27 mg/m3
(addition)
6.19 mg/m3
(weighing)
report.
Overseas measurements were considered
applicable due to similarities in tasks. Use
of the full-shift measurements for
addition is preferred.
The typical exposure concentration was assessed to be
equal to the median of the concentration
measurements referenced under Basis.
HBCD granules and
XPS masterbatch
RWC:
0.37 mg/m3
The basis is the worker exposure
monitoring data for manufacture of XPS
from HBCD granules that are reported in
Abbott (2001) - lb of Table AoxE-l of
this report.
The RWC exposure concentration was assessed to be
equal to the 90th percentile value referenced under
Basis.
Typical:
0.08 mg/m3
The typical exposure concentration was assessed to be
equal to the highest LOD, which is 0.08 mg/m3 the
median concentration is lower than the LOD for a
high proportion of samples.
Exposure Duration
HBCD standard grade
powder
1 hour/day
The basis for this assumption is on the
weighing and addition tasks at plants
producing EPS. The tasks took 10 to 15
minutes per batch. Overall, weighing and
transfer of HBCD took about an hour a
week.
The exposure duration is assumed to be 0.5 hour/day
for addition and 0.5 hour/day for weighing.
HBCD granules
Based on the study conducted by the
European Extruded Polystyrene
Insulation Board Association on the
Page 566 of 723
-------
Assessment Parameter
measured airborne concentration of
HBCD in the production of XPS resin
from HBCD granules. The main relevant
tasks were emptying boxes and cleaning
the feed deck, which took approximately
0.25 hour daily and 1 hour weekly.
Exposure Frequency
HBCD standard grade
powder
1 day/year
This is based on occupational exposure
scenarios for masterbatch compounding
from sites in Australia.
Not applicable
HBCD standard grade
powder and HBCD
granules
180 days/year
This is based on occupational exposure
scenarios for EPS resin compounding
from sites in Australia.
Not applicable
Page 567 of 723
-------
Table Apx E-5. Summary of Approaches from Other Risk Assessment Reports (RARs)
Row
Life Cycle
Stage
Inhalation
Exposures
Oral Exposures
Dermal Exposures
Environmental Releases
1
Repackaging
of import
containers
See TableApx
E-3 and
TableApx E-4
The EURAR and NICNAS RAR
assumed 100% absorption of
inhalable particulates.
Neither the EU nor the NICNAS RARs
included monitoring data for dermal
exposures. These RARs modelled dermal
exposures using the EASE model.
EURAR assessed releases from
manufacturing of HBCD and not Import /
repackaging.
NICNAS RAR assessed releases with the
OECD ESD on Plastic Additives (OECD
2009).
2
Compounding
of polystyrene
to produce
XPS
masterbatch
See Table Apx
E-3 and
Table Apx E-4
The EURAR and NICNAS RAR
assumed 100% absorption of
inhalable particulates.
The methodology described in Row 1 was
also used in this exposure scenario.
EURAR assessed releases with site-
specific data.
NICNAS RAR assessed only dust releases
with the OECD ESD on Plastic Additives
(OECD 2009).
Enviromnental Canada RAR assessed only
dust releases with the OECD ESD on
Plastic Additives (OECD 2009).
3
Manufacture
of XPS foam
from XPS
masterbatch
See Table Apx
E-3 and
Table Apx E-4
The EURAR and NICNAS RAR
assumed 100% absorption of
inhalable particulates.
The methodology described in Row 1 was
also used in this exposure scenario.
EURAR assessed releases with site-
specific data.
NICNAS RAR assessed only dust releases
with the OECD ESD on Plastic Additives
(OECD 2009).
4
Manufacture
of XPS foam
using HBCD
powder
See Table Apx
E-3 and
Table Apx E-4
The EURAR and NICNAS RAR
assumed 100% absorption of
inhalable particulates.
The methodology described in Row 1 was
also used in this exposure scenario.
EURAR assessed releases with site-
specific data.
NICNAS RAR assessed only dust releases
with the OECD ESD on Plastic Additives
(OECD 2009).
5
Manufacture
of EPS foam
from imported
EPS resin
beads
See Table Apx
E-3 and
Table Apx E-4
The EU and NICNAS RARs
assessed exposures from the
production of EPS resin and
indicated that exposures are
expected to be low during the
conversion of these EPS resin beads
The EU and NICNAS RARs assessed
exposures from the production of EPS
resin and indicated that exposures are
expected to be low during the conversion
of these EPS resin beads into EPS foam,
thus were not assessed.
EURAR assessed only dust releases with
the OECD ESD on Plastic Additives
(OECD 2009).
NICNAS RAR assessed only dust releases
with the OECD ESD on Plastic Additives
(OECD 2009).
Page 568 of 723
-------
Row
Life Cycle
Stage
Inhalation
Exposures
Oral Exposures
Dermal Exposures
Environmental Releases
into EPS foam, thus were not
assessed.
6
Manufacture
of SIPs and
Automobile
Replacement
Parts from
XPS orEPS
See TableApx
E-3 and
TableApx E-4
Because of the low inhalation
exposure potential, the EU and
NICNAS RARs did not assess oral
exposures during this exposure
scenario.
The EU and NICNAS RARs indicate that,
because HBCD is incorporated into the
foam matrix, dermal exposure is unlikely
and is not assessed.
EURAR assessed releases with data on
particulate emission rates during cutting
and sawing of EPS and XPS foam.
NICNAS RAR did not assess this release.
7
Installation of
Automobile
Replacement
Parts
See Table Apx
E-3 and
Table Apx E-4
The methodology described in Row
6 was also used in this exposure
scenario.
The methodology described in Row 6 was
also used in this exposure scenario.
The RARs reviewed did not assess this
exposure scenario.
8
Installation of
XPS/EPS
Foam
Insulation in
Residential,
Public and
Commercial
Buildings, and
Other
Structures
See Table Apx
E-3 and
Table Apx E-4
The methodology described in Row
6 was also used in this exposure
scenario.
The methodology described in Row 6 was
also used in this exposure scenario.
EURAR assessed releases with data on
particulate emission rates during cutting
and sawing of EPS and XPS foam.
NICNAS RAR did not assess this release.
9
Demolition
and Disposal
of XPS/EPS
Foam
Insulation in
Residential,
Public and
Commercial
Buildings, and
Other
Structures
The EU and
NICNAS RARs
did not assess
occupational
exposures during
this exposure
scenario.
The EU and NICNAS RARs did not
assess occupational exposures
during this exposure scenario.
The EU and NICNAS RARs did not assess
occupational exposures during this
exposure scenario.
EURAR assessed releases with data on
particulate emission rates during breaking
of EPS and XPS foam. The EURAR did
not quantify disposal releases.
NICNAS RAR assessed a steady-state
scenario, where all HBCD imported is
releases. NICNAS subtracted upstream
losses and assumed the remaining amount
was released in this exposure scenario.
10
Recycling of
EPS Foam
The EU and
NICNAS RARs
did not assess
The EU and NICNAS RARs did not
assess occupational exposures
during this exposure scenario.
The EU and NICNAS RARs did not assess
occupational exposures during this
exposure scenario.
The EU and NICNAS RARs did not assess
releases during this exposure scenario.
Page 569 of 723
-------
Row
Life Cycle
Stage
Inhalation
Exposures
Oral Exposures
Dermal Exposures
Environmental Releases
occupational
exposures during
this exposure
scenario.
11
Formulation
of Flux/
Solder Pastes
This exposure
scenario was not
included in the
identified RARs.
This exposure scenario was not
included in the identified RARs.
This exposure scenario was not included in
the identified RARs.
This exposure scenario was not included in
the identified RARs.
12
Use of Flux /
Solder Pastes
This exposure
scenario was not
included in the
identified RARs.
This exposure scenario was not
included in the identified RARs.
This exposure scenario was not included in
the identified RARs.
This exposure scenario was not included in
the identified RARs.
Page 570 of 723
-------
E.3 Equations for Calculating Acute and Chronic (Non-Cancer)
Inhalation Exposures
This report assesses HBCD exposures to workers in occupational settings, presented as 8-hr time
weighted average (TWA). The 8-hr TWA exposures are then used to calculate acute exposure, average
daily dose (ADD) for chronic, non-cancer risks.
Acute workplace exposures are assumed to be equal to the contaminant concentration in air (8-hr TWA),
per Equation E-4.
Equation E-l:
C x ED xb
AED =
BW
Where:
AED = Acute exposure dose (mg/kg-day)
C = Contaminant concentration in air (TWA) (mg/m3)
ED = exposure duration (8 hr/day)
b = breathing rate (1.25 m3/hr)
BW = body weight (80 kg)
ADD is used to estimate workplace chronic exposures for non-cancer risks. These exposures are
estimated as follows:
Equation E-2:
C x ED xb xEFxWY
ADD = —
BWxAT
Where:
ADD = average daily dose used for chronic non-cancer risk calculations (mg/kg-day)
C = contaminant concentration in air (8-hr TWA) (mg/m3)
ED = exposure duration (8 hr/day)
b = breathing rate (1.25 m3/hr)
EF = exposure frequency (days/yr)
WY = exposed working years per lifetime (50th percentile = 31; 95th percentile = 40)
BW = body weight (80 kg)
AT = averaging time, non-cancer risks (WY x 365 days/yr)
Table Apx E-6 Parameter Values for Ca
culating Inhalation Exposure Estimates
Parameter Name
Symbol
Value
Unit
Exposure Duration
ED
8
hr/day
Breathing Rate
b
1.25a
m3/to-
Exposure Frequency
EF
discussed in Section 2
days/year
Working Years
WY
31 (50th percentile)
years
Page 571 of 723
-------
Parameter Name
Sv ill hoi
Value
I nil
40 (95th percentile)
Body Weight
BW
80 (average adult worker)
72.4 (female of reproductive age)
kg
Averaging Time, non-cancer
AT
11,315 (CT)b
14,600 (HE)0
days
a(U.S. EPA 2011b") provides breathing rates for pregnant and lactating females, breathing rate used is for light activity for
workers which is higher than these specific rates provided for pregnant and lactating females.
b Calculated using the 50th percentile value for working years (WY)
0 Calculated using the 95th percentile value for working years (WY)
Exposure Duration (ED)
EPA uses an exposure duration of 8 hours per day for averaging full-shift exposures.
Breathing Rate (b)
EPA uses a breathing rate of 1.25 m3 per hour for workers (representing adults undergoing light
activity).
Exposure Frequency (EF)
EPA estimated a range of exposure frequency based on the number of operation days that EPA
determined for each exposure scenario, except for The Installation of XPS/EPS Foam Insulation and the
Demolition and Disposal of XPS/EPS Foam Insulation. For these exposure scenarios, EPA estimated a
range of exposure frequency of 1 day/year, based on release frequency, up to 250 days/year, based on
worker schedules as described below. The assessed exposure frequency did not exceed 250 days/year,
based on a worker schedule of 5 days/week over 50 weeks/year. With this range of exposure frequency,
EPA used the midpoint of this range to calculate central tendency average daily dose and the high-end of
this range to calculate high-end average daily dose. EPA's choice of these exposure frequencies are
further described in Section 2.3.
Exposure frequency (EF) is expressed as the number of days per year a worker is exposed to the
chemical being assessed. In some cases, it may be reasonable to assume a worker is exposed to the
chemical on each working day. In other cases, it may be more appropriate to estimate a worker's
exposure to the chemical occurs during a subset of the worker's annual working days. The relationship
between exposure frequency and annual working days can be described mathematically as follows:
EF=fxAWD
Where:
EF = exposure frequency, the number of days per year a worker is exposed to the chemical
(day/yr)
f = fractional number of annual working days during which a worker is exposed to the
chemical (unitless)
AWD = annual working days, the number of days per year a worker works (day/yr)
Page 572 of 723
-------
BLS (2014) provides data on the total number of hours worked and total number of employees by each
industry NAICS code. These data are available from the 3- to 6-digitNAICS level (where 3-digit
NAICS are less granular and 6-digitNAICS are the most granular). Dividing the total, annual hours
worked by the number of employees yields the average number of hours worked per employee per year
for each NAICS.
EPA has identified approximately 140 NAICS codes applicable to the multiple exposure scenarios for
the 10 chemicals undergoing Risk Evaluation. For each NAICS code of interest, EPA looked up the
average hours worked per employee per year at the most granular NAICS level available {i.e., 4-digit, 5-
digit, or 6-digit). EPA converted the working hours per employee to working days per year per
employee assuming employees work an average of eight hours per day. The average number of days per
year worked, or AWD, ranges from 169 to 282 days per year, with a 50th percentile value of 250 days
per year. EPA repeated this analysis for all NAICS codes at the 4-digit level. The average AWD for all
4-digit NAICS codes ranges from 111 to 282 days per year, with a 50th percentile value of 228 days per
year. 250 days per year is approximately the 75th percentile.
In the absence of industry- and HBCD-specific data, EPA assumes the parameter/is equal to one for all
exposure scenarios.
Working Years (WY)
EPA has developed a triangular distribution for working years. EPA has defined the parameters of the
triangular distribution as follows:
• Minimum value: BLS CPS tenure data with current employer as a low-end estimate of the
number of lifetime working years: 10.4 years;
• Mode value: The 50th percentile tenure data with all employers from Survey of Income and
Program Participation (SIPP) as a mode value for the number of lifetime working years: 36
years; and
• Maximum value: The maximum average tenure data with all employers from SIPP as a high-end
estimate on the number of lifetime working years: 44 years.
This triangular distribution has a 50th percentile value of 31 years and a 95th percentile value of 40 years.
EPA uses these values for central tendency and high-end ADD calculations, respectively.
The BLS (2014) provides information on employee tenure with current employer obtained from the
Current Population Survey (CPS). CPS is a monthly sample survey of about 60,000 households that
provides information on the labor force status of the civilian non-institutional population age 16 and
over; CPS data are released every two years. The data are available by demographics and by generic
industry sectors but are not available by NAICS codes.
The U.S. Census' (Census Bureau 2016) Survey of Income and Program Participation (SIPP) provides
information on lifetime tenure with all employers. SIPP is a household survey that collects data on
income, labor force participation, social program participation and eligibility, and general demographic
characteristics through a continuous series of national panel surveys of between 14,000 and 52,000
households (Census Bureau 2.016). EPA analyzed the 2008 SIPP Panel Wave 1, a panel that began in
2008 and covers the interview months of September 2008 through December 2008 (Census Bureau
Page 573 of 723
-------
2016). For this panel, lifetime tenure data are available by Census Industry Codes, which can be cross-
walked with NAICS codes.
SIPP data include fields for the industry in which each surveyed, employed individual works
(TJBIND1), worker age (TAGE), and years of work experience with all employers over the surveyed
individual's lifetime.22 Census household surveys use different industry codes than the NAICS codes
used in its firm surveys, so these were converted to NAICS using a published crosswalk (Census Bureau
2012). EPA calculated the average tenure for the following age groups: 1) workers age 50 and older; 2)
workers age 60 and older; and 3) workers of all ages employed at time of survey. EPA used tenure data
for age group "50 and older" to determine the high-end lifetime working years, because the sample size
in this age group is often substantially higher than the sample size for age group "60 and older." For
some industries, the number of workers surveyed, or the sample size, was too small to provide a reliable
representation of the worker tenure in that industry. Therefore, EPA excluded data where the sample
size is less than five from our analysis.
TableApx E-7 summarizes the average tenure for workers age 50 and older from SIPP data. Although
the tenure may differ for any given industry sector, there is no significant variability between the 50th
and 95th percentile values of average tenure across manufacturing and non-manufacturing sectors.
Table Apx E-7. Overview of Average Worker Tenure from U.S. Census SIPP (Age Group 50+)
Industry Sectors
Working Years
Average
50th Percentile
95th Percentile
Maximum
All industry sectors relevant to the 10
chemicals undergoing Risk Evaluation
35.9
36
39
44
Manufacturing sectors (NAICS 31-33)
35.7
36
39
40
Non-manufacturing sectors (NAICS 42-81)
36.1
36
39
44
Source: (Census Bureau 2016)
Note: Industries where sample size is less than five are excluded from this analysis.
BLS CPS data provides the median years of tenure that wage and salary workers had been with their
current employer. Table Apx E-8 presents CPS data for all demographics (men and women) by age
group from 2008 to 2012. To estimate the low-end value on number of working years, EPA uses the
most recent (U.S. BLS 2014) CPS data for workers age 55 to 64 years, which indicates a median tenure
of 10.4 years with their current employer. The use of this low-end value represents a scenario where
workers are only exposed to the chemical of interest for a portion of their lifetime working years, as they
may change jobs or move from one industry to another throughout their career.
Table Apx E-8. Median Years of Tenure with Current Employer by Age Group
Age
January 2008
January 2010
January 2012
January 2014
16 years and over
4.1
4.4
4.6
4.6
16 to 17 years
0.7
0.7
0.7
0.7
22 To calculate the number of years of work experience we took the difference between the year first
worked (TMAKMNYR) and the current data year (i.e., 2008). We then subtracted any intervening
months when not working (ETIMEOFF).
Page 574 of 723
-------
Age
January 2008
January 2010
January 2012
January 2014
18 to 19 years
0.8
1.0
0.8
0.8
20 to 24 years
1.3
1.5
1.3
1.3
25 years and over
5.1
5.2
5.4
5.5
25 to 34 years
2.7
3.1
3.2
3.0
35 to 44 years
4.9
5.1
5.3
5.2
45 to 54 years
7.6
7.8
7.8
7.9
55 to 64 years
9.9
10.0
10.3
10.4
65 years and over
10.2
9.9
10.3
10.3
Source: (U.S. BLS 2014)
Body Weight (BW)
EPA assumes a body weight of 80 kg for all worker demographics.
E.4 Sample Calculations for Calculating Acute and Chronic (Non-
Cancer) Inhalation Exposure
Sample calculations for high-end and central tendency chronic exposure doses for one setting,
Repackaging of Import Containers, are demonstrated below. The explanation of the equations and
parameters used is provided in Appendix E.3.
Example High-End ADD
Calculate ADDhe:
CHE x b x ED x EF x WY
ADDhf = —
HE BW XAT
1.89 x 1.25 2! x 8^1x60^x40years
ADDhe = ^^^ ymr , = 3.88 X 10-2
80
kg x (40 years x 365 ^ays\ kg day
a V yearJ
Example Central Tendency ADD
Calculate ADDct:
Cct x b x ED x EF x WY
addct =
BW X ATADD
0.89 x 1.25 x 8-7— x 60^^ x 31 years mn
ADDct = ^ K ^ year , =1.83x10-2 m9
80 kg x (31 years x 365——^—^ ^ ^a^
a V yearJ
Page 575 of 723
-------
E.5 Approaches for Estimating Number of Workers
This appendix summarizes the methods and provides an example of the method that EPA used to
estimate the number of workers who are potentially exposed to HBCD in each of its exposure scenarios.
The method consists of the following steps:
• Identify the North American Industry Classification System (NAICS) codes for the industry
sectors associated with each exposure scenario.
• Estimate total employment by industry/occupation combination using the Bureau of Labor
Statistics' occupational employment statistics data ( _201 o).
• Refine the occupational employment statistics estimates where they are not sufficiently granular
by using the U.S. Census' (2015) Statistics of U.S. Businesses (SUSB) data on total employment
by 6-digit NAICS (Census Bureau 2015).
• Estimate the number of potentially exposed employees per site.
• Estimate the number of potentially exposed employees within the exposure scenario using the
estimated number of sites.
Step 1: Identifying Affected NAICS Codes
As a first step, EPA identified NAICS industry codes associated with each exposure scenario. EPA
generally identified NAICS industry codes for an exposure scenario by:
• Querying the Census Bureau's NAICS Search tool using keywords associated with each
exposure scenario to identify NAICS codes with descriptions that match the exposure scenario.
• Referencing EPA/OPPT Generic Scenarios (GS's) and Organisation for Economic Co-operation
and Development (OECD) Emission Scenario Documents (ESDs) for an exposure scenario to
identify NAICS codes cited by the GS or ESD.
• Reviewing Chemical Data Reporting (CDR) data for the chemical, identifying the industrial
sector codes reported for downstream industrial uses, and matching those industrial sector codes
to NAICS codes using Table_Apx D-2 provided in the CDR reporting instructions.
Each exposure scenario section in the main body of this report identifies the NAICS codes EPA
identified for the respective exposure scenario.
Step 2: Estimating Total Employment by Industry and Occupation
BLS's (2016) occupational employment statistics data provide employment data for workers in specific
industries and occupations. The industries are classified by NAICS codes (identified previously), and
occupations are classified by Standard Occupational Classification (SOC) codes.
Among the relevant NAICS codes (identified previously), EPA reviewed the occupation description and
identified those occupations (SOC codes) where workers are potentially exposed. TableApx E-9. shows
the SOC codes EPA classified as occupations potentially exposed. These occupations are classified into
workers (W) and occupational non-users (O). All other SOC codes are assumed to represent occupations
where exposure is unlikely. An example is provided below for an exposure scenario of dry cleaning.
Table Apx E-9. SOCs with Worker and ONU Designations for All Exposure scenarios
After identifying relevant NAICS and SOC codes, EPA/OPPT used BLS data to determine total
employment by industry and by occupation based on the NAICS and SOC combinations. For example,
Page 576 of 723
-------
there are 1,790 employees associated with 4-digit NAICS 3259 (Other Chemical Product and
Preparation Manufacturing) and SOC 49-9070 (Maintenance and Repair Workers, General).
Using a combination of NAICS and SOC codes to estimate total employment provides more accurate
estimates for the number of workers than using NAICS codes alone. Using only NAICS codes to
estimate number of workers typically result in an overestimate, because not all workers employed in that
industry sector will be exposed. However, in some cases, BLS only provide employment data at the 4-
digit or 5-digit NAICS level; therefore, further refinement of this approach may be needed (see next
step).
Step 3: Refining Employment Estimates to Account for lack of NAICS Granularity
The third step in EPA's methodology was to further refine the employment estimates by using total
employment data in the U.S. Census Bureau's SUSB (Census Bureau 2015). In some cases, BLS
occupational employment statistics' occupation-specific data are only available at the 4-digit or 5-digit
NAICS level, whereas the SUSB data are available at the 6-digit level (but are not occupation-specific).
Identifying specific 6-digit NAICS will ensure that only industries with potential exposure are included.
As an example, occupational employment statistics data are available for the 4-digit NAICS 3259 Other
Chemical Product and Preparation Manufacturing, which includes the following 6-digit NAICS:
1. NAICS 325910 Printing Ink Manufacturing;
2. NAICS 325920 Explosives Manufacturing;
3. NAICS 325991 Custom Compounding of Purchased Resins;
4. NAICS 325992 Photographic Film, Paper, Plate, and Chemical Manufacturing; and
5. NAICS 325998 All Other Miscellaneous Chemical Product and Preparation Manufacturing.
In this example, only NAICS 325991 is of interest. The Census data allow EPA to calculate employment
in the specific 6-digit NAICS of interest as a percentage of employment in the BLS 4-digit NAICS.
The 6-digit NAICS 325991 comprises 23.5 percent of total employment under the 4-digit NAICS 3259.
This percentage can be multiplied by the occupation-specific employment estimates given in the BLS
OES data to further refine our estimates of the number of employees with potential exposure.
Table Apx E-10 illustrates this granularity adjustment for NAICS 325991.
Table Apx E-10. Estimated Number of Potentially Exposed Workers and ONUs under NAICS
325991
NAICS
SOC
CODE
SOC Description
Occupation
Designation
Employment
by SOC at 4-
digit NAICS
level
% of Total
Employment
Estimated
Employment
by SOC at 6-
digit NAICS
level
325900
17-2000
Engineers
O
3,010
23.5%
709
325900
17-3000
Drafters, Engineering
Technicians, and Mapping
Technicians
O
860
23.5%
202
325900
19-2031
Chemists
0
1,400
23.5%
330
325900
19-4000
Life, Physical, and Social
Science Technicians
O
1,810
23.5%
426
325900
47-2000
Construction Trades Workers
w
200
23.5%
47
Page 577 of 723
-------
NAICS
SOC
CODE
SOC Description
Occupation
Designation
Employment
by SOC at 4-
digit NAICS
level
% of Total
Employment
Estimated
Employment
by SOC at 6-
digit NAICS
level
325900
49-1000
Supervisors of Installation
Maintenance, and Repair
Workers
O
340
23.5%
80
325900
49-2000
Electrical and Electronic
Equipment Mechanics,
Installers, and Repairers
w
260
23.5%
61
325900
49-9010
Control and Valve Installers
and Repairers
w
60
23.5%
14
325900
49-9040
Industrial Machinery
Installation, Repair, and
Maintenance Workers
w
1,720
23.5%
405
325900
49-9060
Precision Instrument and
Equipment Repairers
w
30
23.5%
7
325900
49-9070
Maintenance and Repair
Workers, General
w
1,790
23.5%
421
325900
49-9090
Miscellaneous Installation,
Maintenance, and Repair
Workers
w
80
23.5%
19
325900
51-1000
Supervisors of Production
Workers
0
3,480
23.5%
819
325900
51-2000
Assemblers and Fabricators
w
5,270
23.5%
1,241
325900
51-4020
Forming Machine Setters,
Operators, and Tenders,
Metal and Plastic
w
1,170
23.5%
275
325900
51-6090
Miscellaneous Textile,
Apparel, and Furnishings
Workers
0
1,320
23.5%
311
325900
51-8020
Stationary Engineers and
Boiler Operators
w
40
23.5%
9
325900
51-8090
Miscellaneous Plant and
System Operators
w
1,530
23.5%
360
325900
51-9000
Other Production
Occupations
w
24,880
23.5%
5,858
Total Potentially Exposed Employees
49,250
11,597
Total Workers
8,719
Total Occupational Non-Users
2,877
Note: numbers may not sum exactly due to rounding.
W = worker
O = occupational non-user
Source: (Census Bureau 2015): (U.S. BLS 2016)
Step 4: Estimating the Percentage of Workers Using HBCD Instead of Other Chemicals
In the final step, EPA accounted for the market share by applying a factor to the number of workers
determined in Step 3. This accounts for the fact that the substance may be only one of multiple
chemicals used for the applications of interest. EPA did not identify market penetration data for any
exposure scenarios. In the absence of market penetration data for a given exposure scenario, EPA/OPPT
assumed HBCD may be used at up to all sites and by up to all workers calculated in this method as a
bounding estimate. This assumes a market penetration of 100%. Market penetration is discussed for each
exposure scenario in the main body of this report.
Page 578 of 723
-------
Step 5: Estimating the Number of Workers per Site
EPA/OPPT calculated the number of workers and occupational non-users in each industry/occupation
combination using the formula below (granularity adjustment is only applicable where SOC data are not
available at the 6-digitNAICS level):
Number of Workers or ONUs in NAICS/SOC (Step 2) x Granularity Adjustment Percentage (Step 3) =
Number of Workers or ONUs in the Industry/Occupation Combination
EPA/OPPT then estimated the total number of establishments by obtaining the number of establishments
reported in the U.S. Census Bureau's SUSB data at the 6-digit NAICS level (Census Bureau 2015).
EPA then summed the number of workers and occupational non-users over all occupations within a
NAICS code and divided these sums by the number of establishments in the NAICS code to calculate
the average number of workers and occupational non-users per site.
Step 6: Estimating the Number of Workers and Sites for an Exposure Scenario
EPA estimated the number of workers and occupational non-users potentially exposed and the number
of sites that use HBCD in a given exposure scenario through the following steps:
• Obtaining the total number of establishments by:
o Obtaining the number of establishments from SUSB at the 6-digit NAICS level (Step 5)
for each NAICS code in the exposure scenario and summing these values (Census Bureau
20.1.5) or
o Obtaining the number of establishments from the Toxics Release Inventory (TRI),
Discharge Monitoring Report (DMR) data, National Emissions Inventory (NEI), or
literature for the exposure scenario.
• Estimating the number of establishments that use HBCD by taking the total number of
establishments from Step 6. A and multiplying it by the market penetration factor from Step
4.
• Estimating the number of workers and occupational non-users potentially exposed to HBCD
by taking the number of establishments calculated in Step 6.B and multiplying it by the
average number of workers and occupational non-users per site from Step 5.
E.6 Evaluation of Occupational Exposure Data Sources
EPA has reviewed acceptable sources for HBCD inhalation exposure data according to the data quality
evaluation criteria found in The Application of Systematic Review in TSCA Risk Evaluations (
2018b). Table Apx E-l 1 summarizes the results of this evaluation. The data quality evaluation of
inhalation monitoring data sources indicated the quality of the sources ranges from unacceptable to high;
however, unacceptable data were excluded from the assessment of occupational inhalation exposure to
HBCD.
Page 579 of 723
-------
Table Apx E-ll. Summary of Inhalation Monitoring Data and Systematic Review Results
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
Searl and
Robertson
(2005)- la
Manufacturing
of HBCD
Standard
grade
HBCD
Personal
Breathing
Zone
Packaging,
compaction,
process
operations,
and working in
the warehouse
Mean: 1.23
Median: 0.89
90 th percentile:
1.89
Max: 3 mg/m3
10
8-hr TWA
(ECHA 2008b)
(ECHA 2009b)
3970747; 3809166
High
Included -
although
manufacturing of
ITBCD is not an
exposure scenario,
these data are
applicable to the
importation of
HBCD
Searl and
Robertson
(2005)- lb
Manufacturing
of HBCD
Fine grade
HBCD
Personal
Breathing
Zone
Packaging,
compaction,
process
operations,
and working in
the warehouse
Mean: 23
90th percentile:
35
4
8-hr TWA
(ECHA 2008b)
3970747
High
Excluded -
manufacturing is
out of scope and,
while this data
may be applicable
to other exposure
scenarios, fine
grade ITBCD is not
preferred
Searl and
Robertson
(2005)- lc
Manufacturing
of HBCD
HBCD of
unknown
grade
NR
Packaging and
compaction of
powders
Respirable,
Mean: 0.18
Inhalable,
Mean: 1.23
NR
NR
(ECHA 2009c)
3970759
High
Excluded -
manufacturing is
out of scope and,
while this data
may be applicable
to other exposure
scenarios, the
grade of ITBCD
and sample time
are unknown
Waindzioch
(2000) - la
Manufacturing
of HBCD
HBCD of
unknown
grade
Area
Reactor
0.00028 -
0.0285
3
Short-term
(ECHA 2008b)
3970747
Unacceptable
Excluded -
manufacturing of
ITBCD is not an
exposure scenario
for this Risk
Evaluation and this
data is not
applicable to other
exposure scenarios
Waindzioch
(2000)- lb
Manufacturing
of HBCD
HBCD of
unknown
grade
Area
Filling Station
0.0094 - 0.097
2
Short-term
(ECHA 2008b)
3970747
High
Excluded -
manufacturing is
out of scope and,
while this data
Page 580 of 723
-------
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
may be applicable
to other exposure
scenarios, area
samples are not
preferred
Biesemeier
(1996)
Manufacturing
of HBCD
HBCD of
unknown
grade
NR
Bagging
HBCD product
4.0-4.5
NR
NR
(ECHA 2008b)
3970747
High
Excluded -
manufacturing is
out of scope and,
while this data
may be applicable
to other exposure
scenarios, sample
type and time are
unknown
Velsicol (1978)
Manufacturing
of HBCD
HBCD of
unknown
grade
Personal
Breathing
Zone
Transfer of the
HBCD in the
hammer-mill
to 28 drums
1.9
1
300 minutes
(Velsicol Chem
Com 1978)
1928232
High
Excluded -
manufacturing is
out of scope and,
while this data
may be applicable
to other exposure
scenarios, the
grade of HBCD
and sample time
are unknown
Yietal. (2016)
Manufacturing
of HBCD
HBCD of
unknown
grade
Personal
Breathing
Zone
NR
0.0102-
0.0283
14
NR
(Yietal. 2016)
3350493
High
Excluded -
manufacturing is
out of scope and,
while this data
may be applicable
to other exposure
scenarios, the
grade of HBCD
and sample time
are unknown
Searl and
Robertson
(2005)-2a
Manufacturing
of EPS Resin
beads
Standard
grade
HBCD
Personal
Manual
addition of
HBCD powder
to reactor each
time a batch of
EPS resin was
produced
Range: 2.89-
21.5
Mean: 7.2
Median: 5.52
90th percentile:
10.5
12
Short-term (13
to 56 mins)
(ECHA 2008b)
(NICNAS
2012b)
3978355
High
Included - These
data are the basis
of the estimates
developed by the
EURAR for
HBCD processing
in the plastics
Page 581 of 723
-------
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
industry, which
were used by EPA
in this Risk
Evaluation
Searl and
Robertson
(2005)-2b
Manufacturing
of EPS Resin
beads
Standard
grade
HBCD
Personal
Manual
addition of
HBCD powder
to reactor each
time a batch of
EPS resin was
produced
Range: 0.12-
3.36
Mean: 1
Median: 0.42
90th percentile:
1.3
12
8-hr TWA -
Note this is the
8-hr TWA of the
data in the above
row
(ECHA 2008b)
(NICNAS
2012b)
3978355
High
Included - These
data are the basis
of the estimates
developed by the
EURAR for
HBCD processing
in the plastics
industry, which
were used by EPA
in this Risk
Evaluation
Searl and
Robertson
(2005)-2c
Manufacturing
of EPS Resin
beads
Standard
grade
HBCD
Personal
Manual
addition of
HBCD powder
to reactor each
time a batch of
EPS resin was
produced
Range: 0.07-
14.7
Mean: 1.2
Median: 0.27
90th percentile:
1.10
18
275 to 504 mins
(NICNAS
2012b)
(ECHA 2008b)
(NICNAS
2012b)
3978355
High
Included - These
data are the basis
of the estimates
developed by the
EURAR for
HBCD processing
in the plastics
industry, which
were used by EPA
in this Risk
Evaluation
Searl and
Robertson
(2005)-2d
Manufacturing
of EPS Resin
beads
Standard
grade
HBCD
Personal
Weighing
powder prior
to addition to
reactor. HBCD
bags were
weighed and
opened
concurrently,
or weighed in
advance, in
which case
HBCD was
transferred
from 25-kg
sacks using
Range: 4.35-
12.1
Mean: 7.2
Median: 6.19
90th percentile:
10.5
4
124 to 350 mins
(NICNAS
2012b)
(ECHA 2008b)
(NICNAS
2012b)
3978355
High
Included - These
data are the basis
of the estimates
developed by the
EURAR for
HBCD processing
in the plastics
industry, which
were used by EPA
in this Risk
Evaluation
Page 582 of 723
-------
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
plastic scoop
(full-shift
measurement).
Searl and
Robertson
(2005)- 3a
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
Area
Weighing and
mixing
Max 7.5 (for 2
hours)
Mean: 1.89
Median: 0.83
90th percentile:
5.4
10
Short-term
(ECHA 2008b)
(ECHA 2009b)
3970747; 3809166
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Searl and
Robertson
(2005)- 3b
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
Area
Weighing and
mixing
Mean: 0.88
90th percentile:
1.36
10
8-hr TWA
(ECHA 2008b)
3970747
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Searl and
Robertson
(2005)- 3c
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
NR
Extruder
Mean: 0.12
Median: 0.10
90th percentile:
0.16
4
5 hours
(ECHA 2008b)
(ECHA 2009b)
3970747; 3809166
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Searl and
Robertson
(2005)- 3d
Compounding
of Polystyrene
resin to
produce XPS
Masterbatch
containing
HBCD
HBCD of
unknown
grade
NR
Automated
handling of
HBCD
Negligible
3
NR
(ECHA 2008b)
3970747
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Page 583 of 723
-------
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
Abbott (2001)-
la
Manufacture of
XPS from
HBCD powder
or granules
Standard
grade
HBCD
Area
At the feed
deck near
typical
operator
positions
Range 0.24 -
1.6
Mean: 0.66
90 th percentile:
1.45
(excluding 10
ND samples)
16(10
ND)
8-hr TWA
(ECHA 2008b)
3970747
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Abbott (2001)-
lb
Manufacture of
XPS from
HBCD powder
or granules
HBCD
granules
Mostly
area and
some
personal
breathing
zone
Feed deck near
typical
operator
positions
Range 0.005-
0.9
Mean: 0.24
90th percentile:
0.47
(excluding 16
ND samples)
43(16
ND)
60- 1435
minutes
(ECHA 2008b)
3970747
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Thomsen
(2007) - la
Manufacture of
XPS from
HBCD powder
or granules
HBCD
powder and
granules
Personal
breathing
zone
Activities in
the mixer area,
including
operating a
closed
automated
process
excluding
potential
contact with
neat HBCD
Range: 0.0002-
0.0009
Mean: 0.0005
Median:
0.0005
6
8-hr TWA
(ECHA 2008b)
(NICNAS
2012b)
3970747; 3978355
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Thomsen
(2007)- lb
Manufacture of
XPS from
HBCD powder
or granules
HBCD
powder and
granules
Personal
breathing
zone
Weighing and
addition of
HBCD to the
reactor and
subsequent
washing,
centrifugation,
sifting, and
transfer of
product to a
silo container
Range: 0.001-
0.15
Mean: 0.015
Median:
0.0027
24
8-hr TWA
(ECHA 2008b)
(NICNAS
2012b)
3970747; 3978355
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Page 584 of 723
-------
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
Searl and
Robertson
(2005)-4
Manufacture of
XPS from
HBCD powder
or granules
HBCD
granules
Area
Logistics,
extruding, and
laboratory
Mean: 0.00003
90th percentile:
0.00004
12
8-hr TWA
(ECHA 2008b)
3970747
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Ransbotyn
(1999)
Manufacturing
of EPS Resin
beads
Respirable
Dust
Inhalable
Dust
Personal
Addition of
HBCDD to
reactor or the
supervising of
the addition.
Respirable
dust: <0.5
Total Inhalable
dust: 2.0
Not specific to
HBCD
5
Max 8-hr TWA
(ECHA 2008b)
3970747
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
NICNAS
(2012)- la
HBCD
importation /
repackaging
sites and all
industrial
polymer
processing
sites
Standard
grade
HBCD
Modelled
with
EASE
Addition of
HBCD into
process
operation
Typical: 2 to 5
Worst-case: 5
to 50
Not
applicable
- this is a
modelled
exposure
8-hr TWA
(NICNAS
2012b)
3978355
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
NICNAS
(2012)- lb
HBCD
importation /
repackaging
sites and all
industrial
polymer
processing
sites
HBCD
granules
Modelled
with
EASE
Repackaging
with the use of
LEV (typical)
and without
LEV (worst-
case)
Typical: 0.2 to
0.5
Worst-case: 0.5
to 5
Not
applicable
- this is a
modelled
exposure
8-hr TWA
(NICNAS
2012b)
3978355
High
Excluded - EPA
used the estimates
for HBCD
processing in the
plastics industry
calculated in the
EURAR from the
available data (See
Appendix E.2)
Searl and
Robertson
(2005)- 5a
Secondary
processing of
XPS foam
(cutting,
XPS foam
NR
Secondary
processing of
XPS foam -
including
cutting,
Mean: 0.08
90th percentile:
0.22
9
8-hr TWA
Original source:
Searl and
Robertson
(2005)
3809166
High
Included - these
data were used to
estimate worker
inhalation
exposure in the
Page 585 of 723
-------
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
sawing,
machining)
sawing, and
machining to
manufacture
shaped
products
Reported in:
fECHA 2008b);
fECHA 2009b)
following
exposure
scenarios:
Manufacturing of
XPS Foam using
XPS Masterbatch;
Manufacturing of
EPS Foam from
Imported EPS
Resin Beads;
Manufacturing of
SIPs and
Automobile
Replacement Parts
from XPS/EPS
Foam; Installation
of XPS/EPS Foam
Insulation in
Residential, Public
and Commercial
Buildings, and
Other Structures;
Demolition and
Disposal of
XPS/EPS Foam
Insulation
Products in
Residential, Public
and Commercial
Buildings, and
Other Structures;
Recycling of EPS
Foam and Reuse
of XPS Foam
Searl and
Robertson
(2005)- 5b
Reclamation of
XPS foam -
including
shredding and
reprocessing of
process waste
XPS foam
NR
Reclamation
of XPS foam -
including
shredding and
reprocessing
of process
waste
Mean: 0.02
90th percentile:
0.02
5
8-hr TWA
Original source:
Searl and
Robertson
(2005)
Reported in:
3809166
High
Excluded - EPA
used the data in
Searl and
Robertson (2005) -
5 a because it
presents a larger
Page 586 of 723
-------
Literature
Study a
Exposure
scenario
Data from source b
Source c
Data Identifier
from Data
Extraction and
Evaluation
Overall
Confidence
Rating from
Data Extraction
and Evaluation
Rationale for
Inclusion /
Exclusion
Form of
HBCD
Handled
Type of
Sample
Worker
Activity or
Sampling
Location
Exposure
Concentration
(mg/m3)
Number
of
Samples
Sample Time /
Type of
Measurement
(ECHA 2008b);
(ECHA 2009b)
range of potential
exposure
Searl and
Robertson
(2005)- 5c
Manufacture of
XPS from XPS
masterbatch
XPS foam
NR
Other process
control
operators
Mean: 0.03
90th percentile:
0.03
4
8-hr TWA
Original source:
Searl and
Robertson
(2005)
Reported in:
(ECHA 2008b);
(ECHA 2009b)
3809166
High
Excluded - worker
activities unknown
Searl and
Robertson
(2005) - 5d
Manufacture of
XPS from XPS
masterbatch
XPS foam
NR
Process
operators
handling XPS
masterbatch
Mean: 0.03
90th percentile:
0.03
24
8-hr TWA
Original source:
Searl and
Robertson
(2005)
Reported in:
(ECHA 2008b);
(ECHA 2009b)
3809166
High
Excluded - EPA
used the data in
Searl and
Robertson (2005) -
5 a because it
presents a larger
range of potential
exposure
Zhang et al.
(2012)- la
Thermal
cutting of XPS
foam
XPS foam
NR
Thermal
cutting of XPS
boards in a
closed
glovebox
Mean: 0.089
NR
NR
(Zhang et al.
2012)
1927576
High
Excluded - sample
time is unknown
Zhang et al.
(2012)- lb
Thermal
cutting of EPS
foam
EPS foam
NR
Thermal
cutting of EPS
boards in a
closed
glovebox
Mean: 0.057
NR
NR
(Zhang et al.
2012)
1927576
High
Excluded - sample
time is unknown
NR = Not Reported; N/A = Not Applicable
a - Where multiple datasets were available from one literature source, EPA distinguished data as la, lb, 2a, 2b, etc.
b - Statistics were calculated by the cited source and are presented here as they were presented in the source.
c - Where information is presented in multiple sources all sources are listed. Information was not combined from these sources but was presented in all sources
independently.
Page 587 of 723
-------
E.7 Data Integration Strategy for Occupational Exposure and
Release Data/ Information
General Approach
Data integration is the stage following the data extraction and evaluation step discussed in the
Application of Systematic Review in TSCA Risk Evaluations (U.S. EPA. 2018b). Data
integration is where the analysis, synthesis and integration of data/ information takes place. For
integration of occupational exposure and environmental release data/information, EPA will
normally use the highest rated quality data among the higher level of the hierarchy of preferences
as described below. TableApx E-12 and TableApx E-13 below present the hierarchy of
preferences among the primary types of data/ information to be analyzed, synthesized and
integrated for the occupational exposure and release assessments in the TSCA Risk Evaluations.
EPA will provide rationale when deviations from the hierarchy occur.
Selection of Data and Approaches
EPA will select data for use from the data extraction and evaluation phase of systematic review.
EPA will only use data/information rated as High, Medium, or Low in the environmental release
and occupational exposure assessments; data/ information rated as unacceptable will not be used.
If need be, data of lower rated quality or approaches in lower levels of the hierarchy may be used
to supplement the analysis. For example, data/ information of high quality could be determined
to be sufficient such that lower quality data may not be included or integrated with the higher
quality data. Also, data/ information of high quality could be determined to be sufficient such
that approaches assigned lower preference levels in the hierarchy may not be pursued even if
they are available and possible. In many cases, EPA does not have robust and/or representative
monitoring data and will augment such data with modeled estimates of exposure.
Assessment Data and Results
EPA will provide occupational exposure and environmental release data and results
representative of central tendency conditions and high-end conditions. A central tendency is
assumed to be representative of occupational exposures and environmental releases in the center
of the distribution for a given condition of use. For Risk Evaluation, EPA may use the 50th
percentile (median), mean (arithmetic or geometric), mode, or midpoint values of a distribution
as representative of the central tendency scenario. EPA's preference is to provide the 50th
percentile of the distribution. However, if the full distribution is not known, EPA may assume
that the mean, mode, or midpoint of the distribution represents the central tendency depending on
the statistics available for the distribution.
A high-end is assumed to be representative of occupational exposures and environmental
releases that occur at probabilities above the 90th percentile but below the exposure of the
individual with the highest exposure (U. S. EPA. 1992) or the highest release. For Risk
Evaluation , EPA plans to provide high-end results at the 95th percentile. If the 95th percentile is
not available, EPA may use a different percentile greater than or equal to the 90th percentile but
less than or equal to the 99.9th percentile, depending on the statistics available for the
Page 588 of 723
-------
distribution. If the full distribution is not known and the preferred statistics are not available,
EPA may estimate a maximum or bounding estimate in lieu of the high-end.
EPA has defined occupational exposure and environmental release scenarios (OEERS) as the
most granular level that EPA will generate results within each condition of use. For some
conditions of use, EPA may define only a single OEERS (e.g., a manufacturing condition of use
for multiple manufacturing sites may be defined by a single manufacturing OEERS). Other
conditions of use have multiple OEERS (e.g., the use of chemical X in vapor degreasing has
OEERS for open-top batch vapor degreasing, conveyorized degreasing, web degreasing, and
closed-system degreasing). EPA will attempt to provide a single set of results (central tendency
and high-end) for each release or exposure assessed for an OEERS.
Integration of Data Sets
To provide the occupational and environmental release results at the central tendency and high-
end descriptors, EPA may integrate data sets representative of different sites, job descriptions, or
process conditions to develop a distribution representative of the entire population of workers
and sites involved in the given OEERS in the United States. Ideally, the distribution would
account for inter-site variability (variability in operations among different sites) and intra-site
variability (variability in operations within a single site).
To integrate data sets together, EPA will review the available metadata for each data set to
ensure the data sets are representative of the same OEERS. EPA will document any uncertainties
in the metadata or if EPA used a data set of a similar scenario as surrogate for the OEERS being
assessed.
Integration of Data for Modeling and Calculations
For occupational exposures, EPA may use measured or estimated air concentrations to calculate
exposure concentration metrics required for risk assessment, such as average daily concentration
and lifetime average daily concentration. These calculations require additional parameter inputs,
such as years of exposure, exposure duration and frequency, and lifetime years. EPA may
estimate exposure concentrations from monitoring data, modeling, or occupational exposure
limits, as identified in Table Apx E-12 and use each of these in its evidence integration to assess
the strength of the evidence.
For the final exposure result metrics, each of the input parameters (e.g., air concentrations,
working years, exposure frequency, lifetime years) may be a point estimate (i.e., a single
descriptor or statistic, such as 50th percentile or 95th percentile) or a full distribution. EPA will
consider three general approaches for estimating the final exposure result metrics:
• Deterministic calculations: EPA will use combinations of point estimates of each
parameter to estimate a central tendency and high-end for each final exposure metric
result. EPA will document the method and rationale for selecting parametric
combinations to be representative of central tendency and high-end.
Page 589 of 723
-------
• Probabilistic (stochastic) calculations: EPA will pursue Monte Carlo simulations using
the full distribution of each parameter to calculate a full distribution of the final exposure
metric results and selecting the 50th and 95th percentiles of this resulting distribution as
the central tendency and high-end, respectively.
• Combination of deterministic and probabilistic calculations: EPA may have full
distributions for some parameters but point estimates of the remaining parameters. For
example, EPA may pursue Monte Carlo modeling to estimate exposure concentrations,
but only have point estimates of working years of exposure, exposure duration and
frequency, and lifetime years. In this case, EPA will document the approach and rationale
for combining point estimates with distribution results for estimating central tendency
and high-end results.
o Probabilistic approaches can also supplement and complement monitoring
estimates by providing sensitivity analysis of parameters for certain conditions
and thus provide greater certainty about the strength of the evidence.
Confidence Statements
For each use, EPA considered the assessment approach, the quality of the data and models, and
uncertainties in assessment results to determine a level of confidence for the 8-hr TWA data and
modeled estimates.
For the inhalation air concentration monitoring data, strength of confidence is improved by the
following factors:
• higher approaches in the inhalation approach hierarchy
• larger numbers of data points
• larger number of sites monitored
• larger broadness of worker population groups included in monitoring
• higher systematic review data quality ratings.
Strength of confidence in monitoring data is reduced by:
• uncertainty of the representativeness of these data toward the true distribution of
inhalation concentrations for the industries and sites covered by the use.
For modeled air concentrations, strength of confidence is improved by the following factors:
• higher approaches in the inhalation approach hierarchy
• model validation
• full distributions of input parameters.
Strength of confidence in modeled air concentration estimates is reduced by:
• uncertainty of the representativeness of the model or parameter inputs toward the true
distribution of inhalation concentrations for the industries and sites covered by the use.
Page 590 of 723
-------
Table Apx E-12. Hierarchy guiding integration of occupational exposure data/information
For occupational exposures, the generic hierarchy of preferences, listed from highest to lowest
levels, is as follows (and may be modified based on the assessment):
1.
Monitoring data:
Highest
a.
Personal and directly applicable
Preferred
b.
Area and directly applicable
c.
Personal and potentially applicable or similar
i
d.
Area and potentially applicable or similar
2.
Modeling approaches:
a.
Surrogate monitoring data: Modeling exposure for chemical ""X"
and condition of use 'A" based on observed monitoring data for
chemical "Y" and condition of use 'A", assuming a known
relationship (e.g., a linear relationship) between observed exposure
and physical property (e.g., vapor pressure).
b.
Fundamental modeling approaches: Modeling exposure for
chemical "X" for condition of use 'A" based on fundamental mass
transfer, thermodynamic, and kinetic phenomena for chemical "X"
and data for condition of use 'A"
c.
Fundamental modeling approaches (with surrogacy): A modeling
approach following item 2b, but using surrogate data in the model,
such as data for condition of use "ET judged to be similar to
condition of use 'A"
d.
Statistical regression modeling approaches: Modeling exposure for
chemical "X" in condition of use 'A" using a statistical regression
model developed based on:
i. Observed monitoring data for chemical ""X" statistically
correlated with observed data specific for condition of
use "B"judged to be similar to condition of use 'A"
such that replacement of input values in the model can
extrapolate exposure results to condition of use 'A"
ii. Observed monitoring data for chemical "Y" statistically
correlated with physical properties and/or molecular
structure such that an exposure prediction for chemical
"X" can be made (e.g., QSAR techniques)
3.
Occupational exposure limits (OELs):
a.
Company-specific OELs (for site-specific exposure assessments,
e.g., there is only one manufacturer who provides to EPA their
internal OEL but does not provide monitoring data)
Lowest
b.
OSHA PEL
Preferred
c.
Voluntary limits (ACGIH TLV, NIOSH REL, OARS WEEL
[formerly by AIHA])
Page 591 of 723
-------
Table Apx E-13. Hierarchy guiding integration of environmental release data/information
For environmental releases, the generic hierarchy of preferences, listed from highest to lowest
levels, is as follows (and may be modified based on the assessment):
Highest
Preferred
Lowest
Preferred
1. Monitoring and measured data:
a.
b.
Releases calculated from site-specific concentration in medium and
flow rate data (e.g., concentration in and flow rate of wastewater
effluent discharged through outfall)
Releases calculated from mass balances or emission factor methods
using site-specific measured data (e.g., process flow rates and
concentrations)
2. Modeling approaches:
a. Surrogate monitoring data: Modeling release for chemical "X" and
condition of use "A" based on observed monitoring data for
chemical "Y" and condition of use "A", assuming a known
relationship (e.g., a linear relationship) between observed release
and physical property (e.g., vapor pressure)
b. Fundamental modeling approaches: Modeling release for chemical
""X" for condition of use 'A" based on fundamental mass transfer,
thermodynamic, and kinetic phenomena for chemical "X" and data
for condition of use 'A"
c. Fundamental modeling approaches (with surrogacy): A modeling
approach following item 2b, but using surrogate data in the model,
such as data for condition of use "ET judged to be similar to
condition of use "A"
d. Statistical regression modeling approaches: Modeling release for
chemical "X" in condition of use 'A" using a statistical regression
model developed based on:
iii. Observed monitoring data for chemical "X" statistically
correlated with observed data specific for condition of
use "B" judged to be similar to condition of use 'A"
such that replacement of input values in the model can
extrapolate exposure results to condition of use 'A"
iv. Observed monitoring data for chemical "Y" statistically
correlated with physical properties and/or molecular
structure such that a release prediction for chemical
"X" can be made (e.g., QSAR techniques)
3. Release limits:
a. Company-specific limits (for site-specific exposure assessments,
e.g., there is only one manufacturer who provides to EPA their
internal limits (e.g., point-source permits) but does not provide
monitoring data)
b. NESHAP or effluent limitations/ requirements
Page 592 of 723
-------
E.8 Information on the Age of Employed Persons
For occupational exposures, EPA assessed exposures to workers and ONUs. TableApx E-14 presents
the percentage of employed workers and ONUs who may be susceptible subpopulations within select
industry sectors relevant to HBCD conditions of use. The percentages were calculated using Current
Population Survey (CPS) data for 2017. CPS is a monthly survey of households conducted by the
Bureau of Census for the Bureau of Labor Statistics (BLS) and provides a comprehensive body of data
on the labor force characteristics. Statistics for the following subpopulations of workers and ONUs are
provided: individuals age 16 to 19, men and women of reproductive age,23 and the elderly. CPS
considers "reproductive age" as age 16 to 54. As shown in Table Apx E-14, men make up the majority
of the workforce in the construction and manufacturing sectors. In other sectors, women (including those
of reproductive age and elderly women) make up nearly half of the workforce.
Adolescents (16 to <21 years old) appear to be generally a small part of the total workforce based on
CPS data for employed individuals between 16 and 19 years of age. Table Apx E-15 presents further
breakdown on this subset of adolescents employed by industry subsectors. As shown in the table, they
comprise less than two percent of the workforce. These data do not cover all adolescents in the HBCD
workforce because of the different age range used by the BLS.
Table Apx E-14. Percentage of Employed Persons by Age, Sex, and Industry Sector
Age Group
Sex
Construction
Manufacturing
Wholesale and
retail trade
Professional and
business services
Adolescent
(16-19 years)
Male
1.7%
0.8%
3.0%
0.7%
Female
0.1%
0.4%
3.2%
0.5%
Reproductive
Age
(16-54 years)
Male
72.2%
52.9%
42.8%
44.4%
Female
6.8%
22.2%
35.4%
32.8%
Elderly (55+)
Male
18.8%
17.5%
12.3%
13.4%
Female
2.3%
7.3%
9.6%
9.4%
Source: (U.S. BLS. 2017). Percentage calculated using CPS table 14, "Employed persons in nonagricultural industries by
age, sex, race, and Hispanic or Latino ethnicity."
Table Apx E-15. Percentage of Employed Persons Age 16-19 Years by Detailed Industry Sector
Sector
Subsector
Adolescents (16-19 years)
Construction (No subsectors)
All
1.82%
Manufacturing
All
1.21%
Wholesale and retail trade
Wholesale trade
1.36%
Professional and business services
Waste management and remediation services
0.93%
Source: (U.S. BLS. 2017). Percentage calculated using CPS table 18b, "Employed persons by detailed industry and age."
The CPS uses 2012 Census industry classification, which was derived from the 2012 NAICS. The
Census classification uses the same basic structure as NAICS but is generally less detailed. HBCD
conditions of use fall under the following Census industry sectors:
23 While statistics on pregnant women are not available, CPS provides data on the number of employed female workers by
age group, which allows for determination of the number of employed women of reproductive age.
Page 593 of 723
-------
Construction - The Construction sector comprises establishments primarily engaged in the
construction of buildings or engineering projects (e.g., highways and utility systems).
Establishments primarily engaged in the preparation of sites for new construction and
establishments primarily engaged in subdividing land for sale as building sites also are included
in this sector. Construction work done may include new work, additions, alterations, or
maintenance and repairs. For HBCD, this sector covers the conditions of use for Installation of
XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings, and Other
Structures and Demolition and Disposal of XPS/EPS Foam Insulation Products in Residential,
Public and Commercial Buildings, and Other Structures.
Manufacturing - The Manufacturing sector comprises establishments engaged in the mechanical,
physical, or chemical transformation of materials, substances, or components into new products.
Establishments in the sector are often described as plants, factories, or mills. For HBCD, this
sector covers conditions of use that occur in an industrial setting, including: Compounding of
Polystyrene Resin to Produce XPS Masterbatch, Processing of HBCD to Produce XPS Foam
using XPS Masterbatch, Processing of HBCD to Produce XPS Foam Using HBCD Powder,
Processing of HBCD to Produce EPS Foam from Imported EPS Resin Beads, Processing of
HBCD to Produce SIPs and Automobile Replacement Parts from XPS/EPS Foam, Recycling of
EPS Foam and Reuse of XPS Foam, Formulation of Flux/Solder Pastes, and Use of Flux/Solder
Paste.
Wholesale and retail trade - The wholesale trade sector comprises establishments engaged in
wholesaling merchandise, generally without transformation, and rendering services incidental to
the sale of merchandise. Wholesalers normally operate from a warehouse or office. This sector
likely covers facilities that are engaged in the importation of HBCD or EPS resin beads
containing HBCD. The retail trade sector comprises establishments engaged in retailing
merchandise and rendering services incidental to the sale of merchandise.
Professional and business services - This sector comprises establishments that specialize in a
wide range of services. This sector covers waste management and remediation services, which
includes Recycling of Electronics Waste (E-Waste) Containing HIPS.
Page 594 of 723
-------
Appendix F ENVIRONMENTAL EXPOSURES
F.l Modeled Exposure Scenarios Across Conditions of Use
F.l.l Water Releases
Table Apx F-l. Scenarios Used Across Conditions of Use for Water Releases of HBCD
Scenario
Label
Condition of Use
Type of
Water
Release3
[SIC]
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
1.1
Repackaging of
import containers
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Dust emissions factor
for coarse particles
(>40 |im)
Lower
Value
29
1.6E+00
1.2
Repackaging of
import containers
On-site
WWT
[Plastic
Resinsl
90
Lower
Value
Dust emissions factor
for coarse particles
(>40 |im)
Higher
Value
300
1.5E-01
1.3
Repackaging of
import containers
On-site
WWT
[Plastic
Resinsl
90
Higher
Value
Dust emissions factor
for fine particles (<40
Hin)
Lower
Value
29
7.8E+00
1.4
Repackaging of
import containers
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Dust emissions factor
for fine particles (<40
Hin)
Higher
Value
300
7.6E-01
1.5
Repackaging of
import containers
POTW
[Ind
POTW]
90
Lower
Value
Dust emissions factor
for coarse particles
(>40 urn)
Lower
Value
29
1.6E+00
1.6
Repackaging of
import containers
POTW
[Ind
POTW1
90
Lower
Value
Dust emissions factor
for coarse particles
(>40 urn)
Higher
Value
300
1.5E-01
1.7
Repackaging of
import containers
POTW
[Ind
POTW1
90
Higher
Value
Dust emissions factor
for fine particles (<40
um)
Lower
Value
29
7.8E+00
1.8
Repackaging of
import containers
POTW
[Ind
POTW1
90
Higher
Value
Dust emissions factor
for fine particles (<40
um)
Higher
Value
300
7.6E-01
2.1
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
Surface
Water
0
Lower
Value
Average calculated
emission factor from
EURAR data
Lower
Value
10
1.5E-01
2.2
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
Surface
Water
0
Lower
Value
Average calculated
emission factor from
EURAR data
Higher
Value
60
2.4E-02
2.3
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
Surface
Water
0
Higher
Value
EURAR's 'worst-
case' emission factor
Lower
Value
10
3.4E-01
Page 595 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
2.4
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
Surface
Water
0
Higher
Value
EURAR's 'worst-
case' emission factor
Higher
Value
60
5.6E-02
2.5
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Lower
Value
10
1.5E-01
2.6
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Higher
Value
60
2.4E-02
2.7
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
On-site
WWT
[Plastic
Resins]
90
Higher
Value
EURAR's 'worst-
case' emission factor
Lower
Value
10
3.4E-01
2.8
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
On-site
WWT
[Plastic
Resins]
90
Higher
Value
EURAR's 'worst-
case' emission factor
Higher
Value
60
5.6E-02
2.9
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
POTW
[Ind
POTW]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Lower
Value
10
1.5E-01
2.10
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
POTW
[Ind
POTW]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Higher
Value
60
2.4E-02
2.11
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
POTW
[Ind
POTW]
90
Higher
Value
EURAR's 'worst-
case' emission factor
Lower
Value
10
3.4E-01
2.12
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
POTW
[Ind
POTW]
90
Higher
Value
EURAR's 'worst-
case' emission factor
Higher
Value
60
5.6E-02
3.1
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
Surface
Water
0
Lower
Value
Average calculated
emission factor from
EURAR data
Lower
Value
1
4.9E-01
3.2
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
Surface
Water
0
Lower
Value
Average calculated
emission factor from
EURAR data
Higher
Value
15
3.2E-02
3.3
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
Surface
Water
0
Higher
Value
EURAR's 'worst-
case' emission factor
Lower
Value
1
1.2E+00
Page 596 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
3.4
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
Surface
Water
0
Higher
Value
EURAR's 'worst-
case' emission factor
Higher
Value
15
8.0E-02
3.5
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Lower
Value
1
4.9E-01
3.6
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Higher
Value
15
3.2E-02
3.7
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
On-site
WWT
[Plastic
Resinsl
90
Higher
Value
EURAR's 'worst-
case' emission factor
Lower
Value
1
1.2E+00
3.8
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
On-site
WWT
[Plastic
Resinsl
90
Higher
Value
EURAR's 'worst-
case' emission factor
Higher
Value
15
8.0E-02
3.9
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
POTW
[Ind
POTW]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Lower
Value
1
4.9E-01
3.10
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
POTW
[Ind
POTW]
90
Lower
Value
Average calculated
emission factor from
EURAR data
Higher
Value
15
3.2E-02
3.11
3. Manufacturing
of XPS Foam
using XPS
Masterbatch
POTW
[Ind
POTW]
90
Higher
Value
EURAR's 'worst-
case' emission factor
Lower
Value
1
1.2E+00
3.12
Manufacturing of
XPS Foam using
XPS Masterbatch
POTW
[Ind
POTW]
90
Higher
Value
EURAR's 'worst-
case' emission factor
Higher
Value
15
8.0E-02
4.1
Manufacturing of
XPS Foam using
HBCD Powder
Surface
Water
0
-
Average calculated
emission factor from
EURAR data
Lower
Value
1
4.6E-01
4.2
Manufacturing of
XPS Foam using
HBCD Powder
Surface
Water
0
-
Average calculated
emission factor from
EURAR data
Higher
Value
12
3.9E-02
4.3
Manufacturing of
XPS Foam using
HBCD Powder
On-site
WWT
[Plastic
Resins]
90
-
Average calculated
emission factor from
EURAR data
Lower
Value
1
4.6E-01
4.4
Manufacturing of
XPS Foam using
HBCD Powder
On-site
WWT
[Plastic
Resins]
90
-
Average calculated
emission factor from
EURAR data
Higher
Value
12
3.9E-02
Page 597 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
4.5
Manufacturing of
XPS Foam using
HBCD Powder
POTW
[Ind
POTW]
90
-
Average calculated
emission factor from
EURAR data
Lower
Value
1
4.6E-01
4.6
Manufacturing of
XPS Foam using
HBCD Powder
POTW
[Ind
POTW]
90
-
Average calculated
emission factor from
EURAR data
Higher
Value
12
3.9E-02
5.1
Manufacturing of
EPS Foam from
Surface
0
Lower
Dust emissions
during converting
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Lower
16
3.1E+01
Imported EPS
Resin beads
Water
Value
Value
Transport Containers
Model emission
factor
5.2
Manufacturing of
EPS Foam from
On-site
WWT
90
Lower
Dust emissions
during converting
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
16
3.1E+01
Imported EPS
Resin beads
[Plastic
Resins]
Value
Value
5.3
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
POTW
[Ind
POTW]
90
Lower
Value
Dust emissions
during converting
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
16
3.1E+01
5.4
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
Surface
Water
0
Lower
Value
Dust emissions
during converting
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Higher
Value
140
3.6E+00
Transport Containers
Model emission
factor
5.5
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Dust emissions
during converting
process emission
factor (lower) and
EPA/OPPT Solid
Higher
Value
140
3.6E+00
Page 598 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
Residuals in
Transport Containers
Model emission
factor
5.6
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
POTW
[Ind
POTW]
90
Lower
Value
Dust emissions
during converting
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
3.6E+00
5.7
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
Surface
Water
0
Higher
Value
Dust emissions
during converting
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
16
4.2E+01
5.8
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Dust emissions
during converting
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
16
4.2E+01
5.9
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
POTW
[Ind
POTW]
90
Higher
Value
Dust emissions
during converting
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
16
4.2E+01
5.10
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
Surface
Water
0
Higher
Value
Dust emissions
during converting
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
4.9E+00
Page 599 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
5.11
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Dust emissions
during converting
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
4.9E+00
5.12
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
POTW
[Ind
POTW]
90
Higher
Value
Dust emissions
during converting
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
4.9E+00
6.1
Manufacturing of
SIPs and
Automobile
Replacement
Parts
Surface
Water
0
Lower
Value
Dust release during
cutting of foam
Lower
Value
16
1.4E-01
6.2
Manufacturing of
SIPs and
Automobile
Replacement
Parts
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Dust release during
cutting of foam
Lower
Value
16
1.4E-01
6.3
Manufacturing of
SIPs and
Automobile
Replacement
Parts
POTW
[Ind
POTW]
90
Lower
Value
Dust release during
cutting of foam
Lower
Value
16
1.4E-01
6.4
Manufacturing of
SIPs and
Automobile
Replacement
Parts
Surface
Water
0
Lower
Value
Dust release during
cutting of foam
Higher
Value
300
7.6E-03
6.5
Manufacturing of
SIPs and
Automobile
Replacement
Parts
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Dust release during
cutting of foam
Higher
Value
300
7.6E-03
6.6
Manufacturing of
SIPs and
Automobile
Replacement
Parts
POTW
[Ind
POTW]
90
Lower
Value
Dust release during
cutting of foam
Higher
Value
300
7.6E-03
6.7
Manufacturing of
SIPs and
Automobile
Replacement
Parts
Surface
Water
0
Higher
Value
Dust release during
sawing of foam
Lower
Value
16
6.4E-01
Page 600 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
6.8
Manufacturing of
SIPs and
Automobile
Replacement
Parts
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Dust release during
sawing of foam
Lower
Value
16
6.4E-01
6.9
Manufacturing of
SIPs and
Automobile
Replacement
Parts
POTW
[Ind
POTW]
90
Higher
Value
Dust release during
sawing of foam
Lower
Value
16
6.4E-01
6.1
Manufacturing of
SIPs and
Automobile
Replacement
Parts
Surface
Water
0
Higher
Value
Dust release during
sawing of foam
Higher
Value
300
3.4E-02
6.11
Manufacturing of
SIPs and
Automobile
Replacement
Parts
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Dust release during
sawing of foam
Higher
Value
300
3.4E-02
6.12
Manufacturing of
SIPs and
Automobile
Replacement
Parts
POTW
[Ind
POTW]
90
Higher
Value
Dust release during
sawing of foam
Higher
Value
300
3.4E-02
8.1
Installation of
Insulation in
Buildings
Surface
water
0
Lower
Value
Dust release during
cutting of foam
Lower
Value
1
8.5E-04
8.2
Installation of
Insulation in
Buildings
POTW
[Ind
POTW]
90
Lower
Value
Dust release during
cutting of foam
Lower
Value
1
8.5E-04
8.3
Installation of
Insulation in
Buildings
Surface
water
0
Higher
Value
Dust release during
sawing of foam
Higher
Value
3
0.10
8.4
Installation of
Insulation in
Buildings
POTW
[Ind
POTW]
90
Higher
Value
Dust release during
sawing of foam
Higher
Value
3
0.10
9.1
Generation of
foam particles
during demolition
Surface
Water
0
Lower
Value
Dust release during
breaking of foam
Lower
Value
1
7.57E-04
9.2
Generation of
foam particles
during demolition
Surface
Water
0
Higher
Value
Dust release during
breaking of foam
Lower
Value
1
0.675
9.3
Generation of
foam particles
during demolition
POTW
[Ind
POTW]
90
Lower
Value
Dust release during
breaking of foam
Lower
Value
1
7.57E-04
9.4
Generation of
foam particles
during demolition
POTW
[Ind
POTW]
90
Higher
Value
Dust release during
breaking of foam
Lower
Value
1
0.675
Page 601 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
10.1
Recycling of EPS
Foam
surface
water
0
Lower
Value
Dust emissions
during recycling
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
1
6.7E-01
10.2
Recycling of EPS
Foam
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Dust emissions
during recycling
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
1
6.7E-01
10.3
Recycling of EPS
Foam
POTW
[Ind
POTW]
90
Lower
Value
Dust emissions
during recycling
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
1
6.7E-01
10.4
Recycling of EPS
Foam
Surface
Water
0
Lower
Value
Dust emissions
during recycling
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
4.8E-03
10.5
Recycling of EPS
Foam
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Dust emissions
during recycling
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
4.8E-03
10.6
Recycling of EPS
Foam
POTW
[Ind
POTW]
90
Lower
Value
Dust emissions
during recycling
process emission
factor (lower) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
4.8E-03
Page 602 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
10.7
Recycling of EPS
Foam
Surface
Water
0
Higher
Value
Dust emissions
during recycling
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
1
7.9E-01
10.8
Recycling of EPS
Foam
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Dust emissions
during recycling
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
1
7.9E-01
10.9
Recycling of EPS
Foam
POTW
[Ind
POTW]
90
Higher
Value
Dust emissions
during recycling
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Lower
Value
1
7.9E-01
10.1
Recycling of EPS
Foam
Surface
Water
0
Higher
Value
Dust emissions
during recycling
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
5.7E-03
10.11
Recycling of EPS
Foam
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Dust emissions
during recycling
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
5.7E-03
10.12
Recycling of EPS
Foam
POTW
[Ind
POTW]
90
Higher
Value
Dust emissions
during recycling
process emission
factor (higher) and
EPA/OPPT Solid
Residuals in
Transport Containers
Model emission
factor
Higher
Value
140
5.7E-03
Page 603 of 723
-------
Scenario
Label
Condition of Use
Type of
Water
Release"
rsici
WW TP
%
Emission
Facto rb
Characterization of
Emission Factor
Number
of Release
Days0
Release
Days
Daily Release
(kg/site/day)
12.1
Use of Solder
On-site
WWT
[Plastic
Resinsl
90
Lower
Value
Equipment cleaning
emission factor
(lower) (OECD
2010a)
Lower
Value
4
2.5E-02
12.2
Use of Solder
POTW
[Ind
POTW]
90
Lower
Value
Equipment cleaning
emission factor
(lower) (OECD
2010a)
Lower
Value
4
2.5E-02
12.3
Use of Solder
On-site
WWT
[Plastic
Resins]
90
Lower
Value
Equipment cleaning
emission factor
(lower) (OECD
2010a)
Higher
Value
300
3.3E-04
12.4
Use of Solder
POTW
[Ind
POTW]
90
Lower
Value
Equipment cleaning
emission factor
(lower) (OECD
2010a)
Higher
Value
300
3.3E-04
12.5
Use of Solder
On-site
WWT
[Plastic
Resinsl
90
Higher
Value
Equipment cleaning
emission factor
(hisher) (OECD
2010a)
Lower
Value
4
5.0E-02
12.6
Use of Solder
POTW
[Ind
POTW]
90
Higher
Value
Equipment cleaning
emission factor
(liisher) (OECD
2010a)
Lower
Value
4
5.0E-02
12.7
Use of Solder
On-site
WWT
[Plastic
Resins]
90
Higher
Value
Equipment cleaning
emission factor
(liisher) (OECD
2010a)
Higher
Value
300
6.7E-04
12.8
Use of Solder
POTW
[Ind
POTW]
90
Higher
Value
Equipment cleaning
emission factor
(liisher) (OECD
2010a)
Higher
Value
300
6.7E-04
Tor each release source, water releases were modeled depending on the potential for the release to go directly to surface
water, to an on-site wastewater treatment or publicly owned treatment works.
bWhere identified in literature, EPA utilized a range of emission factors with the characterization of those emission factor
described in further details in Section 2.2.
°Where identified in literature, EPA utilized a range of release days based on the specific condition of use as discussed further
in Section 2.2.
F.1.2 Air Releases
Table A
px F-2. Scenarios Used Across Conditions of Use for Air Releases of HBCD
Scenario
Label
Conditions of Use
Type of Air
Release
Characterization of
Emission Factor
Emission
Factor
Release
Days
Number of
Release Days
Daily Release
(kg/site/day)
1.1
Import/Repackaging
Fugitive
Dust release during
unloading of HBCD
lower
value
lower value
29
1.6E+00
1.2
Import/Repackaging
Fugitive
Dust release during
unloading of HBCD
lower
value
higher
value
300
1.5E-01
1.3
Import/Repackaging
Fugitive
Dust release during
unloading of HBCD
upper
value
lower value
29
7.8E+00
Page 604 of 723
-------
Scenario
Label
Conditions of Use
Type of Air
Release
Characterization of
Emission Factor
Emission
Factor
Release
Days
Number of
Release Days
Daily Release
(kg/site/day)
1.4
Import/Repackaging
Fugitive
Dust release during
unloading of HBCD
upper
value
higher
value
300
7.6E-01
1.5
Import/Repackaging
Stack
Dust release during
unloading of HBCD
lower
value
lower value
29
1.6E+00
1.6
Import/Repackaging
Stack
Dust release during
unloading of HBCD
lower
value
higher
value
300
1.5E-01
1.7
Import/Repackaging
Stack
Dust release during
unloading of HBCD
upper
value
lower value
29
7.8E+00
1.8
Import/Repackaging
Stack
Dust release during
unloading of HBCD
upper
value
higher
value
300
7.6E-01
1.9
Import/Repackaging
Incineration
Dust release during
unloading of HBCD
lower
value
lower value
29
1.6E+00
1.10
Import/Repackaging
Incineration
Dust release during
unloading of HBCD
lower
value
higher
value
300
1.5E-01
1.11
Import/Repackaging
Incineration
Dust release during
unloading of HBCD
upper
value
lower value
29
7.8E+00
1.12
Import/Repackaging
Incineration
Dust release during
unloading of HBCD
upper
value
higher
value
300
7.6E-01
2.1
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
fugitive
Average calculated
emission factor from
EURAR data
lower
value
lower value
10
2.8E-02
2.2
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
fugitive
Average calculated
emission factor from
EURAR data
lower
value
higher
value
60
4.6E-03
2.3
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
fugitive
Average calculated
emission factor from
EURAR data
upper
value
lower value
10
3.3E-02
2.4
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
fugitive
Average calculated
emission factor from
EURAR data
upper
value
higher
value
60
5.5E-03
2.5
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
stack
Average calculated
emission factor from
EURAR data
lower
value
lower value
10
2.8E-02
Page 605 of 723
-------
Scenario
Label
Conditions of Use
Type of Air
Release
Characterization of
Emission Factor
Emission
Factor
Release
Days
Number of
Release Days
Daily Release
(kg/site/day)
2.6
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
stack
Average calculated
emission factor from
EURAR data
lower
value
higher
value
60
4.6E-03
2.7
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
stack
Average calculated
emission factor from
EURAR data
upper
value
lower value
10
3.3E-02
2.8
Compounding of
Polystyrene Resin to
Produce XPS
Masterbatch
stack
Average calculated
emission factor from
EURAR data
upper
value
higher
value
60
5.5E-03
3.1
Manufacturing of
XPS Foam using XPS
Masterbatch
fugitive
Average calculated
emission factor from
EURAR data
central
value
lower value
1
2.6E+00
3.2
Manufacturing of
XPS Foam using XPS
Masterbatch
fugitive
Average calculated
emission factor from
EURAR data
central
value
higher
value
16
1.6E-01
3.3
Manufacturing of
XPS Foam using XPS
Masterbatch
stack
Average calculated
emission factor from
EURAR data
central
value
lower value
1
2.6E+00
3.4
Manufacturing of
XPS Foam using XPS
Masterbatch
stack
Average calculated
emission factor from
EURAR data
central
value
higher
value
16
1.6E-01
4.1
Manufacturing of
XPS Foam using
HBCD Powder
fugitive
Average calculated
emission factor from
EURAR data
central
value
lower value
1
3.3E-01
4.2
Manufacturing of
XPS Foam using
HBCD Powder
fugitive
Average calculated
emission factor from
EURAR data
central
value
higher
value
16
2.1E-02
4.3
Manufacturing of
XPS Foam using
HBCD Powder
stack
Average calculated
emission factor from
EURAR data
central
value
lower value
1
3.3E-01
4.4
Manufacturing of
XPS Foam using
HBCD Powder
stack
Average calculated
emission factor from
EURAR data
central
value
higher
value
16
2.1E-02
4.5
Manufacturing of
XPS Foam using
HBCD Powder
stack
TRI data
empirical
value
lower value
1
1.8E+00
4.6
Manufacturing of
XPS Foam using
HBCD Powder
stack
TRI data
empirical
value
higher
value
16
1.1E-01
Page 606 of 723
-------
Scenario
Label
Conditions of Use
Type of Air
Release
Characterization of
Emission Factor
Emission
Factor
Release
Days
Number of
Release Days
Daily Release
(kg/site/day)
4.7
Manufacturing of
XPS Foam using
HBCD Powder
incineration
TRI data
empirical
value
lower value
1
3.1E+01
4.8
Manufacturing of
XPS Foam using
HBCD Powder
incineration
TRI data
empirical
value
higher
value
16
1.9E+00
4.9
Manufacturing of
XPS Foam using
HBCD Powder
stack
TRI data
empirical
value
lower value
1
2.1E+01
4.10
Manufacturing of
XPS Foam using
HBCD Powder
stack
TRI data
empirical
value
higher
value
16
1.3E+00
4.11
Manufacturing of
XPS Foam using
HBCD Powder
incineration
TRI data
empirical
value
lower value
1
2.3E+01
4.12
Manufacturing of
XPS Foam using
HBCD Powder
incineration
TRI data
empirical
value
higher
value
16
1.5E+00
5.1
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
stack
Dust release during
converting process
lower
value
lower value
16
2.8E+00
5.2
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
stack
Dust release during
converting process
lower
value
higher
value
140
3.2E-01
5.3
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
stack
Dust release during
converting process
upper
value
lower value
16
1.4E+01
5.4
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
stack
Dust release during
converting process
upper
value
higher
value
140
1.6E+00
5.5
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
fugitive
Dust release during
converting process
lower
value
lower value
16
2.8E+00
5.6
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
fugitive
Dust release during
converting process
lower
value
higher
value
140
3.2E-01
5.7
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
fugitive
Dust release during
converting process
upper
value
lower value
16
1.4E+01
Page 607 of 723
-------
Scenario
Label
Conditions of Use
Type of Air
Release
Characterization of
Emission Factor
Emission
Factor
Release
Days
Number of
Release Days
Daily Release
(kg/site/day)
5.8
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
fugitive
Dust release during
converting process
upper
value
higher
value
140
1.6E+00
5.9
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
incineration
Dust release during
converting process
lower
value
lower value
16
6.0E+01
5.10
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
incineration
Dust release during
converting process
lower
value
higher
value
140
6.8E+00
5.11
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
incineration
Dust release during
converting process
upper
value
lower value
16
1.1E+02
5.12
Manufacturing of
EPS Foam from
Imported EPS Resin
beads
incineration
Dust release during
converting process
upper
value
higher
value
140
1.3E+01
6.1
Manufacturing of
SIPs and Automobile
Replacement Parts
fugitive
Dust release during
sawing / cutting of
foam
lower
value
lower value
16
1.4E-01
6.2
Manufacturing of
SIPs and Automobile
Replacement Parts
fugitive
Dust release during
sawing / cutting of
foam
lower
value
higher
value
300
7.6E-03
6.3
Manufacturing of
SIPs and Automobile
Replacement Parts
fugitive
Dust release during
sawing / cutting of
foam
upper
value
lower value
16
6.4E-01
6.4
Manufacturing of
SIPs and Automobile
Replacement Parts
fugitive
Dust release during
sawing / cutting of
foam
upper
value
higher
value
300
3.4E-02
6.5
Manufacturing of
SIPs and Automobile
Replacement Parts
stack
Dust release during
sawing / cutting of
foam
lower
value
lower value
16
1.4E-01
6.6
Manufacturing of
SIPs and Automobile
Replacement Parts
stack
Dust release during
sawing / cutting of
foam
lower
value
higher
value
300
7.6E-03
6.7
Manufacturing of
SIPs and Automobile
Replacement Parts
stack
Dust release during
sawing / cutting of
foam
upper
value
lower value
16
6.4E-01
Page 608 of 723
-------
Scenario
Label
Conditions of Use
Type of Air
Release
Characterization of
Emission Factor
Emission
Factor
Release
Days
Number of
Release Days
Daily Release
(kg/site/day)
6.8
Manufacturing of
SIPs and Automobile
Replacement Parts
stack
Dust release during
sawing / cutting of
foam
upper
value
higher
value
300
3.4E-02
6.9
Manufacturing of
SIPs and Automobile
Replacement Parts
incineration
Dust release during
sawing / cutting of
foam
lower
value
lower value
16
2.8E+01
6.10
Manufacturing of
SIPs and Automobile
Replacement Parts
incineration
Dust release during
sawing / cutting of
foam
lower
value
higher
value
300
1.5E+00
6.11
Manufacturing of
SIPs and Automobile
Replacement Parts
incineration
Dust release during
sawing / cutting of
foam
upper
value
lower value
16
7.2E+01
6.12
Manufacturing of
SIPs and Automobile
Replacement Parts
incineration
Dust release during
sawing / cutting of
foam
upper
value
higher
value
300
3.8E+00
8.1
Installation of
Insulation in
Buildings
fugitive
Dust release during
sawing / cutting of
foam
lower
value
lower value
1
8.5E-04
8.2
Installation of
Insulation in
Buildings
fugitive
Dust release during
sawing / cutting of
foam
upper
value
higher
value
3
8.5E-04
8.3
Installation of
Insulation in
Buildings
incineration
Dust release during
sawing / cutting of
foam
lower
value
lower value
1
1.0E-02
8.4
Installation of
Insulation in
Buildings
incineration
Dust release during
sawing / cutting of
foam
upper
value
higher
value
3
1.0E-02
9.1
Generation of foam
particles during
demolition
fugitive
Dust release during
breaking of foam
lower
value
lower value
1
7.57E-04
9.2
Generation of foam
particles during
demolition
fugitive
Dust release during
breaking of foam
higher
value
lower value
1
0.675
10.1
Recycling of EPS
Foam
fugitive
Dust release from
grinding of foam
lower
value
lower value
1
3.2E-02
10.2
Recycling of EPS
Foam
fugitive
Dust release from
grinding of foam
lower
value
higher
value
140
2.3E-04
10.3
Recycling of EPS
Foam
fugitive
Dust release from
grinding of foam
upper
value
lower value
1
1.6E-01
10.4
Recycling of EPS
Foam
fugitive
Dust release from
grinding of foam
upper
value
higher
value
140
1.1E-03
10.5
Recycling of EPS
Foam
stack
Dust release from
grinding of foam
lower
value
lower value
1
3.2E-02
10.6
Recycling of EPS
Foam
stack
Dust release from
grinding of foam
lower
value
higher
value
140
2.3E-04
Page 609 of 723
-------
Scenario
Label
Conditions of Use
Type of Air
Release
Characterization of
Emission Factor
Emission
Factor
Release
Days
Number of
Release Days
Daily Release
(kg/site/day)
10.7
Recycling of EPS
Foam
stack
Dust release from
grinding of foam
upper
value
lower value
1
1.6E-01
10.8
Recycling of EPS
Foam
stack
Dust release from
grinding of foam
upper
value
higher
value
140
1.1E-03
10.9
Recycling of EPS
Foam
incineration
Dust release from
grinding of foam
lower
value
lower value
1
6.7E-01
10.10
Recycling of EPS
Foam
incineration
Dust release from
grinding of foam
lower
value
higher
value
140
4.8E-03
10.11
Recycling of EPS
Foam
incineration
Dust release from
grinding of foam
upper
value
lower value
1
7.9E-01
10.12
Recycling of EPS
Foam
incineration
Dust release from
grinding of foam
upper
value
higher
value
140
5.7E-03
11.1
Formulation of solder
fugitive
TRI data
empirical
value
lower value
5
9.1E-02
11.2
Formulation of solder
fugitive
TRI data
empirical
value
higher
value
300
1.5E-03
11.3
Formulation of solder
stack
TRI data
empirical
value
lower value
5
1.3E+00
11.4
Formulation of solder
stack
TRI data
empirical
value
higher
value
300
2.1E-02
12.1
Use of Solder
incineration
Disposal of transport
containers and
overapplied/unused
solder-incineration
higher
value
lower value
4
2.2E-01
12.2
Use of Solder
incineration
Disposal of transport
containers and
overapplied/unused
solder-incineration
higher
value
higher
value
300
3.0E-03
12.3
Use of Solder
incineration
Disposal of transport
containers and
overapplied/unused
solder-incineration
lower
value
lower value
4
2.0E-01
12.4
Use of Solder
incineration
Disposal of transport
containers and
overapplied/unused
solder-incineration
lower
value
higher
value
300
2.7E-03
F.2 E-FAST and VVWM-PSC Modeling
EPA's Exposure and Fate Assessment Screening Tool (E-FAST), Version 2.0, was specifically
developed to support EPA assessments of potential environmental exposures. The E-FAST model
contains default parameter values that allow for exposure estimations of a chemical in the surface water
after a source emits the chemical into a water body considering simple dilution under four stream flow
conditions (harmonic mean, 30Q5, 7Q10, and 1Q10 flow). Details of E-FAST are given in the model
user guide at https://www.epa.gov/tsca-screening-tools/e-fast-exposure-and-fate-assessment-screening-
tool-2014-documentati on-manual.
The Point Source Calculator (PSC) is variation of the Variable Volume Water Model (VVWM) used by
the USEPA for chemical exposure in surface waters. Details of the VVWM are given in the model user
Page 610 of 723
-------
guide at https://www.epa.gov/tsca-screening-tools/point-source-calculator-version-105-psc-vl05. The
PSC is similar to the SWCC and PFAM in that employs a user-friendly interface that generates a
VVWM input file, runs the VVWM, and processes the data. The differences in PSC, SWCC, and PFAM
are essentially in the user interface and in the post processing output. In the case of the PSC, the user
interface and post processing are intended to assess chemicals that flow directly into a water body and to
compare the chemical concentrations to levels of concern.
The conceptualization of the processes in the PSC is given by FigureApx F-l. In this conceptualization,
the VVWM is used to represent a segment of a water body which receives a direct application of a
chemical. The chemical immediately mixes with the water column of the segment. The water column is
coupled to a sediment layer and chemical can move into the sediment by a first-order mass transfer
process. The chemical can degrade in the water column by user-supplied inputs of hydrolysis,
photolysis, and general degradation. Water column chemical can also volatilize according to chemical
properties supplied by the user. In the benthic region, the chemical can degrade by hydrolysis and a
general benthic degradation rate as supplied by the user. Partitioning to suspended sediment as well as
benthic solids occurs according to input values for either an organic carbon portioning linear coefficient
(K OC ) or a linear sorption coefficient (Kd). In all cases, the waterbody is modeled as a single segment
(comprised of a water column and a benthic region), with the appropriate segment being the one that
receives the direct application of the chemical.
Chemical Input
Volatilization
Water Column
Inflow
Degradation due to:
metabolism, hydrolysis,
photolysis, etc.
Water-C'olumn-to- Benthic
.Mass Transfer
e
Degradation due to:
metabolism
hvdrotvsis.
Benthic Region
0
on due to: j
.etc. \ )
Washout, Dispersion
Figure Apx F-l. Depiction of the Chemical Processes in the Point Source Calculator
Table Apx F-3. Estimated HBCD Surface Water (^ig/L) Concentrations Using E-FAST
Scenario Label
Harmonic Mean
SWC
50tli Percentile
Harmonic Mean
SWC
10th Percentile
7Q10 SWC
50th percentile
7Q10 SWC
10th percentile
Wl.l
1.2E-01
3.5E+00
3.9E-01
1.9E+01
W1.2
1.1E-02
3.4E-01
3.7E-02
1.9E+00
W1.3
5.9E-01
1.8E+01
1.9E+00
9.8E+01
W1.4
5.7E-02
1.7E+00
1.9E-01
9.4E+00
W1.5
5.4E-01
3.9E+00
2.0E+00
2.0E+01
W1.6
5.2E-02
3.8E-01
1.9E-01
1.9E+00
Page 611 of 723
-------
Scenario Label
Harmonic Mean
SWC
50th Percentile
Harmonic Mean
SWC
10th Percentile
7Q10 SWC
50th percentile
7Q10 SWC
10th percentile
W1.7
2.7E+00
2.0E+01
1.0E+01
1.0E+02
W1.8
2.6E-01
1.9E+00
9.7E-01
9.7E+00
W2.1
1.1E-01
3.4E+00
3.7E-01
1.9E+01
W2.2
1.9E-02
5.5E-01
6.1E-02
3.0E+00
W2.3
2.5E-01
7.6E+00
8.4E-01
4.2E+01
W2.4
4.2E-02
1.3E+00
1.4E-01
7.0E+00
W2.5
1.1E-02
3.4E-01
3.7E-02
1.9E+00
W2.6
1.9E-03
5.5E-02
6.1E-03
3.0E-01
W2.7
2.6E-02
7.6E-01
8.4E-02
4.2E+00
W2.8
4.2E-03
1.3E-01
1.4E-02
7.0E-01
W2.9
5.2E-02
3.8E-01
1.9E-01
1.9E+00
W2.10
8.5E-03
6.2E-02
3.1E-02
3.1E-01
W2.ll
1.2E-01
8.5E-01
4.3E-01
4.3E+00
W2.12
2.0E-02
1.4E-01
7.2E-02
7.2E-01
W3.1
3.7E-01
1.1E+01
1.2E+00
6.1E+01
W3.2
2.5E-02
7.3E-01
8.0E-02
4.0E+00
W3.3
9.0E-01
2.7E+01
3.0E+00
1.5E+02
W3.4
6.1E-02
1.8E+00
2.0E-01
1.0E+01
W3.5
3.7E-02
1.1E+00
1.2E-01
6.1E+00
W3.6
2.5E-03
7.0E-02
1.0E-02
4.0E-01
W3.7
9.0E-02
2.7E+00
3.0E-01
1.5E+01
W3.8
6.1E-03
1.8E-01
2.0E-02
1.0E+00
W3.9
1.7E-01
1.2E+00
6.2E-01
6.3E+00
W3.10
1.1E-02
8.2E-02
4.1E-02
4.2E-01
W3.ll
4.1E-01
3.0E+00
1.5E+00
1.5E+01
W3.12
2.8E-02
2.0E-01
1.0E-01
1.0E+00
W4.1
3.5E-01
1.0E+01
1.2E+00
5.8E+01
W4.2
3.0E-02
8.8E-01
9.7E-02
4.9E+00
W4.3
3.5E-02
1.0E+00
1.2E-01
5.8E+00
W4.4
3.0E-03
8.8E-02
9.7E-03
4.9E-01
W4.5
1.6E-01
1.2E+00
5.9E-01
6.0E+00
W4.6
1.4E-02
9.9E-02
5.0E-02
5.0E-01
W5.1
2.4E+01
7.0E+02
7.7E+01
3.9E+03
W5.2
2.4E+00
7.0E+01
7.7E+00
3.9E+02
W5.3
1.1E+01
7.9E+01
4.0E+01
4.0E+02
W5.4
2.7E+00
8.0E+01
8.8E+00
4.4E+02
W5.5
2.7E-01
8.0E+00
8.8E-01
4.4E+01
W5.6
1.2E+00
9.0E+00
4.6E+00
4.6E+01
W5.7
3.2E+01
9.5E+02
1.1E+02
5.3E+03
W5.8
3.2E+00
9.5E+01
1.1E+01
5.3E+02
Page 612 of 723
-------
Scenario Label
Harmonic Mean
SWC
50th Percentile
Harmonic Mean
SWC
10th Percentile
7Q10 SWC
50th percentile
7Q10 SWC
10th percentile
W5.9
1.5E+01
1.1E+02
5.4E+01
5.5E+02
W5.1
3.7E+00
1.1E+02
1.2E+01
6.1E+02
W5.ll
3.7E-01
1.1E+01
1.2E+00
6.1E+01
W5.12
1.7E+00
1.2E+01
6.2E+00
6.3E+01
W6.1
1.1E-01
3.2E+00
3.5E-01
1.8E+01
W6.2
1.1E-02
3.2E-01
3.5E-02
1.8E+00
W6.3
5.0E-02
3.6E-01
1.8E-01
1.8E+00
W6.4
5.8E-03
1.7E-01
1.9E-02
9.5E-01
W6.5
5.8E-04
1.7E-02
1.9E-03
9.5E-02
W6.6
2.7E-03
1.9E-02
9.8E-03
9.9E-02
W6.7
4.8E-01
1.4E+01
1.6E+00
8.0E+01
W6.8
4.8E-02
1.4E+00
1.6E-01
8.0E+00
W6.9
2.2E-01
1.6E+00
8.2E-01
8.3E+00
W6.10
2.6E-02
7.6E-01
8.4E-02
4.2E+00
W6.ll
2.6E-03
7.6E-02
8.4E-03
4.2E-01
W6.12
1.2E-02
8.6E-02
4.4E-02
4.4E-01
W8.1
6.8E-03
7.7E-02
3.2E-02
8.0E-01
W8.2
6.8E-04
7.7E-03
3.2E-03
8.0E-02
W8.3
8.0E-01
9.0E+00
3.7E+00
9.4E+01
W8.4
8.0E-02
9.0E-01
3.7E-01
9.4E+00
W9.1
6.0E-03
6.8E-02
2.8E-02
7.1E-01
W9.2
6.0E-04
6.8E-03
2.8E-03
7.1E-02
W9.3
5.4E+00
6.1E+01
2.5E+01
6.4E+02
W9.4
5.4E-01
6.1E+00
2.5E+00
6.4E+01
W10.1
5.0E-01
1.5E+01
1.7E+00
8.3E+01
W10.2
5.0E-02
1.5E+00
1.7E-01
8.3E+00
W10.3
2.3E-01
1.7E+00
8.5E-01
8.6E+00
W10.4
3.6E-03
1.1E-01
1.2E-02
5.9E-01
W10.5
3.6E-04
1.1E-02
1.2E-03
5.9E-02
W10.6
1.7E-03
1.2E-02
6.1E-03
6.2E-02
W10.7
6.0E-01
1.8E+01
2.0E+00
9.9E+01
W10.8
6.0E-02
1.8E+00
2.0E-01
9.9E+00
W10.9
2.8E-01
2.0E+00
1.0E+00
1.0E+01
W10.10
4.3E-03
1.3E-01
1.4E-02
7.1E-01
W10.ll
4.3E-04
1.3E-02
1.4E-03
7.1E-02
W10.12
2.0E-03
1.4E-02
7.3E-03
7.3E-02
W12.1
1.9E-03
5.5E-02
6.2E-03
3.1E-01
W12.2
8.7E-03
6.3E-02
3.2E-02
3.2E-01
W12.3
2.5E-05
7.5E-04
8.3E-05
4.2E-03
W12.4
1.2E-04
8.4E-04
4.3E-04
4.3E-03
Page 613 of 723
-------
Scenario Label
Harmonic Mean
SWC
50th Percentile
Harmonic Mean
SWC
10th Percentile
7Q10 SWC
50th percentile
7Q10 SWC
10th percentile
12.5
3.8E-03
1.1E-01
1.2E-02
6.2E-01
12.6
1.7E-02
1.3E-01
6.4E-02
6.4E-01
12.7
5.0E-05
1.5E-03
1.7E-04
8.3E-03
12.8
2.3E-04
1.7E-03
8.5E-04
8.6E-03
F.3 IIOAC Modeling
The IIOAC modeling methodology is discussed in further detail in Appendix G. The tables below
present a summary of the modeled air deposition and estimated soil concentrations.
Table Apx F-4. Total Annual Particle Deposition from Facility Air Releases
Scenario Name
Range of Total Annual Particle Deposition (g/m2)
Fugitive
Stack
Incineration
Min - Max
Min - Max
Min - Max
1. Import/Repackaging
Fenceline
Community
3.58E-06 - 2.18E-05
1.23E-07 - 6.19E-07
1.52E-06 - 1.13E-05
1.03E-07 - 5.17E-07
6.20E-08 - 5.81E-07
4.35E-08 - 2.18E-07
2. Compounding of Polystyrene Resin to
Fenceline
Community
Jroduce XPS Masterbat
2.55E-08 - 3.48E-08
7.57E-10 - 9.04E-10
ch
1.29E-08 - 1.90E-08
6.32E-10 - 7.57E-10
n/a - n/a
n/a - n/a
3. Manufacturing of XPS Foam using XP
Fenceline
Community
S Masterbatch
2.64E-07 - 3.48E-07
7.15E-09 - 7.16E-09
1.38E-07 - 2.03E-07
5.98E-09 - 5.98E-09
n/a - n/a
n/a - n/a
4. Manufacturing of XPS Foam using HB
Fenceline
Community
CD Powder
3.32E-08 - 4.38E-08
9.00E-10 - 9.01E-10
1.73E-08 - 1.27E-06
7.53E-10 - 6.46E-08
5.85E-08 - 1.02E-07
3.58E-08 - 5.70E-08
5. Manufacturing of EPS Foam from Imp
Fenceline
Community
orted EPS Resin Beads
3.83E-06 - 2.28E-05
1.23E-07 - 6.18E-07
1.89E-06 - 1.19E-05
1.03E-07 - 5.19E-07
1.63E-06 - 5.48E-06
9.18E-07 - 1.75E-06
6. Manufacturing of SIPs and Automobil
Fenceline
Community
e Replacement Parts
1.81E-07 - 1.02E-06
6.20E-09 - 2.78E-08
7.69E-08 - 5.35E-07
5.22E-09 - 2.33E-08
6.23E-07 - 3.45E-06
4.37E-07 - 1.10E-06
8. Installation of XPS/EPS Foam Insulati
Structures
Fenceline
Community
an in Residential, Public
1.13E-10 - 3.48E-08
2.32E-12 - 8.17E-10
, and Commercial Buildii
n/a - n/a
n/a - n/a
ags, and Other
7.72E-10 - 1.65E-07
1.62E-10 - 3.21E-08
9. Demolition of XPS/EPS Foam Insulatic
Fenceline
Community
>n in Residential, Public
1.00E-10 - 8.95E-08
2.06E-12 - 1.84E-09
and Commercial Buildin
n/a - n/a
n/a - n/a
gs, and Other Structures
n/a - n/a
n/a - n/a
10. Recycling of EPS Foam
Fenceline
Community
2.68E-09 - 2.11E-08
8.64E-11 - 4.32E-10
1.33E-09 - 1.22E-08
7.23E-11 - 3.63E-10
1.14E-09 - 3.64E-09
6.42E-10 - 7.65E-10
11. Formulation of Solder
Page 614 of 723
-------
Range of Total Annual Particle Deposition (g/m2)
Scenario Name
Fugitive
Stack
Incineration
Min
Max
Min
Max
Min
Max
Fenceline
7.25E-08
- 8.48E-08
8.12E-07
- 1.17E-06
n/a
n/a
Community
2.81E-09
- 2.81E-09
4.22E-08
- 4.23E-08
n/a
n/a
12. Use of Solder
Fenceline
n/a
n/a
n/a
n/a
1.09E-09 -
4.03E-09
Community
n/a
n/a
n/a
n/a
7.66E-10 -
8.66E-10
Table Apx F-5. Estimated Soil Concentrations from Facility Air Releases
Range of Estimate Soil Concentration (jig/kg)
Scenario Name
Fugitive
Stack
Incineration
Min
Max
Min
Max
Min
Max
1. Import/Repackaging
Fenceline
2.11E-02 -
1.28E-01
8.95E-03
- 6.66E-02
3.64E-04 -
3.42E-03
Community
7.22E-04 -
3.64E-03
6.08E-04
- 3.04E-03
2.56E-04 -
1.29E-03
2. Compounding of Polystyrene Resin to Produce XPS Masterbatch
Fenceline
1.50E-04 -
2.05E-04
7.56E-05
- 1.12E-04
n/a
n/a
Community
4.45E-06 -
5.32E-06
3.72E-06
- 4.45E-06
n/a
n/a
3. Manufacturing of XPS Foam using XPS Masterbatch
Fenceline
1.55E-03 -
2.05E-03
8.10E-04
- 1.19E-03
n/a
n/a
Community
4.20E-05 -
4.21E-05
3.52E-05
- 3.52E-05
n/a
n/a
4. Manufacturing of XPS Foam using HBCD Powder
Fenceline
1.95E-04 -
2.58E-04
1.02E-04
- 7.46E-03
3.44E-04 -
5.98E-04
Community
5.29E-06 -
5.30E-06
4.43E-06
- 3.80E-04
2.11E-04 -
3.35E-04
5. Manufacturing of EPS Foam from Imported EPS Resin Beads
Fenceline
2.25E-02 -
1.34E-01
1.11E-02
- 7.00E-02
9.59E-03 -
3.22E-02
Community
7.26E-04 -
3.64E-03
6.08E-04
- 3.05E-03
5.40E-03 -
1.03E-02
6. Manufacturing of SIPs and Automobile Replacement Parts
Fenceline
1.07E-03 -
6.03E-03
4.53E-04
- 3.15E-03
3.66E-03 -
2.03E-02
Community
3.65E-05 -
1.64E-04
3.07E-05
- 1.37E-04
2.57E-03 -
6.48E-03
8. Installation of XPS/EPS Foam Insulation in Residential, Public, and Commercial Buildings, and Other
Structures
Fenceline
6.64E-07 -
2.05E-04
n/a
n/a
4.54E-06 -
9.68E-04
Community
1.36E-08 -
4.81E-06
n/a
n/a
9.53E-07 -
1.89E-04
9. Demolition of XPS/EPS Foam Insulation in Residential, Public and Commercial Buildings, and Other Structures
Fenceline
5.91E-07 -
5.27E-04
n/a
n/a
n/a
n/a
Community
1.21E-08 -
1.08E-05
n/a
n/a
n/a
n/a
10. Recycling of EPS Foam
Fenceline
1.58E-05 -
1.24E-04
7.80E-06
- 7.20E-05
6.72E-06 -
2.14E-05
Community
5.08E-07 -
2.54E-06
4.25E-07
- 2.14E-06
3.78E-06 -
4.50E-06
11. Formulation of Solder
Fenceline
4.27E-04 -
4.99E-04
4.77E-03
- 6.88E-03
n/a
n/a
Community
1.65E-05 -
1.65E-05
2.48E-04
- 2.49E-04
n/a
n/a
Page 615 of 723
-------
Scenario Name
Range of Estimate Soil Concentration (jig/kg)
Fugitive
Stack
Incineration
Min - Max
Min - Max
Min - Max
12. Use of Solder
Fenceline
Community
n/a - n/a
n/a - n/a
n/a - n/a
n/a - n/a
6.42E-06 - 2.37E-05
4.51E-06 - 5.09E-06
Page 616 of 723
-------
Appendix G GENERAL POPULATION, HIGHLY EXPOSED
AND CONSUMER EXPOSURES
G.l Exposure Factors for General Population, Highly Exposed,
and Consumer Exposure Calculations
Exposure factors in this section were applied to all general population, highly exposed, and
consumer scenarios, as applicable.
Table Apx G-l. Body Weight by Age Group.
Age group
Mean body weight
(kg)1
Infant (<1 year)*
7.83
Young Toddler (l-<2 years)
11.4
Toddler (2-<3 years)
13.8
Small Child (3-<6 years)
18.6
Child (6-
-------
TableApx G-3. High-end (95th Percentile) Dietary Ingestion Rates (Consumers Only) by
Age Group- Fruit, Vegetables, Grains, Meats, Dairy, Fats
Age Group
95th Percentile Ingestion Rates (g/kg-day) Consumers Only
Fruits1
Vegetables1
Grains2
Meat3
Dairy3
Fats3
Infant (<1 year)
27.2
18.70
8.70
8.90
64.2
8.91*
Young Toddler (l-<2 years)
24.00
16.30
12.70
9.60
100.5
7.10
Toddler (2-<3 years)
20.50
14.00
11.70
9.60
78.7
6.40
Small Child (3-<6 years)
16.40
13.30
10.50
9.00
51.1
5.8
Child (6-
-------
Age group
Breast Milk Ingestion
(mL/kg day)1
Breast Milk Lipids
Ingestion (g/kg day)2
Mean
Upper
Mean
Upper
6 to <12 month
3.3
5.2
3.4
5.4
Birth to <1 year
4.1
6.2
4.2
6.4
1 U.S. EPA. Exposure Factors Handbook (U.S. EPA 2011b). Table 15-1.
2 Converted using 1.03 g/inL density of human breastmilk.
Table Apx G-6. Inhalation Rate by Age Group
Age group
Mean (mJ/day)'
95th (mJ/day)'
Infant (<1 year)
5.4
9.2
Young Toddler (l-<2 years)
8
12.8
Toddler (2-<3 years)
8.9
13.7
Small Child (3-<6 years)
10.1
13.8
Child (6-
-------
Time Awake Spent (hr/day)1
Fraction Awake Time Spent
Unitless)
Part-
Full-
Part-
Full-
Microenvironment
SAH
Time
Time
SAH
Time
Time
Adult /
School/
School/
Adult /
School/
School/
Child
COF/
COF/
Child
COF/
COF/
Work
Work
Work
Work
Total
13
13
13
SAH = Stay at home
COF = Child-occupied facility
1 CHAD Database (U.S. EPA 2009b)
Table Apx G-9. Generic Activity Patterns for Time Spent in a Day (24 hours)
Microenvironment
Time Spent Total (hr/day)1
Fraction Time Spent Total
(unitless)
SAH
Adult /
Child
Part-
Time
School/
COF/
Work
Full-
Time
School/
COF/
Work
SAH
Adult /
Child
Part-
Time
School/
COF/
Work
Full-
Time
School /
COF/
Work
Public and Commercial Buildings
1
3
6
0.04
0.13
0.25
Outside
2
2
2
0.08
0.08
0.08
Automobile
1
2
2
0.04
0.08
0.08
Residences
20
17
14
0.83
0.71
0.58
Total
24
24
24
SAH = Stay at home
COF = Child-occupied facility
1 CHAD Database (U.S. EPA 2009b)
Table Apx G-10. Surface Area to Body Weight Ratios (cm2/kg) By Age Group a
Age Group
Surface Area to Body Weight Ratios (cm2/kg)
Hands, 45% of Legs, 50%
of Arms
Hands
Infant (<1 year)
110.6
27.2
Young Toddler (l-<2 years)
104.7
26.3
Toddler (2-<3 years)
102.4
20.3
Small Child (3-<6 years)
95.6
19.9
Child (6-
-------
Table Apx G-ll. Dermal Adherence Factors for Dust By Age Group
Age Group
Dust Adherence Factor By Body
Part (mg/cm2)1
Weighting for Exposed
Surface Area of Body (Hands,
45% of Legs, 50% of Arms)2
Activity
Grouping
Hands
Legs
Arms
Total
Surface Area
Exposed
(cm2)
Weighted Dust
Adherence
Factor (mg/cm2)
Infant (<1 year)
Residential,
indoors
0.011
0.0035
0.0041
0.086
0.006
Young Toddler (l-<2
years)
0.119
0.006
Toddler (2-<3 years)
0.141
0.005
Small Child (3-<6 years)
0.178
0.005
Child (6-
-------
Table Apx G-13. Surface Area of Object Mouthed (cm2)
Age Group
Surface Area of Object Mouthed (cm2)1
Central Tendency
High-End
Young Toddler (l-<2 years)
10
50
1 Series on testing and assessment No. 306. (OECD 2019)
Table Apx G-14. Hourly Mouthing Duration (min)
Age Group
Hourly Mouthing Duration (min)1
Central Tendency
(Mean)
High-End
(95th Percentile)
Infants (0-1 year)
7.1
13.1
Young Toddler (l-<2 years)
3
9.7
1 Values are for "Non-Pacifier" objects, ages 3 to 12 months (infants) and ages 12 to 24 months (young toddler) as
cited in Table 4-20 of U.S. EPA. Exposure Factors Handbook (U.S. EPA 2011b) and originating from (Greene
2002). Non-pacifier objects include all soft plastic items, anatomy, non-soft plastic items, and "other" items.
G.2 Scenario Gl: General Population
The tables in this section provide a detailed breakdown of the dietary doses by food group.
Table Apx G-15. Estimated Average Daily Dose (ADD) by Age Group for Diet
Age Group
Dietary ADR (mg/kg-day)
Fruits
Veggies
Grains
Meats
Dairy
Fats
Fish
Breast
milk
Total from
Diet
Infant (<1 year)
2.6E-07
1.1E-06
3.2E-07
3.4E-07
2.1E-06
8.0E-07
N/A
1.9E-05
2.4E-05
Young Toddler
(l-<2 years)
2.6E-07
1.1E-06
5.3E-07
4.6E-07
7.8E-06
7.0E-07
1.1E-07
N/A
1.1E-05
Toddler (2-<3
years)
2.0E-07
9.7E-07
5.3E-07
4.8E-07
5.8E-06
6.3E-07
8.7E-08
N/A
8.7E-06
Small Child (3-
<6 years)
1.5E-07
8.5E-07
4.9E-07
4.5E-07
3.6E-06
5.9E-07
7.5E-08
N/A
6.2E-06
Child (6-
-------
Table Apx G-16. Percent of Dietary ADD by Food Group
Age Group
Percent of Dietary ADD (mg/kg-day)
Fruits
Veggies
Grains
Meats
Dairy
Fats
Fish
Breast
milk
Total from
Diet
Infant (<1 year)
1.1%
4.5%
1.4%
1.4%
8.8%
3.4%
0.0%
79.4%
100%
Young Toddler
(l-<2 years)
2.4%
9.9%
4.8%
4.2%
71.4%
6.4%
1.0%
0.0%
100%
Toddler (2-<3
years)
2.3%
11.1%
6.1%
5.6%
66.6%
7.2%
1.0%
0.0%
100%
Small Child (3-
<6 years)
2.5%
13.7%
7.9%
7.2%
58.0%
9.5%
1.2%
0.0%
100%
Cliild (6-
-------
Table Apx G-18. Percent of Dietary ADR by Food Group
Age Group
Percent of Dietary ADR
Fruits
Veggies
Grains
Meats
Dairy
Fats
Fish
Breast
milk
Total from
Diet
Infant (<1 year)
1.8%
4.3%
1.2%
2.0%
19.3%
2.5%
0.0%
68.9%
100.0%
Young Toddler
(l-<2 years)
3.7%
8.6%
4.0%
4.9%
69.3%
4.6%
4.8%
0.0%
100.0%
Toddler (2-<3
years)
3.9%
9.1%
4.5%
6.0%
66.5%
5.1%
4.9%
0.0%
100.0%
Small Child (3-
<6 years)
4.2%
11.7%
5.5%
7.6%
58.6%
6.3%
6.0%
0.0%
100.0%
Child (6-
-------
Scenario
Label
Water Column Concentration (jig/L) - 21
day average - Dissolved
Fish Tissue Concentration (mg/kg)1
Harmonic Mean
Flow
(50th percentile)
Harmonic Mean
Flow
(10th percentile)
Harmonic Mean
Flow
(50th percentile)
Harmonic Mean Flow
(10th percentile)
*W2.8
W2.9
1.9E-03
1.3E-02
8.7E-02
6.2E-01
*W2.10
W2.ll
4.2E-03
3.0E-02
2.0E-01
1.4E+00
*W2.12
W3.1
1.3E-02
3.8E-01
6.2E-01
1.7E+01
W3.2
8.9E-04
2.6E-02
4.1E-02
1.2E+00
W3.3
3.2E-02
9.2E-01
1.5E+00
4.3E+01
W3.4
2.2E-03
6.4E-02
1.0E-01
3.0E+00
W3.5
1.3E-03
3.8E-02
6.2E-02
1.7E+00
W3.6
8.9E-05
2.6E-03
4.1E-03
1.2E-01
W3.7
3.2E-03
9.2E-02
1.5E-01
4.3E+00
W3.8
2.2E-04
6.4E-03
1.0E-02
3.0E-01
W3.9
6.0E-03
4.2E-02
2.8E-01
2.0E+00
W3.10
4.0E-04
2.9E-03
1.9E-02
1.4E-01
W3.ll
1.5E-02
1.0E-01
6.9E-01
4.8E+00
W3.12
1.0E-03
7.2E-03
4.6E-02
3.3E-01
W4.1
1.3E-02
3.6E-01
5.9E-01
1.7E+01
W4.2
1.1E-03
3.1E-02
4.9E-02
1.4E+00
W4.3
1.3E-03
3.6E-02
5.9E-02
1.7E+00
W4.4
1.1E-04
3.1E-03
4.9E-03
1.4E-01
W4.5
5.8E-03
4.0E-02
2.7E-01
1.9E+00
W4.6
4.9E-04
3.5E-03
2.3E-02
1.6E-01
W5.1
8.5E-01
2.5E+01
3.9E+01
1.2E+03
W5.2
8.5E-02
2.5E+00
3.9E+00
1.2E+02
W5.3
3.9E-01
2.8E+00
1.8E+01
1.3E+02
W5.4
6.8E-01
2.0E+01
3.2E+01
9.3E+02
W5.5
6.8E-02
2.0E+00
3.2E+00
9.3E+01
W5.6
3.1E-01
2.2E+00
1.4E+01
1.0E+02
W5.7
1.2E+00
3.4E+01
5.4E+01
1.6E+03
W5.8
1.2E-01
3.4E+00
5.4E+00
1.6E+02
W5.9
5.3E-01
3.8E+00
2.5E+01
1.8E+02
W5.10
9.3E-01
2.7E+01
4.3E+01
1.3E+03
W5.ll
9.3E-02
2.7E+00
4.3E+00
1.3E+02
W5.12
4.3E-01
3.1E+00
2.0E+01
1.4E+02
W6.1
3.9E-03
1.1E-01
1.8E-01
5.3E+00
W6.2
3.9E-04
1.1E-02
1.8E-02
5.3E-01
W6.3
1.8E-03
1.3E-02
8.3E-02
5.9E-01
W6.4
3.7E-03
1.1E-01
1.7E-01
5.1E+00
Page 625 of 723
-------
Scenario
Label
Water Column Concentration (jig/L) - 21
day average - Dissolved
Fish Tissue Concentration (mg/kg)1
Harmonic Mean
Flow
(50th percentile)
Harmonic Mean
Flow
(10th percentile)
Harmonic Mean
Flow
(50th percentile)
Harmonic Mean Flow
(10th percentile)
*W6.5
*W6.6
W6.7
1.7E-02
5.1E-01
8.1E-01
2.4E+01
W6.8
1.7E-03
5.1E-02
8.1E-02
2.4E+00
W6.9
7.9E-03
5.7E-02
3.7E-01
2.7E+00
W6.10
1.7E-02
4.9E-01
7.7E-01
2.3E+01
W6.ll
1.7E-03
4.9E-02
7.7E-02
2.3E+00
W6.12
7.6E-03
5.5E-02
3.6E-01
2.6E+00
W8.1
2.4E-05
2.3E-04
1.1E-03
1.1E-02
*W8.2
W8.3
2.8E-02
2.8E-01
1.3E-00
1.3E+01
W8.4
2.8E-03
2.8E-02
1.32E+01
1.29E+00
*W9.1
*W9.2
W9.3
1.91E-01
1.9E+00
8.9E+00
8.6E+01
W9.4
1.91E-02
1.9E-01
8.9E-01
8.6E+00
W10.1
1.8E-02
5.2E-01
8.4E-01
2.4E+01
W10.2
1.8E-03
5.2E-02
8.4E-02
2.4E+00
W10.3
8.3E-03
5.8E-02
3.8E-01
2.7E+00
W10.4
9.1E-04
2.7E-02
4.2E-02
1.2E+00
*W10.5
*W10.6
W10.7
2.2E-02
6.2E-01
1.0E+00
2.9E+01
W10.8
2.2E-03
6.2E-02
1.0E-01
2.9E+00
W10.9
9.8E-03
6.9E-02
4.6E-01
3.2E+00
W10.10
1.1E-03
3.2E-02
5.0E-02
1.5E+00
*W10.11
*W10.12
W12.1
6.8E-05
1.9E-03
3.2E-03
9.0E-02
W12.2
3.1E-04
2.2E-03
1.4E-02
1.0E-01
*W12.3
*W12.4
W12.5
1.4E-04
3.9E-03
6.3E-03
1.8E-01
W12.6
6.2E-04
4.4E-03
2.9E-02
2.0E-01
W12.7
W12.8
* Scenario was not ran in the second tier model (WWM-PSC) because risks were not of concern using the first tier
model (E-FAST).
1 Fish Tissue Concentration (mg/kg) = Surface Water Concentration (|ig/L) * Wet Weight BAF (46,488 L/kg) *
Conversion Factor (0.001 mg/|ig)
Page 626 of 723
-------
Table Apx G-20. Highly Exposed Acute Dose
Rate and Average Daily Doses (mg/kg/day) for Mode
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
Wl.l
4.76E-03
2.1E-05
3.9E-03
1.7E-05
3.6E-03
1.5E-05
2.8E-03
1.4E-05
1.7E-03
7.6E-06
3.2E-03
5.5E-06
1.5E-05
W1.2
3.9E-03
1.7E-05
3.2E-03
1.4E-05
3.0E-03
1.2E-05
2.3E-03
1.1E-05
1.4E-03
6.3E-06
2.6E-03
4.5E-06
1.2E-05
W1.3
2.4E-02
1.0E-04
2.0E-02
8.6E-05
1.8E-02
7.5E-05
1.4E-02
6.9E-05
8.5E-03
3.8E-05
1.6E-02
2.8E-05
7.6E-05
W1.4
1.9E-02
8.6E-05
1.6E-02
7.1E-05
1.4E-02
6.1E-05
1.1E-02
5.6E-05
6.8E-03
3.2E-05
1.3E-02
2.3E-05
6.3E-05
W1.5
5.4E-03
9.5E-05
4.4E-03
7.9E-05
4.0E-03
6.8E-05
3.1E-03
6.3E-05
1.9E-03
3.5E-05
3.6E-03
2.5E-05
7.0E-05
W1.6
4.4E-03
7.8E-05
3.6E-03
6.4E-05
3.3E-03
5.6E-05
2.6E-03
5.1E-05
1.6E-03
2.9E-05
3.0E-03
2.1E-05
5.7E-05
W1.7
2.7E-02
4.8E-04
2.2E-02
3.9E-04
2.0E-02
3.4E-04
1.6E-02
3.1E-04
9.5E-03
1.8E-04
1.8E-02
1.3E-04
3.5E-04
W1.8
2.2E-02
3.9E-04
1.8E-02
3.2E-04
1.7E-02
2.8E-04
1.3E-02
2.6E-04
7.9E-03
1.4E-04
1.5E-02
1.0E-04
2.9E-04
W2.1
2.3E-03
1.0E-05
1.9E-03
8.3E-06
1.7E-03
7.2E-06
1.3E-03
6.6E-06
8.0E-04
3.7E-06
1.5E-03
2.7E-06
7.3E-06
W2.2
1.5E-03
6.5E-06
1.2E-03
5.4E-06
1.1E-03
4.6E-06
8.8E-04
4.3E-06
5.3E-04
2.4E-06
1.0E-03
1.7E-06
4.7E-06
W2.3
5.1E-03
2.2E-05
4.2E-03
1.9E-05
3.9E-03
1.6E-05
3.0E-03
1.5E-05
1.8E-03
8.2E-06
3.4E-03
5.9E-06
1.6E-05
W2.4
3.4E-03
1.5E-05
2.8E-03
1.2E-05
2.6E-03
1.1E-05
2.0E-03
9.8E-06
1.2E-03
5.5E-06
2.3E-03
4.0E-06
1.1E-05
W2.5
2.3E-04
1.0E-06
1.9E-04
8.3E-07
1.7E-04
7.2E-07
1.3E-04
6.6E-07
8.0E-05
3.7E-07
1.5E-04
2.7E-07
7.3E-07
*W2.6
W2.7
5.09E-04
2.24E-06
4.21E-04
1.85E-06
3.85E-04
1.60E-06
2.99E-04
1.47E-06
1.80E-04
8.25E-07
3.42E-04
5.95E-07
1.64E-06
*W2.8
W2.9
2.54E-04
4.57E-06
2.10E-04
3.78E-06
1.92E-04
3.27E-06
1.49E-04
3.01E-06
9.00E-05
1.68E-06
1.71E-04
1.21E-06
3.34E-06
*W2.10
W2.ll
5.72E-04
1.03E-05
4.72E-04
8.49E-06
4.32E-04
7.35E-06
3.36E-04
6.76E-06
2.03E-04
3.78E-06
3.84E-04
2.73E-06
7.50E-06
*W2.12
W3.1
7.21E-03
3.24E-05
5.96E-03
2.68E-05
5.45E-03
2.32E-05
4.23E-03
2.13E-05
2.56E-03
1.19E-05
4.85E-03
8.60E-06
2.37E-05
W3.2
4.98E-04
2.17E-06
4.11E-04
1.79E-06
3.76E-04
1.55E-06
2.92E-04
1.42E-06
1.76E-04
7.97E-07
3.34E-04
5.75E-07
1.58E-06
W3.3
1.77E-02
7.93E-05
1.46E-02
6.55E-05
1.34E-02
5.67E-05
1.04E-02
5.21E-05
6.27E-03
2.92E-05
1.19E-02
2.10E-05
5.79E-05
W3.4
1.23E-03
5.33E-06
1.01E-03
4.41E-06
9.28E-04
3.81E-06
7.21E-04
3.51E-06
4.35E-04
1.96E-06
8.25E-04
1.42E-06
3.89E-06
W3.5
7.21E-04
3.24E-06
5.96E-04
2.68E-06
5.45E-04
2.32E-06
4.23E-04
2.13E-06
2.56E-04
1.19E-06
4.85E-04
8.60E-07
2.37E-06
W3.6
4.98E-05
2.17E-07
4.11E-05
1.79E-07
3.76E-05
1.55E-07
2.92E-05
1.42E-07
1.76E-05
7.97E-08
3.34E-05
5.75E-08
1.58E-07
ed Fish Ingestion Only
Page 627 of 723
-------
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
W3.7
1.77E-03
7.93E-06
1.46E-03
6.55E-06
1.34E-03
5.67E-06
1.04E-03
5.21E-06
6.27E-04
2.92E-06
1.19E-03
2.10E-06
5.79E-06
W3.8
1.23E-04
5.33E-07
1.01E-04
4.41E-07
9.28E-05
3.81E-07
7.21E-05
3.51E-07
4.35E-05
1.96E-07
8.25E-05
1.42E-07
3.89E-07
W3.9
8.07E-04
1.48E-05
6.66E-04
1.22E-05
6.10E-04
1.06E-05
4.74E-04
9.71E-06
2.86E-04
5.44E-06
5.42E-04
3.92E-06
1.08E-05
W3.10
5.59E-05
9.91E-07
4.61E-05
8.18E-07
4.22E-05
7.08E-07
3.28E-05
6.51E-07
1.98E-05
3.65E-07
3.75E-05
2.63E-07
7.23E-07
W3.ll
1.97E-03
3.61E-05
1.63E-03
2.98E-05
1.49E-03
2.58E-05
1.16E-03
2.37E-05
6.99E-04
1.33E-05
1.33E-03
9.59E-06
2.64E-05
W3.12
1.38E-04
2.44E-06
1.14E-04
2.02E-06
1.04E-04
1.75E-06
8.09E-05
1.61E-06
4.88E-05
8.99E-07
9.26E-05
6.49E-07
1.78E-06
W4.1
6.88E-03
3.09E-05
5.68E-03
2.55E-05
5.20E-03
2.21E-05
4.04E-03
2.03E-05
2.44E-03
1.14E-05
4.62E-03
8.21E-06
2.26E-05
W4.2
5.92E-04
2.59E-06
4.89E-04
2.14E-06
4.48E-04
1.85E-06
3.48E-04
1.70E-06
2.10E-04
9.54E-07
3.98E-04
6.88E-07
1.89E-06
W4.3
6.88E-04
3.09E-06
5.68E-04
2.55E-06
5.20E-04
2.21E-06
4.04E-04
2.03E-06
2.44E-04
1.14E-06
4.62E-04
8.21E-07
2.26E-06
W4.4
5.92E-05
2.59E-07
4.89E-05
2.14E-07
4.48E-05
1.85E-07
3.48E-05
1.70E-07
2.10E-05
9.54E-08
3.98E-05
6.88E-08
1.89E-07
W4.5
7.69E-04
1.41E-05
6.35E-04
1.16E-05
5.82E-04
1.01E-05
4.52E-04
9.25E-06
2.73E-04
5.18E-06
5.17E-04
3.74E-06
1.03E-05
W4.6
6.63E-05
1.19E-06
5.48E-05
9.82E-07
5.01E-05
8.50E-07
3.89E-05
7.81E-07
2.35E-05
4.37E-07
4.46E-05
3.16E-07
8.68E-07
W5.1
4.76E-01
2.07E-03
3.93E-01
1.71E-03
3.60E-01
1.48E-03
2.79E-01
1.36E-03
1.69E-01
7.63E-04
3.20E-01
5.51E-04
1.51E-03
W5.2
4.76E-02
2.07E-04
3.93E-02
1.71E-04
3.60E-02
1.48E-04
2.79E-02
1.36E-04
1.69E-02
7.63E-05
3.20E-02
5.51E-05
1.51E-04
W5.3
5.34E-02
9.50E-04
4.41E-02
7.85E-04
4.04E-02
6.79E-04
3.14E-02
6.24E-04
1.89E-02
3.50E-04
3.59E-02
2.52E-04
6.93E-04
W5.4
3.83E-01
1.66E-03
3.16E-01
1.37E-03
2.90E-01
1.19E-03
2.25E-01
1.09E-03
1.36E-01
6.11E-04
2.57E-01
4.41E-04
1.21E-03
W5.5
3.83E-02
1.66E-04
3.16E-02
1.37E-04
2.90E-02
1.19E-04
2.25E-02
1.09E-04
1.36E-02
6.11E-05
2.57E-02
4.41E-05
1.21E-04
W5.6
4.29E-02
7.61E-04
3.55E-02
6.29E-04
3.25E-02
5.44E-04
2.52E-02
5.00E-04
1.52E-02
2.80E-04
2.89E-02
2.02E-04
5.56E-04
W5.7
6.50E-01
2.83E-03
5.37E-01
2.34E-03
4.92E-01
2.03E-03
3.82E-01
1.86E-03
2.30E-01
1.04E-03
4.37E-01
7.52E-04
2.07E-03
W5.8
6.50E-02
2.83E-04
5.37E-02
2.34E-04
4.92E-02
2.03E-04
3.82E-02
1.86E-04
2.30E-02
1.04E-04
4.37E-02
7.52E-05
2.07E-04
W5.9
7.30E-02
1.30E-03
6.03E-02
1.07E-03
5.52E-02
9.28E-04
4.29E-02
8.53E-04
2.59E-02
4.78E-04
4.90E-02
3.45E-04
9.48E-04
W5.10
5.22E-01
2.28E-03
4.31E-01
1.88E-03
3.95E-01
1.63E-03
3.07E-01
1.50E-03
1.85E-01
8.38E-04
3.51E-01
6.05E-04
1.66E-03
W5.ll
5.22E-02
2.28E-04
4.31E-02
1.88E-04
3.95E-02
1.63E-04
3.07E-02
1.50E-04
1.85E-02
8.38E-05
3.51E-02
6.05E-05
1.66E-04
W5.12
5.88E-02
1.04E-03
4.85E-02
8.60E-04
4.44E-02
7.44E-04
3.45E-02
6.84E-04
2.08E-02
3.83E-04
3.95E-02
2.76E-04
7.60E-04
W6.1
2.19E-03
9.54E-06
1.81E-03
7.88E-06
1.66E-03
6.82E-06
1.29E-03
6.27E-06
7.76E-04
3.51E-06
1.47E-03
2.53E-06
6.96E-06
W6.2
2.19E-04
9.54E-07
1.81E-04
7.88E-07
1.66E-04
6.82E-07
1.29E-04
6.27E-07
7.76E-05
3.51E-07
1.47E-04
2.53E-07
6.96E-07
W6.3
2.45E-04
4.37E-06
2.03E-04
3.61E-06
1.85E-04
3.13E-06
1.44E-04
2.87E-06
8.69E-05
1.61E-06
1.65E-04
1.16E-06
3.19E-06
W6.4
2.10E-03
9.17E-06
1.74E-03
7.57E-06
1.59E-03
6.55E-06
1.24E-03
6.02E-06
7.46E-04
3.37E-06
1.41E-03
2.43E-06
6.69E-06
Page 628 of 723
-------
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
*W6.5
*W6.6
W6.7
9.76E-03
4.26E-05
8.07E-03
3.52E-05
7.38E-03
3.05E-05
5.73E-03
2.80E-05
3.46E-03
1.57E-05
6.56E-03
1.13E-05
3.11E-05
W6.8
9.76E-04
4.26E-06
8.07E-04
3.52E-06
7.38E-04
3.05E-06
5.73E-04
2.80E-06
3.46E-04
1.57E-06
6.56E-04
1.13E-06
3.11E-06
W6.9
1.10E-03
1.94E-05
9.06E-04
1.61E-05
8.29E-04
1.39E-05
6.44E-04
1.28E-05
3.89E-04
7.15E-06
7.37E-04
5.16E-06
1.42E-05
W6.10
9.39E-03
4.07E-05
7.75E-03
3.37E-05
7.10E-03
2.91E-05
5.51E-03
2.68E-05
3.33E-03
1.50E-05
6.31E-03
1.08E-05
2.97E-05
W6.ll
9.39E-04
4.07E-06
7.75E-04
3.37E-06
7.10E-04
2.91E-06
5.51E-04
2.68E-06
3.33E-04
1.50E-06
6.31E-04
1.08E-06
2.97E-06
W6.12
1.05E-03
1.87E-05
8.71E-04
1.55E-05
7.98E-04
1.34E-05
6.19E-04
1.23E-05
3.74E-04
6.88E-06
7.09E-04
4.96E-06
1.37E-05
*W8.1
*W8.2
W8.3
5.32E-03
6.93E-05
4.40E-03
5.72E-05
4.03E-03
4.95E-05
3.13E-03
4.55E-05
1.89E-03
2.55E-05
3.58E-03
1.84E-05
5.06E-05
W8.4
5.32E-04
6.93E-06
4.40E-04
5.72E-06
4.03E-04
4.95E-06
3.13E-04
4.55E-06
1.89E-04
2.55E-06
3.58E-04
1.84E-06
5.06E-06
*W9.1
*W9.2
W9.3
1.03E-01
4.67E-04
8.51E-02
3.86E-04
7.79E-02
3.34E-04
6.05E-02
3.07E-04
3.65E-02
1.72E-04
6.93E-02
1.24E-04
3.41E-04
W9.4
5.32E-04
4.67E-05
4.40E-04
3.86E-05
4.03E-04
3.34E-05
3.13E-04
3.07E-05
1.89E-04
1.72E-05
3.58E-04
1.24E-05
3.41E-05
W10.1
9.91E-03
4.44E-05
8.18E-03
3.67E-05
7.49E-03
3.18E-05
5.82E-03
2.92E-05
3.51E-03
1.64E-05
6.66E-03
1.18E-05
3.24E-05
W10.2
9.91E-04
4.44E-06
8.18E-04
3.67E-06
7.49E-04
3.18E-06
5.82E-04
2.92E-06
3.51E-04
1.64E-06
6.66E-04
1.18E-06
3.24E-06
W10.3
1.11E-03
2.02E-05
9.16E-04
1.67E-05
8.38E-04
1.44E-05
6.51E-04
1.33E-05
3.93E-04
7.43E-06
7.45E-04
5.36E-06
1.47E-05
W10.4
5.14E-04
2.22E-06
4.24E-04
1.84E-06
3.88E-04
1.59E-06
3.02E-04
1.46E-06
1.82E-04
8.18E-07
3.45E-04
5.90E-07
1.62E-06
*W10.5
*W10.6
W10.7
1.18E-02
5.30E-05
9.74E-03
4.38E-05
8.92E-03
3.79E-05
6.93E-03
3.48E-05
4.18E-03
1.95E-05
7.93E-03
1.41E-05
3.87E-05
W10.8
1.18E-03
5.30E-06
9.74E-04
4.38E-06
8.92E-04
3.79E-06
6.93E-04
3.48E-06
4.18E-04
1.95E-06
7.93E-04
1.41E-06
3.87E-06
W10.9
1.32E-03
2.41E-05
1.09E-03
1.99E-05
9.97E-04
1.72E-05
7.74E-04
1.58E-05
4.67E-04
8.86E-06
8.86E-04
6.39E-06
1.76E-05
W10.10
6.09E-04
2.65E-06
5.03E-04
2.19E-06
4.61E-04
1.89E-06
3.58E-04
1.74E-06
2.16E-04
9.74E-07
4.10E-04
7.03E-07
1.93E-06
*W10.11
*W10.12
Page 629 of 723
-------
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
W12.1
3.73E-05
1.66E-07
3.08E-05
1.38E-07
2.82E-05
1.19E-07
2.19E-05
1.09E-07
1.32E-05
6.13E-08
2.51E-05
4.42E-08
1.22E-07
W12.2
4.18E-05
7.61E-07
3.45E-05
6.29E-07
3.16E-05
5.44E-07
2.45E-05
5.00E-07
1.48E-05
2.80E-07
2.81E-05
2.02E-07
5.56E-07
*W12.3
*W12.4
W12.5
7.46E-05
3.33E-07
6.16E-05
2.75E-07
5.64E-05
2.38E-07
4.38E-05
2.19E-07
2.64E-05
1.23E-07
5.01E-05
8.85E-08
2.43E-07
W12.6
8.34E-05
1.52E-06
6.89E-05
1.26E-06
6.31E-05
1.09E-06
4.90E-05
1.00E-06
2.96E-05
5.60E-07
5.61E-05
4.04E-07
1.11E-06
*W12.7
*W12.8
All
Scenarios -
Minimum
3.73E-05
1.66E-07
3.08E-05
1.38E-07
2.82E-05
1.19E-07
2.19E-05
1.09E-07
1.32E-05
6.13E-08
2.51E-05
4.42E-08
1.22E-07
All
Scenarios -
Maximum
6.50E-01
2.83E-03
5.37E-01
2.34E-03
4.92E-01
2.03E-03
3.82E-01
1.86E-03
2.30E-01
1.04E-03
4.37E-01
7.52E-04
2.07E-03
ADR = acute dose rate; ADD = average daily dose; HE = high-end residency, CT = central tendency residency
Page 630 of 723
-------
TableApx G-21. Highly Exposed Aggregate Acute Dose Rate and Average Daily Doses (mg/kg/day) for Modeled Fish
ngestion and Background
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
Wl.l
4.79E-03
4.99E-05
3.95E-03
3.52E-05
3.61E-03
2.82E-05
2.80E-03
2.21E-05
1.69E-03
1.19E-05
3.21E-03
8.46E-06
1.81E-05
W1.2
3.96E-03
4.61E-05
3.27E-03
3.21E-05
2.99E-03
2.55E-05
2.32E-03
1.96E-05
1.40E-03
1.06E-05
2.65E-03
7.47E-06
1.54E-05
W1.3
2.40E-02
1.33E-04
1.98E-02
1.04E-04
1.81E-02
8.79E-05
1.41E-02
7.70E-05
8.49E-03
4.27E-05
1.61E-02
3.06E-05
7.91E-05
W1.4
1.92E-02
1.15E-04
1.58E-02
8.89E-05
1.45E-02
7.47E-05
1.13E-02
6.48E-05
6.79E-03
3.58E-05
1.29E-02
2.57E-05
6.55E-05
W1.5
5.38E-03
1.25E-04
4.44E-03
9.69E-05
4.06E-03
8.16E-05
3.15E-03
7.11E-05
1.90E-03
3.94E-05
3.61E-03
2.83E-05
7.26E-05
W1.6
4.44E-03
1.07E-04
3.66E-03
8.25E-05
3.35E-03
6.91E-05
2.60E-03
5.97E-05
1.57E-03
3.30E-05
2.97E-03
2.37E-05
5.99E-05
W1.7
2.69E-02
5.07E-04
2.22E-02
4.13E-04
2.03E-02
3.55E-04
1.58E-02
3.22E-04
9.52E-03
1.80E-04
1.81E-02
1.30E-04
3.52E-04
W1.8
2.22E-02
4.22E-04
1.84E-02
3.42E-04
1.68E-02
2.94E-04
1.30E-02
2.66E-04
7.87E-03
1.49E-04
1.49E-02
1.07E-04
2.90E-04
W2.1
2.29E-03
3.91E-05
1.89E-03
2.64E-05
1.73E-03
2.05E-05
1.34E-03
1.50E-05
8.06E-04
7.98E-06
1.53E-03
5.61E-06
1.03E-05
W2.2
1.52E-03
3.56E-05
1.25E-03
2.35E-05
1.14E-03
1.80E-05
8.86E-04
1.27E-05
5.34E-04
6.69E-06
1.01E-03
4.68E-06
7.70E-06
W2.3
5.12E-03
5.15E-05
4.22E-03
3.66E-05
3.86E-03
2.94E-05
3.00E-03
2.32E-05
1.81E-03
1.25E-05
3.43E-03
8.90E-06
1.93E-05
W2.4
3.47E-03
4.41E-05
2.86E-03
3.05E-05
2.61E-03
2.41E-05
2.03E-03
1.83E-05
1.22E-03
9.80E-06
2.32E-03
6.93E-06
1.39E-05
W2.5
2.55E-04
3.01E-05
2.05E-04
1.89E-05
1.85E-04
1.41E-05
1.41E-04
9.12E-06
8.45E-05
4.67E-06
1.55E-04
3.22E-06
3.69E-06
*W2.6
W2.7
5.38E-04
3.14E-05
4.39E-04
1.99E-05
3.99E-04
1.50E-05
3.08E-04
9.93E-06
1.85E-04
5.12E-06
3.46E-04
3.55E-06
4.59E-06
*W2.8
W2.9
2.83E-04
3.37E-05
2.28E-04
2.19E-05
2.05E-04
1.67E-05
1.58E-04
1.15E-05
9.43E-05
5.98E-06
1.74E-04
4.17E-06
6.29E-06
*W2.10
W2.ll
6.01E-04
3.94E-05
4.90E-04
2.66E-05
4.46E-04
2.07E-05
3.44E-04
1.52E-05
2.07E-04
8.08E-06
3.88E-04
5.68E-06
1.05E-05
*W2.12
W3.1
7.24E-03
6.15E-05
5.97E-03
4.49E-05
5.47E-03
3.66E-05
4.24E-03
2.98E-05
2.56E-03
1.62E-05
4.86E-03
1.16E-05
2.66E-05
W3.2
5.27E-04
3.13E-05
4.29E-04
1.99E-05
3.90E-04
1.49E-05
3.01E-04
9.89E-06
1.81E-04
5.09E-06
3.38E-04
3.53E-06
4.54E-06
W3.3
1.77E-02
1.08E-04
1.46E-02
8.36E-05
1.34E-02
7.01E-05
1.04E-02
6.06E-05
6.28E-03
3.35E-05
1.19E-02
2.40E-05
6.08E-05
W3.4
1.26E-03
3.45E-05
1.03E-03
2.25E-05
9.42E-04
1.72E-05
7.29E-04
1.20E-05
4.39E-04
6.26E-06
8.29E-04
4.37E-06
6.85E-06
W3.5
7.50E-04
3.24E-05
6.14E-04
2.08E-05
5.59E-04
1.57E-05
4.32E-04
1.06E-05
2.60E-04
5.49E-06
4.88E-04
3.82E-06
5.32E-06
W3.6
7.89E-05
2.94E-05
5.92E-05
1.83E-05
5.10E-05
1.35E-05
3.77E-05
8.60E-06
2.19E-05
4.38E-06
3.64E-05
3.01E-06
3.11E-06
Page 631 of 723
-------
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
W3.7
1.80E-03
3.71E-05
1.48E-03
2.46E-05
1.35E-03
1.91E-05
1.05E-03
1.37E-05
6.32E-04
7.21E-06
1.19E-03
5.06E-06
8.74E-06
W3.8
1.52E-04
2.97E-05
1.19E-04
1.85E-05
1.06E-04
1.38E-05
8.05E-05
8.81E-06
4.78E-05
4.49E-06
8.56E-05
3.10E-06
3.35E-06
W3.9
8.36E-04
4.39E-05
6.84E-04
3.03E-05
6.23E-04
2.40E-05
4.82E-04
1.82E-05
2.90E-04
9.73E-06
5.46E-04
6.88E-06
1.37E-05
W3.10
8.50E-05
3.01E-05
6.42E-05
1.89E-05
5.56E-05
1.41E-05
4.13E-05
9.11E-06
2.41E-05
4.66E-06
4.05E-05
3.22E-06
3.68E-06
W3.ll
2.00E-03
6.52E-05
1.65E-03
4.79E-05
1.51E-03
3.92E-05
1.17E-03
3.22E-05
7.04E-04
1.76E-05
1.33E-03
1.25E-05
2.93E-05
W3.12
1.67E-04
3.16E-05
1.32E-04
2.01E-05
1.18E-04
1.51E-05
8.94E-05
1.01E-05
5.31E-05
5.20E-06
9.57E-05
3.60E-06
4.74E-06
W4.1
6.91E-03
6.01E-05
5.70E-03
4.36E-05
5.21E-03
3.55E-05
4.05E-03
2.88E-05
2.44E-03
1.57E-05
4.63E-03
1.12E-05
2.55E-05
W4.2
6.21E-04
3.17E-05
5.07E-04
2.02E-05
4.61E-04
1.52E-05
3.56E-04
1.02E-05
2.14E-04
5.25E-06
4.01E-04
3.64E-06
4.85E-06
W4.3
7.17E-04
3.22E-05
5.86E-04
2.06E-05
5.33E-04
1.56E-05
4.12E-04
1.05E-05
2.48E-04
5.44E-06
4.66E-04
3.78E-06
5.21E-06
W4.4
8.83E-05
2.94E-05
6.70E-05
1.83E-05
5.82E-05
1.36E-05
4.32E-05
8.63E-06
2.53E-05
4.39E-06
4.28E-05
3.02E-06
3.15E-06
W4.5
7.98E-04
4.32E-05
6.53E-04
2.97E-05
5.95E-04
2.35E-05
4.60E-04
1.77E-05
2.77E-04
9.48E-06
5.20E-04
6.69E-06
1.32E-05
W4.6
9.54E-05
3.03E-05
7.29E-05
1.91E-05
6.35E-05
1.42E-05
4.74E-05
9.24E-06
2.78E-05
4.73E-06
4.76E-05
3.27E-06
3.82E-06
W5.1
4.76E-01
2.10E-03
3.93E-01
1.73E-03
3.60E-01
1.50E-03
2.79E-01
1.37E-03
1.69E-01
7.67E-04
3.20E-01
5.54E-04
1.52E-03
W5.2
4.76E-02
2.37E-04
3.93E-02
1.89E-04
3.60E-02
1.62E-04
2.80E-02
1.45E-04
1.69E-02
8.06E-05
3.20E-02
5.80E-05
1.54E-04
W5.3
5.34E-02
9.79E-04
4.41E-02
8.03E-04
4.04E-02
6.93E-04
3.14E-02
6.33E-04
1.89E-02
3.54E-04
3.59E-02
2.55E-04
6.96E-04
W5.4
3.83E-01
1.69E-03
3.16E-01
1.39E-03
2.90E-01
1.20E-03
2.25E-01
1.10E-03
1.36E-01
6.16E-04
2.58E-01
4.44E-04
1.22E-03
W5.5
3.83E-02
1.95E-04
3.17E-02
1.55E-04
2.90E-02
1.32E-04
2.25E-02
1.18E-04
1.36E-02
6.54E-05
2.58E-02
4.70E-05
1.24E-04
W5.6
4.30E-02
7.90E-04
3.55E-02
6.47E-04
3.25E-02
5.58E-04
2.52E-02
5.09E-04
1.52E-02
2.84E-04
2.89E-02
2.05E-04
5.59E-04
W5.7
6.50E-01
2.86E-03
5.37E-01
2.36E-03
4.92E-01
2.04E-03
3.82E-01
1.87E-03
2.30E-01
1.05E-03
4.37E-01
7.55E-04
2.07E-03
W5.8
6.50E-02
3.12E-04
5.37E-02
2.52E-04
4.92E-02
2.16E-04
3.82E-02
1.95E-04
2.30E-02
1.09E-04
4.37E-02
7.82E-05
2.10E-04
W5.9
7.30E-02
1.33E-03
6.03E-02
1.09E-03
5.52E-02
9.42E-04
4.29E-02
8.62E-04
2.59E-02
4.82E-04
4.91E-02
3.48E-04
9.51E-04
W5.10
5.22E-01
2.31E-03
4.31E-01
1.90E-03
3.95E-01
1.64E-03
3.07E-01
1.51E-03
1.85E-01
8.42E-04
3.52E-01
6.08E-04
1.67E-03
W5.ll
5.23E-02
2.57E-04
4.32E-02
2.06E-04
3.95E-02
1.76E-04
3.07E-02
1.58E-04
1.85E-02
8.81E-05
3.52E-02
6.34E-05
1.69E-04
W5.12
5.88E-02
1.07E-03
4.86E-02
8.78E-04
4.44E-02
7.58E-04
3.45E-02
6.92E-04
2.08E-02
3.87E-04
3.95E-02
2.79E-04
7.63E-04
W6.1
2.22E-03
3.87E-05
1.83E-03
2.60E-05
1.67E-03
2.02E-05
1.29E-03
1.47E-05
7.81E-04
7.81E-06
1.48E-03
5.49E-06
9.92E-06
W6.2
2.48E-04
3.01E-05
1.99E-04
1.89E-05
1.79E-04
1.41E-05
1.37E-04
9.09E-06
8.19E-05
4.65E-06
1.50E-04
3.21E-06
3.65E-06
W6.3
2.74E-04
3.35E-05
2.21E-04
2.17E-05
1.99E-04
1.65E-05
1.52E-04
1.13E-05
9.12E-05
5.91E-06
1.68E-04
4.12E-06
6.15E-06
W6.4
2.13E-03
3.83E-05
1.76E-03
2.57E-05
1.60E-03
1.99E-05
1.24E-03
1.45E-05
7.50E-04
7.67E-06
1.42E-03
5.39E-06
9.65E-06
Page 632 of 723
-------
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
*W6.5
*W6.6
W6.7
9.79E-03
7.17E-05
8.08E-03
5.33E-05
7.40E-03
4.38E-05
5.74E-03
3.65E-05
3.46E-03
2.00E-05
6.57E-03
1.43E-05
3.40E-05
W6.8
1.01E-03
3.34E-05
8.25E-04
2.16E-05
7.52E-04
1.64E-05
5.82E-04
1.13E-05
3.50E-04
5.86E-06
6.60E-04
4.09E-06
6.07E-06
W6.9
1.13E-03
4.86E-05
9.24E-04
3.42E-05
8.43E-04
2.73E-05
6.53E-04
2.12E-05
3.93E-04
1.15E-05
7.41E-04
8.12E-06
1.72E-05
W6.10
9.41E-03
6.99E-05
7.77E-03
5.17E-05
7.11E-03
4.25E-05
5.52E-03
3.52E-05
3.33E-03
1.93E-05
6.32E-03
1.38E-05
3.27E-05
W6.ll
9.68E-04
3.32E-05
7.93E-04
2.15E-05
7.23E-04
1.63E-05
5.60E-04
1.11E-05
3.37E-04
5.80E-06
6.35E-04
4.04E-06
5.93E-06
W6.12
1.08E-03
4.78E-05
8.89E-04
3.35E-05
8.11E-04
2.68E-05
6.28E-04
2.08E-05
3.78E-04
1.12E-05
7.13E-04
7.92E-06
1.66E-05
W8.1
*W8.2
W8.3
5.35E-03
9.84E-05
4.42E-03
7.53E-05
4.04E-03
6.29E-05
3.14E-03
5.40E-05
1.89E-03
2.98E-05
3.59E-03
2.13E-05
5.35E-05
*W8.4
5.62E-04
3.61E-05
4.58E-04
2.38E-05
4.16E-04
1.83E-05
3.21E-04
1.30E-05
1.93E-04
6.85E-06
3.61E-04
4.79E-06
8.01E-06
W9.1
W9.2
W9.3
3.56E-02
4.96E-04
2.94E-02
4.04E-04
2.69E-02
3.47E-04
2.09E-02
3.15E-04
1.26E-02
1.76E-04
2.4E-02
1.27E-04
3.44E-04
W9.4
3.58E-03
7.58E-05
2.95E-03
5.66E-05
2.70E-03
4.68E-05
2.10E-03
3.91E-05
1.26E-03
2.15E-05
2.4E-03
1.53E-05
3.70E-05
W10.1
9.94E-03
7.36E-05
8.20E-03
5.48E-05
7.51E-03
4.52E-05
5.83E-03
3.77E-05
3.52E-03
2.07E-05
6.67E-03
1.48E-05
3.54E-05
W10.2
1.02E-03
3.36E-05
8.37E-04
2.18E-05
7.63E-04
1.66E-05
5.90E-04
1.14E-05
3.55E-04
5.93E-06
6.70E-04
4.14E-06
6.20E-06
W10.3
1.14E-03
4.93E-05
9.34E-04
3.48E-05
8.52E-04
2.78E-05
6.59E-04
2.17E-05
3.97E-04
1.17E-05
7.49E-04
8.31E-06
1.77E-05
W10.4
5.43E-04
3.14E-05
4.42E-04
1.99E-05
4.02E-04
1.50E-05
3.10E-04
9.92E-06
1.86E-04
5.11E-06
3.49E-04
3.55E-06
4.58E-06
*W10.5
*W10.6
W10.7
1.18E-02
8.21E-05
9.76E-03
6.18E-05
8.93E-03
5.13E-05
6.94E-03
4.33E-05
4.18E-03
2.38E-05
7.94E-03
1.70E-05
4.16E-05
W10.8
1.21E-03
3.44E-05
9.92E-04
2.25E-05
9.05E-04
1.72E-05
7.01E-04
1.19E-05
4.22E-04
6.25E-06
7.97E-04
4.36E-06
6.82E-06
W10.9
1.35E-03
5.32E-05
1.11E-03
3.80E-05
1.01E-03
3.06E-05
7.83E-04
2.43E-05
4.72E-04
1.32E-05
8.91E-04
9.35E-06
2.05E-05
W10.10
6.38E-04
3.18E-05
5.21E-04
2.03E-05
4.74E-04
1.53E-05
3.66E-04
1.02E-05
2.20E-04
5.27E-06
4.13E-04
3.66E-06
4.89E-06
*W10.11
*W10.12
Page 633 of 723
-------
Scenario
Label
Young Toddler
Doses (mg/kg/day)
Toddler Doses
(mg/kg/day)
Small Child Doses
(mg/kg/day)
Child Doses
(mg/kg/day)
Teenager Doses
(mg/kg/day)
Adult Doses (mg/kg/day)
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD
ADR
ADD -CT
ADD -HE
W12.1
6.64E-05
2.93E-05
4.89E-05
1.82E-05
4.16E-05
1.35E-05
3.04E-05
8.57E-06
1.75E-05
4.36E-06
2.80E-05
3.00E-06
3.08E-06
W12.2
7.09E-05
2.99E-05
5.26E-05
1.87E-05
4.50E-05
1.39E-05
3.30E-05
8.96E-06
1.91E-05
4.58E-06
3.11E-05
3.16E-06
3.51E-06
*W12.3
*W12.4
W12.5
1.04E-04
2.95E-05
7.97E-05
1.84E-05
6.98E-05
1.36E-05
5.23E-05
8.68E-06
3.07E-05
4.42E-06
5.31E-05
3.04E-06
3.20E-06
W12.6
1.13E-04
3.07E-05
8.70E-05
1.93E-05
7.65E-05
1.45E-05
5.75E-05
9.46E-06
3.39E-05
4.86E-06
5.91E-05
3.36E-06
4.07E-06
*W12.7
*W12.8
All
Scenarios -
Minimum
6.64E-05
2.93E-05
4.89E-05
1.82E-05
4.16E-05
1.35E-05
3.04E-05
8.57E-06
1.75E-05
4.36E-06
2.80E-05
3.00E-06
3.08E-06
All
Scenarios -
Maximum
6.50E-01
2.86E-03
5.37E-01
2.36E-03
4.92E-01
2.04E-03
3.82E-01
1.87E-03
2.30E-01
1.05E-03
4.37E-01
7.55E-04
2.07E-03
ADR = acute dose rate; ADD = average daily dose; HE = high-end residency, CT = central tendency residency
Page 634 of 723
-------
G.4 Scenario H2: Near Facility Suspended Particulates in Air —
Inhalation
EPA/OPPT's Integrated Indoor-Outdoor Air Calculation (IIOAC) was used to estimate ambient air
concentrations for highly exposed groups living near facilities. IIOAC is based on a set of pre-run
AERMOD dispersion scenarios at a variety of meteorological and land-use settings. For the source types
of interest in HBCD modeling, users are required to enter: (1) emission parameters - emission source type,
number of emission scenarios, number of releases per scenario, mass released per day, release duration,
number of release days, and release pattern; (2) system parameters - applicable only for fugitive sources
where an area must be specified; and (3) location parameters - urban or rural setting, particle size/vapor,
and climate region. IIOAC outputs of daily-averaged air concentration, annual-averaged air concentration,
and doses are provided as central tendency and high-end estimates at two distances: fenceline (100 m from
source) and community (averaged across 100 to 1,000 m from the source).
IIOAC calculates ambient air concentration based on the release duration and number of days of release
per year entered by the user (e.g., release occurs 4 hrs/day for 52 days in a year). An adjusted emission
rate is first calculated, as shown in EquationApx G-l, to take into account the release duration and
convert the user-defined mass released per day into g/s.
Equation Apx G-l
ERadJ = ™- 0-2778
where ERadj = adjusted emission rate [g/s]
ER = user-defined mass released per day [kg/day]
h = emission duration [hrs/day]
0.2778 = conver si on factor from kg/hr to g/s
Air concentrations are calculated in Equation Apx G-2 by scaling the post-processed AERMOD result,
obtained based on an emission of 1 g/s, by the adjusted emission rate. For fugitive sources, scaling by
just the adjusted emission rate gives an air concentration corresponding to an area size of 100 m2, the
same as that used in the AERMOD runs. To account for a different area size, an area size scaling factor,
SFj, is applied.
Equation Apx G-2
ER w ¦
Coutdoor = —V"' SFj 1 Postprocessed AERMOD result
1Q / s
where Coutdoor = outdoor air concentration [|ig/m3]
ERadj = adjusted emission rate [g/s]
SFj = scaling factor for fugitive area size j [-]; set to 1 for point sources
For point and fugitive sources, three particle size scenarios are available:
a. Fine particles (with a mass-mean aerodynamic diameter of 2.5 |iin),
b. Coarse particles (with a mass-mean aerodynamic diameter of 10 (j,m), and
c. Vapor (no particles).
Page 635 of 723
-------
All calculated air concentrations of fine and coarse particles are capped by an upper limit equal to the
National Ambient Air Quality Standards (NAAQS) for particulate matter (PM) (U.S. EPA 2009a). These
limits are 35 and 150 [j,m/m3 for fine and coarse particles (i.e., the NAAQS for PM2.5 and PM10),
respectively, over 24-hr. For vapors, the chemical is released in gaseous form and therefore there is no
transfer from one phase to another. IIOAC currently does not set an upper limit for point and fugitive
sources in vapor form. Air concentrations are then calculated by multiplying the ambient air concentration
by an indoor-outdoor ratio.
In modeling ambient air concentration for highly exposed groups living near facilities, twelve emission
scenarios were considered, based on the conditions of use defined in the Section 1.4. For scenarios with
site-specific information, this information was used in the IIOAC model runs. When site-specific
information was not unknown, the following default parameters were used:
i. Emission parameters:
a. Source type: Both stack and fugitive.
b. Emission duration: 24 hours.
c. Release pattern: Conservative pattern of release was used for all runs.
ii. System parameters:
a. Fugitive source area: 100 m2
iii. Location parameters:
a. Population setting: Rural
b. Particle size: Coarse - In the United States, standard grade HBCD powder is defined as a
mean particle size of 20 to 150 |im; therefore, coarse particles was selected for use in the
IIOAC runs.
c. Climate region default: Three regions were used:
i. West north central to obtain central tendency estimates for both air concentration
and particle deposition.
ii. South (coastal) to obtain high-end estimates when considering only air
concentration.
iii. East north central to obtain high-end estimates when considering both air
concentration and particle deposition.
Table Apx G-22. Highly Exposed Acute Dose Rate (mg/kg/day) for Modeled Air Only
Label
Source Type
Average Air
Concentration (jig/m3)
ADR - Modeled Air
(mg/kg/d)
Daily
Annual
Infant
Young
Toddler
Toddler
Small
Child
Child
Teen
Adult
1.1
Fugitive
1.17E+00
8.74E-04
1.37E-03
1.31E-03
1.16E-03
8.68E-04
6.10E-04
4.51E-04
3.08E-04
1.2
Fugitive
6.72E-02
8.82E-04
7.90E-05
7.55E-05
6.67E-05
4.99E-05
3.51E-05
2.59E-05
1.77E-05
1.3
Fugitive
5.85E+00
4.37E-03
6.87E-03
6.57E-03
5.81E-03
4.34E-03
3.05E-03
2.25E-03
1.54E-03
1.4
Fugitive
3.36E-01
4.41E-03
3.95E-04
3.77E-04
3.34E-04
2.49E-04
1.75E-04
1.30E-04
8.86E-05
1.5
Stack
1.71E-01
6.67E-04
2.01E-04
1.92E-04
1.70E-04
1.27E-04
8.92E-05
6.59E-05
4.50E-05
1.6
Stack
1.17E-02
6.72E-04
1.38E-05
1.32E-05
1.16E-05
8.70E-06
6.12E-06
4.52E-06
3.09E-06
1.7
Stack
8.54E-01
3.33E-03
1.00E-03
9.59E-04
8.48E-04
6.34E-04
4.46E-04
3.29E-04
2.25E-04
1.8
Stack
5.86E-02
3.36E-03
6.89E-05
6.58E-05
5.82E-05
4.35E-05
3.06E-05
2.26E-05
1.55E-05
1.9
Incineration
6.31E-03
2.56E-04
7.42E-06
7.09E-06
6.27E-06
4.68E-06
3.30E-06
2.43E-06
1.67E-06
1.10
Incineration
3.28E-04
2.56E-04
3.85E-07
3.68E-07
3.25E-07
2.43E-07
1.71E-07
1.26E-07
8.64E-08
1.11
Incineration
3.16E-02
1.28E-03
3.71E-05
3.54E-05
3.13E-05
2.34E-05
1.65E-05
1.22E-05
8.33E-06
Page 636 of 723
-------
Label
Source Type
Average Air
Concentration (jig/m3)
ADR - Modeled Air
(mg/kg/d)
Daily
Annual
Infant
Young
Toddler
Toddler
Small
Child
Child
Teen
Adult
1.12
Incineration
1.64E-03
1.28E-03
1.92E-06
1.84E-06
1.63E-06
1.22E-06
8.55E-07
6.32E-07
4.32E-07
2.1
Fugitive
2.21E-02
5.35E-06
2.59E-05
2.48E-05
2.19E-05
1.64E-05
1.15E-05
8.51E-06
5.82E-06
2.2
Fugitive
3.43E-03
5.35E-06
4.03E-06
3.85E-06
3.41E-06
2.55E-06
1.79E-06
1.32E-06
9.05E-07
2.3
Fugitive
2.64E-02
6.39E-06
3.10E-05
2.96E-05
2.62E-05
1.96E-05
1.38E-05
1.02E-05
6.95E-06
2.4
Fugitive
4.10E-03
6.39E-06
4.82E-06
4.60E-06
4.07E-06
3.04E-06
2.14E-06
1.58E-06
1.08E-06
2.5
Stack
3.18E-03
4.08E-06
3.74E-06
3.57E-06
3.16E-06
2.36E-06
1.66E-06
1.23E-06
8.40E-07
2.6
Stack
4.90E-04
4.09E-06
5.75E-07
5.50E-07
4.86E-07
3.63E-07
2.56E-07
1.89E-07
1.29E-07
2.7
Stack
3.80E-03
4.87E-06
4.47E-06
4.27E-06
3.77E-06
2.82E-06
1.98E-06
1.47E-06
1.00E-06
2.8
Stack
5.85E-04
4.88E-06
6.87E-07
6.57E-07
5.81E-07
4.34E-07
3.05E-07
2.25E-07
1.54E-07
3.1
Fugitive
2.77E+00
5.05E-05
3.25E-03
3.11E-03
2.75E-03
2.05E-03
1.45E-03
1.07E-03
7.30E-04
3.2
Fugitive
1.30E-01
5.06E-05
1.53E-04
1.47E-04
1.30E-04
9.68E-05
6.81E-05
5.03E-05
3.44E-05
3.3
Stack
3.50E-01
3.86E-05
4.11E-04
3.93E-04
3.47E-04
2.59E-04
1.83E-04
1.35E-04
9.22E-05
3.4
Stack
1.85E-02
3.86E-05
2.18E-05
2.08E-05
1.84E-05
1.37E-05
9.67E-06
7.14E-06
4.88E-06
4.1
Fugitive
3.49E-01
6.36E-06
4.10E-04
3.91E-04
3.46E-04
2.59E-04
1.82E-04
1.34E-04
9.19E-05
4.2
Fugitive
1.64E-02
6.37E-06
1.93E-05
1.84E-05
1.63E-05
1.22E-05
8.58E-06
6.33E-06
4.33E-06
4.3
Stack
4.40E-02
4.86E-06
5.17E-05
4.94E-05
4.37E-05
3.27E-05
2.30E-05
1.70E-05
1.16E-05
4.4
Stack
2.33E-03
4.86E-06
2.74E-06
2.62E-06
2.31E-06
1.73E-06
1.22E-06
8.99E-07
6.15E-07
4.5
Stack
1.89E-01
2.52E-05
2.22E-04
2.12E-04
1.87E-04
1.40E-04
9.86E-05
7.28E-05
4.98E-05
4.6
Stack
1.09E-02
2.51E-05
1.28E-05
1.22E-05
1.08E-05
8.06E-06
5.67E-06
4.19E-06
2.87E-06
4.7
Incineration
1.61E-01
1.89E-04
1.90E-04
1.81E-04
1.60E-04
1.20E-04
8.42E-05
6.22E-05
4.26E-05
4.8
Incineration
6.84E-03
1.89E-04
8.04E-06
7.68E-06
6.79E-06
5.07E-06
3.57E-06
2.64E-06
1.80E-06
4.9
Stack
2.90E+00
3.46E-04
3.40E-03
3.25E-03
2.88E-03
2.15E-03
1.51E-03
1.12E-03
7.64E-04
4.10
Stack
1.43E-01
3.46E-04
1.68E-04
1.61E-04
1.42E-04
1.06E-04
7.47E-05
5.52E-05
3.78E-05
4.11
Incineration
2.30E-01
1.78E-04
2.70E-04
2.58E-04
2.28E-04
1.71E-04
1.20E-04
8.86E-05
6.06E-05
4.12
Incineration
7.14E-03
1.79E-04
8.39E-06
8.02E-06
7.09E-06
5.30E-06
3.73E-06
2.75E-06
1.88E-06
5.1
Stack
3.20E-01
6.67E-04
3.76E-04
3.59E-04
3.18E-04
2.37E-04
1.67E-04
1.23E-04
8.44E-05
5.2
Stack
3.16E-02
6.69E-04
3.72E-05
3.55E-05
3.14E-05
2.35E-05
1.65E-05
1.22E-05
8.34E-06
5.3
Stack
1.60E+00
3.33E-03
1.88E-03
1.80E-03
1.59E-03
1.19E-03
8.35E-04
6.17E-04
4.22E-04
5.4
Stack
1.58E-01
3.35E-03
1.86E-04
1.78E-04
1.57E-04
1.17E-04
8.26E-05
6.10E-05
4.17E-05
5.5
Fugitive
2.25E+00
8.74E-04
2.65E-03
2.53E-03
2.24E-03
1.67E-03
1.18E-03
8.69E-04
5.95E-04
5.6
Fugitive
1.97E-01
8.77E-04
2.31E-04
2.21E-04
1.96E-04
1.46E-04
1.03E-04
7.59E-05
5.19E-05
5.7
Fugitive
1.13E+01
4.37E-03
1.32E-02
1.27E-02
1.12E-02
8.36E-03
5.89E-03
4.35E-03
2.97E-03
5.8
Fugitive
9.85E-01
4.38E-03
1.16E-03
1.11E-03
9.78E-04
7.31E-04
5.14E-04
3.80E-04
2.60E-04
5.9
Incineration
2.61E-01
5.36E-03
3.06E-04
2.93E-04
2.59E-04
1.93E-04
1.36E-04
1.00E-04
6.87E-05
5.10
Incineration
2.09E-02
5.37E-03
2.46E-05
2.35E-05
2.08E-05
1.55E-05
1.09E-05
8.07E-06
5.52E-06
5.11
Incineration
4.96E-01
1.02E-02
5.83E-04
5.57E-04
4.93E-04
3.68E-04
2.59E-04
1.91E-04
1.31E-04
5.12
Incineration
3.99E-02
1.02E-02
4.69E-05
4.48E-05
3.96E-05
2.96E-05
2.08E-05
1.54E-05
1.05E-05
6.1
Fugitive
1.14E-01
4.42E-05
1.34E-04
1.28E-04
1.13E-04
8.46E-05
5.95E-05
4.40E-05
3.01E-05
6.2
Fugitive
3.40E-03
4.46E-05
3.99E-06
3.82E-06
3.37E-06
2.52E-06
1.77E-06
1.31E-06
8.96E-07
6.3
Fugitive
5.07E-01
1.97E-04
5.96E-04
5.70E-04
5.04E-04
3.76E-04
2.65E-04
1.96E-04
1.34E-04
6.4
Fugitive
1.51E-02
1.98E-04
1.78E-05
1.70E-05
1.50E-05
1.12E-05
7.90E-06
5.83E-06
3.99E-06
Page 637 of 723
-------
Label
Source Type
Average Air
Concentration (jig/m3)
ADR - Modeled Air
(mg/kg/d)
Daily
Annual
Infant
Young
Toddler
Toddler
Small
Child
Child
Teen
Adult
6.5
Stack
1.62E-02
3.37E-05
1.90E-05
1.82E-05
1.61E-05
1.20E-05
8.45E-06
6.24E-06
4.27E-06
6.6
Stack
5.93E-04
3.40E-05
6.97E-07
6.66E-07
5.89E-07
4.40E-07
3.10E-07
2.29E-07
1.56E-07
6.7
Stack
7.20E-02
1.50E-04
8.46E-05
8.08E-05
7.15E-05
5.34E-05
3.76E-05
2.78E-05
1.90E-05
6.8
Stack
2.64E-03
1.51E-04
3.10E-06
2.96E-06
2.62E-06
1.96E-06
1.38E-06
1.02E-06
6.96E-07
6.9
Incineration
1.25E-01
2.57E-03
1.47E-04
1.40E-04
1.24E-04
9.25E-05
6.51E-05
4.81E-05
3.29E-05
6.10
Incineration
3.29E-03
2.57E-03
3.87E-06
3.70E-06
3.27E-06
2.44E-06
1.72E-06
1.27E-06
8.68E-07
6.11
Incineration
3.13E-01
6.44E-03
3.68E-04
3.51E-04
3.11E-04
2.32E-04
1.63E-04
1.21E-04
8.26E-05
6.12
Incineration
8.26E-03
6.45E-03
9.71E-06
9.28E-06
8.20E-06
6.13E-06
4.31E-06
3.19E-06
2.18E-06
8.1
Fugitive
8.97E-04
1.64E-08
1.05E-06
1.01E-06
8.91E-07
6.66E-07
4.68E-07
3.46E-07
2.37E-07
8.2
Fugitive
8.93E-02
5.78E-06
1.05E-04
1.00E-04
8.87E-05
6.63E-05
4.66E-05
3.44E-05
2.36E-05
8.3
Incineration
1.25E-03
9.47E-07
1.47E-06
1.40E-06
1.24E-06
9.28E-07
6.53E-07
4.82E-07
3.30E-07
8.4
Incineration
6.60E-02
1.88E-04
7.76E-05
7.41E-05
6.55E-05
4.90E-05
3.45E-05
2.54E-05
1.74E-05
9.1
Fugitive
7.98E-04
1.46E-08
9.38E-07
8.97E-07
7.93E-07
5.92E-07
4.17E-07
3.08E-07
2.11E-07
9.2
Fugitive
7.12E-01
1.30E-05
8.37E-04
7.99E-04
7.07E-04
5.28E-04
3.72E-04
2.75E-04
1.88E-04
10.1
Fugitive
3.35E-02
6.11E-07
3.94E-05
3.76E-05
3.32E-05
2.48E-05
1.75E-05
1.29E-05
8.83E-06
10.2
Fugitive
1.38E-04
6.14E-07
1.62E-07
1.55E-07
1.37E-07
1.02E-07
7.20E-08
5.32E-08
3.64E-08
10.3
Fugitive
1.67E-01
3.06E-06
1.97E-04
1.88E-04
1.66E-04
1.24E-04
8.74E-05
6.46E-05
4.42E-05
10.4
Fugitive
6.89E-04
3.07E-06
8.10E-07
7.74E-07
6.84E-07
5.11E-07
3.60E-07
2.66E-07
1.82E-07
10.5
Stack
4.23E-03
4.67E-07
4.97E-06
4.75E-06
4.20E-06
3.14E-06
2.21E-06
1.63E-06
1.12E-06
10.6
Stack
2.21E-05
4.68E-07
2.60E-08
2.49E-08
2.20E-08
1.64E-08
1.16E-08
8.54E-09
5.84E-09
10.7
Stack
2.12E-02
2.33E-06
2.49E-05
2.37E-05
2.10E-05
1.57E-05
1.10E-05
8.16E-06
5.58E-06
10.8
Stack
1.11E-04
2.34E-06
1.30E-07
1.24E-07
1.10E-07
8.22E-08
5.78E-08
4.27E-08
2.92E-08
10.9
Incineration
4.96E-03
3.75E-06
5.82E-06
5.57E-06
4.92E-06
3.68E-06
2.59E-06
1.91E-06
1.31E-06
10.10
Incineration
1.47E-05
3.76E-06
1.72E-08
1.65E-08
1.46E-08
1.09E-08
7.66E-09
5.66E-09
3.87E-09
10.11
Incineration
5.90E-03
4.47E-06
6.93E-06
6.63E-06
5.86E-06
4.38E-06
3.08E-06
2.28E-06
1.56E-06
10.12
Incineration
1.75E-05
4.47E-06
2.05E-08
1.96E-08
1.73E-08
1.29E-08
9.11E-09
6.73E-09
4.60E-09
11.1
Fugitive
3.10E-02
6.71E-06
3.64E-05
3.48E-05
3.08E-05
2.30E-05
1.62E-05
1.20E-05
8.18E-06
11.2
Fugitive
2.93E-04
6.60E-06
3.45E-07
3.29E-07
2.91E-07
2.18E-07
1.53E-07
1.13E-07
7.73E-08
11.3
Stack
1.63E-01
7.62E-05
1.92E-04
1.83E-04
1.62E-04
1.21E-04
8.52E-05
6.29E-05
4.30E-05
11.4
Stack
1.92E-03
7.54E-05
2.26E-06
2.16E-06
1.91E-06
1.43E-06
1.00E-06
7.41E-07
5.07E-07
12.1
Incineration
1.22E-03
5.06E-06
1.44E-06
1.37E-06
1.21E-06
9.07E-07
6.38E-07
4.71E-07
3.22E-07
12.2
Incineration
6.49E-06
5.07E-06
7.63E-09
7.29E-09
6.45E-09
4.82E-09
3.39E-09
2.50E-09
1.71E-09
12.3
Incineration
1.09E-03
4.50E-06
1.28E-06
1.22E-06
1.08E-06
8.06E-07
5.67E-07
4.19E-07
2.87E-07
12.4
Incineration
5.77E-06
4.50E-06
6.78E-09
6.48E-09
5.73E-09
4.28E-09
3.01E-09
2.23E-09
1.52E-09
Table Apx G-23
». Highly Exposet
Average Daily Dose (mg/kg/day) for Modeled Air Only
Label
Source Type
Average Air
Concentration
(Hg/m3)
ADD - Modeled Air
(mg/kg/d)
Daily
Annual
Infant
Young
Toddler
Toddler
Small
Child
Child
Teen
Adult -
CT
Adult -
HE
1.1
Fugitive
1.17E+00
8.74E-04
6.03E-07
6.13E-07
5.64E-07
4.75E-07
3.30E-07
2.34E-07
3.79E-08
1.04E-07
1.2
Fugitive
6.72E-02
8.82E-04
6.08E-07
6.19E-07
5.69E-07
4.79E-07
3.33E-07
2.36E-07
3.82E-08
1.05E-07
Page 638 of 723
-------
Label
Source Type
Average Air
Concentration
(Hg/m3)
ADD - Modeled Air
(mg/kg/d)
Daily
Annual
Infant
Young
Toddler
Toddler
Small
Child
Child
Teen
Adult -
CT
Adult -
HE
1.3
Fugitive
5.85E+00
4.37E-03
3.01E-06
3.07E-06
2.82E-06
2.37E-06
1.65E-06
1.17E-06
1.89E-07
5.21E-07
1.4
Fugitive
3.36E-01
4.41E-03
3.04E-06
3.09E-06
2.84E-06
2.39E-06
1.66E-06
1.18E-06
1.91E-07
5.25E-07
1.5
Stack
1.71E-01
6.67E-04
4.60E-07
4.68E-07
4.30E-07
3.62E-07
2.52E-07
1.78E-07
2.89E-08
7.95E-08
1.6
Stack
1.17E-02
6.72E-04
4.63E-07
4.71E-07
4.33E-07
3.65E-07
2.53E-07
1.80E-07
2.91E-08
8.00E-08
1.7
Stack
8.54E-01
3.33E-03
2.30E-06
2.34E-06
2.15E-06
1.81E-06
1.26E-06
8.92E-07
1.44E-07
3.97E-07
1.8
Stack
5.86E-02
3.36E-03
2.32E-06
2.36E-06
2.17E-06
1.82E-06
1.27E-06
8.99E-07
1.46E-07
4.00E-07
1.9
Incineration
6.31E-03
2.56E-04
1.76E-07
1.79E-07
1.65E-07
1.39E-07
9.64E-08
6.84E-08
1.11E-08
3.04E-08
1.10
Incineration
3.28E-04
2.56E-04
1.76E-07
1.79E-07
1.65E-07
1.39E-07
9.65E-08
6.84E-08
1.11E-08
3.05E-08
1.11
Incineration
3.16E-02
1.28E-03
8.81E-07
8.97E-07
8.24E-07
6.94E-07
4.82E-07
3.42E-07
5.54E-08
1.52E-07
1.12
Incineration
1.64E-03
1.28E-03
8.81E-07
8.97E-07
8.24E-07
6.94E-07
4.82E-07
3.42E-07
5.54E-08
1.52E-07
2.1
Fugitive
2.21E-02
5.35E-06
3.69E-09
3.75E-09
3.45E-09
2.90E-09
2.02E-09
1.43E-09
2.32E-10
6.37E-10
2.2
Fugitive
3.43E-03
5.35E-06
3.69E-09
3.75E-09
3.45E-09
2.90E-09
2.02E-09
1.43E-09
2.32E-10
6.37E-10
2.3
Fugitive
2.64E-02
6.39E-06
4.41E-09
4.48E-09
4.12E-09
3.47E-09
2.41E-09
1.71E-09
2.77E-10
7.61E-10
2.4
Fugitive
4.10E-03
6.39E-06
4.40E-09
4.48E-09
4.12E-09
3.47E-09
2.41E-09
1.71E-09
2.77E-10
7.61E-10
2.5
Stack
3.18E-03
4.08E-06
2.81E-09
2.86E-09
2.63E-09
2.22E-09
1.54E-09
1.09E-09
1.77E-10
4.86E-10
2.6
Stack
4.90E-04
4.09E-06
2.82E-09
2.87E-09
2.64E-09
2.22E-09
1.54E-09
1.09E-09
1.77E-10
4.87E-10
2.7
Stack
3.80E-03
4.87E-06
3.36E-09
3.42E-09
3.14E-09
2.65E-09
1.84E-09
1.30E-09
2.11E-10
5.81E-10
2.8
Stack
5.85E-04
4.88E-06
3.37E-09
3.43E-09
3.15E-09
2.65E-09
1.84E-09
1.31E-09
2.12E-10
5.82E-10
3.1
Fugitive
2.77E+00
5.05E-05
3.49E-08
3.55E-08
3.26E-08
2.74E-08
1.91E-08
1.35E-08
2.19E-09
6.02E-09
3.2
Fugitive
1.30E-01
5.06E-05
3.49E-08
3.55E-08
3.26E-08
2.75E-08
1.91E-08
1.35E-08
2.19E-09
6.03E-09
3.3
Stack
3.50E-01
3.86E-05
2.66E-08
2.71E-08
2.49E-08
2.09E-08
1.46E-08
1.03E-08
1.67E-09
4.60E-09
3.4
Stack
1.85E-02
3.86E-05
2.66E-08
2.71E-08
2.49E-08
2.09E-08
1.46E-08
1.03E-08
1.67E-09
4.60E-09
4.1
Fugitive
3.49E-01
6.36E-06
4.39E-09
4.47E-09
4.10E-09
3.46E-09
2.40E-09
1.70E-09
2.76E-10
7.58E-10
4.2
Fugitive
1.64E-02
6.37E-06
4.39E-09
4.47E-09
4.11E-09
3.46E-09
2.40E-09
1.70E-09
2.76E-10
7.59E-10
4.3
Stack
4.40E-02
4.86E-06
3.35E-09
3.41E-09
3.13E-09
2.64E-09
1.83E-09
1.30E-09
2.10E-10
5.79E-10
4.4
Stack
2.33E-03
4.86E-06
3.35E-09
3.41E-09
3.13E-09
2.64E-09
1.83E-09
1.30E-09
2.10E-10
5.79E-10
4.5
Stack
1.89E-01
2.52E-05
1.74E-08
1.77E-08
1.62E-08
1.37E-08
9.49E-09
6.73E-09
1.09E-09
3.00E-09
4.6
Stack
1.09E-02
2.51E-05
1.73E-08
1.76E-08
1.62E-08
1.36E-08
9.48E-09
6.72E-09
1.09E-09
2.99E-09
4.7
Incineration
1.61E-01
1.89E-04
1.30E-07
1.32E-07
1.22E-07
1.02E-07
7.12E-08
5.05E-08
8.17E-09
2.25E-08
4.8
Incineration
6.84E-03
1.89E-04
1.30E-07
1.33E-07
1.22E-07
1.03E-07
7.14E-08
5.06E-08
8.19E-09
2.25E-08
4.9
Stack
2.90E+00
3.46E-04
2.39E-07
2.43E-07
2.23E-07
1.88E-07
1.31E-07
9.26E-08
1.50E-08
4.12E-08
4.10
Stack
1.43E-01
3.46E-04
2.38E-07
2.43E-07
2.23E-07
1.88E-07
1.30E-07
9.25E-08
1.50E-08
4.12E-08
4.11
Incineration
2.30E-01
1.78E-04
1.23E-07
1.25E-07
1.15E-07
9.68E-08
6.73E-08
4.77E-08
7.72E-09
2.12E-08
4.12
Incineration
7.14E-03
1.79E-04
1.23E-07
1.26E-07
1.15E-07
9.72E-08
6.75E-08
4.79E-08
7.75E-09
2.13E-08
5.1
Stack
3.20E-01
6.67E-04
4.60E-07
4.68E-07
4.30E-07
3.62E-07
2.52E-07
1.78E-07
2.89E-08
7.94E-08
5.2
Stack
3.16E-02
6.69E-04
4.61E-07
4.70E-07
4.32E-07
3.63E-07
2.52E-07
1.79E-07
2.90E-08
7.97E-08
5.3
Stack
1.60E+00
3.33E-03
2.30E-06
2.34E-06
2.15E-06
1.81E-06
1.26E-06
8.92E-07
1.44E-07
3.97E-07
5.4
Stack
1.58E-01
3.35E-03
2.31E-06
2.35E-06
2.16E-06
1.82E-06
1.26E-06
8.95E-07
1.45E-07
3.99E-07
5.5
Fugitive
2.25E+00
8.74E-04
6.03E-07
6.13E-07
5.64E-07
4.75E-07
3.30E-07
2.34E-07
3.79E-08
1.04E-07
5.6
Fugitive
1.97E-01
8.77E-04
6.04E-07
6.15E-07
5.65E-07
4.76E-07
3.31E-07
2.35E-07
3.80E-08
1.04E-07
5.7
Fugitive
1.13E+01
4.37E-03
3.01E-06
3.07E-06
2.82E-06
2.37E-06
1.65E-06
1.17E-06
1.89E-07
5.21E-07
Page 639 of 723
-------
Label
Source Type
Average Air
Concentration
(Hg/m3)
ADD - Modeled Air
(mg/kg/d)
Daily
Annual
Infant
Young
Toddler
Toddler
Small
Child
Child
Teen
Adult -
CT
Adult -
HE
5.8
Fugitive
9.85E-01
4.38E-03
3.02E-06
3.08E-06
2.83E-06
2.38E-06
1.65E-06
1.17E-06
1.90E-07
5.22E-07
5.9
Incineration
2.61E-01
5.36E-03
3.70E-06
3.76E-06
3.46E-06
2.91E-06
2.02E-06
1.44E-06
2.32E-07
6.39E-07
5.10
Incineration
2.09E-02
5.37E-03
3.70E-06
3.77E-06
3.46E-06
2.91E-06
2.03E-06
1.44E-06
2.33E-07
6.40E-07
5.11
Incineration
4.96E-01
1.02E-02
7.05E-06
7.17E-06
6.59E-06
5.55E-06
3.86E-06
2.73E-06
4.43E-07
1.22E-06
5.12
Incineration
3.99E-02
1.02E-02
7.05E-06
7.17E-06
6.59E-06
5.55E-06
3.86E-06
2.74E-06
4.43E-07
1.22E-06
6.1
Fugitive
1.14E-01
4.42E-05
3.05E-08
3.10E-08
2.85E-08
2.40E-08
1.67E-08
1.18E-08
1.91E-09
5.27E-09
6.2
Fugitive
3.40E-03
4.46E-05
3.07E-08
3.13E-08
2.87E-08
2.42E-08
1.68E-08
1.19E-08
1.93E-09
5.31E-09
6.3
Fugitive
5.07E-01
1.97E-04
1.36E-07
1.38E-07
1.27E-07
1.07E-07
7.42E-08
5.26E-08
8.52E-09
2.34E-08
6.4
Fugitive
1.51E-02
1.98E-04
1.37E-07
1.39E-07
1.28E-07
1.08E-07
7.48E-08
5.31E-08
8.59E-09
2.36E-08
6.5
Stack
1.62E-02
3.37E-05
2.32E-08
2.37E-08
2.17E-08
1.83E-08
1.27E-08
9.02E-09
1.46E-09
4.02E-09
6.6
Stack
5.93E-04
3.40E-05
2.34E-08
2.38E-08
2.19E-08
1.84E-08
1.28E-08
9.09E-09
1.47E-09
4.05E-09
6.7
Stack
7.20E-02
1.50E-04
1.03E-07
1.05E-07
9.67E-08
8.14E-08
5.66E-08
4.01E-08
6.50E-09
1.79E-08
6.8
Stack
2.64E-03
1.51E-04
1.04E-07
1.06E-07
9.75E-08
8.21E-08
5.70E-08
4.04E-08
6.55E-09
1.80E-08
6.9
Incineration
1.25E-01
2.57E-03
1.77E-06
1.80E-06
1.66E-06
1.39E-06
9.69E-07
6.87E-07
1.11E-07
3.06E-07
6.10
Incineration
3.29E-03
2.57E-03
1.77E-06
1.80E-06
1.66E-06
1.39E-06
9.69E-07
6.87E-07
1.11E-07
3.06E-07
6.11
Incineration
3.13E-01
6.44E-03
4.44E-06
4.52E-06
4.16E-06
3.50E-06
2.43E-06
1.72E-06
2.79E-07
7.68E-07
6.12
Incineration
8.26E-03
6.45E-03
4.45E-06
4.52E-06
4.16E-06
3.50E-06
2.43E-06
1.73E-06
2.79E-07
7.68E-07
8.1
Fugitive
8.97E-04
1.64E-08
1.13E-11
1.15E-11
1.06E-11
8.89E-12
6.18E-12
4.38E-12
7.10E-13
1.95E-12
8.2
Fugitive
8.93E-02
5.78E-06
3.99E-09
4.06E-09
3.73E-09
3.14E-09
2.18E-09
1.55E-09
2.51E-10
6.89E-10
8.3
Incineration
1.25E-03
9.47E-07
6.53E-10
6.65E-10
6.11E-10
5.14E-10
3.57E-10
2.53E-10
4.10E-11
1.13E-10
8.4
Incineration
6.60E-02
1.88E-04
1.29E-07
1.32E-07
1.21E-07
1.02E-07
7.09E-08
5.02E-08
8.14E-09
2.24E-08
9.1
Fugitive
7.98E-04
1.46E-08
1.01E-11
1.02E-11
9.40E-12
7.91E-12
5.50E-12
3.90E-12
6.32E-13
1.74E-12
9.2
Fugitive
7.12E-01
1.30E-05
8.96E-09
9.12E-09
8.38E-09
7.06E-09
4.90E-09
3.48E-09
5.63E-10
1.55E-09
10.1
Fugitive
3.35E-02
6.11E-07
4.22E-10
4.29E-10
3.94E-10
3.32E-10
2.31E-10
1.64E-10
2.65E-11
7.28E-11
10.2
Fugitive
1.38E-04
6.14E-07
4.23E-10
4.31E-10
3.96E-10
3.33E-10
2.32E-10
1.64E-10
2.66E-11
7.31E-11
10.3
Fugitive
1.67E-01
3.06E-06
2.11E-09
2.14E-09
1.97E-09
1.66E-09
1.15E-09
8.18E-10
1.32E-10
3.64E-10
10.4
Fugitive
6.89E-04
3.07E-06
2.12E-09
2.15E-09
1.98E-09
1.67E-09
1.16E-09
8.21E-10
1.33E-10
3.66E-10
10.5
Stack
4.23E-03
4.67E-07
3.22E-10
3.27E-10
3.01E-10
2.53E-10
1.76E-10
1.25E-10
2.02E-11
5.56E-11
10.6
Stack
2.21E-05
4.68E-07
3.23E-10
3.29E-10
3.02E-10
2.54E-10
1.77E-10
1.25E-10
2.03E-11
5.58E-11
10.7
Stack
2.12E-02
2.33E-06
1.61E-09
1.64E-09
1.50E-09
1.27E-09
8.80E-10
6.24E-10
1.01E-10
2.78E-10
10.8
Stack
1.11E-04
2.34E-06
1.62E-09
1.64E-09
1.51E-09
1.27E-09
8.84E-10
6.27E-10
1.01E-10
2.79E-10
10.9
Incineration
4.96E-03
3.75E-06
2.59E-09
2.63E-09
2.42E-09
2.04E-09
1.42E-09
1.00E-09
1.63E-10
4.47E-10
10.10
Incineration
1.47E-05
3.76E-06
2.59E-09
2.64E-09
2.43E-09
2.04E-09
1.42E-09
1.01E-09
1.63E-10
4.48E-10
10.11
Incineration
5.90E-03
4.47E-06
3.08E-09
3.14E-09
2.88E-09
2.43E-09
1.69E-09
1.20E-09
1.94E-10
5.33E-10
10.12
Incineration
1.75E-05
4.47E-06
3.08E-09
3.14E-09
2.88E-09
2.43E-09
1.69E-09
1.20E-09
1.94E-10
5.33E-10
11.1
Fugitive
3.10E-02
6.71E-06
4.63E-09
4.71E-09
4.33E-09
3.64E-09
2.53E-09
1.80E-09
2.91E-10
8.00E-10
11.2
Fugitive
2.93E-04
6.60E-06
4.55E-09
4.63E-09
4.26E-09
3.59E-09
2.49E-09
1.77E-09
2.86E-10
7.87E-10
11.3
Stack
1.63E-01
7.62E-05
5.25E-08
5.34E-08
4.91E-08
4.14E-08
2.87E-08
2.04E-08
3.30E-09
9.08E-09
11.4
Stack
1.92E-03
7.54E-05
5.20E-08
5.29E-08
4.87E-08
4.10E-08
2.85E-08
2.02E-08
3.27E-09
8.99E-09
12.1
Incineration
1.22E-03
5.06E-06
3.49E-09
3.55E-09
3.27E-09
2.75E-09
1.91E-09
1.35E-09
2.19E-10
6.03E-10
12.2
Incineration
6.49E-06
5.07E-06
3.49E-09
3.56E-09
3.27E-09
2.75E-09
1.91E-09
1.36E-09
2.20E-10
6.04E-10
Page 640 of 723
-------
Label
Source Type
Average Air
Concentration
(Hg/m3)
ADD - Modeled Air
(mg/kg/d)
Daily
Annual
Infant
Young
Toddler
Toddler
Small
Child
Child
Teen
Adult -
CT
Adult -
HE
12.3
Incineration
1.09E-03
4.50E-06
3.10E-09
3.16E-09
2.90E-09
2.44E-09
1.70E-09
1.20E-09
1.95E-10
5.36E-10
12.4
Incineration
5.77E-06
4.50E-06
3.11E-09
3.16E-09
2.90E-09
2.45E-09
1.70E-09
1.21E-09
1.95E-10
5.37E-10
TableApx G-24. Highly Exposed Aggregate Acute Dose Rate (mg/kg/day) for Modeled Air and
Non-Air and Background
Label
Source
Type
ADR - Modeled Air and Non-Air Background (mg/kg/d)
Infant
Young Toddler
Toddler
Small Child
Child
Teen
Adult
1.1
Fugitive
1.41E-03
1.34E-03
1.18E-03
8.81E-04
6.19E-04
4.55E-04
3.11E-04
1.2
Fugitive
1.18E-04
1.04E-04
8.44E-05
6.29E-05
4.33E-05
3.00E-05
2.06E-05
1.3
Fugitive
6.91E-03
6.59E-03
5.82E-03
4.35E-03
3.06E-03
2.26E-03
1.55E-03
1.4
Fugitive
4.34E-04
4.06E-04
3.51E-04
2.62E-04
1.84E-04
1.34E-04
9.16E-05
1.5
Stack
2.40E-04
2.20E-04
1.87E-04
1.40E-04
9.74E-05
7.00E-05
4.80E-05
1.6
Stack
5.31E-05
4.18E-05
2.93E-05
2.17E-05
1.43E-05
8.63E-06
6.01E-06
1.7
Stack
1.04E-03
9.87E-04
8.65E-04
6.47E-04
4.54E-04
3.33E-04
2.28E-04
1.8
Stack
1.08E-04
9.45E-05
7.58E-05
5.65E-05
3.88E-05
2.67E-05
1.84E-05
1.9
Incineration
4.67E-05
3.57E-05
2.39E-05
1.77E-05
1.15E-05
6.54E-06
4.58E-06
1.10
Incineration
3.97E-05
2.90E-05
1.80E-05
1.32E-05
8.38E-06
4.24E-06
3.00E-06
1.11
Incineration
7.64E-05
6.41E-05
4.90E-05
3.64E-05
2.47E-05
1.63E-05
1.12E-05
1.12
Incineration
4.12E-05
3.05E-05
1.93E-05
1.42E-05
9.07E-06
4.74E-06
3.35E-06
2.1
Fugitive
6.52E-05
5.34E-05
3.95E-05
2.94E-05
1.97E-05
1.26E-05
8.74E-06
2.2
Fugitive
4.33E-05
3.25E-05
2.10E-05
1.55E-05
1.00E-05
5.43E-06
3.82E-06
2.3
Fugitive
7.03E-05
5.82E-05
4.38E-05
3.26E-05
2.20E-05
1.43E-05
9.87E-06
2.4
Fugitive
4.41E-05
3.32E-05
2.17E-05
1.60E-05
1.04E-05
5.69E-06
4.00E-06
2.5
Stack
4.30E-05
3.22E-05
2.08E-05
1.54E-05
9.87E-06
5.34E-06
3.75E-06
2.6
Stack
3.99E-05
2.92E-05
1.81E-05
1.34E-05
8.47E-06
4.30E-06
3.04E-06
2.7
Stack
4.38E-05
3.29E-05
2.14E-05
1.58E-05
1.02E-05
5.58E-06
3.92E-06
2.8
Stack
4.00E-05
2.93E-05
1.82E-05
1.34E-05
8.52E-06
4.33E-06
3.07E-06
3.1
Fugitive
3.29E-03
3.14E-03
2.77E-03
2.07E-03
1.45E-03
1.07E-03
7.33E-04
3.2
Fugitive
1.93E-04
1.75E-04
1.47E-04
1.10E-04
7.63E-05
5.44E-05
3.73E-05
3.3
Stack
4.50E-04
4.21E-04
3.65E-04
2.72E-04
1.91E-04
1.39E-04
9.52E-05
3.4
Stack
6.11E-05
4.94E-05
3.60E-05
2.67E-05
1.79E-05
1.12E-05
7.80E-06
4.1
Fugitive
4.49E-04
4.20E-04
3.64E-04
2.72E-04
1.90E-04
1.39E-04
9.49E-05
4.2
Fugitive
5.86E-05
4.71E-05
3.39E-05
2.52E-05
1.68E-05
1.04E-05
7.25E-06
4.3
Stack
9.10E-05
7.81E-05
6.13E-05
4.57E-05
3.12E-05
2.11E-05
1.45E-05
4.4
Stack
4.20E-05
3.13E-05
1.99E-05
1.47E-05
9.43E-06
5.01E-06
3.53E-06
4.5
Stack
2.61E-04
2.41E-04
2.05E-04
1.53E-04
1.07E-04
7.69E-05
5.27E-05
4.6
Stack
5.21E-05
4.08E-05
2.84E-05
2.11E-05
1.39E-05
8.30E-06
5.78E-06
4.7
Incineration
2.29E-04
2.10E-04
1.78E-04
1.33E-04
9.25E-05
6.63E-05
4.55E-05
4.8
Incineration
4.73E-05
3.63E-05
2.44E-05
1.81E-05
1.18E-05
6.75E-06
4.72E-06
4.9
Stack
3.44E-03
3.28E-03
2.89E-03
2.16E-03
1.52E-03
1.12E-03
7.67E-04
4.10
Stack
2.08E-04
1.89E-04
1.60E-04
1.19E-04
8.30E-05
5.93E-05
4.07E-05
Page 641 of 723
-------
Label
Source
Type
ADR - Modeled Air and Non-Air Background (mg/kg/d)
Infant
Young Toddler
Toddler
Small Child
Child
Teen
Adult
4.11
Incineration
3.09E-04
2.87E-04
2.46E-04
1.84E-04
1.28E-04
9.28E-05
6.36E-05
4.12
Incineration
4.77E-05
3.67E-05
2.47E-05
1.83E-05
1.19E-05
6.86E-06
4.80E-06
5.1
Stack
4.15E-04
3.88E-04
3.35E-04
2.50E-04
1.75E-04
1.27E-04
8.73E-05
5.2
Stack
7.65E-05
6.42E-05
4.90E-05
3.65E-05
2.47E-05
1.63E-05
1.13E-05
5.3
Stack
1.92E-03
1.83E-03
1.61E-03
1.20E-03
8.43E-04
6.21E-04
4.25E-04
5.4
Stack
2.25E-04
2.06E-04
1.75E-04
1.30E-04
9.08E-05
6.51E-05
4.46E-05
5.5
Fugitive
2.69E-03
2.56E-03
2.26E-03
1.69E-03
1.19E-03
8.73E-04
5.98E-04
5.6
Fugitive
2.71E-04
2.50E-04
2.13E-04
1.59E-04
1.11E-04
8.00E-05
5.49E-05
5.7
Fugitive
1.33E-02
1.27E-02
1.12E-02
8.38E-03
5.89E-03
4.35E-03
2.98E-03
5.8
Fugitive
1.20E-03
1.13E-03
9.95E-04
7.44E-04
5.22E-04
3.84E-04
2.63E-04
5.9
Incineration
3.45E-04
3.21E-04
2.76E-04
2.06E-04
1.44E-04
1.05E-04
7.16E-05
5.10
Incineration
6.39E-05
5.22E-05
3.84E-05
2.85E-05
1.91E-05
1.22E-05
8.44E-06
5.11
Incineration
6.22E-04
5.86E-04
5.10E-04
3.81E-04
2.67E-04
1.95E-04
1.34E-04
5.12
Incineration
8.62E-05
7.34E-05
5.72E-05
4.26E-05
2.90E-05
1.95E-05
1.34E-05
6.1
Fugitive
1.73E-04
1.57E-04
1.31E-04
9.76E-05
6.77E-05
4.81E-05
3.30E-05
6.2
Fugitive
4.33E-05
3.25E-05
2.10E-05
1.55E-05
9.99E-06
5.42E-06
3.81E-06
6.3
Fugitive
6.35E-04
5.98E-04
5.21E-04
3.89E-04
2.73E-04
2.00E-04
1.37E-04
6.4
Fugitive
5.71E-05
4.56E-05
3.26E-05
2.42E-05
1.61E-05
9.94E-06
6.90E-06
6.5
Stack
5.83E-05
4.68E-05
3.37E-05
2.50E-05
1.67E-05
1.03E-05
7.18E-06
6.6
Stack
4.00E-05
2.93E-05
1.82E-05
1.34E-05
8.52E-06
4.34E-06
3.07E-06
6.7
Stack
1.24E-04
1.09E-04
8.91E-05
6.64E-05
4.58E-05
3.19E-05
2.19E-05
6.8
Stack
4.24E-05
3.16E-05
2.03E-05
1.50E-05
9.59E-06
5.13E-06
3.61E-06
6.9
Incineration
1.86E-04
1.69E-04
1.41E-04
1.06E-04
7.33E-05
5.22E-05
3.58E-05
6.10
Incineration
4.32E-05
3.23E-05
2.09E-05
1.54E-05
9.93E-06
5.38E-06
3.78E-06
6.11
Incineration
4.07E-04
3.80E-04
3.28E-04
2.45E-04
1.72E-04
1.25E-04
8.55E-05
6.12
Incineration
4.90E-05
3.79E-05
2.58E-05
1.91E-05
1.25E-05
7.30E-06
5.09E-06
8.1
Fugitive
4.03E-05
2.97E-05
1.85E-05
1.37E-05
8.68E-06
4.46E-06
3.15E-06
8.2
Fugitive
1.44E-04
1.29E-04
1.06E-04
7.93E-05
5.48E-05
3.85E-05
2.65E-05
8.3
Incineration
4.08E-05
3.01E-05
1.89E-05
1.39E-05
8.86E-06
4.59E-06
3.24E-06
8.4
Incineration
1.17E-04
1.03E-04
8.32E-05
6.20E-05
4.27E-05
2.96E-05
2.03E-05
9.1
Fugitive
4.02E-05
2.95E-05
1.84E-05
1.36E-05
8.63E-06
4.42E-06
3.13E-06
9.2
Fugitive
8.76E-04
8.28E-04
7.24E-04
5.41E-04
3.80E-04
2.79E-04
1.91E-04
10.1
Fugitive
7.86E-05
6.63E-05
5.09E-05
3.79E-05
2.57E-05
1.70E-05
1.17E-05
10.2
Fugitive
3.95E-05
2.88E-05
1.78E-05
1.31E-05
8.28E-06
4.16E-06
2.95E-06
10.3
Fugitive
2.36E-04
2.17E-04
1.84E-04
1.37E-04
9.56E-05
6.87E-05
4.71E-05
10.4
Fugitive
4.01E-05
2.94E-05
1.83E-05
1.35E-05
8.57E-06
4.38E-06
3.10E-06
10.5
Stack
4.43E-05
3.34E-05
2.18E-05
1.61E-05
1.04E-05
5.74E-06
4.03E-06
10.6
Stack
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.22E-06
4.12E-06
2.92E-06
10.7
Stack
6.41E-05
5.24E-05
3.86E-05
2.87E-05
1.93E-05
1.23E-05
8.49E-06
10.8
Stack
3.94E-05
2.88E-05
1.77E-05
1.31E-05
8.27E-06
4.15E-06
2.94E-06
10.9
Incineration
4.51E-05
3.42E-05
2.26E-05
1.67E-05
1.08E-05
6.02E-06
4.22E-06
10.10
Incineration
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.22E-06
4.11E-06
2.92E-06
Page 642 of 723
-------
Label
Source
Type
ADR - Modeled Air and Non-Air Background (mg/kg/d)
Infant
Young Toddler
Toddler
Small Child
Child
Teen
Adult
10.11
Incineration
4.62E-05
3.53E-05
2.35E-05
1.74E-05
1.13E-05
6.38E-06
4.47E-06
10.12
Incineration
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.22E-06
4.12E-06
2.92E-06
11.1
Fugitive
7.57E-05
6.35E-05
4.84E-05
3.60E-05
2.44E-05
1.61E-05
1.11E-05
11.2
Fugitive
3.96E-05
2.90E-05
1.79E-05
1.32E-05
8.36E-06
4.22E-06
2.99E-06
11.3
Stack
2.31E-04
2.12E-04
1.80E-04
1.34E-04
9.34E-05
6.70E-05
4.59E-05
11.4
Stack
4.16E-05
3.08E-05
1.95E-05
1.44E-05
9.21E-06
4.85E-06
3.42E-06
12.1
Incineration
4.07E-05
3.00E-05
1.88E-05
1.39E-05
8.85E-06
4.58E-06
3.24E-06
12.2
Incineration
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
12.3
Incineration
4.06E-05
2.99E-05
1.87E-05
1.38E-05
8.78E-06
4.53E-06
3.20E-06
12.4
Incineration
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
Page 643 of 723
-------
TableApx G-25. Highly Exposed Aggregate Average Daily Dose (mg/kg/day) for Modeled Air
and Non-Air and Background
Label
Source Type
ADD - Modeled Air and Non-Air Background (mg/kg/d)
Infant
Young Toddler
Toddler
Small Child
Child
Teen
Adult
1.1
Fugitive
3.99E-05
2.93E-05
1.82E-05
1.35E-05
8.54E-06
4.34E-06
2.95E-06
1.2
Fugitive
3.99E-05
2.93E-05
1.82E-05
1.35E-05
8.54E-06
4.35E-06
2.95E-06
1.3
Fugitive
4.23E-05
3.17E-05
2.04E-05
1.54E-05
9.86E-06
5.28E-06
3.10E-06
1.4
Fugitive
4.23E-05
3.17E-05
2.05E-05
1.54E-05
9.87E-06
5.29E-06
3.11E-06
1.5
Stack
3.98E-05
2.91E-05
1.81E-05
1.34E-05
8.46E-06
4.29E-06
2.94E-06
1.6
Stack
3.98E-05
2.91E-05
1.81E-05
1.34E-05
8.46E-06
4.29E-06
2.94E-06
1.7
Stack
4.16E-05
3.10E-05
1.98E-05
1.48E-05
9.47E-06
5.00E-06
3.06E-06
1.8
Stack
4.16E-05
3.10E-05
1.98E-05
1.48E-05
9.48E-06
5.01E-06
3.06E-06
1.9
Incineration
3.95E-05
2.88E-05
1.78E-05
1.31E-05
8.31E-06
4.18E-06
2.93E-06
1.10
Incineration
3.95E-05
2.88E-05
1.78E-05
1.31E-05
8.31E-06
4.18E-06
2.93E-06
1.11
Incineration
4.02E-05
2.95E-05
1.85E-05
1.37E-05
8.69E-06
4.45E-06
2.97E-06
1.12
Incineration
4.02E-05
2.95E-05
1.85E-05
1.37E-05
8.69E-06
4.45E-06
2.97E-06
2.1
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
2.2
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
2.3
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
2.4
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
2.5
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
2.6
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
2.7
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
2.8
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
3.1
Fugitive
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.23E-06
4.12E-06
2.92E-06
3.2
Fugitive
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.23E-06
4.12E-06
2.92E-06
3.3
Stack
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.23E-06
4.12E-06
2.92E-06
3.4
Stack
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.23E-06
4.12E-06
2.92E-06
4.1
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
4.2
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
4.3
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
4.4
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
4.5
Stack
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.22E-06
4.12E-06
2.92E-06
4.6
Stack
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.22E-06
4.12E-06
2.92E-06
4.7
Incineration
3.94E-05
2.88E-05
1.78E-05
1.31E-05
8.28E-06
4.16E-06
2.92E-06
4.8
Incineration
3.94E-05
2.88E-05
1.78E-05
1.31E-05
8.28E-06
4.16E-06
2.92E-06
4.9
Stack
3.95E-05
2.89E-05
1.79E-05
1.32E-05
8.34E-06
4.20E-06
2.93E-06
4.10
Stack
3.95E-05
2.89E-05
1.79E-05
1.32E-05
8.34E-06
4.20E-06
2.93E-06
4.11
Incineration
3.94E-05
2.88E-05
1.77E-05
1.31E-05
8.28E-06
4.16E-06
2.92E-06
4.12
Incineration
3.94E-05
2.88E-05
1.77E-05
1.31E-05
8.28E-06
4.16E-06
2.92E-06
5.1
Stack
3.98E-05
2.91E-05
1.81E-05
1.34E-05
8.46E-06
4.29E-06
2.94E-06
5.2
Stack
3.98E-05
2.91E-05
1.81E-05
1.34E-05
8.46E-06
4.29E-06
2.94E-06
5.3
Stack
4.16E-05
3.10E-05
1.98E-05
1.48E-05
9.47E-06
5.00E-06
3.06E-06
Page 644 of 723
-------
Label
Source Type
ADD - Modeled Air and Non-Air Background (mg/kg/d)
Infant
Young Toddler
Toddler
Small Child
Child
Teen
Adult
5.4
Stack
4.16E-05
3.10E-05
1.98E-05
1.48E-05
9.47E-06
5.00E-06
3.06E-06
5.5
Fugitive
3.99E-05
2.93E-05
1.82E-05
1.35E-05
8.54E-06
4.34E-06
2.95E-06
5.6
Fugitive
3.99E-05
2.93E-05
1.82E-05
1.35E-05
8.54E-06
4.34E-06
2.95E-06
5.7
Fugitive
4.23E-05
3.17E-05
2.05E-05
1.54E-05
9.86E-06
5.28E-06
3.10E-06
5.8
Fugitive
4.23E-05
3.17E-05
2.05E-05
1.54E-05
9.87E-06
5.28E-06
3.10E-06
5.9
Incineration
4.30E-05
3.24E-05
2.11E-05
1.59E-05
1.02E-05
5.54E-06
3.15E-06
5.10
Incineration
4.30E-05
3.24E-05
2.11E-05
1.59E-05
1.02E-05
5.55E-06
3.15E-06
5.11
Incineration
4.63E-05
3.58E-05
2.42E-05
1.86E-05
1.21E-05
6.84E-06
3.36E-06
5.12
Incineration
4.63E-05
3.58E-05
2.42E-05
1.86E-05
1.21E-05
6.84E-06
3.36E-06
6.1
Fugitive
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.23E-06
4.12E-06
2.92E-06
6.2
Fugitive
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.23E-06
4.12E-06
2.92E-06
6.3
Fugitive
3.94E-05
2.88E-05
1.78E-05
1.31E-05
8.29E-06
4.16E-06
2.92E-06
6.4
Fugitive
3.94E-05
2.88E-05
1.78E-05
1.31E-05
8.29E-06
4.16E-06
2.92E-06
6.5
Stack
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.22E-06
4.12E-06
2.92E-06
6.6
Stack
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.22E-06
4.12E-06
2.92E-06
6.7
Stack
3.94E-05
2.88E-05
1.77E-05
1.31E-05
8.27E-06
4.15E-06
2.92E-06
6.8
Stack
3.94E-05
2.88E-05
1.77E-05
1.31E-05
8.27E-06
4.15E-06
2.92E-06
6.9
Incineration
4.11E-05
3.04E-05
1.93E-05
1.44E-05
9.18E-06
4.80E-06
3.03E-06
6.10
Incineration
4.11E-05
3.04E-05
1.93E-05
1.44E-05
9.18E-06
4.80E-06
3.03E-06
6.11
Incineration
4.37E-05
3.32E-05
2.18E-05
1.65E-05
1.06E-05
5.83E-06
3.19E-06
6.12
Incineration
4.37E-05
3.32E-05
2.18E-05
1.65E-05
1.06E-05
5.83E-06
3.19E-06
8.1
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
8.2
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
8.3
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
8.4
Incineration
3.94E-05
2.88E-05
1.78E-05
1.31E-05
8.28E-06
4.16E-06
2.92E-06
9.1
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
9.2
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.22E-06
4.11E-06
2.92E-06
10.1
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.2
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.3
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.4
Fugitive
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.5
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.6
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.7
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.8
Stack
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.91E-06
10.9
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
10.10
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
10.11
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
10.12
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
11.1
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
11.2
Fugitive
3.93E-05
2.87E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
11.3
Stack
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.24E-06
4.13E-06
2.92E-06
Page 645 of 723
-------
Label
Source Type
ADD - Modeled Air and Non-Air Background (mg/kg/d)
Infant
Young Toddler
Toddler
Small Child
Child
Teen
Adult
11.4
Stack
3.93E-05
2.87E-05
1.77E-05
1.30E-05
8.24E-06
4.13E-06
2.92E-06
12.1
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
12.2
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
12.3
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
12.4
Incineration
3.93E-05
2.86E-05
1.76E-05
1.30E-05
8.21E-06
4.11E-06
2.92E-06
G.5 Scenarios CI and C2: Consumer Exposure to XPS/EPS
Insulation in Residences and Automobiles Calculations
G.5.1 General Mass Balance Equation Used in IECCU
EPA used the following general mass balance as defined in the user guide of the IECCU model to
estimate the indoor concentrations of HBCD in indoor air and dust of a multi-zone indoor environment
(U.S. EPA 2019r).
EquationApx G-3
v dCi — vni a f — Vn2 n r vn3 n r — Y™4 — Vns p — Vns n
i at ~ A/=l J J ^fc=0 Vifc "I" Zjfc=o Vfci £jm=iJm L,v=i rp L,q=\uq
where V, is volume of zone i (m3)
Ci is air concentration in zone i ((jg/m3)
t is elapsed time (h)
Aj is area of source j in zone i (m2)
Ej is emission factor for source j in zone i ((jg/m2/h)
Oik is air flow from zone i to zone k,ifk (m3/h)
Oki is air flow from zone k to zone i, k f i (m3/h)
Ck is air concentration in zone k ([j,g/m3)
Sm is sorption rate onto interior surface m in zone i ((J-g/h)
Pp is rate of sorption by airborne particulate matterp in zone i ((J-g/h)
Dq is rate of sorption by settled dust q in zone i ((J-g/h)
Subscripts j, k, I, m,p, and q are summation counters
m through fi6 are item numbers for their respective summations.
Equation_Apx G-3 states that the change of the concentration in air in zone i is determined by six
factors: (1) the emissions from the sources in the zone, (2) the rate of chemical removed from zone i by
the ventilation and interzonal air flows (Oik), (3) the rate of chemical carried into zone i by the
infiltration and interzonal air flows (Oki), (4) the rate of chemical sorption by interior surfaces, (5) the
rate of chemical sorption by airborne particles, and (6) the rate of chemical sorption by settled dust.
Given a set of initial conditions, Equation_Apx G-3 can be solved numerically.
Equation_Apx G-3 does not include the term for chemical reactions because HBCD is chemically inert
at normal temperatures. Also the air concentrations in Equation_Apx G-3, G and Ck can be used to
represent either the gas-phase or particle-phase concentrations or both.
Page 646 of 723
-------
G.5.2 Typical Residential House
A three-zone configuration described by (Bevington et al. 2017) was used to represent a generic
residential building, where the insulation is applied to both the attic and crawlspace. The baseline
ventilation and interzonal air flows are shown in Figure_Apx G-lFigure_Apx G-l. The three-zone
configuration for a generic residential setting and baseline ventilation and interzonal air flows. The
ventilation rates for the three zones are shown in FigureApx G-l. In this work, EPA used the
ventilation rates for the "vented" attic and crawlspace.
Qo2 = 300 m3/h (vented)
Qq2 = 105 m3/h (unvented)
Zone 2 (Attic)
V2 = 150 m3
t
supply air
return air
Q3hl = 10 m3/h (leakage to the return flow duct)
HVAC
h(vented)
(unvented)
Zone 3 (Crawlspace)
V3 = 150 m3
Zone 1 (Living Space)
Vj = 300 m3
Q21 = 15 m3/h (vented)
Q21 = 5 m3/h (unvented)
Q31 = 15 m3/h (vented)
Q31 = 5 m3/h (unvented)
Figure Apx G-l. The three-zone configuration for a generic residential setting and baseline
ventilation and interzonal air flows.
Table Apx G-26. Zone Names, Volumes, and Baseline Ventilation Rates
Zone name
Zone volume (ra3)
Ventilation rate (h"1)
Living space
300
0.5
Attic
150
2.0 (vented)
0.7 (unvented)
Crawlspace
150
1.0 (vented)
0.35 (unvented)
G.5.3 Typical Passenger Vehicle
EPA used 3.4 m3 as the typical interior volume of a small SUV (passenger volume plus cargo volume).
The in-vehicle ventilation rate can be drastically different depending on factors such as whether the
vehicle is moving, how the AC operates, and vehicle type and age. A study by (Ott et al. 2008) shows
that, with a vehicle moving, windows closed, and the ventilation system off (or the air conditioner set to
AC Max), the air change rate was less than 6.6 h"1 for speeds ranging from 20 to 72 mph (32 to 116
km/h).
In this work EPA assume the air change rate is 5 h"1 for a moving vehicle with windows closed, and 0.5
h"1 for a stationary vehicle with windows closed.
Page 647 of 723
-------
For a moving vehicle with the AC on, EPA assumes the temperature inside the cabin is constant and at
21 °C.
For a stationary vehicle, EPA assume its temperature is subject to diurnal fluctuation, as defined by the
following parameters:
Daily average 20 °C
Daily fluctuation ±15 °C
Peak temperature occurrence 2:00 pm
G.5.4 Estimation of Key Parameters
Material/air partition coefficient (iT)
EPA has been unable to find experimentally determined material/air partition coefficients for HBCD in
insulation boards. In this evaluation, EPA estimatedKfrom EquationApx G-4 (Guo_2002):
EquationApx G-4
In K = 9.76- 0.785 In P
where P is the vapor pressure, mm Hg.
The K values obtained from Equation_Apx G-5 was then adjusted by the density of the foam material
(Equation 3):
Equation Apx G-5
K' = K —
Po
where
K' is the partition coefficient for the foam board, dimensionless,
K is the partition coefficient for the neat polymer, dimensionless,
p is the density of the foam, g/cm3,
po is the density of the neat polymer, g/cm3; po = 1.05 for polystyrene polymer.
The temperature dependence of the partition coefficient was estimated by the method proposed by (Tian
etal. 2017V
Equation Apx G-6
In— = a —
r \t2 tJ
where
Ki, K2 are partition coefficients at temperatures Ti and T2 (dimensionless),
a is the absolute value of the slope for the ln(/Q-ln(/J) relationship, where P is vapor pressure.
AHV = vaporization enthalpy (J/mol),
'/'/, T2 = absolute temperature corresponding to Ki and K2 (K),
R = gas constant (J/mol/K).
Parameter a is reported to be between 0.753 and 1.05 for open-cell PU foam. In this work, EPA used a =
0.9 and AHV = 8.14x 104 J/mol (Tian etal. 2017Y
Page 648 of 723
-------
• Solid-phase diffusion coefficient (D)
A QSAR model developed by (Huang et al. 2017) was used to estimate the solid-phase diffusion
coefficient for the foam materials (
EquationApx G-7):
EquationApx G-7
T—3486
logD = 6.39 — 2.49logm + b H -—
where
m is the molecular weight of the chemical, g/mol,
b is an empirical constant that reflects the material type,
r is an empirical constant that reflects the temperature effect,
Tis temperature (K).
The values of b and t for polystyrene foams — including both XPS and EPS — are -8.323 and 1676,
respectively. The difference between XPS and EPS is discussed in the main Risk Evaluation document.
• Aerosol/air partition coefficient (Kp)
The aerosol/air partition coefficient was calculated from Equation F-8 (Finizio et al. 1997):
log Kp=m log Koa + b (F-8)
where
m and b are constant for a given chemical,
Koa is the octanol-air partition coefficient (dimensionless).
In this work, EPA used Koa = 2.92 x 1010 for HBCD (from EPA's EP1 Suite ("https://www.epa.gov/tsca-
screening-tools/epi-suitetm-estimation-program-interface). The m and b values for generic organic
compounds are m = 0.55, and b = 8.23 (Finizio et al. 1997). The resulting KP is 3.36 x 109 for HBCD.
• Dust/air partition coefficient (Ka)
The dimensionless dust/air partition coefficient was estimated with the empirical model developed by
(Shoeib et al. 2005):
Kd = 0.411 pf0CK0A (F-9)
where
p is the density of the dust, g/cm3,
foe is the organic carbon content in the dust, fraction,
Koa is the octanol/air partition coefficient, dimensionless.
Page 649 of 723
-------
G.5.5 Model Parameters
HBCD sources - polystyrene foam boards
EPA assume that the source areas are 180 m2 in the attic and 120 m2 in the crawlspace (Bevington et al.
2017). Other parameters are summarized in TableApx G-27. Parameters for the HBCD sources.
Table Apx G-27. Parameters for the HBCD sources.
Parameter
Value
Data source/method
Board thickness (cm)
10
FOAMULAR 400 specs
HBCD content
0.50%
(TJ.S. EPA 2014d)
Board density (kg/m3)
28.9
FOAMULAR 400 specs
Partition coef. (K) at 21 °C
1.70 x 107
Guo (2002); adjusted by foam density
K as a function of temperature
Equation 9
Tian et al. (2017)
Diffusion coef. (D) at 21 °C (m2/h)
3.20 x 10"12
(Huang et al. 2017)
D as a function of temperature
Equation 10
(Huang et al. 2017)
The parameters EPA used to represent the HBCD sources in passenger vehicles are the same as those in
Table Apx G-27 except that the source area is 0.5 m2 and that the HBCD content in the polymer is
2.5%.
• HBCD sinks - gypsum board walls
The indoor sinks in the living space are represented by the gypsum board walls. Parameters used are
shown in Table Apx G-28.
Table Apx G-28. Parameters for the HBCD sinks.
Parameter
Value
Data source/method
Surface area (m2)
800
Bevington et al. (2017)
Thickness (m)
0.01 (-3/8 inch)
Product specs
Partition coefficient (dimensionless)
5.88 x 108
Guo (2002)
Diffusion coefficient (m2/h)
1.08 x 10"9
(Huang et al. 2017)
• Airborne PM
For airborne particulate matter, EPA used the following parameters:
Particle size 2.5 |im
Mass concentration in ambient air 30 |ig/m3
Infiltration factor 0.8
Aerosol/air partition coefficient 3.36 x 109 (by the (Finizio et al. 1997) method)
Deposition rate constant 0.68 h"1 for the living area
0.60 for attic and crawlspace
• Settled dust
The parameters EPA used to model settled dust are presented in Table Apx G-29.
Page 650 of 723
-------
Table Apx G-29. Parameters for Settled Dust
Parameter
Value
Data source/method
Average diameter (|im)
50
Bevinston et al. (2017)
Dust loading (g/m2)
10
Bevinston et al. (2017)
Partition coefficient
2.90 x 109
Shoeib et al. (2005)
Diffusion coefficient (m2/h)
1.0 x 10"13
Estimated [1]
[1] The reported diffusion coefficient values for aerosol particles vary significantly. The value EPA used
is in the middle.
G.5.6 Simulation Results
• HBCD in a "typical" home
Simulation results are presented in FigureApx G-2 through FigureApx G-5. As shown in FigureApx
G-4, the predicted HBCD content in house dust is in line with the measured values in the literature.
TableApx G-30 presents the mass balance results at the 100 elapsed days.
The predicted emission rates (Figure Apx G-5, sorption rates (Figure Apx G-6) and the mass balance
(Table Apx G-30) were obtained with the new features recently added to IECCU.
0.12
0.08
0.034 ng/m
0.04
0.00
0
50
100
150
Elapsed Time (days)
200
250
300
350
400
Figure Apx G-2. Predicted Gas-phase HBCD Concentration in Living Area
Page 651 of 723
-------
300
250
euo
SP 200
150
113 ng/g
u 100
CO
50
0
0
50
100
150
200 250
Elapsed Time (days)
300
350
400
FigureApx G-3. Predicted HBCD Concentration in Airborne PM in Living Area
20
10.4 ng/g
CQ
5
0
0
50
100
150
Elapsed Time (days)
200
250
300
350
400
Figure Apx G-4. Predicted HBCD Concentration in Settled Dust
Page 652 of 723
-------
10,000
m 1,000
(O
cc
£
O
o
u
CO
100
10
PS foam in attic
PS foam in crawlspace
100 200
Elapsed Time (days)
300
400
FigureApx G-5. Predicted HBCD Emission Rates from Polystyrene Foam Boards in Attic and
Crawlspace
OO 20
Elapsed Time (days)
Figure Apx G-6. Rate of HBCD Sorption by Gypsum Board Walls
Table Apx G-30. Mass Balance Results for HBCD in the Simulated Home at 100 Elapsed Days
Emission/Fate
Mass
(US)
Percentage of
of emitted
Total HBCD Emitted
2.2E+06
HBCD
Fate
Vented out
2.1E+06
94.3%
Remaining in air
4.9E+02
0.02%
Absorbed by sinks
8.7E+04
4.0%
PM deposition
7.8E+03
0.4%
In dust
8.1E+03
0.4%
Total
2.2E+06
100%
Page 653 of 723
-------
• HBCD in passenger vehicles
The HBCD concentrations inside the cabin are shown in FigureApx G-7 and the concentrations in the
settled dust are shown in Figure Apx G-8. Note that we have assumed that all the dust particles are
freshly introduced and the initial HBCD concentration in the dust is zero.
Stationary
Moving
50 100 150 200 250
Elapsed Time (h)
Figure Apx G-7. Predicted HBCD Concentrations in Vehicle's Cabin
80
60
40
20
0
0
50
100
150
200
250
Elapsed Time (h)
Figure Apx G-8. Predicted HBCD Concentrations in the Settled Dust in Vehicle's Cabin. The
Dust Contained no HBCD Initially
G.5.7 Discussion
• XPS versus EPS foam boards
Page 654 of 723
-------
Extruded polystyrene (XPS) insulation is manufactured through an extrusion process, which produces a
closed-cell rigid insulation. In contrast, expanded polystyrene (EPS) insulation is manufactured using a
mold to contain small foam beads. Heat or steam is then applied to the mold, which causes the small
beads to expand and fuse together. This manufacturing process produces open-cell insulation (see
https://www.kingspan.com/meati/en-in/product-groups/insulation/knowledge-base/faqs/general/what-is-
the-difference-between-xps-and-eps).
The presence of interconnected voids in the EPS foam facilitates both heat and mass transfers in the
foam. According to website http://www.giasxps.ro/index.php/en/electronic-library-polvstvrene/77-xps-
eps-comparison. the resistances to water vapor diffusion are as follows:
• Air = 1
• EPS = 50-70
• XPS = 50-250
These numbers suggest that the solid-phase diffusion coefficient for the low-performance XPS foam is
about the same as that for the EPS foam and that the diffusion coefficient for the high-performance XPS
foam can be as small as one fourth to one fifth of that for the EPS foam.
In (Huang et al. 20171 the XPS and EPS foams are lumped into a single material type. To evaluate the
difference in HBCD emissions between XPS and EPS, EPA conducted several simulations in a single-
zone setting (i.e., a test chamber) by varying only the solid-phase diffusion coefficient:
Table Apx G-31. Parameters Used in Comparing EPS and XPS Foams
Parameter
Value
Diffusion coef. predicted bv (Huang et al. 2017):
3.2 x 10"12 (m2/h) at 21 °C
Diffusion coef. used in the simulations:
1 x 10"12 and 5 x 10"12 (m2/h)
Chamber volume
30 m3
Ventilation rate
0.5 h"1
Source area
5 m2
Source thickness
10 cm
Board density
28.9 kg/m3
HBCD content
0.50% (equivalent to 1.45 x 108 |ig/m3)
Partition coef.
1.70 x 107 at 21 °C
Gas-phase mass transfer coefficient
1 m/h
As shown in Figure_Apx G-9, when D increases by a factor of 5 from 1 x 10"12 to 5 x 10"12 m2/h, the
average concentration over a year increases from 0.49 to 0.84 |ig/m3, an increase by a factor of 1.7.
These results suggest that, if the XPS and EPS boards have the same HBCD content and the same
density, then the emission from EPS boards can be twice as much as the emissions from high-
performance XPS boards. However, the emission from the low-performance XPS boards is expected to
be similar to that from the EPS boards.
Page 655 of 723
-------
2.5
2.0
D = 1.0E-12 m2/h
D = 5.0E-12 m2/h
1.5
1.0
0.5
0.0
0
50
100
150
200
250
300
350
400
Elapsed Time (h)
FigureApx G-9. Simulated HBCD Concentrations with Different Solid-phase Diffusion
Coefficients
1. Effect of temperature on HBCD emission rates
The temperature dependence of HBCD emission rate from polystyrene foam boards is affected by both
the partition and diffusion coefficients (K and D). In this work, the temperature dependent K and D were
calculated from existing empirical models. To determine whether the models we used can reasonably
predict the temperature dependence of the emission rate, we compared our simulation results with those
in the 2012 report by Chemicals Evaluation and Research Institute, Japan
(http://www.meti.go.ip/meti lib/report/2012fy/E001880.pdf).
To make the data comparable, we normalized the emission rates according to:
RT
JVff =
R
TO
where
Nr = normalized emission/diffusion rate (dimensionless)
Rt = emission rate at temperature |ig/m2/h,
Rto = emission rate at reference temperature To, |ig/m2/h.
The single-zone model described was used to generate the HBCD emission rates. The temperature-
dependent Ks and Ds were estimated.
As shown in Figure Apx G-10, the predicted emission rates in this work are in good agreement with the
data reported by the Japanese researchers (Kataoka et al. 2012).
Page 656 of 723
-------
Japan XPS #12
Japan XPS #13
Japan XPS #14
Japan XPS #15
~ This work
20
30 40
Temperature (°C)
50
60
FigureApx G-10. Comparison of Normalized Emission Rates
The four dotted lines are from Tables 3-2-25 and 3-2-26 in the Japanese report. The reference
temperature is To = 28 °C.
2. "Faced" versus "unfaced" insulation boards
The simulation results presented above are applicable to "unfaced" insulation boards and boards with a
permeable facer (e.g., paper and fabrics). The results are not applicable to the boards with both sides
covered with a nonpermeable facer such as foil. It is our understanding that most sheathing insulation
boards on the market have one side covered by foil. When installed, the foil side faces the exterior of the
building.
Page 657 of 723
-------
Appendix H ENVIRONMENTAL HAZARDS
H.l Supplemental Environmental Hazard Information
See Supplemental Document:
Risk Evaluation for Cyclic Aliphatic Bromide Cluster (HBCD), Sys
File: Data Extraction Tables of Environmental Hazard Studies. ( .019b)
H.2 Calculations Used to Evaluate the Potential Trophic Transfer of
HBCD
The below calculations were used to calculate food and HBCD ingestion, as presented above in Table
3-2 and Table 3-3
Legend:
Cpredator: Amount of food consumed by predator
BWpredatorPredator body weight
Equation 1: Calculation used to quantify food ingestion by a predator
amount of food consumed by predator ^ . . .. , . amount of food consumed by predator
— — * % food type in predator diet * BWnrpdntnr = — —
BWpredator*day ' y predator d
Equation 2: Calculation used to quantify HBCD ingestion by a predator
g food consumed by predator ng HBCD amount of HBCD consumed by predator
* =
d qfood d
Equation 3: Calculation used to quantify allometrically-scaled osprey reproductive LOEC based
on kestrel reproductive LOEC (Fernie et al., 2011)
Kestrel Reproductive LOEC PldJiBCD\ HBCD
—— = Osprey Reproductive LOEC ( —)
Osprey BW (g) g BW per day
nr\ ion HBCD
70,380 ng , 40.8 ng HBCD
1/725^ = gBWperday = °Sprey L0EC
Page 658 of 723
-------
H.3 KABAM Outputs for Aquatic HBCD Bioaccumulation and Bioconcentration
H.3.1 10th Percentile Surface and Pore Water Concentrations
The outputs from KABAM (vl) are provided below, per sub-scenario- and HBCD-specific release information and physiochemical properties, respectively. Both sub-scenarios
(3.3 and 5.7) are modeled with the assumption that the releases and subsequent surface water and pore water concentrations are based on a 75% removal of HBCD from the
direct releases of HBCD into surface water. The outputs below are also based on the HBCD half-life of 128 days. Further information regarding the trophic level designations
and calculations for the output parameters are available at: https://www.epa.gOv/sites/production/files/2015-07/documents/kabam_vl_0_users_guide.pdf.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Trophic
Level
Total
concentration
(Hg/kg-ww)
Lipid
normalized
concentration
(Hg/kg-lipid)
Contribution
due to diet
(Hg/kg-ww)
Total BCF
(M«/kg-
ww)/(|ig/L)
Total BAF
(M«/kg-
ww)/(|ig/L)
BCF
(Mg/kg-
lipid)/(ng/L)
BAF
(M«/kg-
lipid)/(fxg/L)
BMF
(M«/kg-
lipid)/(ng/kg-
lipid)
BSAF
(M«/kg-
lipid)/(ng/kg-
OC)
Phytoplankton
18676.00
933799.77
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
31.98
Zooplankton
16672.34
555744.59
1745.38
14257.78
15581.62
475259.43
519387.47
0.60
19.03
100,000
Benthic
Invertebrates
19749.81
658327.12
4696.47
15004.75
18457.77
500158.39
615258.99
1.34
22.55
Filter Feeders
12943.12
647155.76
3029.45
9862.90
12096.37
493144.94
604818.47
1.32
22.16
Small Fish
38714.05
967851.24
20981.21
19303.06
36181.35
482576.58
904533.87
1.59
33.15
Medium Fish
63361.90
1584047.59
47561.14
19303.06
59216.73
482576.58
1480418.31
1.95
54.25
Processing:
Large Fish
154955.55
3873888.81
140823.26
20031.30
144818.27
500782.60
3620456.83
2.45
132.67
Manufacturing
ofXPS Foam
using XPS
3.3
Phytoplankton
9425.27
471263.44
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
32.06
Zooplankton
8414.08
280469.23
880.84
14257.78
15581.62
475259.43
519387.47
0.60
19.08
Masterbatch
50,000
Benthic
Invertebrates
9966.74
332224.55
2369.99
15004.23
18456.92
500140.87
615230.66
1.34
22.60
Filter Feeders
6531.74
326586.99
1528.76
9862.55
12095.81
493127.67
604790.71
1.32
22.22
Small Fish
19537.35
488433.84
10588.38
19302.39
36180.28
482559.67
904507.12
1.59
33.23
Medium Fish
31975.91
799397.82
24001.97
19302.39
59214.65
482559.67
1480366.33
1.95
54.38
Large Fish
78199.37
1954984.30
71067.19
20031.30
144813.65
500782.60
3620341.30
2.45
132.99
25,000
Phytoplankton
4695.18
234759.01
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
31.94
Zooplankton
4191.46
139715.23
438.79
14257.78
15581.62
475259.43
519387.47
0.60
19.01
Page 659 of 723
-------
The outputs from KABAM (vl) are provided below, per sub-scenario- and HBCD-specific release information and physiochemical properties, respectively. Both sub-scenarios
(3.3 and 5.7) are modeled with the assumption that the releases and subsequent surface water and pore water concentrations are based on a 75% removal of HBCD from the
direct releases of HBCD into surface water. The outputs below are also based on the HBCD half-life of 128 days. Further information regarding the trophic level designations
and calculations for the output parameters are available at: https://www.epa.gOv/sites/production/files/2015-07/documents/kabam_vl_0_users_guide.pdf.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Trophic
Level
Total
concentration
(Hg/kg-ww)
Lipid
normalized
concentration
(Hg/kg-lipid)
Contribution
due to diet
(Hg/kg-ww)
Total BCF
(M«/kg-
ww)/(|ig/L)
Total BAF
(M«/kg-
ww)/(|ig/L)
BCF
(Mg/kg-
lipid)/(ng/L)
BAF
(M«/kg-
lipid)/(fxg/L)
BMF
(M«/kg-
lipid)/(ng/kg-
lipid)
BSAF
(M«/kg-
lipid)/(ng/kg-
OC)
Benthic
Invertebrates
4965.25
165508.48
1180.75
15005.01
18458.19
500167.14
615273.14
1.34
22.52
Filter Feeders
3254.00
162699.89
761.64
9863.07
12096.65
493153.56
604832.32
1.32
22.14
Small Fish
9732.93
243323.21
5274.78
19303.40
36181.89
482585.01
904547.23
1.59
33.11
Medium Fish
15929.58
398239.51
11957.17
19303.40
59217.77
482585.01
1480444.27
1.95
54.18
Large Fish
38956.74
973918.40
35403.85
20031.30
144820.58
500782.60
3620514.51
2.45
132.51
Processing:
Manufacturing
of EPS Foam
from Imported
EPS Resin
beads
5.7
100,000
Phytoplankton
774146.19
38707309.59
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
10.90
Zooplankton
691091.77
23036392.32
72348.24
14257.78
15581.62
475259.43
519387.47
0.60
6.49
Benthic
Invertebrates
848124.80
28270826.57
207064.12
15415.43
19122.15
513847.74
637405.06
1.39
7.96
Filter Feeders
555749.04
27787451.92
133566.50
10132.85
12530.13
506642.33
626506.71
1.36
7.83
Small Fish
1641844.84
41046121.09
886675.31
19831.39
37017.67
495784.71
925441.82
1.60
11.56
Medium Fish
2698514.01
67462850.15
2025623.68
19831.39
60841.75
495784.71
1521043.68
1.95
19.00
Large Fish
6583311.56
164582789.06
5997508.21
20031.30
148429.90
500782.60
3710747.62
2.44
46.36
50,000
Phytoplankton
708116.95
35405847.41
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
17.97
Zooplankton
632146.49
21071549.53
66177.44
14257.78
15581.62
475259.43
519387.47
0.60
10.70
Benthic
Invertebrates
759699.33
25323311.02
182639.83
15170.33
18725.64
505677.82
624188.10
1.36
12.85
Filter Feeders
497845.00
24892249.96
117811.64
9971.74
12271.26
498586.97
613562.98
1.34
12.64
Small Fish
1481557.62
37038940.43
801781.47
19516.08
36518.55
487901.98
912963.78
1.60
18.80
Medium Fish
2429003.97
60725099.21
1823292.57
19516.08
59871.92
487901.98
1496798.11
1.95
30.82
Large Fish
5934354.51
148358862.71
5398516.08
20031.30
146274.45
500782.60
3656861.29
2.44
75.31
Page 660 of 723
-------
The outputs from KABAM (vl) are provided below, per sub-scenario- and HBCD-specific release information and physiochemical properties, respectively. Both sub-scenarios
(3.3 and 5.7) are modeled with the assumption that the releases and subsequent surface water and pore water concentrations are based on a 75% removal of HBCD from the
direct releases of HBCD into surface water. The outputs below are also based on the HBCD half-life of 128 days. Further information regarding the trophic level designations
and calculations for the output parameters are available at: https://www.epa.gOv/sites/production/files/2015-07/documents/kabam_vl_0_users_guide.pdf.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Trophic
Level
Total
concentration
(Hg/kg-ww)
Lipid
normalized
concentration
(Hg/kg-lipid)
Contribution
due to diet
(Hg/kg-ww)
Total BCF
(M«/kg-
ww)/(|ig/L)
Total BAF
(M«/kg-
ww)/(|ig/L)
BCF
(Mg/kg-
lipid)/(ng/L)
BAF
(M«/kg-
lipid)/(fxg/L)
BMF
(M«/kg-
lipid)/(ng/kg-
lipid)
BSAF
(M«/kg-
lipid)/(ng/kg-
OC)
25,000
Phytoplankton
677746.64
33887331.89
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
25.48
Zooplankton
605034.46
20167815.34
63339.17
14257.78
15581.62
475259.43
519387.47
0.60
15.16
Benthic
Invertebrates
720120.12
24004003.88
171865.05
15058.95
18545.46
501965.16
618181.92
1.35
18.05
Filter Feeders
471925.04
23596251.88
110861.37
9898.53
12153.62
494926.38
607680.97
1.32
17.74
Small Fish
1409208.09
35230202.20
763363.57
19372.79
36291.74
484319.84
907293.39
1.60
26.49
Medium Fish
2307713.82
57692845.46
1732237.04
19372.79
59431.21
484319.84
1485780.21
1.95
43.38
Large Fish
5641802.93
141045073.25
5128945.99
20031.30
145294.95
500782.60
3632373.76
2.44
106.05
Page 661 of 723
-------
H.3.2 50th Percentile Surface and Pore Water Concentrations
The outputs from KABAM (vl) are provided below, per sub-scenario- and HBCD-specific release information and physiochemical properties, respectively. Both sub-scenarios
(3.3 and 5.7) are modeled with the assumption that the releases and subsequent surface water and pore water concentrations are based on a 75% removal of HBCD from the
direct releases of HBCD into surface water. The outputs below are also based on the HBCD half-life of 128 days. Further information regarding the trophic level designations
and calculations for the output parameters are available at: https://www.epa.gov/sites/production/files/2015-07/documents/kabam vl 0 users guide.pdf.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Trophic
Level
Total
concentration
(Hg/kg-ww)
Lipid
normalized
concentration
(Hg/kg-lipid)
Contribution
due to diet
(Hg/kg-ww)
Total BCF
(M«/kg-
ww)/(|ig/L)
Total BAF
(Mg/kg-
ww)/(|ig/L)
BCF
(Mg/kg-
lipid)/(ng/L)
BAF
(M«/kg-
lipid)/(fxg/L)
BMF
(M«/kg-
lipid)/(ng/kg-
lipid)
BSAF
(M«/kg-
lipid)/(ng/kg-
OC)
Processing:
Manufacturing
ofXPS Foam
using XPS
Masterbatch
3.3
100,000
Phytoplankton
453.81
22690.46
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
30.66
Zooplankton
405.12
13504.07
42.41
14257.78
15581.62
475259.43
519387.47
0.60
18.25
Benthic
Invertebrates
480.29
16009.52
114.28
15013.87
18472.53
500462.49
615750.96
1.34
21.63
Filter Feeders
314.76
15737.81
73.72
9868.90
12106.01
493444.78
605300.26
1.32
21.27
Small Fish
941.20
23529.96
510.04
19314.80
36199.93
482869.99
904998.33
1.59
31.80
Medium Fish
1540.57
38514.34
1156.40
19314.80
59252.83
482869.99
1481320.79
1.95
52.05
Large Fish
3767.36
94184.03
3423.96
20031.30
144898.50
500782.60
3622462.60
2.45
127.28
50,000
Phytoplankton
232.14
11607.04
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
29.02
Zooplankton
207.24
6907.85
21.69
14257.78
15581.62
475259.43
519387.47
0.60
17.27
Benthic
Invertebrates
245.95
8198.50
58.57
15026.44
18492.85
500881.26
616428.42
1.34
20.50
Filter Feeders
161.19
8059.32
37.78
9877.15
12119.27
493857.67
605963.72
1.32
20.15
Small Fish
481.80
12044.98
261.06
19330.96
36225.52
483274.03
905637.92
1.59
30.11
Medium Fish
788.72
19718.10
592.04
19330.96
59302.54
483274.03
1482563.54
1.95
49.30
Large Fish
1928.62
48215.49
1752.96
20031.30
145008.99
500782.60
3625224.65
2.45
120.54
25,000
Phytoplankton
116.94
5847.16
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
31.44
Zooplankton
104.40
3479.90
10.93
14257.78
15581.62
475259.43
519387.47
0.60
18.71
Benthic
Invertebrates
123.71
4123.56
29.42
15008.42
18463.71
500280.74
615456.93
1.34
22.17
Page 662 of 723
-------
The outputs from KABAM (vl) are provided below, per sub-scenario- and HBCD-specific release information and physiochemical properties, respectively. Both sub-scenarios
(3.3 and 5.7) are modeled with the assumption that the releases and subsequent surface water and pore water concentrations are based on a 75% removal of HBCD from the
direct releases of HBCD into surface water. The outputs below are also based on the HBCD half-life of 128 days. Further information regarding the trophic level designations
and calculations for the output parameters are available at: https://www.epa.gov/sites/production/files/2015-07/documents/kabam vl 0 users guide.pdf.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Trophic
Level
Total
concentration
(Hg/kg-ww)
Lipid
normalized
concentration
(Hg/kg-lipid)
Contribution
due to diet
(Hg/kg-ww)
Total BCF
(M«/kg-
ww)/(|ig/L)
Total BAF
(Mg/kg-
ww)/(|ig/L)
BCF
(Mg/kg-
lipid)/(ng/L)
BAF
(M«/kg-
lipid)/(fxg/L)
BMF
(M«/kg-
lipid)/(ng/kg-
lipid)
BSAF
(M«/kg-
lipid)/(ng/kg-
OC)
Filter Feeders
81.07
4053.58
18.98
9865.31
12100.25
493265.58
605012.31
1.32
21.79
Small Fish
242.47
6061.63
131.40
19307.79
36188.83
482694.63
904720.75
1.59
32.59
Medium Fish
396.85
9921.24
297.89
19307.79
59231.26
482694.63
1480781.42
1.95
53.34
Large Fish
970.50
24262.47
882.01
20031.30
144850.55
500782.60
3621263.84
2.45
130.44
Phytoplankton
16511.67
825583.72
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
10.89
Zooplankton
14740.22
491340.54
1543.11
14257.78
15581.62
475259.43
519387.47
0.60
6.48
Benthic
Invertebrates
18090.59
603019.81
4416.88
15416.11
19123.25
513870.36
637441.65
1.39
7.96
100,000
Filter Feeders
11854.18
592709.25
2849.11
10133.29
12530.85
506664.63
626542.55
1.36
7.82
Small Fish
35020.03
875500.65
18912.39
19832.26
37019.05
495806.53
925476.37
1.60
11.55
Medium Fish
57558.83
1438970.82
43206.20
19832.26
60844.43
495806.53
1521110.81
1.95
18.98
Processing:
Manufacturing
ofEPS Foam
Large Fish
140420.34
3510508.39
127925.80
20031.30
148435.87
500782.60
3710896.82
2.44
46.31
5.7
Phytoplankton
16511.67
825583.72
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
19.20
from Imported
EPS Resin
beads
Zooplankton
14740.22
491340.54
1543.11
14257.78
15581.62
475259.43
519387.47
0.60
11.43
Benthic
Invertebrates
17677.48
589249.35
4243.20
15146.17
18686.55
504872.41
622885.15
1.36
13.70
50,000
Filter Feeders
11584.47
579223.47
2737.07
9955.86
12245.74
497792.86
612286.97
1.33
13.47
Small Fish
34500.00
862500.06
18674.42
19485.00
36469.35
487124.89
911733.67
1.60
20.06
Medium Fish
56548.40
1413709.92
42447.08
19485.00
59776.32
487124.89
1494407.95
1.95
32.88
Large Fish
138174.62
3454365.46
125680.09
20031.30
146061.96
500782.60
3651549.11
2.44
80.33
25,000
Phytoplankton
16511.67
825583.72
N/A
20010.63
17454.20
1000531.52
872710.07
N/A
26.89
Zooplankton
14740.22
491340.54
1543.11
14257.78
15581.62
475259.43
519387.47
0.60
16.00
Page 663 of 723
-------
The outputs from KABAM (vl) are provided below, per sub-scenario- and HBCD-specific release information and physiochemical properties, respectively. Both sub-scenarios
(3.3 and 5.7) are modeled with the assumption that the releases and subsequent surface water and pore water concentrations are based on a 75% removal of HBCD from the
direct releases of HBCD into surface water. The outputs below are also based on the HBCD half-life of 128 days. Further information regarding the trophic level designations
and calculations for the output parameters are available at: https://www.epa.gov/sites/production/files/2015-07/documents/kabam vl 0 users guide.pdf.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Trophic
Level
Total
concentration
(Hg/kg-ww)
Lipid
normalized
concentration
(Hg/kg-lipid)
Contribution
due to diet
(Hg/kg-ww)
Total BCF
(M«/kg-
ww)/(|ig/L)
Total BAF
(Mg/kg-
ww)/(|ig/L)
BCF
(Mg/kg-
lipid)/(ng/L)
BAF
(M«/kg-
lipid)/(fxg/L)
BMF
(M«/kg-
lipid)/(ng/kg-
lipid)
BSAF
(M«/kg-
lipid)/(ng/kg-
OC)
Benthic
Invertebrates
17522.56
584085.43
4178.07
15044.95
18522.79
501498.19
617426.46
1.34
19.03
Filter Feeders
11483.33
574166.30
2695.06
9889.32
12138.82
494465.95
606941.12
1.32
18.70
Small Fish
34304.99
857624.8
18585.18
19354.77
36263.21
483869.3
906580.2
1.594949
27.93566
Medium Fish
56169.48
1404237
42162.41
19354.77
59375.77
483869.3
1484394
1.948016
45.74062
Large Fish
137332.5
3433312
124837.9
20031.3
145171.7
500782.6
3629294
2.444966
111.8343
Page 664 of 723
-------
Appendix I BMD MODELING RESULTS FOR SELECTED PODs
1.1 Noncancer Endpoints for BMD Modeling
The noncancer endpoints that were selected for dose-response modeling are presented in Table Apx 1-1
For each endpoint, the doses and response data used for the modeling are presented.
Table Apx 1-1. Noncancer Endpoints Selected for Dose-response Modeling for HBCD
Endpoint
Species
(strain)/sex
Dose
(mg/kg-d)a
Incidence [%] or mean ± SD
(number of animals or litters)
BMR(s)
Thyroid
|T4
Ema et al. (2008)
F0 rats (CRL
Sprague-
Dawley)/male
0
10
101
1,008
TWA of lifetime exposure,
F0
4.04 : 1.42 (8)
3.98 ±0.89 (8)
2.97 ±0.76 (8)
2.49 ±0.55 (8)
10% RD, 15%
RD, 20% RD, 1
SD
|T4
Ema et al. (2008)
F0 rats (CRL
Sprague-
Dawley)/female
0
14
141
1,363
TWA of lifetime exposure,
F0
2.84 ±0.61 (8)
3.14 ±0.48 (8)
3.00 ±0.77 (8)
1.96 ±0.55 (8)
10% RD, 15%
RD,
20% RD, 1 SD
|T4
Ema et al. (2008)
F1 rats (CRL
Sprague-
Dawley)/female
0
14.3
138
1,363
TWA of lifetime exposure,
F1
3.59 ± 1.08 (8)
3.56 ±0.53 (8)
3.39 ± 1.21 (8)
2.58 ±0.37 (8)
10% RD, 15%
RD,
20% RD, 1 SD
Liver
Relative liver
weight
Ema et al. (2008)
F1 rats (CRL
Sprague-
Dawley)/male
weanlings,
PND26
0
16.5
168
1,570
TWA of F0 gestational and
lactational doses
4.6 ±0.37 (23)
4.6 ±0.32 (21)
5.05 ± 0.32 (20)
6 ±0.44 (17)
10% RD, 1 SD
Relative liver
weight
Ema et al. (2008)
F1 rats (CRL
Sprague-
Dawley)/female
weanlings,
PND26
0
16.5
168
1,570
TWA of F0 gestational and
lactational doses
4.57 ± 0.35 (23)
4.59 ±0.28 (21)
5.02 ±0.32 (20)
6.07 ±0.36 (14)
10% RD, 1 SD
Relative liver
weight
Ema et al. (2008)
F1 rats (CRL
Sprague-
Dawley)/male
adults
0
11.4
115
1,142
3.27 ±0.18 (24)
3.34 ±0.26 (24)
3.37 ±0.25 (22)
3.86 ±0.28 (24)
10% RD, 1 SD
Page 665 of 723
-------
Endpoint
Species
(strain)/sex
Dose
(mg/kg-d)a
Incidence [%] or mean ± SD
(number of animals or litters)
BMR(s)
TWA of lifetime exposure,
F1
Relative liver
weight
Ema et al. (2008)
F1 rats (CRL
Sprague-
Dawley)/female
adults
0
14.3
138
1,363
TWA of lifetime exposure,
F1
4.18 ±0.42 (22)
4.39 ±0.44 (22)
4.38 ±0.47 (20)
5.05 ±0.50 (13)
10% RD, 1 SD
Relative liver
weight
Ema et al. (2008)
F2 rats (CRL
Sprague-
Dawley)/male
weanlings,
PND26
0
14.7
139
1,360
TWA of F1 gestational and
lactational doses
4.72 ±0.59 (22)
4.74 ±0.35 (22)
5.04 ±0.4 (18)
6.0 ±0.25 (13)
10% RD, 1 SD
Relative liver
weight
Ema et al. (2008)
F2 rats (CRL
Sprague-
Dawley)/female
weanlings, PND
26
0
14.7
139
1,360
TWA of F1 gestational and
lactational doses
4.70 ±0.27 (21)
4.70 ± 0.28 (22)
4.94 ±0.32 (20)
5.89 ±0.44 (13)
10% RD, 1 SD
Relative liver
weight
WIL Research
(2001)
Rats (Sprague-
Dawley)/male
0
100
300
1,000
2.709 ±0.1193 (10)
3.175 ± 0.2293 (10)
3.183 ±0.2653 (10)
3.855 ±0.1557 (9)
10% RD, 1 SD
Relative liver
weight WIL
Research (2001)
Rats (Sprague-
Dawley)/female
0
100
300
1,000
2.887 ±0.2062 (10)
3.583 ±0.2734 (10)
3.578 ±0.3454 (10)
4.314 ±0.2869 (10)
10% RD, 1 SD
Reproductive
Primordial follicles
Ema et al. (2008)
(supplemental)
F1 parental rat
(CRL Sprague-
Dawley)/female
0
9.6
96
941
The F0 adult female
gestational doses
316.3 ± 119.5 (10)
294.2 ±66.3 (10)
197.9 ±76.9 (10)
203.4 ±79.5 (10)
1% RD, 5% RD,
10% RD
Developmental
Offspring loss at
PND4
Ema et al. (2008)
F2 offspring rats
(CRL Sprague-
Dawley)
0
9.7
100
995
The F1 adult female
gestational doses
28/132 [21%]
26/135 [19.3%]
23/118 [19.5%]
47/120 [39.2%]
1% ER, 5% ER
Page 666 of 723
-------
Endpoint
Species
(strain)/sex
Dose
(mg/kg-d)a
Incidence [%] or mean ± SD
(number of animals or litters)
BMR(s)
Offspring loss at
PND21
Ema et al. (2008)
F2 offspring rats
(CRL Sprague-
Dawley)
0
19.6
179
1,724
The F1 adult female
lactational doses
11/70 [15.7%]
7/70 [10.0%]
18/64 [28.1%]
32/64 [50.0%]
1% ER, 5% ER
Pup weight during
lactation at PND
21
Ema et al. (2008)
F2 offspring rats
(CRL Sprague-
Dawley)/male
0
19.6
179
1,724
The F1 adult female
lactational doses
53 ± 12.6 (22)
56.2 ±6.7 (22)
54.1 ± 10.1 (18)
42.6 ±8.3 (13)
5% RD, 10%
RD,
0.5 SD, 1 SD
Pup weight during
lactation at PND
21
Ema et al. (2008)
F2 offspring rats
(CRL Sprague-
Dawley)/female
0
19.6
179
1,724
The F1 adult female
lactational doses
52 ± 10(21)
52.8 ±6.6 (22)
51.2 ± 10.8(20)
41.6 ±8.4 (13)
5% RD, 10%
RD,
0.5 SD, 1 SD
Delayed eye
opening,
(Ema et al. 2008)
F2 offspring rats
(CRL Sprague-
Dawley)/female
0
15
139
1,360
82.9 ±26.8 (21)
72.7 ± 37.7 (22)
53.8 ±40.3 (20)
48.1 ±42.0 (13)
5% ER, 10% ER
Delayed eye
opening
(Ema et al. 2008)
F2 offspring rats
(CRL Sprague-
Dawley)/male
0
15
139
1,360
72.7 ± 40.0 (22)
62.5 ± 40.6 (22)
47.2 ±44.8 (18)
33.9 ±34.7 (14)
NOAEL
aDoses were calculated as TWA doses using weekly average doses (in mg/kg-day) as reported in Table 10 of the
SiiDDlcmcntal Materials to Ema et al. (2008).
BMR = benchmark response; ER = extra risk; PND = postnatal day; RD = relative deviation; SD = standard deviation; T4 =
thyroxine; TWA = time-weighted average
1.1.1 Thyroid Effects
TableApx 1-2. Summary of BMD modeling results for T4 in FO parental male CRL Sprague-
Dawley rats exposed to HBCD by diet for 18 weeks (Ema et al. 2008); BMR = 10% RD from
control mean
Model3
Goodness of fit
BMDiord
(mg/kg-d)
BMDLiord
(mg/kg-d)
BMDisrd
(mg/kg-d)
BMDLisrd
(mg/kg-d)
Basis for model
selection
p-value
AIC
Exponential (M2)
Exponential (M3)b
0.0473
33.926
259
177
399
274
Of the models
without saturation
that provided an
adequate fit and a
valid BMDL
estimate, the
Exponential 4
model with
modeled variance
Exponential (M4)
Exponential (M5)c
0.742
29.933
23.9
6.99
39.1
11.5
Hill
0.949
29.829
14.4
3.21
25.6
5.66
Power"1
Polynomial 3oe
Polynomial 2of
0.0418
34.174
303
227
455
341
Page 667 of 723
-------
Model3
Goodness of fit
BMDiord
(mg/kg-d)
BMDLiord
(mg/kg-d)
BMDisrd
(mg/kg-d)
BMDLisrd
(mg/kg-d)
Basis for model
selection
was selected based
on lowest AIC
(BMDLs differed
by <3).
p-value
AIC
Linear
Model3
Goodness of fit
BMDiord
(mg/kg-d)
BMDLxird
(mg/kg-d)
BMDisd
(mg/kg-d)
BMDLi sd
(mg/kg-d)
p-value
AIC
Exponential (M2)
Exponential (M3)b
0.0473
33.926
548
376
866
511
Exponential (M4)
Exponential (M5)°
0.742
29.933
57.9
17.2
101
29.5
Hill
0.949
29.829
42.0
9.11
94.9
Error8
Power"1
Polynomial 3°e
Polynomial 2of
Linear
0.0418
34.174
607
454
906
595
3Modeled variance case presented (BMDS Test 2 /?-valuc = 0.0756, BMDS Test 3 /?-valuc = 0.553), selected model in bold;
scaled residuals for selected model for doses 0, 10.2, 101, and 1,008 mg/kg-day were -0.1665, 0.166, 0.03642, and -0.03619,
respectively.
bFor the Exponential (M3) model, the estimate of d was 1 (boundary). The models in this row reduced to the Exponential
(M2) model.
Tor the Exponential (M5) model, the estimate of d was 1 (boundary). The models in this row reduced to the Exponential
(M4) model.
dFor the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear model.
Tor the Polynomial 3° model, the b3 coefficient estimate was 0 (boundary of parameters space). The models in this row
reduced to the Polynomial 2° model. For the Polynomial 3° model, the b3 and b2 coefficient estimates were 0 (boundary of
parameters space). The models in this row reduced to the Linear model.
fFor the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in this row
reduced to the Linear model.
gBMD or BMDL computation failed for this model.
Data from Ema et al. (2008)
Exponential 4 Model, with BMRof 0.1 Re I. Dev. for the BMD and 0.95 Lower Confidence Limit for the BMDL
5.5
Exponential 4
5
4.5
4
3.5
3
2.5
2
0
200
400
600
800
1000
BMR = 10% RD from control mean; dose shown in mg/kg-day.
FigureApx 1-1. Plot of mean response by dose, with fitted curve for Exponential 4 Model, for T4
in F0 parental CRL Sprague-Dawley male rats exposed to HBCD by diet for 18 weeks (Ema et al.
2008).
Page 668 of 723
-------
Exponential 4 Model (Version: 1.10; Date: 01/12/2015)
The form of the response function is:
Model 4: Y[dose] = a * [c-(c-l) * exp(-b * dose)]
A modeled variance is fit
Benchmark Dose Computation
BMR = 10% RD
BMD = 23.8946
BMDL at the 95% confidence level = 6.99406
Parameter Estimates
Variable
Estimate
Default initial parameter values
lalpha
-3.94284
-3.54227
rho
2.98463
2.72754
a
4.1075
4.242
b
0.0123219
0.00282274
d
1 (specified)
1 (specified)
Table of Data and Estimated Values of Interest
Dose
N
Observed mean
Estimated mean
Observed SD
Estimated SD
Scaled residuals
0
8
4.04
4.11
1.42
1.15
-0.167
10.2
8
3.98
3.92
0.89
1.07
0.166
101
8
2.97
2.961
0.76
0.71
0.036
1,008
8
2.49
2.50
0.59
0.56
-0.036
Likelihoods of Interest
Model
Log (likelihood)
Number of parameters
AIC
A1
-12.76333
5
35.52665
A2
-9.319925
8
34.63985
A3
-9.91228
6
31.82456
fitted
-9.966286
5
29.93257
R
-19.64317
2
43.28634
Tests of Interest
Test
-2*log (likelihood ratio)
Test df
p-value
Test 1
20.65
6
0.002123
Test 2
6.887
3
0.07559
Test 3
1.185
2
0.553
Test 6a
0.108
1
0.7424
df = degree(s) of freedom
Page 669 of 723
-------
1.1.2 Liver Effects
TableApx 1-3. Summary of BMD modeling results for relative liver weight (g/100 g BW) in male
F1 CRL rats exposed to HBCD on GD 0-PND 26, dose TWA gestation through lactation (Ema et
Model3
Goodness of fit
BMDiord
(mg/kg-d)
BMDLiord
(mg/kg-d)
BMDisd
(mg/kg-d)
BMDLisd
(mg/kg-d)
Basis for model
selection
p-value
AIC
Exponential (M2)
Exponential (M3)b
0.00369
-70.405
599
533
488
417
Of the models that
provided an
adequate fit and a
valid BMDL
estimate, the
Exponential M4
constant variance
model was selected
Exponential
(M4)
0.606
-79.345
163
109
120
80.5
Exponential (M5)
N/A°
-77.611
169
111
157
82.0
Hill
N/A°
-77.611
169
104
156
75.4
Powerd
Polynomial 3oe
Polynomial 2of
Linear
0.00590
-71.344
548
480
440
371
based on lowest AIC
and visual fit.
aConstant variance case presented (BMDS Test 2 /j-value = 0.462), selected model in bold; scaled residuals for selected
model for doses 0, 16.5, 168, and 1,570 mg/kg-day were 0.3267, -0.3947, 0.05759, and -0.003788, respectively.
bFor the Exponential (M3) model, the estimate of d was 1 (boundary). The models in this row reduced to the Exponential
(M2) model.
°No available degrees of freedom to calculate a goodness-of-fit value.
dFor the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear model.
Tor the Polynomial 3° model, the b3 and b2 coefficient estimates were 0 (boundary of parameters space). The models in this
row reduced to the Linear model.
fFor the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in this row
reduced to the Linear model.
Data from Ema et al. (2008)
Exponential 4 Model, with BMR of 0.1 Rel. Dev. for the BMD and 0.95 Lower Confidence Limit for the BMDL
5.5
5
12:31 05/20 2016
BMR = 10% RD from control mean; dose shown in mg/kg-day.
Figure Apx 1-2. Plot of mean response by dose with fitted curve for Exponential (M4) model with
constant variance for relative liver weight (g/100 g BW) in F1 weanling male CRL Sprague-
Dawley rats exposed to HBCD on GD 0-PND 26, dose TWA gestation through lactation (Ema et
al. 2008).
Page 670 of 723
-------
Exponential Model (Version: 1.10; Date: 01/12/2015)
The form of the response function is: Y[dose] = a * [c-(c-l) * exp(-b * dose)]
A constant variance model is fit
Benchmark Dose Computation
BMR = 10% RD
BMD = 162.81
BMDL at the 95% confidence level = 108.569
Parameter Estimates
Variable
Estimate
Default initial parameter values
lnalpha
-2.07833
-2.08162
rho
N/A
0
a
4.5759
4.37
b
0.00230233
0.00120199
c
1.3199
1.44165
d
N/A
1
Table of Data and Estimated Values of Interest
Dose
N
Observed mean
Estimated mean
Observed SD
Estimated SD
Scaled residuals
0
23
4.6
4.576
0.37
0.3538
0.3267
16.5
21
4.6
4.63
0.32
0.3538
-0.3947
168
20
5.05
5.045
0.32
0.3538
0.05759
1,570
17
6
6
0.44
0.3538
-0.003788
Likelihoods of Interest
Model
Log (likelihood)
Number of parameters
AIC
A1
43.80548
5
-77.61096
A2
45.09301
8
-74.18602
A3
43.80548
5
-77.61096
R
-5.569318
2
15.13864
4
43.67234
4
-79.34469
Tests of Interest
Test
-2*log (likelihood ratio)
Test df
p-value
Test 1
101.3
6
<0.0001
Test 2
2.575
3
0.4619
Test 3
2.575
3
0.4619
Test 6a
0.2663
1
0.6058
Page 671 of 723
-------
TableApx 1-4. Summary of BMD modeling results for relative liver weight (g/100 g BW) in male
CRL Sprague-Dawley rats exposed to HBCD by gavage for 13 weeks (WIL Research 2001); BMR
= 10% RD from control mean and 1 SD change from control mean
Goodness of fit
BMDiord
(mg/kg-d)
BMDLiord
(mg/kg-d)
BMDisd
(mg/kg-d)
BMDLisd
(mg/kg-d)
Basis for model
Model3
/?-\aluc
AIC
selection
Modeled with constant variance
No model showed
Exponential (M2)
Exponential
(M3)b
3.14 x
10-4
-67.830
328
283
269
219
adequate fit. Dropping
highest dose is not
expected to help in this
case.
Exponential
(M4)°
3.92 x
10-4
-69.396
164
97.7
128
77.9
Exponential
(M5)d
3.92 x
10-4
-69.396
164
97.7
128
77.9
Hill
4.91 x
10-4
-69.815
145
74.8
113
59.7
Power6
Polynomial 3of
Polynomial 2og
Linear
5.14 x
10-4
-68.817
290
244
234
187
Modeled with modeled variance
Exponential (M2)
Exponential
(M3)b
0.00119
-68.721
337
295
320
245
Exponential
(M4)°
5.50 x
10-4
-68.244
204
103
187
67.5
Exponential
(M5)d
5.50 x
10-4
-68.244
204
103
187
67.5
Hill
5.84 x
10-4
-68.355
192
35.9
173
106
Power6
Polynomial 3of
Polynomial 2og
Linear
0.00161
-69.324
299
256
282
210
aConstant variance (BMDS Test 2 /j-value = 0.0644, BMDS Test 3 /rvalue = 0.0644) and nonconstant variance cases
presented, no model was selected as a best-fitting model.
bFor the Exponential (M3) model, the estimate of d was 1 (boundary). The models in this row reduced to the Exponential
(M2) model.
The Exponential (M4) model may appear equivalent to the Exponential (M5) model; however, differences exist in digits not
displayed in the table.
dThe Exponential (M5) model may appear equivalent to the Exponential (M4) model; however, differences exist in digits not
displayed in the table.
Tor the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear model.
fFor the Polynomial 3° model, the b3 and b2 coefficient estimates were 0 (boundary of parameters space). The models in this
row reduced to the Linear model.
gFor the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in this row
reduced to the Linear model.
Page 672 of 723
-------
TableApx 1-5. Summary of BMD modeling results for relative liver weight (g/100 g BW) in
female CRL Sprague-Dawley rats exposed to HBCD by gavage for 13 weeks (WIL Research
2001): BMR =
0% RD from con
trol mean and 1 SD c
lange from control mean
Model3
Goodness of fit
BMDLiord
(mg/kg-d)
BMDisd
(mg/kg-d)
BMDLisd
(mg/kg-d)
Basis for model selection
/?-\aluc
AIC
(mg/kg-d)
Modeled with constant variance
No model showed
adequate fit. Dropping
highest dose is not
expected to help in this
case
Exponential (M2)
Exponential
(M3)b
<0.0001
-39.545
310
261
332
267
Exponential (M4)
Exponential
(M5)°
2.59 x
10-4
-44.035
101
56.0
106
61.8
Hill
5.71 x
10-4
-45.515
69.3
30.6
73.3
34.6
Power"1
Polynomial 3°e
Polynomial 2of
Linear
<0.0001
-40.679
270
220
287
226
Modeled with modeled variance
Exponential (M2)
Exponential
(M3)b
<0.0001
-38.793
319
269
374
282
Exponential (M4)
Exponential
(M5)°
1.72 x
10-4
-42.217
53.4
28.5
38.3
16.0
Hill
0.00115
-45.763
39.2
20.7
26.0
11.6
Powerd
Polynomial 3oe
Polynomial 2of
Linear
<0.0001
-39.727
278
227
327
237
aConstant variance (BMDS Test 2 /j-value = 0.461, BMDS Test 3 /j-value = 0.461) and nonconstant variance presented; no
model was selected as a best-fitting model.
bFor the Exponential (M3) model, the estimate of d was 1 (boundary). The models in this row reduced to the Exponential
(M2) model.
Tor the Exponential (M5) model, the estimate of d was 1 (boundary). The models in this row reduced to the Exponential
(M4) model.
dFor the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear model.
Tor the Polynomial 3° model, the b3 and b2 coefficient estimates were 0 (boundary of parameters space). The models in this
row reduced to the Linear model.
fFor the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in this row
reduced to the Linear model.
Page 673 of 723
-------
1.1.3 Reproductive Effects
Reduced Primordial Follicles
TableApx 1-6. Summary of BMD modeling results for primordial follicles in F1 parental female
CRL Sprague-Dawley rats exposed to HBCD by diet for 18 weeks (Ema et al. 2008); BMR = 1%
RD from control mean, 5% RD from control mean, and 10% RD from control mean
Model3
Goodness of fit
BMDird
(mg/kg-d)
BMDLird
(mg/kg-d)
BMDsrd
(mg/kg-d)
BMDLsrd
(mg/kg-d)
BMDiord
(mg/kg-d)
BMDLiord
(mg/kg-d)
Basis for
model
selection
/?-\aluc
AIC
Exponential
(M2)
Exponential
(M3'b
0.0130
408.57
26.8
13.9
137
71.0
281
146
Exponential
M4 constant
variance
selected as
only model
with
adequate fit.
Exponential
(M4)
0.688
402.05
0.883
0.252
4.67
1.33
10.1
2.87
Exponential
(M5)
N/A°
403.91
4.09
0.259
8.23
1.37
11.4
2.95
Hill
N/A°
403.91
8.00
error"1
9.28
1.10
9.99
2.50
Power6
Polynomial 2of
Linear
Polynomial 3°g
0.0117
408.78
33.1
19.8
165
99.0
331
198
"¦Constant variance case presented (BMDS Test 2 /j-value = 0.242), selected model in bold; scaled residuals for selected
model for doses 0, 9.6, 96.3, and 940.7 mg/kg-day were -0.129, 0.1915, -0.2611, and 0.1987, respectively.
bFor the Exponential (M3) model, the estimate of d was 1 (boundary). The models in this row reduced to the Exponential
(M2) model.
°No available degrees of freedom to calculate a goodness-of-fit value.
dBMD or BMDL computation failed for this model.
Tor the Power model, the power parameter estimate was 1. The models in this row reduced to the Linear model.
fFor the Polynomial 2° model, the b2 coefficient estimate was 0 (boundary of parameters space). The models in this row
reduced to the Linear model.
gThe Polynomial 3° model may appear equivalent to the Linear model; however, differences exist in digits not displayed in
the table.
Page 674 of 723
-------
Data from Ema et al. (2008)
Exponential 4 Model, with BMR of 0.1 Rel. Dev. for the BMD dnd 0.95 Lower Confidence Limit for the EMIX
0 200 400 600 800 1000
dose
BMR = 10% RD from control mean; dose shown in mg/kg-day.
FigureApx 1-3. Plot of mean response by dose, with fitted curve for Exponential M4, for
primordial follicles in F1 parental female CRL Sprague-Dawley rats exposed to HBCD by diet for
18 weeks (Ema et al. 2008).
Exponential Model (Version: 1.9; Date: 01/29/2013)
The form of the response function is: Y[dose] = a * [c-(c-l) * exp(-b * dose)]
A constant variance model is fit
Benchmark Dose Computation
BMR = 10% RD
BMD = 10.1143
BMDL at the 95% confidence level = 2.86589
Parameter Estimates
Variable
Estimate
Default initial parameter values
lnalpha
8.85121
8.84717
rho(S)
N/A
0
a
319.71
332.115
b
0.0301725
0.0026785
c
0.619779
0.567503
d
1
1
Table of Data and Estimated Values of Interest
Dose
N
Observed mean
Estimated mean
Observed SD
Estimated SD
Scaled residuals
0
10
316.3
319.7
119.5
83.56
-0.129
9.6
10
294.2
289.1
66.3
83.56
0.1915
96.3
10
197.9
204.8
76.9
83.56
-0.2611
940.7
10
203.4
198.1
79.5
83.56
0.1987
Page 675 of 723
-------
Likelihoods of Interest
Model
Log (likelihood)
Number of parameters
AIC
Al
-196.9435
5
403.8869
A2
-194.8505
8
405.701
A3
-196.9435
5
403.8869
R
-203.7104
2
411.4207
4
-197.0241
4
402.0483
Tests of Interest
Test
-2*log (likelihood ratio)
Test df
p-value
Test 1
17.72
6
0.006972
Test 2
4.186
3
0.2421
Test 3
4.186
3
0.2421
Test 6a
0.1613
1
0.6879
1.1.4 Developmental Effects
Offspring Loss
TableApx 1-7. Summary of BMD modeling results for offspring loss from PND 4 through PND 21
in F2 offspring CRL Sprague-Dawley rats; lactational doses of F1 dams (Ema et al. 2008); BMR =
1% ER and 5% ER
Model3
Goodness of Fit
p-value | AIC
BMDier
(mg/kg-d)
BMDLier
(mg/kg-d)
BMDser
(mg/kg-d)
BMDLser
(mg/kg-d)
Basis for model selection
Litter-specific covariate = implantation size; intra-litter correlations estimated
Of the models that provided an
adequate fit, a valid BMDL
estimate and BMD/BMDL <5,
the Nested Logistic model
(litter-specific covariate not
used; intra-litter correlations
estimated) was selected based
on lowest AIC (BMDLs
differed by <3).
Nested Logistic
0.4417
561.04
20.4
10.1841
106.295
53.0644
NCTR
0.4114
561.816
25.079
12.5395
127.994
63.997
Rai and Van Ryzin
0.4056
564.38
25.8561
1.00024
131.96
5.9492
Litter-specific covariate = implantation size; intra-litter correlations assumed to be zero
Nested Logistic
0.0000
643.52
36.1762
22.5296
188.497
117.391
NCTR
0.0000
650.146
33.8744
16.9372
172.883
86.4414
Rai and Van Ryzin
0.0000
660.111
35.975
17.9875
183.603
91.8017
Litter-specific covariate not used; intra-litter correlations estimated
Nested Logistic
0.3944
559.472
16.9114
9.03491
88.1172
47.0766
NCTRb
Rai and Van Ryzin
0.4051
560.38
25.8566
12.9283
131.963
65.9814
Litter-specific covariate not used; intra-litter correlations assumed to be zero
Nested Logistic
0.0000
654.556
26.3666
18.3313
137.384
95.5159
NCTRb
Rai and Van Ryzin
0.0000
656.111
35.975
17.9875
183.603
91.8017
Page 676 of 723
-------
aBecause the individual animal data were available, the BMDS nested models were fitted, with the selected model in bold.
For the selected model, the proportion of litters with scaled residuals above 2 in absolute value for doses 0, 19.6, 179, and
1,724 mg/kg-d were 2/22, 0/22, 2/20, and 0/20, respectively.
bWith the litter-specific covariate not used, the NCTR and Rai and van Ryzin models yielded identical results.
Data from Ema et al. (2008)
Nested Logistic Model, with BMR of 1% Extra Risk for the BMD and 0.95 Lower Confidence Limit for the BMDL
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 200 400 600 800 1000 1200 1400 1600 1800
dose
BMR = 1% ER; dose shown in mg/kg-day.
Nested Logistic
FigureApx 1-4. Plot of incidence rate by dose, with fitted curve for the nested logistic model
where the litter specific covariate was not used and the intra-litter correlations were estimated, for
incidence of offspring loss from PND 4 through PND 21 in F2 offspring CRL Sprague-Dawley
rats; lactational doses of F1 dams (Ema et al. 2008).
Nested Logistic Model (Version: 2.20; Date: 04/27/2015)
The form of the probability function is:
Prob. = alpha + thetal*Rij + [1 - alpha - thetal*Rij]/
[l+exp(-beta-theta2*Rij-rho*log(Dose))],
where Rij is the litter specific covariate.
Restrict Power rho >= 1.
Benchmark Dose Computation
To calculate the BMD and BMDL, the litter specific covariate is fixed at the mean litter specific
covariate of all the data: 14.654762
BMR = 1% ER
BMD = 16.9114
BMDL at the 95% confidence level = 9.03491
Parameter Estimates
Variable
Estimate
(Default) Initial Parameter Values
Alpha
0.133513
0.133513
Beta
-7.42311
-7.42311
Rho
1
1
Page 677 of 723
-------
phil
0.229222
0.229222
phi2
0.152985
0.152985
phi3
0.247495
0.247495
phi4
0.586386
0.586386
Log-likelihood:-273.736 AIC: 559.472
Goodness-of-Fit Table
Lit.-Spec. Litter Scaled
Dose Cov. Est. Prob. Size Expected Observed Residual
0.0000
9.0000
0.134
6
0.801
0
-0.6563
0.0000
10.0000
0.134
6
0.801
1
0.1630
0.0000
11.0000
0.134
8
1.068
0
-0.6880
0.0000
11.0000
0.134
6
0.801
0
-0.6563
0.0000
12.0000
0.134
8
1.068
1
-0.0439
0.0000
13.0000
0.134
8
1.068
6
3.1766
0.0000
13.0000
0.134
8
1.068
0
-0.6880
0.0000
13.0000
0.134
8
1.068
3
1.2443
0.0000
13.0000
0.134
8
1.068
0
-0.6880
0.0000
14.0000
0.134
8
1.068
1
-0.0439
0.0000
14.0000
0.134
8
1.068
0
-0.6880
0.0000
15.0000
0.134
4
0.534
0
-0.6043
0.0000
16.0000
0.134
8
1.068
1
-0.0439
0.0000
16.0000
0.134
8
1.068
1
-0.0439
0.0000
16.0000
0.134
8
1.068
0
-0.6880
0.0000
16.0000
0.134
8
1.068
2
0.6002
0.0000
16.0000
0.134
8
1.068
1
-0.0439
0.0000
16.0000
0.134
8
1.068
4
1.8884
0.0000
17.0000
0.134
8
1.068
0
-0.6880
0.0000
17.0000
0.134
8
1.068
0
-0.6880
0.0000
17.0000
0.134
8
1.068
5
2.5325
0.0000
18.0000
0.134
8
1.068
0
-0.6880
19.6000
12.0000
0.144
7
1.005
2
0.7747
19.6000
13.0000
0.144
8
1.148
1
-0.1039
19.6000
13.0000
0.144
8
1.148
0
-0.8046
19.6000
13.0000
0.144
8
1.148
3
1.2975
19.6000
14.0000
0.144
8
1.148
2
0.5968
19.6000
14.0000
0.144
8
1.148
0
-0.8046
19.6000
14.0000
0.144
8
1.148
0
-0.8046
19.6000
14.0000
0.144
8
1.148
0
-0.8046
19.6000
14.0000
0.144
8
1.148
0
-0.8046
19.6000
15.0000
0.144
8
1.148
1
-0.1039
19.6000
15.0000
0.144
8
1.148
3
1.2975
19.6000
15.0000
0.144
8
1.148
0
-0.8046
19.6000
15.0000
0.144
8
1.148
1
-0.1039
19.6000
16.0000
0.144
8
1.148
0
-0.8046
19.6000
16.0000
0.144
8
1.148
0
-0.8046
19.6000
16.0000
0.144
8
1.148
0
-0.8046
19.6000
16.0000
0.144
8
1.148
0
-0.8046
19.6000
17.0000
0.144
8
1.148
1
-0.1039
19.6000
17.0000
0.144
8
1.148
0
-0.8046
19.6000
17.0000
0.144
8
1.148
3
1.2975
19.6000
18.0000
0.144
8
1.148
1
-0.1039
19.6000
21.0000
0.144
8
1.148
0
-0.8046
Page 678 of 723
-------
179.0000
11.0000
0.217
8
1.738
4
1.1735
179.0000
11.0000
0.217
8
1.738
2
0.1361
179.0000
12.0000
0.217
8
1.738
2
0.1361
179.0000
13.0000
0.217
8
1.738
0
-0.9013
179.0000
14.0000
0.217
8
1.738
2
0.1361
179.0000
14.0000
0.217
8
1.738
5
1.6922
179.0000
14.0000
0.217
8
1.738
3
0.6548
179.0000
14.0000
0.217
8
1.738
1
-0.3826
179.0000
14.0000
0.217
8
1.738
4
1.1735
179.0000
14.0000
0.217
8
1.738
1
-0.3826
179.0000
14.0000
0.217
8
1.738
6
2.2109
179.0000
15.0000
0.217
8
1.738
0
-0.9013
179.0000
15.0000
0.217
8
1.738
0
-0.9013
179.0000
15.0000
0.217
8
1.738
1
-0.3826
179.0000
15.0000
0.217
8
1.738
6
2.2109
179.0000
16.0000
0.217
8
1.738
0
-0.9013
179.0000
16.0000
0.217
8
1.738
4
1.1735
179.0000
17.0000
0.217
8
1.738
0
-0.9013
179.0000
17.0000
0.217
8
1.738
0
-0.9013
179.0000
19.0000
0.217
8
1.738
5
1.6922
1.724.0000
10.0000
0.573
8
4.585
4
-0.1850
1.724.0000
11.0000
0.573
8
4.585
2
-0.8178
1.724.0000
12.0000
0.573
8
4.585
1
-1.1341
1.724.0000
12.0000
0.573
6
3.439
0
-1.4313
1.724.0000
13.0000
0.573
4
2.292
1
-0.7865
1.724.0000
14.0000
0.573
8
4.585
8
1.0805
1.724.0000
14.0000
0.573
8
4.585
1
-1.1341
1.724.0000
14.0000
0.573
8
4.585
0
-1.4505
1.724.0000
14.0000
0.573
4
2.292
4
1.0392
1.724.0000
15.0000
0.573
7
4.012
3
-0.3637
1.724.0000
15.0000
0.573
8
4.585
0
-1.4505
1.724.0000
15.0000
0.573
6
3.439
6
1.0662
1.724.0000
15.0000
0.573
4
2.292
4
1.0392
1.724.0000
16.0000
0.573
1
0.573
1
0.8631
1.724.0000
16.0000
0.573
8
4.585
5
0.1313
1.724.0000
16.0000
0.573
8
4.585
0
-1.4505
1.724.0000
17.0000
0.573
8
4.585
3
-0.5014
1.724.0000
17.0000
0.573
8
4.585
8
1.0805
1.724.0000
17.0000
0.573
8
4.585
3
-0.5014
1.724.0000
20.0000
0.573
8
4.585
8
1.0805
Observed Chi-square = 86.7400 Bootstrap Iterations per run = 10,000
p-value = 0.3944
Page 679 of 723
-------
Reduced Pup Body Weight
TableApx 1-8. Summary of BMD modeling results for pup weight during lactation in F2 male
offspring CRL Sprague-Dawley rats (PND 21) exposed to HBCD by diet for 3 weeks, lactational
dose (Ema et al. 2008); BMR = 5% RD from control mean, 10% RD from control mean, 0.5 SD
Model3
Goodness of fit
BMDL5RD
(mg/kg-d)
BMDiord
(mg/kg-d)
BMDLiord
(mg/kg-d)
Basis for model selection
p-value
AIC
(mg/kg-d)
Exponential (M2)
0.486
420.90
354
240
727
494
Of the models that provided
an adequate fit, a valid
BMDL estimate and
BMD/BMDL <5, the
Exponential M4 constant
variance model was selected
based on lowest BMDL
(BMDLs differed by >3).
Exponential (M3)
0.266
422.69
651
244
1016
500
Exponential (M4)
0.486
420.90
354
89.6
727
206
Exponential (M5)
N/Ab
424.68
230
94.0
258
181
Hill
N/Ab
424.68
230
89.2
264
error0
Power
0.266
422.69
676
282
1,049
565
Polynomial 3°
Polynomial 2°
0.264
422.70
817
282
1,161
564
Linear
0.497
420.85
389
280
779
560
Model3
Goodness of fit
BMDq.ssd
(mg/kg-d)
BMDLo.ssd
(mg/kg-d)
BMDisd
(mg/kg-d)
BMDLisd
(mg/kg-d)
p-value
AIC
Exponential (M2)
0.486
420.90
634
419
1,332
879
Exponential (M3)
0.266
422.69
937
425
1,483
891
Exponential (M4)
0.486
420.90
634
172
1,332
468
Exponential (M5)
N/Ab
424.68
252
176
296
189
Hill
N/Ab
424.68
256
176
324
error0
Power
0.266
422.69
969
482
1,503
965
Polynomial 3°
Polynomial 2°
0.264
422.70
1,091
482
1,549
964
Linear
0.497
420.85
684
478
1,368
956
"¦Constant variance case presented (BMDS Test 2 p-value = 0.0278), selected model in bold; scaled residuals for selected
model for doses 0, 19.6, 179, and 1,724 mg/kg-day were -0.92, 0.71, 0.27, and -0.06, respectively.
bNo available degrees of freedom to calculate a goodness-of-fit value.
°BMD or BMDL computation failed for this model.
Page 680 of 723
-------
Exponential 4 Model, with BMR of 0.05 Rel. Dev. for the BMD and 0.95 Lower Confidence Limit for the BMDL
Exponential 4
60
55
50
45
40
0
200
400
600
800
1000
1200
1400
1600
1800
BMR = 5% RD from control mean; dose shown in mg/kg-day.
FigureApx 1-5. Plot of mean response by dose with fitted curve for Exponential (M4) model with
constant variance for pup weight during lactation in F2 male offspring CRL Sprague-Dawley rats
(PND 21) exposed to HBCD by diet multigenerationally, lactational dose (Ema et al. 2008).
Exponential Model (Version: 1.10; Date: 01/12/2015)
The form of the response function is: Y[dose] = a * [c-(c-l) * exp(-b * dose)]
A constant variance model is fit
Benchmark Dose Computation
BMR = 5% RD
BMD = 353.728
BMDL at the 95% confidence level = 89.5935
Parameter Estimates
Variable
Estimate
Default initial parameter values
lnalpha
4.53195
4.51269
rho
N/A
0
a
54.8883
59.01
b
0.000145008
0.00128594
c
0
0.687535
d
N/A
1
Table of Data and Estimated Values of Interes
Dose
N
Observed mean
Estimated mean
Observed SD
Estimated SD
Scaled residuals
0
22
53
54.89
12.6
9.64
-0.9187
19.6
22
56.2
54.73
6.7
9.64
0.714
179
18
54.1
53.48
10.1
9.64
0.272
1,724
13
42.6
42.75
8.3
9.64
-0.0551
Page 681 of 723
-------
Likelihoods of Interest
Model
Log (likelihood)
Number of parameters
AIC
A1
-206.7258
5
423.4517
A2
-202.1665
8
420.333
A3
-206.7258
5
423.4517
R
-214.7267
2
433.4535
4
-207.4482
3
420.8963
Tests of Interest
Test
-2*log (likelihood ratio)
Test df
p-value
Test 1
25.12
6
0.0003244
Test 2
9.119
3
0.02775
Test 3
9.119
3
0.02775
Test 6a
1.445
2
0.4856
Delayed Eye Opening
The benchmark dose (BMD) modeling of nested dichotomous data was conducted with the EPA's BMD
software (BMDS version 3.11). The only model currently available in the software for use with nested
data is the nested logistic model. The nested logistic model was applied to the both male and female F2
pup eye opening datasets with and without a litter specific covariate to account for intra-litter similarity
(litter effects) based on non-treatment-related condition and with and without modeling of intra-litter
correlation to account for intra-litter similarity based on treatment-related effects in the two-generation
reproduction study. The number of implantations in F1 dams was found to not vary with treatment and
was therefore used as the litter-specific covariate for the modeling of the F2 pup eye opening
(Table Apx 1-9). F1 dam GDO body weight, F2 pup PND4 viability index, and F2 pup PND21 viability
index were also considered as litter-specific covariates. However, all these endpoints were affected by
treatment at the highest dose, and therefore, not suitable for use as a covariate.
Because BMDS can only model increasing dose-response trends for quantal data, the data were inverted
for modeling, as per the following example: 4 open/4 total (100%) -> 0 not open/4 total (0%).
Table Apx 1-9. Effect of Dose on Potential Litter-Specific Covariates
Dose
(m»/k;i-(l;i>)
l-'l (liim (.1)11 li\\ (li)
1-2 pup PND4
\ i;ihilil\ index (ni>)
1-2 pup PND2I
\ i;ihilil\ index ("»)
l-'l (him
iinpliiiils
0
297.7±28.1 (23)a
86.9±24.8 (23)
85.0±22.0 (22)
14.3±2.5 (23)
15
299.3±20.6 (23)
87.3±21.1 (23)
89.6±13.9 (22)
14.7±3.4 (23)
139
290.2±19.6 (21)
92.1±12.8 (20)
71.3±26.9 (20)
14.0±3.2 (21)
1360
272.9±22.2 (21)**
68.4±33.5 (21)*
49.7±41.1 (20)**
14.3±2.8 (21)
"Mean ± standard deviation (n)
* Statistically significant difference reported by study authors (p<0.05); **(p<0.01)
Page 682 of 723
-------
Female Offspring
Results for the F2 female pups are shown in TableApx 1-10. Significant model fit (p>0.1) was achieved
only when intra-litter correlation was modeled. With intra-litter correlation included in the model,
inclusion of the covariate (number of implantations) in the model increased model fit slightly (p=0.53 vs
p=0.49), but not enough to justify inclusion of the extra parameters in the model, as shown by the lower
AIC when the covariate was not modeled (270.4 vs 272.8). BMDLs were sufficiently close (<3-fold
difference), so model selection was based on AIC. The selected model (lowest AIC) included parameters
for intra-litter correlation but not for the covariate. Visual inspection of model fit and review of scaled
residuals confirmed adequate fit of the selected model to the data. Modeling was performed using BMR
= 10% or 5% extra risk. For both BMRs, the BMD results for the female pups are within the range of
observation (15-1360 mg/kg-day), and the BMDL results reflect acceptable levels of uncertainty
(BMD/BMDL ratio -2.6). For comparison to the BMDL values in Table Apx 1-10, the NOAEL and
LOAEL values for this endpoint were 15 and 139 mg/kg-day, respectively, based on statistical
significance.
Table Apx 1-10. Summary of BMD modeling results for delayed eye opening F2 female offspring
CRL Sprague-Dawley rats (PND 14); F2 generation doses (Ema et al. 2008); BMR = 5% ER and
10% ER
Model3
Goodness of Fit
BMDser
(mg/kg-d)
BMDL5ER
(mg/kg-d)
BMDioer
(mg/kg-d)
BMDLioer
(mg/kg-d)
Basis for model selection
p-value
AIC
Litter-specific covariate = implantation size; intra-litter correlations estimatea
Of the models that provided an
adequate fit, a valid BMDL
estimate and BMD/BMDL <5,
the Nested Logistic model with
litter-specific covariate not used
and intra-litter correlations
estimated was selected based on
lowest AIC (BMDLs differed
by <3).
Nested Logistic
0.5286
272.82
69.65
27.49
147.05
58.03
Litter-specific covariate = implantation size; intra-litter correlations assumed to be zero
Nested Logistic
<0.0001
302.36
57.77
29.62
121.97
62.53
Litter-specific covariate not used; intra-litter correlations estimated
Nested Logistic
0.4893
270.44
75.61
28.73
159.62
60.66
Litter-specific covariate not used; intra-litter correlations assumed to be zero
Nested Logistic
<0.0001
300.89
61.03
30.31
300.89
128.84
Estimated Probability
BMDL
BMR = 10% RD from control mean; dose shown in mg/kg-day.
Page 683 of 723
-------
0.9
0.8
0.7
Estimated Probability
0.6
0.5
Response at BMD
0.4
O Data
0.3
BMD
0.2
BMDL
0.1
0
0 200 400 600 800 1000 1200
Dose
BMR = 5% RD from control mean; dose shown in mg/kg-day.
FigureApx 1-6 and FigureApx 1-7. Plot of mean response by dose with fitted curve for
Frequentist Nested Logistic Model without litter-specific covariate and with intra-litter
correlation; and 0.95 Lower Confidence Limit for the BMDL in F2 female offspring CRL
Sprague-Dawley rats (PND 14) exposed to HBCD multigenerationally (Ema et al. 2008). Plots
display results for BMRs of 10% and 5% ER, respectively.
Male Offspring
Results for the F2 male pups are shown in TableApx 1-11. Significant model fit (p>0.1) was achieved
only when intra-litter correlation was modeled. With intra-litter correlation included in the model,
inclusion of the covariate (number of implantations) in the model increased model fit slightly and
slightly lowered the AIC (274.3 vs 274.4). BMDLs were sufficiently close (<3-fold difference), so initial
model selection was based on AIC. The model with lowest AIC included parameters for intra-litter
correlation and for the covariate. Review of scaled residuals confirmed adequate fit of this model to the
data, but visual inspection showed that model fit was problematic, with the lower doses not influencing
the shape and the high dose having outsized influence. Modeling was performed using BMR = 10% or
5% extra risk. For both BMRs, the BMD results for the male pups are within the range of observation,
but the BMDL results reflect very high levels of uncertainty in the modeling results (BMD/BMDL ratio
= 16-31). Log transformation of the doses produced a curve that appeared visually to better fit the data,
but p-value was not improved, and associated BMDs were below the range of observation. Dropping the
high dose was considered but not done because the only statistically significant change was at the high
dose. For comparison to the BMDL values in Table Apx 1-11, the NOAEL and LOAEL values for this
endpoint were 139 and 1360 mg/kg-day, respectively, based on statistical significance.
Page 684 of 723
-------
TableApx 1-11. Summary of BMD modeling results for delayed eye opening F2 female offspring
CRL Sprague-Dawley rats (PND 14); F2 generation doses (Ema et al. 2008); BMR = 5% ER and
10% ER
Model3
Goodness of Fit
p-value | AIC
BMDser
(mg/kg-d)
BMDLser
(mg/kg-d)
BMD icier
(mg/kg-d)
BMDLicier
(mg/kg-d)
Basis for model selection
Litter-specific covariate = implantation size; intra-litter correlations estimated
No model selected due to high
uncertainty in modeling results,
as indicated by BMD/ BMDL
ratio = 16-36 and poor visual fit
for models with adequate
statistical fit (p>0.1).
Nested Logistic
0.5223
274.29
842.06
27.50
954.73
58.05
Litter-specific covariate = implantation size; intra-litter correlations assumed to be zero
Nested Logistic
<0.0001
315.88
58.68
28.39
123.87
59.94
Litter-specific covariate not used; intra-litter correlations estimated
Nested Logistic
0.5220
274.37
917.36
25.30
1031.43
53.42
Litter-specific covariate not used; intra-litter correlations assumed to be zero
Nested Logistic
<0.0001
317.48
74.46
34.00
157.19
71.79
0.9
0.8
0.7
0.6
(u
c 0.5
o
Q.
V, 0.4
0.3
0.2
0.1
0
O
200
400
600 800
Dose
1000
1200
• Estimated
Probability
¦ Response at BMD
O Data
BMR = 10% RD from control mean; dose shown in mg/kg-day.
Figure Apx 1-8. Plot of mean response by dose with fitted curve for Frequentist Nested Logistic
Model without litter-specific covariate and with intra-litter correlation; and 0.95 Lower
Confidence Limit for the BMDL in F2 male offspring CRL Sprague-Dawley rats (PND 14)
exposed to HBCD multigenerationally (Ema et al. 2008). Plot displays results for BMRs of 10%
and 5% ER.
Page 685 of 723
-------
Appendix J ENVIRONMENTAL RISK
J.l Aquatic Environment
J.l.l Risk Quotients based on a Production Volume of 100,000 lbs/yr and 0% Removal from Direct Releases
J.l.1.1 E-FAST Initial Screening for Surface Water Concentrations
Table Apx J-l. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations (jig/L) Using E-FAST (0%
Removal)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration of concern
(COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 |ig/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th Percentile 7Q10
50th percentile: 7Q10
SWC
(Mg/L)
Acute
RQ
(COC:
0.4 jig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
SWC
(Mg/L)
Acute
RQ
(COC:
0.4
fig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
Section 2.4.1.2 - Repackaging of Import
Containers (1)
1.1
29
19.45
48.63
46.64
19.45
0.39
0.98
0.94
0.39
1.2
300
1.87
4.68
4.48
1.87
0.04
0.09
0.09
0.04
1.3
29
97.51
243.78
233.84
97.51
1.94
4.85
4.65
1.94
1.4
300
9.43
23.58
22.61
9.43
0.19
0.48
0.46
0.19
1.5
29
20.10
50.25
48.20
20.10
2.00
5.00
4.80
2.00
1.6
300
1.93
4.83
4.63
1.93
0.19
0.48
0.46
0.19
1.7
29
100.77
251.93
241.65
100.77
10.00
25.00
23.98
10.00
1.8
300
9.74
24.35
23.36
9.74
0.97
2.43
2.33
0.97
Section 2.4.1.3 - Compounding of Polystyrene
Resin to Produce XPS Masterbatch (2)
2.1
10
18.70
46.75
44.84
18.70
0.37
0.93
0.89
0.37
2.2
60
3.04
7.60
7.29
3.04
0.06
0.15
0.15
0.06
2.3
10
42.02
105.05
100.77
42.02
0.84
2.10
2.01
0.84
2.4
60
7.00
17.50
16.79
7.00
0.14
0.35
0.34
0.14
2.5
10
1.87
4.68
4.48
1.87
0.04
0.09
0.09
0.04
Page 686 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration of concern
(COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 |ig/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th Percentile 7Q10
50th percentile: 7Q10
SWC
(Mg/L)
Acute
RQ
(COC:
0.4 ng/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
SWC
(Mg/L)
Acute
RQ
(COC:
0.4
fig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
2.6
60
0.30
0.75
0.72
0.30
0.01
0.02
0.01
0.01
2.7
10
4.20
10.50
10.07
4.20
0.08
0.21
0.20
0.08
2.8
60
0.70
1.75
1.68
0.70
0.01
0.03
0.03
0.01
2.9
10
1.93
4.83
4.63
1.93
0.19
0.48
0.46
0.19
2.10
60
0.31
0.78
0.74
0.31
0.03
0.08
0.07
0.03
2.11
10
4.34
10.85
10.41
4.34
0.43
1.08
1.03
0.43
2.12
60
0.72
1.80
1.73
0.72
0.07
0.18
0.17
0.07
Section 2.4.1.4 - Processing of HBCD to produce
XPS Foam using XPS Masterbatch (3)
3.1
1
60.60
151.50
145.32
60.60
1.20
3.00
2.88
1.20
3.2
15
4.04
10.10
9.69
4.04
0.08
0.20
0.19
0.08
3.3
1
148.38
370.95
355.83
148.38
2.95
7.38
7.07
2.95
3.4
15
9.98
24.95
23.93
9.98
0.20
0.50
0.48
0.20
3.5
1
6.06
15.15
14.53
6.06
0.12
0.30
0.29
0.12
3.6
15
0.40
1.01
0.97
0.40
0.01
0.02
0.02
0.01
3.7
1
14.84
37.10
35.58
14.84
0.30
0.74
0.71
0.30
3.8
15
1.00
2.50
2.39
1.00
0.02
0.05
0.05
0.02
3.9
1
6.26
15.66
15.02
6.26
0.62
1.56
1.49
0.62
3.10
15
0.42
1.05
1.00
0.42
0.04
0.10
0.10
0.04
3.11
1
15.34
38.34
36.77
15.34
1.52
3.81
3.65
1.52
3.12
15
1.03
2.58
2.47
1.03
0.10
0.25
0.24
0.10
Section 2.2.5 - Processing of HBCD to produce
XPS Foam using HBCD Powder (4)
4.1
1
57.73
144.33
138.44
57.73
1.15
2.88
2.76
1.15
4.2
12
4.86
12.15
11.65
4.86
0.10
0.24
0.23
0.10
4.3
1
5.77
14.43
13.84
5.77
0.12
0.29
0.28
0.12
4.4
12
0.49
1.22
1.17
0.49
0.01
0.02
0.02
0.01
Page 687 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration of concern
(COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 |ig/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th Percentile 7Q10
50th percentile: 7Q10
SWC
(Mg/L)
Acute
RQ
(COC:
0.4 ng/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
SWC
(Mg/L)
Acute
RQ
(COC:
0.4
fig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
4.5
1
5.97
14.93
14.32
5.97
0.59
1.48
1.41
0.59
4.6
12
0.50
1.25
1.20
0.50
0.05
0.12
0.12
0.05
Section 2.4.1.6 - Processing of HBCD to produce
EPS Foam from Imported EPS Resin Beads (5)
5.1
16
3881.55
9703.88
9308.27
3881.55
77.16
192.90
185.04
77.16
5.2
16
388.16
970.39
930.83
388.16
7.72
19.29
18.50
7.72
5.3
16
401.16
1002.90
962.01
401.16
39.82
99.55
95.49
39.82
5.4
140
444.39
1110.98
1065.68
444.39
8.83
22.08
21.18
8.83
5.5
140
44.44
111.10
106.57
44.44
0.88
2.21
2.12
0.88
5.6
140
45.93
114.83
110.14
45.93
4.56
11.40
10.94
4.56
5.7
16
5295.51
13238.78
12699.06
5295.51
105.26
263.15
252.42
105.26
5.8
16
529.55
1323.88
1269.91
529.55
10.53
26.32
25.24
10.53
5.9
16
547.29
1368.23
1312.45
547.29
54.32
135.80
130.26
54.32
5.10
140
605.99
1514.98
1453.21
605.99
12.05
30.13
28.90
12.05
5.11
140
60.60
151.50
145.32
60.60
1.21
3.01
2.89
1.21
5.12
140
62.63
156.58
150.19
62.63
6.22
15.55
14.92
6.22
Section 2.4.1.7 - Processing of HBCD to produce
SIPs and Automobile Replacement Parts from
XPS/EPS Foam (6)
6.1
16
17.83
44.58
42.76
17.83
0.35
0.88
0.84
0.35
6.2
16
1.78
4.46
4.28
1.78
0.04
0.09
0.08
0.04
6.3
16
1.84
4.60
4.41
1.84
0.18
0.45
0.43
0.18
6.4
300
0.95
2.38
2.28
0.95
0.02
0.05
0.05
0.02
6.5
300
0.10
0.24
0.23
0.10
0.00
0.00
0.00
0.00
6.6
300
0.10
0.25
0.24
0.10
0.01
0.02
0.02
0.01
6.7
16
79.60
199.00
190.89
79.60
1.60
4.00
3.84
1.60
6.8
16
7.96
19.90
19.09
7.96
0.16
0.40
0.38
0.16
6.9
16
8.25
20.63
19.78
8.25
0.82
2.05
1.97
0.82
Page 688 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration of concern
(COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 |ig/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th Percentile 7Q10
50th percentile: 7Q10
SWC
(Mg/L)
Acute
RQ
(COC:
0.4 ng/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
SWC
(Mg/L)
Acute
RQ
(COC:
0.4
fig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
6.10
300
4.20
10.50
10.07
4.20
0.08
0.21
0.20
0.08
6.11
300
0.42
1.05
1.01
0.42
0.01
0.02
0.02
0.01
6.12
300
0.44
1.10
1.06
0.44
0.04
0.11
0.10
0.04
Section 2.4.1.9 - Installation of XPS/EPS Foam
Insulation in Residential, Public, and Commercial
Buildings, and Other Structures (8)
8.1
1
0.80
2.00
1.92
0.80
0.03
0.08
0.08
0.03
8.2
1
0.08
0.20
0.19
0.08
0.00
0.01
0.01
0.00
8.3
3
94.00
235.00
225.42
94.00
3.70
9.25
8.87
3.70
8.4
3
9.40
23.50
22.54
9.40
0.37
0.93
0.89
0.37
Section 2.4.1.10 - Demolition of XPS/EPS Foam
Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures (9)
9.1
1
0.71
1.78
1.70
0.71
0.03
0.07
0.07
0.03
9.2
1
0.07
0.18
0.17
0.07
0.00
0.01
0.01
0.00
9.3
3
636.79
1591.98
1527.07
636.79
25.19
62.98
60.41
25.19
9.4
3
63.68
159.20
152.71
63.68
2.52
6.30
6.04
2.52
Section 2.4.1.11- Recycling of EPS Foam and
Reuse of XPS Foam (10)
10.1
1
83.14
207.85
199.38
83.14
1.65
4.13
3.96
1.65
10.2
1
8.31
20.79
19.94
8.31
0.17
0.41
0.40
0.17
10.3
1
8.59
21.48
20.60
8.59
0.85
2.13
2.04
0.85
10.4
140
0.59
1.48
1.41
0.59
0.01
0.03
0.03
0.01
10.5
140
0.06
0.15
0.14
0.06
0.00
0.00
0.00
0.00
10.6
140
0.06
0.15
0.15
0.06
0.01
0.02
0.01
0.01
10.7
1
99.00
247.50
237.41
99.00
1.97
4.93
4.72
1.97
10.8
1
9.90
24.75
23.74
9.90
0.20
0.49
0.47
0.20
10.9
1
10.23
25.58
24.53
10.23
1.02
2.55
2.45
1.02
10.10
140
0.71
1.77
1.70
0.71
0.01
0.04
0.03
0.01
10.11
140
0.07
0.18
0.17
0.07
0.00
0.00
0.00
0.00
10.12
140
0.07
0.18
0.18
0.07
0.01
0.02
0.02
0.01
Page 689 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration of concern
(COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 |ig/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th Percentile 7Q10
50th percentile: 7Q10
SWC
(Mg/L)
Acute
RQ
(COC:
0.4 ng/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
SWC
(Mg/L)
Acute
RQ
(COC:
0.4
fig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Algae
RQ
(COC:
lfig/L)
Section 2.4.1.13 -Use of Flux/Solder Pastes (12)
12.1
4
0.31
0.78
0.74
0.31
0.01
0.02
0.01
0.01
12.2
4
0.32
0.80
0.77
0.32
0.03
0.08
0.08
0.03
12.3
300
0.00
0.01
0.01
0.00
0.00
0.00
0.00
0.00
12.4
300
0.00
0.01
0.01
0.00
0.00
0.00
0.00
0.00
12.5
4
0.62
1.55
1.49
0.62
0.01
0.03
0.03
0.01
12.6
4
0.64
1.60
1.53
0.64
0.06
0.16
0.15
0.06
12.7
300
0.01
0.02
0.02
0.01
0.00
0.00
0.00
0.00
12.8
300
0.01
0.02
0.02
0.01
0.00
0.00
0.00
0.00
Page 690 of 723
-------
J.l.1.2
PSC Predicted Surface Water and Sediment Concentrations
Table Apx J-2. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations (jig/L) Using PSC (0%
Removal)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration
of concern (COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-
scenario exceeds the water solubility of HBCD (66 ng/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th percentile
50th percentile
1-day
SWC
(HS/L)
RQs based on 1-
d SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(HS/L)
RQs based on
1-d SWC
21-
day
SWC:
fig/L
RQs
based
on 21-d
SWC
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
1
fig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Section 2.4.1.2- Repackaging of Import
Containers (1)
1.1
29
14.70
36.75
14.70
1.71
4.10
0.38
0.96
0.38
0.04
0.09
1.2
300
1.72
4.30
1.72
1.46
3.50
0.04
0.09
0.04
0.03
0.07
1.3
29
73.70
184.25
73.70
8.59
20.60
1.93
4.83
1.93
0.18
0.44
1.4
300
8.69
21.73
8.69
7.35
17.63
0.19
0.47
0.19
0.15
0.36
1.5
29
15.10
37.75
15.10
1.77
4.24
1.93
4.83
1.93
0.19
0.45
1.6
300
1.78
4.45
1.78
1.51
3.62
0.19
0.48
0.19
0.16
0.37
1.7
29
75.60
189.00
75.60
8.85
21.22
9.68
24.20
9.68
0.94
2.26
1.8
300
8.96
22.40
8.96
7.59
18.20
0.96
2.40
0.96
0.78
1.87
Section 2.4.1.3 - Compounding of Polystyrene
Resin to Produce XPS Masterbatch (2)
2.1
10
13.90
34.75
13.90
0.79
1.88
0.37
0.92
0.37
0.02
0.04
2.2
60
2.36
5.90
2.36
0.54
1.30
0.06
0.15
0.06
0.02
0.04
2.3
10
31.30
78.25
31.30
1.76
4.22
0.83
2.08
0.83
0.04
0.10
2.4
60
5.43
13.58
5.43
1.25
3.00
0.14
0.35
0.14
0.03
0.06
2.5
10
1.39
3.48
1.39
0.08
0.19
0.04
0.09
0.04
0.00
0.00
2.6
60
0.00
0.00
0.00
0.00
0.00
0.00
2.7
10
3.13
7.83
3.13
0.18
0.42
0.08
0.21
0.08
0.00
0.01
2.8
60
0.00
0.00
0.00
0.00
0.00
0.00
Page 691 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration
of concern (COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-
scenario exceeds the water solubility of HBCD (66 ng/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
10th percentile
50th percentile
Sub-
Scenario
Days of
Release
1-day
SWC
(fig/L)
RQs based on 1-
d SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(fig/L)
RQs based on
1-d SWC
21-
RQs
based
on 21-d
SWC
Exposure Scenario
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
1
fig/L)
day
SWC:
fig/L
Chronic
RQ
(COC:
0.417
fig/L)
2.9
10
1.43
3.58
1.43
0.08
0.19
0.19
0.46
0.19
0.01
0.02
2.10
60
0.00
0.00
0.00
0.00
0.00
0.00
2.11
10
3.21
8.03
3.21
0.18
0.44
0.42
1.04
0.42
0.02
0.05
2.12
60
0.00
0.00
0.00
0.00
0.00
0.00
3.1
1
44.90
112.25
44.90
2.31
5.54
1.20
3.00
1.20
0.06
0.14
3.2
15
3.02
7.55
3.02
0.18
0.43
0.08
0.20
0.08
0.00
0.01
3.3
1
110.00
275.00
110.00
5.65
13.55
2.93
7.33
2.93
0.14
0.34
3.4
15
7.46
18.65
7.46
0.45
1.07
0.20
0.49
0.20
0.01
0.02
3.5
1
4.49
11.23
4.49
0.23
0.55
0.12
0.30
0.12
0.01
0.01
Section 2.4.1.4 - Processing of HBCD to
produce XPS Foam using XPS Masterbatch (3)
3.6
15
0.30
0.76
0.30
0.02
0.04
0.01
0.02
0.01
0.00
0.00
3.7
1
11.00
27.50
11.00
0.57
1.35
0.29
0.73
0.29
0.01
0.03
3.8
15
0.75
1.87
0.75
0.04
0.11
0.02
0.05
0.02
0.00
0.00
3.9
1
4.60
11.50
4.60
0.24
0.57
0.60
1.50
0.60
0.03
0.07
3.10
15
0.31
0.78
0.31
0.02
0.04
0.04
0.10
0.04
0.00
0.00
3.11
1
11.30
28.25
11.30
0.58
1.39
1.47
3.68
1.47
0.07
0.17
3.12
15
0.77
1.91
0.77
0.05
0.11
0.10
0.25
0.10
0.00
0.01
4.1
1
42.80
107.00
42.80
2.19
5.25
1.14
2.85
1.14
0.05
0.13
Section 2.2.5 - Processing of HBCD to produce
4.2
12
3.63
9.08
3.63
0.21
0.49
0.10
0.24
0.10
0.00
0.01
XPS Foam using HBCD Powder (4)
4.3
1
4.28
10.70
4.28
0.22
0.53
0.11
0.29
0.11
0.01
0.01
4.4
12
0.36
0.91
0.36
0.02
0.05
0.01
0.02
0.01
0.00
0.00
Page 692 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration
of concern (COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-
scenario exceeds the water solubility of HBCD (66 ng/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
10th percentile
50th percentile
Sub-
Scenario
Days of
Release
1-day
SWC
(fig/L)
RQs based on 1-
d SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(fig/L)
RQs based on
1-d SWC
21-
RQs
based
on 21-d
SWC
Exposure Scenario
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
1
fig/L)
day
SWC:
fig/L
Chronic
RQ
(COC:
0.417
fig/L)
4.5
1
4.38
10.95
4.38
0.23
0.54
0.57
1.43
0.57
0.03
0.07
4.6
12
0.37
0.93
0.37
0.02
0.05
0.05
0.12
0.05
0.00
0.01
5.1
16
2900.00
7250.00
2900.00
172.00
412.47
76.60
191.50
76.60
3.67
8.80
5.2
16
290.00
725.00
290.00
17.20
41.25
7.66
19.15
7.66
0.37
0.88
5.3
16
297.00
742.50
297.00
17.70
42.45
38.50
96.25
38.50
1.88
4.51
5.4
140
358.00
895.00
358.00
140.00
335.73
8.78
21.95
8.78
2.94
7.05
Section 2.4.1.6 - Processing of HBCD to
produce EPS Foam from Imported EPS Resin
Beads (5)
5.5
140
35.80
89.50
35.80
14.00
33.57
0.88
2.20
0.88
0.29
0.71
5.6
140
36.80
92.00
36.80
14.40
34.53
4.44
11.10
4.44
1.51
3.62
5.7
16
3960.00
9900.00
3960.00
235.00
563.55
105.00
262.50
105.00
5.01
12.01
5.8
16
396.00
990.00
396.00
23.50
56.35
10.50
26.25
10.50
0.50
1.20
5.9
16
406.00
1015.00
406.00
24.20
58.03
52.50
131.25
52.50
2.57
6.16
5.10
140
489.00
1222.50
489.00
191.00
458.03
12.00
30.00
12.00
4.01
9.62
5.11
140
48.90
122.25
48.90
19.10
45.80
1.20
3.00
1.20
0.40
0.96
5.12
140
50.30
125.75
50.30
19.70
47.24
6.06
15.15
6.06
2.06
4.94
6.1
16
13.30
33.25
13.30
0.79
1.89
0.35
0.88
0.35
0.02
0.04
6.2
16
1.33
3.33
1.33
0.08
0.19
0.04
0.09
0.04
0.00
0.00
Section 2.4.1.7 - Processing of HBCD to
produce SIPs and Automobile Replacement Parts
from XPS/EPS Foam (6)
6.3
16
1.37
3.43
1.37
0.08
0.20
0.18
0.44
0.18
0.01
0.02
6.4
300
0.87
2.17
0.87
0.77
1.84
0.02
0.05
0.02
0.02
0.04
6.5
300
0.00
0.00
0.00
0.00
0.00
0.00
6.6
300
0.00
0.00
0.00
0.00
0.00
0.00
Page 693 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration
of concern (COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-
scenario exceeds the water solubility of HBCD (66 ng/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th percentile
50th percentile
1-day
SWC
(fig/L)
RQs based on 1-
d SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(fig/L)
RQs based on
1-d SWC
21-
day
SWC:
fig/L
RQs
based
on 21-d
SWC
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
1
fig/L)
Chronic
RQ
(COC:
0.417
fig/L)
6.7
16
59.50
148.75
59.50
3.53
8.47
1.57
3.93
1.57
0.08
0.18
6.8
16
5.95
14.88
5.95
0.35
0.85
0.16
0.39
0.16
0.01
0.02
6.9
16
6.10
15.25
6.10
0.36
0.87
0.79
1.97
0.79
0.04
0.09
6.10
300
3.87
9.68
3.87
3.41
8.18
0.08
0.21
0.08
0.07
0.17
6.11
300
0.39
0.97
0.39
0.34
0.82
0.01
0.02
0.01
0.01
0.02
6.12
300
0.40
1.00
0.40
0.35
0.84
0.04
0.11
0.04
0.04
0.09
Section 2.4.1.9 - Installation of XPS/EPS Foam
Insulation in Residential, Public, and
Commercial Buildings, and Other Structures (8)
8.1
1
0.02
0.05
0.02
0.00
0.00
0.00
0.01
0.00
0.00
0.00
8.2
1
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
8.3
3
23.70
59.25
23.70
1.71
4.10
3.38
8.45
3.38
0.02
0.04
8.4
3
2.37
5.93
2.37
0.17
0.41
0.34
0.85
0.34
0.00
0.00
Section 2.4.1.10 - Demolition of XPS/EPS Foam
Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures (9)
9.1
1
0.02
0.05
0.02
0.00
0.00
0.00
0.01
0.00
0.00
0.00
9.2
1
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
9.3
3
23.70
59.25
23.70
1.71
4.10
3.38
8.45
3.38
0.02
0.04
9.4
3
2.37
5.93
2.37
0.17
0.41
0.34
0.85
0.34
0.00
0.00
Section 2.4.1.11- Recycling of EPS Foam and
Reuse of XPS Foam (10)
10.1
1
61.60
154.00
61.60
3.16
7.58
1.64
4.10
1.64
0.08
0.19
10.2
1
6.16
15.40
6.16
0.32
0.76
0.16
0.41
0.16
0.01
0.02
10.3
1
6.31
15.78
6.31
0.33
0.78
0.82
2.06
0.82
0.04
0.09
10.4
140
0.48
1.20
0.48
0.19
0.45
0.01
0.03
0.01
0.00
0.01
10.5
140
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
10.6
140
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Page 694 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the concentration
of concern (COC) for acute, algae or chronic environmental hazard. Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-
scenario exceeds the water solubility of HBCD (66 ng/L), resulting in acute, algae or chronic RQs greater than 165, 66, and 158.3, respectively.
10th percentile
50th percentile
Sub-
Scenario
Days of
Release
1-day
SWC
(fig/L)
RQs based on 1-
d SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(fig/L)
RQs based on
1-d SWC
21-
RQs
based
on 21-d
SWC
Exposure Scenario
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
1
fig/L)
day
SWC:
fig/L
Chronic
RQ
(COC:
0.417
fig/L)
10.7
1
73.30
183.25
73.30
3.76
9.02
1.95
4.88
1.95
0.09
0.22
10.8
1
7.33
18.33
7.33
0.38
0.90
0.20
0.49
0.20
0.01
0.02
10.9
1
7.51
18.78
7.51
0.39
0.93
0.98
2.45
0.98
0.05
0.11
10.10
140
0.57
1.43
0.57
0.22
0.53
0.01
0.04
0.01
0.00
0.01
10.11
140
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
10.12
140
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
12.1
4
0.23
0.58
0.23
0.01
0.03
0.01
0.02
0.01
0.00
0.00
12.2
4
0.24
0.59
0.24
0.01
0.03
0.03
0.08
0.03
0.00
0.00
12.3
300
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Section 2.4.1.13 - Use of Flux/Solder Pastes (12)
12.4
300
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
12.5
4
0.46
1.16
0.46
0.02
0.06
0.01
0.03
0.01
0.00
0.00
12.6
4
0.47
1.19
0.47
0.03
0.06
0.06
0.15
0.06
0.00
0.01
12.7
300
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
12.8
300
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Page 695 of 723
-------
Table Apx J-3. Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (^ig/kg) Using PSC (0% Removal)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC) for chronic environmental hazard. The shaded values in yellow denote predicted sediment concentrations by PSC. Blank spaces denote scenarios where HBCD
sediment concentrations are <100 jug/kg for segments of a water body within 100 meters of the facility; these scenarios were not run because the surface water concentrations
that less than the chronic COC of 1,570 Jig/kg.
10th percentile
50th percentile
11-d half-life
128-d half-life
11-d half-life
128-d half-life
Exposure Scenario
Sub-
Scenario
Days of
Release
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
1.1
29
1400
0.89
3620
2.31
34.4
0.02
77
0.05
1.2
300
1380
0.88
3600
2.29
33.8
0.02
76.7
0.05
1.3
29
7040
4.48
18200
11.59
172
0.11
386
0.25
Section 2.4.1.2 - Repackaging of Import
1.4
300
6980
4.45
18200
11.59
170
0.11
385
0.25
Containers (1)
1.5
29
1440
0.92
3730
2.38
174
0.11
395
0.25
1.6
300
1420
0.90
3720
2.37
171
0.11
393
0.25
1.7
29
7230
4.61
18700
11.91
872
0.56
1980
1.26
1.8
300
7170
4.57
18700
11.91
862
0.55
1980
1.26
2.1
10
537
0.34
1300
0.83
13.3
0.01
27.9
0.02
2.2
60
471
0.30
1220
0.78
11.5
0.01
26
0.02
2.3
10
1210
0.77
2920
1.86
29.8
0.02
62.8
0.04
2.4
60
1080
0.69
2810
1.79
26.5
0.02
59.7
0.04
Section 2.4.1.3 - Compounding of Polystyrene
Resin to Produce XPS Masterbatch (2)
2.5
10
53.7
0.03
130
0.08
1.33
0.00
2.79
0.00
2.6
60
2.7
10
121
0.08
292
0.19
2.98
0.00
6.28
0.00
2.8
60
2.9
10
55.1
0.04
134
0.09
6.72
0.00
14.3
0.01
2.1
60
2.11
10
124
0.08
301
0.19
15.1
0.01
32.2
0.02
Page 696 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC) for chronic environmental hazard. The shaded values in yellow denote predicted sediment concentrations by PSC. Blank spaces denote scenarios where HBCD
sediment concentrations are <100 jug/kg for segments of a water body within 100 meters of the facility; these scenarios were not run because the surface water concentrations
that less than the chronic COC of 1,570 Jig/kg.
10th percentile
50th percentile
11-d half-life
128-d half-life
11-d half-life
128-d half-life
Exposure Scenario
Sub-
Scenario
Days of
Release
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
2.12
60
3.1
1
1430
0.91
1910
1.22
36.4
0.02
48.3
0.03
3.2
15
163
0.10
414
0.26
4.01
0.00
8.86
0.01
3.3
1
3490
2.22
4670
2.97
89.1
0.06
118
0.08
3.4
15
403
0.26
1020
0.65
9.9
0.01
21.9
0.01
3.5
1
143
0.09
191
0.12
3.64
0.00
4.83
0.00
Section 2.4.1.4 - Processing of HBCD to
3.6
15
16.3
0.01
41.4
0.03
0.4
0.00
0.89
0.00
produce XPS Foam using XPS Masterbatch (3)
3.7
1
349
0.22
467
0.30
8.91
0.01
11.8
0.01
3.8
15
40.3
0.03
102
0.06
0.99
0.00
2.19
0.00
3.9
1
146
0.09
196
0.12
18.3
0.01
24.4
0.02
3.10
15
16.8
0.01
42.7
0.03
2.03
0.00
4.54
0.00
3.11
1
358
0.23
479
0.31
44.9
0.03
59.7
0.04
3.12
15
41.4
0.03
105
0.07
5.01
0.00
11.2
0.01
4.1
1
1360
0.87
1820
1.16
34.7
0.02
46
0.03
4.2
12
152
0.10
385
0.25
3.73
0.00
8.22
0.01
Section 2.2.5 - Processing of HBCD to produce
4.3
1
136
0.09
182
0.12
3.47
0.00
4.6
0.00
XPS Foam using HBCD Powder (4)
4.4
12
15.2
0.01
38.5
0.02
0.37
0.00
0.82
0.00
4.5
1
139
0.09
186
0.12
17.5
0.01
23.2
0.01
4.6
12
15.6
0.01
39.7
0.03
1.89
0.00
4.22
0.00
5.1
16
165000
105.10
417000
265.61
4050
2.58
8910
5.68
5.2
16
16500
10.51
41700
26.56
405
0.26
891
0.57
Page 697 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC) for chronic environmental hazard. The shaded values in yellow denote predicted sediment concentrations by PSC. Blank spaces denote scenarios where HBCD
sediment concentrations are <100 jug/kg for segments of a water body within 100 meters of the facility; these scenarios were not run because the surface water concentrations
that less than the chronic COC of 1,570 Jig/kg.
10th percentile
50th percentile
11-d half-life
128-d half-life
11-d half-life
128-d half-life
Exposure Scenario
Sub-
Scenario
Days of
Release
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
5.3
16
16900
10.76
42900
27.32
2050
1.31
4560
2.90
5.4
140
137000
87.26
356000
226.75
3340
2.13
7560
4.82
5.5
140
13700
8.73
35600
22.68
334
0.21
756
0.48
5.6
140
14100
8.98
36800
23.44
1690
1.08
3880
2.47
Section 2.4.1.6 - Processing of HBCD to
produce EPS Foam from Imported EPS Resin
Beads (5)
5.7
16
225000
143.31
568000
361.78
5530
3.52
12200
7.77
5.8
16
22500
14.33
56800
36.18
553
0.35
1220
0.78
5.9
16
23100
14.71
58600
37.32
2800
1.78
6230
3.97
5.1
140
187000
119.11
487000
310.19
4560
2.90
10300
6.56
5.11
140
18700
11.91
48700
31.02
456
0.29
1030
0.66
5.12
140
19200
12.23
50200
31.97
2330
1.48
5300
3.38
6.1
16
758
0.48
1910
1.22
18.6
0.01
40.9
0.03
6.2
16
75.8
0.05
191
0.12
1.86
0.00
4.09
0.00
6.3
16
77.8
0.05
197
0.13
9.42
0.01
21
0.01
6.4
300
735
0.47
1910
1.22
17.9
0.01
40.6
0.03
Section 2.4.1.7 - Processing of HBCD to
produce SIPs and Automobile Replacement Parts
6.5
300
6.6
300
from XPS/EPS Foam (6)
6.7
16
3380
2.15
8540
5.44
83
0.05
183
0.12
6.8
16
338
0.22
854
0.54
8.3
0.01
18.3
0.01
6.9
16
347
0.22
880
0.56
42
0.03
93.6
0.06
6.10
300
3270
2.08
8510
5.42
79.8
0.05
181
0.12
6.11
300
327
0.21
851
0.54
7.98
0.01
18.1
0.01
Page 698 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC) for chronic environmental hazard. The shaded values in yellow denote predicted sediment concentrations by PSC. Blank spaces denote scenarios where HBCD
sediment concentrations are <100 jug/kg for segments of a water body within 100 meters of the facility; these scenarios were not run because the surface water concentrations
that less than the chronic COC of 1,570 Jig/kg.
10th percentile
50th percentile
11-d half-life
128-d half-life
11-d half-life
128-d half-life
Exposure Scenario
Sub-
Scenario
Days of
Release
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
6.12
300
336
0.21
878
0.56
40.4
0.03
92.7
0.06
8.1
1
0.76
0.00
1.05
0.00
0.09
0.00
0.12
0.00
Section 2.4.1.9 - Installation of XPS/EPS Foam
Insulation in Residential, Public, and
Commercial Buildings, and Other Structures (8)
8.2
1
8.3
3
898
0.57
2010
1.28
105
0.07
161.0
0.10
8.4
3
89.8
0.06
201
0.13
10.5
0.01
16.1
0.01
9.1
1
0.76
0.00
1.05
0.00
0.09
0.00
0.12
0.00
Section 2.4.1.10 - Demolition of XPS/EPS Foam
Insulation Products in Residential, Public and
Commercial Buildings, and Other Structures (9)
9.2
1
9.3
3
159
0.10
10.2
0.01
22.8
0.01
1.1
0.00
9.4
3
15.9
0.01
1.02
0.00
2.28
0.00
0.1
0.00
10.1
1
1960
1.25
2620
1.67
49.9
0.03
66.3
0.04
10.2
1
196
0.12
262
0.17
4.99
0.00
6.63
0.00
10.3
1
201
0.13
268
0.17
25.2
0.02
33.4
0.02
10.4
140
184
0.12
478
0.30
4.48
0.00
10.1
0.01
10.5
140
Section 2.4.1.11- Recycling of EPS Foam and
10.6
140
Reuse of XPS Foam (10)
10.7
1
2330
1.48
3110
1.98
59.5
0.04
95.6
0.06
10.8
1
233
0.15
311
0.20
5.95
0.00
9.56
0.01
10.9
1
239
0.15
320
0.20
30
0.02
39.8
0.03
10.10
140
218
0.14
568
0.36
5.32
0.00
12
0.01
10.11
140
10.12
140
Page 699 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC) for chronic environmental hazard. The shaded values in yellow denote predicted sediment concentrations by PSC. Blank spaces denote scenarios where HBCD
sediment concentrations are <100 jug/kg for segments of a water body within 100 meters of the facility; these scenarios were not run because the surface water concentrations
that less than the chronic COC of 1,570 Jig/kg.
Exposure Scenario
Sub-
Scenario
Days of
Release
10th percentile
50th percentile
11-d half-life
128-d half-life
11-d half-life
128-d half-life
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Section 2.4.1.13 -Use of Flux/Solder Pastes (12)
12.1
4
7.37
0.00
12.5
0.01
0.19
0.00
0.29
0.00
12.2
4
7.56
0.00
12.9
0.01
0.95
0.00
1.47
0.00
12.3
300
12.4
300
12.5
4
14.7
0.01
25
0.02
0.38
0.00
0.58
0.00
12.6
4
15.1
0.01
25.7
0.02
1.89
0.00
2.94
0.00
12.7
300
12.8
300
Page 700 of 723
-------
J.1.2 Targeted Sensitivity Analysis
J.l.2.1 Exposure Scenario 1: Repackaging of Import Containers
Table Apx J-4. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations (jig/L) Using PSC (Targeted
Sensitivity Analysis Parameter: Production Volume)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. For this exposure scenario, HBCD is
released following treatment via an onsite WWT or POTW and there are not any sub-scenarios where direct release of HBCD into surface water is expected.
Exposure
Scenario
Sub-
Scenario
Production
Volume
(lbs/yr)
10th percentile
50th percentile
1-day
SWC
(Mg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based on
21-d
SWC
1-day
SWC
(Mg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based on
21-d
SWC
Acute
RQ
(COC:
0*4 Hg/L)
Algae
RQ
(COC: 1
fig/L)
Chronic
RQ
(COC:
0.417
Hg/L)
Acute
RQ
(COC:
0.4 ng/L)
Algae
RQ
(COC: 1
fig/L)
Chronic
RQ
(COC:
0.417
Hg/L)
Section
2.4.1.2-
Repackaging
of Import
Containers
(1)
1.1
100,000
14.70
36.75
14.70
1.71
4.10
0.38
0.96
0.38
0.04
0.09
1.2
1.72
4.30
1.72
1.46
3.50
0.04
0.09
0.04
0.03
0.07
1.3
73.70
184.25
73.70
8.59
20.60
1.93
4.83
1.93
0.18
0.44
1.4
8.69
21.73
8.69
7.35
17.63
0.19
0.47
0.19
0.15
0.36
1.5
15.10
37.75
15.10
1.77
4.24
1.93
4.83
1.93
0.19
0.45
1.6
1.78
4.45
1.78
1.51
3.62
0.19
0.48
0.19
0.16
0.37
1.7
75.60
189.00
75.60
8.85
21.22
9.68
24.20
9.68
0.94
2.26
1.8
8.96
22.40
8.96
7.59
18.20
0.96
2.40
0.96
0.78
1.87
Section
2.4.1.2-
Repackaging
of Import
Containers
(1)
1.1
50,000
14.10
35.25
14.10
0.83
1.99
0.37
0.93
0.37
0.02
0.04
1.2
1.57
3.93
1.57
0.92
2.20
0.04
0.09
0.04
0.02
0.05
1.3
70.50
176.25
70.50
4.15
9.95
1.86
4.65
1.86
0.09
0.21
1.4
7.94
19.85
7.94
4.62
11.08
0.19
0.47
0.19
0.10
0.24
1.5
14.40
36.00
14.40
0.85
2.05
1.87
4.68
1.87
0.09
0.22
1.6
1.62
4.05
1.62
0.95
2.27
0.19
0.47
0.19
0.10
0.24
1.7
72.20
180.50
72.20
4.27
10.24
9.35
23.38
9.35
0.46
1.10
Page 701 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. For this exposure scenario, HBCD is
released following treatment via an onsite WWT or POTW and there are not any sub-scenarios where direct release of HBCD into surface water is expected.
Exposure
Scenario
Sub-
Scenario
Production
Volume
(lbs/yr)
10th percentile
50th percentile
1-day
SWC
(Mg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based on
21-d
SWC
1-day
SWC
(Mg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based on
21-d
SWC
Acute
RQ
(COC:
0*4 Hg/L)
Algae
RQ
(COC: 1
fig/L)
Chronic
RQ
(COC:
0.417
Hg/L)
Acute
RQ
(COC:
0.4 ng/L)
Algae
RQ
(COC: 1
fig/L)
Chronic
RQ
(COC:
0.417
Hg/L)
1.8
8.16
20.40
8.16
4.77
11.44
0.95
2.37
0.95
0.50
1.21
Section
2.4.1.2
Repackaging
of Import
Containers
(1)
1.1
25,000
15.00
37.50
15.00
0.81
1.94
0.40
1.00
0.40
0.02
0.05
1.2
1.46
3.65
1.46
0.41
0.97
0.04
0.09
0.04
0.01
0.02
1.3
75.00
187.50
75.00
4.06
9.74
1.99
4.98
1.99
0.10
0.23
1.4
7.35
18.38
7.35
2.05
4.92
0.19
0.47
0.19
0.04
0.11
1.5
15.40
38.50
15.40
0.83
2.00
2.00
5.00
2.00
0.10
0.23
1.6
1.50
3.75
1.50
0.42
1.00
0.19
0.47
0.19
0.05
0.11
1.7
76.90
192.25
76.90
4.17
10.00
10.00
25.00
10.00
0.48
1.16
1.8
7.54
18.85
7.54
2.11
5.06
0.94
2.35
0.94
0.23
0.55
Page 702 of 723
-------
TableApx J-5. Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (jig/kg) Using PSC (Targeted
Sensitivity Analysis Parameter: Production Volume)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC). Sub-scenarios were removed if there were no RQs calculated based on either 10th or 50th percentile sediment concentration predictions that are >1. For this exposure
scenario, HBCD is released following treatment via an onsite WWT or POTW and there are not any sub-scenarios where direct release of HBCD into surface water is
expected.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
11-d half-life: 10th
percentile
128-d half-life: 10th
percentile
11-d half-life: 50th
percentile
128-d half-life: 50th
percentile
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Section 2.4.1.2 -
Repackaging of
Import
Containers (1)
1.1
100,000
1400
0.89
3620
2.31
34.4
0.02
77
0.05
1.2
1380
0.88
3600
2.29
33.8
0.02
76.7
0.05
1.3
7040
4.48
18200
11.59
172
0.11
386
0.25
1.4
6980
4.45
18200
11.59
170
0.11
385
0.25
1.5
1440
0.92
3730
2.38
174
0.11
395
0.25
1.6
1420
0.9
3720
2.37
171
0.11
393
0.25
1.7
7230
4.61
18700
11.91
872
0.56
1980
1.26
1.8
7170
4.57
18700
11.91
862
0.55
1980
1.26
Section 2.4.1.2 -
Repackaging of
Import
Containers (1)
1.1
50,000
760
0.48
1930
1.23
18.7
0.01
41.3
0.03
1.2
865
0.55
2250
1.43
21.1
0.01
47.8
0.03
1.3
3810
2.43
9660
6.15
93.5
0.06
207
0.13
1.4
4360
2.78
11400
7.26
106
0.07
241
0.15
1.5
781
0.5
1990
1.27
94.5
0.06
212
0.14
1.6
888
0.57
2320
1.48
107
0.07
245
0.16
1.7
3910
2.49
9960
6.34
473
0.3
1060
0.68
1.8
4480
2.85
11700
7.45
538
0.34
1240
0.79
Section 2.4.1.2 -
Repackaging of
Import
Containers (1)
1.1
25,000
512
0.33
1120
0.71
12.8
0.01
24.7
0.02
1.2
347
0.22
902
0.57
8.47
0.01
19.2
0.01
1.3
2560
1.63
5580
3.55
63.9
0.04
123
0.08
1.4
1750
1.11
4550
2.9
42.7
0.03
96.5
0.06
1.5
525
0.33
1150
0.73
64.5
0.04
126
0.08
Page 703 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC). Sub-scenarios were removed if there were no RQs calculated based on either 10th or 50th percentile sediment concentration predictions that are >1. For this exposure
scenario, HBCD is released following treatment via an onsite WWT or POTW and there are not any sub-scenarios where direct release of HBCD into surface water is
expected.
Production
Volume
(lbs/yr)
11-d half-life: 10th
percentile
128-d half-life: 10th
percentile
11-d half-life: 50th
percentile
128-d half-life: 50th
percentile
Condition of
Use
Sub-
Scenario
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
1.6
356
0.23
930
0.59
42.9
0.03
98.3
0.06
1.7
2630
1.68
5740
3.66
323
0.21
630
0.4
1.8
1800
1.15
4690
2.99
216
0.14
495
0
Page 704 of 723
-------
J.l.2.2 Exposure Scenario 3: Processing of HBCD to produce XPS Foam using XPS Masterbatch
Table Apx J-6. Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations (jig/L) Using PSC (Targeted
Sensitivity Analysis Parameters: Production Volume)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. N/A indicates sub-scenarios where HBCD
is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green indicate that these HBCD releases
are due to direct release.
10th percentile
50th percentile
Exposure
Sub-
Production
Volume
(lbs/yr)
%
WW TP
Removal
1-day
SWC
(fig/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(fig/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
Scenario
Scenario
for
Direct
Releases"
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
3.1
0
44.90
112.25
44.90
2.31
5.54
1.20
3.00
1.20
0.06
0.14
3.2
0
3.02
7.55
3.02
0.18
0.43
0.08
0.20
0.08
0.00
0.01
Section
2.4.1.4-
3.3
0
110.00
275.00
110.00
5.65
13.55
2.93
7.33
2.93
0.14
0.34
3.4
0
7.46
18.65
7.46
0.45
1.07
0.20
0.49
0.20
0.01
0.02
Processing
3.5
N/A
4.49
11.23
4.49
0.23
0.55
0.12
0.30
0.12
0.01
0.01
of HBCD to
produce
XPS Foam
3.6
100,000
N/A
0.30
0.76
0.30
0.02
0.04
0.01
0.02
0.01
0.00
0.00
3.7
N/A
11.00
27.50
11.00
0.57
1.35
0.29
0.73
0.29
0.01
0.03
using XPS
3.8
N/A
0.75
1.87
0.75
0.04
0.11
0.02
0.05
0.02
0.00
0.00
Masterbatch
(3)
3.9
N/A
4.60
11.50
4.60
0.24
0.57
0.60
1.50
0.60
0.03
0.07
3.10
N/A
0.31
0.78
0.31
0.02
0.04
0.04
0.10
0.04
0.00
0.00
3.11
N/A
11.30
28.25
11.30
0.58
1.39
1.47
3.68
1.47
0.07
0.17
3.12
N/A
0.77
1.91
0.77
0.05
0.11
0.10
0.25
0.10
0.00
0.01
Section
3.1
0
22.40
56.00
22.40
1.15
2.76
0.60
1.50
0.60
0.03
0.07
2.4.1.4-
Processing
of HBCD to
3.2
50,000
0
1.51
3.78
1.51
0.09
0.21
0.04
0.10
0.04
0.00
0.00
3.3
0
55.40
138.50
55.40
2.84
6.81
1.48
3.70
1.48
0.07
0.17
produce
3.4
0
3.73
9.33
3.73
0.22
0.53
0.10
0.25
0.10
0.00
0.01
Page 705 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. N/A indicates sub-scenarios where HBCD
is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green indicate that these HBCD releases
are due to direct release.
10th percentile
50th percentile
Exposure
Sub-
Production
Volume
(lbs/yr)
%
WW TP
Removal
1-day
SWC
(fig/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(fig/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
Scenario
Scenario
for
Direct
Releases3
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
XPS Foam
3.5
N/A
2.24
5.60
2.24
0.12
0.28
0.06
0.15
0.06
0.00
0.01
using XPS
Masterbatch
(3)
3.6
N/A
0.15
0.38
0.15
0.01
0.02
0.00
0.01
0.00
0.00
0.00
3.7
N/A
5.54
13.85
5.54
0.28
0.68
0.15
0.37
0.15
0.01
0.02
3.8
N/A
0.37
0.93
0.37
0.02
0.05
0.01
0.02
0.01
0.00
0.00
3.9
N/A
2.30
5.75
2.30
0.12
0.28
0.30
0.75
0.30
0.01
0.03
3.10
N/A
0.16
0.39
0.16
0.01
0.02
0.02
0.05
0.02
0.00
0.00
3.11
N/A
5.68
14.20
5.68
0.29
0.70
0.74
1.85
0.74
0.04
0.09
3.12
N/A
0.38
0.96
0.38
0.02
0.05
0.05
0.12
0.05
0.00
0.01
3.1
0
11.20
28.00
11.20
0.57
1.38
0.30
0.75
0.30
0.01
0.03
3.2
0
0.76
1.89
0.76
0.04
0.11
0.02
0.05
0.02
0.00
0.00
Section
3.3
0
27.70
69.25
27.70
1.42
3.41
0.74
1.85
0.74
0.04
0.08
2.4.1.4-
Processing
of HBCD to
3.4
0
1.86
4.65
1.86
0.11
0.26
0.05
0.12
0.05
0.00
0.01
3.5
N/A
1.12
2.80
1.12
0.06
0.14
0.03
0.07
0.03
0.00
0.00
produce
3.6
25,000
N/A
0.08
0.19
0.08
#REF!
#REF!
0.00
0.00
0.00
0.00
0.00
XPS Foam
using XPS
Masterbatch
3.7
N/A
2.77
6.93
2.77
0.00
0.01
0.07
0.18
0.07
0.00
0.01
3.8
N/A
0.19
0.47
0.19
#REF!
#REF!
0.00
0.01
0.00
0.00
0.00
(3)
3.9
N/A
1.14
2.85
1.14
0.06
0.14
0.15
0.37
0.15
0.01
0.02
3.10
N/A
0.08
0.19
0.08
0.00
0.01
0.01
0.03
0.01
0.00
0.00
3.11
N/A
2.84
7.10
2.84
0.15
0.35
0.37
0.93
0.37
0.02
0.04
Page 706 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. N/A indicates sub-scenarios where HBCD
is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green indicate that these HBCD releases
are due to direct release.
Exposure
Scenario
Sub-
Scenario
Production
Volume
(lbs/yr)
%
WW TP
Removal
for
Direct
Releases3
10th percentile
50th percentile
1-day
SWC
(fig/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(fig/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
3.12
N/A
0.19
0.48
0.19
0.01
0.03
0.02
0.06
0.02
0.00
0.00
TableApx 3-1. Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (jig/kg) Using PSC (Targeted
Sensitivity Analysis Parameters: Production Volume)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC). N/A indicates sub-scenarios where HBCD is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green
indicate that these HBCD releases are due to direct release. Sub-scenarios were removed if there were no RQs calculated based on either 10th or 50th percentile sediment
concentration predictions that are >1.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
% WW TP
Removal
for Direct
Releasesa
11-d half-life: 10th
percentile
128-d half-life: 10th
percentile
11-d half-life: 50th
percentile
128-d half-life: 50th
percentile
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
Jig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
Jig/kg
RQ
(COC:
1,570
jig/kg)
Section 2.4.1.4
- Processing of
HBCD to
produce XPS
Foam using
XPS
Masterbatch (3)
3.1
100,000
0
1430
0.91
1910
1.22
36.4
0.02
48.3
0.03
3.2
0
163
0.1
414
0.26
4.01
0
8.86
0.01
3.3
0
3490
2.22
4670
2.97
89.1
0.06
118
0.08
3.4
0
403
0.26
1020
0.65
9.9
0.01
21.9
0.01
3.5
N/A
143
0.09
191
0.12
3.64
0
4.83
0
3.6
N/A
16.3
0.01
41.4
0.03
0.4
0
0.89
0
3.7
N/A
349
0.22
467
0.3
8.91
0.01
11.8
0.01
Page 707 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC). N/A indicates sub-scenarios where HBCD is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green
indicate that these HBCD releases are due to direct release. Sub-scenarios were removed if there were no RQs calculated based on either 10th or 50th percentile sediment
concentration predictions that are >1.
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
% WW TP
Removal
for Direct
Releasesa
11-d half-life: 10th
percentile
128-d half-life: 10th
percentile
11-d half-life: 50th
percentile
128-d half-life: 50th
percentile
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
3.8
N/A
40.3
0.03
102
0.06
0.99
0
2.19
0
3.9
N/A
146
0.09
196
0.12
18.3
0.01
24.4
0.02
3.10
N/A
16.8
0.01
42.7
0.03
2.03
0
4.54
0
3.11
N/A
358
0.23
479
0.31
44.9
0.03
59.7
0.04
3.12
N/A
41.4
0.03
105
0.07
5.01
0
11.2
0.01
Section 2.4.1.4
- Processing of
HBCD to
produce XPS
Foam using
XPS
Masterbatch (3)
3.1
50,000
0
713
0.45
953
0.61
18.2
0.01
24.2
0.02
3.2
0
81.6
0.05
207
0.13
2
0
4.43
0
3.3
0
1760
1.12
2350
1.5
44.9
0.03
59.6
0.04
3.4
0
201
0.13
511
0.33
4.95
0
10.9
0.01
3.5
N/A
71.3
0.05
95.3
0.06
1.82
0
2.42
0
3.6
N/A
8.16
0.01
20.7
0.01
0.2
0
0.44
0
3.7
N/A
176
0.11
235
0.15
4.49
0
5.96
0
3.8
N/A
20.1
0.01
51.1
0.03
0.5
0
1.09
0
3.9
N/A
73.2
0.05
97.8
0.06
9.17
0.01
12.2
0.01
3.10
N/A
8.38
0.01
21.3
0.01
1.01
0
2.27
0
3.11
N/A
181
0.12
241
0.15
22.6
0.01
30.1
0.02
3.12
N/A
20.7
0.01
52.7
0.03
2.5
0
5.61
0
Section 2.4.1.4
- Processing of
HBCD to
produce XPS
Foam using
XPS
Masterbatch (3)
3.1
25,000
0
355
0.23
475
0.3
9.06
0.01
12
0.01
3.2
0
40.8
0.03
104
0.07
1
0
2.21
0
3.3
0
881
0.56
1180
0.75
22.5
0.01
29.8
0.02
3.4
0
101
0.06
256
0.16
2.47
0
5.47
0
3.5
N/A
35.5
0.02
47.5
0.03
0.91
0
1.2
0
3.6
N/A
4.08
0
10.4
0.01
0.1
0
0.22
0
Page 708 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC). N/A indicates sub-scenarios where HBCD is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green
indicate that these HBCD releases are due to direct release. Sub-scenarios were removed if there were no RQs calculated based on either 10th or 50th percentile sediment
concentration predictions that are >1.
% WW TP
Removal
for Direct
Releasesa
11-d half-life: 10th
percentile
128-d half-life: 10th
percentile
11-d half-life: 50th
percentile
128-d half-life: 50th
percentile
Condition of
Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
3.7
N/A
88.1
0.06
118
0.08
2.25
0
2.98
0
3.8
N/A
10.1
0.01
25.6
0.02
0.25
0
0.55
0
3.9
N/A
36.4
0.02
48.7
0.03
4.57
0
6.07
0
3.10
N/A
4.19
0
10.7
0.01
0.51
0
1.14
0
3.11
N/A
90.3
0.06
121
0.08
11.3
0.01
15
0.01
3.12
N/A
10.3
0.01
26.4
0.02
1.25
0
2.8
0
Page 709 of 723
-------
J.l.2.3 Exposure Scenario 5: Processing of HBCD to Produce EPS Foam from Imported EPS Resin Beads
Table Apx J-8 Calculated Risk Quotients based on Estimated HBCD Surface Water Concentrations (jig/L) Using PSC (Targeted
Sensitivity Analysis Parameters: Production Volume)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. N/A indicates sub-scenarios where HBCD
is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green indicate that these HBCD releases
are due to direct release.
Exposure
Scenario
Sub-
Scenario
Production
Volume
(lbs/yr)
%
WW TP
Removal
for
Direct
Releases3
10th percentile
50th percentile
1-day
SWC
(Hg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(Mg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Section
2.4.1.6-
Processing
of HBCD
to produce
EPS Foam
from
Imported
EPS Resin
Beads (5)
5.1
100,000
0
2900.00
7250.00
2900.00
172.00
412.47
76.60
191.50
76.60
3.67
8.80
5.2
N/A
290.00
725.00
290.00
17.20
41.25
7.66
19.15
7.66
0.37
0.88
5.3
N/A
297.00
742.50
297.00
17.70
42.45
38.50
96.25
38.50
1.88
4.51
5.4
0
358.00
895.00
358.00
140.00
335.73
8.78
21.95
8.78
2.94
7.05
5.5
N/A
35.80
89.50
35.80
14.00
33.57
0.88
2.19
0.88
0.29
0.70
5.6
N/A
36.80
92.00
36.80
14.40
34.53
4.44
11.10
4.44
1.51
3.62
5.7
0
3960.00
9900.00
3960.00
235.00
563.55
105.00
262.50
105.00
5.01
12.01
5.8
N/A
396.00
990.00
396.00
23.50
56.35
10.50
26.25
10.50
0.50
1.20
5.9
N/A
406.00
1015.00
406.00
24.20
58.03
52.50
131.25
52.50
2.57
6.16
5.10
0
489.00
1222.50
489.00
191.00
458.03
12.00
30.00
12.00
4.01
9.62
5.11
N/A
48.90
122.25
48.90
19.10
45.80
1.20
3.00
1.20
0.40
0.96
5.12
N/A
50.30
125.75
50.30
19.70
47.24
6.06
15.15
6.06
2.06
4.94
Section
2.4.1.6-
Processing
of HBCD
5.1
50,000
0
2880.00
7200.00
2880.00
157.00
376.50
76.60
191.50
76.60
3.66
8.78
5.2
N/A
289.00
722.50
289.00
15.70
37.65
7.66
19.15
7.66
0.37
0.88
5.3
N/A
296.00
740.00
296.00
16.20
38.85
38.50
96.25
38.50
1.86
4.46
Page 710 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. N/A indicates sub-scenarios where HBCD
is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green indicate that these HBCD releases
are due to direct release.
10th percentile
50th percentile
Exposure
Sub-
Production
Volume
(lbs/yr)
%
WW TP
Removal
1-day
SWC
(Hg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(Mg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
Scenario
Scenario
for
Direct
Releases3
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
to produce
5.4
0
179.00
447.50
179.00
69.90
167.63
4.39
10.98
4.39
1.47
3.53
EPS Foam
from
Imported
5.5
N/A
17.90
44.75
17.90
6.99
16.76
0.44
1.10
0.44
0.15
0.35
5.6
N/A
18.40
46.00
18.40
7.21
17.29
2.22
5.55
2.22
0.76
1.81
EPS Resin
Beads (5)
5.7
0
3940.00
9850.00
3940.00
215.00
515.59
105.00
262.50
105.00
5.00
11.99
5.8
N/A
392.00
980.00
392.00
19.90
47.72
10.50
26.25
10.50
0.50
1.20
5.9
N/A
402.00
1005.00
402.00
20.50
49.16
52.50
131.25
52.50
2.54
6.09
5.10
0
245.00
612.50
245.00
95.50
229.02
5.99
14.98
5.99
2.01
4.82
5.11
N/A
23.20
58.00
23.20
8.27
19.83
0.60
1.50
0.60
0.22
0.52
5.12
N/A
23.80
59.50
23.80
8.49
20.36
3.03
7.58
3.03
1.03
2.47
5.1
0
2880.00
7200.00
2880.00
151.00
362.11
76.60
191.50
76.60
3.66
8.78
Section
5.2
N/A
288.00
720.00
288.00
15.10
36.21
7.66
19.15
7.66
0.37
0.88
2.4.1.6-
5.3
N/A
295.00
737.50
295.00
15.50
37.17
38.50
96.25
38.50
1.85
4.44
Processing
of HBCD
to produce
5.4
0
89.70
224.25
89.70
35.00
83.93
2.20
5.50
2.20
0.74
1.76
5.5
25,000
N/A
8.97
22.43
8.97
3.50
8.39
0.22
0.55
0.22
0.07
0.18
EPS Foam
from
Imported
EPS Resin
5.6
N/A
9.21
23.03
9.21
3.61
8.66
1.11
2.78
1.11
0.38
0.91
5.7
0
3930.00
9825.00
3930.00
205.00
491.61
105.00
262.50
105.00
4.99
11.97
5.8
N/A
7.29
18.23
7.29
1.86
4.46
0.94
2.35
0.94
0.23
0.55
Beads (5)
5.9
N/A
402.00
1005.00
402.00
20.30
48.68
52.50
131.25
52.50
2.53
6.07
5.10
0
122.00
305.00
122.00
47.50
113.91
2.98
7.45
2.98
1.00
2.40
Page 711 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the surface water concentration (SWC) exceeds the
concentration of concern (COC). Underlined SWCs indicate when the predicted surface water release of HBCD for a sub-scenario exceeds the water
solubility of HBCD (66 jig/L), resulting in acute, algae or chronic RQs greater than 165,66, and 158.3, respectively. N/A indicates sub-scenarios where HBCD
is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green indicate that these HBCD releases
are due to direct release.
Exposure
Scenario
Sub-
Scenario
Production
Volume
(lbs/yr)
%
WW TP
Removal
for
Direct
Releases3
10th percentile
50th percentile
1-day
SWC
(fig/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
1-day
SWC
(Mg/L)
RQs based on 1-d
SWC
21-day
SWC:
fig/L
RQs
based
on 21-d
SWC
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
Acute
RQ
(COC:
0.4
fig/L)
Algae
RQ
(COC:
lfig/L)
Chronic
RQ
(COC:
0.417
fig/L)
5.11
N/A
11.50
28.75
11.50
4.12
9.88
0.30
0.75
0.30
0.10
0.24
5.12
N/A
11.80
29.5
11.8
4.23
10.14
1.51
3.775
1.51
0.51
1.23
Page 712 of 723
-------
TableApx J-9 Calculated Risk Quotients based on Estimated HBCD Sediment Concentrations (jig/kg) Using PSC (Targeted
Sensitivity Analysis Parameters: Production Volume)
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC). N/A indicates sub-scenarios where HBCD is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green
indicate that these HBCD releases are due to direct release. Sub-scenarios were removed if there were no RQs calculated based on either 10th or 50th percentile sediment
concentration predictions that are >1.
% WW TP
Removal
for Direct
Releasesa
11-d half-life: 10th
percentile
128-d half-life: 10th
percentile
11-d half-life: 50th
percentile
128-d half-life: 50th
percentile
Condition
of Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
us/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
5.1
0
165000
105.1
417000
265.61
4050
2.58
8910
5.68
5.2
N/A
16500
10.51
41700
26.56
405
0.26
891
0.57
Section
5.3
N/A
16900
10.76
42900
27.32
2050
1.31
4560
2.9
2.4.1.6 -
Processing
of HBCD to
produce
5.4
0
137000
87.26
356000
226.75
3340
2.13
7560
4.82
5.5
N/A
13700
8.73
35600
22.68
334
0.21
756
0.48
5.6
100,000
N/A
14100
8.98
36800
23.44
1690
1.08
3880
2.47
EPS Foam
from
Imported
EPS Resin
5.7
0
225000
143.31
568000
361.78
5530
3.52
12200
7.77
5.8
N/A
22500
14.33
56800
36.18
553
0.35
1220
0.78
5.9
N/A
23100
14.71
58600
37.32
2800
1.78
6230
3.97
Beads (5)
5.10
0
187000
119.11
487000
310.19
4560
2.9
10300
6.56
5.11
N/A
18700
11.91
48700
31.02
456
0.29
1030
0.66
5.12
N/A
19200
12.23
50200
31.97
2330
1.48
5300
3.38
Section
24 16-
5.1
0
102000
64.97
231000
147.13
2560
1.63
5050
3.22
5.2
N/A
10300
6.56
23100
14.71
256
0.16
505
0.32
Processing
5.3
N/A
10500
6.69
23800
15.16
1290
0.82
2590
1.65
of HBCD to
produce
EPS Foam
5.4
0
68500
43.63
178000
113.38
1670
1.06
3780
2.41
5.5
50,000
N/A
6850
4.36
17800
11.34
167
0.11
378
0.24
from
5.6
N/A
7030
4.48
18400
11.72
846
0.54
1940
1.24
Imported
EPS Resin
Beads (5)
5.7
0
140000
89.17
316000
201.27
3500
2.23
6900
4.39
5.8
N/A
13600
8.66
29700
18.92
341
0.22
655
0.42
5.9
N/A
13900
8.85
30600
19.49
1720
1.1
3350
2.13
Page 713 of 723
-------
The bolded and gray highlighted values denote a risk (RQ>1) to the aquatic environment where the sediment HBCD concentration exceeds the concentration of concern
(COC). N/A indicates sub-scenarios where HBCD is released following treatment via an onsite WWT or POTW (not direct release). The sub-scenarios that are shaded green
indicate that these HBCD releases are due to direct release. Sub-scenarios were removed if there were no RQs calculated based on either 10th or 50th percentile sediment
concentration predictions that are >1.
% WW TP
Removal
for Direct
Releasesa
11-d half-life: 10th
percentile
128-d half-life: 10th
percentile
11-d half-life: 50th
percentile
128-d half-life: 50th
percentile
Condition
of Use
Sub-
Scenario
Production
Volume
(lbs/yr)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
Sediment:
fig/kg
RQ
(COC:
1,570
jig/kg)
5.10
0
93500
59.55
243000
154.78
2280
1.45
5160
3.29
5.11
N/A
9350
5.96
24300
15.48
228
0.15
516
0.33
5.12
N/A
9600
6.11
25100
15.99
1160
0.74
2650
1.69
5.1
0
91800
58.47
156000
99.36
2340
1.49
3600
2.29
5.2
N/A
9180
5.85
15600
9.94
234
0.15
360
0.23
Section
5.3
N/A
9420
6
16000
10.19
1180
0.75
1830
1.17
2.4.1.6-
Processing
of HBCD to
5.4
0
34300
21.85
89200
56.82
836
0.53
1890
1.2
5.5
N/A
3430
2.18
8920
5.68
83.6
0.05
189
0.12
produce
5.6
25,000
N/A
3520
2.24
9200
5.86
424
0.27
972
0.62
EPS Foam
from
Imported
5.7
0
125000
79.62
212000
135.03
3190
2.03
4920
3.13
5.8
N/A
1800
1.15
4690
2.99
216
0.14
495
0.32
EPS Resin
Beads (5)
5.9
N/A
12900
8.22
21900
13.95
1610
1.03
2500
1.59
5.10
0
46600
29.68
121000
77.07
1140
0.73
2570
1.64
5.11
N/A
4660
2.97
12100
7.71
114
0.07
257
0.16
5.12
N/A
4780
3.04
12500
7.96
575
0.37
1320
0.84
Page 714 of 723
-------
J.l.2.4 Trophic Transfer: Risk Quotients for Terrestrial Mammals based on KABAM
Table Apx J-10. Chemical Properties: Input Parameters for KABAM (vl) based on Estimated HBCD Surface Water and Sediment
Concentrations (^ig/kg) Using PSC
Condition of Use
Sub-
Scenario
Production
Volume
(lbs/yr)
% WW TP
Removal for
Direct
Releases
Physiochemical
Properties
21-day SWC: jig/L:
10th percentile
128-d half-life:
10th percentile
21-day SWC: jig/L:
50th percentile
128-d half-life:
50th percentile
Log
Kow
Koc
(L/kg
OC)
Surface Water
Concentration:
Dissolved Fraction
(Hg/L)
Pore Water
Concentration
(Hg/L)
Surface Water
Concentration:
Dissolved Fraction
(Hg/L)
Pore Water
Concentration
(Mg/L
Processing:
Manufacturing of
XPS Foam using
XPS Masterbatch
3.3
100,000
75
5.62
100,000
1.067
0.292
0.0264
0.0074
50,000
75
5.62
100,000
0.538
0.147
0.0133
0.00373
25,000
75
5.62
100,000
0.269
0.0735
0.00667
0.00186
Processing:
Manufacturing of
EPS Foam from
Imported EPS
Resin beads
5.7
100,000
75
5.62
100,000
44.353
35.5
0.946
0.758
50,000
75
5.62
100,000
40.56902
19.7
0.946
0.43
25,000
75
5.62
100,000
38.828185
13.3
0.946
0.307
Table Apx J-ll. HBCD Hazard Data: Input Parameters for KABAM (vl)
Avian Toxicity Data
Mammalian Toxicity Data
Avian
Species
Avian
NOAEC
(mg/kg-diet)
Endpoint
References
Data
Evaluation
Score
Mammalian
Species
Mammalian
LOEC (mg/kg-
bw)
Endpoint
References
Data
Evaluation
Score
Japanese
quail
125
Development
(MOEJ
2009)
High
Rat
10
Thyroid
(Ema et al.,
2008)
High
Page 715 of 723
-------
TableApx J-12. Calculated Risk Quotients based on KABAM (vl) based on Estimated HBCD Surface Water and Sediment
Concentrations (^ig/kg) Using PSC
The bolded values denote a risk (RQ>1) to the terrestrial environment, based on input parameters for KABAM (vl).
10th Percentile Surface Water and Sediment
Concentrations
50th Percentile Surface Water and Sediment
Concentrations
Wildlife Species
(COU 3.3) Processing:
Manufacturing of XPS
Foam using XPS
Masterbatch
(COU 5.7) Processing:
Manufacturing of EPS
Foam from Imported EPS
Resin beads
(COU 3.3) Processing:
Manufacturing of XPS
Foam using XPS
Masterbatch
(COU 5.7) Processing:
Manufacturing of EPS
Foam from Imported EPS
Resin beads
Production Volume (lbs/year)
100,000
50,000
25,000
100,000
50,000
25,000
100,000
50,000
25,000
100,000
50,000
25,000
fog/water shrew
0.6
0.3
0.1
23.6
21.2
20.1
0.0
0.0
0.0
0.5
0.5
0.5
Mammalian
Species
rice rat/star-nosed mole
0.8
0.4
0.2
34.4
31.0
29.4
0.0
0.0
0.0
0.7
0.7
0.7
small mink
2.0
1.0
0.5
84.3
75.9
72.1
0.0
0.0
0.0
1.8
1.8
1.8
large mink
2.2
1.1
0.5
93.1
83.8
79.6
0.1
0.0
0.0
2.0
2.0
1.9
small river otter
2.4
1.2
0.6
100.2
90.2
85.7
0.1
0.0
0.0
2.1
2.1
2.1
large river otter
6.2
3.1
1.6
264.7
238.6
226.8
0.2
0.1
0.0
5.6
5.6
5.5
Page 716 of 723
-------
J.1.3
Terrestrial Environment
J.l.3.1 IIOAC Predicted Soil Concentrations via Air Deposition
Table Apx J-13. Calculated Risk Quotients based on Estimated HBCD Soil Concentrations
Hg/kg) Using IIOAC
There are no instances of risk quotients (RQ) that are >1 for the terrestrial soil environment (indicating risk)
where the predicted soil HBCD concentration exceeds the hazard effect concentration for earthworms (173,000 jig
/kg).
Exposure
Scenario
Sub-
Scenario
Fugitive Range
Stack Range
Incineration Range
Soil
Concentration
(jig/kg)
RQ
Soil
Concentration
(US/kg)
RQ
Soil
Concentration
(US/kg)
RQ
Section 2.4.1.2 -
Repackaging of
Import
Containers (1)
Fenceline
1.28E-01
7.40E-07
6.66E-02
3.85E-07
3.42E-03
1.98E-08
Community
3.64E-03
2.10E-08
3.04E-03
1.76E-08
1.29E-03
7.46E-09
Section 2.4.1.3 -
Compounding of
Polystyrene
Resin to Produce
XPS Masterbatch
(2)
Fenceline
2.05E-04
1.18E-09
1.12E-04
6.47E-10
N/A
N/A
Community
5.32E-06
3.08E-11
4.45E-06
2.57E-11
N/A
N/A
Section 2.4.1.4 -
Processing of
HBCD to
produce XPS
Foam using XPS
Masterbatch (3)
Fenceline
2.05E-03
1.18E-08
1.19E-03
6.88E-09
N/A
N/A
Community
4.21E-05
2.43E-10
3.52E-05
2.03E-10
N/A
N/A
Section 2.2.5 -
Processing of
HBCD to
produce XPS
Foam using
HBCD Powder
(4)
Fenceline
2.58E-04
1.49E-09
7.46E-03
4.31E-08
5.98E-04
3.46E-09
Community
5.30E-06
3.06E-11
3.80E-04
2.20E-09
3.35E-04
1.94E-09
Section 2.4.1.6 -
Processing of
HBCD to
produce EPS
Foam from
Imported EPS
Resin Beads (5)
Fenceline
1.34E-01
7.75E-07
7.00E-02
4.05E-07
3.22E-02
1.86E-07
Community
3.64E-03
2.10E-08
3.05E-03
1.76E-08
1.03E-02
5.95E-08
Section 2.4.1.7 -
Processing of
HBCD to
produce SIPs and
Automobile
Replacement
Parts from
XPS/EPS Foam
(6)
Fenceline
6.03E-03
3.49E-08
3.15E-03
1.82E-08
2.03E-02
1.17E-07
Community
1.64E-04
9.48E-10
1.37E-04
7.92E-10
6.48E-03
3.75E-08
Section 2.4.1.9 -
Installation of
XPS/EPS Foam
Insulation in
Fenceline
2.05E-04
1.18E-09
N/A
N/A
9.68E-04
5.60E-09
Community
4.81E-06
2.78E-11
N/A
N/A
1.89E-04
1.09E-09
Page 717 of 723
-------
There are no instances of risk quotients (RQ) that are >1 for the terrestrial soil environment (indicating risk)
where the predicted soil HBCD concentration exceeds the hazard effect concentration for earthworms (173,000 jig
/kg).
Exposure
Scenario
Residential,
Public, and
Commercial
Buildings, and
Other Structures
(8)
Sub-
Scenario
Fugitive Range
Stack Range
Incineration Range
Soil
Concentration
(jig/kg)
RQ
Soil
Concentration
(Mg/kg)
RQ
Soil
Concentration
(Mg/kg)
RQ
Section 2.4.1.10
- Demolition of
XPS/EPS Foam
Insulation
Products in
Residential,
Public and
Commercial
Buildings, and
Other Structures
(9)
Fenceline
5.27E-04
3.05E-09
N/A
N/A
N/A
N/A
Community
1.08E-05
6.24E-11
N/A
N/A
N/A
N/A
Section 2.4.1.11
- Recycling of
EPS Foam and
Reuse of XPS
Foam (10)
Fenceline
1.24E-04
7.17E-10
7.20E-05
4.16E-10
2.14E-05
1.24E-10
Community
2.54E-06
1.47E-11
2.14E-06
1.24E-11
4.50E-06
2.60E-11
Section 2.4.1.12
- Formulation of
Flux/Solder
Pastes (11)
Fenceline
4.99E-04
2.88E-09
6.88E-03
3.98E-08
N/A
N/A
Community
1.65E-05
9.54E-11
2.49E-04
1.44E-09
N/A
N/A
Section2.4.1.13
- Use of
Flux/Solder
Pastes (12)
Fenceline
N/A
N/A
N/A
N/A
2.37E-05
1.37E-10
Community
N/A
N/A
N/A
N/A
5.09E-06
2.94E-11
Page 718 of 723
-------
Appendix K Human Health Risk
K.1 Targeted Sensitivity Analysis
A targeted sensitivity analyses on the impact of import volumes on environmental risk estimates was
performed. The exposure scenarios considered in the sensitivity analysis represent the exposure
scenarios that resulted in the highest estimates of releases on a daily basis and include scenarios that rely
on both industry data and OECD ESDs.
Manufacturing of EPS Foam from Imported EPS Resin beads
Estimation of the risk is below the benchmark MOE for all lifestages only following acute exposure
from the highest exposure sub-scenario (5.7) assuming 100,000 lbs PV and 0% WWT removal. Reduced
PV has essentially no effect on acute exposures and associated risk estimates. Therefore, sensitivity
analysis demonstrates that differing assumptions of production volume has minimal effect on the risk
estimate conclusions for the highly exposed population.
Page 719 of 723
-------
TableApx K-l. Targeted Sensitivity Analysis Based on Production Volume for the Highly Exposed Population Following Acute
Exposure
SCENARIO NAME
Production
Volume
(lbs / year)
1- <2
years
2- <3
years
3-<6
years
6 -<11
years
11-<16
years
16- <70
years
5.7 Manufacturing of EPS Foam from Imported EPS Resin
beads (Highest Exposure)
100,000
14
17
18
24
39
21
5.7 Manufacturing of EPS Foam from Imported EPS Resin
beads (Highest Exposure)
50,000
14
17
19
24
40
21
5.7 Manufacturing of EPS Foam from Imported EPS Resin
beads (Highest Exposure)
25,000
14
17
19
24
40
21
MOEs represent risk from aggregate exposure values from fish ingestion ADR and background general population exposure.
Page 720 of 723
-------
Appendix L Dermal Absorption Estimate Method Comparison
L.l Fraction Absorbed Method As Used in Risk Evaluation
Dexp ~ S x ( Qu x fobs) x Yderm x FT
Where:
S = Surface area of contact (cm2)
Qu = Quantity remaining on the skin (mg/cm2-event)
fabs = Fraction absorbed through the skin
Yderm = Weight fraction of the chemical of interest in the liquid (0 < Yderm < 1)
FT = Frequency of events (integer number per day) - assumed to be 1
Based on three identified studies examining dermal absorption, the highest fractional absorbed value
(fabs) was used and applied to all dermal exposure estimates (6.5% from (Abdallah et al. 2015)).
AAD = PDR x 0.065 -h 80kg = 2.52 mg/kg
Characterization
Yd.™
a
M = Qu xS
Frequency
of Events
Potential
Dose Rate
Acute
Absorbed [>ose
wt
fraction
mg/cm*-
event
mg/event
FT
Img/dayt
AAD^
(mg/kg-dayf
High-end: 90th
percentile
Central Tendency:
Median
1.0
3,100
1
3,100
2.52
FigureApx L-l. Excerpt of Dermal Exposure Results from Repackaging of Import Containers
L.2 Permeability Method
KP is a constant with units cm/hr.
KP X p (density in g/cm3) = maximum flux (Jmax, neat) in g/(cm2-hr).
KP X maximum solubility (in g/cm3) = maximum flux (Jmax, in solvent) in g/(cm2-hr).
KP X experimental concentration (in g/cm3) = steady state flux (Jss, in solvent) in g/(cm2-hr).
According to (Kissel 2011). one should always consider flux instead of simple fractional absorption
when possible in order to account for surface loading and limited time for absorption.
Abdallah et al., 2015
From (Abdallah et al. 2015). the highest reported value of KP in acetone = 2.74E-4 cm/hr (for a
diastereomer).
Jss (steady state flux) = 1.33 ng/(cm2-hr) (or 1.33E-9 g/(cm2-hr)).
Flux can also be calculated from % absorbed dose if the three variables in the equation are known: the
amount of chemical added to the surface, the surface area, and the time allowed for penetration (Kissel
2011).
Jss = (% absorbed) x (Qu, quantity deposited (ng/cm2)) x duration of exposure
Based on the experimental methods of (Abdallah et al. 2015).
Page 721 of 723
-------
Jss = 0.065 x {^^n^/icm2) + 24hr = 1.35 ng/(cm2 — hr), almost identical to the provided
value.
Based on the formula, Jss = KP * C (concentration), with C = 500 ng/ 100 ju.1 = 5 ng/|il.
Jss = 2.74E-4 X 5 ng/|il = 1.5 ng/(cm2-hr), which is also almost identical in value.
Roper et al., 2007
Roper et al., (200?) reports a much lower % absorbed dose value of 0.01%, however when considering
the amount of chemical applied, the calculated steady state flux (Jss) = 4.2 ng/(cm2-hr), over 3x higher
than the flux from Abdallah 2015.
Jss = 0.001 x (lmgA 2) h- 24/ir x IE + 6— = 4.2 ng
\ ' 1C771 J mg cm2-hr
KP = Jss/C;
£ =640nfly^^ =21.3 |ig/ul (as 5 applications of6ul) = 21333333 ng/ml;
Rp= 4.2 ng/(cm2 - hr)/^^ ^ = 1-97E.7 cm/hr
This value could potentially be underestimating flux, because absorption continues after removal of load
(I'rnsch et al. 2014). especially for non-volatile compounds. Therefore, it might be better to assume that
the dermal delivery load could contribute additionally to systemic absorption over time. (Roper et at.
2007) estimated 34.6% of the original dose initially retained in the skin, with 1.35% dermal delivery
remaining after 24h following washing and drying. Additionally, the KP value may be inaccurate, as a
specific concentration was not provided and instead needed to be approximated by adding 5 separate
aliquots of HBCD dissolved in acetone.
L.3 Method Comparison
L.3.1 Occupational Exposure Using Flux
Because dermal load may contribute to additional absorption over time, it is reasonable to use 24hr as a
higher-end estimate on the time duration variable to account for continued absorption.
ChemSTEER uses a value of 1070 cm2 as the surface area of both hands for calculating exposure.
Based on these two factors, 4.2 ng/(cm2-hr) X 1070cm2 X 24hr = 107856 ng = 107.86 jug = 0.108 mg as
the amount of HBCD absorbed at steady-state flux.
PDR = 0.108 mg / 80 kg = 1.35E-03 mg/kg , which is over 1800-fold less than the amount of HBCD
absorbed by the fraction absorbed method (see Figure Apx L-l).
One cannot calculate Jmax, maximum flux, without the maximum solubility of HBCD in acetone.
According to a commercial SDS,
(https://www.siemaaldrich.com/catalog/product/aldrich 2?lang=en®ion=US) and
ChemicalBook.com (https://www.chemicalbook.com/ChemicalProductPropertv EN CB73 htm),
the solubility of HBCD in acetone is 25 mg/ml. The permeability of HBCD would be significantly lower
in water, however permeability could be higher in certain formulations or through oily skin (Pawar et
al. 2.016).
Page 722 of 723
-------
Using the higher KP value of the two studies (and the one directly measured, from (Abdallah et al.
2015). we can calculate J max.
KP in acetone = 2.74E-4 cm/hr;
KP X 25 mg/ml = 6.85E-3 mg/(cm2-hr) or 6.85 |ig/(cm2-hr). This is over lOOOx higher than the steady
state flux Jss calculated from either study, indicating that testing a higher concentration of HBCD would
have resulted in greater measured flux.
6.85 |ig/(cm2-hr) X 1070cm2 X 24hr = 1.76E5 jug = 175.9 mg absorbed.
PDR = 175.9 mg / 80 kg = 2.2 mg/kg, which is very similar to the originally estimated dose of 2.52
mg/kg (Figure Apx L-l). Therefore, while both of these calculations represent very high-end
conservative estimates, it can be concluded that the upper bound of dermal absorption estimates is
consistent between the fraction absorbed and permeability methods.
L.3.2 General Population Considerations
Use of fractional absorption is appropriate for general population or consumer exposure estimates,
where exposure is assumed to be continuous and sustained over time. In that case, there would be an
infinite time variable and the flux rate would be irrelevant. Therefore, the steady state fraction absorbed
is suitable for this use.
Page 723 of 723
------- |