1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
v>EPA
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
EPA Document# EPA-740-D-23-001
July 2023
United States Office of Chemical Safety and
Environmental Protection Agency Pollution Prevention
Draft Supplement to the Risk Evaluation for
1,4-Dioxane
CASRN 123-91-1
July 2023
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
29 TABLE OF CONTENTS
30 ACKNOWLEDGEMENTS 17
31 EXECUTIVE SUMMARY 19
32 1 INTRODUCTION 25
33 1.1 Regulatory Context 25
34 1.2 Scope 27
35 1.3 Use Characterization 28
36 1.3,1 Conceptual Models 28
37 1.3.1.1 1,4-Dioxane as a Byproduct 30
38 1.3.1.2 Occupational Exposures 32
39 1.3.1.3 General Population Exposures 34
40 1.3.1.3.1 Drinking Water 36
41 1.3.1.3.2 Air 37
42 1.3.1.3.3 Aggregate Exposure 37
43 1,3,2 Potentially Exposed or Susceptible Subpopulations 37
44 1.4 Systematic Review 38
45 1.5 Document Outline 38
46 2 RELEASES AND CONCENTRATIONS 40
47 2.1 Approach and Methodology 40
48 2.1,1 Industrial and Commercial Releases 40
49 2.1.1.1 General Approach and Methodology for Environmental Releases 42
50 2.1.1.2 Water Release Estimates 42
51 2.1.1.3 Land Release Estimates 43
52 2.1.1.4 Air Release Estimates 44
53 2.1.1.4.1 Pre-screening Analysis 44
54 2.1.1.4.2 Single-Year Fenceline Analysis 44
55 2.1.1.4.3 Multi-year Analysis 45
56 2.2 Environmental Releases 45
57 2.2,1 Industrial and Commercial Releases 45
58 2.2.1.1 Release Estimates Summary 45
59 2.2.1.2 Weight of the Scientific Evidence Conclusions for Environmental Releases 49
60 2.2.1.3 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
61 Environmental Release Assessment 50
62 2.3 1,4-Dioxane Environmental Concentrations 52
63 2.3,1 Surface Water Pathway 52
64 2.3.1.1 Monitoring Data 52
65 2.3.1.2 Surface Water and Drinking Water Modeling 58
66 2.3.1.2.1 Modeling Methodology 58
67 2.3.1.2.2 EstimatingDown-the-DrainReleases 60
68 2.3.1.2.3 Hydraulic Fracturing 60
69 2.3.1.2.4 Proximity to Drinking Water Sources 60
70 2.3.1.3 Modeling Results 60
71 2.3.1.3.1 Facility-Specific Results 60
72 2.3.1.3.2 Concentrations from Down-the-Drain Loading 66
73 2.3.1.3.3 Concentrations from Hydraulic Fracturing 67
74 2.3.1.3.4 Aggregate Probabilistic Results 68
Page 2 of 484
-------
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2.3.1.4 Comparison of Modeled and Monitored Surface Water Concentrations 71
2.3.1.5 Strengths, Limitations, and Sources of Uncertainty in Assessment Results for
Monitored and Modeled Drinking Water and Surface Water Concentrations 72
2.3.2 Land Pathway (Groundwater) 73
2.3.2.1 Groundwater Monitoring Data 73
2.3.2.2 Disposal via Underground Injection 75
2.3.2.2.1 Summary of Assessment for Disposal to Underground Injection 76
2.3.2.2.2 Strengths, Limitations, and Sources of Uncertainty in Assessment of Disposal to
Underground Injection Wells 76
2.3.2.3 Disposal to Landfills 76
2.3.2.3.1 Summary of Assessment for Disposal to Landfills 80
2.3.2.3.2 Strengths, Limitations, and Sources of Uncertainty in Assessment Results for
Disposal to Landfills 80
2.3.2.4 Disposal of Hydraulic Fracturing Produced Water to Surface Impoundments 81
2.3.2.4.1 Summary of Assessment for Disposal of Hydraulic Fracturing Produced Water.... 81
2.3.2.4.2 Strengths, Limitations, and Sources of Uncertainty in Assessment Results for
Disposal from Hydraulic Fracturing Operations 82
2.3.3 Ambient Air Pathway 82
2.3.3.1 Measured Concentrations in Air 83
2.3.3.2 Modeled Concentrations in Air 83
2.3.3.2.1 Ambient Air: Screening Methodology 84
2.3.3.2.2 Ambient Air: Single Year Methodology (AERMOD) 84
2.3.3.2.3 Ambient Air: Multi-Year Analysis (IIOAC) 88
2.3.3.2.4 Ambient Air: IIO AC Methodology for COUs without Site-Specific Data
(Hydraulic Fracturing, Industrial, and Institutional Laundry Facilities) 88
2.3.3.3 Strengths, Limitations, and Sources of Uncertainty for Modeled Air Concentrations.... 89
3 HUMAN EXPOSURES 91
3.1 Occupational Exposures 91
3.1.1 Approach and Methodology 92
3.1.1.1 Process Description, Number of Sites, Number of Workers, and ONUs 92
3.1.1.2 Inhalation Exposures Approach and Methodology 93
3.1.1.3 Dermal Exposures Approach and Methodology 94
3.1.1.4 Engineering Controls and Personal Protective Equipment 94
3.1.2 Occupational Exposure Estimates 95
3.1.2.1 Summary of Inhalation Exposure Assessment 95
3.1.2.2 Summary of Dermal Exposures Assessment 96
3.1.2.3 Weight of the Scientific Evidence Conclusions for Occupational Exposure
Information 97
3.1.2.4 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Occupational Exposure Assessment 99
3.1.2.4.1 Number of Workers 99
3.1.2.4.2 Analysis of Inhalation Exposure Monitoring Data 99
3.1.2.4.3 Modeled Inhalation Exposures 100
3.1.2.4.4 Modeled Dermal Exposures 100
3.2 General Population Exposures 101
3.2.1 Approach and Methodology 102
3.2.2 Drinking Water Exposure Assessment 103
3.2.2.1 Surface Water Exposure Assessment 103
Page 3 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
123 3.2.2.1.1 Exposures from Individual Facility Releases 103
124 3.2.2.1.2 Exposures from Down-the-Drain Releases 106
125 3.2.2.1.3 Disposal of Hydraulic Fracturing Produced Waters 106
126 3.2.2.1.4 Aggregate Exposure 107
127 3.2.2.2 Groundwater Exposure Assessment 108
128 3.2.2.2.1 Disposal to Landfills 108
129 3.2.2.2.2 Disposal of Hydraulic Fracturing Produced Waters 109
130 3.2.3 Air Exposure Assessment 110
131 3.2.3.1 Industrial COUs Reported to TRI 110
132 3.2.3.2 Hydraulic Fracturing 112
133 3.2.3.3 Industrial and Institutional Laundry Facilities 115
134 3.3 Weight of the Scientific Evidence Conclusions 116
135 3.3.1 Occupational Exposures 116
136 3.3.1.1 Inhalation Exposure 116
137 3.3.1.2 Dermal Exposure 117
138 3.3,2 Drinking Water 118
139 3.3.2.1 Drinking Water Exposure Estimates Based on Surface Water Concentrations 118
140 3.3.2.2 Drinking Water Exposure Estimates Based on Groundwater Concentrations 122
141 3.3.2.2.1 Groundwater Concentrations Resulting from Disposal to Landfill 123
142 3.3.2.2.2 Groundwater Concentrations Resulting from Disposal of Hydraulic Fracturing
143 Waste 124
144 3.3.3 Air 124
145 3.3.3.1 Modeled Air Concentrations for Industrial COUs Reported to TRI 125
146 3.3.3.2 Air Concentrations Modeled near Hydraulic Fracturing Operations and
147 Industrial/Institutional Laundries 126
148 4 HUMAN HEALTH HAZARD 128
149 4.1 Summary of Hazard Endpoints Previously Identified in the 2020 Risk Evaluation 128
150 4.2 Summary of Adjustments to Previously Established Hazard Values 128
151 4.2.1 Derivation of Acute/Short-Term Hazard Values 131
152 4.2.1.1 Inhalation HEC 131
153 4.2.1.2 Oral and Dermal HEDs 131
154 4.2.2 Derivation of Chronic Hazard Values 131
155 4.2.2.1 Inhalation HEC 131
156 4.2.2.2 Oral HEDs 132
157 4.2.2.3 Dermal HEDs 132
158 4.2.3 Derivation of Cancer Hazard Values 132
159 4.3 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty in the Hazard and
160 Dose-Response Analysis 133
161 5 HUMAN HEALTH RISK CHARACTERIZATION 134
162 5.1 Risk Characterization Approach 135
163 5,1.1 Estimation of Non-cancer Risks 136
164 5.1.2 Estimation of Cancer Risks 136
165 5.2 Human Health Risk Characterization 136
166 5.2.1 Summary of Risk Estimates for Occupational Exposures 136
167 5.2,2 Summary of Risk Estimates for the General Population 140
168 5.2.2.1 Drinking Water - Surface Water Pathway 140
Page 4 of 484
-------
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
5.2.2.1.1 Risks from Exposure to Drinking Water Concentrations Indicated in Finished
Drinking Water Monitoring Data 141
5.2.2.1.2 Risks from Exposures to Water Concentrations Modeled from Industrial
Releases 141
5.2.2.1.3 Risks from Exposures to Water Concentrations Modeled from DTD Releases
(from POTWs), Assuming No Downstream Dilution 146
5.2.2.1.4 Risks from Exposure to Drinking Water Concentrations Modeled from Disposal
of Hydraulic Fracturing Produced Waters to Surface Water, Assuming No
Downstream Dilution 147
5.2.2.1.5 Aggregate Risks from Drinking Water Exposures Modeled from Multiple
Sources Releasing to Surface Water, Assuming No Downstream Dilution 148
5.2.2.1.6 Integrated Summary of Drinking Water Risk Estimates across Multiple Lines of
Evidence for Surface Water 151
5.2.2.2 Drinking Water - Groundwater and Disposal Pathways 152
5.2.2.3 Air Pathway 154
5.2.2.3.1 Industrial COUs Reported to TRI 154
5.2.2.3.2 Hydraulic Fracturing 159
5.2.2.3.3 Industrial and Institutional Laundry Facilities 161
5.2.2.4 Potentially Exposed or Susceptible Subpopulations 162
5.2.2.5 Aggregate and Sentinel Exposures 164
5.2.2.6 Summary of Overall Confidence and Remaining Uncertainties in Human Health Risk
Characterization 165
5.2.2.6.1 Risks from Occupational Exposures 165
5.2.2.6.2 Risks from General Population Exposures through Drinking Water 166
5.2.2.6.3 Risks from General Population Exposures through Groundwater and Land
Disposal Pathways 167
5.2.2.6.4 Risks from General Population Exposures through Air 167
REFERENCES 169
APPENDICES 177
Appendix A ABBREVIATIONS AND ACRONYMS 177
Appendix B LIST OF SUPPLEMENTAL DOCUMENTS 180
Appendix C SYSTEMATIC REVIEW PROTOCOL FOR THE DRAFT SUPPLEMENT TO
THE RISK EVALUATION FOR 1,4-DIOXANE 183
C.l Clarifications and Updates to the 2021 Draft Systematic Review Protocol 184
C.l.l Clarifications and Updates 184
C.l Data Search 187
C.2.1 Multi-Disciplinary Updates to the Data Search 187
C.2.2 Additional Data Sources Identified 188
C.2.2.1 Additional Data Sources Identified for Environmental Release and Occupational
Exposure 188
C.2.2.2 Additional Data Sources Identified for General Population, Consumer, and
Environmental Exposure 189
C.2.3 Search Strings 189
C.2.3.1 Environmental Release and Occupational Exposure Search Strings 189
C.2.3.2 General Population, Consumer, and Environmental Exposure Search Strings 190
C.3 Data Screening 191
Page 5 of 484
-------
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
C.3.1 Environmental Release and Occupational Exposure 192
C.3.1.1 Environmental Release and Occupational Exposure Literature Tree 193
C.3.2 General Population, Consumer, and Environmental Exposure 193
C.3.2.1 General Population, Consumer, and Environmental Exposure Literature Tree 194
C.4 Data Evaluation and Data Extraction 194
C.4.1 Environmental Release and Occupational Exposure 195
C.4.2 General Population, Consumer, and Environmental Exposure 195
C.4.2.1 Data Quality Evaluation Metric Updates 196
C.4.2.2 Data Evaluation Criteria for Monitoring Data, as Revised 197
C.4.2.3 Data Evaluation Criteria for Experimental Data, as Revised 204
C.4.2.4 Data Evaluation Criteria for Databases, as Revised 210
C. 5 Evi dence Integrati on 214
C.5.1 Environmental Release and Occupational Exposure 219
C.5.2 General Population 219
C.5.2.1 General Population: Surface Water 219
C.5.2.2 General Population: Groundwater 219
C.5.2.3 General Population Exposure: Ambient Air 219
Appendix D COU-OES MAPPING AND CROSSWALK 221
D.l COU-OES Mapping 221
D,2 COU-OES Crosswalk 223
Appendix E INDUSTRIAL AND COMMERCIAL ENVIRONMENTAL RELEASES 226
E.l Estimates of the Number of Industrial and Commercial Facilities with Environmental
Releases 226
E.2 Estimates of Number of Release Days for Industrial and Commercial Releases 229
E.3 Water Release Assessment 231
E.3.1 Assessment Using TRI and DMR 232
E.3.2 Assessment for OES without TRI and DMR 234
E.3.3 Water Release Estimates Summary 237
E.3.4 Summary of Weight of the Scientific Evidence Conclusions in Water Release Estimates. 243
E,4 Land Release Assessment 251
E.4.1 Assessment Using TRI 251
E.4.2 Assessment for OES without TRI 253
E.4.3 Land Release Estimates Summary 259
E.4.4 Summary of Weight of the Scientific Evidence Conclusions in Land Release Estimates.. 263
E.5 Air Release Assessment 271
E.5.1 Assessment Using TRI 271
E.5.2 Assessment for OESd without TRI 273
E.5.3 Air Release Estimates Summary 279
E.5.4 Summary of Weight of the Scientific Evidence Conclusions in Air Release Estimates 285
E.6 Comparison to PET Life Cycle Analysis 291
E.7 Detailed Strengths, Limitations, Assumptions and Key Sources of Uncertainties for the
Environmental Release Assessment 292
E.8 Weight of the Scientific Evidence Conclusions for Environmental Releases 296
E,9 TRI to CDR Crosswalk 300
E, 10 Developing Models that Use Monte Carlo Methods 318
E.10.1 Background on Monte Carlo Methods 318
E.10.2 Implementation of Monte Carlo Methods 318
Page 6 of 484
-------
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
319
319
319
323
323
323
323
324
326
329
329
329
329
329
329
329
330
330
330
330
330
330
331
333
339
343
343
343
343
343
344
344
344
344
344
345
345
345
345
345
345
345
346
346
346
347
351
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.10.3 Building the Model
E.10.3.1 Build the Deterministic Model
E.10.3.2 Define Probability Distributions for Input Parameters
E.10.3.3 Select Model Outputs for Aggregation of Simulation Results
E.10.3.4 Select Simulation Settings and Run Model
E.10.3.5 Aggregate the Simulation Results and Produce Output Statistics
E, 11 Textile Dye Modeling Approach and Parameters for Estimating Environmental Releases
E. 11.1 Model Equations
E. 11.2 Model Input Parameters
E. 11.3 Number of Sites
E.11.4 Mass Fraction of Dye Containing 1,4-Dioxane
E. 11.5 Operating Days
E.l 1.6 Mass Fraction of 1,4-Dioxane in Dye Formulation
E. 11.7 Textile Production Rate
E. 11.8 Mass Fraction of Textiles Treated with Dye
E.l 1.9 Mass Fraction of Dye Used per Mass of Textile Dyed
E. 11.10 Mass Fraction of the Dye Formulation in the Dyebath
E. 11.11 Container Size for Dye Formulation
E. 11.12 Container Residual Fraction for Totes
E. 11.13 Container Residual Fraction for Drums
E. 11.14 Container Residual Fraction for Pails
E. 11.15 Fraction of Dye Product Affixed to Textile during Dyeing Process Substrate
E, 12 Laundry Detergent Modeling Approach and Parameters for Estimating Environmental
Releases
E.12.1 Model Equations
E.12.2 Model Input Parameters
E.12.3 Operating Days
E.12.4 Mass Fraction of 1,4-Dioxane in Laundry Detergent
E.12.5 Daily Use Rate of Detergent
E.12.6 Container Size
E.12.7 Indoor Air Speed
E.12.8 Container Residual Fraction for Totes
E. 12.9 Container Residual Fraction for Drums
E. 12.10 Container Residual Fraction for Pails
E. 12.11 Container Residual Fraction for Powders
E. 12.12 Fraction of Laundry Detergents Containing 1,4-Dioxane
E.12.13 Duration of Release for Container Unloading
E. 12.14 Fraction of Chemical Lost during Transfer of Solid Powders
E. 12.15 Control Efficiency for Dust Control Methods
E. 12.16 Capture Efficiency for Dust Capture Methods
E.12.17 Number of Sites
E. 12.18 Diameter of Container Opening
E.12.19 Diameter of Wash Opening
E.12.20 Dilution Factor
E. 12.21 Container Fill Rate
E.l3 Hydraulic Fracturing Modeling Approach and Parameters for Estimating Environmental
Releases
E.13.1 Model Equations
E.13.2 Model Input Parameters
Page 7 of 484
-------
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
354
354
354
354
354
354
354
354
355
355
355
355
355
356
356
357
357
357
360
363
363
369
372
376
380
381
387
392
393
397
401
405
408
411
411
413
416
416
416
416
416
416
417
417
417
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.13.3 Number of Sites
E.13.4 Operating Days
E.13.5 Container Size
E.13.6 Diameter of Container Opening
E.13.7 Diameter of Equipment Opening
E.13.8 Air Speed during Equipment Cleaning
E.13.9 Equipment Cleaning Loss Fraction
E. 13.10 Container Fill Rate
E. 13.11 Equipment Cleaning Operating Hours
E.13.12 Annual Use Rate of Fracturing Fluids Containing 1,4-Dioxane
E.13.13 Mass Fraction of 1,4-Dioxane in Hydraulic Fracturing Additive/Fluid
E.13.14 Saturation Factor
E.13.15 Container Residual Fraction for Totes
E.13.16 Container Residual Fraction for Drums
E. 13.17 Fraction of Inj ected Fracturing Fluid that Returns to the Surface
Appendix F OCCUPATIONAL EXPOSURES
F.l Calculating Acute and Chronic Inhalation Exposures and Dermal Doses
F.2 Approach for Estimating Number of Workers and Occupational Non-users
F.3 Occupational Dermal Exposure Assessment Method
F.4 Occupational Exposure Scenarios
F.4.1 Textile Dye
F.4.2 Antifreeze
F.4.3 Surface Cleaner
F.4.4 Dish Soap
F. 4.5 Di shwasher D etergent
F.4.6 Laundry Detergent (Industrial and Institutional)
F.4.7 Paint and Floor Lacquer
F.4.8 Spray Foam Application
F.4.9 Polyethylene Terephthalate Byproduct
F.4.10 Ethoxylation Process Byproduct
F.4.11 Hydraulic Fracturing
F,5 Summary of Occupational Inhalation Exposures
F.6 Summary of Weight of the Scientific Evidence Conclusions in Inhalation Exposure
Estimates
F.7 Antifreeze Modeling Approach and Parameters for Estimating Occupational Inhalation
Exposures
F.7.1 Model Equations
F.7.2 Modeling Input Parameters
F.7.3 Container Size
F.7.4 Jobs per Day
F.7.5 Concentration of 1,4-Dioxane in Antifreeze
F.7.6 Ventilation Rate
F.7.7 Mixing Factor
F. 7.8 S aturati on Factor
F.7.9 Consumer Use Rate of Antifreeze
F.7.10 Container Fill Rate
F,8 Laundry Detergent Modeling Approach and Parameters for Estimating Occupational
Inhalation Exposures
Page 8 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
359 F.8.1 Model Equations 419
360 F.8.2 Model Input Parameters 422
361 F.8.3 Ventilation Rate 424
362 F.8.4 Mixing Factor 424
363 F.8.5 Total Particulate Concentration 424
364 F.8.6 Respirable Particulate Concentration 424
365 F.9 Hydraulic Fracturing Modeling Approach and Parameters for Estimating Occupational
366 Inhalation Exposures 425
367 F.9.1 Model Equations 426
368 F.9.2 Model Input Parameters 430
369 F.9.3 Ventilation Rate 432
370 F.9.4 Mixing Factor 432
371 Appendix G SURFACE WATER CONCENTRATIONS 433
372 G.l Surface Water Monitoring Data 433
373 G. 1.1 Monitoring Data Retrieval and Processing 433
374 G. 1.2 Raw and Finished Drinking Water 434
375 G.2 Surface Water Modeling 435
376 G.2.1 Hydrologic Flow Data 435
377 G.2.2 Facility-Specific Release Modeling 436
378 G.2.3 Aggregate and Probabilistic Modeling 437
379 G.2.3.1 The Fit-For-Purpose Aggregate Surface Water Model 437
380 G.2.3.2 Case Studies to Validate Aggregate Model 440
381 G.2.3.3 The Probabilistic Model 444
382 G.2.3.4 Modeling Ranges of DTD Contributions 446
383 G.2.3.5 Modeling Concentrations in Surface Water from Hydraulic Fracturing 448
384 G.2.4 Assessing Downstream Drinking Water Intakes 450
385 Appendix H GROUNDWATER CONCENTRATIONS AND DISPOSAL PATHWAYS
386 FROM LAND RELEASES 455
387 H. 1 Groundwater Monitoring Data Retrieval and Processing 455
388 H,2 Review of Land Release Permits 455
389 H.3 Landfill Analysis Using DRAS 457
390 H.4 Landfill Analysis Using EPACMTP 460
391 H.5 Surface Impoundment Analysis for the Disposal of Hydraulic Fracturing Produced Water
392 Using DRAS 463
393 Appendix I DRINKING WATER EXPOSURE ESTIMATES 466
394 1.1 Surface Water Sources of Drinking Water 467
395 1.2 Groundwater Sources of Drinking Water 467
396 Appendix J AIR EXPOSURE PATHWAY 468
397 J.l Ambient Air Concentrations and Exposures 468
398 J. 1.1 Ambient Air: Screening Methodologies and Results Summary - Fenceline 468
399 J. 1.2 Ambient Air: IIOAC Methodology and Results for COUs without Site-Specific Data
400 (Hydraulic Fracturing, Industrial, and Institutional Laundry Facilities) 471
401 J. 1.3 Ambient Air: Single Year Methodology (AERMOD) 473
402 J. 1.4 Ambient Air: Multi-Year Analysis Methodology (IIOAC) 476
403 J,2 Inhalation Exposure Estimates for Fenceline Communities 477
404 J.3 Land Use Analysis 478
Page 9 of 484
-------
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
J,4 Aggregate Analysis across Facilities 479
LIST OF TABLES
Table 2-1. Additional Categories and Subcategories of COUs and Associated OESs Included in the
Scope of the Draft Supplement Due to the Presence of 1,4-Dioxane Produced as a
Byproduct 41
Table 2-2. Summary of PWS Monitoring Datasets of 1,4-Dioxane Monitoring in PWSs Using
Surface Water as a Source 55
Table 2-3. Summary of Surface Water Concentration Results by OES from Facility-Specific
Modeling of Annual Maximum Releases between 2013 and 2019 for 1 Operating Day
per Year 62
Table 2-4. Summary of Surface Water Concentration Results by OES for Facility-Specific Modeling
of Annual Maximum Releases between 2013 and 2019 for the Maximum Operating
Days per Year 63
Table 2-5. OES-COU Crosswalk for Identified Facilities Releasing to Surface Water 64
Table 2-6. Summary by OES of Data Sources for Releases and Receiving Water Body Flow 65
Table 2-7. Hypothetical Mean Annual Concentrations (|ig/L) for a Range of Annual Release and
Flow Rate Combinations, for a Facility with 250 Days of Release per Year 66
Table 2-8. Occurrence of Facilities for Distributions of Maximum Annual 1,4-Dioxane Release
Amounts and Receiving Water Body Flow 66
Table 2-9. Estimated Surface Water Concentrations (|ig/L) Due to DTD Loading for a Range of
Populations and Hydrologic Flows 67
Table 2-10. Estimated Percent Occurrence of Combinations of Contributing Population to POTWs
and Receiving Water Body Flow, from Combined ICIS-NPDES and 2020 Census
Data 67
Table 2-11. Distribution of Potential Concentrations in Surface Water Resulting from Hydraulic
Fracturing Operations from a Single Site Reporting 1,4-Dioxane as an Ingredient 68
Table 2-12. Aggregate Probabilistic Results Showing Distribution of Total 1,4-Dioxane
Concentration in Surface Water (Release Plus Background) 69
Table 2-13. Potential Groundwater Concentrations (|ig/L) of 1,4-Dioxane Found in Wells within 1
Mile of a Disposal Facility Determined by Using the DRAS Model 79
Table 2-14. Total Annual Release Summary 82
Table 2-15. Summary of Select Statistics for the 95th Percentile Estimated Annual Average
Concentrations from the "Full-Screening" Analysis for 1,4-Dioxane Releases
Reported to TRI 86
Table 3-1. Estimated Dermal Absorbed Dose (mg/day) for Workers in Various Conditions of Use 96
Table 3-2. Summary of the Weight of the Scientific Evidence for Occupational Exposure Estimates
by OES 98
Table 3-3. Adult and Infant Exposures Estimated from Facility-Specific Releases 104
Table 3-4. Adult LADD Exposures (mg/kg/day) Estimated from 1,4-Dioxane DTD Consumer and
Commercial Releases 106
Table 3-5. Adult ADR, ADD, and LADD Exposures Estimated from Disposal of Hydraulic
Fracturing Produced Waters to Surface Water 107
Table 3-6. Adult LADD Exposures from Aggregate Concentrations Estimated Downstream of
Release Sites (Including DTD Releases and Direct and Indirect Industrial Releases)... 107
Table 3-7. Adult LADD Exposures Estimated from Groundwater Contamination from Landfills
under Varying Landfill Conditions 109
Page 10 of 484
-------
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 3-8. Estimated Exposures Resulting from Groundwater Contamination from Disposal of
Hydraulic Fracturing Produced Water 109
Table 3-9. Lifetime Average Daily Concentrations Estimated within 10 km of 1,4-Dioxane Releases
to Air Ill
Table 3-10. Exposures from Fugitive Emissions Estimated within 1,000 m of Hydraulic Fracturing
Operations 113
Table 3-11. Exposures from Fugitive Emissions Estimated near Industrial and Institutional Laundry
Facilities 115
Table 4-1. Hazard Values Used for 1,4-Dioxane in this Draft Supplement 130
Table 5-1. Use Scenarios, Populations of Interest, and Toxicological Endpoints Used for Acute and
Chronic Exposures 135
Table 5-2. Lifetime Cancer Risk Estimates for 1,4-Dioxane Concentrations Detected in Finished
Drinking Water 141
Table 5-3. Proximity of Nearest Downstream Drinking Water Intakes to Facilities Resulting in
Cancer Risk Greater than 1 x 10~6 143
Table 5-4. Lifetime Cancer Risk Estimates from DTD Releases Alone (at the Point of Release) under
a Range of Population and Flow Rate Scenarios 147
Table 5-5. Lifetime Cancer Risks Estimated from Hydraulic Fracturing Produced Waters Disposed .. 148
Table 5-6. Lifetime Cancer Risks Estimated for Modeled Groundwater Concentrations Estimated
under Varying Landfill Conditions 153
Table 5-7. Lifetime Cancer Risks Estimated for Modeled Groundwater Concentrations Resulting
from Disposal of Hydraulic Fracturing Produced Water 154
Table 5-8. Inhalation Lifetime Cancer Risksa within 10 km of Industrial Air Releases Based on 95th
Percentile Modeled Exposure Concentrations 156
Table 5-9. Lifetime Cancer Risk Estimates for Fugitive Emissions from Hydraulic Fracturing 160
Table 5-10. Lifetime Cancer Risk Estimates for Fugitive Emissions from Industrial and Institutional
Laundry Facilities 161
Table 5-11. Summary of PESS Considerations Incorporated throughout the Analysis and Remaining
Sources of Uncertainty 162
LIST OF FIGURES
Figure 1-1. 1,4-Dioxane Life Cycle Diagram 29
Figure 1-2. Production of 1,4-Dioxane as a Byproduct and Potential Exposure Pathways 30
Figure 1-3. Conceptual Model for Occupational Exposures from Industrial and Commercial
Activities 33
Figure 1-4. Conceptual Model for Environmental Releases and General Population Exposures 35
Figure 1-5. Overview of Analyses Included in this Draft Supplement to the Risk Evaluation for 1,4-
Dioxane 39
Figure 2-1. Overview of EPA's Approach to Estimate Daily Releases for Each OES 42
Figure 2-2. 1,4-Dioxane Annual Water Releases as Reported to TRI and DMR, 2013-2019 46
Figure 2-3. 1,4-Dioxane Annual Releases to Land as Reported to TRI, 2013-2019 47
Figure 2-4. 1,4-Dioxane Annual Releases to Air as Reported by TRI, 2013-2019 48
Figure 2-5. Locations of Hydraulic Fracturing Operations that Report 1,4-Dioxane in Produced
Waters 49
Figure 2-6. Frequency of Nationwide Measured 1,4-Dioxane Surface Water Concentrations Retrieved
from the Water Quality Portal, 1997-2022 53
Figure 2-7. Frequency of Detection Limits for Nationwide Non-detect 1,4-Dioxane Surface Water
Samples Retrieved from the Water Quality Portal, 1997-2022 54
Page 11 of 484
-------
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Figure 2-8. Detectable Concentrations of 1,4-Dioxane in Surface Water from the Water Quality
Portal, 1997-2022 54
Figure 2-9. Frequency of 1,4-Dioxane Concentrations Monitored in Raw (Untreated) Drinking Water
Derived from Surface Water 56
Figure 2-10. Frequency of 1,4-Dioxane Concentrations Monitored in Finished (Treated) Drinking
Water Derived from Surface Water 56
Figure 2-11. Map of Counties Containing PWSs that Reported Monitoring of Finished Drinking
Water Drawn from Surface Water for 1,4-Dioxane under UCMR3 57
Figure 2-12. Schematic of the EWISRD-XL Model Inputs and Outputs 59
Figure 2-13. Distributions of Surface Water Concentrations Estimated by Aggregate Probabilistic
Model for Each OES 70
Figure 2-14. Case Study Comparison of Modeled and Monitored Concentrations in Brunswick
County 72
Figure 2-15. Frequency of Nationwide Detected 1,4-Dioxane Groundwater Concentrations (n =
2,284) Retrieved from the Water Quality Portal, 1997-2022 74
Figure 2-16. Detectable Concentrations of 1,4-Dioxane in Groundwater from the Water Quality
Portal, 1997-2022 74
Figure 2-17. Groundwater Concentrations of 1,4-Dioxane vs. Sample Collection Date for Data
Collected between 1997 and 2022 75
Figure 2-18. Brief Description of Methodologies and Analyses Used to Estimate Ambient Air
Concentrations and Exposures 83
Figure 3-1. Potential Human Exposure Pathways to 1,4-Dioxane for the General Populationa 102
Figure 5-1. Distribution of Adult Lifetime Cancer Risk across all Facilities, Assuming No Additional
Dilution Occurs between the Point of Release and the Location of Drinking Water
Intakes 142
Figure 5-2. Distribution of Adult Lifetime Cancer Risk across Facilities with High Quality Release
Data, Assuming No Additional Dilution Occurs between the Point of Release and the
Location of Drinking Water Intakes 143
Figure 5-3. Distribution of Dilution of 1,4-Dioxane Concentrations at Downstream Drinking Water
Intakes 144
Figure 5-4. Distribution of Adult Lifetime Cancer Risk across all Facilities, Assuming Dilution to 1%
of Initial Concentrations in the Receiving Water Body 144
Figure 5-5. Histograms of Lifetime Cancer Risk Estimates for Aggregate Water Concentrations
Estimated Downstream of COUs with Vertical Lines Showing the Median and 95th
Percentile (P95) Values 150
LIST OF APPENDIX TABLES
TableApx C-l. Terminology Clarifications between the 2021 Draft Systematic Review Protocol and
the 1,4-Dioxane Systematic Review Protocol 185
Table_Apx C-2. Evaluation Criteria for Sources of Monitoring Data 197
Table Apx C-3. Evaluation Criteria for Sources of Experimental Data 204
Table Apx C-4. Evaluation Criteria for Sources of Database Data 210
Table Apx C-5. Considerations that Inform Evaluations of the Strength of the Evidence 214
Table Apx C-6. Evaluation of the Weight of the Scientific Evidence for Exposure Assessments 216
TableApx D-l. Categories and Subcategories of Conditions of Use Included in the Scope of the
Risk Evaluation 223
Table Apx E-l. Summary of EPA's Estimates for the Number of Facilities for Each OES 226
Page 12 of 484
-------
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TableApx E-2. Summary of EPA's Estimates for Air and Water Release Days Expected for Each
OES 230
TableApx E-3. Summary of Daily Industrial and Commercial Water Release Estimates for 1,4-
Dioxane 238
Table Apx E-4. Summary of Weight of the Scientific Evidence Conclusions in Water Release
Estimates by OES 243
Table Apx E-5. Summary of Daily Industrial and Commercial Land Release Estimates for 1,4-
Dioxane 260
Table Apx E-6. Summary of Weight of the Scientific Evidence Conclusions in Land Release
Estimates by OES 264
Table Apx E-7 Summary of Daily Industrial and Commercial Air Release Estimates for 1,4-
Dioxane 280
Table Apx E-8 Summary of Weight of the Scientific Evidence Conclusions in Air Release Estimates
by OES 285
Table Apx E-9. Comparison of TRI/DMR Release Data to LCA Study for PET Byproduct 292
Table Apx E-10. Summary of Overall Weight of the Scientific Evidence Conclusions for
Environmental Release Estimates by OES 297
Table_Apx E-l 1. TRI-CDR Use Code Crosswalk 300
Table Apx E-12. Summary of Parameter Values and Distributions Used in the Textile Release
Model 327
Table_Apx E-13. Triangular Distributions Ffixation 331
Table Apx E-14. Summary of Parameter Values and Distributions Used in the Industrial and
Institutional Laundry Release Model 340
TableApx E-15. Summary of Parameter Values and Distributions Used in the Hydraulic Fracturing
Release Model 352
TableApx F-l. Summary of Total Number of Workers and ONUs Potentially Exposed to 1,4-
Dioxane for Each Supplemental OESa 358
Table Apx F-2. Glove Protection Factors for Different Dermal Protection Strategies from ECETOC
TRA v3 360
Table Apx F-3. Textile Dye Worker Exposure Data Evaluation 365
Table Apx F-4. Inhalation Exposures of Workers for the Use of Textile Dye Based on Monitoring
Data 367
Table Apx F-5. Occupational Inhalation Monitoring Data for Textile Dyes 368
Table_Apx F-6. Antifreeze Data Source Evaluation 370
Table Apx F-7. Modeled Occupational Inhalation Exposures for Antifreeze 371
Table Apx F-8. Inhalation Exposures of Workers for the Use of Antifreeze Based on Modeling 371
Table Apx F-9. Surface Cleaner Worker Exposure Data Evaluation 373
Table Apx F-10. Inhalation Exposures of Workers for the Use of Surface Cleaner Based on
Monitoring Data 374
Table Apx F-l 1. Occupational Inhalation Monitoring Data for Surface Cleaner 375
Table Apx F-12. Dish Soap Worker Exposure Data Evaluation 377
TableApx F-13. Inhalation Exposures of Workers for the Use of Dish Soap Based on Monitoring
Data 378
Table Apx F-14. Occupational Inhalation Monitoring Data for Dish Soap 379
Table Apx F-15. Laundry Detergent Worker Exposure Data Evaluation 384
Table Apx F-16. Modeled Occupational Inhalation Exposures for Industrial Laundries 385
Table Apx F-17. Modeled Occupational Inhalation Exposures for Institutional Laundries 385
Page 13 of 484
-------
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TableApx F-18. Inhalation Exposures of Workers for the Use of Laundry Detergent in Industrial
Laundries Based on Modeling 386
Table Apx F-19. Acute and Chronic Inhalation Exposures of Workers for the Use of Laundry
Detergent in Institutional Laundries Based on Modeling 386
Table Apx F-20. Paint and Floor Lacquer Worker Exposure Data Evaluation 388
Table Apx F-21. Inhalation Exposures of Workers for the Use of Paint and Floor Lacquer Based on
Monitoring Data 389
Table Apx F-22. Occupational Inhalation Monitoring Data for Paint and Floor Lacquer 390
Table Apx F-23. Polyethylene Terephthalate (PET) Byproduct Worker Exposure Data Evaluation ... 394
TableApx F-24. Inhalation Exposures of Workers for PET Byproduct Based on Monitoring Data.... 395
Table Apx F-25. Occupational Inhalation Monitoring Data for Polyethylene Terephthalate (PET)
Byproduct 396
Table Apx F-26. Ethoxylation Process Byproduct Worker Exposure Data Evaluation 398
Table Apx F-27. Inhalation Exposures of Workers for the Ethoxylation Process Byproduct Based on
Monitoring Data 399
Table Apx F-28. Occupational Inhalation Monitoring Data for Ethoxylation Process Byproduct 400
Table Apx F-29. Hydraulic Fracturing Worker Exposure Data Evaluation 403
Table Apx F-30. Modeled Occupational Inhalation Exposures for Hydraulic Fracturing 404
Table Apx F-31. Inhalation Exposures of Workers for Hydraulic Fracturing Based on Modeling 404
Table Apx F-32. Estimated Inhalation Exposure (mg/m3) for Workers during Various Conditions of
Use 406
TableApx F-33. Summary of Weight of the Scientific Evidence Conclusions in Inhalation Exposure
Estimates by OES 408
Table Apx F-34. Summary of Parameter Values and Distributions Used in the Antifreeze Exposure
Modeling 415
TableApx F-35. Summary of Parameter Values and Distributions Used in the Laundry Detergent
Exposure Modeling 423
TableApx F-36. Summary of Parameter Values and Distributions Used in the Hydraulic Fracturing
Exposure Modeling 431
Table Apx G-l. Summary of Community Water Systems with Treatment Processes Capable of
Removing 1,4-Dioxane 435
Table Apx G-2. Summary of per Capita DTD Loading Estimates from SHEDS-HT Modeling 439
Table Apx G-3. Summary of Case Study Locations Including Modeled and Observed Surface Water
Concentrations 440
Table Apx G-4. Distribution of per Capita DTD Loading, in G/Day, by Product, for Non-commercial
Uses Modeled by SHEDS-HT 447
Table Apx G-5. Proportions of Population Expected to Contribute to DTD Loading through
Commercial Activities and Product Uses 447
Table Apx G-6. Summary of Proximity of Downstream Drinking Water Intakes to Releasing
Facilities Resulting in Modeled Risk above 1E-06 452
Table Apx G-l. Ranges of Dilution and Diluted 1,4-Dioxane Concentrations Modeled at Drinking
Water Intakes Downstream of Industrial Releases 453
Table Apx G-8. Ranges of LADD and Adult Lifetime Cancer Risk Estimates for Diluted 1,4-
Dioxane Concentrations Modeled at Drinking Water Intakes Downstream of Industrial
Releases 454
Table Apx H-l. Release Year, TRI Facility ID, Facility Name, State, Registry Number, Disposal
Type, and Disposal Weight for On-Site Class I Underground Injection Wells
According to TRI 455
Page 14 of 484
-------
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TableApx H-2. Release Year, Source TRI Facility ID, Source State, Receiving Facility RCRA ID,
State, Disposal Type, and Disposal Weight for Off-Site Class I Underground Injection
Wells According to TRI and RCRAInfo Databases 456
Table Apx H-3. Release year, TRI Facility ID, Facility Name, State, CERCLIS ID, Disposal Type,
and Disposal Weight for RCRA Subtitle C Landfills According to TRI 456
Table Apx H-4. Release Year, Source TRI Facility ID, Source State, Receiving Facility RCRA ID,
State, Disposal Type, and Disposal Weight for Off-Site Class I Underground Injection
Wells According to TRI and RCRAInfo Databases 456
Table_Apx H-5. Input Variables for Chemical of Concern 458
Table_Apx H-6. Waste Management Unit (WMU) Properties 459
Table Apx H-7. Potential Groundwater Concentrations (mg/L) Based on Disposal of 1,4-Dioxane to
Unlined and Clay-Lined Landfills as Assessed by Applying the EPACMTP Model.... 461
Table Apx H-8. Input Variables for Chemical of Concern 463
Table_Apx H-9. Waste Management Unit 465
Table Apx J-l. Release Estimates from 2019 TRI Used for Ambient Air: Screening Methodology for
1,4-Dioxane 469
Table Apx J-2. Exposure and Risk Estimates from the Ambient Air: Screening Methodology for 1,4-
Dioxane Releases Reported to TRI 471
TableApx J-3. Exposure Scenarios and Inputs Utilized for Pre-screening Analysis of Hydraulic
Fracturing, Industrial Laundry, and Institutional Laundry COU 472
Table Apx J-4. Description of Daily or Period Average and Air Concentration Statistics 476
Table Apx J-5. Summary of Fenceline Community Exposures Expected Near Facilities Where
Modeled Air Concentrations Indicated Risk for 1,4-Dioxane 478
Table_Apx J-6. Summary of Groups of Facilities Considered in Aggregate Analysis 482
LIST OF APPENDIX FIGURES
FigureApx C-l. Overview of the TSCA Risk Evaluation Process with Identified Systematic Review
Steps 183
Figure Apx C-2. Literature Inventory Tree - Environmental Releases and Occupational Exposure
Search Results for 1,4-Dioxane 193
Figure Apx C-3. Literature Inventory Tree - General Population, Consumer, and Environmental
Exposure Search Results for 1,4-Dioxane 194
Figure_Apx D-l. COU and OES Mapping 222
FigureApx E-l. Flowchart of a Monte Carlo Method Implemented in a Microsoft Excel-Based
Model Using a Monte Carlo Add-In Tool 319
Figure Apx E-2. Environmental Release and Occupational Exposure Points during Textile Dying .... 324
Figure Apx E-3. Environmental Release and Occupational Exposure Points during
Industrial/Institutional Laundering Operations 332
Figure Apx E-4. Environmental Release and Occupational Exposure Points during Hydraulic
Fracturing 346
Figure Apx F-l. Environmental Release and Occupational Exposure Points during Textile Dying .... 364
Figure Apx F-2. Environmental Release and Occupational Exposure Points during
Industrial/Institutional Laundering Operation 382
Figure Apx F-3. Environmental Release and Occupational Exposure Points during Hydraulic
Fracturing 402
Figure Apx F-4. Environmental Release and Occupational Exposure Points during
Industrial/Institutional Laundering Operations 418
Page 15 of 484
-------
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
FigureApx F-5. Environmental Release and Occupational Exposure Points during Hydraulic
Fracturing 425
Figure Apx G-l. Example Raw and Finished Water Concentrations from a PWS without Processes
to Remove 1,4-Dioxane 434
Figure Apx G-2. Schematic of the General Fit-for-Purpose EWISRD-XL Model 438
Figure_Apx G-3. Map of Brunswick County, NC Model Case Study 441
Figure Apx G-4. Plot Comparing Results from Brunswick County Case Study Modeling with
Observed Concentrations 442
Figure_Apx G-5. Map of the Columbia, TN, Case Study 443
Figure_Apx G-6. Map of the East Liverpool, OH, Case Study 444
Figure Apx G-l. Schematic of the Flow of Data within the EWISRD-XL-R Probabilistic Model 446
Figure Apx G-8. Distribution of Mean Annual Modeled Flow Rates for NHDPlus V2.1 Reaches
Identified within 5 km of Hydraulic Fracturing Wells Reporting 1,4-Dioxane 449
Figure Apx G-9. Distribution of Modeled Ranges of 1,4-Dioxane Concentrations in Streams near
Hydraulic Fracturing Wells Reporting 1,4-Dioxane 450
Figure Apx G-10. Generic Schematic of Hypothetical Release Point with Surface Water Intakes for
Drinking Water Systems Located Downstream 451
FigureApx G-l 1. Summary Distribution of Mean Annual Flow at Stream Reaches Matched with
Drinking Water Intakes 453
Figure Apx J-l. Summary of Methodologies Used to Estimate Ambient Air Concentrations and
Exposures 468
Figure Apx J-2. Exposure Scenarios Modeled for Max and Mean Release Using IIOAC Model for
Ambient Air: Screening Methodology 470
Figure Apx J-3. Modeled Receptor Locations for Finite Distance Rings 474
Figure_Apx J-4. Modeled Receptor Locations for Area Distance 475
Figure Apx J-5. Example of Group of Air Releasing Facilities with Overlapping 10 km Buffers for
Aggregate Air Risk Screening 479
Figure Apx J-6. Decision Tree for Characterizing Aggregate Air Risk for Multiple Facilities 481
Figure_Apx J-7. Map of Aggregated Air Facilities, Group 1 482
Figure_Apx J-8. Map of Aggregated Air Facilities, Group 2 483
Figure_Apx J-9. Map of Aggregated Air Facilities, Group 3 483
Figure_Apx J-10. Map of Aggregated Air Facilities, Group 4 484
Figure_Apx J-l 1. Map of Aggregated Air Facilities, Group 5 484
Page 16 of 484
-------
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
ACKNOWLEDGEMENTS
This report was developed by the United States Environmental Protection Agency (U.S. EPA or the
Agency), Office of Chemical Safety and Pollution Prevention (OCSPP), Office of Pollution Prevention
and Toxics (OPPT).
Acknowledgements
The Assessment Team gratefully acknowledges the participation, input, and review comments from
OPPT and OCSPP senior managers and advisors. Acknowledgement is also given for the contributions
of interagency reviewers that included multiple federal agencies and assistance provided from EPA
contractors ERG (Contract No. 68HERD20A0002), ICF (Contract No. EP-W-12-010), and Versar
(Contract No. EP-W-17-006). Special acknowledgement is given for the contributions of technical
experts from EPA's Office of Research and Development, including Daniel Dawson for initial
development of the 1,4-dioxane water model and Caroline Ring for input on probabilistic modeling
approaches for the water pathway.
Docket
Supporting information can be found in the public docket (Docket ID: EPA-HQ-QPPT-2022-0905Y
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/Contributors
Jeffrey Morris (Division Director), Yvette Selby-Mohamadu (Deputy Division Director), Rochelle
Bohaty (Branch Chief/Management Lead), Susanna Wegner (Assessment Lead), Sarah Au, Rebecca
Feldman, Mark Gibson, Bryan Groza, Franklyn Hall, Lauren Knapp Shannon Rebersak, Shawn Shifflett,
Adam Theising, Jason Todd, Kevin Vuilleumier, Cindy Wheeler, Daniel Whitby
Executive Team
This draft risk evaluation was reviewed by OPPT and OCSPP leadership, including senior advisors Stan
Barone, Jeff Dawson, Anna Lowit, and Ryan Schmit and senior leaders, including Mark Hartman
(Deputy Office Director, OPPT) and Denise Keehner (Office Director, OPPT).
Internal Reviewers
This assessment was provided for review to scientists in EPA's Program and Region Offices. Comments
were submitted by:
• Office of the Administrator/Office of Children's Health Protection
• Office of Air and Radiation
• Office of Chemical Safety and Pollution Prevention/Office of Pesticide Programs
• Office of General Council
• Office of Land and Emergency Management
• Office of Research and Development
• Office of Water
Page 17 of 484
-------
767
768
769
770
771
772
773
774
775
776
777
778
779
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
External Reviewers
This assessment was provided for review to other federal agencies and Executive Offices of the
President, including:
• Consumer Product Safety Commission
• Department of Defense
• Department of Health and Human Services/National Institute of Environmental Health Sciences
• Department of Health and Human Services/National Institute for Occupational Safety and Health
• Department of Justice
• Department of Labor/Occupational Safety and Health Administration
• Executive Office of the President/Office of Management and Budget
• Food and Drug Administration
• National Aeronautics and Space Administration
• Small Business Administration Office of Advocacy
Page 18 of 484
-------
780
781
782
783
784
785
786
787
788
789
790
791
792
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Summary of Risk Findings and Support for Risk Determination
1,4-Dioxane is a solvent used in a variety of commercial and industrial applications in the United
States. It is also produced as a byproduct in several manufacturing processes and may remain present
as a byproduct in consumer and commercial products, including soaps, detergents, and cleaning
products. Health effects of concern for 1,4-dioxane include cancer and effects in liver and olfactory
tissue. People may be exposed to 1,4-dioxane through occupational exposure, consumer products, or
contact with water, land, or air where 1,4-dioxane has been released to the environment from
industrial and commercial sources or from consumer and commercial products washed down the
drain or disposed of in landfills.
The risk evaluation for 1,4-dioxane published in 2020 evaluated risks from a range of occupational
and consumer uses of 1,4-dioxane, risks to aquatic species, and risks to the general population
resulting from incidental recreational contact with water. It did not evaluate general population
exposures to 1,4-dioxane in drinking water or air and did not evaluate the full range of exposure that
may result from 1,4-dioxane produced as a byproduct.
This draft supplement completes the Toxics Substances Control Act (TSCA) risk evaluation for 1,4-
dioxane by (1) more comprehensively evaluating risks from 1,4-dioxane present as a byproduct; and
(2) evaluating risks from general population exposures to 1,4-dioxane released to water, air, and
land. This analysis identified cancer risk estimates higher than 1 in 10,000 (1 x 10~4) for a range of
typical and high-end occupational exposures to 1,4-dioxane produced as a byproduct. It also
identified cancer risk estimates higher than 1 in 1 million (1 x 10~6) for a range of general population
exposure scenarios associated with 1,4-dioxane in drinking water sourced downstream of release
sites and in air within 1 km of releasing facilities. Although these risk estimates include inherent
uncertainties and the overall confidence in specific risk estimates varies, the analysis provides
support for the Agency to make a determination about whether 1,4-dioxane poses an unreasonable
risk and to identify drivers of unreasonable risk among exposures for people (1) with occupational
exposure to 1,4-dioxane under some conditions of use, (2) who rely on sources of drinking water
downstream of release sites, and (3) breathing air near release sites.
Subsequent to this draft supplement, EPA is releasing a draft revised risk determination for 1,4-
dioxane. The updated risk determination considers the results presented in this draft supplement as
well as those published in the 2020 Risk Evaluation for 1,4-Dioxane.
EXECUTIVE SUMMARY
This draft document is a supplement to the Final Risk Evaluation for 1,4-Dioxane. published December
2020 (2020 RE). EPA conducted this supplemental analysis because contrary to the law's requirement
for TSCA risk evaluations to be carried out on the "chemical substance" under the conditions of use
(COUs), the 2020 RE excluded certain known human exposure pathways that are important to
understanding the health implications of exposure to 1,4-dioxane. This supplement effectively completes
EPA's risk evaluation on the chemical substance and positions the Agency to comprehensively address
identified unreasonable risks.
1,4-Dioxane is primarily used as a solvent in commercial and industrial applications. It can also be
produced as a byproduct of several common manufacturing processes, including but not limited to
ethoxylation processes used in the production of surfactants used in soaps and detergents and production
Page 19 of 484
-------
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
of polyethylene terephthalate (PET) plastics. Even though it is not intentionally added, 1,4-dioxane
produced as a byproduct may remain present in consumer and commercial products, including soaps and
detergents, cleaning products, antifreeze, textile dyes, and paints/lacquers. 1,4-Dioxane is released to the
environment from industrial and commercial releases and from consumer and commercial products that
are washed down the drain or disposed of in landfills. People may be exposed to 1,4-dioxane through
occupational exposure, consumer products, or contact with water, land, or air where 1,4-dioxane has
been released to the environment.
The 2020 RE did not evaluate risks from two critical areas: (1) general population exposures to 1,4-
dioxane in drinking water or air, and (2) the full range of exposure that may result from 1,4-dioxane
produced as a byproduct. During review of the draft risk evaluation, peer reviewers and public
commenters raised concerns that failure to consider these exposure pathways could leave portions of the
population at risk. These concerns include the fact that 1,4-dioxane has been detected in drinking water
and is not readily removed through traditional treatment. In addition, 1,4-dioxane produced as a
byproduct results in occupational exposures that were not evaluated in the 2020 RE. Finally, 1,4-dioxane
produced as a byproduct also contributes to 1,4-dioxane in drinking water through industrial releases
and down-the-drain (DTD) disposal of consumer and commercial products.
This draft supplement expands on the analysis of COUs in which 1,4-dioxane is present as a byproduct
to include additional COUs for which information is reasonably available and consider associated
occupational exposures. This draft supplement also evaluates risks to the general population—including
potentially exposed or susceptible subpopulations (PESS)—from exposure to 1,4-dioxane through
drinking water or air resulting from all industrial releases (including those resulting from 1,4-dioxane
produced as a byproduct) as well as DTD releases of consumer and commercial products.
EPA's evaluation of these additional human exposure pathways included new methods and novel
applications of existing methods that will be subject to peer review at a Science Advisory Committee on
Chemicals (SACC) meeting in September 2023. Following review by the public and the SACC, the
Agency will finalize this supplemental risk evaluation. At that time, EPA will initiate steps to address
unreasonable risks identified through its complete evaluation of 1,4-dioxane.
Approach
For this draft supplemental risk evaluation, EPA is relying on the physical and chemical properties,
chemical lifecycle information, environmental fate and transport information, and the hazard
identification and dose-response analysis presented in the 2020 RE. EPA evaluated cancer and non-
cancer risks from occupational and general population exposure scenarios using available modeling
and/or monitoring information. The Agency also considered site-specific exposures, including combined
or additive releases from multiple releasing facilities within a single air or water exposure pathway. This
draft supplement considers PESS throughout the human health exposure assessment and risk
characterization.
Exposure
Occupational Exposure: EPA estimated both high-end and central tendency occupational exposures
through inhalation and dermal absorption. High-end exposure estimates were used because they attempt
to capture potential variability in exposure across facilities and individuals and may be representative of
PESS and "sentinel" exposures. In some cases, high-end estimates reflect uncertainty around the extent
of this variability. EPA estimated occupational exposure for most COUs based on available monitoring
data. For COUs without occupational monitoring data, EPA applied Monte Carlo methods to estimate
Page 20 of 484
-------
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
exposures using generic scenarios and emission scenario documents. These methods are generally
consistent with Monte Carlo approaches used in previous TSCA risk evaluations.
General Population Exposure: EPA evaluated general population exposures to 1,4-dioxane through
drinking water and air that could result from releases to surface water, groundwater, land, and air. The
Agency evaluated a range of exposure scenarios for each pathway, including (but not limited to) high-
end exposure scenarios. To be protective of PESS and sentinel exposures, EPA developed risk estimates
on the scenarios, populations, and life stages with the highest levels of exposure. For drinking water,
EPA evaluated life stage-specific exposures for adults, formula-fed infants, and children. For air
exposures, because the impacts of lifestage differences could not be quantified adequately, air
concentrations are used for all lifestages. To address exposure to fenceline communities, EPA
considered air exposures within 10 km of a release site. For water releases, EPA considered exposures to
communities relying on drinking water sourced near release sites.
Hazard
All hazard values used in this draft supplement were derived from the points of departure (PODs)
previously peer-reviewed by the SACC and published in the 2020 RE. Some of the exposure scenarios
evaluated in this analysis required duration adjustments to the previously established hazard values;
however, the underlying hazard endpoints and PODs remain the same. Health effects of concern for 1,4-
dioxane include cancer and adverse, non-cancer effects in liver and olfactory tissue.
Risk Characterization
EPA evaluated both cancer and non-cancer risks for each exposure pathway. Because cancer is the
primary risk driver for 1,4-dioxane, results presented here are cancer risk estimates. Overall confidence
in the risk estimates varies across exposure pathways and COUs, depending on the data, models, and
assumptions used. For risk estimates, EPA has medium to high confidence in the underlying hazard
PODs used as the basis for risk characterization. Therefore, exposure-related considerations drive
differences in confidence among risk estimates. Differences in central tendency and high-end risk
estimates may reflect variability in exposure across the population (e.g., due to differences in the
frequency or intensity of occupational exposures) and/or uncertainty in the exposure assessment (e.g.,
due to incomplete information on release amounts or variations in flow rates of receiving water bodies
and/or drinking water intake locations).
Estimates of Occupational Risks: EPA estimated cancer and non-cancer risks for a set of new
occupational COUs where 1,4-dioxane is present as a byproduct.
• Dermal Exposure: Dermal occupational exposure is expected to occur as a result of worker
activities such as transfer operations, application of 1,4-dioxane containing formulations, and the
cleaning of equipment. COU-specific weight fractions and evaporation drive the variability in
results. Cancer risk estimates for dermal exposures range from 8.1 x 10~7 to 8,6/ 10 4 for central
tendency exposure and from 5,Ox 10 6 to 1.5/10 2 for high-end exposures across COUs. Overall
confidence in risk estimates for occupational dermal exposures is medium for all occupational
exposure scenarios.
• Inhalation Exposure: Inhalation exposure to 1,4-dioxane is expected to occur as a result of the
scenario-specific considerations described in the bullets below. Cancer risk estimates for
inhalation exposure range from 8.3*10~12 to 1.8x10 3 for central tendency exposures and from
5,4/10 " to 2,3/10 2 for high-end exposures across COUs. Occupational exposure scenarios
with the highest estimates of risk from inhalation exposure are summarized below:
Page 21 of 484
-------
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
o Polyethylene Terephthalate (PET) Manufacturing. Workers may inhale 1,4-dioxane
generated as a byproduct of PET plastic manufacturing. Cancer risk estimates for
inhalation exposure range from 1.8x10 3 for central tendency exposures to 2.3/10 2 for
high-end exposures. There is uncertainty regarding the risk estimates because the analysis
relied on decades-old monitoring data and the extent to which the monitoring data reflect
current practices is unknown. Overall confidence in risk estimates for PET plastic
manufacturing is medium.
o Hydraulic Fracturing Operations. 1,4-Dioxane inhalation exposures may occur during
hydraulic fracturing operations due to its documented presence in scale inhibitors,
additives, friction reducers, and surfactants used in fracturing fluid formulations. Cancer
risk estimates for inhalation exposure range from 7,Ox 10 5 for central tendency exposures
to 9.5x10 3 for high-end exposures. There is uncertainty regarding the model inputs used
to estimate exposures and the extent to which they reflect the actual distribution of
hydraulic fracturing occupational exposures and workplace practices. Overall confidence
in risk estimates for hydraulic fracturing operations is medium to high.
o Ethoxylation Processes. 1,4-Dioxane may be generated as a byproduct in ethoxylation
reactions during the manufacture of common surfactants that result in worker inhalation
exposure. Cancer risk estimates for inhalation exposure range from 4.6x 10~4 for central
tendency exposures to 5.9xl0~4 for high-end exposures. There are numerous uncertainties
due to the limited monitoring data, unknown concentrations, and the mass of 1,4-dioxane
generated as a byproduct during ethoxylation. There is also uncertainty in the worker
activities covered by the monitoring data and whether all foreseeable activities,
corresponding exposures, and workplace operations are represented. Overall confidence
in risk estimates for ethoxylation processes is low to medium.
o Industrial/Commercial Use of Dish Soap and Dishwasher Detergent. 1,4-Dioxane
inhalation exposures are expected during the use of dish soap and dishwashing detergent
from unloading and transferring detergent formulation, transport container cleaning, and
washing operations due to the presence of 1,4-dioxane as a surfactant byproduct. Cancer
risk estimates for inhalation exposure range from 4,Ox 10 4 for central tendency exposures
to 1.0x10 3 for high-end exposures. There is uncertainty as to the representativeness of
these estimates due to the age of the monitoring data, number of non-detects, and the
limited sample size. Overall confidence in risk estimates for dish soap and dishwasher
detergent use is low to medium.
Estimates of Risk to the General Population Exposure through Drinking Water Sourcedfrom Surface
Water: EPA estimated cancer and non-cancer risks for a range of general population exposures to
surface water used as drinking water. 1,4-Dioxane is not readily removed through typical wastewater or
drinking water treatment processes. Sources of 1,4-dioxane in surface water include direct and indirect
industrial releases from COUs where 1,4-dioxane is manufactured, processed, or used; industrial COUs
where 1,4-dioxane is present due to production as a byproduct (including PET manufacturing,
ethoxylation processes and hydraulic fracturing operations); and DTD releases of 1,4-dioxane present in
consumer and commercial products.
Monitoring data demonstrates that 1,4-dioxane is present in some source water and finished drinking
water samples. Measured concentrations in finished drinking water samples resulted in cancer risk
estimates greater than 1 in 1 million at the high end of the distribution of monitoring samples. However,
available surface water monitoring datasets are not designed to reflect source water impacts of direct and
indirect releases into water bodies. Therefore, EPA estimated concentrations modeled for a range of
Page 22 of 484
-------
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
specific release scenarios. The Agency evaluated the performance of the models against monitoring data
from site-specific locations serving as cases studies. This evaluation demonstrated strong concordance
between modeled concentrations and monitoring data, thereby increasing confidence in risk estimates
based on modeled concentrations.
EPA used modeled water concentrations to evaluate risks from a range of sources individually and in
aggregate {i.e., by evaluating risks from water concentrations resulting from multiple sources of 1,4-
dioxane releasing to the same water bodies). For each of the sources assessed, risk estimates from
modeled concentrations in receiving water bodies at the point of release may be greater than 1 in
1,000,000 or 1 in 100,000 under some conditions.
• Industrial Releases to Surface Water: Risk from individual facilities vary substantially within and
across COUs, with cancer risk estimates ranging from 5,4/10 13 to 0.025. Overall confidence in
risk estimates for specific facilities depends on confidence in facility-specific release data, but
confidence in the overall analysis is medium-high
• Down-the-drain Releases to Surface Water: EPA evaluated the conditions under which down-
the-drain releases contribute to different levels of risk and identified plausible scenarios in which
risks from down-the-drain releases result in risks greater than 1 in 1 million. Risk estimates from
modeled down-the-drain releases are highest in locations where large populations contribute to
these releases and where they are discharged to streams with low flow. Overall confidence in this
analysis is medium.
• Hydraulic Fracturing Releases to Surface Water: Cancer risk estimates from modeled hydraulic
fracturing waste releases to surface water are 3,57/10 8 for median modeled releases and
1,45/10 6 for 95th percentile modeled releases. Overall confidence in this analysis is medium.
• Aggregate Releases to Surface Water: Probabilistic modeling provides a distribution of risk
estimates reflecting a range of drinking water scenarios that account for aggregate sources of 1,4-
dioxane in water. Overall confidence in risk estimates for specific facilities depends on
confidence in facility-specific release data used as model inputs.
The relative contribution from different sources varies under different conditions and is likely to be
driven by site-specific factors including the amounts released from each source, flow rates of receiving
water bodies, and proximity of releases to drinking water intakes. In locations where industrial releases
are particularly high, industrial releases are likely to drive risk. In locations where large populations
contribute to down-the-drain releases and receiving water bodies have relatively low flow rates, down-
the-drain releases on their own may drive risks. Each of the sources evaluated may contribute to
drinking water risks under some conditions. Furthermore, this analysis illustrates the fact that sources of
1,4-dioxane produced as a byproduct, including those from industrial releases and DTD releases, can
contribute to risks from 1,4-dioxane in water. The analyses in this draft supplement describe the
conditions under which different levels of risk may occur.
Estimates of Risk to the General Population Exposure through Drinking Water Sourcedfrom
Groundwater: EPA estimated cancer and non-cancer risks for a range of general population exposures to
groundwater used as drinking water. Sources of 1,4-dioxane in groundwater include leachate from
landfills and disposal of hydraulic fracturing waste. Down the drain releases to septic fields from
consumer and commercial products containing 1,4-dioxane as well as historical disposals of 1,4-dioxane
are other potential sources of groundwater contamination but were not considered in this assessment.
Overall confidence in these risk estimates is low to medium.
Page 23 of 484
-------
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• 1,4-Dioxane in Groundwater from Hydraulic Fracturing: Cancer risk estimates for modeled
groundwater concentrations are 4.Ox 10~7 for median modeled releases and 8,6/ 10 6 for 95th
percentile modeled releases.
• 1,4-Dioxane in Groundwater Resulting from Landfill Leachate: Cancer risk estimates increase
under scenarios with higher leachate concentrations and loading rates. Monitoring data for
groundwater contamination surrounding landfills was not readily available for comparison.
Estimates of Risk to the General Population through Air: EPA estimated cancer and non-cancer risks for
a range of general population exposures to 1,4-dioxane in air. 1,4-Dioxane concentrations in air depend
on the facility-specific release amount, stack height(s), topography, and meteorological conditions—not
on specific COUs.
Potential sources of 1,4-dioxane in air include industrial releases reported to Toxics Release Inventory
(TRI), fugitive emissions from hydraulic fracturing, and emissions from institutional and industrial
laundries. The highest estimated risks occurred within 1,000 m of industrial release sites. EPA also
estimated risk from the aggregate exposures from multiple facilities releasing 1,4-dioxane in proximity
to fenceline communities. This aggregate analysis did not identify locations with aggregate cancer risk
greater than 1 x 10~6 that did not already have cancer risk above that level from an individual facility and
therefore did not have a substantial impact on the overall findings.
1,4-Dioxane concentrations in air depend on the facility-specific release amount, stack height(s),
topography, and meteorological conditions.
• Air Releases Reported to TRI: Cancer risk estimates for 95th percentile modeled air
concentrations with 1,000 m (approximately 0.6 mile) of the highest risk facilities in each COU
range from 1,0x 10~10 to 1.1x 10~4. Cancer risk estimates based on 50th percentile modeled
exposure concentrations within 1,000 m of the highest risk facilities range from 2.5/10 " to
8.3xl0~5. Although individual risk estimates for specific locations should be interpreted with
caution, most estimates are informed by moderate to robust modeling approaches and input data.
Overall confidence in risk estimates for inhalation exposures resulting for air concentrations
modeled for industrial releases ranges from low to high, depending on the level of confidence in
release information underlying risk estimates for specific facilities and COUs.
• Fugitive Air Emissions from Hydraulic Fracturing Operations: Cancer risk estimates within
1,000 m of hydraulic fracturing operations range from 2.2x 10~8 to 7.1 / 10 5 for a range of model
air model scenarios across a range of high-end (95th percentile) and central tendency release
scenarios. Overall confidence in risk estimates for inhalation exposures resulting for air
concentrations modeled based on releases from hydraulic fracturing operations is medium.
• Emissions from Industrial and Institutional Laundries: Cancer risk estimates within 1,000 m of
industrial and institutional laundries range from 1.5 xlO-11 to 3.8 xlO-8 across a range of high-end
exposure scenarios. Overall confidence in risk estimates from inhalation exposures resulting
from industrial and institutional laundries is medium.
Page 24 of 484
-------
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
1 INTRODUCTION
1,4-Dioxane is one of the first 10 chemicals undergoing the Toxics Substances Control Act (TSCA) risk
evaluation process after passage of the Frank R. Lautenberg Chemical Safety for the 21st Century Act,
which amended TSCA. 1,4-Dioxane is primarily used as a solvent in a variety of commercial and
industrial applications such as the manufacture of other chemicals (e.g., adhesives, sealants) or as a
processing aid or laboratory chemical. It is produced as a byproduct in several manufacturing processes,
including ethoxylation, sulfonation, sulfation, and esterification. Although there are no direct consumer
uses, 1,4-dioxane produced as a byproduct in the aforementioned processes can be present in
commercial and consumer products, including soaps, detergents, and cleaning products. Use of these
products may result in direct occupational and consumer exposures. Disposal of these products down-
the-drain (DTD) may contribute to general population exposure to 1,4-dioxane present in some surface
water.
1.1 Regulatory Context
In the 2019 draft 1,4-dioxane risk evaluation, EPA reviewed the exposures, hazards, and risks of 1,4-
dioxane from occupational exposures and surface water exposures to environmental organisms. It also
included the physical and chemical properties, lifecycle information, environmental fate and transport
information, and hazard identification and dose-response analysis. However, the 2019 draft risk
evaluation excluded general population exposures through drinking water and air and conditions of use
(COUs) in which 1,4-dioxane is present as a byproduct. These exclusions were based in part on an
interpretation that EPA had broad discretionary authority under TSCA to categorically exclude
conditions of use from the scope of its evaluations, and, as described in the 2018 1,4-Dioxane Problem
Formulation} that certain exposure pathways need not be considered if they were under the jurisdiction
of other EPA regulatory programs or analytical processes.
These analyses were reviewed by the Science Advisory Committee on Chemicals (SACC) in 2019. The
SACC raised a number of concerns regarding the evaluation and approach, but particularly noted its
concerns about the Problem Formulation straying from "basic principles of risk assessment," the
omission of well-known exposure routes, and that general lack of comprehensiveness undermining
EPA's ability to protect against risks to human health and the environment. As stated in the meeting
minutes and final report2 from the July 2019 SACC meeting, "there was general dissatisfaction in the
Committee that the human health risk characterization did not extend to the general population since
there was no indication in the Evaluation that other offices in the EPA had plans to conduct such a
characterization." Furthermore, "several committee members also observed that failure to assess 1,4-
dioxane exposure in the general population may leave substantial portions of the population at risk. This
is particularly concerning for drinking water." The SACC also raised concerns about potential risks from
1,4-dioxane produced as a byproduct, recommending that "EPA should provide a detailed discussion of
the scientific basis for the exclusion of impurity or byproduct formation of 1,4-dioxane."
Public stakeholders also raised concerns about water monitoring data demonstrating the presence of 1,4-
dioxane in drinking water. Commenters also identified additional sources of 1,4-dioxane that had not
been assessed, including COUs in which 1,4-dioxane is produced as a byproduct, and raised concerns
that some of these COUs may be an important source of exposure to 1,4-dioxane in water.
1 The 1,4-Dioxane Problem Formulation is available at https://www.epa.. gov/assessing-and-managing-chemicals-under-
tsca/14-dioxane-problem-fonnulation.
2 The SACC July 2019 meeting minutes and final report (Document ID EPA-HQ-OPPT-2019-0237-0064) are available in the
docket at https://www.regulations.gov/document/EPA-HO-OPPT-2Q19-uzj /~wo4.
Page 25 of 484
-------
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
In November of 2020, EPA released a supplement to the draft 1,4-dioxane risk evaluation for public
comment. The November 2020 supplement assessed eight additional COUs of 1,4-dioxane as a
byproduct in consumer products and general population exposure from incidental contact with surface
water. The EPA determined that the additional analysis did not warrant SACC review as no novel
science was utilized.
In December 2020, the Agency published the Final Risk Evaluation for 1,4-Dioxane ( )20c)
(referred hereafter as the "2020 RE") ( 20c). The 2019 draft and 2020 supplement were
both incorporated into the 2020 RE, which assessed risks for
• worker and occupational non-user (ONU) exposures to 1,4-dioxane through 16 industrial and
commercial COUs;
• consumer and bystander exposures to 1,4-dioxane present as a byproduct3 in eight consumer
product categories;
• general population exposure via incidental/recreational contact with 1,4-dioxane present in
surface water from industrial releases; and
• aquatic species' exposures to 1,4-dioxane present in surface water.
In January 2021, the White House issued Executive Order 13990 instructing that the federal government
be guided by the best science and be protected by processes that ensure the integrity of Federal decision-
making, and established the Administration's policy of, among other concerns, following the science,
improving public health and protecting the environment, limiting exposure to dangerous chemicals, and
prioritizing environmental justice when delivering on these concerns. Executive Order 13990 also
instructs agencies to (1) review actions issued between January 20, 2017, and January 20, 2021, which
may be inconsistent with or present obstacles to implementing the policy established in the order and;
(2) consider suspending, revising, or rescinding such actions.
Upon further review, EPA determined that the approach taken in the 2020 RE {i.e., the exclusion of
reasonably foreseeable exposures to workers, as well as exposures to the general population from air,
water, and disposal) was inconsistent with the plain language of TSCA section 6 and left potential
risks—including risks to potentially exposed and susceptible subpopulations (PESS)—unaccounted for.
The law's requirement that EPA conduct risk evaluation on a "chemical substance" under the COUs
requires the Agency to determine the chemical's COUs and to not otherwise exclude those COUs from
the scope of the risk evaluation.
In June of 2021, EPA announced that additional analysis was needed to consider critical exposure
pathways not assessed in the final risk evaluations for the first 10 chemicals (including, but not limited
to, ambient air, ambient water, and drinking water). For many of the first 10 risk evaluations, EPA
applied the Draft Screening Level Approach for Assessing Ambient Air and Water Exposures to
Fenceline Communities Version 1.0,4 published in January 2022 to determine whether further analysis
was needed. For 1,4-dioxane, however, EPA determined that a more in-depth analysis was needed to
address concerns about known drinking water contamination (described in Section 1.3.1.3) and to more
fully evaluate COUs in which 1,4-dioxane is present as a byproduct (described in Section 1.3.1.1), and
signaled its intention to re-open and formally supplement the 1,4-dioxane risk evaluation.
3 Byproduct means a chemical substance produced without a separate commercial intent during the manufacture, processing,
use, or disposal of another chemical substance(s) or mixture(s).
4 The draft Screening Level Approach for Assessing Ambient Air and Water Exposures to Fenceline Communities Version 1.0
is available at https://www.epa.gov/assessing~aiMl~managing~chemicals~under~tsca/tsca~screening~level~approach~assessing~
ambient-air-and.
Page 26 of 484
-------
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
This draft supplement to the 2020 RE is intended to complete the risk evaluation on the chemical
substance 1,4-dioxane as required under TSCA by (1) expanding the analysis of COUs in which 1,4-
dioxane is present as a byproduct to include additional COUs and consider associated occupational
exposures; and (2) evaluating risks from general population exposures to 1,4-dioxane released to surface
and groundwater, air, and land.
EPA's evaluation of these additional human exposure pathways included new methods and novel
applications of existing methods. This draft supplement is the first under amended TSCA to evaluate:
exposures and risks from a chemical produced as a byproduct, aggregate risks for communities relying
on drinking water sourced from surface water receiving a chemical from multiple sources, risks for
communities relying on drinking water sourced from groundwater, aggregate risks for communities
exposed through air near multiple release sites, and consideration of multiple years of environmental
release data.
1.2 Scope
This draft supplement is intended to evaluate risks from exposure pathways and COUs for 1,4-dioxane
that were not assessed in the 2020 RE. Additional exposure pathways and new COUs included in this
draft supplement were identified based on information submitted in previous public comments and other
reasonably available information. For the current analysis, EPA is relying on the physical and chemical
properties, as well as lifecycle information, environmental fate and transport information, and hazard
identification and dose-response analysis presented in the 2020 RE (Sections 1.1, 1.4, 2.1, and 3.2 of the
2020 RE, respectively). Furthermore, this supplement does not reevaluate the occupational, consumer,
or ecological exposure pathways and risks that were previously assessed in the 2020 RE.
This draft supplement to the 2020 RE evaluates risks for the following exposure pathways:
• Occupational exposure to
o 1,4-dioxane present as a byproduct in commercial products (corresponding to consumer
products considered in the 2020 RE); and
o 1,4-dioxane produced or present as a byproduct in additional industrial COUs for which
information on the presence of 1,4-dioxane is reasonably available, including
ethoxylation processing, polyethylene terephthalate (PET) manufacturing, and hydraulic
fracturing (Sections 3.1, 5.2.1).
• General population exposures to
o 1,4-dioxane present in drinking water sourced from surface water as a result of all direct
and indirect industrial releases and DTD releases of consumer and commercial products
(Sections 2.3.1, 3.2.2 and 5.2.2.1);
o 1,4-dioxane present in drinking water sourced from groundwater contaminated as a result
of disposals (Sections 2.3.2, 3.2.2.2 and 5.2.2.1.6); and,
o 1,4-dioxane released to air from industrial and commercial sources (Sections 2.3.3, 3.2.3,
and 5.2.2.3).
Many of the COUs assessed in this draft supplement contribute to more than one exposure pathway. For
example, 1,4-dioxane present as a byproduct of PET manufacturing may contribute to occupational
exposures during manufacturing as well as general population exposures through releases to water and
air. In addition, for many of the exposure pathways assessed, multiple COUs contribute to 1,4-dioxane
exposure. For example, many COUs can contribute to general population exposures to 1,4-dioxane in
surface water, including industrial releases from a range of COUs and DTD releases of consumer and
commercial products. In this draft supplement, EPA evaluated general population exposures resulting
from all known releases, including releases associated with COUs evaluated in the 2020 RE and releases
Page 27 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
1152 associated with new COUs introduced in this draft supplement due to the presence of 1,4-dioxane
1153 produced as a byproduct.
1154 1.3 Use Characterization
1155 1.3.1 Conceptual Models
1156 The life cycle diagram for 1,4-dioxane in Figure 1-1 summarizes the conditions of use that are within the
1157 combined scope of the 2020 RE and the current draft supplement. The life cycle diagram has been
1158 updated from the 2020 RE to highlight additional sources of 1,4-dioxane produced as a byproduct,
1159 including commercial products and industrial uses, releases, and disposals (e.g., PET manufacturing,
1160 ethoxylation byproducts, disposal of hydraulic fracturing produced waters).
Page 28 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
MFG/IMPORT PROCESSING INDUSTRIAL, COMMERCIAL, RELEASES and
CONSUMER USES WASTE DISPOSAL
Manufacture
(includes import)
(i million lbs.)
Processing as a
Rea eta nt/lnter mediate
(Not reported in 2016 CDR)
Industrial Uses
Repackaging
(270,000 lbs.)
Non-incorporative
Activities
(270,00 lbs.)
Recycling
Processing Aids, Not
Otherwise Listed
(270,000 lbs.)
e.g., wood pulping, etching
of fluropolymers
Functional Fluids
(Open and Closed Systems)
(<150,000 lbs.)
e.g., hydraulic fluid.
Laboratory Chemicals
(<150,000 lbs.)
e.g., laboratory reagent
Adhesive Sealants
e.g., film cement
Other Industrial Uses
e.g., spray polyurethane
foam, printing and printing
compositions; dry film
lubricant
Consumer Uses
{Present only as Byproduct)
Paints and Coatings
e.g., Latex Wall Paint or
Floor Lacquer
Cleaning and Furniture
Care Products
e.g.. Surface Cleaner
Laundry and Dishwashing
Products
e.g., Dish Soap, Dishwasher
Detergent, Laundry
Pete rgent
Arts, Crafts and Hobby
Materials
e.g.. Textile Dye
Automotive Care Products
e.g., Antifreeze
Other Consumer Uses
e.g., Spray Polyurethane
Foam, Antifreeze
Commercial Uses
(Present only as Byproduct)
Paints and Coatings
e.g.. Latex Wall Paint or
Floor Lacquer
Clea ning a nid Fu miture
Care Products
e.g., Surface Cleaner
Laundry and Dishwashing
Products
e.g.. Dish Soap, Dishwasher
Detergent, Laundry
Detergent
Arts, Crafts and Hobby
Materials
e.g.. Textile Dye
Automotive Care Products
e.g.. Antifreeze
Other Consumer Uses
e.g.. Spray Polyurethane
Foam, Antifreeze
Industrial uses including 1,4-dioxane as a byproduct
(e.g., ethoxylation processes, PET manufacturing, and hydraulic fracturing operations)
Disposal
Byproduct
U ses from
Manufacturing
1161
1162 Figure 1-1. 1,4-Dioxane Life Cycle Diagram
1163 Note: This life cycle diagram has been expanded from what was published in the 2020 RE to include additional sources of 1,4-dioxane produced as a
1164 byproduct (indicated in blue boxes). See Appendix D for a complete table of COUs considered in the 2020 RE and the current supplement.
1 Assessed in the Supplement to
I the Risk Evaluation
Page 29 of 484
-------
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
1.3.1.1 1,4-Dioxane as a Byproduct
1,4-Dioxane produced as a byproduct may be a source of exposure for several of the occupational and
general population exposure pathways evaluated in this draft supplement. Byproduct refers to a chemical
substance produced without a separate commercial intent during the manufacture, processing, use, or
disposal of another chemical substance(s) or mixture(s). In this assessment, the term byproduct is used
to refer to 1,4-dioxane produced during manufacturing or industrial processes, including 1,4-dioxane
that remains present in downstream processes or in consumer and commercial products.
In the 2020 RE, EPA evaluated risks to consumers and bystanders from 1,4-dioxane present as a
byproduct in consumer products. In this draft supplement, EPA expanded on the previous evaluation to
consider risks from all other pathways of exposure to 1,4-dioxane produced as a byproduct for which
information is reasonably available. Figure 1-2 summarizes both what is known about the processes that
may result in 1,4-dioxane production and how it may contribute to human exposures through a range of
exposure pathways.
Source
chemicals
Occupational
exposure
Chemical
processes
..in many types of
manufacturing
facilities
(ethoxylation,
esterification,
sulfonation,
sulfation)
Direct and indirect*
industrial releases to
air and water
General population
exposure through air
or drinking water
1,4-dioxane
produced as a
byproduct of
manufacturing
'1
Removal of 1,4-dioxane
prior to product
formulation and releases
Consumer and
occupational
exposure a
t
Consumer and
commercial products
Down-the-drain
releases*
*these releases may also go to wastewater treatment plants
Figure 1-2. Production of 1,4-Dioxane as a Byproduct and Potential Exposure Pathways
1,4-Dioxane is produced as a byproduct in several common manufacturing reactions, including in
manufacturing of PET plastics and in ethoxylation reactions during the manufacture of common
surfactants. In some facilities, additional processing steps may remove 1,4-dioxane produced as a
byproduct prior to product formulation and environmental releases, but the full extent of this practice
across industries is not known. Occupational exposure to 1,4-dioxane produced as a byproduct may
occur at manufacturing facilities and hydraulic fracturing operations. Releases of 1,4-dioxane from
manufacturing and industrial sites may also contribute to general population exposures through drinking
water and air.
1,4-Dioxane produced as a byproduct has also been detected in consumer and commercial products,
resulting in potential exposure to consumers and bystanders (evaluated in the 2020 RE) or workers and
ONUs (evaluated as described in Section 3.1.2). For example, 1,4-dioxane exposures are expected
during the use of dish soap and dishwashing detergent from unloading and transferring detergent
formulation, transport container cleaning, and washing operations due to the presence of 1,4-dioxane as
Page 30 of 484
-------
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
a surfactant byproduct. In addition, consumer and commercial products containing 1,4-dioxane may
contribute to general population exposures when released down the drain.
In this draft supplement, EPA evaluated pathways of exposure to 1,4-dioxane produced as a byproduct
that were not previously assessed. Specifically, EPA considered 1,4-dioxane present as a byproduct in
commercial products corresponding to the consumer products considered in the 2020 RE. EPA
considered the direct occupational exposures that result from use of these commercial products as well
as the DTD releases of consumer and commercial products, which contribute to general population
exposures through surface water. EPA also identified a new set of COUs where 1,4-dioxane is produced
or present as a byproduct, based on information submitted by public commenters and other reasonably
available information. For each of these new COUs, EPA evaluated occupational exposure as well as
industrial releases that contribute to general population exposures. The available information supporting
inclusion of each of the new COUs is described below. A more detailed list of the new COUs and COU
subcategories considered in this draft supplement is presented in Section 2.1.1. The following COUs are
known to produce 1,4-dioxane as a byproduct based on reasonably available information, but 1,4-
dioxane produced as a byproduct may also be present in other industries that have not yet been
identified:
• Industrial/Commercial Use of Products Containing 1,4-Dioxane as a Byproduct. 1,4-
Dioxane is present in a range of commercial products (including textile dyes, antifreeze, surface
cleaners, dish soaps, laundry detergents, and paint and floor lacquer) because it is produced as a
byproduct during the manufacture of ingredients such as ethoxylated surfactants. While 1,4-
dioxane present as a byproduct in consumer products was previously assessed, evaluation of 1,4-
dioxane in these corresponding commercial products is new in this draft supplement. 1,4-
Dioxane present in commercial products can result in occupational exposure in commercial
settings as well as DTD releases that contribute to general population exposures.
• PET Manufacturing. 1,4-Dioxane has been identified as a byproduct in the manufacture of PET
(l c> ' i1 \ 1 ). EPA does not have information on the byproduct concentration of 1,4-
dioxane in PET. PET is produced by the esterification of terephthalic acid to form
bishydroxyethyl terephthalate (BHET) (Forkner et ai. 2004). BHET polymerizes in a
transesterification reaction catalyzed by antimony oxide to form PET (Forkner et ai. 2004). In
2014, 20.6 million metric tons of PET were used in the United States (McDaniel and
DesLauriers. 2015). 1,4-Dioxane produced during PET manufacturing may result in occupational
exposures and may contribute to general population exposures via releases to water and air.
• Ethoxylation Processing. 1,4-Dioxane may be formed as a byproduct of reactions based on
condensing ethylene oxide or ethylene glycol during manufacture of detergents, shampoos,
surfactants, some food additives, and certain pharmaceuticals (HHS. 2016). In cosmetic
ethoxylated raw materials and ethoxylated alkyl sulfates, 1,4-dioxane has been detected at
concentrations of 0.48 to 1,410 ppm (U.S. EPA. 2020c; Saraii and Shirvani. 2017; Davarani et
al. 2012; Black et al. 2001). Polyethoxylated raw materials are widely used in cosmetic
products as emulsifiers, foaming agents, and dispersants (Black et al.. 2001). They are produced
by polymerizing ethylene oxide, usually with a fatty alcohol, to form polyethoxylated alcohols
which may be used to synthesize other products such as sulfated surface-active agent. During the
ethoxylation process, 1,4-dioxane can be formed as a byproduct by the dimerization of ethylene
oxide (Black et al.. 2001). The volume of 1,4-dioxane produced as a byproduct of ethoxylation
reactions is unknown. 1,4-Dioxane produced during ethoxylation processing may result in
occupational exposures and may contribute to general population exposures via releases to water
and air. It also contributes to the presence of 1,4-dioxane in consumer and commercial products.
Page 31 of 484
-------
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• Hydraulic Fracturing. Hydraulic fracturing stimulates an existing oil or gas well by injecting a
pressurized fluid containing chemical additives into the well ( I022d). 1,4-Dioxane is
measured in fracturing fluid, a water-based fluid that contains several chemical additives and in
waste fluid (produced waters). FracFocus 3.0 contains self-reported information indicating that
1,4-dioxane is present in hydraulic fracturing fluid additives, as scale inhibitors, additives,
biocides, friction reducers, and surfactants (GWPC and IOGCC. 2022). According to the
FracFocus 3.0 database, 1,4 dioxane is present in weight fractions ranging from 2.3 xl0~n to 0.05
within hydraulic fracturing additives and 1.00/10 12 to 4.30><10"6 in hydraulic fracturing fluids
(GWPC and IOGCC. 2022). 1.4-Dioxane has been documented to have a concentration of 60
|.ig/L in hydraulic fracturing produced waters (Lester et al. ). 1,4-Dioxane present in
fracturing fluid may result in occupational exposures. It may also contribute to general
population exposures via discharge to surface water, groundwater, or fugitive air emissions from
fracturing operations.
1.3.1.2 Occupational Exposures
The conceptual model Figure 1-3 Figure l-3presents the exposure pathways, exposure routes, and
hazards to people from industrial and commercial releases and uses of 1,4-dioxane. Blue shading
highlights the exposures evaluated in this draft supplement. Workers and ONUs may have acute (8 hr)
or chronic (annual to lifetime) exposures to 1,4-dioxane produced as a byproduct during PET
manufacturing, ethoxylation processes, or hydraulic fracturing operations. Workers and ONUs may also
have acute or chronic exposures to 1,4-dioxane present as a byproduct in commercial products,
including detergents, cleaners, and lacquers.
Page 32 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
INDUSTRIAL AND COMMERCIAL ACTIVITIES/USE EXPOSURE PATHWAY EXPOSURE ROUTE & RECEPTORS HAZARDS
DURATION
INDUSTRIAL USES
Manufacture
(Including Import)
Processing:
• Processing as a
reactant/i ntermediate
• Repackaging
• Non-incorporative
activities
Recycling
Processing Aid, Not
Otherwise Listed
Functional Fluids
(Open and Closed Systems)
Laboratory Chemicals
Other Industrial Uses
1,4-Dioxane as a Byproduct
COMMERCIAL USES
(Present only as Byproduct)
Paints and Coatings
e.g., Latex Wall Paint or
Floor Lacquer
Cleaning and Furniture
Care Products
e.g., Surface Cleaner
Laundry and Dishwashing
Products
e.g., Dish Soap, Dishwasher
Detergent, Laundry
Detergent
Arts, Crafts and Hobby
Materials
e.g., Textile Dye
Automotive Care Products
e.g., Antifreeze
Other Consumer Uses
e.g., Spray Polyurethane
Foam, Antifreeze
T
Waste Handling,
Treatment, and Disposal
-~ Liquid Contact
Dermal
(acute dose, chronic
ADD, LADD)
Hazards Potentially
Associated with
Acute and/or
Chronic Exposures
Vapor/Mist
i
L
l Fug
Emis
tive
sions
Inhalation
(8hrTWA, chronic
ADC, LA DC)
1265
1266
Wastewater, Liquid Wastes, Solid
Wastes
Assessed in the Supplement to
the Risk Evaluation
Figure 1-3. Conceptual Model for Occupational Exposures from Industrial and Commercial Activities
Page 33 of 484
-------
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1211
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
In this draft supplement, EPA evaluated acute and chronic cancer and non-cancer risks from
occupational inhalation and dermal exposures to 1,4-dioxane produced as a byproduct during PET
manufacturing and ethoxylation processes, hydraulic fracturing fluids and waste containing 1,4-dioxane,
and commercial products containing 1,4-dioxane.
1.3.1.3 General Population Exposures
The conceptual model in Figure 1-4 presents general population exposure pathways and hazards from
environmental releases and wastes associated with COUs (red, blue, and purple shading for each source
in the figure corresponds to the environmental media to which they release). The disposal and release
scenarios illustrated in Figure 1-1, Figure 1-2, and Figure 1-3 all contribute to the releases to air, water
and land that may result in the general population exposures illustrated in Figure 1-4. The general
population may be exposed to 1,4-dioxane released to surface water (blue shading), groundwater (red
shading), and air (purple shading). Drinking water exposures are evaluated based on releases to both
surface water and groundwater and these are each described in the drinking water subsection below.
EPA's evaluation of general population exposures considers potentially exposed or susceptible
subpopulations (PESS). Exposures to 1,4-dioxane through air and water could result in risk to fenceline
communities. As defined in the Draft Screening Level Approach for Assessing Ambient Air and Water
Exposures to Fenceline Communities Version 1.0,5 fenceline communities are members of the general
population that are in proximity to air emitting facilities or a receiving water body, and who therefore
may be disproportionately exposed to a chemical undergoing risk evaluation under TSCA section 6. For
the air pathway, proximity goes out to 10,000 m from an air emitting source. For the water pathway,
proximity does not refer to a specific distance measured from a receiving water body, but rather to those
members of the general population that may interact with the receiving water body and thus may be
exposed.
5 The draft Screening Level Approach for Assessing Ambient Air and Water Exposures to Fenceline Communities Version 1.0
is available at https://www.epa.gov/assessing~aiMl~managing~chemicals~iinder~tsca/tsca~screening~level~approach~assessing~
ambient-air-and.
Page 34 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
1292
1293
RELEASES AND WASTES FROM
INDUSTRIAL / COMMERCIAL / CONSUMER
USES
Industrial Pre-
Treatment or
Industrial WWT
EXPOSURE PATHWAYS
Indirect discharge
~
Wastewater or
Liquid Wastes
POTW
Down-the-Drain
Consumer Product Use
Hydraulic
Fracturing
Produced Water!
Underground
Injection
Hazardous and
Municipal
Waste Landfill
Solid Wastes
Liquid Wastes
Hazardous an
Municipal
Waste
Off-site Waste
T ransfer
Recycling, Other
Treatment
Emissions to Air
EXPOSURE ROUTES
RECEPTORS
HAZARDS
Hazards Potentially
Associated with
Acute and/or Chronic
Exposures
| Land Pathway
| Air Pathway
H Water Pathway
Included in 2020 RE (No Revisit)
Figure 1-4. Conceptual Model for Environmental Releases and General Population Exposures
Page 35 of 484
-------
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
The 2020 RE included an assessment of acute incidental/recreational general population exposure to 1,4-
dioxane in surface water resulting from industrial releases. It did not evaluate risks from other general
population exposure pathways such as drinking water or air.
In this draft supplement, EPA evaluated additional general population exposure via air and drinking
water. EPA evaluated acute (24-hour) and chronic (annual to lifetime) cancer and non-cancer risks from
these exposure pathways. Where data were reasonably available, EPA incorporated releases of 1,4-
dioxane produced as a byproduct into these pathways. The Agency also considered aggregate 1,4-
dioxane exposures and risks from multiple releasing facilities or COUs for each pathway.
1.3.1.3.1 Drinking Water
1,4-Dioxane may enter surface water through direct and indirect industrial releases, DTD releases from
consumer and commercial products via wastewater treatment facilities, and releases of wastewater from
hydraulic fracturing sites. Similarly, 1,4-dioxane released or disposed of through various land pathways
may reach groundwater under some conditions. There is potential for general population exposures to
1,4-dioxane if contaminated surface water or groundwater are used as drinking water. 1,4-Dioxane is
mobile in water and does not readily degrade in water. Available data indicate that typical wastewater
treatment and drinking water treatment methods are not effective at removing 1,4-dioxane. The
subsections below explain how general population exposures through surface and groundwater were
considered.
Surface Water Pathway
1,4-Dioxane was included in the third unregulated contaminant monitoring rule (UCMR3) (U.S. EPA.
2017d) published in May of 2012 requiring community water systems to monitor for 1,4-dioxane
between 2013 and 2015. National and state water monitoring programs have detected 1,4-dioxane in
drinking water and drinking water sources (as described in Section 2.3.1.10). In the absence of
monitoring data, estimating 1,4-dioxane surface water concentrations can be complex because in
addition to direct and indirect industrial and commercial releases, upstream sources from releasing
facilities and DTD releases of consumer and commercial products contribute to surface water
contamination.
EPA evaluated surface water concentrations (Section 2.3.1) and drinking water exposures (Section
3.2.2.1) that could result from direct and indirect industrial releases, DTD releases of consumer and
commercial products, and disposal of wastewater from hydraulic fracturing sites. EPA used a novel
aggregate model to predict water concentrations of 1,4-dioxane that could result from multiple sources
that release 1,4-dioxane to the same receiving water bodies. Surface water concentrations estimated for
each source in isolation and from multiple sources in aggregation were used to evaluate potential
exposures (Section 3.2.2.1) and risks (Section 5.2.2.1) from general population oral exposure to 1,4-
dioxane in drinking water.
Land Pathway to Groundwater
Groundwater contamination with 1,4-dioxane presents a potential risk when the chemical substance is
released to landfills, underground injection wells, or surface impoundments. Due to its physical-
chemical properties (e.g., water solubility, Henry's law constant) and fate characteristics (e.g.,
biodegradability, half-life in groundwater), 1,4-dioxane is anticipated to persist in groundwater for
months to years. This persistence has resulted in higher 1,4-dioxane concentrations in groundwater
relative to other media (A.TSD ). EPA considered potential for groundwater contamination
following disposal of waste containing 1,4-dioxane to landfills, underground injection of 1,4-dioxane
waste, and disposals of hydraulic fracturing waste containing 1,4-dioxane. Groundwater concentrations
Page 36 of 484
-------
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
estimated for each source (Section 2.3.2) were used to evaluate exposure (Section 3.2.2.2) and risks
(Section 5.2.2.1.6) for communities who rely on groundwater as a source of drinking water.
1.3.1.3.2 Air
Industrial releases to air include those from sites where 1,4-dioxane is manufactured intentionally as
well as those where it is produced or present as a byproduct. In this draft supplement, EPA evaluated
exposures and risks for communities located near release sites (fenceline communities) because they are
the members of the general population that are expected to be PESS due to their greater exposure. EPA
applied the methodology presented in the Draft TSCA Screening Level Approach for Assessing Ambient
Air and Water Exposures to Fenceline Communities6 to evaluate risks from industrial air releases to
fenceline communities. In addition to considering risks from individual facilities, EPA evaluated risks
from aggregate exposures in cases where multiple facilities reporting 1,4-dioxane releases to air were in
proximity. The Agency also evaluated potential risks to fenceline communities from air emissions of
1,4-dioxane modeled for hydraulic fracturing operations and industrial and commercial laundries.
1.3.1.3.3 Aggregate Exposure
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 § 702.33)." In this
draft supplement, EPA considered the combined 1,4-dioxane exposure an individual may experience due
to releases to air or water from multiple sources. For general population drinking water exposure
scenarios, EPA evaluated combined exposure and risks from multiple sources of 1,4-dioxane in surface
water, including direct and indirect industrial releases, DTD releases, and upstream background
contamination (Section 5.2.2.1). For general population air exposure scenarios, EPA evaluated
combined exposure and risk across multiple facilities in proximity releasing to air (Section 5.2.2.3 and
Appendix J.4). EPA qualitatively considered aggregate exposures across exposure routes {i.e., across
oral and inhalation) and across exposure pathways {i.e., across air and water) but did not quantitatively
aggregate these exposures due to uncertainties around the additivity of effects across routes. The
rationale for the scope of aggregate analysis in this draft supplement and remaining sources of
uncertainty are further discussed in Section 5.2.2.5.
1.3.2 Potentially Exposed or Susceptible Subpopulations
TSCA section 6(b)(4)(A) requires that risk evaluations "determine whether a chemical substance
presents an unreasonable risk of injury to health or the environment, without consideration of costs or
other non-risk factors, including an unreasonable risk to a potentially exposed or susceptible
subpopulation 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 subpopulatioif
[PESS] 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."
Considerations related to PESS can influence the selection of relevant exposure pathways, the sensitivity
of derived hazard values, the inclusion of particular populations, and the discussion of uncertainties
throughout the assessment. Factors that may contribute to increased exposure or biological susceptibility
to a chemical include lifestage, pre-existing disease, lifestyle activities {e.g., smoking, physical activity),
occupational and consumer exposures (including workers and occupational non-users, consumers and
6 The draft Screening Level Approach for Assessing Ambient Air and Water Exposures to Fenceline Communities Version 1.0
is available at https://www.epa.gov/assessing~aiMl~managing~chemicals~iinder~tsca/tsca~screening~level~approach~assessing~
ambient-air-and.
Page 37 of 484
-------
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
bystanders), geographic factors (e.g., fenceline communities), socio-demographic factors, nutrition,
genetics, unique activities (e.g., subsistence fishing), aggregate exposures, and other chemical and non-
chemical stressors.
This draft supplement considers PESS throughout the human health exposure assessment and risk
characterization. The hazard assessment and dose-response analysis used in this draft supplement
incorporate all PESS considerations described previously in the 2020 RE. Section 5.2.2.4 provides a
summary of how specific factors contributing to exposure and susceptibility were addressed in this
assessment and identifies remaining sources of uncertainty for PESS.
1.4 Systematic Review
EPA used the TSCA systematic review process described in the Draft Systematic Review Protocol
Supporting TSCA Risk Evaluations for Chemical Substances ( ) (hereinafter referred to
as "2021 Draft Systematic Review Protocol") to identify information needed to evaluate additional
COUs and exposure pathways considered in this draft supplement. Appendix C provides additional
information on the literature search strategy, data screening, evaluation, extraction, and evidence
integration steps performed in support of this assessment—including clarifications and updates made to
the 2021 Draft Systematic Review Protocol to better address assessment needs for this draft supplement.
1.5 Document Outline
This draft supplement to the risk evaluation for 1,4-dioxane comprises the following sections and
appendices:
• Section 1 presents information on the scope of the draft supplement. It also includes an overview
of the systematic review process used in this analysis. Appendix A provides a list of
abbreviations and acronyms used throughout this report while Appendix B provides the full
name and links to all supplemental documents associated with this draft risk evaluation. A more
detailed description of the systematic review protocol for this assessment is presented in
Appendix C while Appendix D provides a crosswalk of COUs with occupational exposure
scenarios.
• Section 2 presents an overview of releases and concentrations of 1,4-dioxane in the environment.
A more detailed description of the industrial and commercial environmental release assessment
is presented in Appendix E. Methods for estimating environmental concentrations of 1,4-dioxane
are described in more detail in Appendix G (surface water), Appendix H (groundwater), and
Appendix J (air).
• Section 3 presents the human exposure assessment for occupational and general population
exposure pathways. Details of the occupational exposure assessment are presented in Appendix
F and details of the general population exposure assessment are presented in Appendix I and
Appendix J.
• Section 4 provides a summary of the human health hazard and dose-response assessment
previously published in the 2020 RE and describes duration adjustments made for the current
analysis.
• Section 5 presents risk characterization based on the conditions of use and exposure pathways
evaluated in this draft supplement. Details of risk calculations and risk estimates are provided in
the set of supplemental risk calculator files listed in Appendix B. Section 5 also includes a
discussion of PESS based on both greater exposure and susceptibility, a description of aggregate
and sentinel exposures, and a discussion of assumptions and uncertainties and the impact on the
supplemental risk evaluation.
Page 38 of 484
-------
1430
1431
1432
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Figure 1-5 provides an overview of how the analyses presented in each section are integrated into risk
characterization.
Release Assessment
Section 2
Releases to
surface water
Releasesto
land & groundwater
Releasesto air
Exposure Assessment
Section 3
x N
Occupational
Exposure Scenarios
Inhalation exposure
Dermal exposure
a
General Population
Exposure Scenarios
Oral exposures
through drinking
water
Inhalation
exposure
through air
Risk Characterization
Section 5
Occupational Risk
Characterization
Inhalation risk
Dermal risk
General Population
Risk Characterization
Drinking water risk
Inhalation risk
Hazard Assessment
Section 4
Occupational
hazard values for
acute and chronic
non-cancerand
cancer (consistent
with 2020 RE)
General population
hazard valuesfor
acute, chronic non-
cancer and cancer
hazard values
(derived from PODs
in the 2020 RE, with
some duration
adjustments)
1433
1434 Figure 1-5. Overview of Analyses Included in this Draft Supplement to the Risk Evaluation for
1435 1,4-Dioxane
Page 39 of 484
-------
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2 RELEASES AND CONCENTRATIONS
2.1 Approach and Methodology
2.1.1 Industrial and Commercial Releases
Releases to the environment are one component of potential exposure and may be derived from reported
data that are obtained through direct measurement via monitoring, calculations based on empirical data,
and/or assumptions and models.
The original COUs for 1,4-dioxane are summarized in Table 1-4 of the Final Risk Evaluation for 1,4-
Dioxane ( s20c). Additional COUs included in this draft supplement due to 1,4-dioxane
produced as a byproduct are presented in Table 2-1. For general population exposures, this draft
supplement considers releases from all COUs (including the original COUs included in the 2020 RE and
the additional COUs associated with 1,4-dioxane produced as a byproduct). For occupational exposures,
this draft supplement focuses on the additional COUs associated with 1,4-dioxane produced as a
byproduct. For additional information and context on the inclusion of these COUs in the draft
supplement, refer to Sections 1.1 and 1.2. For the full table of COUs, including those previously
assessed in the Final Risk Evaluation for 1,4-Dioxane ( 2020c). see Appendix D.
A COU is a combination of life cycle stage, category, and subcategory, as shown in Table 2-1. The COU
subcategory is the most granular description of the use. EPA mapped each COU to an occupational
exposure scenario (OES). The purpose of an OES is to group, where appropriate, COUs based on
similarity of the operations and data availability for each COU. For each OES, EPA estimated air, land,
and water releases and occupational dermal and inhalation exposures. EPA mapped OESs to COUs
using professional judgment based on reasonably available data and information that describe how
releases and exposures take place within an occupational COU. EPA may group multiple COUs into an
OES if the release and exposure potential is similar across the COUs and there is insufficient data to
differentiate the COUs. This grouping minimized repetitive assessments. Alternatively, EPA may assign
multiple OESs to one COU if there are several ways in which release and exposure takes place for the
given COU and sufficient data exist to separately assess the OES. Appendix D.l shows mapping
between COUs and OESs. A crosswalk of the COUs with the OESs assessed is provided in Table 2-1.
As shown in Table 2-1, most COU life cycle stage, category, and subcategory combinations map to a
single OES with a similar or identical name to the COU subcategory. However, for the COU
subcategory of dish soap, dishwasher detergent, and laundry detergent, EPA assigned four OESs: (1)
dish soap, (2) dishwasher detergent, (3) laundry detergent (industrial), and (4) laundry detergent
(institutional). Institutional use of laundry detergent equates to commercial use.
EPA assessed environmental releases (air, water, and land) and occupational exposures (inhalation and
dermal) to 1,4-dioxane for each of the OESs listed in Table Apx D-l. EPA used the environmental
release estimates for each OES for subsequent environmental concentrations and general population
exposure calculations.
Page 40 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
1476 Table 2-1. Additional Categories and Subcategories of COUs and Associated OESs Included in the Scope of the Draft Supplement
1477 Due to the Presence of 1,4-Dioxane Produced as a Byproduct"
Condition of Use
OES Mapped to COU
Life Cycle Stage
Category''
Subcategory'
Processing
Byproduct
Byproduct produced during the ethoxylation processes
Ethoxylation Process Byproduct
Byproduct produced during the production of
polyethylene terephthalate
Polyethylene Terephthalate (PET)
Byproduct
Industrial use,
commercial use
Other uses
Hydraulic fracturing
Hydraulic Fracturing
Consumer use,
commercial use
Paints and coatings
Latex Wall Paint or Floor Lacquer
Paint and Floor Lacquer
Cleaning and furniture care
products
Surface Cleaner
Surface Cleaner
Laundry and dishwashing products
Dish soap
Dishwasher detergent
Laundry detergent
Dish Soap
Dishwasher Detergent
Laundry Detergent (Industrial)''
Laundry Detergent (Institutional)''
Arts, crafts, and hobby materials
Textile dye
Textile Dye
Consumer use,
commercial use
Automotive care products
Antifreeze
Antifreeze
Disposal
Disposal
Industrial pre-treatment
Disposal
Industrial wastewater treatment
Publicly owned treatment works (POTW)
Underground injection
Municipal landfill
Hazardous landfill
Other land disposal
Municipal waste incinerator
Hazardous waste incinerator
Off-site waste transfer
aNew COUs and associated OESs where 1,4-dioxane is produced as a byproduct.
b These categories of COU reflect CDR codes and broadly represent conditions of use for 1,4-dioxane in industrial and/or commercial settings.
c These subcategories reflect more specific uses of 1,4-dioxane.
dLaundry detergent use may occur in industrial, commercial, or consumer settings. Sufficient information was available to separately assess each use setting.
Thus, there are two OESs—one for industrial and one for institutional, which equates to commercial use.
1478
Page 41 of 484
-------
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2.1.1.1 General Approach and Methodology for Environmental Releases
Data reported to the Toxics Release Inventory (TRI)7 and discharge monitoring reports (DMRs) are the
primary sources of release data that EPA used for the release assessments. Under the Emergency
Planning and Community Right-to-Know Act (EPCRA) section 313, 1,4-dioxane has been a TRI-
reportable substance since 1987. The TRI database includes information on disposal and other releases
of 1,4-dioxane to air, water, and land, in addition to how it is being managed through recycling,
treatment, and burning for energy recovery. Under the Clean Water Act (CWA), surface water
discharges reported in DMRs are based on required monitoring as part of a facility's National Pollutant
Discharge Elimination System (NPDES) permit.
For each OES, a daily air, land, and water release was estimated based on annual releases, release days,
and the number of facilities (Figure 2-1). The blue boxes represent primary sources of release data that
are used to develop annual releases, release days, and number of facilities. The information in the green
boxes is aggregated by OES to provide a daily release estimate.
Figure 2-1. Overview of EPA's Approach to Estimate Daily Releases for Each OES
TRI = Toxics Release Inventory; DMR = discharge monitoring report; ESD = emission scenario
document; GS = generic scenario
2.1.1.2 Water Release Estimates
EPA followed a similar approach for estimating industrial and commercial water releases as it did in the
2020 RE, with one key difference. Here, EPA evaluated multiple years of data using data from 2013 to
2019 TRI (U.S. EPA. 2022g) and 2013 to 2019 DMR (U.S. EPA. 2022c\ as opposed to utilizing 1 year
of data.
Where water releases are expected for an OES but TRI and DMR data were not available, EPA
estimated industrial, and commercial water releases using two approaches. If available, Stochastic
Human Exposure and Dose Simulation for High Throughput (SHEDS-HT) DTD modeling was used.
SHEDS-HT predicts a per capita DTD loading of 1,4-dioxane, which is combined with an estimation of
the population contributing to publicly owned treatment works (POTW) effluent on the modeled water
7 TRI page: https://www.epa.gov/toxics-releasE-inventorv-tri-program.
Page 42 of 484
-------
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
body stream to produce an estimated DTD loading. If SHEDS-HT DTD modeling was not available for
an OES without TRI or DMR data, EPA used data from literature, ESDs, and GSs in conjunction with
Monte Carlo simulation where sufficient data were available to vary calculation input parameters to
estimate industrial and commercial water releases (see Appendix E.3 for additional information).
For the following OESs, EPA either could not estimate water releases due to lack of reasonably
available data or information or did not expect water releases based on volatility and use patterns:
• Functional Fluids (Closed-Systems). Water release data were not available for this OES.
However, EPA expects that the sources of release for this OES to be similar to those for the
Industrial Uses OES (per process information in the 2020 RE, Appendix G.6.4). Therefore, EPA
grouped the water release assessment for Functional Fluids (Closed-Systems) into that for
Industrial Uses. However, there is uncertainty in this assumption of similar release sources
between these OESs.
• Laboratory Chemical, Film Cement, and Dry Film Lubricant. Wastewater discharges
containing 1,4-dioxane were not expected for these OESs; releases from these OESs are expected
to be to air from volatilizations and landfill/incineration from disposal of empty containers and
other waste (2020 RE, Appendix G).
• Antifreeze. Wastewater discharges containing 1,4-dioxane were not expected for this OES;
releases from this OES are expected to be to air from volatilizations during antifreeze changeouts
and to landfill/incineration from disposal of empty antifreeze containers and spent antifreeze.
• Paints and Floor Lacquer. Wastewater discharges containing 1,4-dioxane were not expected
for this OES; releases from this OES are expected to be to air from volatilizations during
painting/drying and to landfill/incineration from disposal of empty paint containers, used paint
brushes/rollers, or solvent washes of paint brushes/rollers.
2.1.1.3 Land Release Estimates
EPA used data from 2019 TRI ( ) to estimate industrial and commercial land releases
that were mapped to each OES with the exception of the Disposal OES. For that OES, EPA performed a
more detailed analysis using data from 2013 to 2019 TRI ( Z022g). Where land releases are
expected for an OES, but TRI data were not available, releases were estimated using reasonably
available data from literature, ESDs, and GSs in conjunction with Monte Carlo simulation (Palisade.
2022a) to allow for variability in calculation input parameters where sufficient data were available to
inform such variability.
EPA did not estimate daily land releases due to the high level of uncertainty in the number of release
days. This uncertainty is because facility operating days does not correlate directly to releases. For
example, a facility may wait until a dumpster or other waste receptacle is full before disposing of it.
Because these releases may occur on a daily, weekly, and even monthly scale, EPA used the annual land
releases reported in TRI data or modeled without estimating land releases for a different frequency
(daily, weekly, monthly). See Appendix E.4 for additional information. Annual land release estimates
were used to estimate potential groundwater contamination from landfill releases as described in Section
2.3.2.
For the following OESs, EPA was not able to estimate land releases due to lack of data or information or
did not expect land releases due to physical form, use patterns, and lack of data:
• Manufacturing, Import and Repackaging, and Functional Fluids (Open-Systems). Data
from 2019 TRI ( 22e) indicated that there were no releases of 1,4-dioxane to land
Page 43 of 484
-------
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
from facilities that EPA mapped to these OESs. EPA did not have additional reasonably
available information to estimate land releases from these OESs.
• Functional Fluids (Closed-Systems). See previous explanation in the "Water Release
Estimates" section above.
• 3D Printing. Industrial applications of this OES are expected to be accounted for in the
Industrial Uses TRI data. Per Appendix G.6.8 of the 2020 RE, 3D printing ink containing 1,4-
dioxane is used in research labs to print biomedical products. Because the 2019 TRI data for the
Industrial Uses OES include medicinal and pharmaceutical manufacturing NAICS codes,
medical research labs that conduct 3D printing with 1,4-dioxane inks may be captured in that
OES. Therefore, EPA grouped the land release assessment for 3D Printing into that OES for
Industrial Uses. However, the extent to which all potential 3D printing sites that use 1,4-dioxane
are captured in the Industrial Uses TRI data is unknown.
EPA also notes that the Hydraulic Fracturing OES is associated with certain specific land releases that
may not apply to other OESs, such as the releases of wastewater containing 1,4-dioxane to deep well
injection or surface impoundments, which are considered land releases in this assessment. The Agency
estimated these deep well injection and surface impoundment releases, which were used in addition to
landfill releases, to estimate potential groundwater contamination from hydraulic fracturing described in
Section 2.3.2.
2.1.1.4 Air Release Estimates
EPA applied the following tiered approach to developing air release, exposure, and risk estimates:
1. Pre-screening analysis,
2. Single-year fenceline analysis, and
3. Multi-year fenceline analysis.
2.1.1.4.1 Pre-screening Analysis
This analysis is described in the Draft TSCA Screening Level Approach for Assessing Ambient Air and
'Water Exposures to Fenceline Communities and consisted of extracting data for all facilities reporting
1,4-dioxane air releases to the 2019 TRI ( 22g). The extracted data were reviewed to
identify the maximum single facility release reported across all reporting facilities. Additionally, the
arithmetic average (mean) value of all reported releases was calculated. These two release values were
used for further analysis to estimate exposure concentrations at select distances from the releasing
facility.
2.1.1.4.2 Single-Year Fenceline Analysis
Where available, EPA used data from 2019 TRI to estimate industrial and commercial air releases in
accordance with the Draft TSCA Screening Level Approach for Assessing Ambient Air and 'Water
Exposures to I7emetine Communities. Facilities are only required to report to TRI if the facility has 10
or more full-time employees; is included in an applicable North American Industry Classification
System (NAICS) code; and manufactures, processes, or uses the chemical in quantities greater than a
certain threshold. Due to these limitations, some sites that manufacture, process, or use 1,4-dioxane may
not report to TRI and are therefore not included in these datasets.
Where air releases are expected for an OES, but TRI data were not available, industrial, and commercial
air releases were estimated using data from literature, ESDs, and GSs in conjunction with Monte Carlo
simulation (Palisade. 2022a) to allow for variability in calculation input parameters where sufficient data
were available to inform such variability. See Appendix E.5 for additional information.
Page 44 of 484
-------
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
For the following OESs, EPA was not able to estimate air releases due to lack of data or information:
• Functional Fluids (Closed-Systems). See previous explanation in the "Water Release
Estimates" section above.
• 3D Printing. Industrial applications of this OES are expected to be accounted for in the
Industrial Uses TRI data. Per Appendix G.6.8 of the 2020 RE, 3D printing ink containing 1,4-
dioxane is used in research labs to print biomedical products. Because the 2019 TRI data for the
Industrial Uses OES include medicinal and pharmaceutical manufacturing NAICS codes,
medical research labs that conduct 3D printing with 1,4-dioxane inks may be captured in that
OES. Therefore, EPA grouped the air release assessment for 3D Printing into that OES for
Industrial Uses. However, the extent to which all potential 3D printing sites that use 1,4-dioxane
are captured in the Industrial Uses TRI data is unknown.
• Textile Dyes, Dish Soap, and Dishwasher Detergent. EPA did not find relevant reasonably
available 1,4-dioxane or surrogate TRI data, literature sources, sufficient process information,
nor ESD or GS with air release estimation approaches to estimate air releases for these OESs.
Therefore, EPA was not able to estimate air releases for these OESs.
2.1.1.4.3 Multi-Year Analysis
The multi-year analysis incorporates Science Advisory Committee on Chemicals (SACC)8
recommendations on the Draft TSCA Screem pproach for Assessing Ambient Air and Water
Exposures to Fenceline Communities to evaluate multiple years of chemical release data to estimate
exposures and associated risks to fenceline communities. This is achieved by conducting a facility-by-
facility evaluation of all 1,4-dioxane releases reported to TRI from 2015 through 2020. Data for these 6
years were obtained from the TRI database (TRI basic plus files downloaded on August 5, 2022).
Annual release data for 1,4-dioxane were extracted from the entire TRI data set for all facilities
reporting air releases of 1,4-dioxane for one or more years between 2015 and 2020. Facilities were
categorized into occupational exposure scenarios for exposure modeling purposes and later cross-walked
to COUs for risk management purposes.
2.2 Environmental Releases
2,2.1 Industrial and Commercial Releases
This section summarizes the estimated air, water, and land releases for each OES; the weight of the
scientific evidence conclusions for these estimates; and the strengths, limitations, assumptions, and key
sources of uncertainty for these estimates.
2.2.1.1 Release Estimates Summary
EPA estimated air, water, and land releases of 1,4-dioxane using various methods and information
sources—including TRI and DMR data, GSs and ESDs modeling with and without Monte Carlo,
process information, and SHEDS-HT DTD Modeling. EPA combined its estimates for annual releases,
release days, and number of facilities to estimate a range of daily releases for all OESs, including those
presented in the December 2020 RE. The COUs associated with each OES are summarized in
TableApx D-l. A summary of these industrial and commercial releases for air, water, and land are
presented in Table Apx E-3, Table Apx E-5, and Table Apx E-7, respectively. These release estimates
are for total releases from a facility and may include multiple points of release, such as multiple outfalls
for discharges to surface water or multiple point sources for air emissions. Note that for some release
8 Additional information about SACC is available at https://www.epa.gov/tsca-peer-review/science-advisore-cotninittee-
chemicals-basic-information.
Page 45 of 484
-------
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
estimates, there is uncertainty and variability in the potential media of release. In such cases, EPA did
not have sufficient information to partition the release estimates between all potential media of release
and they are replicated between the air, land, and water subsections if there is overlap in the potential
media of release.
EPA mapped these releases by media, state, and tribal territory for the conterminous United States.
Alaska, American Samoa, Guam, Hawaii, N. Mariana Islands, Puerto Rico, and the U.S. Virgin Islands
are not mapped since no estimated releases are known.
Surface water releases as reported by TRI and DMR are presented in Figure 2-2. For surface water
releases, the data are divided based on the source of data (i.e., DMR, TRI) and whether the release is
from a direct (on-site) source, including on-site wastewater treatment systems, or indirect (offsite)
source where the chemical substance was taken to a different location for potential release, such as a
POTW. The largest releases have been from PET manufacturing in Tennessee (2,512,434 kg in 2019
and 15,168 kg in 2018), Alabama (170,526 kg in 2015; 125,903 kg in 2014; and 111,924 kg in 2017),
and West Virginia (14,134 kg in 2016 and 12,229 kg in 2014).
(7-P Tribal Lands
Source of Water Release
O Direct Water Releases (TRI)
• Indirect Water Releases (TRI)
• Direct Releases (DMR)
Annual Mass Released (kg/yr)
o 0.00-10.0
O >10.0-100
O >100- 1,000
O >1,000- 10,000
O >10,000 - 100,000
O >100,000- 1,000,000
o >1,000,000-2,512,434
0 200 400 800 1,200 1,600
¦ Kilometers
Figure 2-2.1,4-Dioxane Annual Water Releases as Reported to TRI and DMR, 2013-2019
Note: Some symbols for individual years may overlap and obscure annual releases at each site.
Alaska, American Samoa, Guam, Hawaii, N. Mariana Islands, Puerto Rico, and the U.S. Virgin Islands are not
shown due to no known modeled or estimated releases.
Page 46 of 484
-------
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Land disposals as reported by TRI are available in Figure 2-3. The largest disposals have been to on-site
Class I Underground Injection Wells in Texas (169,035 kg in 2013; 42,865 kg in 2015; 10,729 kg in
2018), On-site Subtitle C Landfills in Oregon (7,321 kg in 2014; 7,000 kg in 2013; and 6,076 kg in
2015), and Offsite Other Landfills in Indiana (862 kg in 2019; 603 kg in 2018; and 354 kg in 2017). Air
release as reported by TRI are available in Figure 2-4.
Tribal Lands
On-site Disposal
• Disposal to RCRASubtitle C Landfill
¦ Underground I njection to Class 1 Wells
Offsite Disposal
a. Disposal to Other Landfills
c Disposal to RCRASubtitle C Landfills
¦ Underground Injection to Class 1 Wells
Mass of Disposal (kg)
0.001 - 1.00
o >1.00-10.0
O >10.0-100
O >100-862
200 400 800 1,200 1,600
l Kilometers
Figure 2-3.1,4-Dioxane Annual Releases to Land as Reported to TRI, 2013-2019
Note: Some symbols for individual years may overlap and obscure annual releases at each site.
Alaska, American Samoa, Guam, Hawaii, N. Mariana Islands, Puerto Rico, and the U.S. Virgin Islands are not
shown as there are no known modeled or estimated releases.
For air releases, the largest emissions have been in Illinois (9,943 kg/year), South Carolina (3,495
kg/year), and Texas (2,097 kg/year). Collectively, these figures give insight into the spatial distribution
of releases and corresponding amount across the contiguous United States. A full summary of these
estimates can be found in Appendix E.
Page 47 of 484
-------
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Tribal Lands
Total Air Emissions (kg/yr)
O 0.00-10.0
• >10.0-100
• >100 - 1,000
• >1,000 - 10,000
• >10,000 - 100,000
200 400 800 1,200 1,600
I Kilometers
Figure 2-4. 1,4-Dioxane Annual Releases to Air as Reported by TRI, 2013-2019
Note: Some symbols for individual years may overlap and obscure annual releases at each site.
Alaska, American Samoa, Guam, Hawaii, N. Mariana Islands, Puerto Rico, and the U.S. Virgin Islands are not
shown as there are no known modeled or estimated releases.
In addition to mapping releases from TRI and DMR, EPA also mapped hydraulic fracturing sites
reporting the presence of 1,4-dioxane in hydraulic fracturing operation fluids according to FracFocus 3.0
(GWPC and IOGCC. 2022). These operations are primary sited in a wide range of shale plays across the
country (as indicated by the multi-colored plays mapped in Figure 2-5). The Delaware play in Texas has
the largest number of operations (n = 158) followed by the Niobrara in Colorado (n = 86) and the Utica
play that spreads across Pennsylvania and Ohio (n = 70).
Page 48 of 484
-------
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Figure 2-5. Locations of Hydraulic Fracturing Operations that Report 1,4-Dioxane in
Produced Waters
Note: Some symbols for individual years may overlap and obscure annual releases at each site.
Alaska, American Samoa, Guam, Hawaii, N. Mariana Islands, Puerto Rico, and the U.S. Virgin Islands are
not shown as there are no known modeled or estimated releases.
The basis for overall data quality determinations is also included in the water, air, and land summary
subsections below. Each source is evaluated on multiple metrics based on defined criteria. For air, water,
and land releases, all monitoring data had data quality ratings of medium/high. Modeled data had data
quality ratings of medium or high.
2.2.1.2 Weight of the Scientific Evidence Conclusions for Environmental Releases
EPA's judgment on the weight of the scientific evidence is based on the strengths, limitations, and
uncertainties associated with the release estimates. The Agency considers factors that increase or
decrease the strength of the evidence supporting the release estimate—including quality of the
data/information, applicability of the release data to the COU (including considerations of temporal
relevance, locational relevance) and the representativeness of the estimate for the whole industry. The
best professional judgment is summarized using the descriptors of robust, moderate, slight, or
indeterminant, according to EPA's Application of Systematic Review in TSCA Risk Evaluations (U.S.
EPA. 2018c). For example, a conclusion of moderate weight of the scientific evidence is appropriate
where there is measured release data from a limited number of sources such that there is a limited
number of data points that may not cover most or all of the sites within the COU. A conclusion of slight
weight of the scientific evidence is appropriate where there is limited information that does not
Page 49 of 484
-------
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
sufficiently cover all sites within the COU, and the assumptions and uncertainties are not fully known or
documented. See EPA's Application of Systematic Review in TSCA Risk Evaluations (U.S. EPA.. 2018c)
for additional information on weight of the scientific evidence conclusions.
For air, water, and land releases, all monitoring data had data quality ratings of medium/high. Modeled
data had data quality ratings of medium or high. For supplemental releases modeled with TRI/DMR
(PET Byproduct, Ethoxylation Byproduct, Disposal), the weight of the scientific evidence conclusion
was moderate to robust because the reasonably available information relevant for the conditions of use
of 1,4-dioxane at facilities in TRI and DMR is limited. For releases that used SHEDS-HT modeling
(Surface Cleaner, Dish Soap, Dishwasher Detergent), the weight of the scientific conclusion was slight
since there is uncertainty in the application of this modeling for a commercial setting, and this case study
does not represent all sites in this OES. For supplemental releases that used GS/ESDs, the weight of the
scientific conclusion was moderate when used in tandem with Monte Carlo modeling (Textile Dye,
Laundries), and slight/moderate when used alone (Antifreeze, Paint and Floor Lacquer). For Hydraulic
Fracturing, the weight of the scientific conclusion was moderate to robust since FracFocus 3.0, an ESD,
and Monte Carlo modeling were used. See Appendix E.8 for a summary of EPA's overall weight of the
scientific evidence conclusions for its release estimates for each of the assessed OESs.
2.2.1.3 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Environmental Release Assessment
EPA estimated air, water, and land releases of 1,4-dioxane using various methods and information
sources, including TRI and DMR data, GSs and ESDs modeling with and without Monte Carlo, process
information, and SHEDS-HT DTD Modeling.
TRI and DMR were determined to have the overall data quality determination of medium through
EPA's systematic review process. Uncertainties for using TRI and DMR data are discussed in the Final
Risk Evaluation for 1,4-Dioxane ( :0c). In summary, these uncertainties include
underestimation of the number of sites for a given OES due to reporting thresholds in TRI, the accuracy
of EPA's mapping of sites reporting to TRI and DMR to a specific OES, and quality of the data reported
to TRI and DMR. There is additional uncertainty for sites that report to TRI with Form A because these
sites do not report release quantities if the quantity did not exceed 500 lb for the total annual reportable
release amount. For these sites, EPA assessed a "what-if' scenario, which assumes the entire 500 lb
going to single media of release, noting that the 500 lb should not be added over all release media.
Additional information on TRI uncertainties is provided in Appendix E.5. In addition, as discussed in
Section 2.2.1, EPA used data from the 2019 reporting year to estimate air and land releases. A key
source of uncertainty in the assessment of air and land releases is whether 2019 TRI data is
representative of releases from other reporting years. This does not apply to the water release estimates
because EPA used data from reporting years 2013 to 2019. A strength of using TRI is that it compiles
the best readily available release data for all facilities that reported to EPA. However, not all facilities
are required to report to TRI.
Some uncertainties of using DMR data include the accuracy of EPA's mapping of sites reporting to
DMR to a specific OES, and quality of the data reported to DMR. Also, an uncertainty of using the
ECHO Pollutant Loading Tool Advanced Search option is that average measurements may be reported
as a quantity (kg/day) or a concentration (mg/L). Calculating annual loads from concentrations requires
adding wastewater flow to the equation, which increases the uncertainty of the calculated annual load. In
addition, for facilities that reported having zero pollutant loads to DMR, the EZ Search Load Module
uses a combination of setting non-detects equal to zero and as one-half the detection limit to calculate
the annual pollutant loadings. This method could cause overestimation or underestimation of annual and
Page 50 of 484
-------
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Daily pollutant loads. A strength of using DMR data and the Pollutant Loading Tool is that the tool
calculates an annual pollutant load by integrating monitoring period release reports provided to the EPA
and extrapolating over the course of the year. However, this approach assumes average quantities,
concentrations, and hydrologic flows for a given period are representative of other times of the year.
Where TRI and DMR data were not reasonably available, EPA used GS and ESDs. One uncertainty for
this method is lack of specific 1,4-dioxane data. Because GS/ESDs are generic, assessed parameter
values may not always be representative of applications specific to 1,4-dioxane use in each OES.
Another uncertainty is lack of consideration for release controls. The GS/ESDs assume that all activities
occur without any release controls, and in an open-system environment where vapor and particulates
freely escape (U.S. EPA. 2022d; QECn 1 , J01 la. b). Actual releases may be less than estimated if
facilities utilize pollution control methods. Although 1,4-dioxane monitoring data are preferred to
modeled data, EPA strengthened modeled estimates by using Monte Carlo modeling to allow for
variation in environmental release calculation input parameters according to the GS/ESD and other
literature sources. However, EPA did not utilize Monte Carlo modeling for all GS/ESD, which is a
limitation of this assessment. TableApx E-8 includes information on which GS/ESDs were used in
tandem with Monte Carlo modeling.
EPA used process information to quantify environmental releases for the film cement and dry film
lubricant OESs. This process information is from the 2020 RE ( 20c) and the underlying
sources were determined to have high overall data quality determinations through EPA's systematic
review process. To develop these release estimates, EPA made assumptions on the likely media of
release for various releases sources and, in some cases, used standard EPA models in conjunction with
process information to estimate the release quantity. A source of uncertainty in this approach is the
representativeness of these estimates regarding all sites that use 1,4-dioxane for this OES.
Where no other data or information was reasonably available, EPA used SHEDS-HT down the drain
(DTD) modeling to estimate commercial use environmental releases to surface water or land
(Appendices E.3.2 and E.4.2). The results for this analysis are included in Table Apx E-3. The main
source of uncertainty is that the modeling EPA performed to estimate the total release amounts from
each COU to surface water or land (via disposal to landfills) is based on information for SHEDS-HT in
combination with information from a single case study location, Liverpool, OH. It is uncertain whether
the release estimates generated from this case study are applicable to other areas of the country. EPA
was unable to estimate the number of sites in Liverpool, OH, for the OES where this modeling approach
was used; therefore, the release estimates were presented as totals for all sites as opposed to per-site
estimates. To estimate land release, EPA used the modeled water releases from SHEDS-HT and back-
calculating a 1,4-dioxane use rate based on the expected loss fraction to water for the OES. Finally, a
land release loss fraction was applied to the back-calculated use rate. The uncertainty in this approach is
due to the standard models and assumptions used to estimate loss fractions to water and land.
Additionally, the same uncertainties listed above for the use of SHEDS-HT to estimate water releases
are applicable to the approach for estimating land releases. EPA is unsure whether the use of SHEDS-
HT results in a high-end or typical exposure scenario, so the use of this data may lead to over or
underestimates of releases.
To assess daily air and water discharges, EPA divided annual release loads by the number of facility
release days to estimate the daily release load for the facility. There is uncertainty if the assumed release
duration is applicable to all sites for a given OES; therefore, the average daily releases may be higher if
sites have fewer release days or lower if they have greater release days. Furthermore, 1,4-dioxane
concentrations in air emissions and wastewater release to receiving water bodies at each facility may
Page 51 of 484
-------
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
vary from day-to-day such that on any given day the actual daily releases may be higher or lower than
the estimated average daily discharge. Thus, this approach minimizes variations in emissions and
discharges from day to day. EPA did not estimate daily land releases due to the high level of uncertainty
in the number of release days associated with land releases. The Agency expects that sites may not send
waste to landfills every day and are more likely to accumulate waste for periodic shipments to landfills.
However, sites that release to municipal landfills may have more frequent release days based on the
frequency of shipments.
2.3 1,4-Dioxane Environmental Concentrations
2.3.1 Surface Water Pathway
Surface water contamination from 1,4-dioxane can occur from direct releases of wastewater from
industrial operations, discharges from wastewater treatment plants containing DTD releases of 1,4-
dioxane from consumer and commercial product usage {i.e., dish soap, laundry detergent, etc.), and
other activities where 1,4-dioxane may be present as a byproduct, such as in hydraulic fracturing
operations. To understand possible exposure scenarios from these practices, EPA assessed exposures to
the general population from ambient surface and drinking water. These exposures are due to 1,4-dioxane
being directly or indirectly discharged to receiving water bodies.
The evaluation of these exposures considered both the review of reasonably available monitoring data to
both ambient surface water and drinking water as well as the modeling of estimated exposures due to
releases. Although EPA identified a robust set of surface and drinking water monitoring data (Section
2.3.1.1) indicating the presence of 1,4-dioxane in these pathways, it was collected independent of release
data, and cannot be attributed to specific sources (Section 2.2). Therefore, EPA relied primarily on a
series of modeling approaches to estimate concentrations of 1,4-dioxane in surface water near known
release locations (Sections 2.3.1.2 and 2.3.1.3). For this assessment, EPA modeled concentrations
resulting from industrial releases for all COUs releasing to surface water, including those assessed in the
2020 RE, as well as those producing 1,4-dioxane as a byproduct. To the degree possible, the relationship
between monitoring and modeled data is further evaluated in Section 2.3.1.4.
2.3.1.1 Monitoring Data
Environmental concentration data for 1,4-dioxane in ambient surface water {i.e., measured in rivers,
streams, lakes, and ponds, rather than within industrial operations or drinking water systems) across the
country, as well as routine monitoring conducted by public water systems (PWSs) of raw (untreated)
source water and finished (treated) drinking water were collected from readily available public databases
and publications. The methods for retrieving and processing ambient surface water and PWS data are
described in detail in Appendix G. 1.
Ambient Surface Water Monitoring
Data were retrieved from the Water Quality Portal (WQP) to characterize observed concentrations in
ambient surface water (NWQMC. 2022). These monitored values may or may not represent locations
used as a source for drinking water and are analyzed to characterize the observed ranges of 1,4-dioxane
concentrations in ambient surface water—irrespective of the reasons for sample collection—and to
provide context for the modeled surface water concentrations presented in Section 2.3.1.3. Data
retrieved in July 2022 included sampling dates from 1997 to 2022 and resulted in 12,471 available
sample results. Full details of the retrieval and processing of ambient surface water monitoring data from
the WQP are presented in Appendix G. Table 2-2 shows the range of 1,4-dioxane concentrations
detected in surface water samples. Most {i.e., 92.3 percent) of the sample records available had no level
of 1,4-dioxane detected above the reported detection limit for the analysis (referred to as "non-detects"),
Page 52 of 484
-------
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
with limits of detection ranging from 0.001 to 28,000 |ig/L across all samples. The 105 detected values
ranged from 0.016 to 470 |ig/L, with a median of 1.10 |ig/L. Figure 2-6 and Figure 2-7 show the
distribution of detected concentrations and reported detection limits of non-detect samples, respectively.
The highest concentrations reported in this dataset are noted in the metadata to have been collected at
the point of discharge from an industrial facility, while for most samples, the reason for sampling, or
sample location in relation to expected releases is not included in the metadata. Figure 2-8 shows the
spatial distribution of detected samples. For the entire dataset, including non-detects, approximately 70
percent of the samples were collected from the states of North Carolina, New Mexico, and New Jersey.
Of the 105 detected values, 46 percent are located in Pennsylvania, 21 percent in North Carolina, and 14
percent in Illinois. In the absence of a national standardized study of 1,4-dioxane in ambient surface
water (analogous to the UCMR monitoring in drinking water), and without more national coverage and
metadata, it is difficult to characterize the national occurrence of 1,4-dioxane in surface water. It is
apparent from the available monitoring data that certain areas may be more likely to have higher
concentrations, while many others have little or no detected 1,4-dioxane. Over-representation of certain
states or regions may reflect targeted sampling campaigns of specific locations expected to have higher
concentrations, and conclusions about areas without monitoring data cannot be drawn without further
exploration through modeling.
Detected 1,4-Dioxane Concentration (|jg/L)
Figure 2-6. Frequency of Nationwide Measured 1,4-Dioxane Surface Water Concentrations
Retrieved from the Water Quality Portal, 1997-2022
Note: Detectable levels of 1,4-dioxane may vary by location.
Page 53 of 484
-------
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
c
CD
250-
c
o
OJ
?no-
1—
c
CI)
o
c
150 -
o
o
4—
o
100-
>N
o
c
CI)
50-
cr
0
LL
i i 11 in|" i i 1111n| i i i I nn| i i i 111n| i i i I iiii| i i i 11 ii'i| ' i i i 11 iii^ i ill 1111| i ^ i I ii
10~3 10~2 1CT1 10° 101 102 103 104
Detection Limit for 1,4-Dioxane Non-Detect (pg/L)
Figure 2-7. Frequency of Detection Limits for Nationwide Non-detect 1,4-Dioxane
Surface Water Samples Retrieved from the Water Quality Portal, 1997-2022
Figure 2-8. Detectable Concentrations of 1,4-Dioxane in Surface Water from the Water
Quality Portal, 1997-2022
Note: Detectable levels of 1,4-dioxane may vary by sampling location.
Alaska, American Samoa, Guam, Hawaii, N. Mariana Islands, Puerto Rico, and the U.S. Virgin
Islands are not shown as there are no known monitoring data above detection limits.
Page 54 of 484
-------
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Drinking Water Monitoring Data
The Safe Drinking Water Act (SDWA) authorizes the United States Environmental Protection Agency
(US EPA) to set national health-based standards for drinking water to protect against both naturally
occurring and man-made contaminants that may be found in drinking water. The National Primary
Drinking Water Regulations (NPDWR) are legally enforceable primary standards and treatment
techniques that apply to PWSs. Although states, tribes or territories that have been approved as the
primary implementation authority for drinking water may require monitoring or impose limits for
contaminants beyond those regulated under SDWA, there are not currently national requirements to
routinely monitor or limit 1,4-dioxane in finished water from PWSs. In support of the SDWA, EPA
often relies on data from the Unregulated Contaminant Monitoring Rule (UCMR) program as the best
available occurrence information to support its regulatory determinations (i.e., to judge whether a
particular contaminant is known to occur or there is substantial likelihood the contaminant will occur in
public water systems with a frequency and at levels of public health concern). UCMR monitoring is
designed to produce a data set that is nationally representative of public water systems (PWSs) across
the country, but its focus is on finished water (rather than source water) and it may not capture worst-
case conditions. PWS monitoring data of finished drinking water were collected for 1,4-dioxane via
EPA's published Third UCMR (UCMR3) dataset from 2013 to 2015, as well as raw and finished
drinking water monitoring from additional individual state databases (CA, MA, and NY) from 2008 to
2022 (CA. Water Board. 2022; NY DOR 2022; Commonwealth of Massachusetts. 201N; 1 ^ \
201 ?d). UCMR3 data were filtered to only include facilities flagged as using surface water, while the
individual state data were filtered down to only those systems with surface water listed as the primary
source in SDWIS. Datasets were processed to ensure that no samples were repeated in multiple datasets.
These PWS monitoring data were collected to assess possible exposures to the general population
through drinking water. Descriptions of the data retrieval and processing methods are presented in
Appendix G.2.
The combined datasets resulted in 16,972 samples from 2,847 PWSs across 50 states (Table 2-2).
Reported detection limits across the PWS datasets ranged from 0.0001 to 3 |ig/L, with 81 percent of
samples reporting a detection limit of 0.07 |ig/L. To the extent that it could be determined from the
database records, samples were separated into raw (untreated) water from the PWS intake or finished
(treated) water being sent to the distribution system. The distribution of raw water monitoring
concentrations is presented in Figure 2-9, and the distribution of finished drinking water concentrations
is presented in Figure 2-10.
Table 2-2. Summary of PWS Monitoring Datasets of 1,4-Dioxane Monitoring in PWSs Using
Surface Water as a Source
Dataset of
Origin
Number of
Samples
Minimum
Concentration
(Hg/L)
Median
Concentration
fag/L)
Maximum
Concentration
(Hg/L)
Start
Year
End
Year
CA
1,797
0.25
0.5
1.5
2011
2022
MA
949
0.049
0.22
3.8
2008
2022
NY
615
7.20E-05
0.035
1
2015
2022
UCMR3
13,611
0.035
0.035
13.3
2013
2016
Note: for the summary presented in this table, results reported as below their respective detection limit were
assigned a value of half of the detection limit.
Page 55 of 484
-------
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Detected 1,4-Dioxane Concentration in Raw Water (ug/L)
Figure 2-9. Frequency of 1,4-Dioxane Concentrations Monitored in Raw (Untreated)
Drinking Water Derived from Surface Water
Data retrieved from state databases (CA, MA, and NY) between 2008-2022.
Detected 1,4-Dioxane Concentration in Finished Drinking Water (ug/L)
Figure 2-10. Frequency of 1,4-Dioxane Concentrations Monitored in Finished (Treated)
Drinking Water Derived from Surface Water.
Data retrieved from the UCMR3 and state databases (CA, MA, and NY) between 2008-2022
Note: the detection limit for the method used in UCMR3, and the most common detection limit reported
in state databases is 0.07 (ig/L.
Water treatment systems may vary widely across the country based on available and utilized water
treatment processes that depend on whether source water is groundwater or surface water. These
processes typically include disinfection, coagulation/flocculation, sedimentation, and filtration (U.S.
EPA. 2006a). In assessing drinking water exposures, the ability to treat and remove or transform
chemicals in possible drinking water supplies should be considered. Typical treatment processes do not
remove 1,4-dioxane from ambient surface water and groundwater prior to possible general population
consumption as drinking water and treatment processes that do effectively remove 1,4-dioxane are
uncommon. EPA therefore assumes zero removal in the following analyses to provide a conservative
estimate of general population drinking water exposures. Even without treatment processes that remove
Page 56 of 484
-------
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
1,4-dioxane, multiple sources of water may be mixed within the same drinking water system which may
result in finished water with lower concentrations than one of the higher contributing source water
concentrations. EPA acknowledges that the surface water concentration at a single intake location may
be higher than the finished drinking water once mixed with other sources. Further discussion of the
prevalence of treatment processes across water systems, and the methodology for identifying raw and
finished drinking water monitoring samples is presented in Appendix G.1.2.
Figure 2-11 shows the spatial distribution of UCMR3 samples at the county level, with 1,4-dioxane
detected in 25 percent or 240 of 943 counties with participating water systems.
Maximum Reported 1,4-dioxane
Concentration
(M9/L)
¦¦ 6.21 - 13.30
2.49 - 6.20
1.21 - 2.48
0.62 - 1.20
0.07 - 0.61
I I < 0.07 (Not Detected)
I I No Data
125 250
500
750
1,000
I Miles
Figure 2-11. Map of Counties Containing PWSs that Reported Monitoring of Finished
Drinking Water Drawn from Surface Water for 1,4-Dioxane under UCMR3
Note: UCMR3 monitoring of 1,4-dioxane required four sampling events, one for each season, to capture
temporal variability. Each county highlighted may include one or multiple PWSs reporting data.
Monitored drinking water data were also included in exposure and risk estimates to assess the human
health implications of drinking water concentrations in this range. Since the UCMR program and state
monitoring datasets are not designed to reflect source water impacts of direct and indirect releases into
water bodies, EPA's TSCA program relied on estimated concentrations modeled for a range of specific
release scenarios to characterize risks from the water pathway. EPA evaluated the performance of the
models used to estimate water concentrations with monitoring data from site-specific locations serving
as cases studies. These case study comparisons demonstrated strong concordance between modeled
Page 57 of 484
-------
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
concentrations and monitoring data, thereby increasing confidence in risk estimates based on modeled
concentrations.
Although monitoring data confirm that 1,4-dioxane is present in drinking water in some locations,
samples collected under the UCMR program are designed to be nationally representative of drinking
water occurrence and not specifically associated with industrial releases of 1,4-dioxane. Since these
monitoring data may not reflect the 1,4-dioxane concentrations that result from industrial releases, EPA
relied on modeling to estimate 1,4-dioxane concentrations that occur near release sites.
2.3.1.2 Surface Water and Drinking Water Modeling
To assess possible general population exposures to 1,4-dioxane via industrial releases to surface water,
concentrations of 1,4-dioxane in surface water were modeled using two separate approaches. First, a
facility-specific approach aimed to quantify the maximum expected aqueous concentrations resulting
from reported 1,4-dioxane discharges from individual facilities in isolation. Second, a probabilistic
model was applied to assess the range of expected aqueous concentrations resulting from reported 1,4-
dioxane discharges across a COU, with consideration of expected ranges of background concentrations
of 1,4-dioxane from DTD loading and other unreported releases.
2.3.1.2.1 Modeling Methodology
A detailed description of modeling methods is presented in Appendix G.2.
As described in Section 2.2, annual releases of 1,4-dioxane to surface water from regulated dischargers
were retrieved from TRI and DMR. To the extent possible, modeled hydrologic flow data {i.e., stream
flow) associated with the receiving water body to which each facility released was retrieved from the
NHDPlus V2.1 dataset ( 2016). The receiving water body was identified either
through NPDES permit information for the releasing facility, or the nearest identified NFIDPlus V2.1
flowline. Detailed methods for the retrieval and processing of flow data are presented in Appendix
G.2.1.
Facility-Specific Modeling
Facility-specific modeling was conducted to estimate concentrations in receiving water bodies resulting
from the greatest facility-specific annual release reported between 2013 through 2019. This modeling
approach employed the equations used to model releases from facilities in the E-FAST 2014 model
(I E014) and is described in Appendix G.2.2. For each facility and annual release amount, three
different scenarios for days of release per year were considered: 1 day, 30 days, and expected number of
days of operation reported in Table Apx E-2 (referred to as the "maximum" number of days and ranges
from 250 to 365 days depending on OES). These additional scenarios with lower numbers of days of
operation provide more conservative estimates of resulting surface water concentrations and are
intended to evaluate the full range of possible facility release patterns based on the best available
information. Two flow metrics were evaluated: the lowest monthly average flow from NHDPlus, and the
harmonic mean flow derived from E-FAST 2014 methodology. The resulting concentrations from the
facility-specific modeling are used in calculations of general population exposure and human health
outcomes.
Probabilistic Modeling
The probabilistic modeling approach was conducted to consider multiple years of release data per
facility and multiple modeled flow metrics from NHDPlus V2.1 ( ) per facility
to generate a distribution of potential surface water concentrations resulting from releases across each
COU. The underlying model for the probabilistic approach is a fit-for-purpose model developed by EPA
Page 58 of 484
-------
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
in Microsoft Excel, the EWISRD-XL model (Estimating Water Industrial Surface Release and Down
the Drain in Excel). The EWISRD-XL model was designed to model 1,4-dioxane inputs to a stream
segment, including existing in-stream concentrations, DTD loading from consumer and commercial
products, and industrial releases, as a steady-state snapshot of a single point in time (Figure 2-12).
Outputs from the model include the resulting downstream concentration and the relative contribution
from each input source to that concentration. Examples of the EWISRD-XL model applied to three
specific case study locations (Brunswick County, NC, Columbia, TN, and Liverpool, OH) are included
in three Supplemental Information Files (U.S. EPA. 2023 o. p, g).
Inputs: -
POTW Loading:
Down-the-drain
from upstream
population
Indirect industrial
releases
Figure 2-12. Schematic of the EWISRD-XL Model Inputs and Outputs
For the probabilistic 1,4-dioxane COU modeling, an R script (R Core Team. 2022) was developed to
rapidly run multiple iterations of the EWISRD-XL model. In this configuration, called the EWISRD-
XL-R model, the underlying calculations were performed by EWISRD-XL model, and an R script
wrapper managed the processing of input and output data. For the probabilistic COU modeling, the
EWISRD-XL-R model developed to calculate the receiving water body concentrations at the point-of-
release by a facility. The EWISRD-XL-R results include the concentrations due only to releases from
facilities, as well as an estimated background concentration of DTD and unmonitored releases. The full
details of the underlying EWISRD-XL model and the probabilistic implementation are presented in
Appendix G.2.3. Distributions of total concentrations resulting from combinations of facility releases
and background concentrations were used for calculations of general population exposure and human
health outcomes.
A series of case studies was developed with the EWISRD-XL model to evaluate its performance across
various 1,4-dioxane release settings. These cases are presented in Appendix G.2.3.2.
Page 59 of 484
-------
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2.3.1.2.2 Estimating Down-the-Drain Releases
To evaluate the anticipated ranges of DTD contributions of 1,4-dioxane to water bodies receiving
POTW effluent, a range of combinations of hydrologic flows and populations served by a POTW were
evaluated using the EWISRD-XL-R model. For this modeling exercise, only contributions from the
DTD component were used to calculate resulting surface water contributions {i.e., no facility releases or
existing background concentrations were included). Hypothetical combinations of hydrologic flows and
populations contributing to wastewater loading derived from the national distribution of hydrologic
flows and populated places were selected to represent a range of results, which were then compared with
concentrations expected from industrial releases and used to calculate ranges of human exposure and
risk. More detailed methodology for this calculation is presented in Appendix G.2.3.4.
2.3.1.2.3 Hydraulic Fracturing
Hydraulic fracturing is a process used to extract oil and gas from shale plays. After hydraulic fracturing
operations inject fluids to extract oil and gas, a substantial volume of water may be produced through
flowback. The composition of these produced waters depends both on the geochemistry of the injected
area and the injected fluids. 1,4-Dioxane has been reported to EPA as one of the chemicals present in
these produced waters by 411 facilities via FracFocus 3.0 (GWPC and IOGCC. 2022). Estimated 1,4-
dioxane loadings of produced water to surface water from hydraulic fracturing activities (described in
Appendix E.9) were evaluated for expected ranges of resulting concentrations in receiving water bodies
using the EWISRD-XL-R model. Hydraulic fracturing wells reporting 1,4-dioxane use by FracFocus 3.0
were mapped, and flow data from nearby water bodies were collected from NHDPlus V2.1. A Monte
Carlo analysis was used to generate loadings to receiving water bodies from the distribution of modeled
releases and to pair them with hydrologic flows, resulting in a distribution of possible surface water
concentrations. Methodology for this analysis is presented in Appendix G.
2.3.1.2.4 Proximity to Drinking Water Sources
Drinking water exposures from facility-specific results assumed that the exposure occurs at the receiving
water body to provide a conservative estimation of drinking water exposures. However, the evaluated
water bodies may not be used as, or proximate to, actual drinking water sources and intakes. To give a
more robust characterization of possible drinking water exposures, known facility-specific releases were
mapped to drinking water sources using public water systems data stored in EPA's Safe Drinking Water
Information System Federal Data Warehouse (U.S. EPA. 2022f). This dataset is updated quarterly, and
the 2nd quarter 2022 version was used for this analysis. Following mapping, the colocation of and
proximity of releases to drinking water sources were evaluated. Locations of raw water intakes for
PWSs are considered sensitive by EPA Office of Water due to public safety concerns. Geospatial
analysis and the NHDPlus V2.1 flowline network were used to assess whether any known drinking
water intakes are located downstream of 1,4-dioxane releasing facilities. Methodology for this analysis
is presented in Appendix G.2.4.
2.3.1.3 Modeling Results
2.3.1.3.1 Facility-Specific Results
The facility-specific results show the expected concentration at the point of release from the facility
discharging 1,4-dioxane to receiving water bodies, without consideration of the contribution from other
sources. The total number of modeled releases within a given OES may be greater than the number of
1,4-dioxane releasing facilities in cases where facilities indirectly dispose of 1,4-dioxane by transferring
to another facility in addition to directly discharging 1,4-dioxane. Surface water concentrations resulting
from facility-specific modeling for one day of release are summarized in Table 2-3 and represent the
highest expected concentrations in receiving water bodies, due to the annual release amount being
Page 60 of 484
-------
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
discharged in a single day. Surface water concentrations resulting from facility-specific modeling for
maximum days of release are summarized in Table 2-4, and represent the lowest expected
concentrations in receiving water bodies due to the annual release spread out over the most days. The
single day release scenario allows consideration of a "worst-case scenario" given the available annual
release information and can inform an upper limit of concentrations resulting from releases. The
maximum days release scenario can inform a lower limit of expected concentrations from the available
annual release data. Full discussion on the evaluation of multiple release days is given in Section
2.3.1.2.1, but the range of evaluated release days is intended to provide to full range of expected surface
water concentrations resulting from possible facility release patterns and available information. As
described in Section 5, these variations in concentration due to days of release do not affect chronic
cancer risk estimates resulting from a particular releasing facility, due to annual averaging of exposure.
Resulting concentrations varied widely, both across and within OESs, due to variability in facility
release amounts as well as receiving water body flow magnitudes. Facility-specific releases are
organized around their identified OES as fully described in Section 2.1 and Appendix D.
Page 61 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2103 Table 2-3. Summary of Surface Water Concentration Results by OES from Facility-Specific Modeling of Annual Maximum Releases
2104 between 2013 and 2019 for 1 Operating Day per Year
OES
No. of
Releases
Modeled
Sum of
Annual
Releases
Modeled
(kg/year)
Annual Release by Facility
(kg/site-vear)
Surface Water Concentration
(Lowest Monthly Flow)
(jig/L)
Surface Water Concentration
(Harmonic Mean Flow)
(Hg/L)
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Disposal
25
16,997
1.36E-04
6.80E02
7.95E03
1.50E-02
6.45E05
9.52E06
1.50E-02
4.77E05
7.34E06
Ethoxylation
Byproduct
8
112,076
4.54E-01
1.40E04
1.12E05
5.39E-03
2.58E06
2.07E07
3.01E-03
1.22E06
9.73E06
Functional
Fluids (Open-
System)
6
17,711
3.80E-01
2.95E03
1.75E04
1.39E01
1.57E03
4.78E03
6.07E00
7.40E02
2.21E03
Import and
Repackaging
12
2,722
2.27E02
2.27E02
2.27E02
1.08E01
8.15E06
9.28E07
4.39E00
1.01E06
7.40E06
Industrial Uses
31
70,343
2.07E-01
2.27E03
2.62E04
1.33E-02
5.11E05
4.64E06
6.52E-03
4.53E05
5.15E06
Manufacture
2
7,034
1.67E03
3.52E03
5.36E03
8.31E04
1.63E06
3.18E06
8.31E04
1.63E06
3.18E06
PET
Manufacturing
19
2,773,355
3.40E-01
1.46E05
2.51E06
2.77E00
1.07E06
1.66E07
1.28E00
1.05E06
1.66E07
Printing Inks
1
5
5.45E00
5.45E00
5.45E00
2.05E03
2.05E03
2.05E03
2.05E03
2.05E03
2.05E03
Remediation
16
46
3.40E-05
2.91E00
2.39E01
1.50E-03
1.83E03
1.79E04
3.54E-04
1.52E03
1.37E04
Overall
120
3,000,290
3.40E-05
2.50E04
2.51E06
1.50E-03
1.45E06
9.28E07
3.54E-04
5.92E05
1.66E07
2105
Page 62 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2106 Table 2-4. Summary of Surface Water Concentration Results by OES for Facility-Specific Modeling of Annual Maximum Releases
2107 between 2013 and 2019 for the Maximum Operating Days per Year
OES
No. of
Releases
Modeled
Sum of
Annual
Releases
Modeled
(kg/year)
Annual Release by Facility
(kg/site-vear)
Surface Water Concentration
(Lowest Monthly Flow)
(lig/L)
Surface Water Concentration
(Harmonic Mean Flow)
(Hg/L)
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Disposal
25
16,997
1.36E-04
6.80E02
7.95E03
5.99E-05
2.57E03
3.81E04
5.99E-05
1.90E03
2.94E04
Ethoxylation
byproduct
8
112,076
4.54E-01
1.40E04
1.12E05
2.16E-05
1.03E04
8.26E04
1.20E-05
4.87E03
3.89E04
Functional
Fluids (Open-
System)
6
17,711
3.80E-01
2.95E03
1.75E04
5.63E-02
6.37E00
1.93E01
2.46E-02
3.00E00
8.95E00
Import and
Repackaging
12
2,722
2.27E02
2.27E02
2.27E02
4.32E-02
3.26E04
3.71E05
1.76E-02
4.04E03
2.96E04
Industrial Uses
31
70,343
2.07E-01
2.27E03
2.62E04
5.31E-05
2.04E03
1.86E04
2.61E-05
1.81E03
2.06E04
Manufacture
2
7,034
1.67E03
3.52E03
5.36E03
3.32E02
6.52E03
1.27E04
3.32E02
6.52E03
1.27E04
PET
Manufacturing
19
2,773,355
3.40E-01
1.46E05
2.51E06
1.11E-02
4.29E03
6.63E04
5.12E-03
4.20E03
6.63E04
Printing Inks
1
5
5.45E00
5.45E00
5.45E00
8.21E00
8.21E00
8.21E00
8.21E00
8.21E00
8.21E00
Remediation
16
46
3.40E-05
2.91E00
2.39E01
4.11E-06
5.01E00
4.90E01
9.69E-07
4.15E00
3.75E01
Overall
120
3,000,290
3.40E-05
2.50E04
2.51E06
4.11E-06
5.80E03
3.71E05
9.69E-07
2.37E03
6.63E04
2108
Page 63 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2109 Table 2-5 describes the crosswalk between identified OESs and relevant COUs under each for the
2110 identified facility releases to surface water. The full facility-specific analysis is included in 1,4-Dioxane
2111 Supplemental Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Release
2112 to Surface Water from Individual Facilities (U.S. EPA. 2023h).
2113
Table 2-5. OES-COU Crosswalk for Identified Facilities Releasing to Sur
'ace Water"
COU*
OES'
Life Cycle Stage
Category
Subcategory'
Manufacturing
Domestic manufacture
Domestic manufacture
Manufacturing
Import
Import
Repackaging
Import and Repackaging
Processing as a reactant
Polymerization catalyst
Industrial Uses
Non-incorporative
Basic organic chemical manufacturing
(process solvent)
Processing
Byproduct
Byproduct produced during the
ethoxylation process to make
ethoxylated ingredients for personal
care products
Ethoxylation Process
Byproduct
Byproduct produced during the
production of polyethlene terephtalate
PET Byproduct
Intermediate use
Plasticizer intermediate
Catalysts and reagents for anhydrous
acid reactions, brominations, and
sulfonations
Industrial Uses
Industrial Usea
Processing aids, not
otherwise listed
Wood pulping
Extraction of animal and vegetable
oils
Wetting and dispersing agent in
textile processing
Polymerization catalyst
Purification of process intermediates
Etching of fluoropolymers
Industrial Uses
Functional fluids (open
and closed systems)
Polyalkylene glycol lubricant
Synthetic metalworking fluid
Cutting and tapping fluid
Functional Fluids (Open
System)
Industrial Use,
Commercial Use
Other Uses
Spray polyurethane foam
Printing and printing compositions,
including 3D printing
Dry film lubricant
Hydraulic fracturing
Printing Inks (3D)
Disposal
Disposal
Remediation
Remediation
Disposal
Disposal
Industrial pre-treatment
Industrial wastewater treatment
Publicly owned treatment works
(POTW)
Underground injection
Municipal landfill
Disposal
Page 64 of 484
-------
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
COU''
OESrf
Life Cycle Stage
Category
Subcategory'
Hazardous landfill
Other land disposal
Municipal waste incinerator
Hazardous waste incinerator
Off-site waste transfer
" 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.
b As mapped to COU Life Cycle Stage, Category, and Subcategory in Table Apx D-l.
c Evaluated facilities within an OES may not encompass all listed COU subcategories.
d Note that identified OESs can encompass multiple COUs across different life cycle stages and categories.
To put the modeled releases in the context of the underlying data sources for release amounts and
receiving water body flow, Table 2-6 presents the results of the process of assigning the receiving water
body (by reach code in the NHDPlus 2.1 dataset) to each releasing facility. Those facilities with reach
code information in their NPDES permit were regarded as the highest confidence in an accurate match
to the actual discharging water body, followed by facilities matched geospatially to the nearest reach
code within 1 km of the facility. Facilities matched to reaches beyond 1 km from the facility but within 2
km provided lower confidence, and those without reach code matches were substituted with the lowest
non-zero flow within the OES as a conservative estimate. The full details of the flow matching process
are presented in Appendix G.2.1. Due to the assumptions described in Section 2.2.1.3 required to model
releases from facilities reporting only via TRI Form A, the percent of facilities within an OES using
Form A is also reported.
Table 2-6. Summary by OES of Data Sources for Releases and Receiving Water Body Flow
Method of Matching to Receiving Water Body
OES
Total
Number of
Releases
NPDES
Permit
Contains
Reach Code
Nearest
Reach
(within 1
km)
Nearest
Reach
within 2 km
Lowest Non-
zero Flow
within OES
Substituted
%of
Releases
Estimated
from TRI
Form A
Disposal
25
22
0
1
1
8
Ethoxylation
Byproduct
8
1
1
1
0
0
Functional Fluids
(Open-System)
6
5
0
0
0
0
Import and
Repackaging
12
1
2
2
7
100
Industrial Uses
31
11
3
5
7
45
Manufacture
2
1
0
0
0
0
PET Manufacturing
19
11
0
1
0
0
Printing Inks
1
1
0
0
0
0
Remediation
16
14
2
0
0
0
Total
120
67
8
10
15
23
A generic table of annual facility release and average flow rates for the receiving water body is
presented in Table 2-7, which demonstrates the relationship between the facility and water body
Page 65 of 484
-------
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
characteristics with regard to the resulting surface water concentrations. Table 2-8 shows the relative
occurrence of each of the releases modeled for this assessment within binned ranges of releases and
flows. Combined, these tables demonstrate that most facilities releasing 1,4-dioxane are initially
discharging to smaller water bodies, even in some cases where large annual release amounts result in
very high modeled concentrations.
Table 2-7. Hypothetical Mean Annual Concentrations (jig/L) for a Range of Annual Release and
Flow Rate Combinations, for a Facility with 250 Days of Release per Year
Annual Release Amount (kg)
1
10
100
1,000
10,000
100,000
1,000,000
Mean
Annual
Receiving
Water Body
Flow (cfs)
1
1.6
16
160
1.6E03
1.6E04
1.6E05
1.6E06
10
0.16
1.6
16
160
1.6E03
1.6E04
1.6E05
100
0.016
0.16
1.6
16
160
1.6E03
1.6E04
1,000
1.6E-03
0.016
0.16
1.6
16
160
1.6E03
10,000
1.6E-04
1.6E-03
0.016
0.16
1.6
16
160
100,000
1.6E-05
1.6E-04
1.6E-03
0.016
0.16
1.6
16
Table 2-8. Occurrence of Facilities for Distributions of Maximum Annual 1,4-Dioxane Release
Amounts and Receiving Water Body Flow
Annual Release Amount (kg)
<10
10 to 100
100 to
1,000
1,000 to
10,000
10,000 to
100,000
>100,000
Mean Annual
Receiving Water
Body Flow (cfs)
<10
14%
8%
11%
6%
2%
<1%
10 to 100
9%
7%
2%
<1%
1%
3%
100 to 1,000
3%
6%
2%
1%
<1%
<1%
1,000 to 10,000
1%
2%
2%
2%
2%
<1%
10,000 to 100,000
3%
1%
3%
4%
2%
<1%
2.3.1.3.2 Concentrations from Down-the-Drain Loading
Water concentrations of 1,4-dioxane resulting from DTD releases depend on the population size (an
indicator of the number of people using products and contributing to the releases) and the stream flows
of the receiving water bodies. The representative per capita DTD loading developed from modeling
results from SHEDS-HT was applied to a range of population sizes (100 to 1,000,000 people) and
stream flows (300 to 30,000 cfs) to develop a distribution of potential surface water concentrations.
Estimated surface water at the point of discharge by POTWs resulting from DTD releases ranged from
less than 0.0001 to 110 |ig/L (Table 2-9). The typical ranges of results from this analysis are comparable
to the range of minimum to mean concentrations calculated from individual facility releases in Section
2.3.1.3.1.
Page 66 of 484
-------
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 2-9. Estimated Surface Water Concentrations (jig/L) Due to DTD Loading
for a Range of Populations and Hydrologic Flows
Population Contributing to DTD Loading
100
1,000
10,000
100,000
1,000,000
Receiving
Water Body
Flow (cfs)
100
0.011
0.11
1.1
11
110
300
3.6E-03
0.036
0.36
3.6
36
1,000
1.1E-03
0.011
0.11
1.1
11
3,000
3.6E-04
3.6E-03
0.036
0.36
3.6
10,000
1.1E-04
1.1E-03
0.011
0.11
1.1
The occurrence of POTWs processing wastewater from various populations and the associated flows of
the receiving water bodies were investigated using data from the ICIS-NPDES database (U.S. EPA.
2013). to inform the interpretation of the above ranges of DTD loading concentrations. For communities
with a single POTW treating wastewater, most fell into the range of 100 to 10,000 people, with the
annual average flow of the receiving water body less than 300 cfs (Table 2-10).
Table 2-10. Estimated Percent Occurrence of Combinations of Contributing Population to
Population Contributing to DTD Loading
<100
100 to 1,000
1,000 to
10,000
10,000 to
100,000
100,000 to
1,000,000
Mean Annual
Receiving
Water Body
Flow (cfs)
<100
5%
44%
26%
4%
<1%
100 to 300
<1%
3%
4%
1%
<1%
300 to 1,000
<1%
2%
2%
1%
<1%
1,000 to 3,000
<1%
1%
2%
<1%
<1%
3,000 to 10,000
<1%
<1%
1%
<1%
<1%
>10,000
<1%
1%
1%
1%
<1%
2.3.1.3.3 Concentrations from Hydraulic Fracturing
The Monte Carlo distribution of potential surface water concentrations resulting from hydraulic
fracturing operations is presented in Table 2-11. Hydrologic flows in water bodies near hydraulic
fracturing wells reporting 1,4-dioxane as a constituent of wastewater ranged from less than 10 to 44,300
cfs. Due to the very low flows in many nearby streams, resulting concentrations were sensitive to the
receiving water body flow rate. The distribution of loading to surface water from hydraulic fracturing
represents the loading from a single site of hydraulic fracturing operations (described in Appendix
G.2.3.5) at the immediate point of discharge to the receiving water body. Concentrations estimated at the
highest end are comparable to mean to high-end facility releases presented in Section 2.3.1.3.1. More
than half of the modeled concentrations fell below the typical detection limit in drinking water of 0.07
Hg/L-
Page 67 of 484
-------
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 2-11. Distribution of Potential Concentrations in Surface
Water Resulting from Hydraulic Fracturing Operations from a
Single Site Reporting 1,4-Dioxane as an Ingredient
Monte Carlo Distribution
Concentration (jig/L)
Maximum
198
99th Percentile
7.79
95th Percentile
2.60
Median
0.064
5th Percentile
2.31E-04
Minimum
6.08E-12
2.3.1.3.4 Aggregate Probabilistic Results
The aggregate probabilistic model predicts surface water concentrations at the point of facility releases
when incorporating potential contributions from DTD and other unmonitored sources. The model
incorporates multiple years of release data and was run with 10,000 iterations for each OES using different
combinations of direct and indirect facility releases, DTD releases, flows, and background concentrations.
This results in a more descriptive distribution of the potential releases. At the highest end, the results of
the aggregate probabilistic model are similar to those from the facility-specific modeling. This is due to
both the facility-specific modeling and the highest end of the probabilistic modeling being based on the
maximum reported releases from the modeled facilities. Additionally, the loading from facilities far
outweighs the contribution from background sources at the higher end. The shape of the resulting
distribution can be informative in its representation of the frequency of concentrations exceeding a certain
threshold.
Resulting surface water concentrations ranged from 1,45 x ] 0 4 to 7.34><103 |ig/L. Summaries of the
resulting concentrations by OES are presented in Table 2-12 and Figure 2-13. Overall, releases from
facilities tended to result in greater 1,4-dioxane concentrations in surface water than the expected ranges
of background concentrations. Background concentrations were derived from values of 1,4-dioxane
measured by drinking water systems using surface water as a source that were not downstream of known
1,4-dioxane releases (Figure 2-11). The "% of Releases Greater than Background" column in Table 2-12
refers to the frequency of model runs (out of the 10,000 per OES) in which the resulting concentration
from the facility release was greater than the generated background concentration resulting from DTD
and other unregulated surface water loading. A low percentage for this metric may suggest that releases
by a particular OES are typically outweighed by these other unreported releases with respect to their
contribution to surface water concentrations.
Page 68 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2203 Table 2-12. Aggregate Probabilistic Results Showing Distribution of Total 1,4-Dioxane Concentration in Surface Water (Release
2204 Plus Background)
OES
Min
(^g/L)
5th
Percentile
(jig/L)
25th
Percentile
(jLLg/L)
Median
(jig/L)
75th
Percentile
(jig/L)
95th
Percentile
(^g/L)
Max
(^g/L)
% of Releases
Greater than
Background
Disposal
1.88E-03
1.51E-01
1.98E-01
3.50E-01
8.52E-01
1.94E00
2.02E01
81
Ethoxylation Process Byproduct
2.25E-03
9.81E-02
1.35E-01
2.74E-01
4.65E-01
2.55E00
1.46E01
72
Functional Fluids (Open-System)
1.78E-04
8.20E-02
1.11 E—01
1.58E-01
2.79E-01
1.60E00
6.10E00
48
Import and Repackaging
5.70E-03
1.32E-01
2.83E-01
6.60E01
3.25E02
1.42E03
2.12E03
90
Industrial Uses
1.45E-04
5.15E-02
8.95E-02
1.22E-01
2.52E-01
1.33E01
2.26E02
44
Manufacture
1.10E02
3.32E02
3.32E02
7.19E02
2.32E03
5.48E03
7.34E03
100
PET Byproduct
5.44E-03
1.42E-01
2.13E-01
4.65E-01
4.36E00
1.30E02
2.23E03
84
Printing Inks (3D)
1.51E-01
3.17E-01
1.10E00
4.00E00
6.41E00
7.48E00
8.26E00
96
Remediation
5.34E-04
5.89E-02
9.04E-02
1.35E-01
2.73E-01
6.66E00
1.46E01
47
Page 69 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2205
2206
2207
50
40
30
20
10-
50
40-
(/>
1 30
0
a:
•5 20
2?
10
0
50-
40-
30
20
10
Disposal
TO
T>
-------
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2.3.1.4 Comparison of Modeled and Monitored Surface Water Concentrations
At the higher end, the modeled concentrations from facility releases are several orders of magnitude
greater than those observed in the 1,4-dioxane monitoring data (Figure 2-6 and Table 2-3). This
difference in concentrations may be due to monitoring data being collected further downstream
(allowing for additional dilution), or on reaches that are not impacted by releasing facilities. Many of the
direct releasing facilities, and POTWs assessed for the DTD component, had a receiving water body
specified on their NPDES permits that was associated with a very small stream or industrial canal. These
small receiving water bodies, combined with larger loading values from the releases, resulted in high
modeled concentrations in surface water at the point of release. As this water travels downstream, it is
expected to eventually join with larger water bodies, where some decrease in concentration due to
dilution would occur.
Because most of the reasonably available monitoring data were generally not co-located with 1,4-
dioxane release sites, EPA relied primarily on modeling to estimate water concentrations that could
result from releases. Where co-located monitoring data were available, EPA compared modeled
concentrations to reasonably available monitoring data in the limited set of specific locations to evaluate
the performance of the model. Comparisons of modeled vs. monitoring water concentrations for this
limited set of "case study" locations demonstrate that modeled mean concentrations are generally
consistent with mean concentrations reported in monitoring data. For example, the Cape Fear River
upstream of the Brunswick County, NC drinking water intake was selected as a case study to test the
model due to abundant monitoring data in the region. Water concentrations modeled based on upstream
releases from an industrial facility in Fayetteville in combination with other upstream sources. As
illustrated in Figure 2-14, modeled surface water concentrations generally fell within the ranges reported
from monitored concentrations. Wide ranges of both monitored and modeled values were noted,
indicating variability among inputs to the system. Details of the case study comparisons for Brunswick
County and other locations are described in Appendix G.2.3.2. The concordance of monitoring and
modeled concentrations increases confidence in the model used to estimate water concentrations from
DTD releases and hydraulic fracturing, and to perform probabilistic modeling of aggregate
concentrations from multiple sources.
Page 71 of 484
-------
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Date
Figure 2-14. Case Study Comparison of Modeled and Monitored Concentrations in
Brunswick County
2.3.1.5 Strengths, Limitations, and Sources of Uncertainty in Assessment Results for
Monitored and Modeled Drinking Water and Surface Water Concentrations
The evaluation of general population drinking water exposure scenarios are impacted by uncertainties
and assumptions surrounding inputs and the approaches used for modeling surface water concentrations
and estimation of the drinking water doses. In Appendix E.8, EPA assesses the overall confidence of
estimated releases for various OESs. For those OESs releasing to surface water, confidence is rated as
medium to high depending on an individual OES.
The modeling used and the associated default and user-selected inputs have the ability to affect overall
strength in evaluated general population exposures. The facility-specific releases methodology described
in Section 2.3.1.2.1, and the results in 2.3.1.3.1, rely on a modeling framework that does not consider
downstream fate or transport. However, the physical-chemical properties of 1,4-dioxane are expected to
moderate this limitation due to its likelihood to stay in the water column, and due to the lack of removal
during typical drinking water treatment process. To reduce uncertainties, EPA incorporated an updated
flow network and flow data into this assessment that allowed a more site-specific consideration of
release location and associated receiving water body flows. These facility-specific releases are also
evaluated on a per facility basis that does not account for additional sources of 1,4-dioxane that may be
present in the evaluated waterways. To help address these limitations in this risk evaluation, EPA
conducted additional aggregate and probabilistic approaches, evaluated in Section 2.3.1.2.1 and Section
2.3.1.3.4, that give a more complete overall estimation of possible 1,4-dioxane concentrations. Finally,
drinking water exposures from facility-specific results assume that the exposure occurs at the receiving
water body. The water bodies evaluated may or may not be used as drinking water sources. To address
this limitation, EPA evaluated the proximity of known 1,4-dioxane releases to known drinking water
sources as well as known drinking water intakes as described in Section 2.3.1.2.4.
To evaluate the accuracy of the aggregate model, case studies described in Appendix G.2.3.2 compared
modeled results to observed monitored concentrations. The three evaluated case studies give good
general agreement between available monitoring with modeled values. Overall, this gives strength to the
modeling assumptions, inputs and output calculations for areas that are lacking robust monitoring data.
The model is able to effectively capture the general influences of both DTD loading, facility loading and
Page 72 of 484
-------
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
upstream contributions to create an aggregation of possible ambient surface water concentrations of 1,4-
dioxane. The monitored data encompassed both ambient surface water monitoring as well as drinking
water system monitoring data. For the ambient surface water data, data is limited geographically and
temporally with many states having no reported data and even those areas reporting measured values
having limited samples over time. Monitored concentrations in close proximity to modeled releases were
rare, often making direct comparisons of modeled results unavailable. In most cases, monitoring data
represented water bodies without identified releases of 1,4-dioxane nearby.
The hydraulic fracturing analysis relies on a Monte Carlo distribution of loading values with some level
of uncertainty and is itself a Monte Carlo simulation with potential receiving water body flows. The
precision of such an analysis is lower at the most extreme (minimum and maximum) values.
2.3.2 Land Pathway (Groundwater)
Any activities where chemicals or wastes might be released to the environment has the potential to
pollute groundwater. To understand possible exposure scenarios from these practices, EPA assessed
drinking water exposure resulting from use of 1,4-dioxane contaminated groundwater due to chemical
injection to Underground Class I Wells, leaching from landfills where 1,4-dioxane or products
containing 1,4-dioxane have been disposed, and disposal of hydraulic fracturing produced water to
surface impoundments. Sections 2.3.2.1 through 2.3.2.4 provide a description and an assessment of each
disposal practice. Figure 2-15 and Figure 2-16 provide a visual summary of groundwater monitoring
data available through the WQP fNWOMC. 20221
2.3.2.1 Groundwater Monitoring Data
Measured, field-collected, data from environmental samples representing groundwater 1,4-dioxane
concentrations across the country were collected as direct groundwater monitoring results. These results
are collated by the National Water Quality Monitoring Council and stored in the WQP fNWOMC.
2022). Some monitoring results reported to the WQP included locations expected to be directly
impacted by 1,4-dioxane releases. Data were available from 1997 to 2022, resulting in 8,110 available
sample results. The distribution and detection percentages are presented in Figure 2-15 and mapped in
Figure 2-16. The process for identifying this data is provided in Appendix H.l. This analysis is intended
to characterize the observed ranges of 1,4-dioxane concentrations in groundwater, irrespective of the
reasons for sample collection, and to provide context for the modeled groundwater concentrations
presented in Sections 2.3.2.1 through 2.3.2.4. In order to better understand where highest groundwater
concentrations are occuring, EPA arbitrarily portioned the data based on order of magnitude differences
to best describe where and when data differences could be observed.
Page 73 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Detected 1,4-Dioxane Concentration (|jg/L)
2306 Figure 2-15. Frequency of Nationwide Detected 1,4-Dioxane Groundwater
2307 Concentrations (n = 2,284) Retrieved from the Water Quality Portal, 1997-2022
2308
2309
2310 Figure 2-16. Detectable Concentrations of 1,4-Dioxane in Groundwater from the Water
2311 Quality Portal, 1997-2022
2312 Note: Alaska, American Samoa, Guam, Hawaii, N. Mariana Islands, Puerto Rico, and the U.S. Virgin Islands
2313 are not shown as there are no known monitoring data above detection limits.
2314
Page 74 of 484
-------
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Figure 2-15 shows the range of detected concentrations of 1,4-dioxane in groundwater from 1997 to
2022. During this period the detection limits ranged from 0.028 |ig/L to 320 |ig/L. The maximum
detected concentration (31,000 |ig/L) occurred in Westville, IN, in 1997 at a former waste-oil refinery.
This site and many others identified in this monitoring data have ongoing remediation projects to
address these contamination plumes.
Recent changes in industrial activities and disposal may have largely reduced groundwater
contamination with 1,4-dioxane. As shown in Figure 2-17, samples collected prior to 2000 tended to be
substantially higher in concentration relative to those collected after 2003. This finding may be an
artifact of historical uses and industrial practices related to 1,4-dioxane. Although several samples are
still above 10 |ig/L, particularly in 2007, the bulk of data tend to fall between 1 and 10 |ig/L. Without a
thorough investigation of what practices have changed in industry, it is difficult to attribute this decline
to a single event but indicates continued work to prevent groundwater contamination.
Figure 2-17. Groundwater Concentrations of 1,4-Dioxane vs. Sample
Collection Date for Data Collected between 1997 and 2022
Figure 2-16 shows the spatial distribution of detected 1,4-dioxane concentration across the contiguous
states. This map shows nine locations with concentrations of 1,4-dioxane greater than 10 |ig/L. These
tend to be attributed to past industrial activities causing extensive groundwater contamination. In
addition to this monitoring data, groundwater contamination from disposing 1,4-dioxane to landfills has
been documented in Alaska (Li et al.. 2013). California (Li et al.. 2015; Adamson et al.. 20141
Michigan (Mohr and DiGuiseppi. 20101 New York (Lee et al.. 2020). and recently in Ohio
(https://cumulis.epa.gov/supercpad/cursites/csitinfo.cfm?id=0504014). EPA was not able to identify
reasonably available information specific to groundwater concentrations near or around underground
injection sites, landfills, or surface impoundments that received hydraulic fracturing produced water.
2.3.2.2 Disposal via Underground Injection
Underground injection is a method of disposal for hazardous wastes.9 There are generally six different
classes of underground wells, and only Class I Wells may be permitted to receive hazardous waste.
Oversight of these wells requires that they are designed and constructed to prevent the movement of
9 Additional information about underground injection can be found at https://www.epa.gov/uic.
Page 75 of 484
-------
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
injected waste streams into drinking water systems. Wells typically consist of three or more concentric
layers of pipe including surface casing, long string casing, and injection tubing. In addition, wells must
be sited at locations with geologies that mitigate any movement of contaminants outside of a confined
layer in case of a well failure. Extensive pre-siting geological tests confirm that the injection zone is of
sufficient lateral extent and thickness and is sufficiently porous so that fluids injected through the well
can enter the rock formation without extensive buildup of pressure or possible displacement of injected
fluids outside of the intended zone.
Potential pathways through which injected fluids can migrate to underground sources of drinking water
include failure of the well or improperly plugged or completed wells near the well. Well failures can be
detected by continuous monitoring systems or mechanical integrity tests, at which point the wells would
be shut-in until they are repaired. EPA's extensive technical requirements for Class I wells (40 CFR
148) are designed to prevent contamination of underground sources of drinking water through these
pathways. Operators must conduct appropriate mechanical integrity tests yearly for hazardous wells and
every 5 years for nonhazardous wells to ensure wells are fit for operation. Note that the loss or failure of
mechanical integrity does not necessarily mean that wastewater will escape the injection zone. This
added security can be attributed to redundant safety systems to protect against loss of waste
confinement.
2.3.2.2.1 Summary of Assessment for Disposal to Underground Injection
According to EPA's TRI database, there are two locations where 1,4-dioxane has been disposed of via
underground injection to Class I Wells. On-site disposals to Class I underground injection wells are
provided in TableApx H-l. On-site Class I underground injection wells may be owned and operated by
the producer of the waste. Off-site disposals to Class I underground injection wells are provided in
Table Apx H-2. Offsite Class I underground injection wells may be secondary entities that own and
operate the well. Both on-site and offsite underground injection wells must be permitted and regularly
inspected. Careful review of the permits and state databases corroborates that both sites are permitted
and compliant. These sites have implemented groundwater migration controls and the Enforcement and
Compliance History Online (ECHO) database ( !022e) indicates the site is currently in
compliance.
In addition to reviewing these permits, EPA reviewed reasonably available groundwater monitoring data
available via state databases as well as via the WQP (see Figure 2-15 and Figure 2-16) and found no
evidence of groundwater contamination near the facilities. Because underground injection is not
expected to result in groundwater contamination based on the reasonably available information, EPA did
not quantitatively estimate groundwater concentrations, exposures, or risks from underground injection.
2.3.2.2.2 Strengths, Limitations, and Sources of Uncertainty in Assessment of
Disposal to Underground Injection Wells
Because EPA did not quantitatively evaluate the potential exposure from disposing 1,4-dioxane via
underground injection, the major source of uncertainty is limited to the accuracy of state databases
providing monitoring data surrounding these wells. EPA believes these databases are reporting
accurately where contaminations are known, but only explored states where the TRI database indicated
there were disposals via underground injection. Disposals below the reporting requirement for TRI may
not be captured.
2.3.2.3 Disposal to Landfills
Landfills may have various levels of engineering controls to prevent groundwater contamination. These
can include industrial liners, leachate capturing systems, and routine integration of waste. However,
Page 76 of 484
-------
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
groundwater contamination from disposal of consumer, commercial, and industrial waste streams
continues to be a prominent issue for many landfills throughout the United States (Li et al.. JO I < t
at.. 2013; Mohr and DiGuiseppi. 2010). These contaminations may be attributed to perforations in the
liners, failure of the leachate capturing system, or improper management of the landfills. 1,4-Dioxane
persists in groundwater and can migrate away from landfills into nearby communities at the same rate as
hydraulic flow (Mohr and DiGuiseppi. 2010). When these communities rely on groundwater as their
primary drinking water source, there is a potential for exposure via oral ingestion if that water is
contaminated with 1,4-dioxane and does not undergo treatment. Depending on the distance between the
landfill and a drinking water well, as well as the potential rate of release of landfill leachate into
groundwater, the concentration of this exposure can vary substantially.
Landfills are regulated under the Resource Conservation and Recovery Act (RCRA) or TSCA. RCRA
landfills can be classified as Subtitle C (hazardous waste landfills) or Subtitle D (municipal solid
nonhazardous waste landfills). Subtitle C establishes a federal program to manage hazardous wastes
from cradle to grave. The objective of the Subtitle C program is to ensure that hazardous waste is
handled in a manner that protects human health and the environment. When waste generators produce
greater than 100 kg per month of non-acutely hazardous waste, those hazardous wastes, including 1,4-
dioxane, meeting the U108 waste code description in 40 CFR 261.33, must be treated to meet the land
disposal restriction levels in 40 CFR part 268 and be disposed in RCRA subtitle C landfills. These
disposals are captured partially through the Toxics Release Inventory and are reported for onsite
facilities (TableApx H-3) and offsite facilities (TableApx H-4). Recent violations of permits are
reported in the footnotes of each table.
Review of state databases does not suggest any readily available evidence of groundwater contamination
near or coinciding with these operations that could affect a drinking water supply. Similar review of the
data available via the WQP suggests that there are no known contaminations from RCRA Subtitle C
Landfills as reported to the TRI program (see Figure 2 14 and Figure 2 15). The absence of groundwater
contamination near RCRA Subtitle C Landfills may be attributed to many of the ongoing engineering
controls built into these facilities as well as active monitoring of groundwater wells around facilities. As
a result, EPA did not assess Subtitle C landfills further than understanding their permit violations.
Regulations established under Subtitle D ban open dumping of waste and set minimum federal criteria
for the operation of municipal waste and industrial waste landfills, including design criteria, location
restrictions, financial assurance, corrective action (clean up), and closure requirements. States play a
lead role in implementing these regulations and may set more stringent requirements. National
requirements for Subtitle D landfills are most specific for MSW landfills. MSW landfills built after 1990
must be constructed with composite liner systems and leachate collection systems in place. Composite
landfill liners consist of a minimum of two feet of compacted soil covered by a flexible membrane liner,
which work in concert to create a low hydraulic conductivity barrier and prevent leachate from being
released from the landfill and infiltrating to groundwater. A leachate collection system typically consists
of a layer of higher conductivity material above the composite liner that funnels leachate to centralized
collection points where it is removed from the landfill for treatment and disposal. Despite these controls,
releases may still occur due to imperfections introduced during construction or that form over time (Li et
al., 2015; Li et al., 2013; Mohr and DiGuiseppi, 2010); thus, groundwater monitoring is required to
identify and address any releases before there can be harm to human health and the environment. RCRA
Subtitle D requirements for non-MSW landfills are less stringent. In particular, nonhazardous industrial
landfills and C&D debris landfills do not have specified national requirements for construction and
operation and certain landfills are entirely exempt from RCRA criteria. Under the Land Disposal
Program Flexibility Act of 1996 (Pub.L. 104-119), some villages in Alaska that dispose of less than 20
Page 77 of 484
-------
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
tons of municipal solid waste daily (based on an annual average) may dispose of waste in unlined or
clay-lined landfills or waste piles for open burning or incineration.
There are a several potential sources of 1,4-dioxane to Subtitle D landfills. Waste generators that
produce less than 100 kg per month of non-acutely hazardous waste, including 1,4-dioxane meeting the
U108 waste code, may dispose of this waste in these landfills. Nonhazardous industrial wastes also have
the potential to contain 1,4-dioxane at variable concentrations. Consumer and commercial products may
also contain 1,4-dioxane in relatively low amounts. The greatest potential for release of disposed 1,4-
dioxane to groundwater is from landfills that do not have an adequate liner system. Thus, an objective of
this assessment is to evaluate the potential for groundwater contamination in the absence of landfill
controls.
This assessment was completed using the Hazardous Waste Delisting Risk Assessment Software
(DRAS). DRAS was specifically designed to address the Criteria for Listing Hazardous Waste identified
in Title 40 Code of Federal Regulations (40 CFR) Section 261.11(a)(3), a requirement for evaluating
proposed hazardous waste delistings. In this assessment, DRAS is being utilized to determine potential
groundwater concentrations of 1,4-dioxane after they have been disposed of into a non-hazardous waste
landfill. The results of this assessment are found in Table 2-13. This assessment relied on the default
waste loading rates for RCRA Subtitle C Landfills available in DRAS. Similarly, the assessment relied
on the default values for 1,4-dioxane as the chemical of concern. Lastly, leachate concentrations were
estimated for a range of possibilities until no risk could be identified at the lower end of those
concentrations. Because DRAS calculates a weight adjusted dilution attenuation factor (DAF) rather
than a groundwater concentration, a back of the envelop computation was used to convert the DAF to a
potential concentration that people living within 1 mile of a landfill might be exposed if the release were
not identified and remediated.
Page 78 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2466 Table 2-13. Potential Groundwater Concentrations (jig/L) of 1,4-Dioxane Found in Wells within 1 Mile of a Disposal Facility
2467 Determined by Using the DRAS Model
Leachate Concentration
(jig/L)
Loading Rate (kg)
4.55 E-04
4.55E-03
4.55E-02
4.55E-01
4.55E00
4.55 E01
4.55E02
4.55E03
4.55E04
4.55E05
1.00E-07
7.81E-13
7.46E-12
5.46E-11
5.21E-10
6.49E-09
6.17E-08
5.88E-07
5.62E-06
5.38E-05
5.13E-04
1.00E-06
7.81E-12
7.46E-11
5.46E-10
5.21E-09
6.49E-08
6.17E-07
5.88E-06
5.62E-05
5.38E-04
5.13E-03
1.00E-05
7.81E-11
7.46E-10
5.46E-09
5.21E-08
6.49E-07
6.17E-06
5.88E-05
5.62E-04
5.38E-03
5.13E-02
1.00E-04
7.81E-10
7.46E-09
5.46E-08
5.21E-07
6.49E-06
6.17E-05
5.88E-04
5.62E-03
5.38E-02
5.13E-01
1.00E-03
7.81E-09
7.46E-08
5.46E-07
5.21E-06
6.49E-05
6.17E-04
5.88E-03
5.62E-02
5.38E-01
5.13E00
1.00E-02
7.81E-08
7.46E-07
5.46E-06
5.21E-05
6.49E-04
6.17E-03
5.88E-02
5.62E-01
5.38E00
5.13E01
1.00E-01
7.81E-07
7.46E-06
5.46E-05
5.21E-04
6.49E-03
6.17E-02
5.88E-01
5.62E00
5.38E01
5.13E02
1.00E00
7.81E-06
7.46E-05
5.46E-04
5.21E-03
6.49E-02
6.17E-01
5.88E00
5.62E01
5.38E02
5.13E03
1.00E01
7.81E-05
7.46E-04
5.46E-03
5.21E-02
6.49E-01
6.17E00
5.88E01
5.62E02
5.38E03
5.13E04
Concentrations organized by potential loading rates (kg) and potential leachate concentrations (ng /L).
2468
Page 79 of 484
-------
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2.3.2.3.1 Summary of Assessment for Disposal to Landfills
EPA determined through modeling that groundwater concentration of 1,4-dioxane increased with
increasing landfill load rate and increasing leachate concentration. With each progressive iteration of
loading rate or leachate concentration, potential groundwater concentrations increase by an order of
magnitude. When both loading rate and leachate increase by one order of magnitude, potential
groundwater concentration increase by two orders of magnitude. These increases can largely be
attributed to the increasing weight adjusted dilution attenuation factor and are what would be expected
for a chemical substances with 1,4-dioxane's physical-chemical properties (water solubility, Henry's
law constant) and fate characteristics (biodegradability, half-life in groundwater). 1,4-Dioxane migrates
in groundwater at the rate of hydraulic flow and can persist for greater than 30 days in anaerobic
environments (Adamson et al.. 2014; Mohr and DiGuiseppi. 2010) as described in the 2020 RE. Thus,
these concentrations are likely to represent the range of potential groundwater concentrations for
sentinel PESS living within a 1-mile radius of a RCRA Subtitle D landfills and other non-Subtitle C
landfills.
EPA also determined that the modeled concentrations are within the range of concentrations of 1,4-
dioxane found in groundwater monitoring studies. A survey of monitoring studies in California has
demonstrated that 1,4-dioxane concentrations in groundwater can range from 9 |ig/L at 10th percentile
to 13,460 |ig/L at the 90th percentile (Adamson et al.. 2014). Monitoring data from EPA's Third
Unregulated Contaminant Monitoring Rule (UCMR3) reported 1,4-dioxane concentrations in
groundwater ranging from 0.07 |ig/L to 34 |ig/L (Adamson et al. 201 ; 1 c. « i1 \ J017d). Though
many of the corresponding sites in these monitoring surveys are not specifically tied to the disposal of
1,4-dioxane to landfills, they provide context for what concentrations may be expected when
contamination occurs. When focusing on groundwater concentrations of 1,4-dioxane surrounding
landfills based on reasonably available information, EPA found concentrations of 1,4-dioxane ranging
from 6.4 to 25 mg/L (Cordone et al.. 2016). Leaching from unlined lagoons in Michigan resulted in
groundwater concentrations highs ranging from 1,000 to 20,000 |ig/L (Jackson ai ke. 2019; Mohr
and DiGuiseppi. 2010); four decades later concentrations are now reaching 2 mg/L or less after active
treatment and natural attenuation. Mean concentrations of 1,4-dioxane in landfill leachate in the United
States has ranged from 11.8 |ig/L for municipal landfills to 44.6 |ig/L for hazardous waste landfills (as
described in (Mohr and DiGuiseppi. 2010)). These concentrations further support that the modeled
concentrations are within the range of those reported in the literature.
2.3.2.3.2 Strengths, Limitations, and Sources of Uncertainty in Assessment Results
for Disposal to Landfills
Uncertainties and limitations are inherent in the modeling of groundwater concentrations from disposing
chemical substances into RCRA Subtitle D landfills and other non-Subtitle C landfills. These
uncertainties include, but are not limited to, determining the total and leachable concentrations of waste
constituents; estimating the release of pollutants from the waste management units to the environment;
and, estimating transport of pollutants in a range of variable environments by process that often are not
completely understood or are too complex to quantify accurately. To address some of these uncertainties
and add strength to the assessment, EPA considered multiple loading rates and multiple leachate
concentrations. These considerations add value to estimate exposure that falls at an unknown percentile
of the full distribution of exposures.
A strength of the assessment is that the modeled data are within the range of monitoring data that have
been evaluated at both the national scale (Adamson et a! JO I ; 1, c. « ^ \ :0l
-------
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
represent current conditions of waste management units in the United States. Both the DRAS model and
EPACMTP are based on a survey of drinking water wells located downgradient from a waste
management unit ( |). Due to the age of the survey, it is unclear how the survey represents
current conditions and proximity of drinking water wells to disposal units. Similarly, it is not clear if the
surveyed waste management units are representative of current waste management practices.
2.3.2.4 Disposal of Hydraulic Fracturing Produced Water to Surface Impoundments
After hydraulic fracturing operations inject fluids to extract oil and gas, a substantial volume of water
may be produced through flowback. Otherwise known as produced waters, the composition of this water
depends both on the geochemistry of the injected area and the injected fluids ( ). 1,4-
Dioxane has been reported to EPA as one of the chemicals present in produced waters by 411 facilities
via FracFocus 3.0 (GWPC and IOGCC. 2022) (Table 2-14). A variety of options exist for these
produced waters after use in hydraulic fracturing operations ranging from underground injection,
treatment and subsequent use, treatment and discharge, or evaporation in surface impoundments. Each
of these options are subject to state and federal regulations ( ). When produced waters
are released to unlined surface impoundments, there is potential for groundwater contamination and
subsequent human exposure via drinking water. Thus, EPA conducted an assessment to determine the
range of groundwater concentrations within a 1-mile radius of surface impoundments receiving
produced water from hydraulic fracturing operations.
Based on the results of Monte Carlo analysis presented in Table Apx E-5, disposal to these surface
impoundments could account for up to 3 percent of all produced waters. 1,4-Dioxane has been
documented to have a concentration of 60 |ig/L in these produced waters (Lester et at.. 2015). Thus,
EPA assessed the potential for disposing of hydraulic fracturing produced water at the 5th, 50th, 95th,
and 99th percentiles as well as at the min, mean, and max to a managed surface impoundment assuming
these loading rates and concentration using DRAS. The results are presented in Table 2-14.
2.3.2.4.1 Summary of Assessment for Disposal of Hydraulic Fracturing Produced
Water
In general, EPA determined that groundwater concentrations of 1,4-dioxane would increase as more
produced water was released to surface impoundments. The values presented in Table 2-14 represent the
maximum 3 3-year receptor well concentration within a 1-mile radius of a hypothetical surface
impoundment that leaches into groundwater. With each progressive iteration of summary statistic for
loading rate, potential groundwater concentrations increase accordingly. This increase can be attributed
to the decrease in the weight-adjusted dilution attenuation factor. As the mass of 1,4-dioxane entering an
aquifer increases, it is less diluted and higher concentrations will be found downgradient. Due to its
physical-chemical properties (e.g., water solubility, Henry's law constant) and fate characteristics (e.g.,
biodegradability, half-life in groundwater), 1,4-dioxane migrates in groundwater and can persist for
greater than 30 days in anaerobic environments (Adamson et at.. 2014; Mohr and DiGuiseppi. 2010) as
described in the 2020 RE. Thus, these concentrations are likely to represent the range of potential
groundwater concentrations for people living within a 1-mile radius of a surface impoundment,
representing a sentinel PESS exposure.
Page 81 of 484
-------
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 2-14. Total Annua
Release Summary
Total Annual Release
(kg/site-year)
Monte Carlo
Analysis Summary
Statistic
Weight Adjusted
Dilution Attenuation
Factor
Potential
Groundwater
Concentration (jig/L)
1.68E00
Max
3.18
1.89E-05
01.87E-01
99th Percentile
3.91
1.54E-05
6.52E-02
95th Percentile
3.91
1.54E-05
1.47E-02
Mean
84
7.10E-07
3.83E-03
50th Percentile
495
1.20E-07
3.24E-05
5th Percentile
495
1.20E-07
1.06E-11
Min
135,000
0.00E00
2.3.2.4.2 Strengths, Limitations, and Sources of Uncertainty in Assessment Results
for Disposal from Hydraulic Fracturing Operations
Although it is well understood that 1,4-dioxane is present in produced waters from hydraulic fracturing
as reported in FracFocus (GWPC and IOGCC. 2022). the number of studies reporting the concentration
of the chemical substances in produced waters is limited (Lester et ai. 2015). FracFocus is generally
considered a moderately reliable source of information as it is based on data from thousands of fracking
wells across the United States. Further, both the release assessment (as discussed in Section 2.2.1.2; see
also Table Apx E-7) and the groundwater concentration assessment (Table 2-14) are modeled using a
Monte Carlo simulation. These conditions lower the confidence in the overall assessment.
2.3.3 Ambient Air Pathway
EPA developed and applied tiered methodologies and analyses to estimate ambient air concentrations
and exposures to members of the general population. These methodologies and analyses focus on
inhalation exposures to a sub-set of the general population referred to as fenceline communities.
Fenceline communities are defined as a subset of the general population that are in proximity to air
emitting facilities or a receiving water body, and who therefore may be disproportionately exposed to a
chemical undergoing risk evaluation under TSCA section 6(b). For the air pathway, proximity goes out
to 10,000 meters from an air emitting source. The methodology and analyses were first presented in the
Draft TSCA Screem pproach for Assessing Ambient Air and Water Exposures to Fenceline
Communities referred to here as the 2022 Fenceline Report.1" The specific methodologies used in this
assessment to evaluate general population exposures to 1,4-dioxane in air are briefly described in Figure
2-18. Additional details on the methodologies are provided in Appendix J.
10 The 2022 Fenceline Report is available at https://www.epa.gov/assessing~and~managing~chemieals~iinder~tsea/tsea~
screening~level~approach~assessing~ambient~air~and.
Page 82 of 484
-------
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Figure 2-18. Brief Description of Methodologies and Analyses Used to Estimate Ambient
Air Concentrations and Exposures
EPA used the air release estimates obtained using the methodology described in Section 2.1.1.3 as direct
inputs for the models used to estimate exposure concentrations at various distances from a releasing
facility. EPA expanded upon the methods described in the 2022 Fenceline Report in response to SACC
comments/recommendations by evaluating potential aggregate concentrations from multiple facilities.
2.3.3.1 Measured Concentrations in Air
EPA did not identify quantitative outdoor air monitoring data for 1,4-dioxane.
2.3.3.2 Modeled Concentrations in Air
Because there is no air monitoring data for 1,4-dioxane, the Agency relied upon modeling to estimate
exposure concentrations to fenceline communities at various distances from a releasing facility.
Modeling was used for each analysis described in Figure 2-18 for 1,4-dioxane. For scenarios where the
screening methodology indicated a need for further analysis, EPA performed a full analysis using the
American Meteorological Society/Environmental Protection Agency Regulatory Model (AERMOD)
and/or EOA's Integrated Indoor/Outdoor Air Calculator (IIOAC).11 IIOAC analysis was performed for
three COUs where no site specific data were available (Hydraulic Fracturing, Industrial Laundry
Facilities, Institutional Laundry Facilities) and is briefly described in Section 2.3.3.2.4 with results
presented and discussed in Sections 5.2.2.3.2 and 5.2.2.3.3. An expanded analysis to consider aggregate
exposures was performed for 1,4-dioxane in response to SACC comments/recommendations on the
2022 Fenceline Report.
11 The IIOAC website is available at https://www.epa.gov/tsca-screening-tools/iioac-integrated-indoor-ontdoor-air-calailator.
Page 83 of 484
-------
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2.3.3.2.1 Ambient Air: Screening Methodology
The Ambient Air: Screening Methodology utilizes EPA's IIOAC model to estimate high-end and central
tendency (mean) 1,4-dioxane exposure concentrations in ambient air at three distances from an emitting
facility: 100, 100 to 1,000, and 1,000 m. EPA developed and evaluated a range of exposure scenarios for
each of two categorical release amounts12 designed to capture a variety of release types, topography,
meteorological conditions, and release scenarios. A diagram of these exposure scenarios is provided in
Appendix J. Findings from the Ambient Air: Screening Methodology were used to inform the need for a
higher-tier analysis as well as provide insight into whether risk estimates above the benchmarks are or
are not expected for 1,4-dioxane.
The Ambient Air: Screening Methodology design inherently includes both estimates of exposures as
well as estimates of risks to inform the need, or potential need, for further analysis. If findings from the
Ambient Air: Screening Methodology estimate risk (acute non-cancer, chronic non-cancer, or cancer)
for a given chemical above (or below as applicable) typical Agency benchmarks, EPA generally will
conduct a higher-tier analysis of exposures and associated risks for that chemical. If findings from the
Ambient Air: Screening Methodology estimate risks that do not exceed (or fall below as applicable)
benchmarks, EPA may still conduct a limited higher-tier analysis at distances very near a releasing
facility (less than 100 m) to ensure potential risks are not missed.
A more detailed description of the Ambient Air: Screening Methodology for 1,4-dioxane is provided in
Appendix J, along with summarized results. In general, for 1,4-dioxane, the results of this analysis
identified risk estimates above screening benchmarks for cancer at multiple distances and for multiple
releases (max and mean). In accordance with the tiered methodology presented to the SACC in the 2022
Fenceline Report, because risk estimates exceeded the benchmark for cancer based on the Ambient Air:
Screening Methodology, EPA conducted multiple higher-tier analyses of all facilities reporting releases
of 1,4-dioxane to TRI.
2.3.3.2.2 Ambient Air: Single Year Methodology (AERMOD)
The Ambient Air: Single Year Methodology (AERMOD) utilizes the EPA's American Meteorological
Society/Environmental Protection Agency Regulatory Model (AERMOD)13 to estimate 1,4-dioxane
concentrations in ambient air at eight finite distances (5, 10, 30, 60, 100, 2,500, 5,000, and 10,000 m)
and one area distance from an emitting facility.14 EPA modeled two different types of release estimates,
as applicable, for 1,4-dioxane: (1) facility-specific chemical releases with source attribution when 2019
TRI data was available, and (2) alternative release estimates representing a generic facility when 2019
TRI data was not available for an OES. Daily and period average outputs were obtained via modeling,
and post-processing scripts were used to extract a variety of statistics from the modeled concentration
distribution, including the 95th (high-end), 50th (central tendency), and 10th (low-end) percentile 1,4-
dioxane concentrations at each distance modeled.
A summary of the concentration ranges estimated using the Ambient Air: Single Year Methodology
(AERMOD) is provided in Table 2-15. The summary includes 11 OESs and select statistics (maximum,
12 The pre-screening methodology from the 2022 fenceline analysis evaluated two categorical release values across all
facilities reporting releases to the 2019 TRI. The first is the maximum single facility release reported across all facilities; the
second is the mean (arithmetic average) of all releases reported across all facilities reporting.
13 See https://www.epa.gOv/scram/air-qnalitY-dispersion-modeling-preferred-and-recotn.mended-models#aermod for more
information.
14 For the one "area distance" evaluated, receptors are placed in a cartesian grid between approximately 200 and 900 m, at
100 m spacing. This results in a total of 456 receptors. The exposure estimates for the area distance represent the arithmetic
average (mean) exposure concentration across all 456 receptors within the "area distance" for each day.
Page 84 of 484
-------
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
mean, median, and minimum) calculated from the modeled concentration distributions within each OES
at each distance modeled. The associated range of estimated concentrations is based on the maximum
95th percentile annual average exposure concentrations for each distance. Although the range of
concentrations are provided, there are many instances where the range extends as many as 12 orders of
magnitude from minimum to maximum concentration. This occurs because within each OES there are
several individual facilities evaluated and, in most cases, the reported release values from each
individual facility can vary widely (from 500 lb to several hundred thousand pounds), which in turn
affects the range of estimated exposure concentrations at a given distance. Therefore, in trying to
summarize the wide variety of releases into a single range, the variation in estimated concentrations will
also appear extensive. This is not indicative of an inadequate analysis or methodology, but solely based
on the variability of releases across facilities within a given OES.
Page 85 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2654 Table 2-15. Summary of Select Statistics for the 95th Percentile Estimated Annual Average Concentrations from the "Full-
2655 Screening" Analysis for 1,4-Dioxane Releases Reported to TRI
OES
# Facilities
Evaluated
in OES
Statistic
Annual Average Concentration (ppm) Estimated within 5 to 10,000 m of Releasing Facilities
5
10
30
60
100
100 to
1,000
2,500
5,000
10,000
Disposal
15
Max
4.26E-03
5.05E-03
1.80E-03
6.90E-04
3.15E-04
2.95E-05
1.89E-06
6.28E-07
2.09E-07
Mean
4.76E-04
6.92E-04
2.79E-04
1.12E-04
5.39E-05
6.00E-06
346E-07
1.18E-07
4.12E-08
Median
8.44E-06
1.65E-05
9.35E-06
8.70E-06
5.81E-06
7.64E-07
4.53E-08
1.48E-08
4.81E-09
Min
3.31E-15
9.85E-14
5.17E-11
9.72E-10
2.03E-09
1.10E-09
1.21E-10
3.81E-11
1.22E-11
Dry Film
Lubricant
8
Max
1.61E-10
7.14E-09
5.10E-07
3.88E-06
6.29E-06
9.92E-07
2.79E-08
8.44E-09
3.68E-09
Mean
2.06E-11
9.46E-10
1.90E-07
2.28E-06
4.05E-06
8.14E-07
1.95E-08
5.94E-09
2.45E-09
Median
2.46E-13
3.58E-11
1.59E-07
2.21E-06
4.00E-06
7.75E-07
1.88E-08
6.02E-09
2.66E-09
Min
4.05E-18
2.19E-13
5.64E-08
9.23E-07
2.39E-06
7.39E-07
1.36E-08
4.02E-09
1.40E-09
Ethoxylation
Byproduct
6
Max
6.53E-03
1.36E-02
7.33E-03
3.09E-03
1.64E-03
3.81E-04
2.20E-05
9.00E-06
3.45E-06
Mean
1.74E-03
3.05E-03
149E-03
6.18E-04
3.23E-04
7.03E-05
4.02E-06
1.62E-06
6.13E-07
Median
2.44E-04
4.40E-04
2.08E-04
8.70E-05
440E-05
6.36E-06
3.09E-07
1.03E-07
3.40E-08
Min
4.08E-14
6.32E-13
4.29E-10
5.22E-09
1.15E-08
4.99E-09
7.17E-10
3.39E-10
1.40E-10
Film Cement
1
Max
1.25E-04
1.31E-04
441E-05
2.28E-05
1.25E-05
2.29E-06
1.38E-07
4.60E-08
1.52E-08
Mean
3.90E-05
4.87E-05
2.04E-05
9.36E-06
4.82E-06
7.95E-07
444E-08
1.48E-08
4.87E-09
Median
2.02E-05
2.93E-05
1.74E-05
7.33E-06
3.57E-06
5.68E-07
248E-08
8.10E-09
2.65E-09
Min
3.17E-06
6.36E-06
4.97E-06
1.95E-06
8.99E-07
1.32E-07
444E-09
1.42E-09
4.61E-10
Functional
Fluids (Open-
System)
2
Max
1.28E-05
2.36E-05
1.03E-05
1.08E-05
1.82E-05
7.42E-06
6.78E-07
2.47E-07
8.81E-08
Mean
6.40E-06
1.18E-05
5.74E-06
7.71E-06
1.08E-05
4.24E-06
3.88E-07
1.45E-07
5.28E-08
Median
6.40E-06
1.18E-05
5.74E-06
7.71E-06
1.08E-05
4.24E-06
3.88E-07
1.45E-07
5.28E-08
Min
1.66E-11
1.93E-10
1.18E-06
4.61E-06
3.37E-06
1.06E-06
9.70E-08
4.30E-08
1.74E-08
Import and
Repackaging
1
Single Facility
2.70E-11
5.57E-10
5.52E-08
4.17E-07
8.70E-07
3.21E-07
6.72E-08
4.12E-08
2.23E-08
Page 86 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
# Facilities
Evaluated
in OES
Statistic
Annual Average Concentration (ppm) Estimated within 5 to 10,(MM) m of Releasing Facilities
5
10
30
60
100
100 to
1,000
2,500
5,000
10,000
Industrial Uses
12
Max
4.19E-03
4.78E-03
1.54E-03
5.67E-04
2.80E-04
7.15E-05
8.79E-06
3.22E-06
1.13E-06
Mean
8.76E-04
1.14E-03
4.08E-04
1.70E-04
9.63E-05
1.69E-05
1.48E-06
5.45E-07
1.94E-07
Median
8.76E-05
1.14E-04
3.83E-05
1.65E-05
9.94E-06
2.23E-06
3.19E-07
1.17E-07
4.04E-08
Min
7.75E-13
1.69E-12
2.40E-09
2.50E-08
1.23E-08
1.10E-09
6.36E-11
1.97E-11
6.14E-12
Laboratory
Chemical Use
1
Max
2.06E-03
2.15E-03
7.26E-04
3.75E-04
2.06E-04
3.76E-05
2.27E-06
7.57E-07
2.50E-07
Mean
6.84E-04
8.52E-04
3.58E-04
1.64E-04
8.46E-05
1.39E-05
7.77E-07
2.59E-07
8.55E-08
Median
4.30E-04
5.65E-04
3.15E-04
1.36E-04
6.68E-05
1.08E-05
4.82E-07
1.59E-07
5.24E-08
Min
7.39E-05
1.48E-04
1.16E-04
4.55E-05
2.09E-05
3.08E-06
1.03E-07
3.30E-08
1.07E-08
Manufacturing
1
Single Facility
8.73E-03
1.63E-02
7.69E-03
3.22E-03
1.59E-03
1.42E-04
8.21E-06
2.54E-06
7.92E-07
PET
Manufacturing
13
Max
8.01E-03
9.57E-03
3.50E-03
1.40E-03
6.43E-04
1.07E-04
2.07E-05
1.24E-05
6.58E-06
Mean
1.41E-03
1.89E-03
7.83E-04
3.36E-04
1.85E-04
3.31E-05
4.23E-06
2.08E-06
9.60E-07
Median
8.00E-04
1.64E-03
5.21E-04
2.27E-04
1.42E-04
2.64E-05
2.48E-06
1.09E-06
3.94E-07
Min
6.04E-12
8.54E-11
3.01E-08
2.43E-07
5.56E-07
3.02E-07
4.33E-08
2.07E-08
9.30E-09
Spray Foam
Application
1
Max
7.79E-07
8.40E-07
2.85E-07
1.50E-07
8.55E-08
1.55E-08
1.72E-09
6.30E-10
2.45E-10
Mean
2.68E-07
3.30E-07
1.34E-07
6.21E-08
3.28E-08
5.29E-09
4.78E-10
1.67E-10
5.97E-11
Median
1.41E-07
1.95E-07
1.14E-07
4.88E-08
2.36E-08
3.64E-09
2.25E-10
7.40E-11
2.40E-11
Min
2.51E-08
4.43E-08
3.45E-08
1.36E-08
6.07E-09
8.42E-10
3.26E-11
1.10E-11
3.74E-12
2656
Page 87 of 484
-------
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Based on the air concentrations estimated through the Ambient Air: Single Year Methodology, EPA also
estimated potential aggregate air concentrations resulting from the combined releases of multiple
facilities in proximity to each other. Details of the methods used to aggregate exposure and
corresponding risk are presented in Appendix J.4.
2.3.3.2.3 Ambient Air: Multi-Year Analysis (IIOAC)
The multi-year analysis utilizes EPA's IIOAC model to estimate high-end and central tendency (mean)
1,4-dioxane concentrations in ambient air at three distances from an emitting facility: 100, 100 to 1,000,
and 1,000 m. The multi-year analysis incorporates SACC recommendations on the 2022 Fenceline
Report by evaluating multiple years of chemical release data to estimate exposures and associated risks
to fenceline communities. This is achieved for 1,4-dioxane by expanding upon methodologies described
in the 2022 Fenceline Report and conducting a facility-by-facility evaluation of all 1,4-dioxane releases
reported to TRI (2015 through 2020). Data for these 6 years were obtained from the TRI database (TRI
basic plus files downloaded on August 5, 2022). Annual release data for 1,4-dioxane were extracted
from the entire TRI data set for all facilities reporting air releases of 1,4-dioxane for one or more years
between 2015 and 2020. Facilities were categorized into occupational exposure scenarios for modeling
purposes to inform the release scenarios evaluated.
The multi-year analysis provides highlights of the year-to-year variability that exists in the release data
and illustrates the potential impact of considering multiple years of TRI data on exposure and risk
estimates. The findings from the multi-year analysis can also be used in a comparative manner to
determine how representative the single year of data used for the Ambient Air: Single Year
Methodology (AERMOD) presented in the 2022 Fenceline Report is or to provide additional confidence
in the findings from the Ambient Air: Single Year Methodology (AERMOD) described in the 2022
Fenceline Report for purposes of estimating exposures and associated risks to fenceline communities. In
broader terms, the multi-year analysis provides both a broad analysis of multiple years of release data
and enables a general comparison to the Ambient Air: Single Year Methodology (AERMOD) results
described above and in the 2022 Fenceline Report.
2.3.3.2.4 Ambient Air: IIOAC Methodology for COUs without Site-Specific Data
(Hydraulic Fracturing, Industrial, and Institutional Laundry Facilities)
For COUs without site-specific data, EPA's IIOAC model was used to estimate high-end and central
tendency (mean) 1,4-dioxane concentrations in ambient air at three distances from an emitting facility
(100, 100 to 1,000, and 1,000 m). This methodology was applied for three unique COUs (hydraulic
fracturing, and industrial, and institutional laundry facilities) where there was no site-specific data
available for modeling. Environmental releases (fugitive and stack) along with other data (like days of
release) for these COUs were estimated using Monte Carlo modeling. As such, the Ambient Air: IIOAC
Methodology for COUs without site-specific data was developed to allow modeling all possible
iterations of releases provided across eight different exposure scenarios, including consideration of
source attribution as well as actual days of release. Additionally, the product form for laundry detergent
was provided, allowing for analysis of releases associated with detergent in vapor only form, as well as
solid form (particulate) either coarse (PM10) or fine (PM2.5). A description of this methodology is
provided in Appendix J along with a summary of the model inputs and exposure scenarios evaluated. A
full list of the inputs, exposure scenarios, and results is provided in 1,4-Dioxane Supplemental
Information File: Air Exposure and Risk Estimates for 1,4-Dioxane Emissions from Hydraulic
Fracturing Operations ( s23b) and 1,4-Dioxane Supplemental Information File: Air
Exposures and Risk Estimates for Industrial Laundry ( '023c).
Page 88 of 484
-------
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2.3.3.3 Strengths, Limitations, and Sources of Uncertainty for Modeled Air
Concentrations
EPA has medium to high confidence in the air concentrations estimated from TRI release data using
IIOAC and AERMOD.
IIOAC
IIOAC estimates air concentrations at three pre-defined distances (100, 100 to 1,000, and 1,000 m). The
inherent distance limitations of IIOAC do not allow estimation of exposures closer to a facility (less than
100 m from the facility) where we expect to see higher exposures from fugitive releases.
IIOAC uses meteorological data from 14 pre-defined meteorological stations representing large regions
across the United States. This generalizes the meteorological data used to estimate exposure
concentrations where competing conditions can influence the exposure concentrations modeled upwind
and downwind of a releasing facility. To reduce the uncertainties associated with using regional
meteorological data, EPA conducted a sensitivity analysis of all 14 pre-defined meteorological stations
to identify which two within IIOAC tended to result in a high-end and central tendency estimate of
exposure concentrations. This maintained a more conservative exposure concentration estimate which is
then used in calculations to estimate risks. This approach adds confidence to the findings by ensuring,
potential risks would be captured under a high-end exposure scenario, while also providing insight into
potential risks under a less conservative exposure scenario (central tendency).
AERMOD
AERMOD is an EPA regulatory model and has been thoroughly peer reviewed; therefore, the general
confidence in results from the model is high but relies on the integrity and quality of the inputs used and
interpretation of the results. For the full analysis, EPA used releases reported to the 2019 TRI as direct
inputs to AERMOD. Although there is some uncertainty around the representativeness of using only a
single year of data, AERMOD successfully estimated exposure concentrations to fenceline communities.
Furthermore, in response to SACC recommendations to use multiple years of data to estimate exposures
and associated risks, EPA developed the IIOAC and conducted a multi-year analysis using 6 years of
TRI data and compared the results to those of AERMOD and found exposure concentration estimates
from the 2019 data is generally representative of other years.
AERMOD relied upon the latitude/longitude information reported by each facility to TRI as the location
for the point of release. While this may generally be a close approximation of the release point for a
small facility (for example a single building), it may not represent the release point within a much larger
facility. Therefore, there is some uncertainty associated with the modeled distances from each release
point and the associated exposure concentrations to which fenceline communities may be exposed. For
small facilities where the latitude/longitude may closely approximate the release point, there is a less
uncertainty that the estimated exposure at the associated distance is representative of exposure to
fenceline communities at that distance. For larger facilities, where the latitude/longitude may be several
hundred meters away from the actual release point, there is a higher uncertainty that the estimated
exposure at the associated distance is representative of exposure to fenceline communities at that
distance.
The TRI datasets used for both AERMOD and IIOAC do not include source specific stack parameters
that can affect plume characteristics and associated dispersion of the plume. Therefore, EPA used pre-
defined stack parameters within IIOAC to represent stack parameters of all facilities modeled using each
of these methodologies. Those stack parameters include a stack height 10 m above ground with a 2-
meter inside diameter, an exit gas temperature of 300° Kelvin, and an exit gas velocity of 5 m per
Page 89 of 484
-------
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
second (see Table 6 of the IIOAC User Guide). These parameters were selected since they represent a
slow-moving, low-to-the-ground plume with limited dispersion which results in a more conservative
estimate of exposure concentrations at the distances evaluated. As such, these parameters may result in
some overestimation of emissions for certain facilities modeled.
Additionally, the assumption of a 10x 10 area source for fugitive releases may impact the exposure
estimates very near a releasing facility (5 and 10 m from a fugitive release). This assumption places the
receptor at 5 m directly on top of the release point which may result in an over or underestimation of
exposure. This assumption places the 10-meter receptor just off the release point that may again result in
either an over or underestimation of exposure depending on other factors like meteorological data,
release heights, and plume characteristics.
For facilities reporting releases to TRI via a TRI Form A (which is allowed for use by those facilities
releasing less than 500 lb of the chemical reported), EPA assumed the maximum release value of 500 lb
for exposure modeling purposes. TRI Form A reporters do not provide source attribution (fugitive or
stack releases) so EPA modeled each facility associated with a Form A submittal twice—once assuming
all 500 lb of the reporting threshold was fugitive and once assuming all 500 lb of the reporting threshold
was stack. There is no way to attribute a certain portion of the releases to each release type, so this
modeling approach represents a conservative estimate, in terms of total release, but may overestimate
exposure concentrations associated with each release type if a facility did not actually release all 500 lb
via a single release type or even combined release type. To avoid the potential double counting of
facility releases for TRI Form A reporters, when presenting potential exposures EPA presented only the
highest (more conservative) exposure concentration estimated for either of the two release types for
purposes of evaluating potential risks to fenceline communities. Given the exposure scenarios modeled,
this tended to result from the exposure scenario which assumed all 500 lb of the release were fugitive
releases.
Page 90 of 484
-------
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3 HUMAN EXPOSURES
1,4-Dioxane - Human Exposures (Section 3):
Key Points
EPA evaluated reasonably available information for occupational exposures to 1,4-dioxane present
as a byproduct, and general population, including PESS, exposures to 1,4-dioxane present in the
environment.
• Occupational exposures for workers and ONUs to 1,4-dioxane present as a byproduct occur
during manufacturing, through use of commercial products, or in hydraulic fracturing
operations.
• General population exposures to 1,4-dioxane occur when 1,4-dioxane is present in potential
drinking water sources or ambient air, particularly in fenceline communities.
• EPA considered the potential for increased exposures across PESS factors throughout the
exposure assessment. PESS categories incorporated into this supplemental exposure
assessment include
o Lifestage (including formula-fed infant exposures),
o Occupational exposures (including high-end exposure scenarios), and
o Geography/site-specific factors (i.e., fenceline community exposures)
3.1 Occupational Exposures
1,4-Dioxane - Occupational Exposures (Section 3.1):
Key Points
EPA considered the reasonably available information to evaluate occupational exposures.
• EPA estimated occupational exposures to 1,4-dioxane through air and skin. EPA estimated
both high-end and central tendency exposures for occupational exposure scenarios
associated with each COU.
• Exposure for most COUs was estimated based on monitoring data. For COUs without
monitoring data, EPA applied Monte Carlo statistical modeling approaches to estimate
exposures.
The following sections describe EPA's approach to assessing occupational exposures for OESs
involving industrial and commercial products containing 1,4-dioxane as a byproduct. The assessed OESs
include textile dye, antifreeze, surface cleaner, dish soap, dishwasher detergent, institutional and
industrial laundries, paints and floor lacquer, PET byproducts, ethoxylation process byproducts, and
hydraulic fracturing. For a crosswalk linking COUs to OESs, see Table 2-1. The remaining OESs have
occupational exposure assessments in Section 2.4.1 of the Final Risk Evaluation for 1,4-Dioxane (U.S.
EPA. 2020c).
EPA distinguishes between exposures to workers and exposures to ONUs. Normally, workers may
handle 1,4-dioxane and have direct contact with the chemical, while ONUs are working in the general
Page 91 of 484
-------
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
vicinity of workers but do not handle 1,4-dioxane and do not have direct contact with 1,4-dioxane being
handled by the workers.
EPA evaluated acute and chronic inhalation exposures to workers and ONUs, and dermal exposures to
workers. EPA did not assess dermal exposures to ONUs as EPA does not expect ONUs to have routine
dermal exposures in the course of their work.
The occupational exposure assessment for each COU comprises the following components:
• Process Description of the COU, including the role of the chemical in the use; process vessels,
equipment, and tools used during the COU; and descriptions of the worker activities, including
an assessment for potential points of worker exposure.
• Number of Sites that use the chemical for the given COU.
• Number of Workers and ONUs potentially exposed to the chemical for the given COU. Unless
mentioned otherwise in this report, the total number of workers and ONUs are number of
personnel per site per day. The details on estimation of the number of workers and ONUs are
discussed below for each COU.
• Central Tendency and High-End Estimates of Inhalation Exposure to workers and
occupational non-users. See "General Approach and Methodology for Environmental Releases"
for a discussion of EPA's statistical analysis approach for assessing inhalation exposure.
• Dermal Exposure estimates for multiple scenarios, accounting for simultaneous absorption and
evaporation, and different protection factors of glove use.
• Users include adult workers (greater than 16 years old) exposed to 1,4-dioxane for 8-hour
exposure.
• ONUs include female and male adult workers (greater thanl6 years old) exposed to 1,4-dioxane
indirectly by being in the same work area of the building.
3.1.1 Approach and Methodology
EPA developed occupational exposure values representative of central tendency (50th percentile, mean)
conditions and high-end (90th and 99.9th percentiles). Additional explanation of central tendency and
high-end conditions are described in the Final Risk Evaluation for 1,4-Dioxane ( 2020c).
3.1.1.1 Process Description, Number of Sites, Number of Workers, and ONUs
EPA performed a literature search to find descriptions of processes involving 1,4-dioxane and
worker activities that could potentially result in occupational exposures. This literature search was
specific to the scope of this draft supplement and is described in Section 1.4. A summary of the data
quality evaluation results for the 1,4-dioxane occupational exposure sources are presented in the
attachment Systematic Review Supplemental File: Data Quality Evaluation and Data Extraction
Information for Environmental Release and Occupational Exposure ( 2023t).
EPA used a variety of sources to supplement the data found through the Systematic Review
process. The additional sources included relevant NIOSH Health Hazard Evaluations, Generic
Scenarios, and ESDs. These sources were sometimes used to provide process descriptions of the COUs
as well as estimates for the number of sites and number of workers. Because CDR data were not
available for the COUs included in this occupational exposure assessment, EPA used data from the
Bureau of Labor Statistics (BLS) and the U.S. Census' Statistics of US Businesses (SUSB) to estimate
the number of sites, workers, and ONUs for each OES. This approach involved the identification of
relevant Standard Occupational Classification (SOC) codes within the BLS data for the identified
NAICS codes for each OES. First, EPA identified the affected NAICS codes. Then, EPA reviewed
Page 92 of 484
-------
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
occupation descriptions to designate which SOC codes contained potentially exposed workers and
ONUs. The Agency refined the estimates by using U.S. Census Bureau data. Next, EPA estimated the
percentage of workers using 1,4-dioxane instead of other chemicals to calculate number of workers per
site. Finally, this data was separated by COU. Additional details on this approach can be found in
Appendix G.5 of the Final Risk Evaluation for 1,4-Dioxane ( 020c). These estimates were
utilized in Monte Carlo modeling of environmental releases and occupational exposures.
3.1.1.2 Inhalation Exposures Approach and Methodology
EPA used inhalation monitoring data from literature sources having high or medium data quality ratings
during data evaluation. EPA used modeling approaches to estimate potential inhalation exposures where
inhalation monitoring data were not available.
The Agency reviewed workplace inhalation monitoring data collected by government agencies such as
OSHA and NIOSH, and monitoring data found in published literature {i.e., personal exposure
monitoring data and area monitoring data). Central tendency and high-end exposure values were
calculated from the monitoring data provided in the sources depending on the size of the dataset {i.e.,
number of data points). Where discrete sampling points were not provided in the source and EPA was
unable to calculate central tendency and high-end values, the Agency used values of central tendency
and high-end that were provided in the source. EPA's approach for evaluating central tendency and
high-end estimates from inhalation monitoring data is further discussed in the Final Risk Evaluation for
1,4-Dioxane ( )20c). EPA used the following types of monitoring data of 1,4-dioxane from
various sources to estimate occupational inhalation exposure:
• Personal sample monitoring data from directly applicable scenarios {e.g., personal breathing
zone [PBZ]). This type of monitoring data was used for the textile dye, surface cleaner, dish
soap, paint and floor lacquer, PET byproduct, and the Ethoxylation Process Byproduct OESs.
• Personal sample monitoring data from potentially applicable or similar scenarios. Specifically,
PBZ data from the dish soap OES was also used for the dishwasher detergent OES because these
OESs are expected to be similar.
EPA used the following models and modeling approaches to estimate occupational inhalation exposure
where no monitoring data were found:
• Monte-Carlo statistical modeling approaches, which was used for the antifreeze, laundry
detergent, and hydraulic fracturing OES. EPA developed these models for the purposes of this
assessment. The models and the associated sources of data used in the modeling are described in
detail in Appendices F.7, F.8, and F.9, respectively.
• Additional modeling approaches, including the use of surrogate data and fundamental modeling
approaches for the spray polyurethane foam OES in the Final Risk Evaluation for 1,4-Dioxane
(U.S. EPA. 2020c). While this OES is included in the scope of this draft supplement, EPA
evaluated occupational exposure estimates for this OES in the published risk evaluation and
these estimates remain unchanged in this draft supplement.
• EPA AP-42 Loading Model estimates vapor releases that occur when vapor is displaced by liquid
during container loading. It calculates a vapor generation rate (G) using the physio-chemical
properties of the chemical.
• EPA Mass Balance Inhalation Model estimates occupational inhalation exposures assuming the
air immediately around the source of exposure behaves as a well-mixed zone. EPA used the
vapor generation rate (G), calculated using the EPA AP-42 Loading Model, in conjunction with
this model to develop estimates of inhalation exposure.
Page 93 of 484
-------
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• Generic Model for Central Tendency and High-End Inhalation Exposure to Total and Respirable
Particulates Not Otherwise Regulated (PNOR) estimates occupational inhalation exposures to
particulates containing the chemical using OSHA PNOR data.
EPA did not utilize occupational exposure limits to estimate occupational inhalation exposures in this
assessment because sufficient monitoring data or modeling approaches were available for all OES.
EPA then used measured or modeled air concentrations to calculate exposure concentration metrics
essential for risk assessment. These exposures are presented as 8-hour time weighted averages (TWAs)
and used to calculate average daily concentrations (ADCs) and lifetime average daily concentrations
(LADCs). The ADC is used to estimate chronic, non-cancer risks and the LADC is used to estimate
chronic, cancer risks. These calculations required additional parameter inputs, such as years of exposure,
exposure duration and frequency, and lifetime years. See Appendix F.l for more information about
parameters and equations used to calculate acute and chronic exposures.
3.1.1.3 Dermal Exposures Approach and Methodology
EPA modeled dermal doses using the EPA Dermal Exposure to Volatile Liquids Model. This model
determines a dermal potential dose rate based on an assumed amount of liquid on skin during one
contact event per day and the steady-state fractional absorption for 1,4-dioxane. The amount of liquid on
the skin is adjusted by the weight fraction of 1,4-dioxane in the liquid to which the worker is exposed.
This is the same approach that EPA used in the Final Risk Evaluation for 1,4-Dioxane (U.S. EPA.
2020c).
3.1.1.4 Engineering Controls and Personal Protective Equipment
EPA's occupational exposure estimates do not assume the use of engineering controls or PPE.
Additionally, as discussed in Section 5.2.1, the risk estimates are based on exposures to workers in the
absence of PPE such as gloves or respirators. Reasonably available monitoring data or information on
effectiveness of engineering control and PPE for reducing occupational exposures to 1,4-dioxane during
the assessed OES were not available. This section presents a general discussion on engineering controls
and PPE for informative purposes only.
OSHA recommends employers utilize the hierarchy of controls for reducing or removing hazardous
exposures. The most effective controls are elimination, substitution, or engineering controls. Respirators,
and any other personal protective equipment (PPE), are the last means of worker protection in the
hierarchy of controls and should only be considered when process design and engineering controls
cannot reduce workplace exposure to acceptable levels. OSHA's Respiratory Protection Standard (29
CFR 1910.134) provides a summary of respirator types by their assigned protection factor (APF). OSHA
defines the APF to mean the workplace 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 according to the requirements of the OSHA Respiratory Protection
Standard. Exposure limits, respirator requirements, worker respirator use rates, and a table of APFs for
different types of respirators are provided in the 2020 RE ( 2020c).
OSHA's hand protection standard (29 CFR 1910.138) states that employers must select and require
employees to use appropriate hand protection when employees are expected to be exposed to hazards
such as those from skin absorption of harmful substances; severe cuts or lacerations; severe abrasions;
punctures; chemical burns; thermal burns; and harmful temperature extremes. Dermal protection
selection provisions are provided in § 1910.138(b) and require that appropriate hand protection is
selected based on the performance characteristics of the hand protection relative to the task(s) to be
performed, conditions present, duration of use, and the hazards to which employees will be exposed.
Page 94 of 484
-------
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Unlike respiratory protection, OSHA standards do not provide protection factors (PFs) associated with
various hand protections. Data regarding the frequency of effective glove use in industrial and
commercial settings is limited. While there are no regulatory standards containing dermal protection
factors, the European Center for Ecotoxicity and Toxicology of Chemicals (ECETOC) targeted risk
assessment (TRA) tool includes fixed, assigned protection factors equal to 5, 10, or 20 for various
dermal protection strategies. These are discussed in Appendix F.3 and further explained in the 2020 RE
(I >020cY
3.1.2 Occupational Exposure Estimates
In this section, EPA provides a summary of the exposure estimates for each OES, including estimates
for number of workers and ONUs, inhalation exposures, and dermal exposures. For the crosswalk
linking COU to OES, see Table 2-1. Note that EPA assessed dermal exposures for all OESs with the
same methodology, which is described at the end of this section.
3.1.2.1 Summary of Inhalation Exposure Assessment
EPA estimated central tendency and high-end occupational inhalation exposures using various methods
and information sources—including OSHA data, NIOSH health hazard evaluation data, and GSs and
ESDs with Monte Carlo modeling. EPA estimated inhalation exposures as 8-hour TWA values for the
COUs included in this draft supplement per Table 2-1. Using the estimated central tendency and high-
end inhalation exposures with the estimated exposure frequency, EPA then calculated the cancer and
non-cancer exposures using the calculations described in Appendix F.l.
A summary of the occupational inhalation exposures is presented Table Apx F-32. EPA used
monitoring data to estimate occupational inhalation exposures to workers for the textile dye, surface
cleaner, dish soap, dishwasher detergent, paint and floor lacquer, PET byproduct, and ethoxylation
byproduct conditions of use. This monitoring data was found to be relevant to these scenarios and based
on medium to high data quality. However, several of the scenarios had a low number of samples and
may have preceded changes in current industry practices. Additionally, sufficient representation of the
entire industry is uncertain due to the limited number of sites. For the remaining conditions of use
included in this draft supplement, which are antifreeze, laundry detergent, and hydraulic fracturing, EPA
did not find reasonably available monitoring data and estimated worker inhalation exposure using GSs
and ESDs with Monte Carlo modeling. The applied models are directly relevant to these conditions of
use, but the underlying distributions may not sufficiently capture variability across entire industry
sectors. For both measured and modeled data, the degree of certainty to which these data represent the
true distribution of exposure and the potential over- or underestimation of exposure is unknown.
Monitoring data and modeling approaches were not available to estimate occupational inhalation
exposures for ONUs. The ONU exposures are anticipated to be lower than worker exposures since
ONUs do not typically directly handle the chemical.
The PET byproduct and textile dyes conditions of use had the highest central tendency and high-end
worker inhalation exposure values, respectively. For PET byproduct, worker inhalation exposures were
estimated using OSHA monitoring data, which resulted in central tendency exposure of 4.7 mg/m3 and
high-end exposure of 47 mg/m3. For textile dyes, worker inhalation exposures were also estimated using
OSHA monitoring data, which resulted in central tendency exposure of 0.066 mg/m3 and high-end
exposure of 74 mg/m3.
The monitoring data sources and GSs and ESDs used to estimate occupational inhalation exposures all
had overall data quality determinations of either medium or high. The basis for determining overall data
Page 95 of 484
-------
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
quality is described in this section. In summary, each source is evaluated on multiple metrics based on
defined criteria. The individual metric ratings are used to obtain an overall study rating. All metrics have
an equal weight in determining the overall study rating. The resulting values are converted to an overall
data quality determination of "high," "medium," "low," "critically deficient," or "not rated/not
applicable." For more details on this process, see Section 5 of the 2021 Draft Systematic Review
Protocol
This section also includes information on the weight of the scientific evidence conclusions for these
estimates, and a summary of the strengths, limitations, assumptions, and key sources of uncertainty for
these estimates.
3.1.2.2 Summary of Dermal Exposures Assessment
Table 3-1 presents the estimated dermal absorbed dose for workers in various OES. The dose estimates
assume one dermal exposure event (applied dose) per workday and that approximately 78 or 86 percent
of the applied dose is absorbed through the skin (depending on whether the OES is industrial or
commercial). The exposure estimates are provided for each OES, where the OES are "binned" based on
characteristics known to effect dermal exposure such as the maximum weight fraction of 1,4-dioxane
that could be present in that OES, open or closed system use of 1,4-dioxane, and large or small-scale
use. For a more detailed description of EPA's dermal assessment approach and each bin, see Appendix
F.3.
As shown in the Table 3-1, the calculated dermal absorbed dose for workers is lower in comparison to
those presented in the December 2020 Final Risk Evaluation for 1,4-Dioxane ( )20c). This is
due to the relatively lower concentrations of 1,4-dioxane found for the OES included in this draft
supplement than for those included in the 2020 RE. As noted previously, EPA did not assess dermal
exposures to ONUs as the Agency does not expect ONUs to have routine dermal exposures in the course
of their work. Depending on the OES, ONUs may have incidental dermal exposures due to surface
contamination. However, data (e.g., frequency and amount of liquid on the skin after contact) were not
identified to assess this exposure.
Table 3-1. Estimated Dermal Absorbed Dose (mg/day) for Workers in Various Conditions of Use
Weight
Fraction
(Max
Yilm,,)
No Gloves
(PF = 1)
Exposures Due to Glove Permeation/Chemical
Breakthrough (mg/day)
OES
Bin
Use Setting
Protective
Gloves
(PF = 5)
Protective
Gloves
(PF= 10)
Protective Gloves
(Industrial Uses
Only, PF = 20)
Textile Dye
7
Industrial
and
Commercial
4.7E-06
0.003 (CT)
0.009 (HE)
0.001 (CT)
0.002 (HE)
3.0E-4 (CT)
0.001 (HE)
1.5E-4 (CT)
4.5E-4 (HE)
Antifreeze
8
Commercial
8.6E-05
0.055 (CT)
0.165 (HE)
0.011 (CT)
0.033 (HE)
0.006 (CT)
0.017 (HE)
N/A
Surface
Cleaner
9
Commercial
9.0E-06
0.006 (CT)
0.017 (HE)
0.001 (CT)
0.003 (HE)
0.001 (CT)
0.002 (HE)
N/A
Dish Soap
10
Commercial
2.04E-04
0.131 (CT)
0.393 (HE)
0.026 (CT)
0.079 (HE)
0.013 (CT)
0.039 (HE)
N/A
Dishwasher
Detergent
11
Commercial
5.1E-05
0.033 (CT)
0.098 (HE)
0.007 (CT)
0.020 (HE)
0.003 (CT)
0.010 (HE)
N/A
Page 96 of 484
-------
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Weight
Fraction
(Max
Yilcrm)
No Gloves
(PF = 1)
Exposures Due to Glove Permeation/Chemical
Breakthrough (mg/dav)
OES
Bin
Use Setting
Protective
Gloves
(PF = 5)
Protective
Gloves
(PF= 10)
Protective Gloves
(Industrial Uses
Only, PF = 20)
Laundry
Detergent
(Industrial
and
Institutional)
12
Industrial
and
Commercial
1.4E-05
0.009 (CT)
0.027 (HE)
0.002 (CT)
0.005 (HE)
0.001 (CT)
0.003 (HE)
4.5E-4 (CT)
0.001 (HE)
Paint and
Floor
Lacquer
13
Industrial
and
Commercial
3.0E-05
0.019 (CT)
0.058 (HE)
0.004 (CT)
0.012 (HE)
0.002 (CT)
0.006 (HE)
0.001 (CT)
0.003 (HE)
Polyethylene
Terephthalate
(PET)
Byproduct
14
Industrial
1.4E-03
0.827 (CT)
2.48 (HE)
0.165 (CT)
0.496 (HE)
0.083 (CT)
0.248 (HE)
0.041 (CT)
0.124 (HE)
Ethoxylation
Process
Byproduct
Hydraulic
Fracturing
15
Industrial
and
Commercial
0.05
32.1 (CT)
96.2 (HE)
6.41 (CT)
19.2 (HE)
3.21 (CT)
9.62 (HE)
1.60 (CT)
4.81 (HE)
CT = central tendency; HE = high-end; PF = protection factor
3.1.2.3 Weight of the Scientific Evidence Conclusions for Occupational Exposure
Information
Table 3-2 provides a summary of EPA's overall weight of the scientific evidence conclusions for its
occupational exposure estimates for each of the assessed OES. These determinations are OES-specific.
For a description of overall confidence in all inhalation exposures, see Section 3.3.1.1. For an
explanation of EPA's judgement on the weight of the scientific evidence conclusion, see Section 2.2.1.2.
Factors that increase and decrease the strength of the weight of the scientific evidence are listed in
Table Apx C-5.
Due to a lack of data, EPA was not able to estimate ONU inhalation exposure from monitoring data or
models, so a qualitative assessment of potential ONU exposures was made. Similarly, EPA did not
assess dermal exposures to ONUs as EPA does not expect ONUs to have routine dermal exposures in
the course of their work. Depending on the COU, ONUs may have incidental dermal exposures due to
surface contamination. However, data (e.g., frequency and amount of liquid on the skin after contact)
were not identified to assess this exposure. Finally, due to the absence of dermal monitoring data, these
columns were omitted from Table 3-2.
Page 97 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3023 Table 3-2. Summary of the Weight of the Scientific Evidence for Occupational Exposure Estimates by PES
OES
Inhalation Exposure
Dermal Exposure
Monitoring
Modeling
Weight of the
Scientific Evidence
Modeling
Weight of the
Scientific Evidence
Worker
# Data
Points
Data Quality
Rating"
Worker
Data
Quality
Rating"
Worker
Worker6
Worker
Textile Dye
~
14
H
X
N/A
Moderate
~
Moderate
Antifreeze
X
N/A
N/A
~
H
Moderate
~
Moderate
Surface Cleaner
~
49
H
X
N/A
Moderate to Robust
~
Moderate
Dish Soap
~
29
M
X
N/A
Slight to Moderate
~
Moderate
Dishwasher Detergent
~
29
M
X
N/A
Slight to Moderate
~
Moderate
Laundry Detergent
(Industrial and
Institutional)
X
N/A
N/A
~
M
Moderate
~
Moderate
Paint and Floor
Lacquer
~
17
H
X
N/A
Moderate
~
Moderate
PET Byproduct
~
11
H
X
N/A
Moderate
~
Moderate
Ethoxylation Process
Byproduct
~
1
H
X
N/A
Slight to Moderate
~
Moderate
Hydraulic Fracturing
X
N/A
N/A
~
M
Moderate to Robust
~
Moderate
"Data quality ratings of modeling approaches are based on the GS/ESD that was used in tandem with Monte Carlo modeling.
b Data quality ratings are not applicable for the dermal modeling approach because this modeling was conducted with an already-developed EPA model.
3024
Page 98 of 484
-------
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3.1.2.4 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Occupational Exposure Assessment
3.1.2.4.1 Number of Workers
There are uncertainties surrounding the estimated number of workers potentially exposed to 1,4-dioxane.
First, BLS employment data for each industry/occupation combination are only available at the 3-, 4-, or
5-digit NAICS level, rather than at the full 6-digit NAICS level. This lack of specificity 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 likely to use 1,4-dioxane for the assessed applications. EPA
addressed this issue by refining the OES 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 1,4-dioxane exposure differs from the overall distribution of
workers in each NAICS, then this approach will result in inaccuracy. The effects of this uncertainty on
the number of worker estimates are unknown, as the uncertainties may result in either over or
underestimation of the estimates depending on the actual distribution.
Second, EPA's determinations of industries (represented by NAICS codes) and occupations (represented
by SOC codes) that are associated with the OES assessed in this report are based on EPA's
understanding of how 1,4-dioxane is used in each industry. The designations of which industries and
occupations have potential exposures is a matter of professional judgement; therefore, the possibility
exists for the erroneous inclusion or exclusion of some industries or occupations. This may result in
inaccuracy but would be unlikely to systematically either overestimate or underestimate the count of
exposed workers.
3.1.2.4.2 Analysis of Inhalation Exposure Monitoring Data
The principal limitation of the monitoring data is the uncertainty in the representativeness of the data
due to some scenarios having limited exposure monitoring data in literature. Where few data are
available, the assessed exposure levels are unlikely to be representative of worker exposure across the
entire job category or industry. This may particularly be the case when monitoring data were available
for only one site. Differences in work practices and engineering controls across sites can introduce
variability and limit the representativeness of monitoring data. Age of the monitoring data can also
introduce uncertainty due to differences in workplace practices and equipment used at the time the
monitoring data were collected compared those currently in use. Therefore, older data may overestimate
or underestimate exposures, depending on these differences. The effects of these uncertainties on the
occupational exposure assessment are unknown, as the uncertainties may result in either overestimation
or underestimation of exposures depending on the actual distribution of 1,4-dioxane air concentrations
and the variability of work practices among different sites.
In some scenarios where monitoring data were available, EPA did not find sufficient data to determine
complete statistical distributions. Ideally, EPA will present 50th and 95th percentiles for each exposed
population. In the absence of percentile data for monitoring, the mean or midpoint of the range may
serve as a substitute for the 50th percentile of the actual distributions. Similarly, the highest value of a
range may serve as a substitute for the 95th percentile of the actual distribution. However, these
substitutes are uncertain. The effects of these substitutes on the occupational exposure assessment are
unknown, as the substitutes may result in either overestimation or underestimation of exposures
depending on the actual distribution.
Page 99 of 484
-------
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3.1.2.4.3 Modeled Inhalation Exposures
EPA addressed variability in inhalation models by identifying key model parameters to apply a
statistical distribution that mathematically defines the parameter's variability. EPA defined statistical
distributions for parameters using documented statistical variations where available. Where the
statistical variation was unknown, assumptions were made to estimate the parameter distribution using
available literature data, such as GSs and ESDs. However, there is uncertainty as to the
representativeness of the parameter distributions with respect to the modeled scenario because the data
are often not specific to sites that use 1,4-dioxane. In general, the effects of these uncertainties on the
exposure estimates are unknown, as the uncertainties may result in either overestimation or
underestimation on exposures depending on the actual distributions of each of the model input
parameters.
There is also uncertainty as to whether the model equations generate results that represent actual
workplace air concentrations. Some activity-based modeling does not account for exposures from other
activities. Another uncertainty is lack of consideration for engineering controls. The GS/ESDs assume
that all activities occur without any engineering controls or PPE, and in an open-system environment
where vapor and particulates freely escape and can be inhaled. Actual exposures may be less than
estimated depending on engineering control and PPE use.
A strength of the assessment is the variation of the model input parameters as opposed to using a single
static value. This parameter variation increases the likelihood of true occupational inhalation exposures
falling within the range of modeled estimates. An additional strength is that all data that EPA used to
inform the modeling parameter distributions have overall data quality determinations of either high or
medium from EPA's systematic review process.
3.1.2.4.4 Modeled Dermal Exposures
The Dermal Exposure to Volatile Liquids Model used to estimate dermal exposure to 1,4-dioxane in
occupational settings assumes a fixed fractional absorption of the applied dose; however, fractional
absorption may be dependent on skin loading conditions. The model also assumes a single exposure
event per day based on existing framework of the EPA/OPPT 2-HandDermal Exposure to Liquids
Model and does not address variability in exposure duration and frequency. Additionally, dermal
exposures to 1,4-dioxane vapor that may penetrate clothing and the potential for associated direct skin
contact with clothing saturated with 1,4-dioxane vapor are not included in quantifying exposures, which
could potentially result in underestimates of exposures. Although the extent of saturation of clothing
with 1,4-dioxane vapors is unknown, it is expected to be minimal given the low concentrations of 1,4-
dioxane in formulations for the conditions of use in the draft supplement.
A strength of the dermal assessment approach is the estimation of two different fractional absorption
values specific to industrial and commercial use settings as opposed to applying only one fractional
absorption value to both settings.
Page 100 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3109 3.2 General Population Exposures
3110
1,4-Dioxane - General Population Exposures (Section 3.2):
Key Points
• EPA estimated oral and inhalation exposures to the concentrations of 1,4-dioxane in drinking
water and air estimated in Section 2.3 using equations and exposure factors described in
Appendix G.l and Appendix J.2.
• EPA estimated general population exposures to 1,4-dioxane in air and water with a particular
focus on populations that may be highly exposed
0 Fenceline communities. For exposures through air, EPA considered potential
exposures for communities within 10 km of a release site. For drinking water, EPA
considered potential exposures for communities relying on drinking water collected
downstream of release sites.
0 Lifestage. For drinking water, EPA evaluated lifestage-specific exposures for adults,
formula-fed infants, and children. For air exposures, the impacts of lifestage
differences were not able to be adequately quantified and so the air concentrations are
used for all lifestages.
0 High-end exposure estimates. EPA evaluated exposures based on high-end exposure
scenarios (e.g., air exposures include a range of modeled concentration predictions
[low-end, central tendency, and high-end]), although only high-end model predictions
of air concentrations are presented in this section).
3111
3112 General population exposures occur when 1,4-dioxane is released into the environment and the media is
3113 then a pathway for exposure. Figure 3-1 below provides a graphic representation of where and in which
3114 media 1,4-dioxane may be found and the corresponding route of exposure.
3115
Page 101 of 484
-------
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Ambient Air
Inhalation
Landfills
(Industrial or
Muncipal)
Drinking
Water
Oral
Drinking
Water
Treatment
Wastewater
Facility
>undw£
Water
Recreation
Oral, Dermal
Figure 3-1. Potential Human Exposure Pathways to 1,4-Dioxane for the General Population"
" The diagram presents the media (white text boxes) and routes of exposure (italics for oral, inhalation, or dermal)
for the general population. Sources of drinking water from surface or water pipes is depicted w ith grey arrows.
3.2.1 Approach and Methodology
EPA used modeled concentrations of 1,4-dioxane in air, surface water, and groundwater estimated in
Section 2.3 to estimate acute and chronic general population exposures that could result from contact
with environmental media. These acute and chronic exposure estimates are used to evaluate cancer and
non-cancer risk described in Section 5. To estimate oral exposures to 1,4-dioxane in drinking water
(including groundwater used as drinking water), EPA used equations and exposure factors described in
Appendix G.l. To estimate inhalation exposures from 1,4-dioxane in air, EPA used equations and
exposure factors described in Appendix J.2. Where possible, available monitored data within these
environmental media were used to provide context for modeled results.
To estimate potential acute and chronic exposures through drinking water EPA calculated Acute Dose
Rates (ADR) and Average Daily Doses (ADD) for adults, formula-fed infants, and children. To estimate
lifetime exposures through drinking water, EPA calculated a Lifetime Average Daily Dose (LADD)
based on 33 years1' of exposure starting from birth or 33 years of exposure as an adult, averaged over a
78-year lifetime. Assumptions about drinking water intake and body weight for each age group were
based on information in the Exposure Factors Handbook. EPA calculated ADs, ADDs, and LADDs
based on the drinking water concentrations estimated under a range of conditions in Section 2.3.1.3.
Details of these calculations are presented in Appendix I and 1,4-Dioxane Supplemental Information
File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Release to Surface Water fr om
Individual Facilities (U.S. EPA. 2023h).
Thirty-three years is the 95th percentile residential occupancy period (U.S. EPA Exposure Factors Handbook. Chapter 16,
Table 16-5).
Page 102 of 484
-------
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
To estimate potential acute and chronic exposures through air, EPA calculated ADs and ADCs based on
modeled air concentrations described in Section 2.3.3 To estimate potential lifetime exposures, EPA
calculated LADCs based on 33 years of exposure. Methods adequate to quantify the impact of lifestage
differences on 1,4-dioxane exposure are not available (see Section 4.3) and air concentration is used as
the exposure metric for all lifestages per EPA guidance (I v << \ i Mb). Specific equations,
inputs and assumptions are described in detail in Appendix IJ.2.
3.2.2 Drinking Water Exposure Assessment
EPA assessed general population drinking water exposures that could result from surface water or
groundwater used as drinking water. Exposures estimates presented below are based on surface water
concentrations modeled in Section 2.3.1 or groundwater concentrations modeled in Section 2.3.2.
Exposure estimates presented throughout this section focus on adults and formula-fed infants because
these are lifestages with the greatest drinking water intake relative to body weights and therefore the
greatest potential exposures.
3.2.2.1 Surface Water Exposure Assessment
EPA evaluated acute, chronic, and lifetime general population, exposures to 1,4-dioxane that may be
present in drinking water based on modeled surface water concentrations estimated in Section 2.3.1. For
drinking water exposures in this assessment, EPA focused on exposures in fenceline communities,
defined in this context as members of the general population who rely on drinking water from water
bodies receiving 1,4-dioxane releases from any industrial or DTD source.
Drinking water exposures were evaluated using a series of parallel analyses that provide information
about the individual contributions of specific COUs as well as information about aggregate exposures
that could result from multiple sources releasing to the same water body.
3.2.2.1.1 Exposures from Individual Facility Releases
To evaluate the individual contributions of releases associated with specific industrial and commercial
COUs to general population exposures, EPA calculated ADRs, ADDs and LADDs based on modeled
water concentrations estimated in Section 2.3.1.3.1 (Table 3-3). A total of 125 release scenarios were
evaluated based on water concentrations estimated for annual releases that occur over a single day (a
peak exposure scenario), over 30 days of release, or over 250 to 365 days of release. Exposure estimates
are presented for both adults and formula-fed infants because these are lifestages with greatest drinking
water intake relative to body weights and therefore greatest exposures. ADRs based on a single day
release scenario range from 6.Ox 10~8 to 3,730 mg/kg for adults and 2.1 /10 1 to 1.3x 104 mg/kg for
infants. ADDs range from 1.1/10 " to 0.5 mg/kg/day for adults and 2,7/10 " to 1.3 mg/kg/day for
infants and are not influenced by the days of release. LADDS range from 4.2x 10~12 to 0.2 mg/kg/day for
adults exposed for 33 years and 3,5x10 13 to 1,6x 10 2 mg/kg/day for infants exposed for 1 year.
Complete exposure calculations are available in 1,4-Dioxane Supplemental Information File: Drinking
Water Exposure and Risk Estimates for 1,4-Dioxane Release to Surface Water from Individual Facilities
(U.S. EPA. 2023h\
Page 103 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 3-3. Adult and Infant Exposures Estimated from Facility-Specific Releases
OES
No. of
Releases
Modeled
Age Group'1
ADR (mg/kg-dav)
ADD (mg/kg-dav)
LADD (mg/kg-day)
Min
Exposure'
Mean
Exposure''
Max
Exposure1'
Min
Exposure'
Mean
Exposure''
Max
Exposure1'
Min
Exposure'
Mean
Exposure''
Max
Exposure1'
Disposal
25
Adult (21+
years)
6.03E-07
2.60E01
3.83E02
4.51E-10
1.44E-02
2.21E-01
1.91E—10
6.08E-03
9.36E-02
Infant (birth to
<1 year)
2.11E-06
9.11E01
1.34E03
1.15E-09
3.67E-02
5.65E-01
1.48E-11
4.71E-04
7.25E-03
Ethoxylation
Byproduct
8
Adult (21+
years)
2.17E-07
1.04E02
8.31E02
9.06E-11
3.66E-02
2.93E-01
3.83E-11
1.55E-02
1.24E-01
Infant (birth to
<1 year)
7.61E-07
3.65E02
2.92E03
2.31E-10
9.36E-02
7.48E-01
2.97E-12
1.20E-03
9.59E-03
Functional
Fluids (Open-
System)
6
Adult (21+
years)
5.59E-04
6.33E-02
1.92E-01
1.83E-07
2.23E-05
6.66E-05
7.73E-08
9.44E-06
2.82E-05
Infant (birth to
<1 year)
1.96E-03
2.22E-01
6.75E-01
4.67E-07
5.70E-05
1.70E-04
5.99E-09
7.30E-07
2.18E-06
Import and
Repackaging
12
Adult (21+
years)
4.35E-04
3.28E02
3.73E03
1.32E-07
3.05E-02
2.23E-01
5.59E-08
1.29E-02
9.43E-02
Infant (birth to
<1 year)
1.53E-03
1.15E03
1.31E04
3.38E-07
7.78E-02
5.69E-01
4.33E-09
9.97E-04
7.30E-03
Industrial
Uses
31
Adult (21+
years)
5.34E-07
2.05E01
1.87E02
1.97E-10
1.36E-02
1.55E-01
8.3 IE—11
5.77E-03
6.56E-02
Infant (birth to
<1 year)
1.87E-06
7.21E01
6.55E02
5.02E-10
3.48E-02
3.96E-01
6.44E-12
4.47E-04
5.08E-03
Manufacture
2
Adult (21+
years)
3.35E00
6.56E01
1.28E02
2.50E-03
4.91E-02
9.57E-02
1.06E-03
2.08E-02
4.05E-02
Infant (birth to
<1 year)
1.17E01
2.30E02
4.48E02
6.40E-03
1.25E-01
2.44E-01
8.20E-05
1.61E-03
3.13E-03
PET
Manufacturing
19
Adult (21+
years)
1.11 E—04
4.32E01
6.67E02
3.86E-08
3.16E-02
5.00E-01
1.63E-08
1.34E-02
2.11 E—01
Infant (birth to
<1 year)
3.91E-04
1.52E02
2.34E03
9.86E-08
8.07E-02
1.28E00
1.26E-09
1.04E-03
1.64E-02
Printing Inks
1
Adult (21+
years)
8.26E-02
8.26E-02
8.26E-02
6.18E-05
6.18E-05
6.18E-05
2.62E-05
2.62E-05
2.62E-05
Infant (birth to
<1 year)
2.90E-01
2.90E-01
2.90E-01
1.58E-04
1.58E-04
1.58E-04
2.02E-06
2.02E-06
2.02E-06
Remediation
16
Adult (21+
years)
6.04E-08
7.36E-02
7.19E-01
1.07E-11
4.56E-05
4.12E-04
4.51E-12
1.93E-05
1.74E-04
Infant (birth to
<1 year)
2.12E-07
2.58E-01
2.52E00
2.72E-11
1.17E-04
1.05E-03
3.49E-13
1.49E-06
1.35E-05
Page 104 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
No. of
Releases
Modeled
Age Group*
ADR (mg/kg-dav)
ADD (mg/kg-dav)
LADD (mg/kg-day)
Min
Exposure1'
Mean
Exposure''
Max
Exposure1'
Min
Exposure1'
Mean
Exposure''
Max
Exposure1'
Min
Exposure1'
Mean
Exposure''
Max
Exposure"
Overall
120
Adult (21+
years)
6.04E-08
5.84E01
3.73E03
1.07E-11
1.78E-02
5.00E-01
4.51E-12
7.55E-03
2.11E-01
Infant (birth to
<1 year)
2.12E-07
2.05E02
1.31E04
2.72E-11
4.56E-02
1.28E00
3.49E-13
5.84E-04
1.64E-02
Adult LADDs presented in this table were used to derive cancer risk estimates presented in Figure 5-1, Figure 5-2 and Figure 5-4.
ADRs presented here are calculated based on the assumption that all releases could occur on a single day of release (peak exposure scenario); ADDs and LADDs are based on
chronic exposure scenarios and are the same regardless of the number of days of release assumed. LADDs for adults are based on 33 years of exposure averaged over a 78-year
lifetime while LADDs for infant-specific exposures are based on 1 year of exposure averaged over a 78-year lifetime.
"Adult refers to 21+years; infant refers to birth to <1 year.
4 These COUs are added since the 2020 RE was published.
c The minimum exposure for the identified days of release, within the identified OES, and for the identified age group.
dThe arithmetic mean exposure for the identified days of release, within the identified OES, and for the identified age group.
e The maximum exposure for the identified days of release, within the identified OES, and for the identified age group.
3180
Page 105 of 484
-------
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3.2.2.1.2 Exposures from Down-the-Drain Releases
To evaluate the potential contribution of DTD consumer and commercial releases, EPA calculated
ADRs, ADDs, and LADDs using modeled water concentrations estimated as described in Section
2.3.1.2.2. Water concentrations of 1,4-dioxane resulting from DTD releases depend on the population
size (an indicator of the number of people using products and contributing the releases) and the stream
flows of the receiving water bodies. Therefore, the adult LADDs presented in Table 3-4 are based on the
range of water concentrations estimated by Monte Carlo modeling of DTD release scenarios with
varying population size and stream flows. LADDs range from 1.7/10 10 to 5,1 / ] 0 4 mg/kg/day for
adults exposed for 33 years. Complete exposure calculations for adults and infants are available in 1,4-
Dioxane Supplemental Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane
Surface Water Concentrations Predicted with Probabilistic Modeling (U.S. EPA. 20230.
Table 3-4. Adult LADD Exposures (mg/kg/day) Estimated from 1,4-Dioxane DTD Consumer and
Commercial Releases
Population Contributing to DTD Releases
100
1,000
10,000
100,000
1,000,000
Stream Flow
(cfs)
100
5.1E-08
5.1E-07
5.1E-06
5.1E-05
5.1E-04
300
1.7E-08
1.7E-07
1.7E-06
1.7E-05
1.7E-04
1,000
5.1E-09
5.1E-08
5.1E-07
5.1E-06
5.1E-05
3,000
1.7E-09
1.7E-08
1.7E-07
1.7E-06
1.7E-05
10,000
5.1E-10
5.1E-09
5.1E-08
5.1E-07
5.1E-06
30,000
1.7E-10
1.7E-09
1.7E-08
1.7E-07
1.7E-06
The frequencies of each of these combinations of population size and flow rate are presented Table 2-10. Adult
LADDs presented in this table were used to derive the cancer risk estimates presented in Table 5-4.
LADDs for adults are based on 33 years of exposure averaged over a 78-year lifetime.
3.2.2.1.3 Disposal of Hydraulic Fracturing Produced Waters
To evaluate the potential contribution of disposal of hydraulic fracturing produced waters to surface
water, EPA calculated ADRs, ADDs, and LADDs using the range of modeled water concentrations
estimated in Section 2.3.1.2.2. (Table 3-5). Water concentrations of 1,4-dioxane resulting from disposal
of hydraulic fracturing produced water vary substantially across sites. The estimated exposures
presented here are based on the range of water concentrations estimated by Monte Carlo modeling for a
range of site-specific factors. For this range of estimated surface water concentrations, Adult ADRs
range from 2.45xl0~16 to 7.96xl0~3 mg/kg and adult ADDs range from 6.69xl0~17 to 2.18><10~3
mg/kg/day. LADDs for adults exposed over 33 years over a 78-year lifetime range from 2,8x10 17 to
9.2xl0~4 mg/kg/day. Complete exposure calculations for adults and infants are available in 1,4-Dioxane
Supplemental Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Surface
Water Concentrations Predicted with Probabilistic Modeling (U.S. EPA. 2023i).
Page 106 of 484
-------
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 3-5. Adult ADR, ADD, and LADD Exposures Estimated from Disposal of Hydraulic
Fracturing Produced Waters to Surface Water
Monte Carlo
Distribution
Adult Acute Dose
Rate (mg/kg)
Adult Average Daily
Dose (mg/kg/day)
Adult Lifetime Average
Daily Dose (mg/kg/day)
Maximum
7.96E-03
2.18E-03
9.2E-04
99th percentile
3.13E-04
8.56E-05
3.6E-05
95th percentile
1.05E-04
2.86E-05
1.2E-05
Median
2.57E-06
7.03E-07
3.0E-07
5th percentile
9.31E-09
2.54E-09
1.1E-09
Minimum
2.45E-16
6.69E-17
2.8E-17
Adult LADDs presented in this table were used to derive cancer risk estimates presented in Table 5-5.
3.2.2.1.4 Aggregate Exposure
Because multiple sources of 1,4-dioxane contribute to surface water and drinking water concentrations,
EPA also estimated aggregate general population exposures that could occur because of combined
contributions from DTD releases from consumer and commercial uses, upstream sources, and direct and
indirect industrial releases. EPA calculated ADRs, ADDs, and LADDs based on modeled water
concentrations estimated in Section 2.3.1.3.4 using probabilistic modeling of aggregate 1,4-dioxane
surface water concentrations that could occur downstream of industrial release sites for each COU.
LADDs estimated for adults exposed over 33 years over a 78-year lifetime range from 8.07x 10~7 to
7.4x10 3 mg/kg/day based on median modeled water concentrations across COUs (Table 3-6). Complete
exposure calculations for adults and infants are available in 1,4-Dioxane Supplemental Information File:
Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Surface Water Concentrations Predicted
with Probabilistic Modeling ( Z0231).
Table 3-6. Adult LADD Exposures from Aggregate Concentrations Estimated Downstream of
Release Sites (Including DTD Releases and Direct and Indirect Industrial Releases)
COU
LADDs (mg/kg/day) Based on Modeled Aggregate Surface Water Concentrations
Estimated across the Monte Carlo Distribution
Min
5th
Percentile
25th
Percentile
Median
75th
Percentile
95th
Percentile
Max
Disposal
6.00E-09
6.82E-07
1.63E-06
3.93E-06
9.00E-06
9.64E-04
1.21E-01
Ethoxylation
byproduct
7.17E-09
3.72E-07
1.05E-06
1.93E-06
8.11E-06
1.98E-02
2.63E-01
Functional
Fluids (Open-
System)
5.62E-10
2.58E-07
4.99E-07
8.91E-07
7.20E-06
4.13E-05
6.22E-05
Import and
Repackaging
1.82E-08
4.21E-07
2.10E-04
1.03E-03
4.53E-03
1.34E-02
1.18E00
Industrial Uses
4.61E-10
1.65E-07
3.90E-07
8.07E-07
4.66E-05
4.90E-03
5.91E-02
Page 107 of 484
-------
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
cou
LADDs (mg/kg/day) Based on Modeled Aggregate Surface Water Concentrations
Estimated across the Monte Carlo Distribution
Min
5th
Percentile
25th
Percentile
Median
75th
Percentile
95th
Percentile
Max
Manufacture
3.51E-04
1.06E-03
2.29E-03
7.40E-03
1.75E-02
4.04E-02
4.04E-02
PET
Manufacturing
1.73E-08
4.54E-07
1.48E-06
1.43E-05
6.06E-04
2.53E-02
2.11E-01
Printing Inks
4.80E-07
1.01E-06
1.27E-05
2.04E-05
2.38E-05
2.66E-05
2.71E-05
Remediation
2.48E-09
2.74E-07
6.29E-07
1.27E-06
3.10E-05
9.61E-05
2.29E-04
Adult LADDs presented in this table were used to derive cancer risk estimates presented in Figure 5-5.
Percentiles reflect concentrations estimated at various points in the Monte Carlo distribution.
3.2.2.2 Groundwater Exposure Assessment
EPA evaluated general population exposures that could occur from disposals of 1,4-dioxane that
contaminate groundwater used as a primary source of drinking water. To estimate chronic exposures
through this drinking water pathway, EPA calculated ADDs and LADDs for adults and formula-fed
infants based on modeled groundwater concentrations of 1,4-dioxane estimated in Section 2.3.2. EPA
did not evaluate acute exposures because methods used to estimate groundwater concentrations provide
an indication of potential concentrations occuring over many years, rather than peak concentrations.
3.2.2.2.1 Disposal to Landfills
To evaluate general population exposure, EPA calculated ADDs and LADDs based on modeled
groundwater concentrations estimated in Section 2.3.2.3. Potential groundwater concentrations resulting
from disposal of 1,4-dioxane to municipal solid waste landfills vary across landfill loading rates and
concentrations of 1,4-dioxane in leachate. Estimated exposures presented here are therefore based on the
range of groundwater concentrations estimated under varying landfill conditions. Table 3-7 summarizes
LADD exposure estimates estimated for 33 years of exposure as an adult. Under the range of landfill
scenarios considered, adult LADDs range from 2.5><106 to 2,4/ 10 2 mg/kg/day. The highest LADDs
occur when leachate concentrations are above 100 mg/L and loading rates are above 10,000 lb. The
complete set of exposure estimates for adults and infants relying on groundwater as a primary drinking
water source are presented in 1,4-Dioxane Supplemental Information File: Drinking Water Exposure
and Risk Estimates for 1,4-Dioxane Land Releases to Landfills ( )23f).
Page 108 of 484
-------
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 3-7. Adult LADD Exposures Estimated from Groundwater Contamination from Landfills
under Varying Landfill Conditions
Loading Rate (lb)
Leachate
Concentration
(mg/L)
0.1
1
10
100
1,000
10,000
100,000
1,000,000
0.0001
2.5E-16
2.4E-15
3.0E-14
2.9E-13
2.7E-12
2.6E-11
2.5E-10
2.4E-09
0.001
2.5E-15
2.4E-14
3.0E-13
2.9E-12
2.7E-11
2.6E-10
2.5E-09
2.4E-08
0.01
2.5E-14
2.4E-13
3.0E-12
2.9E-11
2.7E-10
2.6E-09
2.5E-08
2.4E-07
0.1
2.5E-13
2.4E-12
3.0E-11
2.9E-10
2.7E-09
2.6E-08
2.5E-07
2.4E-06
1
2.5E-12
2.4E-11
3.0E-10
2.9E-09
2.7E-08
2.6E-07
2.5E-06
2.4E-05
10
2.5E-11
2.4E-10
3.0E-09
2.9E-08
2.7E-07
2.6E-06
2.5E-05
2.4E-04
100
2.5E-10
2.4E-09
3.0E-08
2.9E-07
2.7E-06
2.6E-05
2.5E-04
2.4E-03
1,000
2.5E-09
2.4E-08
3.0E-07
2.9E-06
2.7E-05
2.6E-04
2.5E-03
2.4E-02
10,000
2.5E-08
2.4E-07
3.0E-06
2.9E-05
2.7E-04
2.6E-03
2.5E-02
2.4E-01
Adult LADDs presented in this table were used to derive cancer risk estimates presented in Table 5-6.
3.2.2.2.2 Disposal of Hydraulic Fracturing Produced Waters
To evaluate general population exposure resulting from disposal of hydraulic fracturing produced waters
to groundwater, EPA calculated ADDs and LADDs estimated in Section 2.3.2.4 (Table 3-8). Potential
groundwater concentrations resulting from disposal of hydraulic fracturing produced waters vary
substantially across sites. Estimated exposures presented here are based on the range of groundwater
concentrations estimated through Monte Carlo modeling. Under the range of hydraulic fracturing
scenarios considered, adult LADDs range from 4,9/ 10 9 to 2,1 /10 4 mg/kg/day. The complete set of
exposure estimates for adults and infants relying on groundwater as a primary drinking water source are
presented in 1,4-Dioxcme Supplemental Information File: Drinking Water Exposure and Risk Estimates
for 1,4-Dioxane Land Releases to Surface Impoundments (U.S. EPA. 2023 g).
Table 3-8. Estimated Exposures Resulting from Groundwater Contamination from Disposal of
Monte Carlo
Distribution
Modeled
Groundwater
Concentration (mg/L)
Adult ADD
(mg/kg/day)
Adult LADD
(mg/kg/day)
Infant ADD
(mg/kg/day)
Max
1.9E-02
2.1E-04
8.8E-05
5.3E-04
99th
1.5E-02
1.7E-04
7.1E-05
4.3E-04
95th
1.5E-02
1.7E-04
7.1E-05
4.3E-04
Mean
7.1E-04
7.9E-06
3.3E-06
2.0E-05
50th
1.2E-04
1.3E-06
5.6E-07
3.4E-06
5th
1.2E-04
1.3E-06
5.6E-07
3.4E-06
Min
4.4E-07
4.9E-09
2.1E-09
1.2E-08
Adult LADDs presented in this table were used to derive cancer risk estimates presented in Table 5-7.
Page 109 of 484
-------
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3,2,3 Air Exposure Assessment
EPA evaluated acute, chronic and lifetime general population, exposures to 1,4-dioxane in air. This
analysis focuses on potential fenceline community exposures that may occur within 10 km of release
sites.
3.2.3.1 Industrial COUs Reported to TRI
To evaluate general population exposures from industrial fugitive and stack emissions, EPA calculated
ACs, ADCs, and LADCs based on modeled air concentrations estimated in Section 2.3.3. The LADCs
presented in Table 3-9 are based on the maximum 95th percentile air concentrations estimated for the
facilities within each COU. LADCs within 10 km of release types considered here range from 1.1x10 "
to 6.9x10 3 ppm. These lifetime exposure estimates are based on 33 years of exposure over a 78-year
lifetime and are relevant to all lifestages. The complete set of inhalation exposure estimates for fenceline
communities are presented in 1,4-Dioxane Supplemental Information File: Air Exposures and Risk
Estimates for Single Year Analysis ( 023 e).
Page 110 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3273 Table 3-9. Lifetime Average Dai y Concentrations Estimated within 10 km of 1,4-Dioxane Releases to Air
1,4-Dioxane OES
#
Facilities
LADCs Estimated within 5-10,000 m of Facilities with Greatest Exposures (ppm)
5 m
III in
30 m
60 m
Kill in
100 to
1,000 m
2,500 m
5,000 m
10,000 m
Disposal
15
1.8E-03
2.1E-03
7.6E-04
2.9E-04
1.3E-04
1.3E-05
8.0E-07
2.7E-07
8.8E-08
Dry Film Lubricant
8
6.8E-11
3.0E-09
2.2E-07
1.6E-06
2.7E-06
4.2E-07
1.2E-08
3.6E-09
1.6E-09
Ethoxylation
Byproduct
6
2.8E-03
5.8E-03
3.1E-03
1.3E-03
6.9E-04
1.6E-04
9.3E-06
3.8E-06
1.5E-06
Film Cement
1
5.3E-05
5.5E-05
1.9E-05
9.7E-06
5.3E-06
9.7E-07
5.8E-08
2.0E-08
6.4E-09
Functional Fluids
(Open-System)
2
5.4E-06
1.0E-05
4.4E-06
4.6E-06
7.7E-06
3.1E-06
2.9E-07
1.1E-07
3.7E-08
Import and
Repackaging
1
1.1E-11
2.4E-10
2.3E-08
1.8E-07
3.7E-07
1.4E-07
2.8E-08
1.7E-08
9.4E-09
Industrial Uses
12
1.8E-03
2.0E-03
6.5E-04
2.4E-04
1.2E-04
3.0E-05
3.7E-06
1.4E-06
4.8E-07
Laboratory Chemical
Use
1
8.7E-04
9.1E-04
3.1E-04
1.6E-04
8.7E-05
1.6E-05
9.6E-07
3.2E-07
1.1E-07
Manufacturing
1
3.7E-03
6.9E-03
3.3E-03
1.4E-03
6.7E-04
6.0E-05
3.5E-06
1.1E-06
3.4E-07
PET Manufacturing
13
3.4E-03
4.0E-03
1.5E-03
5.9E-04
2.7E-04
4.5E-05
8.8E-06
5.3E-06
2.8E-06
Spray Foam
Application
1
3.3E-07
3.6E-07
1.2E-07
6.4E-08
3.6E-08
6.6E-09
7.3E-10
2.7E-10
1.0E-10
LADCs are based on the maximum 95th percentile air concentration predictions for the facility in each COU with the greatest exposures. Adult LADCs
presented in this table were used to derive the cancer risk estimates presented in Table 5-8.
3274
3275
Page 111 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3276 3.2.3.2 Hydraulic Fracturing
3277 To evaluate general population exposures to fugitive emissions from hydraulic fracturing operations,
3278 EPA calculated ACs, ADCs, and LADCs based on modeled air concentrations estimated in Section
3279 2.3.3.2.4 under a range of different release scenarios and topographical conditions (Table 3-10). LADCs
3280 within 1,000 m of hydraulic fracturing operations range from 8.7/10 4 to 5.2 ppm. These lifetime
3281 exposure estimates are based on 33 years of exposure over a 78-year lifetime and are relevant to all
3282 lifestages. The complete set of inhalation exposure estimates from fugitive emissions of hydraulic
3283 fracturing operations are presented in 1,4-Dioxane Supplemental Information File: Air Exposure and
3284 Risk Estimates for 1,4-Dioxane Emissions from Hydraulic Fracturing Operations (U.S. EPA. 2023b).
Page 112 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3285 Table 3-10. Exposures from Fugitive Emissions Estimated within 1,000 in of Hydraulic Fracturing Operations"
Air Concentrations for 95th Percentile Modeled Releases
Air Concentrations for 50th Percentile Modeled Releases
Fugitive
(ppm)
(ppm)
Emissions
Exposure
High-End Modeled Air
Central Tendency (Mean)
High-End Modeled Air
Central Tendency (Mean)
Release
Duration
Concentrations
Modeled Air Concentrations
Concentrations
Modeled Air Concentrations
Scenario
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
AC
5.3E-02
2.4E-03
6.2E-03
4.2E-02
1.8E-03
4.8E-03
3.1E-03
1.4E-04
3.6E-04
2.4E-03
1.0E-04
2.7E-04
South
(Coastal)-
ADC
1.1E-02
4.8E-04
1.2E-03
8.3E-03
3.6E-04
9.4E-04
6.0E-04
2.7E-05
7.0E-05
4.7E-04
2.0E-05
5.4E-05
Rural-24
LADC
4.5E-03
2.0E-04
5.2E-04
3.5E-03
1.5E-04
4.0E-04
2.6E-04
1.2E-05
3.0E-05
2.0E-04
8.6E-06
2.3E-05
AC
4.1E-02
2.4E-03
5.5E-03
3.1E-02
1.5E-03
3.7E-03
2.3E-03
1.3E-04
3.1E-04
1.8E-03
8.3E-05
2.1E-04
West North
Central-
ADC
8.0E-03
4.6E-04
1.1E-03
6.1E-03
2.9E-04
7.3E-04
4.6E-04
2.7E-05
6.2E-05
3.5E-04
1.6E-05
4.2E-05
Rural-24
LADC
3.4E-03
2.0E-04
4.6E-04
2.6E-03
1.2E-04
3.1E-04
1.9E-04
1.1E-05
2.6E-05
1.5E-04
6.9E-06
1.8E-05
AC
2.6E-02
5.8E-04
1.8E-03
2.3E-02
5.0E-04
1.6E-03
1.5E-03
3.3E-05
1.0E-04
1.3E-03
2.9E-05
9.0E-05
South
(Coastal)-
ADC
5.1E-03
1.1E-04
3.5E-04
4.5E-03
9.9E-05
3.1E-04
2.9E-04
6.5E-06
2.0E-05
2.6E-04
5.7E-06
1.8E-05
Urban-24
LADC
2.2E-03
4.8E-05
1.5E-04
1.9E-03
4.2E-05
1.3E-04
1.2E-04
2.8E-06
8.5E-06
1.1E-04
2.4E-06
7.5E-06
AC
2.4E-02
6.2E-04
1.9E-03
1.9E-02
4.6E-04
1.4E-03
1.4E-03
3.6E-05
1.1E-04
1.1E-03
2.6E-05
8.1E-05
West North
Central-
ADC
4.8E-03
1.2E-04
3.7E-04
3.8E-03
9.0E-05
2.8E-04
2.7E-04
7.0E-06
2.1E-05
2.2E-04
5.2E-06
1.6E-05
Urban-24
LADC
2.0E-03
5.2E-05
1.6E-04
1.6E-03
3.8E-05
1.2E-04
1.2E-04
3.0E-06
8.9E-06
9.3E-05
2.2E-06
6.7E-06
AC
9.6E-03
8.8E-05
3.8E-04
8.3E-03
6.8E-05
3.2E-04
5.5E-04
5.1E-06
2.2E-05
4.8E-04
3.9E-06
1.8E-05
South
(Coastal)-
ADC
1.9E-03
1.7E-05
7.6E-05
1.6E-03
1.3E-05
6.2E-05
1.1E-04
1.0E-06
4.3E-06
9.4E-05
7.6E-07
3.6E-06
Rural-8
LADC
8.0E-04
7.4E-06
3.2E-05
7.0E-04
5.6E-06
2.6E-05
4.6E-05
4.2E-07
1.8E-06
4.0E-05
3.2E-07
1.5E-06
AC
2.0E-02
7.7E-04
1.9E-03
1.1E-02
2.1E-04
6.6E-04
6.1E-04
1.2E-05
3.8E-05
6.1E-04
1.2E-05
3.8E-05
West North
Central-
ADC
4.0E-03
1.5E-04
3.7E-04
2.1E-03
4.2E-05
1.3E-04
1.2E-04
2.4E-06
7.4E-06
1.2E-04
2.4E-06
7.4E-06
Rural-8
LADC
1.7E-03
6.4E-05
1.5E-04
8.9E-04
1.8E-05
5.5E-05
5.1E-05
1.0E-06
3.1E-06
5.1E-05
1.0E-06
3.1E-06
Page 113 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Fugitive
Emissions
Release
Scenario
Exposure
Duration
Air Concentrations for 95th Percentile Modeled Releases
(ppm)
Air Concentrations for 50th Percentile Modeled Releases
(ppm)
High-End Modeled Air
Concentrations
Central Tendency (Mean)
Modeled Air Concentrations
High-End Modeled Air
Concentrations
Central Tendency (Mean)
Modeled Air Concentrations
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
South
(Coastal)-
Urban-8
AC
8.7E-03
6.8E-05
3.2E-04
8.0E-03
6.0E-05
2.9E-04
4.6E-04
3.4E-06
1.7E-05
4.6E-04
3.4E-06
1.7E-05
ADC
1.7E-03
1.3E-05
6.4E-05
1.6E-03
1.2E-05
5.8E-05
9.0E-05
6.8E-07
3.3E-06
9.0E-05
6.8E-07
3.3E-06
LADC
7.3E-04
5.6E-06
2.7E-05
6.7E-04
5.0E-06
2.4E-05
3.8E-05
2.9E-07
1.4E-06
3.8E-05
2.9E-07
1.4E-06
West North
Central-
Urban-8
AC
1.5E-02
2.9E-04
9.4E-04
9.2E-03
1.2E-04
4.5E-04
5.3E-04
6.8E-06
2.6E-05
5.3E-04
6.8E-06
2.6E-05
ADC
2.9E-03
5.8E-05
1.8E-04
1.8E-03
2.4E-05
8.9E-05
1.0E-04
1.4E-06
5.1E-06
1.0E-04
1.4E-06
5.1E-06
LADC
1.2E-03
2.4E-05
7.8E-05
7.7E-04
1.0E-05
3.8E-05
4.4E-05
5.7E-07
2.2E-06
4.4E-05
5.7E-07
2.2E-06
" Lifetime Average Daily Concentrations (LADCs) presented in this table correspond to the cancer risk estimates presented in Table 5-7.
AC = Acute Concentration; ADC = Average Daily Concentration; LADC = Lifetime Average Daily Concentration
3286
Page 114 of 484
-------
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3.2.3.3 Industrial and Institutional Laundry Facilities
To evaluate exposures to emissions from industrial and institutional laundry facilities, EPA calculated
ACs, ADCs, and LADCs based on vapor and particular air concentrations estimated in Section 2.3.3.2.4.
High-end and central tendency air exposures estimated under the more conservative exposure scenario
evaluated (rural south coastal topography, assuming 24 hours of releases each day) are presented for
each type of laundry in Table 3-11. LADCs estimated within 1,000 m of laundry facilities operations
range from 8.7><10~4 to 2.4><10H5 ppm. These lifetime exposure estimates are based on 33 years of
exposure over a 78-year lifetime and are relevant to all lifestages. The complete set of inhalation
exposure estimates from fugitive emissions of commercial laundry facilities are presented in 1,4-
Dioxane Supplemental Information File: Air Exposures and Risk Estimates for Industrial Laundry (U.S.
I 23c).
Table 3-11. Exposures from Fugitive Emissions Estimated near Industrial and Institutional
Laundry Facilities"
Facility
Type
Detergent
and
Emission
Type
Exposure
Duration
Modeled Air Concentrations for Maximum Release Estimates
(ppm)
High-End
Central Tendency (Mean)
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
Industrial
laundry
Liquid-
Vapor
AC
6.9E-06
3.3E-07
8.2E-07
5.0E-06
2.1E-07
5.6E-07
ADC
5.4E-06
2.5E-07
6.3E-07
4.9E-06
2.1E-07
5.6E-07
LADC
2.3E-06
1.1E-07
2.7E-07
2.1E-06
9.0E-08
2.4E-07
Powder-
Vapor
AC
6.9E-06
3.3E-07
8.2E-07
4.9E-06
2.1E-07
5.6E-07
ADC
5.4E-06
2.5E-07
6.2E-07
4.9E-06
2.1E-07
5.6E-07
LADC
2.3E-06
1.1E-07
2.6E-07
2.1E-06
8.9E-08
2.3E-07
Powder-
PM10
AC
7.2E-06
1.6E-07
5.9E-07
5.1E-06
1.2E-07
4.2E-07
ADC
5.6E-06
1.3E-07
4.7E-07
5.0E-06
1.2E-07
4.2E-07
LADC
2.4E-06
5.5E-08
2.0E-07
2.1E-06
4.9E-08
1.8E-07
Powder-
PM2.5
AC
6.9E-06
3.1E-07
8.0E-07
4.9E-06
2.0E-07
5.5E-07
ADC
5.4E-06
2.4E-07
6.1E-07
4.9E-06
2.0E-07
5.4E-07
LADC
2.3E-06
1.0E-07
2.6E-07
2.1E-06
8.5E-08
2.3E-07
Institutional
laundry
Liquid-
Vapor
AC
3.6E-06
1.6E-07
4.1E-07
3.1E-06
1.3E-07
3.5E-07
ADC
3.4E-06
1.6E-07
4.0E-07
3.1E-06
1.3E-07
3.5E-07
LADC
1.4E-06
6.7E-08
1.7E-07
1.3E-06
5.7E-08
1.5E-07
Powder-
vapor
AC
1.1E-07
4.8E-09
1.2E-08
9.2E-08
4.0E-09
1.0E-08
ADC
1.0E-07
4.7E-09
1.2E-08
9.2E-08
3.9E-09
1.0E-08
LADC
4.2E-08
2.0E-09
4.9E-09
3.9E-08
1.7E-09
4.4E-09
Page 115 of 484
-------
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Detergent
Modeled Air Concentrations for Maximum Release Estimates
(ppm)
Facility
Type
and
Emission
Exposure
Duration
High-End
Central Tendency (Mean)
Type
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
AC
1.1E-07
2.5E-09
8.9E-09
9.4E-08
2.2E-09
7.9E-09
Powder-
PM10
ADC
1.0E-07
2.4E-09
8.7E-09
9.4E-08
2.2E-09
7.9E-09
LADC
4.4E-08
1.0E-09
3.7E-09
4.0E-08
9.2E-10
3.3E-09
AC
1.1E-07
4.6E-09
1.2E-08
9.2E-08
3.8E-09
1.0E-08
Powder-
PM2.5
ADC
1.0E-07
4.5E-09
1.1E-08
9.2E-08
3.8E-09
1.0E-08
LADC
4.2E-08
1.9E-09
4.8E-09
3.9E-08
1.6E-09
4.3E-09
" LADCs presented in this table correspond to the cancer risk estimates presented in Table 5-10.
AC = Acute Concentration; ADC = Average Daily Concentration; LADC = Lifetime Average Daily Concentration
3.3 Weight of the Scientific Evidence Conclusions
As described in the 2021 Draft Systematic Review Protocol ( ), the weight of the
scientific evidence supporting exposure assessments is evaluated based on the availability and strength
of exposure scenarios and exposure factors, measured and monitored data, estimation methodology and
model input data, and, if appropriate, comparisons of estimated and measured exposures. The strength of
each of these evidence streams can be ranked as either robust, moderate, slight, or indeterminate. For
each component of this exposure assessment, EPA evaluated the weight of the scientific evidence for
individual evidence streams and then used that information to evaluate the overall weight of evidence
supporting each set of exposure estimates. General considerations for evaluating the strength of evidence
for each evidence stream are summarized in TableApx C-5. Specific examples of how these
considerations are applied to overall weight of evidence conclusions are provided in Table Apx C-6.
Overall confidence descriptions of high, medium, or low are assigned to the exposure assessment based
on the strength of the underlying scientific evidence. When the assessment is supported by robust
evidence, overall confidence in the exposure assessment is high; when supported by moderate evidence,
overall confidence is medium; when supported by slight evidence, overall confidence is low.
3.3.1 Occupational Exposures
The weight of the scientific evidence for occupational exposure estimates is determined by several
different evidence streams, including:
• Evidence supporting the exposure scenarios (Section 3.1.1 and Appendix F.4)
• The quality and representativeness of available monitoring data (Appendix F.4)
• Evidence supporting modeling approaches (Section 3.1.1 and Appendix F.4)
• Evidence supporting model input data (Appendix F.4)
3.3.1.1 Inhalation Exposure
Occupational inhalation exposure estimates are supported by moderate to robust evidence (see Appendix
F.6).
• Exposure Scenarios and Exposure Factors. The exposure scenarios and exposure factors
underlying the inhalation assessment are supported by moderate to robust evidence.
Page 116 of 484
-------
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Occupational inhalation exposure scenarios and exposure factors, including duration of exposure,
body weight, and breathing rate, were informed by sources of data with medium to high data
quality ratings, increasing the strength of evidence. For most OESs/COUs, EPA used
information directly relevant to the evaluated exposure scenarios; however, for some
OESs/COUs, EPA used information from surrogate scenarios, decreasing the strength of
evidence for those scenarios. Additionally, there is uncertainty in the extent to which the entire
population of workers within an OES/COU are represented by the available data.
• Measured and Monitored Data. Measured/monitored data are supported by moderate to robust
evidence. EPA used sources of data such as OSHA and NIOSH, which have medium to high data
quality ratings, increasing the strength of the evidence. For the OESs/COUs with available
monitoring data, the data was directly applicable to the assessed exposure scenario, as opposed to
from a surrogate exposure scenario. However, the available monitoring data was limited to a
single source for each OES/COU and often consisted of a small or dated dataset. Additionally,
these data often only included one or a limited number of sites at which the data were measured,
decreasing the strength of evidence for those OESs/COUs.
• Modeling Methodologies. The modeling methodologies are supported by moderate to robust
evidence. Modeling was implemented to assess occupational inhalation exposures for three of
the OESs/COUs, using methodologies from GS/ESD that are generally well described. The
modeling incorporates Monte Carlo simulation to allow for variation in the model input data,
which increases the representativeness of the approach towards the true population of potentially
exposed workers and increases the strength of the evidence. However, EPA was unable to
develop distributions for all input parameters, increasing the uncertainty in the parameterization
and applicability.
• Model Input Data. Model input data are supported by moderate to robust evidence. For some
model input data, EPA used 1,4-dioxane-specific data from sources such as process information,
product concentration information, and FracFocus 3.0. For other model input parameters, generic
data from the GS/ESD used for the modeling methodology was used due to lack of 1,4-dioxane
data.
• Comparison of Modeled and Monitored Data. The comparison of modeled and measured
occupational inhalation exposures is not rated because no comparisons between modeled and
measured exposures were made.
Overall Confidence in Occupational Inhalation Exposure Estimates
The overall confidence in the occupational inhalation exposure estimates (Section 3.3.1.1) ranges from
low to high, depending on the OES/COU. Measured/monitored data are supported by moderate to robust
evidence. Additionally, the modeling methodologies and underlying model input data is supported by
moderate to robust evidence. However, there is uncertainty in the representativeness of the assessed
exposure scenarios towards all potential exposures for the given OES/COU, limitations in the amount
and age of monitoring data, and limitations in the modeling approaches towards 1,4-dioxane-specific use
within the OES/COU. Therefore, while the underlying data and methods used to estimate occupational
inhalation exposures is supported by moderate to robust evidence, the overall confidence of these
estimates is low to high depending on the OES/COU. OES/COU-specific discussions of the available
inhalation exposure data and overall confidence are presented in Appendix F.6.
3.3.1.2 Dermal Exposure
Occupational dermal exposure estimates are supported by slight to robust evidence (see Appendix F.3).
Page 117 of 484
-------
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• Exposure Scenarios and Exposure Factors. The exposure scenarios and exposure factors
underlying the dermal assessment are supported by moderate to robust evidence. Dermal
exposure scenarios were informed by process information and GS/ESD with medium to high
data quality ratings, increasing the strength of evidence. Exposure factors, including amount of
material on skin, surface area of skin exposed, and absorption of 1,4-dioxane through the skin,
were informed by literature sources, the ChemSTEER User Guide (x v ii \ i) for
standard exposure parameters, and a European model, which have medium to high data quality
ratings. EPA used information directly relevant to the evaluated exposure scenarios; however,
there is uncertainty in the extent to which the entire population of workers within an OES/COU
are represented by the available data.
• Measured and Monitored Data. No measured/monitored dermal exposure data were used in
the occupational dermal exposure assessment. EPA did use measured data on 1,4-dioxane
concentrations in various products from process information and other literature sources, which
have medium to high data quality ratings, depending on the data source.
• Modeling Methodologies. The modeling methodologies are supported by moderate evidence.
EPA used the EPA Dermal Exposure to Volatile Liquids to calculate the dermal retained dose for
each OES/COU. This model modifies the EPA/OPPT 2-HandDermal Exposure to Liquids
Model by incorporating a "fraction absorbed (fabs)" parameter to account for the evaporation of
volatile chemicals and a "protection factor (PF)" to account for glove use. These modifications
improve the modeling methodology and allow EPA to differentiate dermal exposures between
commercial and industrial settings by varying the absorption and dermal protection factors.
However, the modeling approach is still limited by the low variability for different worker
activities/exposure scenarios.
• Model Input Data. Model input data are supported by slight to moderate evidence. As discussed
above, model parameters were informed by sources with medium to high data quality ratings.
However, a limitation is that some of the model input data was generic and not specific to 1,4-
dioxane OES/COU.
• Comparison of Modeled and Monitored Data. All occupational dermal exposures were
modeled, and no measured dermal exposures were used in this assessment, therefore there is no
comparison.
Overall Confidence in Occupational Dermal Exposure Estimates
The overall confidence in the occupational dermal exposure estimates (Section 3.1.2.2) is medium for all
OES/COU because the same modeling approach was used for all OES/COU. The modeling
methodology is supported by moderate evidence, with model input parameters from literature sources, a
European model, standard defaults from the ChemSTEER User Guide ( i), and 1,4-
dioxane product concentration data from process information. These sources range from slight to robust,
depending on factors such as age and applicability to OES/COU. The modeling is limited by the use of
standard input parameters that are not specific to 1,4-dioxane and a lack of variability in dermal
exposure for different worker activities. Therefore, EPA's overall confidence in the occupational dermal
exposure estimates is medium.
3.3.2 Drinking Water
3.3.2.1 Drinking Water Exposure Estimates Based on Surface Water Concentrations
The weight of evidence for drinking water exposure estimates is determined by several different
evidence streams, including:
Page 118 of 484
-------
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• Evidence supporting the general population exposure scenarios (Section 3.2.1)
• The quality and representativeness of available surface water and drinking water monitoring
data (Section 2.3.1.1)
• Evidence supporting modeling approaches (Section 2.3.1.3 and Appendix G.2)
• Evidence supporting release data used as model input data (Section 2.2 and Appendix E.3)
• Concordance between modeled and monitored water concentrations (Section 2.3.1.4)
As described in Section 2.3.1, multiple approaches were used to predict surface water concentrations
resulting from several sources. These included the evaluation of facility-specific releases, down the
drain releases to surface water, hydraulic fracturing releases and aggregation of surface water releases.
The associated strengths, limitations and confidence in these estimated environmental concentrations are
described in Section 2.3.1.4. The general population drinking water exposure scenarios and exposure
factors used to estimate exposures that could result from estimated water concentrations are described in
Section 3.2.
Drinking water exposure estimates based on modeled surface water concentrations are supported by
overall moderate to robust evidence, with the strength of the evidence varying across analysis
approaches and COUs/OESs.
• Exposure Scenarios and Exposure Factors. The exposure scenarios and exposure factors
underlying all drinking water exposure estimates are supported by moderate to robust evidence.
Exposure factors for drinking water are based on robust data on drinking water intake rates and
body weights as derived from exposure factors from the EPA's Exposure Factors Handbook
( 2011). However, the drinking water exposure scenarios generally rely on the
assumption that little or no dilution occurs prior to drinking water intakes. That assumption may
not be representative of exposures at all locations. While there are locations where this
assumption is expected to the accurate, the extent of downstream dilution that occurs prior to
drinking water intakes is highly variable across locations. The proximity of facility releases to
actual drinking water intakes is evaluated in Section 2.3.1.2.4 and Appendix G.2.4. Uncertainties
related to downstream dilution decrease the overall strength of evidence for these exposure
scenarios. However, EPA has performed several analyses that calculate exposures and risks
under alternate assumptions about downstream dilution and illustrate the quantitative impact of
those assumptions (see Section 5.2.2.1.2), increasing the overall strength of evidence. Drinking
water exposure scenarios also rely on the data-driven assumption that 1,4-dioxane is not removed
through treatment. Moderate to robust data provide support for this assumption under many
treatment scenarios. These assumptions may over-estimate exposure for some locations, but
provide an overall distribution that is generally expected to be representative of exposure
scenarios.
• Measured and Monitored Data. The measured/monitored data are supported by moderate
evidence. The high number of monitoring data points for surface water and drinking water from
high quality sources in multiple locations over multiple years increases the strength of the
evidence from monitoring data. Monitoring data confirm that 1,4-dioxane is present in some
surface water and drinking water in some locations. However, evidence from monitoring data
may not be representative of all sites where 1,4-dioxane is released to surface water from TSCA
sources, decreasing the strength of evidence from monitoring data. The lack of temporal and/or
spatial concurrence between most monitoring data and reported release locations makes direct
comparison challenging for most locations. However, a limited number of sites with monitoring
data are co-located with sites where 1,4-dioxane releases are reported, supporting comparisons of
monitoring and modeled estimates that increase the overall strength of the evidence. In addition,
Page 119 of 484
-------
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
as described in Section 2.3.1.4, monitoring data for surface water directly downstream from
releases show concentrations multiple orders of magnitude greater than typical ambient surface
water concentrations, aligning with patterns of modeled results.
• Modeling Methodologies. The modeling methodologies are supported by moderate to robust
evidence.
o The methodology for deriving exposure estimates for facility releases is moderate and is
applicable to the populations included in the exposure scenarios. This approach makes
some conservative assumptions about flow rates and release frequency and amount.
Additionally, the modeling does not take into account downstream fate or transport, but
the physical chemical properties of 1,4-dioxane are expected to moderate the impact
these influences could have on the modeled instream concentrations. The model is
designed to estimate possible higher end water concentrations expected at specific
locations.
o The probabilistic methodology used for deriving exposure estimates for DTD releases,
hydraulic fracturing releases, and aggregate releases from all sources is robust. This
approach incorporates the full distribution of facility releases over multiple years and
corresponding instream flow rate data rather than relying on the most conservative model
inputs. It is designed to provide a nationally representative distribution of estimated water
concentrations under varying conditions.
• Model Input Data. Model input data are supported by slight to robust evidence, with the
strength of the evidence varying across individual COUs/OESs. The strength of evidence
supporting modeled water concentrations relies heavily on the quality of the facility or OES-
specific release data used as inputs for the model, including both the amount of release, location
of the release, and the corresponding flow in the receiving water body. A summary of sources of
flow and release data for facility release modeling is presented in Table 2-6. A more detailed
OES-specific discussion of the confidence in sources of release information is presented in
Appendix E.3.4.
o For overall distributions of industrial releases across sites, model input data are supported
by robust evidence. As illustrated in Section 5.2.2.1.2, EPA estimated exposures and risks
across the full distribution of facility releases both for the whole dataset and for a subset
of facilities with high quality reporting information. Comparison of these distributions
demonstrates that inclusion of locations relying on more limited release information had
limited impact on the overall distributions of exposures.
o For COUs/OESs that rely primarily on release data reported to TRI via Form R, or
reported to ICIS-NPDES via DMR, site-specific release estimates are supported by
moderate to robust evidence. As described in Appendix E.3.1, these release estimates are
based on release amounts reported by facilities. Most COUs/OESs are included in this
group.
o For COUs/OESs that rely primarily on release data reported to TRI via Form A, site-
specific release estimates are supported by moderate evidence. As described in Appendix
E.3.1, Form A simply indicates that releases are below the reporting thresholds and
specific release estimates require assumptions about amounts, locations, and media of
release. The Import and Repackaging OES releases used in this analysis are entirely
based on Form A reporting of releases, and just under half of the Industrial Uses OES
releases were reported via Form A.
Page 120 of 484
-------
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
o For COUs/OESs that rely primarily on other sources of release information or generic
scenarios, site-specific release estimates are supported by slight to moderate evidence.
For these scenarios, EPA estimated daily wastewater discharges by using various
modeling approaches, including the use of surrogate TRI and DMR data and modeling
using data from literature, GSs, and ESDs.
¦ For DTD sources, release information is supported by slight to moderate
evidence. Although confidence in the individual contribution from some specific
COUs (ie specific consumer or commercial product categories) is lower,
confidence in estimates of overall DTD releases is moderate. The presented model
is intended to inform the total contribution of DTD releases to overall aggregate
instream concentration as well as providing evidence of individual COUs that
may be most influential. Presented results should be taken in relation to one
another qualitatively rather than discrete quantitative values. Distributions of
DTD releases of consumer and commercial products were estimated for each
COU on a per capita basis using the SHEDS-HT model.
¦ For hydraulic fracturing releases, release information is supported by moderate
evidence. Releases were estimated using Monte Carlo modeling with information
from the Draft ESD on Hydraulic Fracturing and FracFocus 3.0.
• Comparison of Modeled and Monitored Data. Comparisons of estimated and measured
exposures provide moderate evidence. Because most of the available monitoring data are not
collected in locations with known releases temporally or spatially, it is difficult to make direct
comparisons in most locations. However, in case study locations where monitoring data re
located near release sites, comparisons demonstrate that there is concordance between measured
and/or reported and modeled estimates (Section 2.3.1.4), increasing the overall strength of the
evidence. Monitoring data confirm that 1,4-dioxane is present in some surface water and
drinking water. Uncertainty as to whether trends observed in case study locations are
representative of all of the sites decreases overall confidence in these comparisons.
Overall Confidence in Exposure Estimates
Overall confidence in drinking water exposure estimates for surface water concentrations modeled from
facility releases (Section 3.2.2.1.1) is high across the overall distribution, particularly when limited to
sites with high quality sources of release data. For individual facilities and COUs, overall confidence in
exposure estimates varies depending on the confidence in source-specific release data. The modeling
methodology used for this analysis is supported by moderate evidence. This approach makes some
conservative assumptions about flow rates and release frequency and amount. It is designed to estimate
water concentrations expected at specific locations. Available monitoring data confirm that 1,4-dioxane
is present in some surface water and drinking water, though most of the available data were not collected
near release sites are therefore not directly comparable. The overall level of confidence in OES/COU-
specific exposure estimates depends on the source of OES/COU-specific release data described in
Appendix E.3:
• Overall confidence in drinking water exposure estimates is medium to high for OESs/COUs that
rely primarily on site-specific release data reported to DMR or to TRI via Form R.
• Overall confidence in site-specific drinking water exposure estimates is medium for OESs/COUs
for which site-specific release estimates are based on reporting to TRI via Form A
• Overall confidence in drinking water exposure estimates is low to medium for OESs/COUs for
which site-specific release estimates are based on surrogate or modeled information.
Page 121 of 484
-------
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Overall confidence in drinking water exposure estimates for DTD releases under varying conditions
(Section 3.2.2.1.2), is medium. The modeling methodology used for this analysis is supported by robust
evidence. This analysis is designed to provide a nationally representative distribution of estimated water
concentrations under varying conditions. This analysis defines the conditions under which exposures are
higher, but is not designed to predict the specific levels of exposure resulting from DTD releases at
specific locations with precision. Exposure estimates rely on estimated distributions of DTD releases of
specific consumer and commercial products categories associated with each COU. Distributions of DTD
releases of consumer and commercial products were estimated for each COU on a per capita basis using
the SHEDS-HT model. While confidence in the individual contribution from some specific COUs is
lower, confidence in estimates of overall DTD releases is moderate.
Overall confidence in drinking water exposure estimates for hydraulic fracturing releases (Section
3.2.2.1.3) is medium. The modeling methodology used for this analysis is supported by robust evidence
and is designed to provide a nationally representative distribution of estimated water concentrations
under varying conditions. Releases used as inputs in the model were estimated using Monte Carlo
modeling that captures variability across sites. However, the modeled exposure estimates are not directly
tied to specific releases at known locations, decreasing the strength of the evidence related to the
representativeness of the exposure estimates for actual exposures.
Overall confidence in drinking water exposure estimates for aggregate surface water concentrations
predicted by probabilistic modeling (Section 3.2.2.1.4) is high across the overall distribution. For
individual facilities and COUs, overall confidence in exposure estimates varies depending on the
confidence in source-specific release data. The modeling methodology used for this analysis is
supported by robust evidence and is designed to provide a nationally representative distribution of
estimated water concentrations under varying conditions. The estimated drinking water concentrations
modeled in this analysis incorporate contributions from direct and indirect industrial releases, DTD
releases, and other upstream sources. Available monitoring data confirm that 1,4-dioxane is present in
some surface water and drinking water, though most of the available data were not collected near release
sites and are therefore not directly comparable. The overall level of confidence in resulting exposure
estimates depends on the source of OES/COU-specific release data described in Appendix E.3:
• Overall confidence in drinking water exposure estimates is medium to high for OESs/COUs that
rely primarily on release data reported to DMR or to TRI via Form R. Most COUs/OESs are
included in this group
• Overall confidence in drinking water exposure estimates is medium for OESs/COUs for which
release estimates are based on reporting to TRI via Form A. The Import and Repackaging OES
releases used in this analysis are entirely based on Form A reporting of releases, and just under
half of the Industrial Uses OES releases were reported via Form A.
• Overall confidence in drinking water exposure estimates is low to medium for OESs/COUs for
which release estimates are based on surrogate or modeled information.
3.3.2.2 Drinking Water Exposure Estimates Based on Groundwater Concentrations
The weight of evidence for exposure estimates presented in this section is determined by several
different evidence streams, including:
• Evidence supporting the exposure scenarios (Section 3.2.1)
• The quality and representativeness of available groundwater monitoring data (Section 2.3.2.1),
• Evidence supporting modeling approaches and input data (Sections 2.3.2.3.1 and 2.3.2.4.1)
• Evidence supporting release data used as model input data (Section 2.2 and Appendix E.4)
• Concordance between modeled and monitored water concentrations.
Page 122 of 484
-------
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3.3.2.2.1 Groundwater Concentrations Resulting from Disposal to Landfill
Drinking water exposure estimates based on groundwater concentrations modeled for landfill disposal
scenarios are supported by overall slight to moderate evidence.
• Exposure Scenarios and Exposure Factors. The exposure scenarios and exposure factors
underlying these drinking water exposure estimates are supported by slight to moderate
evidence. Exposure factors for drinking water are based on robust data on drinking water intakes,
body weight, and other standard exposure factors from the EPA's Exposure Factors Handbook
(\ v < < \ 1 I). However, the drinking water exposure scenario relies on the assumption that
the groundwater concentrations estimated with the DRAS model may occur in locations where
groundwater is used as a primary drinking water source. While there is uncertainty around this
assumption, this scenario represents a sentinel exposure scenario
• Measured and Monitored Data. Measured/monitored data are supported by moderate evidence.
Monitoring data were available to sufficiently cover most or all of the population groups
included within the exposure scenarios but there are a limited number of studies to corroborate
findings. Since little data is readily available on the concentration of 1,4-dioxane near or around
landfills in groundwater, some caution is required when interpreting monitoring data as it may
not be fully representative of conditions around all landfills.
• Modeling Methodology. The modeling methodology is supported by robust evidence. The
DRAS methodology for deriving the estimate is well described. The underlying computational or
scientific basis is robust and has an empirical basis considering chemical specific properties.
• Model Input Data. The release data relied on as a model input is supported by slight evidence.
Model inputs for the DRAS model include chemical properties of 1,4-dioxane that are well-
defined and reviewed and therefore supported by robust evidence. However, model inputs for
leachate concentrations and loading rates are more uncertain. EPA does not have reasonably
available information on actual concentrations of 1,4-dioxane in leachate for most landfills and
therefore selected landfill leachate concentrations are based on potential for risk to human health.
Loading rates are based on the range reported in TRI for RCRA subtitle C landfills and therefore
may not be representative of nonhazardous landfills evaluated in this analysis. These
uncertainties around landfill leachate concentrations and loading rates decrease the strength of
the evidence for model input data.
• Comparison of Modeled and Monitored Data. Comparison of estimated and measured
exposures provides moderate evidence because monitoring data confirm the presence of 1,4-
dioxane in groundwater in some locations and modeled estimates and measured exposure values
are comparable, however differences in methodology, collection, or context make it difficult to
arrive at full concordance.
Overall Confidence in Exposure Estimates
Overall confidence in drinking water exposure estimates resulting from disposal to landfills (Section
3.2.2.2.1) is low to medium. The modeling methodology is robust. However, the release information
relied on as model input data is supported by slight to moderate evidence, decreasing overall confidence.
In addition, this drinking water exposure scenario relies on the assumption that the groundwater
concentrations estimated with the DRAS model may occur in locations where groundwater is used as a
primary drinking water source. While the substantial uncertainty around the extent to which these
exposures occur decreases overall confidence in the exposure scenario, this scenario represents a
sentinel exposure.
Page 123 of 484
-------
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3.3.2.2.2 Groundwater Concentrations Resulting from Disposal of Hydraulic
Fracturing Waste
Drinking water exposure estimates based on modeled groundwater concentrations estimated under a
range of hydraulic fracturing waste disposal scenarios are supported by slight to moderate evidence.
• Exposure Scenarios and Exposure Factors. The exposure scenario factors underlying these
exposure estimates are supported by slight to moderate evidence. Exposure factors for drinking
water are based on robust data on drinking water intakes, body weight, and other standard
exposure factors from the Exposure Factors Handbook. However, the drinking water exposure
scenario relies on the assumption that the estimated groundwater concentrations may occur in
locations where groundwater is used as a primary drinking water source.
• Measured and Monitored Data. The measured/monitored data are supported by indeterminate
evidence. Available groundwater monitoring data are not located near hydraulic fracturing
operations and do not provide information about the potential for hydraulic fracturing operations
to contribute to groundwater contamination.
• Modeling Methodologies. The modeling methodology and input data are supported by robust
evidence. The methodology for deriving the estimate is well described, the underlying
computational or scientific basis is robust, and has an empirical basis considering chemical
specific properties.
• Model Input Data. Hydraulic fracturing releases are supported by moderate evidence. As
described in Appendix E.4.4, releases were estimated using Monte Carlo modeling with
information from the Draft ESD on Hydraulic Fracturing and FracFocus 3.0. DRAS modeling
was based on very limited data on concentrations of 1,4-dioxane in produced water as reported in
the literature. Reliance on limited data and uncertainty around the representativeness of that data
decrease the strength of the evidence for model input data.
• Comparison of Modeled and Monitored Data. The comparison of estimated and measured
exposures is not rated because no comparisons between estimated and measured exposures were
made.
Overall Confidence in Exposure Estimates
Overall confidence in drinking water exposure estimates resulting from disposal of hydraulic fracturing
waste (Section 3.2.2.2.2) is low to medium. The modeling methodology is robust and the release
information relied on as model input data is supported by moderate evidence. However, no monitoring
data are available to confirm detection of 1,4-dioxane in groundwater near hydraulic fracturing
operations. This drinking water exposure scenario relies on the assumption that the estimated
groundwater concentrations may occur in locations where groundwater is used as a primary drinking
water source. While the substantial uncertainty around the extent to which these exposures occur
decreases overall confidence in the exposure scenario, this scenario represents a sentinel exposure.
3,3.3 Air
The weight of the scientific evidence for exposure estimates presented in this section is determined by
several different evidence streams, including:
• Evidence supporting the exposure scenarios (Section 3.2.1)
• The quality and representativeness of available groundwater monitoring data (Section 2.3.3.1),
• Evidence supporting modeling approaches and input data (Section 2.3.3.2)
• Evidence supporting release data used as model input data (Section 2.3.3.2)
• Concordance between modeled and monitored water concentrations.
Page 124 of 484
-------
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
As described in Section 2.3.3, 1,4-dioxane concentrations in air were estimated for areas around
industrial COUs reported to TRI, hydraulic fracturing operations, and institutional and industrial laundry
facilities. The associated strengths and limitations of these estimated environmental concentrations are
described in Section 2.3.3.3. The general population air exposure scenarios and exposure factors used to
estimate exposures are described in Section 3.2.3.
3.3.3.1 Modeled Air Concentrations for Industrial COUs Reported to TRI
Inhalation exposure estimates resulting from 1,4-dioxane releases for industrial COUs reported to TRI
are supported by overall moderate evidence.
• Exposure Scenarios and Exposure Factors. Exposure scenarios underlying these exposure
estimates are supported by moderate evidence. The exposure factors used to build the exposure
scenarios are directly relevant to general population exposures for communities living in close
proximity to releasing facilities. While the long-term exposure scenarios are most directly
relevant for individuals who reside in fenceline communities for many years, these scenarios are
expected to be within the range of normal habits and exposure patterns expected in the general
population. However, there is uncertainty around the extent to which people actually live and
work around the specific facilities where exposures are highest, decreasing the overall strength of
evidence for these exposure scenarios, particularly at the distances nearest to facilities.
• Measured and Monitored Data. The measured/monitored data are supported by indeterminate
evidence. No measured or monitored data were available.
• Modeling Methodologies. The modeling methodology used to estimate exposure concentrations
via the ambient air pathway is supported by robust evidence. Air concentrations were estimated
using AERMOD and IIOAC. AERMOD is EPA's regulatory model and has been thoroughly
peer reviewed; therefore, the general confidence in results from the model is high but reliant on
the integrity and quality of the inputs used and interpretation of the results. Confidence in
modeled air concentrations resulting from stack releases is lower at distances less than 100 m of
release sites, but confidence in modeled concentrations for fugitive emissions is higher near
release sites. While this is a source of uncertainty, air concentrations from fugitive emissions
tend to peak within 10m of release sites while stack releases were found to peak around 100 m,
indicating that air concentrations modeled at distances less than 100 m of release sites are
generally driven by fugitive emissions. IIOAC is an Excel-based model with results based on
pre-run AERMOD exposure scenarios under a variety of environmental and release conditions.
There is a moderate to high confidence in air concentrations estimated using IIOAC because,
although IIOAC results are based on pre-run AERMOD exposure scenarios (high confidence),
some key sources of uncertainty identified in Section 2.3.3.3 (like limited set of distances
evaluated (100, 100 to 1,000, and 1,000 m) and assumptions made about meteorological
conditions necessary to provide a more conservative exposure estimate) can lead to a slightly
lower confidence (moderate).
• Model Input Data. Model input data on air releases are supported by slight to robust evidence,
with the strength of the evidence varying across COUs/OESs. A more detailed OES-specific
discussion of the confidence in sources of release information is presented in Appendix E.5.4.
o For COUs/OESs that rely primarily on release data reported to TRI via Form R, site-
specific release estimates are supported by moderate to robust evidence. As described in
Appendix E.5.4, these release estimates are based on specific release amounts and other
source-specific information reported by facilities as a regulatory requirement.
Page 125 of 484
-------
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
o For COUs/OESs that rely primarily on release data reported to TRI via Form A, site-
specific release estimates are supported by moderate evidence. As described in Appendix
E.5.4, Form A simply indicates that releases are below the reporting thresholds and
specific release estimates require assumptions about exact amounts and locations of
releases.
o For COUs/OESs that rely primarily on other sources of release information or generic
scenarios, release estimates are supported by evidence ranging from slight to moderate
evidence. For these scenarios, EPA estimated daily and annual air releases using various
modeling approaches, including the use of surrogate TRI data and modeling using data
from literature, GSs, and ESDs.
• Comparison of Modeled and Monitored Data. Comparison of estimated and measured
exposures provide indeterminate evidence. No measured or monitored data were available for
comparison.
Overall Confidence in Exposure Estimates
Overall confidence in inhalation exposure estimates resulting for air concentrations modeled based on
industrial releases (Section 3.2.3.1) varies across COUs. The AERMOD modeling methodology used for
this analysis is robust and considers contributions from both stack and fugitive emissions. The exposure
scenarios considered are most relevant to long-term residents in fenceline communities. There is
uncertainty around the extent to which people live and work in the specific locations where exposures
are highest, decreasing confidence in the exposure scenarios, particularly at distances nearest to
facilities. Overall confidence varies due to variable levels of confidence in underlying release
information used to the support the analysis:
• Overall confidence in site-specific inhalation exposure estimates is medium to high for
OESs/COUs that rely primarily on release data reported to TRI via Form R.
• Overall confidence in site-specific inhalation exposure estimates is medium for OESs/COUs for
which release estimates are based on data reported to TRI via Form A.
• Overall confidence in inhalation exposure estimates is low to medium for OESs/COUs for which
release estimates are based on surrogate or modeled information.
3.3.3.2 Air Concentrations Modeled near Hydraulic Fracturing Operations and
Industrial/Institutional Laundries
Inhalation exposure estimates resulting from 1,4-dioxane released to air from hydraulic fracturing
operations and industrial/institutional laundries are supported by overall moderate evidence.
• Exposure Scenarios and Exposure Factors. Exposure scenarios underlying these exposure
estimates are supported by moderate evidence. The factors used to build the exposure scenarios
are directly relevant to general population exposures for communities living in close proximity to
releasing facilities. While the long-term exposure scenarios are most directly relevant for
individuals who reside in fenceline communities for many years, these scenarios are expected to
be within the range of normal habits and exposure patterns expected in the general population.
However, there is some uncertainty around the extent to which people actually live and work
around the specific locations where exposures are highest, decreasing the overall strength of
evidence for these exposure scenarios.
• Measured and Monitored Data. The measured/monitored data are supported by indeterminate
evidence. No measured or monitored data were available.
• Modeling Methodologies. The modeling methodology used to estimate exposure concentrations
via the ambient air pathway is supported by robust evidence. Air concentrations were estimated
Page 126 of 484
-------
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
using IIOAC. IIOAC is an Excel-based model with results based on pre-run AERMOD exposure
scenarios under a variety of environmental and release conditions. There is a moderate to high
confidence in air concentrations estimated using IIOAC because, although IIOAC results are
based on pre-run AERMOD exposure scenarios (high confidence), some key sources of
uncertainty identified in Section 2.3.3.3 (like limited set of distances evaluated (100, 100 to
1,000, and 1,000 m) and assumptions made about meteorological conditions necessary to provide
a more conservative exposure estimate) can lead to a slightly lower confidence (moderate).
• Model Input Data. Input data used for modeling exposures from hydraulic fracturing operations
and industrial/institutional laundries are supported by moderate evidence. As described in
Appendix E.5.4, these modeled exposure estimates are based on alternative release estimates and
scenario conditions found in the literature and derived with Monte Carlo models of release
estimate, some of which have been peer reviewed, others which may not be peer reviewed. Since
the modeled exposures are based on alternative release estimates, which in turn are based on
modeled data and outputs, there is a lower overall confidence in the modeled exposures from
such input data. Additionally, exposure estimates using this input data requires certain
assumptions which can lead to a lower overall confidence in the estimated exposure
concentrations.
• Comparison of Modeled and Monitored Data. Comparison of estimated and measured
exposures provide indeterminate evidence. No measured or monitored data were available for
comparison.
Overall Confidence in Exposure Estimates
Overall confidence in inhalation exposure estimates resulting for air concentrations modeled based on
releases from hydraulic fracturing operations (Section 3.2.3.2) is medium. The modeling methodologies
used to estimate air concentrations are robust. The distribution of air releases used as model input data
were estimated using Monte Carlo modeling and rely on assumptions. No air monitoring data were
available to confirm detection of 1,4-dioxane in air near hydraulic fracturing operations. There is
uncertainty around the extent to which people live and work in the specific locations where exposures
are highest, decreasing confidence in the exposure scenarios.
Overall confidence in inhalation exposure estimates resulting from air concentrations modeled based on
releases from industrial and institutional laundries ion 0) is medium. The modeling methodologies are
robust. The distribution of air releases used as model input data were estimated using Monte Carlo
modeling and rely on assumptions. No air monitoring data were available to determine whether 1,4-
dioxane is detected near industrial and institutional laundry facilities.
Page 127 of 484
-------
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4 HUMAN HEALTH HAZARD
1,4-Dioxane - Human Health Hazards (Section 4):
Key Points
EPA previously evaluated reasonably available information for human health hazards and
identified hazard endpoints for non-cancer effects and cancer effects following acute and chronic
exposures. This section describes adjustments made to previously published hazard values to
align with the exposure scenarios evaluated in this supplemental evaluation.
• EPA considered the potential for increased susceptibility across PESS factors throughout
the hazard assessment and dose-response analysis. PESS categories identified in the
assessment include lifestage, genetics, and preexisting disease.
• The primary acute/short-term, non-cancer endpoint for 1,4-dioxane is liver toxicity
following inhalation exposure.
• The primary chronic, non-cancer endpoints for 1,4-dioxane are liver toxicity and systemic
effects on the olfactory epithelium.
• Inhalation cancer endpoint for 1,4-dioxane is based on combined tumor risk at multiple
sites.
• Oral and dermal cancer endpoints for 1,4-dioxane are based on liver tumors following
oral exposures.
4.1 Summary of Hazard Endpoints Previously Identified in the 2020 Risk
Evaluation
This draft supplement relies on the Hazard Identification and Dose-Response Assessment that was
previously described in the 2020 RE. All hazard values used to calculate risks for 1,4-dioxane in this
draft supplement were derived from the previously peer-reviewed PODs published in the 2020 RE and
amended in the recent correction memo.
Hazard values used in the 2020 RE include human equivalent concentrations (HECs) and human
equivalent doses (HEDs) for non-cancer endpoints. Additionally, an inhalation unit risk (IUR) and
cancer slope factor (CSF) for lifetime cancer risk were derived for both occupational and consumer
scenarios for COUs where it was applicable. The hazard values published in the 2020 RE and used as
the basis for hazard values in this draft supplement were developed with consideration for potentially
susceptible subpopulations. Several potential sources of susceptibility were discussed qualitatively
including lifestage, genetic variability, liver disease, and other chronic diseases that may influence
metabolism or target organ susceptibility. EPA applied a 10x uncertainty factor to non-cancer hazard
values to account for these sources of human variability.
4.2 Summary of Adjustments to Previously Established Hazard Values
For many of the exposure scenarios evaluated in this draft supplement, the previously established peer-
reviewed hazard values were applied without modification. For example, risks from occupational
exposures to products containing 1,4-dioxane as a byproduct can be evaluated using the acute, chronic,
and cancer hazard values previously developed for OESs.
Some of the exposure scenarios included in this draft supplement require duration adjustments to the
previously established PODs. For example, to evaluate risks from ambient air exposures for fenceline
Page 128 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3837 communities, EPA assumes continuous exposure to air for 24 hours/day, 7 days/week. As described in
3838 more detail below, EPA adjusted the previously established HEC and IUR values (originally developed
3839 for 8 hours/day, 5 days/week exposures) to identify hazard values appropriate for continuous exposure
3840 scenarios.
3841
3842 In addition, acute and chronic non-cancer oral and dermal HEDs extrapolated from occupational HECs
3843 were corrected to apply consistent breathing rates assumptions.
3844
3845 The full set of hazard values used to evaluate risk from the exposure scenarios in this draft supplement
3846 are presented in Table 4-1.
Page 129 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3847 Table 4-1. Hazard Values Used for 1,4-Dioxane in this Draft Supplement
Scenario
(Population)
Endpoints
Inhalation
HEC/IUR
Dermal
HED/CSF
Oral HED/CSF
T otal
Uncertainty
Factors
Reference(s)
Acute non-cancer
(general population)
Systemic liver
toxicity
26.2 ppm
(94.5 mg/m3)
24 hours
17.4 mg/kg-d
(extrapolated from HEC)
17.4 mg/kg-d
(extrapolated from HEC)
300
(Pntz et a )
(Mattic et al, 2012)
Acute
non-cancer
(occupational)
Systemic liver
toxicity
78.7 ppm
(284 mg/m3)
8 hours
17.4 mg/kg-d
(extrapolated from HEC)a
17.4 mg/kg-d
(extrapolated from
HEC)fl
300
(Mattic et al., 2012)
Chronic
non-cancer
(general population)
Olfactory epithelium
effects attributed to
systemic delivery
(inhalation) a; liver
toxicity (oral)
0.846 ppm
(3 mg/m3)
24 hours, 7
days/week
0.56 mg/kg-d
(extrapolated from HEC)
2.6 mg/kg-d
30
(Kano et al., 2009;
Kasai et al., 2009)
Chronic
non-cancer
(occupational)
Olfactory epithelium
effects attributed to
systemic delivery
(inhalation)a; liver
toxicity (oral)
3.6 ppm
(12.8 mg/m3)
8 hours, 5
days/week
0.56 mg/kg-d
(extrapolated from HEC)''
2.6 mg/kg-d
30
(Kano et al., 2009;
Kas£u_etjyLv2009)
Cancer
(general population)
Inhalation cancer
risk based on
combined tumor risk
at multiple sites;
oral/dermal cancer
risk based on liver
tumors
IUR:
1.6E-02 per ppm
4.3E-06
(Hg/m3)"1
24 hours, 365
days/ year
CSF:
1.2E-01 (mg/kg-d)1
(extrapolated from oral
CSF)
CSF:
1.2E-01 (mg/kg-d)1
(Kano et al., 2009;
Kasai et al., 2009;
NIP. 1986)
Cancer
(occupational)
Inhalation cancer
risk based on
combined tumor risk
at multiple sites;
oral/dermal cancer
risk based on liver
tumors
IUR:
3.7E-03 per ppm
1.0E-06
(Hg/m3)-1
8 hours, 5
days/week
CSF: 1.2E-01 (mg/kg-d)1
(extrapolated from oral
CSF)
CSF:
1.2E-01 (mg/kg-d)1
(Kano et al., 2009;
Kasai et al.. 2009)
" Due to the uniform distribution of lesions (rather than a distribution consistent with airflow), EPA concluded that effects in the olfactory epithelium may be due to
systemic delivery rather than portal of entry effects due to the (see discussion in the 2020 RE on p 183, p. 188).
b Occupational HEDs extrapolated from occupational HECs were corrected as described in the correction memo.
HEC = Human Equivalent Concentration; HED = Human Equivalent Dose; CSF = Cancer Slope Factor; 1UR = Inhalation Unit Risk
3848
Page 130 of 484
-------
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4,2,1 Derivation of Acute/Short-Term Hazard Values
4.2.1.1 Inhalation HEC
The acute/short-term HECs are based on the lowest observed adverse effect concentration (LOAEC) for
systemic liver toxicity observed in a short-term inhalation toxicity study in rats exposed 6 hours/day for
5 days/week. In the 2020 RE, EPA derived an HEC for 8-hour occupational exposures by applying a
duration adjustment for an 8-hour exposure and a dosimetric adjustment factor of 1 (the default value
when the calculated ratio of animal to human blood:air partition coefficients is greater than 1 (
1994b). The occupational HEC derived in the 2020 RE is based on default breathing rate assumptions
and did not use adjustments for occupational breathing rates. For this draft supplement, EPA also
derived an HEC for continuous general population exposures by applying a 24-hour duration adjustment
to the original HEC.
4.2.1.2 Oral and Dermal HEDs
In the absence of acute oral or dermal toxicity studies, the acute/short-term HED was derived from the
acute HEC using route-to-route extrapolation. An acute HED for the general population was derived
from the duration-adjusted 24-hour HEC using the following equation:
dermal or oral HED (mg/kg-d) = PODhec (mg/m3) x inhalation volume x 100% inhalation
absorption ^ body weight
where the inhalation volume for the general population is 14.7 m3/day and body weight is 80 kg, based
on EPA's Exposure Factors Handbook ( 2011). Inhalation absorption was estimated based on
experimental data from inhalation exposures in humans (Young et ai. 1977; Young et ai. 1976) that
indicated that 1,4-dioxane is readily absorbed; however, the available studies did not measure the
parameters needed to generate a quantitative estimate of the fraction absorbed. Given this qualitative
indication of rapid systemic uptake and the absence of quantitative inhalation absorption data, 100
percent inhalation absorption is assumed.
In the 2020 RE, an occupational acute HED was derived from the occupational HEC using the same
equation but with an inhalation volume for workers based on higher breathing rates. As described in the
correction memo, that derivation was incorrect. Because the occupational HEC was derived based on a
normal general population breathing rate, the HED derivation should apply the same breathing rate
assumptions. This draft supplement for 1,4-dioxane uses the revised acute occupational HED, which is
equal to the general population HED.
4.2,2 Derivation of Chronic Hazard Values
4.2.2.1 Inhalation HEC
The chronic HECs are based on BMCLio {i.e., the lower confidence limit of the benchmark
concentrations associated with a benchmark response of 10 percent) for effects in the olfactory
epithelium following inhalation exposures to rats for 6 hours/day, 5 days/week for 2 years. In the 2020
RE, EPA derived an HEC for chronic worker exposures by applying a duration adjustment for 8
hours/day and a dosimetric adjustment factor of 1 (the default value using the RGDR approach for
systemic effects when the calculated ratio of animal to human blood:air partition coefficients is greater
than 1 (U.S. EPA. 1994b)). The occupational HEC derived in the 2020 RE used default breathing rate
assumptions and did not adjust for occupational breathing rates. For this draft supplement, EPA also
Page 131 of 484
-------
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
derived an HEC for continuous general population exposures by applying an alternate duration
adjustment for 24 hours/day, 7 days/week.
4.2.2.2 Oral HEDs
A chronic oral HED was calculated based on a benchmark dose level (BMDL) for liver toxicity
observed following chronic drinking water exposures to male rats and a nearly identical NOAEL value
for liver toxicity in male rats in a similar chronic toxicity study. In the 2020 RE, EPA derived an HED
by multiplying the nearly identical rodent BMDL and NOAEL values by (BWa/BWh)0'25, where BWa is
the bioassay-specific rodent body weight, and BWh is the default human body weight of 70 kg. Because
the chronic HED is based on a daily dose rate (as opposed to an intermittent exposure concentration), it
is equally applicable to both occupational and general population exposures and no additional
conversion is required.
4.2.2.3 Dermal HEDs
In the absence of chronic dermal toxicity studies, chronic dermal HEDs were derived from both the
chronic HEC and from the oral HED using route-to-route extrapolation. In the 2020 RE, the dermal
HED used for occupational risk calculations was extrapolated from the chronic worker HEC. For this
draft supplement, EPA also derived an HED from the HEC for continuous general population exposure.
The duration-adjusted chronic HEC for general populations was converted to a chronic HED for the
general population using the following equation:
dermal HED (mg/kg-d) = inhalation BMDLhec (mg/m3) x inhalation volume x 100% inhalation
absorption ^ body weight
where the inhalation volume for the general population is 14.7 m3/day ( ) for a 24-hour
general population exposure and the body weight is 80 kg. As described above for the acute hazard
values, EPA assumed 100 percent inhalation absorption. In the 2020 RE, an occupational HED was
derived from the occupational HEC using the same basic equation but with an inhalation volume for
workers based on higher breathing rates. The difference in the HEDs derived from occupational and
general population HECs reflect differences in breathing rate assumptions for the two populations.
In the 2020 RE, an occupational chronic dermal HED was derived from the occupational HEC using the
same equation but with an inhalation volume for workers based on higher breathing rates. As described
in the correction memo, that derivation was incorrect. Because the occupational HEC was derived based
on a normal general population breathing rate, the HED derivation should apply the same breathing rate
assumptions. This assessment uses the revised occupational chronic dermal HED, which is equal to the
general population HED.
4.2.3 Derivation of Cancer Hazard Values
For cancer, the inhalation unit risk (MR) value was derived using the MS-Combo model to evaluate the
combined cancer risk for multiple tumor sites observed in male rats following inhalation exposure for 6
hours/day, 5 days/week for 2 years. Tumor types included in the MS-Combo model include nasal cavity
squamous cell carcinoma, Zymbal gland adenoma, hepatocellular adenoma or carcinoma, renal cell
carcinoma, peritoneal mesothelioma, mammary gland fibroadenoma, and subcutis fibroma. In the 2020
RE, EPA derived an IUR for chronic worker exposures by applying a dosimetric adjustment factor of 1
and a duration adjustment for 8 hours/day. The occupational IUR derived in the 2020 RE applied default
breathing rate assumptions and did not use adjustments for occupational breathing rates. The
occupational IUR was rounded to 1 x 10~6 (|ig/m3) 1 for application in risk calculations. For this draft
Page 132 of 484
-------
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
supplement, EPA used that rounded occupational IUR to derive an IUR for continuous general
population exposures to 1,4-dioxane by applying a duration adjustment for 24 hours/day, 7 days/week.
The oral and dermal cancer slope factor was derived using the Multistage Weibull Model for the liver
tumors in female mice that had been exposed continuously via drinking water. In the 2020 RE, EPA
calculated an HED for each tumor type by multiplying rodent doses by (BWa/BWh)0,25, where BWa is
the bioassay-specific rodent body weight and BWh is the default human body weight of 70 kg. The CSF
was then calculated by dividing the benchmark response rate (0.5) by the HED. This CSF was applied to
both occupational and consumer/general population scenarios using scenario-specific risk benchmarks
and lifetime exposure estimates.
4.3 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty
in the Hazard and Dose-Response Analysis
All assumptions or uncertainties inherent to the human health hazard assessment and dose-response
analysis peer-reviewed in the 2020 RE are still applicable for this draft supplement. As described in the
2020 RE, EPA has medium confidence in the acute non-cancer PODs and high confidence in the chronic
non-cancer PODs for oral, dermal, and inhalation exposures. EPA has high confidence in the cancer
inhalation unit risk and medium-high confidence in the oral and dermal cancer slope factor. These
conclusions are based on the fact that there is a robust set of high quality chronic and sub-chronic
inhalation and oral exposure studies in rats and mice. The available evidence demonstrates consistent
systemic toxicity and tumor formation in rats exposed via inhalation and in both rats and mice exposed
via drinking water. Key sources of uncertainty include limited data on some sensitive reproductive and
developmental endpoints, reliance on route-to-route extrapolation, uncertainty around the mode of
action for 1,4-dioxane carcinogenicity, and the potential for subpopulations or lifestages with increased
biological susceptibility to 1,4-dioxane. Available methods indicate potential higher inhaled doses in
young children than adults, consistent with 1,4-dioxane specific studies integrating lifestage differences
in ventilation, anatomy and metabolism via CYP2E1 ( ). The preferred method to
quantify these lifestage differences is a 1,4-dioxane specific PBPK model; however, the available PBPK
models for 1,4-dioxane are not adequate and there are not generally accepted default methods not
specific to 1,4-dioxane. Therefore, the air concentration is used as the exposure metric for all lifestages
and the 10x uncertainty factor accounts for these lifestage differences per EPA guidance (
2012. 1994bY
Page 133 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3968 5 HUMAN HEALTH RISK CHARACTERIZATION
1,4-Dioxane - Human Health Risk Characterization (Section 5):
Key Points
EPA estimated cancer and non-cancer risks for each exposure pathway for a range of central tendency and high-
end exposure scenarios. Overall confidence in risk estimates varies across exposure pathways and COUs,
depending on the data and assumptions used to derive exposure and risk estimates. Differences in estimates
between central tendency and high-end exposure scenarios may reflect both variability across the population and
uncertainty in the exposure assessment.
• Cancer and non-cancer risks were evaluated for occupational inhalation and dermal exposures to 1,4-
dioxane present as a byproduct.
O Cancer risk estimates for inhalation exposure range from 8.3x 10 12 to 1,8Ex 10 3 for central
tendency exposures and 5.4x 10" to 2.3x 10 2 for high-end exposures.
O Cancer risk estimates for dermal exposure range from 8.1 x 10 7 to 8.6x 1 () 4 for central tendency
exposures and from 5.Ox 10 " to 1,5x 10 2 for high-end exposures.
• Cancer and non-cancer risks were evaluated for drinking water exposures resulting from releases to
surface water, including facility releases, down-the-drain releases, hydraulic fracturing releases, and
aggregate releases from multiple sources.
O Risk from individual facilities vary substantially within and across COUs, with cancer risk
estimates ranging from 5.4x10 12 to 0.025.
O Cancer risk estimates from modeled do wn-the-drain releases are highest in locations where large
populations are contributing to these releases and where they are ultimately discharged to
streams with low flow.
O Cancer risk estimates from modeled hydraulic fracturing waste releases to surface water are
3.57x 10~8 for median modeled releases and 1.45x 10~6 for 95th percentile modeled releases.
O Probabilistic modeling provides a distribution of risk estimates reflecting a range of drinking
water scenarios that account for aggregate sources of 1,4-dioxane in water.
• Cancer risks were evaluated for drinking water exposures resulting from releases to land with potential
to reach groundwater.
O Risk estimates from landfill leachate are highest under disposal scenarios resulting in higher 1,4-
dioxane concentrations in leachate and higher landfill loading rates.
O Cancer risk estimates for drinking water exposures resulting from hydraulic fracturing waste
released to land/groundwater range from 4.Ox 10 7 for median modeled releases to 8.6x 10 " for
95th percentile modeled releases.
• Cancer and non-cancer risks were evaluated for general population exposure to 1,4-dioxane in air.
O Cancer risk estimates for industrial air releases reported to TRI were generally highest within
1,000 m of the facilities and lower at greater distances.
O Cancer risk estimates within 1,000 m of hydraulic fracturing operations range from 0.2x 10~8 to
7.1x10 5 for a range of model scenarios across a range of high-end and central tendency release
scenarios.
O Cancer risk estimates within 1,000 m of industrial and institutional laundries range from
1.5xl0~n to 3.8 xlO-8 across a range of high-end and central tendency air concentrations
modeled for maximum release scenarios.
3969
Page 134 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3970 5.1 Risk Characterization Approach
3971 The exposure scenarios, populations of interest, and toxicological endpoints used for evaluating risks
3972 from acute and chronic exposures are summarized below in Table 5-1. To estimate risks from
3973 occupational and general population exposure scenarios evaluated in this draft supplement, EPA used
3974 the same methods described in the 2020 RE, as summarized below.
3975
3976 Table 5-1. Use Scenarios, Populations of Interest, and Toxicological Endpoints Used for Acute and
3977 Chronic Exposures
Workers"
Acute - Adolescent (>16 years old) and adult workers exposed to 1,4-dioxane for a single 8-hour exposure
Chronic - Adolescent (>16 years old) and adult workers exposed to 1,4-dioxane for the entire 8-hour
workday for 260 days per year for 40 working years
General Population Drinking Water Exposures b
Acute - Adults, children, and formula-fed infants exposed to 1,4-dioxane through drinking water over a 24-
hour period
Chronic - Adults, children, and formula-fed infants exposed to 1,4-dioxane through drinking water up to 33
years d
General Population Ambient Air Exposurec
Acute - People exposed to 1,4-dioxane through ambient air over a 24-hour period
Chronic - People exposed to 1,4-dioxane through ambient air continuously up to 33 years d
Non-cancer Acute/Short-term Hazard Values
Sensitive acute/short-term health effect: liver toxicity
Acute Uncertainty Factors (Benchmark MOE) = 300 (UFA = 3; UFH = 10; UFL = 10)
• 8-hour HEC (occupational exposure) = 78.7 ppm
• 24-hour HEC (continuous general population exposure) = 26.2 ppm
• Acute Oral and Dermal HED (occupational and general population exposure) = 17.4 mg/kg
Non-cancer Chronic Hazard Values
Sensitive chronic health effects:
• Liver toxicity (oral)
• Effects on the olfactory epithelium due to systemic exposures (inhalation and dermal)
Chronic Uncertainty Factors (Benchmark MOE) = 30 (UFa = 3; UFh = 10)
• HEC (8-hour occupational exposure) = 3.6 ppm
• HEC (continuous exposure general population exposure) = 0.846 ppm
• Oral HED (for both occupational and general population scenarios) = 2.6 mg/kg/day
• Dermal HED (extrapolated from HECs for both occupational and general population scenarios) =
0.56 mg/kg/day
Cancer Hazard Values
Inhalation cancer hazard for 1,4-dioxane is based on combined tumor hazard at multiple sites
• IUR (occupational) = 3.7E-03 per ppm
• IUR (continuous) = 1.6E-02 per ppm
Oral and dermal cancer hazards for 1,4-dioxane are based on liver tumors following oral exposures
• Oral/dermal slope factor = 1.2E-01 (mg/kg/day)'1
" Adult workers (>16 years old) include both female and male workers. Risks to ONUs were not calculated separately because
exposure data were not available for ONUs for the OESs being evaluated. Risks to ONUs are assumed to be equal to or less
than risks to workers who handle materials containing 1,4-dioxane as part of their job.
b These scenarios are used to evaluate potential risks from 1,4-dioxane in surface water, drinking water sources and
groundwater that may be used as drinking water.
c Inhalation exposures are described in terms of air concentrations and do not include lifestage-specific adjustments; risk
estimates based on air concentrations are intended to address risks to all lifestages (see Section 4.3).
d33 years is the 95th percentile residential occupancy period. U.S. EPA Exposure Factors Handbook (U.S. EPA, 2011).
Chapter 16, Table 16-5.
MOE = margin of exposure; UFA = Interspecies uncertainty factor for animal-to-human extrapolation; UFH = Intraspecies
uncertainty factor for human variability; UFL = LOAEC-to-NOAEC uncertainty factor for reliance on a LOAEC as the POD
3978
Populations
of Interest
and Exposure
Scenarios
Health
Effects,
Hazard
Values and
Benchmarks
Page 135 of 484
-------
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
.1.1 Estimation of Non-cancer Risks
EPA used a margin of exposure (MOE) approach to identify potential non-cancer risks. The MOE is the
ratio of the non-cancer POD divided by a human exposure dose. Acute and chronic MOEs for non-
cancer inhalation and dermal risks were calculated using the following equation:
Non — cancer Hazard value (POD)
MOEacute or chronic —
Human Exposure
Where:
MOE = Margin of exposure (unitless)
Hazard value (POD)= HEC (ppm) or HED (mg/kg-d)
Human Exposure = Exposure estimate (in ppm or mg/kg-d)
MOE risk estimates may be interpreted in relation to benchmark MOEs. Benchmark MOEs are typically
the total UF for each non-cancer POD. The MOE estimate is interpreted as indicating a human health
risk if the MOE estimate is less than the benchmark MOE {i.e., the total UF). On the other hand, if the
MOE estimate is equal to or exceeded the benchmark MOE, risk is not indicated. Typically, the larger
the MOE, the more unlikely it is that a non-cancer adverse effect occurs relative to the benchmark.
When determining whether a chemical substance presents unreasonable risk to human health or the
environment, calculated risk estimates are not "bright-line" indicators of unreasonable risk, and EPA has
discretion to consider other risk-related factors apart from risks identified in risk characterization.
5.1.2 Estimation of Cancer Risks
Extra cancer risks for repeated exposures to a chemical were estimated using the following equations:
Inhalation Cancer Risk = Human Exposure x IUR
or
Dermal/Oral Cancer Risk = Human Exposure x CSF
Where:
Risk = Extra cancer risk (unitless)
Human exposure = Exposure estimate (LADC in ppm)
IUR. = Inhalation unit risk
CSF = Cancer slope factor
Estimates of extra cancer risks are interpreted as the incremental probability of an individual developing
cancer over a lifetime following exposure {i.e., incremental, or extra individual lifetime cancer risk).
5.2 Human Health Risk Characterization
5.2,1 Summary of Risk Estimates for Occupational Exposures
EPA estimated cancer and non-cancer risks for workers exposed to 1,4-dioxane based on the
occupational exposure estimates that were described in Section 3.1. Risks to ONUs were not calculated
separately because exposure data were not available for ONUs for the OESs being evaluated. Risks to
ONUs are assumed to be equal to or less than risks to workers who handle materials containing 1,4-
dioxane as part of their job.
Page 136 of 484
-------
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Below are summaries of the cancer risk estimates for the inhalation and dermal exposures as well as key
sources of uncertainty for all occupational exposure scenarios assessed in this draft supplement. These
risk estimates are based on exposures to workers in the absence of PPE such as gloves or respirators.
Section 3.1.2.4 contains an overall discussion on strengths, limitations, assumptions, and key sources of
uncertainty for the occupational exposure assessment. Additionally, Appendix F contains a
comprehensive weight of the scientific evidence summary table which presents an OES-by-OES
discussion of the key factors that contributed to each weight of the scientific evidence conclusion.
Complete risk calculations and results for occupational OES/COUs from the current analysis as well as
those previously presented in the 2020 RE are available in 1,4-Dioxane Supplemental Information File:
Occupational Exposure and Risk Estimates (U.S. EPA. 2023r).
Risk estimates vary across OES/COUs. Because cancer risk is the primary risk driver in most exposure
scenarios, this summary of results focuses on cancer risk estimates. For 7 of the 10 COU subcategories
evaluated, high-end cancer risk estimates were above 1 in 10,000. For many of those COUs, acute
and/or chronic non-cancer risk estimates were below the corresponding benchmark MOEs, indicating
that non-cancer risks may also be a concern. Cancer risk estimates for inhalation exposure range from
8.3/10 12to 1.8/10 3 for central tendency exposures and 5.4/10 " to 2.3/10 2 for high end exposures.
Cancer risk estimates for dermal exposure range from 8.1 x 10~7 to 8.6 / 10 4 for central tendency
exposures and from 5.Ox 10~6 to 1.5/10 2 for high end exposures. Risks are highest for PET
manufacture, hydraulic fracturing operations, ethoxylation processes, and dish soap/dishwashing
detergent. For these OESs, cancer risk estimates were greater than 1 in 10,000 for both central tendency
and high-end exposures. For these OESs, the key uncertainties include limited exposure monitoring data,
age of data, representativeness of key modeling parameters, and the extent to which the data collected
under past practices and operations are representative of modern practice and operations.
Overall confidence in risk estimates for occupational inhalation exposures ranges from low to high,
depending on the confidence in exposure assessment for each OES/COU. As described in Section 4.3,
overall confidence in the cancer inhalation unit risk underlying these risk estimates is high. As described
in Section 3.3.1.1, the measured and monitored inhalation exposure data are supported by moderate to
robust evidence. Additionally, the exposure modeling methodologies and underlying model input data is
supported by moderate to robust evidence. However, there is uncertainty in the representativeness of the
assessed exposure scenarios towards all potential exposures for the given OES/COU, limitations in the
amount and age of monitoring data, and limitations in the modeling approaches towards 1,4-dioxane-
specific use within the OES/COU. Therefore, while the underlying data and methods used to estimate
occupational inhalation risk is supported by moderate to robust evidence, the overall confidence of these
estimates ranges from low to high depending on the OES/COU. Key exposure considerations along with
the corresponding risk estimates are below.
• Industrial/Commercial Use of Textile Dye. Risk estimates were derived using personal
breathing zone and area monitoring data collected from 1991 to 2010 at four facilities linked to
the use of textile dyes. Cancer risk estimates for inhalation exposure range from 1.6/10 5 for
central tendency exposures to 3.6* 10~2 for high-end exposures. However, there is uncertainty in
the risk estimates. The monitoring data used in this analysis are limited {i.e., 14 samples from
four sites). It also is not known how manufacturing processes and workplace conditions have
changed since the 1990s, when approximately half of the data was collected. For instance, EPA
does not have information available about the actual activities of the sampled workers and the
representativeness of the facility engineering controls to the modern practice.
• Industrial/Commercial Use of Antifreeze. Risk estimates were derived from occupational
exposures modeled using Monte Carlo simulations for the worker activity of container
Page 137 of 484
-------
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
unloading. Cancer risk estimates for inhalation exposure range from 8.3x10 12 for central
tendency exposures to 5.4xl0~n for high-end exposures. However, there is uncertainty in the
risk estimates. Specifically, there is uncertainty as to the representativeness of some of the model
input data and, therefore, subsequent calculated exposures to the actual distribution of antifreeze
occupational exposures. This is due to limitations of using generic industry values identified for
the automotive industry. Also contributing to the uncertainty is that EPA used consumer
antifreeze use rates that were scaled up for commercial/industrial use in the Monte Carlo
modeling.
• Industrial/Commercial Use of Surface Cleaner. Risk estimates were derived using 49 personal
breathing zone samples taken in 2019 during the use of surface cleaners in domestic kitchens and
bathrooms. Cancer risk estimates for inhalation exposure range from 1.1 x 10~7 for central
tendency exposures to 1,8x 10~6 for high-end exposures. However, there is uncertainty in the risk
estimates. Specifically, the monitoring data summary did not provide discrete monitoring points
and only provided summary statistics such as the geometric mean and maximum. Therefore,
EPA could not calculate the 50th and 95th percentile exposures. Also, it is uncertain the extent to
which the cleaning activities captured in this study reflect all occupational surface cleaning
scenarios, as they were measured in a consumer setting.
• Industrial/Commercial Use of Dish Soap and Dishwasher Detergent. Risk estimates were
derived using 29 personal breathing zone and area exposure monitoring samples from 1980 taken
during the transfer of detergents to/from storage, liquid mixing, and detergent bottling. Cancer
risk estimates for inhalation exposure range from 4,Ox 10 4 for central tendency exposures to
1.0x10 3 for high-end exposures. However, there is uncertainty in these risk estimates. Although
the percent of 1,4 dioxane in the detergents used at the site where monitoring was conducted was
up to 0.423 percent, all monitoring samples were non-detect for 1,4-dioxane. Therefore, EPA
used the estimated LOD and LOD/2 for the worker high-end and central-tendency exposure
estimates. These values are two to four orders of magnitude (for dish soap and dishwasher
detergent, respectively) greater than the 2020 RE consumer exposure inhalation estimate. This
difference can reasonably be expected considering occupational users of dish soap are potentially
exposed to higher concentrations of 1,4-dioxane in concentrated commercial formulations and
use these dish soaps at higher frequencies for longer durations than consumers. However, there is
uncertainty as to the representativeness of these estimates due to the age of the monitoring data,
number of non-detects, and the limited sample size.
• Industrial/Commercial Use of Laundry Detergent. Risk estimates were derived from
occupational exposures modeled using Monte Carlo simulations for the worker activity of
unloading detergent into machines, container cleaning, and laundry operations. For industrial
laundries, cancer risk estimates for vapor inhalation exposure range from 2.Ox 10~7 for central
tendency exposures to 9.2x10 7 for high-end exposures. For institutional laundries, cancer risk
estimates for vapor inhalation exposure range from 1,6x 10~7 for central tendency exposures to
7.1 x 10~7 for high-end exposures. In both cases, cancer risk estimates for total particulates
inhalation range from 4.0xl0~8 for central tendency exposures to 9.8x ] 0 8 for high-end
exposures. Cancer risk estimates for respirable particulates inhalation range from 1.3 x 10~8 for
central tendency exposures to 3.3x10-8 for high-end exposures. However, there is uncertainty in
the risk estimates. Specifically, there is uncertainty as to the representativeness of some of the
model inputs and, therefore, subsequent calculated exposures to the actual distribution of laundry
Page 138 of 484
-------
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
detergent occupational exposures. This is due to limitations of using generic industry values
identified for institutional and industrial laundries.
• Industrial/Commercial Use of Paint and Floor Lacquer. Risk estimates were derived using 17
personal breathing zone samples collected by NIOSH in 1987 at a military vehicle painting site.
Cancer risk estimates for inhalation exposure range from 8.Ox 10~5 for central tendency exposures
to 5.9x 10~4 for high-end exposures. However, there is uncertainty in the risk estimates. The
monitoring data used in this analysis are limited {i.e., 17 samples taken at one site). It also is not
known how processes and workplace conditions have changed since 1987. For instance, EPA
does not have information available about the actual activities of the sampled workers and the
representativeness of the facility engineering controls to modern practice.
• Polyethylene Terephthalate (PET) Manufacturing. Risk estimates were derived using
personal breathing zone monitoring data collected from 1985 to 1994 at five facilities linked to
PET manufacturing. Cancer risk estimates for inhalation exposure range from 1.8x10 3 for
central tendency exposures to 2.3xl0~2 for high-end exposures. However, there is uncertainty in
the risk estimates. The monitoring data used in this analysis are limited {i.e., 11 samples from
five sites). It also is not known how manufacturing processes and workplace conditions have
changed since 1994. For instance, EPA does not have information available about the actual
activities of the sampled workers and the representativeness of the facility engineering controls
to modern practice.
• Ethoxylation Processes. Risk estimates were derived using one composite 8-hour time-weighted
average personal breathing zone sample collected from one worker in 2000 at a soap and
detergent manufacturing facility. Cancer risk estimates for inhalation exposure range from
4,6x 10 4 for central tendency exposures to 5,9x 10 4 for high-end exposures. However, there is
uncertainty in the risk estimates. Specifically, EPA is unable to model these occupational
exposures and cannot determine the statistical representativeness of the one monitoring data
point {e.g., high-end, central tendency) towards potential exposures from this OES. Additionally,
there is uncertainty as to the worker activities covered by this monitoring data and whether all
foreseeable activities, corresponding exposures, and workplace operations are represented.
• Hydraulic Fracturing Operations. Risk estimates were derived from occupational exposures
modeled using Monte Carlo simulations for the worker activities of container unloading,
container cleaning, and equipment cleaning. Cancer risk estimates for inhalation exposure range
from 7,Ox 10 5 for central tendency exposures to 9,5x ] 0 3 for high-end exposures. However,
there is uncertainty in the risk estimates. Specifically, there is uncertainty as to the
representativeness of some of the model input data and, therefore, the subsequent calculated
exposures to the actual distribution of hydraulic fracturing occupational exposures. This is due to
limitations of using generic industry values identified for the hydraulic fracturing industry as
well as self-reported values from FracFocus as model parameters. FracFocus data may not fully
represent operations across multiple sites throughout the United States as only certain sites
volunteered to submit data.
Overall confidence in risk estimates for occupational dermal exposures is medium for all OES/COUs
because the same modeling approach was used for all OES/COUs. As described in Section 4.3 overall
confidence in the oral and dermal cancer slope factor underlying these risk estimates is medium-high. As
described in Section 3.3.1.2, the dermal exposure modeling methodology is supported by moderate
evidence, with model input parameters from literature sources, a European model, standard defaults
from the ChemSTEER User Guide ( 015a). and 1,4-dioxane product concentration data from
process information. These sources range from slight to robust, depending on factors such as age and
applicability to OES/COU. The modeling is limited by the use of standard input parameters that are not
Page 139 of 484
-------
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
specific to 1,4-dioxane and a lack of variability in dermal exposure for different worker activities.
Differences in the dermal exposure modeling across COUs are driven primarily by COU-specific weight
fractions of 1,4 dioxane and the independent assessment of evaporative impacts in commercial and
industrial settings. Therefore, EPA's overall confidence in the occupational dermal risk estimates is
medium.
5.2.2 Summary of Risk Estimates for the General Population
5.2.2.1 Drinking Water - Surface Water Pathway
Risks from drinking water exposures were evaluated using a series of analyses that provide information
about the specific contributions of releases associated with individual OESs as well as information about
aggregate exposures and risks that could result from multiple sources releasing to the same water body.
Because most reasonably available surface water and drinking water monitoring data are not co-located
with 1,4-dioxane release sites, this analysis relies primarily on drinking water concentrations modeled
based on reasonably available release information. Risks predicted based on reasonably available
monitoring data are presented in 5.2.2.1.1.
EPA estimated cancer and non-cancer risks for adults, children, and formula-fed infants exposed to 1,4-
dioxane in drinking water. All risk estimates presented in this summary focus on the scenario with the
greatest potential exposure and risk. Because adult drinking water exposures relative to body weight are
greater than exposures relative to body weight averaged over the course of childhood (as illustrated in
1,4-Dioxane Supplemental Information File: Drinking Water Exposure and Risk Estimates for 1,4-
Dioxane Release to Surface Water from Individual Facilities ( 2023hV) the drinking water
exposure scenario with the greatest lifetime cancer risk is 33 years of drinking water exposure as an
adult. Similarly, because cancer risk is the primary risk driver in most exposure scenarios, this summary
of results focuses on cancer risk estimates. More comprehensive sets of risk estimates for non-cancer
effects and other exposure scenarios are presented in the supplemental files referenced throughout this
section.
Drinking water exposure and risk estimates are highly dependent on the amount of 1,4-dioxane released
and the flow of the receiving water body. Both of these factors vary substantially across facilities within
each COU/OES, making release amount and flow much more important predictors of risk than a
facility's identified COU/OES. Exposure and risk estimates are also influenced by whether there is a
drinking water intake downstream of a release and the degree of dilution that occurs between the point
of release and the drinking water intake. Many of the risk estimates presented in the sections that follow
(for facility-specific releases, DTD, hydraulic fracturing, and aggregate modeling) assume that no
additional downstream dilution occurs prior to reaching drinking water intakes. This represents an upper
end estimate of exposure and risk based on the available data and the potential for intakes to be directly
downstream of a releasing facility. EPA conducted further analysis of the facility-specific releases to
consider the potential impact of downstream dilution on actual concentrations at drinking water intakes
and resulting risk estimates. Even when accounting for dilution between known releases and identified
drinking water intake locations, water concentrations estimated at drinking water intakes, instances of
cancer risks greater than 1 in 1 million for some public water systems are identified. Proximity of
releases to drinking water intakes and dilution are further discussed in Section 2.3.1.2.4/Appendix G.2.4
and Section 5.2.2.1.2.
1,4-Dioxane is not readily removed through typical wastewater or drinking water treatment processes.
Therefore, the drinking water risk estimates presented below are derived based on the assumptions that
drinking water intakes are located near 1,4-dioxane release sites and that no 1,4-dioxane is removed by
Page 140 of 484
-------
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
POTWs or through drinking water treatment. Use of source water estimated concentrations of 1,4-
dioxane to calculate cancer risk estimates is considered protective of all systems. These assumptions are
further discussed in Section 2.3.1.1/Appendix G.1.2).
5.2.2.1.1 Risks from Exposure to Drinking Water Concentrations Indicated in
Finished Drinking Water Monitoring Data
EPA evaluated risks for 1,4-dioxane concentrations reported in the reasonably available finished
(treated) drinking water monitoring data. Monitoring data included in this analysis were from
generalized, broad monitoring strategies, rather than targeted efforts to assess areas of known
contamination. As previously illustrated in Figure 2-10, 1,4-dioxane was below limits of detection for 89
percent of finished drinking water samples included in UCMR3 and state databases. Table 5-2
summarizes the distribution of lifetime cancer risk estimates from 1,4-dioxane concentrations detected
in finished drinking water reported in these databases (described in Section 2.3.1.1). This drinking water
monitoring data provides evidence that 1,4-dioxane is present in some finished drinking water and may
contribute to cancer risks in locations at the high end of monitored drinking water concentrations.
Monitoring data may not include the full range of 1,4-dioxane concentrations that result from industrial
releases. As discussed in Section 2.3.1.1, available drinking water monitoring data do not necessarily
capture locations that are most impacted by releases temporally or spatially and they often reflect
concentrations at a single point in time rather than average concentrations. However, as described in
Appendix G.2.3.2, in locations where monitoring data are available near release sites, comparisons
demonstrate strong concordance between modeled concentrations and monitoring data. EPA's
evaluation of drinking water risks therefore primarily relied on modeled estimates of 1,4-dioxane
concentrations that occur near release sites.
Table 5-2. Lifetime Cancer Risk Estimates for 1,4-Dioxane Concentrations Detected in Finished
Drinking Water
Percentile Drinking Water Monitoring Data
Min
5%
25%
Median
75%
90%
95%
Max
Water conc.
(Mg/L)
2.00E-03
3.50E-02
3.50E-02
3.50E-02
3.50E-02
7.93E-02
0.16
13.3
Lifetime
Cancer Risk
4.02E-11
1.95E-08
1.95E-08
1.95E-08
8.37E-08
2.79E-07
3.46E-07
7.42E-06
Lifetime cancer risks based on 33 years of oral exposure through drinking water as an adult. Percentiles reflect
concentrations across the distribution of available drinking water monitoring data (this distribution includes non-
detects as half the detection limit).
5.2.2.1.2 Risks from Exposures to Water Concentrations Modeled from Industrial
Releases
To estimate the contribution of industrial releases to general population risks from drinking water, EPA
calculated cancer and non-cancer risk estimates based on modeled surface water concentrations in
receiving water bodies described in Section 2.3.1.3.1 and the resulting drinking water exposures
calculated as described in Section 3.2.2. Because there is substantial variation and uncertainty around
the extent of dilution that may occur in the receiving water body between the point of release and the
locations of drinking water intakes, EPA calculated cancer risk estimates under a range of reasonable
downstream dilution assumptions.
Page 141 of 484
-------
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Figure 5-1 shows the distribution of cancer risk estimates for industrial releases reported to TRI and
DMR, assuming that concentrations at drinking water intakes are the same as concentrations that occur
at the point of release after initial mixing in the receiving water body. Based on available data, this is a
plausible scenario in some locations. Lifetime cancer risk estimates are based on 33 years of exposure as
an adult and range from 5.41 x 10~13 to 2.54x 10~2. The median cancer risk estimate for these modeled
concentrations is 2.32/ 10 6 and the 95th percentile risk estimate is 4.92/ 10 3. Acute and chronic non-
cancer risk estimates for some facilities (data not shown) also indicate potential for non-cancer risk
relative to benchmark MOEs. This analysis represents an upper bound drinking water exposure scenario
in which intakes are located near the point of release or in which minimal additional dilution occurs
downstream. Complete cancer and non-cancer risk estimates for facility and OES-specific releases are
presented in 1,4-Dioxcme Supplemental Information File: Drinking Water Exposure and Risk Estimates
for 1,4-Dioxane Release to Surface Water from Individual Facilities (U.S. EPA. 2023h).
1
c
.2
1
1
1
1
1 IO
CD
1 ~
1
1
1
1
.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
m
i 1 1 1 1 1 1 1 1 1 1 —i r
io~13 icr12 io_" icr10 10"S io_s 10"7 icr6 icr5 10"4 «r3 io~2 io~1
Adult Lifetime Cancer Risk
Figure 5-1. Distribution of Adult Lifetime Cancer Risk across all Facilities, Assuming No
Additional Dilution Occurs between the Point of Release and the Location of Drinking
Water Intakes
There is substantial variation in cancer risk estimates both within and across OESs. The large ranges of
modeled water concentrations and corresponding risk estimates reflect the large differences in the
amount of 1,4-dioxane released from facilities, the magnitude of flow within the receiving water body or
both.
For facilities where specific release amounts or locations are not reported, release amounts and flow
rates are based on conservative assumptions that may result in high risk estimates. There is uncertainty
around risk estimates for those facilities with limited release information, but facility-specific
information on release amounts and locations was available for most facilities. Therefore, while facility-
specific risk estimates based on facilities with limited information should be interpreted with caution,
most estimates are informed by moderate to robust modeling approaches and input data. To determine
the extent to which inclusion of facilities with limited release information influences the overall
distribution, EPA repeated this risk estimate analysis presented in Figure 5-1 using only facilities for
which high quality release data are available (Figure 5-2). Specifically, this additional analysis is limited
to facilities for which the annual release amount was sourced from either TRI Form R or DMR, and the
receiving water body reach code was identified in the facility's NPDES permit. Out of the 120 total
Page 142 of 484
-------
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
direct and indirect releases evaluated in this section, 80 met these strict data criteria. The resulting
distribution of risk estimates are similar to the results of the analysis including all facilities, ranging
from 5.41 xlO-13 to 2.54x10-2, with a median of 8.51><10~7 and 95th percentile of 4.92xl0~3.
c
.2
LO
CD
5
O-
n
i
10 12 10 11 1(T10 1CT9 1CTa 1(T7 10" 10 s 10~4 10"s 10~2 10"'
Adult Lifetime Cancer Risk
Figure 5-2. Distribution of Adult Lifetime Cancer Risk across Facilities with High
Quality Release Data, Assuming No Additional Dilution Occurs between the Point of
Release and the Location of Drinking Water Intakes
The risk estimates summarized in Figure 5-1 and Figure 5-2 rely on the assumption that concentrations
at drinking water intakes are the same as concentrations estimated near the point of release. To evaluate
the validity of that assumption, EPA considered the proximity of release sites to downstream drinking
water intake locations for community and non-community non-transient PWSs. As shown in Table 5-3,
of the 69 facilities with cancer risk greater than 1 x 10~6, 22 (32 percent) have a downstream drinking
water intake within 250 km and 4 of those have a drinking water intake within 10 km. A detailed
description of this analysis is provided in Appendix G.2.4.
Table 5-3. Proximity of Nearest Downstream Drinking Water Intakes to Facilities Resulting in
Cancer Risk Greater than lxlO-6
Total
Facilities
Evaluated
Facilities with
Cancer Risk
above 1E-06
DWI within
250 km
DWI within
100 km
DWI within
50 km
DWI within
25 km
DWI within
10 km
120
69
22
17
11
7
4
The portion of 1,4-dioxane that remains after the additional dilution that occurs as it travels downstream
is highly variable based on site-specific characteristics, ranging from less than 1 percent to nearly 100
percent of the original concentrations (Figure 5-3). The site-specific factors that influence this additional
downstream dilution may not be fully captured in a national-scale assessment. Based on available site-
specific information for each facility, the mean modeled dilution predicted at downstream drinking
water intakes is diluted to 1 percent of original concentrations estimated in receiving water bodies near
the point of release.
Page 143 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
I ^<9
LO LJ ^
CM _c 0 40
•E | g
(/J o .<£
d) (0 (Y
1 0
£ CO
o £ CO 20
- rod
0 0 t_
-Q £= 0
E <2 ro
Q
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
0 25 50 75 100
Diluted Concentration at Intake as a Percent
of the Concentration at the Point of Release
Figure 5-3. Distribution of Dilution of 1,4-Dioxane Concentrations at Downstream
Drinking Water Intakes
Figure 5-4 shows the distribution of cancer risk estimates for industrial releases, assuming that
concentrations at drinking water intake locations are diluted to 1 percent of the original 1,4-dioxane
concentrations in surface water estimated at the point of release. Lifetime cancer risk estimates for these
modeled concentrations range from 5.41 x 10~15 to 2.54x 10~4. The median cancer risk estimate is
8.51><10~9 and the 95th percentile risk estimate is 4.92xl0~5. This represents a plausible drinking water
exposure scenario consistent with mean modeled downstream dilution predicted across all facilities
based on available site-specific information.
20-
16-
cn
0
u)
cu
0
0 .£
Cd CQ
O W
c
0
3
4-
0-
10
-14
10",J 10",z 10 " 10~IU 10"M 10"° 10" 10"° 10"° 10
Adult Lifetime Cancer Risk
-4 "10 3
Figure 5-4. Distribution of Adult Lifetime Cancer Risk across all Facilities,
Assuming Dilution to 1% of Initial Concentrations in the Receiving Water Body
In addition to estimating how the overall distribution of cancer risk estimates would shift based on a
standard assumption of downstream dilution to 1 percent of original concentrations (as shown in Figure
5-4), EPA also estimated water concentrations and risks that may occur at specific drinking water
Page 144 of 484
-------
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
intakes located downstream of releasing facilities, based on location-specific estimates of dilution. For
the 22 facilities with cancer risk greater than 1 in 1 million and drinking water intakes located within
250 km downstream, EPA identified downstream intakes associated with 73 distinct PWSs.
Even when accounting for site-specific influences on dilution, EPA modeled concentrations that would
result in adult lifetime cancer risk in excess of 1 in 1 million at intakes for 20 of the PWSs identified
through this assessment, serving a combined population of 2,124,000 people. Adult lifetime cancer risk
estimates were greater than 1 in 100,000 for 5 of these public water systems, serving a combined
population of 834,000 people. This analysis also identified locations with multiple releasing facilities
upstream of the same drinking water intake; however, in all such cases the aggregated adult lifetime
cancer risk calculated at the intake for the aggregated diluted concentration was less than 1 in a million.
A detailed description of this analysis is provided in Appendix G.2.4. Overall confidence in these
dilution-adjusted risk estimates is high for drinking water intakes located at or near the point of release,
but confidence decreases substantially with increasing distance downstream. This analysis does not
provide a comprehensive survey of modeled 1,4-dioxane concentrations at all drinking water intakes.
There may be additional drinking water intakes downstream of facilities releasing 1,4-dioxane that are
not accounted for in the intake database used in this analysis.
Overall, these analyses indicate that in many locations, downstream dilution may be expected to
substantially reduce 1,4-dioxane concentrations at the point of drinking water intakes. However, even
when accounting for dilution, upstream industrial releases reported to TRI or DMR contribute to cancer
risk estimates greater than 1 in a million or 1 in 100,000 at known drinking water intake locations.
The set of distributions presented in Figure 5-1, Figure 5-2, and Figure 5-4 indicate that high risks can
occur in specific locations downstream of release sites due to factors such as the size of the releasing
event(s), stream flow volume, proximity of the release site to drinking water intake, and limited drinking
water treatment removal from typical treatment methods.
Overall confidence in the overall distribution of risk estimates for drinking water exposures resulting
from facility releases is medium-high. Overall confidence in site-specific risk estimates for individual
facility releases varies both within and across OES, depending on the confidence in the source-specific
release data. As described in Section 4.3 overall confidence in the oral and dermal cancer slope factor
underlying these risk estimates is medium-high. As described in Section 3.3.2.1, the overall exposure
modeling methodology used for this analysis is supported by moderate evidence. It is designed to
estimate water concentrations expected at specific locations. Exposure estimates for this scenario are
based on some conservative assumptions about flow rates and release frequency and amount. For most
COUs, this analysis is limited to facilities that report via TRI and/or DMR. Other sources releasing
smaller amounts of 1,4-dioxane are not directly captured. Available monitoring data confirm that 1,4-
dioxane is present in some surface water and drinking water, though most of the available data were not
collected near release sites are therefore not directly comparable.
The overall level of confidence in facility-specific release estimates and resulting risk estimates depends
on the source of the release data described in Appendix E.3:
• Overall confidence in drinking water exposure estimates is medium to high for OESs/COUs that
rely primarily on release data reported to DMR or to TRI via Form R. Most COUs/OESs are
included in this group.
• Overall confidence in drinking water exposure estimates is medium for OESs/COUs for which
release estimates are based on reporting to TRI via Form A. The Import and Repackaging OES
Page 145 of 484
-------
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
releases used in this analysis are entirely based on Form A reporting of releases, and just under
half of the Industrial Uses OES releases were reported via Form A.
While confidence in facility-specific risk estimates varies, estimates for most facilities are informed by
moderate to robust modeling approaches and input data. Furthermore, the overall distribution is not
meaningfully altered by exclusion of facility-specific data based on more limited release information (as
illustrated in Figure 5-1 and Figure 5-2). There is some uncertainty around the proximity of releases to
drinking water intake locations and the extent to which 1,4-dioxane is further diluted prior to reaching
intake locations. EPA therefore estimated distributions of cancer risk estimates under a range of
assumptions about downstream dilution, reflecting the range of plausible drinking water intake
scenarios, as indicated by available site-specific information.
5.2.2.1.3 Risks from Exposures to Water Concentrations Modeled from DTD
Releases (from POTWs), Assuming No Downstream Dilution
EPA evaluated the potential contribution of DTD releases of consumer and commercial products to
drinking water exposure and risk. Surface water concentrations at the point of DTD releases via POTWs
are primarily determined by the size of the population contributing to DTD releases and the flow rates of
receiving water bodies. Risk estimates presented in this section are not tied to known releases at specific
locations. Rather, this analysis defines the conditions under which DTD releases would result in varying
levels of risk.
Cancer risk estimates shown in Table 5-4 were calculated based on drinking water exposure estimates
presented in Section 3.2.2.1.2, which correspond to surface water concentrations estimated by
probabilistic modeling of DTD releases under varying population sizes and stream flows. The resulting
risk estimates indicate that risk is highest in locations where large populations are contributing to DTD
releases and those releases are ultimately discharged to streams with low flow. Cancer risk estimates
greater than 1 in a million were seen in combinations of population size and flow rates that are likely to
occur in some locations. As described in Section 2.3.1.3.2, EPA considered the frequency of the varying
combinations of population sizes and flow rates. For communities with single POTWs treating
wastewater, most fell into the range of 100 to 10,000 people, with the annual average flow of the
receiving water body less than 300 cfs (Table 2-10). Cancer risk estimates for communities in this range
of population sizes are as low as 2.04x 10~8 at flows of 300 cfs and increase at lower flows. For example,
cancer risk estimates for releases from a population size of 10,000 could be as high as 2.04x 10~6 at a
flow of 30 cfs. Acute and chronic non-cancer risk estimates for these scenarios do not indicate non-
cancer risk relative to benchmark MOEs. For reference, stream flows of 100 cfs might be considered a
small river, while anything less than 100 cfs would be considered a stream or creek. Complete cancer
and non-cancer risk estimates for the range of water concentrations from DTD releases estimated under
varying conditions using probabilistic modeling are presented in 1,4-Dioxane Supplemental Information
File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Surface Water Concentrations
Predicted with Probabilistic Modeling ( ,0231).
Overall confidence in risk estimates for drinking water exposures resulting from DTD releases under
varying population and stream flow conditions is medium. As described in Section 4.3 overall
confidence in the oral and dermal cancer slope factor underlying these risk estimates is medium-high. As
described in Section 3.3.2.1, the exposure modeling methodology used for this analysis is supported by
robust evidence and is designed to provide a nationally representative distribution of estimated water
concentrations under varying conditions. Exposure estimates rely on estimated distributions of DTD
releases of consumer and commercial products for each COU. Distributions of DTD releases of
consumer and commercial products were estimated for each COU on a per capita basis using the
Page 146 of 484
-------
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
SHEDS-HT model. Because this analysis is not tied to specific sites, there is uncertainty around the
proximity of releases to drinking water intake locations and the extent to which 1,4-dioxane is further
diluted prior to reaching intake locations. For this analysis, EPA assumed that no additional dilution
occurs prior to reaching drinking water intakes. While confidence in the individual contribution from
some specific COUs is lower, confidence in estimates of overall DTD releases is moderate.
Table 5-4. Lifetime Cancer Risk" Estimates from DTD Releases Alone (at the Point of Release)
under a Range of Population and Flow Rate Scenarios
Population Contributing to Down-the-Drain Releases
100
1,000
10,000
100,000
1,000,000
Stream Flow
(cfs)
100
6.1 1E 09
6.1 1E-08
6.1 1E-07
6.1 1E 06
6.1 1E-05
300
2.04E-09
2.04E-08
2.04E-07
2.04E-06
2.04E-05
1,000
6.11E-10
6.11E-09
6.11E-08
6.11E-07
6.11E-06
3,000
2.04E-10
2.04E-09
2.04E-08
2.04E-07
2.04E-06
10,000
6.11 E— 11
6.11E-10
6.11E-09
6.11E-08
6.11E-07
30,000
2.04E-11
2.04E-10
2.04E-09
2.04E-08
2.04E-07
"Lifetime cancer risks based on 33 years of oral exposure through drinking water as an adult.
The frequencies of each of these combinations of population size and flow rate are presented in Table 2-10.
5.2.2.1.4 Risks from Exposure to Drinking Water Concentrations Modeled from
Disposal of Hydraulic Fracturing Produced Waters to Surface Water,
Assuming No Downstream Dilution
EPA evaluated the potential contribution of the disposal of hydraulic fracturing produced waters to
surface water by aggregating exposures and risks. The range of water concentrations that may result
from releases of hydraulic fracturing waste to surface water were estimated using probabilistic
modeling. Risk estimates presented in this section are not tied to known releases at specific locations.
Rather, this analysis defines the conditions under which releases from hydraulic fracturing would result
in varying levels of risk. These risk estimates are based on the assumption that 1,4-dioxane is not
removed by POTWs or through drinking water treatment.
Cancer risk estimates across the full distribution of modeled releases are presented in Table 5-5. Cancer
risk estimates are 3.57><10~8 for median modeled releases and 1.45><10~6 for 95th percentile modeled
releases. Acute and chronic non-cancer risk estimates based on 95th percentile modeled releases do not
indicate risk relative to benchmark MOEs. The maximum water concentration estimated by the model
reflects a scenario in which waste is released to a stream with very low flow. EPA does not have site-
specific information to indicate that such a scenario actually occurs. Complete risk estimates for the
range of water concentrations from DTD releases estimated under varying conditions using probabilistic
modeling are presented in 1,4-Dioxane Supplemental Information File: Drinking Water Exposure and
Risk Estimates for 1,4-Dioxane Surface Water Concentrations Predicted with Probabilistic Modeling
( >02311
Overall confidence in risk estimates for drinking water exposures resulting from hydraulic fracturing
releases is medium. As described in Section 4.3 overall confidence in the oral and dermal cancer slope
factor underlying these risk estimates is medium-high. As described in Section 3.3.2.1, the exposure
modeling methodology used for this analysis is supported by robust evidence and is designed to provide
a nationally representative distribution of estimated water concentrations under varying conditions.
Releases used as inputs in the model were estimated using Monte Carlo modeling that captures
Page 147 of 484
-------
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
variability across sites. However, the modeled exposure estimates are not directly tied to specific
releases at known locations, decreasing the strength of the evidence related to the representativeness of
the exposure estimates for actual exposures. There is some uncertainty around the proximity of releases
to drinking water intake locations and the extent to which 1,4-dioxane is further diluted prior to reaching
intake locations. For this analysis, EPA assumed that no additional dilution occurs prior to reaching
drinking water intakes.
Table 5-5. Lifetime Cancer Risks Estimated from
Hydraulic Fracturing Produced Waters Disposed
to Surface Water under a Range of Scenarios
Monte Carlo
Distribution
Adult Lifetime Cancer Risk
Maximum
1.10E-04
99th Percentile
4.35E-06
95th Percentile
1.45E-06
Median
3.57E-08
5th Percentile
1.29E-10
Minimum
3.40E-18
LADDs used to calculate these cancer risk estimates are
presented in Table 3-5.
5.2.2.1.5 Aggregate Risks from Drinking Water Exposures Modeled from Multiple
Sources Releasing to Surface Water, Assuming No Downstream Dilution
Multiple sources may contribute to 1,4-dioxane concentrations in drinking water sourced from surface
water in a single location. EPA therefore estimated aggregate general population exposures and risks
that could occur as a result of combined contributions from multiple sources. As described in Section
2.3.1.3.4, EPA used probabilistic modeling to predict aggregate surface water concentrations that could
occur when accounting for DTD releases, indirect releases, and other upstream sources. EPA estimated
cancer and non-cancer risks for the drinking water exposure estimates in Section 3.2.2, which
correspond to the modeled aggregate surface water concentrations described in Section 2.3.1.3.4 and
assume that no 1,4-dioxane is removed through treatment. This analysis also assumes that
concentrations at drinking water intakes are not further diluted from the concentrations modeled near the
point of release. There is wide variation in both cancer and non-cancer risk within and across
OESs/COUs when taking into account aggregate contributions from other sources. This variation is
illustrated in the cancer risk estimates shown in the distributions of cancer risk estimates for exposures
modeled for each OES/COU in Figure 5-5. The large ranges of risk estimates for some OESs/COUs
reflect substantial variation in releases and characteristics of receiving water bodies across the set
facilities associated with those OESs. High-end cancer risk estimates in this analysis are very similar to
high-end risk estimates for individual facility releases alone, indicating that high-end estimates are
driven primarily by high-end industrial releases. Complete cancer and non-cancer risk estimates for the
range of aggregate water concentrations estimated for each COU using probabilistic modeling are
presented in 1,4-Dioxane Supplemental Information File: Drinking Water Exposure and Risk Estimates
Page 148 of 484
-------
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
for 1,4-Dioxane Surface Water Concentrations Predicted with Probabilistic Modeling (
20231)-
Overall confidence in distributions of risk estimates for drinking water exposures resulting from for
aggregate surface water concentrations predicted by probabilistic modeling varies across OES/COU.
While confidence is not uniform for all facilities within an OES, overall confidence ratings for each
OES are intended to communicate how the factors that contribute to confidence and uncertainty vary
across COUs. As described in Section 4.3 overall confidence in the oral and dermal cancer slope factor
underlying these risk estimates is medium-high. As described in Section 3.3.2.1, the exposure modeling
methodology used for this analysis is supported by robust evidence and is designed to provide a
nationally representative distribution of estimated water concentrations under varying conditions. For
most COUs, this analysis is limited to facilities that report releases via TRI and/or DMR. Other sources
releasing smaller amounts of 1,4-dioxane are not directly captured, though the distribution of surface
water monitoring data used to represent background concentrations in the model is intended to capture
these other upstream sources. Available monitoring data confirm that 1,4-dioxane is present in some
surface water and drinking water, though most of the available data were not collected near release sites
and are therefore not directly comparable. In release locations where monitoring data are available, case
studies demonstrate strong concordance between modeled estimates and measured surface water
concentrations.
There is some uncertainty around the proximity of releases to drinking water intake locations and the
extent to which 1,4-dioxane is further diluted prior to reaching intake locations. For this analysis, EPA
assumed that no additional dilution occurs prior to reaching drinking water intakes. The characterization
of downstream dilution presented in 5.2.2.1.2 for individual facility releases illustrates the extent to
which downstream dilution may impact overall risk estimates.
The overall level of confidence in resulting exposure estimates depends on the source of OES/COU-
specific release data described in Appendix E.3:
• Overall confidence in drinking water exposure estimates is medium to high for OESs/COUs that
rely primarily on release data reported to DMR or to TRI via Form R. Most COUs/OESs are
included in this group.
• Overall confidence in drinking water exposure estimates is medium for OESs/COUs for which
release estimates are based on reporting to TRI via Form A. The Import and Repackaging OES
releases used in this analysis are entirely based on Form A reporting of releases, and just under
half of the Industrial Uses OES releases were reported via Form A.
• Overall confidence in drinking water exposure estimates is low to medium for OESs/COUs for
which release estimates are based on surrogate or modeled information.
Page 149 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4523
4524
4525
4526
30-
20-
10-
40-
w 30 -
3
v.)
CL
Ethoxylation byproduct
c
gj
a>
-=d
lO
CJ>
a.
Functional Fluids (Open-System)
c
<5
TJ
CD
J
LO
o
Q.
illli]
Import and Repackaging
_dn tH
LO
CT>
Q_
Industrial Uses
LO
o
CL
Manufacture
LO
CT>
CL
PET Manufacturing
~u
0)
LO
a>
CL
Printing Inks
Remediation
-=~£1
LO
O)
CL
imlrL
10~J 10"^ 10"1 10"1U 10"3 10"B 10"' 10~b 10"& 10"4 10"J 10"* 10"1 10~1U 10"
Modeled Adult Lifetime Cancer Risk from Total 1,4-Dioxane (Releases + Background)
Figure 5-5. Histograms of Lifetime Cancer Risk Estimates for Aggregate Water Concentrations Estimated Downstream of COUs
with Vertical Lines Showing the Median and 95th Percentile (P95) Values
Page 150 of 484
-------
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
5.2.2.1.6 Integrated Summary of Drinking Water Risk Estimates across Multiple
Lines of Evidence for Surface Water
Risks from drinking water exposures were evaluated using a series of analyses that provided information
about the specific contributions of releases associated with individual sources as well as aggregate
exposures and risks to the general population. This analysis finds cancer risk estimates greater than 1 in
1 million from drinking water exposures informed by both monitoring data and modeled surface water
concentrations. Modeled concentrations result in cancer risk estimates greater than 1 in 1 million across
a range of individual sources and aggregate sources utilizing plausible drinking water exposure
scenarios.
Monitoring data demonstrates that 1,4-dioxane is present in some source water and finished drinking
water samples. Measured concentrations in finished drinking water samples resulted in cancer risk
estimates greater than 1 in 1 million at the high end of the distribution of monitoring samples. Most
drinking water treatment systems are not expected to remove 1,4-dioxane from water, suggesting that
concentrations detected in source water can also be an indication of concentrations in drinking water.
Available monitoring data provided information about general population exposures but did not capture
high concentrations occuring in specific locations or at specific times from direct and indirect releases
into water bodies. Therefore, EPA relied on estimated concentrations modeled for a range of specific
release scenarios, including direct and indirect industrial releases, DTD releases, disposal of hydraulic
fracturing waste, and aggregate concentrations resulting from varying combinations of multiple sources
to characterize risks from the water pathway. EPA evaluated the performance of the models used to
estimate water concentrations with monitoring data from site-specific locations serving as cases studies.
These case study comparisons demonstrated strong concordance between modeled concentrations and
monitoring data, thereby increasing confidence in risk estimates based on modeled concentrations.
Across all modeled scenarios, 1,4-dioxane concentrations in water are primarily determined by the
amount of release from varying sources and the flow of the receiving water body. These two factors are
highly location and source-specific, resulting in very wide ranges of modeled water concentrations and
risk estimates for each set of analyses presented in the previous section above.
Risk estimates based on 1,4-dioxane concentrations modeled in the receiving water bodies at the point of
release show potential for risk greater than 1 in 1 million or 1 in 100,000 from each of the sources
assessed.
As described in Section 5.2.2.1.2, dilution that occurs between the point of release and drinking water
intake locations may be expected to reduce 1,4-dioxane concentrations in some locations. However,
even when accounting for dilution, upstream releases contribute to cancer risk estimates greater than 1 in
a million or 1 in 100,000 at some drinking water intake locations. EPA evaluated risks based on
modeled water concentrations for a sample of drinking water intake locations downstream of releases
where risk was greater than 1 in 1 million. After accounting for additional dilution, cancer risk estimates
remained greater than 1 in 1 million for 27 percent of the public water systems evaluated, serving a
combined population of over 2 million people.
The potential relative contribution from different sources varies under different conditions and is likely
to be site-specific. For example, high-end risk estimates in the aggregate model (presented in Section
5.2.2.1.5) are very similar to high end risk estimates for facility releases alone (presented in Section
5.2.2.1.2), suggesting that in cases where industrial releases are high, those releases will be the dominant
Page 151 of 484
-------
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
source of 1,4-dioxane in water. On the other hand, under some conditions (presented in Section
5.2.2.1.3), plausible DTD release scenarios may present risk greater than 1 in 1 million in the absence of
industrial releases. Taken together, the analyses presented throughout this section demonstrate that each
of these sources may contribute to drinking water risks under some conditions. These analyses define the
conditions under which different levels of risk may occur.
5.2.2.2 Drinking Water - Groundwater and Disposal Pathways
EPA estimated risks from general population exposures that could occur if groundwater containing 1,4-
dioxane is used as a source of drinking water. Risk estimates presented in this section are not tied to
known releases at specific locations. Rather, this analysis defines the conditions under which 1,4-
dioxane disposal to landfills or from hydraulic fracturing operations could result in varying levels of
risk.
Cancer and non-cancer risk estimates were calculated based on modeled groundwater concentrations
described in Section 2.3.1.4 and corresponding drinking water exposures estimates described in Section
3.2.2.2. All risk estimates presented in this summary focus on the scenario with the greatest potential
exposure and risk. Because adult drinking water exposures relative to body weight are greater than
exposures relative to body weight that occur over the course of childhood (as illustrated in 1,4-Dioxane
Supplemental Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Release
to Surface Water from Individual Facilities ( 2023h)), the scenario with the greatest lifetime
cancer risk is 33 years of exposure as an adult.
For potential groundwater concentrations resulting from landfill leachate, EPA estimated cancer and
non-cancer risks for adults and formula-fed infants at concentrations estimated under varying
hypothetical combinations of leachate concentrations and loading rates. As shown in Table 5-6, lifetime
cancer risk estimates increase under scenarios with higher leachate concentrations and loading rates.
Chronic non-cancer risk estimates (not shown) indicate risk relative to the benchmark MOE only at the
highest leachate concentrations and loading rates. These concentrations and loading rates represent a
scenario where 1,4-dioxane is either delisted and released to a municipal solid waste landfill or when
trace concentrations present in consumer and commercial products are disposed to those same landfills.
Though the higher concentrations of 1,4-dioxane in leachate and higher loading rates are less likely, they
may represent a sentinel PESS exposure. Complete results for cancer and non-cancer risk are available
in 1,4-Dioxane Supplemental Information File: Drinking Water Exposure and Risk Estimates for 1,4-
Dioxane Land Releases to Landfills ( 2023f).
Overall confidence in risk estimates for drinking water exposures resulting from disposal to landfills is
low to medium. As described in Section 4.3 overall confidence in the oral and dermal cancer slope
factor underlying these risk estimates is medium-high. As described in Section 3.3.2.2.1 the modeling
methodology is robust. However, the release information relied on as model input data is supported by
slight to moderate evidence, decreasing overall confidence. In addition, this drinking water exposure
scenario relies on the assumption that the estimated groundwater concentrations may occur in locations
where groundwater is used as a primary drinking water source. While the substantial uncertainty around
the extent to which these exposures occur decreases overall confidence in the exposure scenario, this
scenario represents a sentinel exposure.
Page 152 of 484
-------
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 5-6. Lifetime Cancer Risks" Estimated for Modeled Groundwater Concentrations Estimated
under Varying Landfill Conditions
Leachate
Concentration
(mg/L)
Loading Rate (lb)
0.1
1
10
100
1,000
10,000
100,000
1,000,000
0.0001
3.1E-17
2.9E-16
3.6E-15
3.4E-14
3.3E-13
3.1E-12
3.0E-11
2.9E-10
0.001
3.1E-16
2.9E-15
3.6E-14
3.4E-13
3.3E-12
3.1E-11
3.0E-10
2.9E-09
0.01
3.1E-15
2.9E-14
3.6E-13
3.4E-12
3.3E-11
3.1E-10
3.0E-09
2.9E-08
0.1
3.1E-14
2.9E-13
3.6E-12
3.4E-11
3.3E-10
3.1E-09
3.0E-08
2.9E-07
1
3.1E-13
2.9E-12
3.6E-11
3.4E-10
3.3E-09
3.1E-08
3.0E-07
2.9E-06
10
3.1E-12
2.9E-11
3.6E-10
3.4E-09
3.3E-08
3.1E-07
3.0E-06
2.9E-05
100
3.1E-11
2.9E-10
3.6E-09
3.4E-08
3.3E-07
3.1E-06
3.0E-05
2.9E-04
1,000
3.1E-10
2.9E-09
3.6E-08
3.4E-07
3.3E-06
3.1E-05
3.0E-04
2.9E-03
10,000
3.1E-09
2.9E-08
3.6E-07
3.4E-06
3.3E-05
3.1E-04
3.0E-03
2.9E-02
a Lifetime cancer risks based on 33 years of oral exposure through drinking water as an adult.
For groundwater concentrations resulting from disposal of hydraulic fracturing produced water, EPA
calculated cancer and non-cancer risks for adults and formula-fed infants. Cancer risk estimates across
the full distribution of modeled releases are presented in Table 5-7. Cancer risk estimates are 4.0E-07 for
median modeled releases and 8.6xl0~6 for 95th percentile modeled releases. Chronic non-cancer risk
estimates are above the corresponding benchmark MOE for all modeled groundwater concentrations,
indicating lower non-cancer risk from non-cancer effects. Complete cancer and noncancer risk
calculations are available in 1,4-Dioxane Supplemental Information File: Drinking Water Exposure and
Risk Estimates for 1,4-Dioxane Land Releases to Surface Impoundments (U.S. EPA. 2023e).
Overall confidence in risk estimates for drinking waters resulting from disposal of hydraulic fracturing
waste is low to medium. As described in Section 4.3 overall confidence in the oral and dermal cancer
slope factor underlying these risk estimates is medium-high. As described in Section 3.3.2.2.2, the
modeling methodology is robust and the release information relied on as model input data is supported
by moderate evidence. However, no monitoring data are available to confirm detection of 1,4-dioxane in
groundwater near hydraulic fracturing operations. This drinking water exposure scenario relies on the
assumption that the estimated groundwater concentrations may occur in locations where groundwater is
used as a primary drinking water source. While the substantial uncertainty around the extent to which
these exposures occur decreases overall confidence in the exposure scenario, this scenario represents a
sentinel exposure.
Page 153 of 484
-------
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table 5-7. Lifetime Cancer Risks" Estimated for Modeled Groundwater Concentrations Resulting
from Disposal of Hydraulic Fracturing Produced Water
Monte Carlo
Distribution
Modeled Groundwater
Concentration (mg/L)
Adult LADD
(mg/kg/day)
Adult Cancer Risk
Estimate
Max
1.9E-02
8.8E-05
1.1E-05
99th
1.5E-02
7.1E-05
8.6E-06
95th
1.5E-02
7.1E-05
8.6E-06
Mean
7.1E-04
3.3E-06
4.0E-07
50th
1.2E-04
5.6E-07
6.8E-08
5th
1.2E-04
5.6E-07
6.8E-08
Min
4.4E-07
2.1E-09
2.5E-10
a Lifetime cancer risks based on 33 years of oral exposure through drinking water as an adult.
5.2.2.3 Air Pathway
EPA estimated risks from general population exposures to 1,4-dioxane released to air, with a focus on
exposures in fenceline communities. Risks were evaluated for air releases from industrial COUs,
hydraulic fracturing operations, and industrial and institutional laundry facilities based on exposure
estimates in Section 3.2.3.
5.2.2.3.1 Industrial COUs Reported to TRI
EPA estimated risks from general population exposures that could occur in communities neighboring
industrial releases associated with stack and fugitive emissions. Cancer and non-cancer risk estimates for
general population exposures within 10,000 m of industrial releases were calculated for the 10th, 50th,
and 95th percentiles of modeled exposure concentrations estimated in Section 3.2.3.1. Table 5-8
summarizes the cancer risk estimates for 95th percentile exposure concentrations within 1,000 m of the
facilities with the greatest risk in each OES/COU, ranging from 1,05/10 10 to 1.1 x ] o 4, Cancer risk
estimates based on 50th percentile modeled exposure concentrations within 1,000 m of the highest risk
facilities range from 2,5x 10 " to 8.3 x 10 5 (data not shown). Risk estimates were generally highest
within 1,000 m of the facilities and lower at greater distances. As discussed in Section 2.3.3.3, exposure
estimates very near facilities (5 to 10 m) may be impacted by assumptions made for modeling around an
area source (10x10 area source places people at 5 m on top of the release point). This, in combination
with other factors like meteorological data, release heights, and plume characteristics can result in lower
or higher exposures at 5 m than just off the release point at 10 m. Air concentrations from fugitive
emissions tend to peak within 10 m of release sites while contributions from stack releases generally
peak around 100 m, meaning that risks nearest to release sites are often driven by fugitive releases.
Acute and chronic non-cancer risk estimates (not shown) do not indicate risk relative to benchmark
MOES for any of the estimated exposure concentrations at any facilities evaluated. Complete cancer and
non-cancer risk results for air concentrations modeled from stack, fugitive and combined air emissions
are provided in 1,4-Dioxane Supplemental Information File: Air Exposures and Risk Estimates for
Single Year Analysis ( :023e).
Page 154 of 484
-------
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Air exposure and risk estimates are dependent on release amounts, stack heights, contributions from
stack releases and fugitive emissions, topography, and meteorological conditions. These factors vary
substantially across facilities within each OES/COU, making release amount, stack height, and
meteorological conditions more important predictors of risk than a facility's identified OES/COU.
Overall confidence in site-specific risk estimates for inhalation exposure resulting from industrial
releases varies across OES/COUs. As described in Section 4.3, overall confidence in the cancer
inhalation unit risk underlying these risk estimates is high. As described in Section 3.3.3.1, the
AERMOD modeling methodology used for this analysis is robust and accounts for both stack and
fugitive emissions. The exposure scenarios considered are most relevant to long-term residents in
fenceline communities. There is some uncertainty around the extent to which people actually live and
work around the specific facilities where risks are highest, decreasing overall confidence in the exposure
scenario, particularly at distances nearest release sites. Overall confidence varies due to variable levels
of confidence in underlying release information used to estimate exposures. An OES-specific discussion
of the confidence in sources of release information is presented in Appendix E.5E.5.4, but in general
terms:
• Overall confidence in risk estimates is medium to high for OESs/COUs that rely primarily on
release data reported to TRI via Form R.
• Overall confidence in risk estimates is medium for OESs/COUs for which release estimates are
based on data reported to TRI via Form A.
• Overall confidence in risk estimates is low to medium for OESs/COUs for which release
estimates are based on surrogate or modeled information.
Page 155 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4693 Table 5-8. Inhalation Lifetime Cancer Risks" within 10 km of Industrial Air Releases Based on 95th Percentile Modeled Exposure
4694 Concentrations
OES
Corresponding COUs
# Facilities
Distance from Facility with Greatest Risk (m)6
Overall
Confidence
Life Cvelc
Stage
Category
Subcategory
Total
Risk
>1E—06
5
10
30
60
100
100-
1,000
Disposal
Disposal
Disposal
Hazardous waste
incinerator
Off-site waste
transfer
Underground
injection
Hazardous landfill
15
5
2.88E-05
3.42E-05
1.22E-05
4.67E-06
2.13E-06
2.00E-07
Medium
to High
Dry Film
Lubricant
Industrial use,
commercial
use
Other uses
Dry film lubricant
8
0
1.09E-12
4.83E-11
346E-09
2.62E-08
4.26E-08
6.72E-09
Low to
Medium
Ethoxylation
Byproduct
Processing
Byproduct
Byproduct
produced during
the ethoxylation
process to make
ethoxylated
ingredients for
personal care
products
6
3
4.42E-05
9.21E-05
4.96E-05
2.09E-05
1.11E-05
2.58E-06
Medium
to High
Film Cement
Industrial use,
commercial
use
Adhesives
and sealants
Film cement
1
0
846E-07
8.86E-07
2.99E-07
1.54E-07
8.46E-08
1.55E-08
Low to
Medium
Functional
Fluids
(Open-
System)
Industrial use
Functional
fluids (open
and closed
systems)
Polyalkylene
glycol lubricant
Synthetic
metalworking
fluid
Cutting and
tapping fluid
2
0
8.67E-08
1.60E-07
6.98E-08
7.31E-08
1.23E-07
5.02E-08
Medium
to High
Import and
Repackaging
Manufacturing
Import
Import
Repackaging
1
0
1.82E-13
3.78E-12
3.74E-10
2.82E-09
5.89E-09
2.18E-09
Medium
to High
Page 156 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Corresponding COUs
# Facilities
Distance from Facility with Greatest Risk (m)6
Overall
OES
Life Cvelc
Stage
Category
Subcategory
Total
Risk
>1E—06
5
10
30
60
100
100-
1,000
Confidence
Processing
Processing as
a reactant
Polymerization
Catalyst
Industrial
Uses
Processing
Non-
incorporative
Basic organic
chemical
manufacturing
(process solvent)
Medium
to High
Industrial use
Intermediate
use
Plasticizer
intermediate
Catalysts and
reagents for
anhydrous acid
reactions,
brominations,
and sulfonations
12
6
2.84E-05
3.24E-05
1.04E-05
3.84E-06
1.89E-06
4.85E-07
Laboratory
Chemical
Use
Industrial use,
commercial
use
Laboratory
chemicals
Chemical reagent
Reference material
Spectroscopic and
photometric
measurement
Liquid scintillation
counting medium
Stable reaction
medium
Cryoscopic solvent
for molecular
mass
determinations
Preparation of
histological
sections for
microscopic
examination
1
1
1.40E-05
1.46E-05
4.91E-06
2.54E-06
1.40E-06
2.54E-07
Low to
Medium
Manufactur-
ing
Manufacturing
Domestic
manufacture
Domestic
manufacture
1
1
5.91E-05
1.10E-04
5.20E-05
2.18E-05
1.08E-05
9.62E-07
Medium
to High
Page 157 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Corresponding COUs
# Facilities
Distance from Facility with Greatest Risk (m)6
Overall
Confidence
Life Cvelc
Stage
Category
Subcategory
Total
Risk
>1E—06
5
10
30
60
100
100-
1,000
PET
Manufactur-
ing
Processing
Byproduct
Byproduct
produced during
the production of
polyethlene
terephtalate
13
10
5.42E-05
5.48E-05
2.37E-05
9.47E-06
4.35E-06
7.25E-07
Medium
to High
Spray Foam
Application
Industrial use,
commercial
use
Other uses
Spray polyurethane
foam
1
0
5.28E-09
5.68E-09
1.94E-09
1.02E-09
5.79E-10
1.05E-10
Low to
Medium
"Lifetime cancer risks based on 33 years of continuous inhalation exposure averaged over a 78-year lifetime.
h Cancer risks were also calculated at 2,500, 5,000 and 10,000 m from all facilities.
4695
Page 158 of 484
-------
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Land Use Analysis
For locations where lifetime cancer risk is greater than 1 x 10~6, EPA evaluated land use patterns to
determine whether fenceline community exposures may be reasonably anticipated. Detailed results of
this analysis are described in Appendix J.3 and are consistent with the methods described in the draft
2022 Fenceline Report. In short, EPA determined whether residential, industrial/commercial businesses,
or other public spaces are present within those radial distances where cancer risk estimates for 95th
percentile modeled air concentrations are greater than 1 x 10~6 for each facility. In all cases, risks greater
than 1 x 10~6 were within 1,000 m or less of releasing facilities. This analysis was limited to facilities that
could be mapped to a GIS location. Based on this characterization of land use patterns, fenceline
community exposures have the potential to occur at 50 percent of facilities (11 of 22 GIS-mapped
facilities) where cancer risk is greater than 1 x 10~6 based on modeled fenceline air concentrations.
Aggregate Risk
EPA also evaluated potential risks from aggregate exposures from multiple neighboring facilities using a
conservative screening methodology. EPA identified five groups of two to four facilities reporting 1,4-
dioxane releases in proximity to each other, i.e., within 10 km. Aggregating risks estimated for these
groups of facilities were generally dominated by the facility with the greatest risk. This aggregate
analysis did not identify locations with cancer risk greater than 1 x 10~6 that did not already have cancer
risk above that level from an individual facility. Details of the methods and results of this aggregate
analysis are described in Appendix J.4.
5.2.2.3.2 Hydraulic Fracturing
Cancer and non-cancer risk estimates for potential general population exposures within 1,000 m of
hydraulic fracturing operations were calculated for a range of air concentrations modeled across the
distribution of release estimates, as described in Section 3.3.3.2. Table 5-9 presents lifetime cancer risk
estimates for exposure to high-end and central tendency air concentrations modeled for both high end
(95th percentile) and central tendency (50th percentile) modeled releases for a range of topographical
and meteorological scenarios. Lifetime cancer risk estimates for distances within 1,000 m of hydraulic
fracturing operations range from 3.9x 10~7 to 7.1 x 10 5 for high end release estimates and 2,2/ 10 8 to
4.1xio-6 for central tendency release estimates across a range of model scenarios. Acute and chronic
non-cancer risk estimates (not shown) do not indicate risk relative to benchmark MOEs for any exposure
concentrations estimated for hydraulic fracturing operations. Complete results are provided in 1,4-
Dioxane Supplemental Information File: Air Exposure and Risk Estimates for 1,4-Dioxane Emissions
from Hydraulic Fracturing Operations (U.S. EPA. 2023b).
Overall confidence in risk estimates for inhalation exposures resulting for air concentrations modeled
based on releases from hydraulic fracturing operations is medium. As described in Section 4.3, overall
confidence in the cancer inhalation unit risk underlying these risk estimates is high. As described in
Section 3.3.3.2 the modeling methodologies used to estimate air concentrations are robust. The
distribution of air releases used as model input data were estimated using Monte Carlo modeling and
rely on assumptions. No air monitoring data were available to confirm detection of 1,4-dioxane is air
near hydraulic fracturing operations. Because the air concentrations underlying this analysis are based
on releases estimated using probabilistic modeling, they are not tied to specific locations that can be
evaluated for land use patterns. There is therefore substantial uncertainty around the extent to which
people actually live and work around the specific locations where risks are highest, decreasing overall
confidence in the exposure scenario.
Page 159 of 484
-------
4741
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Fugitive
Emissions Release
Scenario
Cancer Risk Estimates for 95th Percentile Modeled Releases
Cancer Risk Estimates for 50th Percentile Modeled Releases
High-End Modeled Air
Concentrations
Central Tendency Modeled Air
Concentrations
High-End Modeled Air
Concentrations
Central Tendency Modeled
Air Concentrations
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
South (Coastal)-
Rural-24
7.1E-05
3.2E-06
8.3E-06
5.6E-05
2.4E-06
6.4E-06
4.1E-06
1.8E-07
4.7E-07
3.2E-06
1.4E-07
3.6E-07
West North
Central-Rural-24
5.4E-05
3.1E-06
7.3E-06
4.1E-05
1.9E-06
4.9E-06
3.1E-06
1.8E-07
4.2E-07
2.4E-06
1.1E-07
2.8E-07
South (Coastal)-
Urban-24
3.4E-05
7.7E-07
2.4E-06
3.0E-05
6.7E-07
2.1E-06
2.0E-06
4.4E-08
1.4E-07
1.7E-06
3.8E-08
1.2E-07
West North
Central-Urban-24
3.2E-05
8.3E-07
2.5E-06
2.6E-05
6.1E-07
1.9E-06
1.8E-06
4.8E-08
1.4E-07
1.5E-06
3.5E-08
1.1E-07
South (Coastal)-
Rural-8
1.3E-05
1.2E-07
5. IE—07
1.1E-05
9.0E-08
4.2E-07
7.3E-07
6.8E-09
2.9E-08
6.4E-07
5.2E-09
2.4E-08
West North
Central-Rural-8
2.7E-05
1.0E-06
2.5E-06
1.4E-05
2.9E-07
8.8E-07
1.5E-06
5.9E-08
1.4E-07
8.2E-07
1.6E-08
5.0E-08
South (Coastal)-
Urban-8
1.2E-05
9.0E-08
4.3E-07
1.1E-05
8.0E-08
3.9E-07
6.7E-07
5.2E-09
2.5E-08
6.1E-07
4.6E-09
2.2E-08
West North
Central-Urban-8
1.9E-05
3.9E-07
1.2E-06
1.2E-05
1.6E-07
6.0E-07
1.1E-06
2.2E-08
7.1E-08
7.0E-07
9. IE—09
3.4E-08
" Lifetime cancer risks based on 33 years of continuous inhalation exposure averaged over a 78-year lifetime.
h Cancer risk estimates shown here are based on modeled releases and air concentrations estimated for 72 days of release.
¦.ab
4742
Page 160 of 484
-------
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
5.2.2.3.3 Industrial and Institutional Laundry Facilities
Cancer and non-cancer risk estimates for potential general population exposures within 1,000 m of
industrial and institutional laundry facilities were calculated for a range of air concentrations modeled
for a range of releases, as described in Section 3.2.3.3. Table 5-10 presents lifetime cancer risk estimates
for exposures estimated from both high-end and central tendency air concentrations modeled based on
the maximum release scenario for each type of laundry under the most conservative exposure scenario
evaluated (rural south coastal topography, assuming 24 hours of releases each day). Lifetime cancer risk
estimates for distances within 1,000 m of laundry facilities range from 1.5/10 " to 3,8/10 8 across a
range of high end and central tendency exposure scenarios. Acute and chronic non-cancer risk estimates
(not shown) do not indicate risk for any estimated exposure concentrations for laundries relative to the
benchmark MOEs. Complete results are provided in 1,4-Dioxane Supplemental Information File: Air
Exposures and Risk Estimates for Industrial Laundry 0 1 S IT \ 2023c).
Overall confidence in risk estimates from inhalation exposures resulting from industrial and institutional
laundries is medium. As described in Section 4.3, overall confidence in the cancer inhalation unit risk
underlying these risk estimates is high. As described in Section 3.3.3.2, the modeling methodologies are
robust. The distribution of air releases used as model input data were estimated using Monte Carlo
modeling and rely on assumptions. No air monitoring data were available to determine whether 1,4-
dioxane is detected near industrial and institutional laundry facilities. Because the air concentrations
underlying this analysis are based on probabilistic modeling, they are not tied to specific locations that
can be evaluated for land use patterns. There is therefore substantial uncertainty around the extent to
which people actually live and work around the specific locations where risks are highest, decreasing
overall confidence in the exposure scenario.
Table 5-10. Lifetime Cancer Risk Estimates for Fugitive Emissions from Industrial and
Institutional Laundry Facilities"
Facility Type
Detergent and
Emissions Type
Cancer Risk Estimates for Maximum Modeled Releases
High-End Modeled Air
Concentrations
Central Tendency Modeled Air
Concentrations
100 m
1,000 m
100 to
1,000 m
100 m
1,000 m
100 to
1,000 m
Industrial
Laundry
Liquid - vapor
3.7E-08
1.7E-09
4.3E-09
3.3E-08
1.4E-09
3.8E-09
Powder - vapor
3.6E-08
1.7E-09
4.2E-09
3.3E-08
1.4E-09
3.8E-09
Powder - PM10
3.8E-08
8.8E-10
3.2E-09
3.4E-08
7.9E-10
2.9E-09
Powder - PM2.5
3.6E-08
1.6E-09
4.1E-09
3.3E-08
1.4E-09
3.7E-09
Institutional
Laundry
Liquid - vapor
2.3E-08
1.1E-09
2.7E-09
2.1E-08
9.0E-10
2.4E-09
Powder - vapor
6.8E-10
3.2E-11
7.9E-11
6.2E-10
2.7E-11
7.0E-11
Powder - PM10
7.1E-10
1.6E-11
5.9E-11
6.4E-10
1.5E-11
5.3E-11
Powder - PM2.5
6.8E-10
3.0E-11
7.7E-11
6.2E-10
2.6E-11
6.9E-11
a Lifetime cancer risks based on 33 years of continuous inhalation exposure averaged over a 78-year lifetime.
Page 161 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4769 5.2.2.4 Potentially Exposed or Susceptible Subpopulations
4770 EPA considered PESS throughout the exposure assessment presented in this draft supplement and
4771 throughout the hazard identification and dose-response analysis described in the 2020 RE. Table 5-11
4772 summarizes how PESS were incorporated into the draft supplement through consideration of increased
4773 exposures and/or increased biological susceptibility. The table also summarizes the remaining sources of
4774 uncertainty related to consideration of PESS.
4775
4776 Table 5-11. Summary of PESS Considerations Incorporated throughout the Analysis and
4777 Remaining Sources of Uncertainty
PESS
Categories
Potential Exposures Identified in Specific
Subpopulations and Incorporated into
Exposure Assessment
Potential Sources of Biological
Susceptibility Identified and
Incorporated into Hazard Assessment
Lifestage
General population drinking water exposure
scenarios include lifestage-specific exposure
factors for adults, children, and formula-fed
infants (Section 5.2.2.1); Inhalation exposures
are based on air concentrations and are therefore
consistent across lifestages (Section 5.2.2.3).
Based on pchem properties and a lack of studies
evaluating potential for accumulation in milk,
EPA did not quantitatively evaluate the milk
pathway and this is a source of uncertainty.
EPA qualitatively described the potential
for biological susceptibility due to lifestage
differences and developmental toxicity but
did not identify quantitative evidence of
lifestage-specific susceptibilities to 1,4-
dioxane; A 10/ UF was applied for human
variability. The magnitude of potential
lifestage differences in metabolism and
toxicity are not well quantified and are a
remaining source of uncertainty.
Pre-existing
Disease
EPA did not identify health conditions that may
influence exposure. The potential for pre-
existing disease to influence exposure (due to
altered metabolism, behaviors, or treatments
related to the condition) is a source of
uncertainty.
EPA qualitatively described the potential
for pre-existing health conditions, such as
liver disease, to increase susceptibility or
alter toxicokinetics, but did not identify
direct quantitative evidence. A 10* UF
was applied for human variability. The
potential impact of pre-existing diseases on
susceptibility to 1,4-dioxane is a remaining
source of uncertainty.
Lifestyle
Activities
EPA did not identify specific lifestyle activities
that expected to increase 1,4-dioxane exposure.
This is a remaining source of uncertainty.
EPA did not identify lifestyle factors that
influence biological susceptibility to 1,4-
dioxane. This is a remaining source of
uncertainty.
Occupational
Exposures
EPA evaluated a range of occupational exposure
scenarios in manufacturing, hydraulic fracturing
and use of commercial products that increase
exposure to 1,4-dioxane present as a byproduct.
EPA evaluated risks for high-end exposure
estimates for each of these scenarios (Section
5.2.1).
EPA did not identify occupational factors
that increase biological susceptibility to
1,4-dioxane. This is a remaining source of
uncertainty.
Geographic
Factors
EPA evaluated risks to fenceline communities
from 1,4-dioxane in ambient air (Section 5.2.2.3)
and in drinking water downstream of release
sites (Section 5.2.2.1). EPA mapped tribal lands
in relation to air, surface water and ground water
releases of 1,4-dioxane to identify potential for
EPA did not identify geographic factors
that increase biological susceptibility to
1,4-dioxane. This is a remaining source of
uncertainty.
Page 162 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
PESS
Categories
Potential Exposures Identified in Specific
Subpopulations and Incorporated into
Exposure Assessment
Potential Sources of Biological
Susceptibility Identified and
Incorporated into Hazard Assessment
increased exposures for tribes due to geographic
proximity (Section 2.3).
Socio-
demographic
Factors
EPA did not identify specific sociodemographic
factors that influence exposure to 1,4-dioxane.
This is a remaining source of uncertainty.
EPA did not identify sociodemographic
factors that influence biological
susceptibility to 1,4-dioxane. This is a
remaining source of uncertainty.
Nutrition
EPA did not identify nutritional factors
influencing exposure to 1,4-dioxane. This is a
remaining source of uncertainty.
EPA did not identify nutritional factors
that influence biological susceptibility to
1,4-dioxane. This is a remaining source of
uncertainty.
Genetics
EPA did not identify genetic factors influencing
exposure to 1,4-dioxane. This is a remaining
source of uncertainty.
Indirect evidence that genetic variants may
increase susceptibility of the target organ
was addressed through a 10* UF for
human variability. The magnitude of the
impact of genetic variants is unknown and
is a source of uncertainty.
Unique
Activities
Some tribes may have increased exposure to
drinking water due to tribal activities such as
sweat lodges. EPA has identified upper bound
drinking water estimates of 2 to 4 L/day
associated with tribal lifeways for some tribes
(Harper, : iarper and Ranco, 2009; Harper
et aL 2007; Harper et aL 2002). Risk
calculations in this draft supplement assume an
acute adult drinking water intake of 3.2 L/day
and a chronic drinking water intake of 0.88
L/day. Other potential sources of increased
exposure to 1,4-dioxane due to specific tribal
lifeways or other unique activity patterns are a
source of uncertainty.
EPA did not identify unique activities that
influence susceptibility to 1,4-dioxane.
This is a remaining source of uncertainty.
Aggregate
Exposures
EPA evaluated risk from aggregate sources of
exposure contributing to 1,4-dioxane in water
(Section 5.2.2.1) or from multiple sources in
proximity releasing to air (Section 5.2.2.3,
Appendix J.4). Risks from aggregate exposures
across routes or pathways were evaluated
qualitatively and are a remaining source of
uncertainty.
EPA does not identify ways that aggregate
exposures would influence susceptibility to
1,4-dioxane. This is a remaining source of
uncertainty.
Other
Chemical and
Non-chemical
Stressors
EPA did not identify chemical and nonchemical
factors influencing exposure to 1,4-dioxane. This
is a remaining source of uncertainty.
EPA did not identify chemical or
nonchemical factors that influence
susceptibility to 1,4-dioxane. There is
insufficient data to quantitatively address
potential increased susceptibility due to
chemical or nonchemical stressors and this
is a remaining source of uncertainty.
Page 163 of 484
-------
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
5.2.2.5 Aggregate and Sentinel Exposures
In this draft supplement, EPA considers the combined 1,4-dioxane exposure an individual may experience
due to releases from multiple facilities in proximity releasing to air or multiple releases contributing to
drinking water concentrations in a particular location. For general population drinking water exposure
scenarios, EPA evaluated combined exposure and risks from multiple sources of 1,4-dioxane in surface
water, including direct and indirect industrial releases, DTD releases, and upstream background
contamination (Section 5.2.2.1). For general population air exposure scenarios, EPA evaluated
combined exposure and risk across multiple facilities in proximity releasing to air (Section 5.2.2.3 and
Appendix J.4).
EPA considered aggregating cancer risks across inhalation, oral, and/or dermal routes of exposure.
There is uncertainty around the extent to which cancer risks across routes are additive for 1,4-dioxane.
Liver tumors are the primary site of cancer risk from oral exposures. Inhalation exposure in rats is
associated with multiple tumor types, including liver. The IUR used to calculate inhalation cancer risk
reflects combined risks from multiple tumor types. While EPA concluded that nasal cavity lesions are
likely to be primarily the result of systematic delivery (as discussed on p. 192 of the 2020 RE), there is
uncertainty around the degree to which those effects could be partially due to portal of entry effects
following inhalation exposure. It is therefore unclear the extent to which it is appropriate to
quantitatively aggregate cancer risks based on the IUR with liver tumor risks associated with oral or
dermal exposures. EPA considers the potential aggregate cancer risk across routes to be a source of
uncertainty for 1,4-dioxane cancer risk estimates.
EPA also considered aggregating cancer risks across dermal and oral exposures. The dermal cancer
slope factor is derived from the oral cancer slope factor by route-to-route extrapolation. Because the
systemic effect is assumed to be the same for both routes, EPA determined that it could be biologically
appropriate to aggregate risk from dermal and oral exposures. General population scenarios included
inhalation and oral not dermal exposures and occupational and consumer exposure scenarios included
inhalation and dermal not oral exposures. However, this draft supplement does not include COUs or
pathways in which both oral and dermal exposure routes are considered.
EPA also considered potential for aggregate exposures across groups. For example, there may be some
individuals who are exposed at work as well as through general population air and drinking water
pathways or through consumer product use. This as a source of uncertainty. These types of aggregate
risks were not quantified and risks for individual exposure scenarios should be interpreted with an
appreciation for potential aggregate exposures and risks.
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 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 § 702.33)." In this draft supplement, EPA considered sentinel
exposures by considering risks to populations who may have upper bound exposures, including workers
and ONUs who perform activities with higher exposure potential and fenceline communities. EPA
characterized high-end exposures in evaluating exposure using both monitoring data and modeling
approaches. Where statistical data are available, EPA typically uses the 95th percentile value of the
available dataset to characterize high-end exposure for a given COU.
Page 164 of 484
-------
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
5.2.2.6 Summary of Overall Confidence and Remaining Uncertainties in Human
Health Risk Characterization
The overall level of confidence in each set of risk estimates depends on the level of confidence in the
underlying hazard values summarized in Section 4.3 and the level of confidence in exposure estimates
described in more detail in Section 3.3.
For all risk estimates, EPA has medium to high confidence in the underlying hazard PODs used as the
basis for this risk characterization. Sources of confidence in each of the hazard values were described in
the 2020 RE and are summarized in Section 4.3. Cancer risk is the primary risk driver for each of the
scenarios evaluated in this draft supplement and is therefore the basis of overall confidence levels
described herein. There is remaining uncertainty for all risk estimates around the potential impact of 1,4-
dioxane on potentially exposed or susceptible subpopulations (as discussed in Section 5.2.2.4). EPA
applied an intraspecies uncertainty factor of 10 to all non-cancer PODs to account for variation across
gender, age, health status, or genetic makeup, and other factors that may increase susceptibility, but the
actual magnitude of the impact of these factors on susceptibility is unknown. Similarly, EPA evaluated
risks from exposure scenarios that are intended to reflect the most highly exposed populations (including
fenceline communities and highly exposed workers), but the potential for other highly exposed
populations that were not identified in this analysis is a source of uncertainty. Potential for aggregate
risks across routes or pathways that are not quantitatively evaluated in this assessment is another source
of uncertainty.
5.2.2.6.1 Risks from Occupational Exposures
Overall confidence in risk estimates for occupational inhalation exposures ranges from low to high,
depending on the confidence in exposure assessment for each OES/COU. As described in Section 4.3,
overall confidence in the cancer inhalation unit risk underlying these risk estimates is high. As described
in Section 3.3.1.1, the measured and monitored inhalation exposure data are supported by moderate to
robust evidence. Additionally, the exposure modeling methodologies and underlying model input data is
supported by moderate to robust evidence. However, there is uncertainty in the representativeness of the
assessed exposure scenarios towards all potential exposures for the given OES/COU, limitations in the
amount and age of monitoring data, and limitations in the modeling approaches towards 1,4-dioxane-
specific use within the OES/COU. Therefore, while the underlying data and methods used to estimate
occupational inhalation risk is supported by moderate to robust evidence, the overall confidence of these
estimates ranges from low to high depending on the OES/COU. OES/COU-specific discussions of the
available inhalation exposure data and overall confidence are presented in Appendix F.6 and
summarized in Table 3-2.
Overall confidence in risk estimates for occupational dermal exposures is medium for all OES/COUs
because the same modeling approach was used for all OES/COUs. As described in Section 4.3 overall
confidence in the oral and dermal cancer slope factor underlying these risk estimates is medium-high. As
described in Section 3.3.1.2, the dermal exposure modeling methodology is supported by moderate
evidence, with model input parameters from literature sources, a European model, standard defaults
from the ChemSTEER User Guide ( 015a). and 1,4-dioxane product concentration data from
process information. These sources range from slight to robust, depending on factors such as age and
applicability to OES/COU. The modeling is limited by the use of standard input parameters that are not
specific to 1,4-dioxane and a lack of variability in dermal exposure for different worker activities.
Therefore, EPA's overall confidence in the occupational dermal risk estimates is medium
Page 165 of 484
-------
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
5.2.2.6.2 Risks from General Population Exposures through Drinking Water
Overall confidence in the overall distribution of risk estimates for drinking water exposures resulting
from facility releases is medium-high. Overall confidence in site-specific risk estimates for individual
facility releases varies both within and across OES, depending on the confidence in the source-specific
release data. As described in Section 4.3 overall confidence in the oral and dermal cancer slope factor
underlying these risk estimates is medium-high. As described in Section 3.3.2.1, the exposure modeling
methodology used for this analysis is supported by moderate evidence. It is designed to estimate water
concentrations expected at specific locations. Exposure estimates for this scenario are based on some
conservative assumptions about flow rates and release frequency and amount. A summary of sources of
flow and release data for facility release modeling is presented in Table 2-6. Available monitoring data
confirm that 1,4-dioxane is present in some surface water and drinking water, though most of the
available data were not collected near release sites are therefore not directly comparable. The overall
level of confidence depends on the source of OES/COU-specific release data described in Appendix E.3:
• Overall confidence in drinking water exposure estimates is medium to high for OESs/COUs that
rely primarily on release data reported to DMR or to TRI via Form R. Most COUs/OESs are
included in this group.
• Overall confidence in drinking water exposure estimates is medium for OESs/COUs for which
release estimates are based on reporting to TRI via Form A. The Import and Repackaging OES
releases used in this analysis are entirely based on Form A reporting of releases, and just under
half of the Industrial Uses OES releases were reported via Form A.
• Overall confidence in drinking water exposure estimates is low to medium for OESs/COUs for
which release estimates are based on surrogate or modeled information.
Overall confidence in risk estimates for drinking water exposures resulting from DTD releases under
varying population and stream flow conditions is medium. As described in Section 4.3 overall
confidence in the oral and dermal cancer slope factor underlying these risk estimates is medium-high. As
described in Section 3.3.2.1, the exposure modeling methodology used for this analysis is supported by
robust evidence and is designed to provide a nationally representative distribution of estimated water
concentrations under varying conditions. Exposure estimates rely on estimated distributions of DTD
releases of consumer and commercial products for each COU.
Overall confidence in risk estimates for drinking water exposures resulting from hydraulic fracturing
releases is medium. As described in Section 4.3 overall confidence in the oral and dermal cancer slope
factor underlying these risk estimates is medium-high. As described in Section 3.3.2.1, the exposure
modeling methodology used for this analysis is supported by robust evidence and is designed to provide
a nationally representative distribution of estimated water concentrations under varying conditions.
Releases used as inputs in the model were estimated using Monte Carlo modeling that captures
variability across sites. However, the modeled exposure estimates are not directly tied to specific
releases at known locations, decreasing the strength of the evidence related to the representativeness of
the exposure estimates for actual exposures.
Overall confidence in risk estimates for drinking water exposures resulting from for aggregate surface
water concentrations predicted by probabilistic modeling varies across OES/COU. As described in
Section 4.3 overall confidence in the oral and dermal cancer slope factor underlying these risk estimates
is medium-high. As described in Section 3.3.2.1, the exposure modeling methodology used for this
analysis is supported by robust evidence and is designed to provide a nationally representative
distribution of estimated water concentrations under varying conditions. Available monitoring data
confirm that 1,4-dioxane is present in some surface water and drinking water, though most of the
available data were not collected near release sites and are therefore not directly comparable. In release
Page 166 of 484
-------
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
locations where monitoring data are available, case studies demonstrate strong concordance between
modeled estimates and measured surface water concentrations. There is some uncertainty around the
proximity of releases to drinking water intake locations and the extent to which 1,4-dioxane is further
diluted prior to reaching intake locations. The characterization of downstream dilution presented in
5.2.2.1.2 for individual facility releases illustrates the extent to which downstream dilution may impact
overall risk estimates. The overall level of confidence in resulting exposure estimates depends on the
source of OES/COU-specific release data described in Appendix E.3:
• Overall confidence in drinking water exposure estimates is medium to high for OESs/COUs that
rely primarily on release data reported to DMR or to TRI via Form R. Most COUs/OESs are
included in this group.
• Overall confidence in drinking water exposure estimates is medium for OESs/COUs for which
release estimates are based on reporting to TRI via Form A. The Import and Repackaging OES
releases used in this analysis are entirely based on Form A reporting of releases, and just under
half of the Industrial Uses OES releases were reported via Form A.
• Overall confidence in drinking water exposure estimates is low to medium for OESs/COUs for
which release estimates are based on surrogate or modeled information.
5.2.2.6.3 Risks from General Population Exposures through Groundwater and
Land Disposal Pathways
Overall confidence in risk estimates for drinking water exposures resulting from disposal to landfills is
low to medium. As described in Section 4.3 overall confidence in the oral and dermal cancer slope
factor underlying these risk estimates is medium-high. As described in Section 3.3.2.2.1 the modeling
methodology is robust. However, the release information relied on as model input data is supported by
slight to moderate evidence. In addition, this drinking water exposure scenario relies on the assumption
that the estimated groundwater concentrations may occur in locations where groundwater is used as a
primary drinking water source. While the substantial uncertainty around the extent to which these
exposures occur decreases overall confidence in the exposure scenario, this scenario represents a
sentinel exposure.
Overall confidence in risk estimates for drinking waters resulting from disposal of hydraulic fracturing
waste is low to medium. As described in Section 4.3 overall confidence in the oral and dermal cancer
slope factor underlying these risk estimates is medium-high. As described in Section 3.3.2.2.2, the
modeling methodology is robust, and the release information relied on as model input data is supported
by moderate evidence. However, no monitoring data are available to confirm detection of 1,4-dioxane in
groundwater near hydraulic fracturing operations. This drinking water exposure scenario relies on the
assumption that the estimated groundwater concentrations may occur in locations where groundwater is
used as a primary drinking water source. While the substantial uncertainty around the extent to which
these exposures occur decreases overall confidence in the exposure scenario, this scenario represents a
sentinel exposure.
5.2.2.6.4 Risks from General Population Exposures through Air
Overall confidence in risk estimates for inhalation exposure resulting from industrial releases varies
across COUs. As described in Section 4.3, overall confidence in the cancer inhalation unit risk
underlying these risk estimates is high. As described in Section 3.3.3.1, the AERMOD modeling
methodology used for this analysis is robust. The exposure scenarios considered are most relevant to
long-term residents in fenceline communities. There is some uncertainty around the extent to which
people actually live and work around the specific facilities where risks are highest, decreasing overall
confidence in the exposure scenario. Overall confidence varies due to variable levels of confidence in
Page 167 of 484
-------
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
underlying release information used to estimate exposures. An OES-specific discussion of the
confidence in sources of release information is presented in Appendix E.5.4, but in general terms:
• Overall confidence in risk estimates is medium to high for OESs/COUs that rely primarily on
release data reported to TRI via Form R.
• Overall confidence in risk estimates is medium for OESs/COUs for which release estimates are
based on data reported to TRI via Form A.
• Overall confidence in risk estimates is low to medium for OESs/COUs for which release
estimates are based on surrogate or modeled information.
Overall confidence in risk estimates for inhalation exposures resulting for air concentrations modeled
based on releases from hydraulic fracturing operations is medium. As described in Section 4.3, overall
confidence in the cancer inhalation unit risk underlying these risk estimates is high. As described in
Section 3.3.3.2 the modeling methodologies used to estimate air concentrations are robust. The
distribution of air releases used as model input data were estimated using Monte Carlo modeling and
rely on assumptions. No air monitoring data were available to confirm detection of 1,4-dioxane is air
near hydraulic fracturing operations. Because the air concentrations underlying this analysis are based
on probabilistic modeling, they are not tied to specific locations that can be evaluated for land use
patterns. There is therefore substantial uncertainty around the extent to which people actually live and
work around the specific locations where risks are highest, decreasing overall confidence in the
exposure scenario.
Overall confidence in risk estimates from inhalation exposures resulting from industrial and institutional
laundries is medium. As described in Section 4.3, overall confidence in the cancer inhalation unit risk
underlying these risk estimates is high. As described in Section 3.3.3.2, the modeling methodologies are
robust. The distribution of air releases used as model input data were estimated using Monte Carlo
modeling and rely on assumptions. No air monitoring data were available to determine whether 1,4-
dioxane is detected near industrial and institutional laundry facilities. Because the air concentrations
underlying this analysis are based on probabilistic modeling, they are not tied to specific locations that
can be evaluated for land use patterns. There is therefore substantial uncertainty around the extent to
which people actually live and work around the specific locations where risks are highest, decreasing
overall confidence in the exposure scenario.
Page 168 of 484
-------
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
REFERENCES
Adam son. DT; Mahendra. S; Walker. KL. Jr.; Ranch. SR; Sengupta. S; Newell. CI. (2014). A multisite
survey to identify the scale of the 1,4-dioxane problem at contaminated groundwater sites.
Environ Sci Technol Lett 1: 254-258. http://dx.doi.ore/10.1021/ez500092u
Adam sob iH Pifia. EA; Cartwrigln \« Kauch. SR; Anderson. RH; Moln « t minor. J A. (2017). 1.4-
Dioxane drinking water occurrence data from the third unregulated contaminant monitoring rule.
Sci Total Environ 596-597: 236-245. http://dx.doi.o /i.scitotenv.2017.04.085
Analvtica. (2015). Choosing an appropriate distribution [Website],
http://wiki.analytica.com/Choosine an appropriate distribution
AT SDR. (2012). Toxicological profile for 1,4 dioxane [ATSDR Tox Profile], Atlanta, GA: U.S.
Department of Health and Human Services, Public Health Service.
http://www.atsdr.cdc. eov/toxprofiles/tp.asp?id=955&tid=l 99
Baldwin. PE; Mayn (1998). A survey of wind speed in indoor workplaces. Ann Occup Hyg 42:
303-313. http://dx.doi.c 5/80003-4878(98)00031-3
Bel anger. PL. (1980). Health hazard evaluation report no. HHE-80-21-721, Colgate-Palmolive Co.,
Berkeley, California. (HHE-80-21-721). Cincinnati, OH: National Institue for Occupational
Safety and Health.
Black. RE; Hurley. FJ; Havery. DC. (2001). Occurrence of 1,4-dioxane in cosmetic raw materials and
finished cosmetic products. J AOAC Int 84: 666-670.
Brouehtoii Sepulveili \ I < ter. K; Kail I' x ickelsen. MG; Gillan. M; Mohr. TKG. (2019). 1.4-
Dioxane: Emerging technologies for an emerging contaminant. Remediation Journal 29: 49-63.
http://dx.doi.ore/10.1002/rem.. I I
Brunswick County. (2022). Brunswick County, North Carolina: Annual reports/facilities information
[Website], https://www.bmnswickcountync.eov/utilities/annual-report.s-facilities-information/
CA Water Board. (2022). EDT library and water quality analyses data and download page [Website],
https://www.waterboards.ca.eov/drinkine water/certlic/drinkinewater/EDTlibrary.html
Commonwealth of Massachusetts. (2018). Energy & Environmental Affairs Data Portal: Search for
drinking water [Website], https://eeaonline.eea.state.ma.us/portal#!/search/drinkine-water
Cordone. L; Carlson (*, Plaehn. W; Shanera^\ « Wilmoth. D. (2016). Case study and retrospective:
Aerobic fixed film biological treatment process for 1,4-dioxane at the Lowry landfill superfund
site. Remediation Journal 27: 159-172. http://dx.doi.ore em.21502
Davarani. SSH; Masoomi. L; Banitaba. IV id. HRLZ; Sadeehi. O; Samiei. A. (2012). A New
Aluminium Hydroxide Coating on Fused Silica Fiber for the Determination of 1,4-Dioxane in
Surfactants and Detergents Using HS-SPME-GC. Chromatographia 75: 371-377.
http://dx.doi.orE
Forkner. MW; Robson. J.H; Snellines. WM; Martin. AE; Mum rsons. TE. (2004). Kirk-
Othmer Encyclopedia of Chemical Technology
Glycols. Hoboken, New Jersey: John Wiley & Sons, Inc.
http://dx.doi.ore 10 100. 0471238961.05200V-0 hi\| j.a0i.pub2
Franklin Associates. (2020). Cradle-to-resin life cycle analysis of polyethylene terephthalate resin (final
revised report). Charlotte, NC: The National Association for PET Container Resources
(NAPCOR).
Franz nnett. S; DeLeo. PC; Collatz. M; Kelly. K; Nekoomaram. J; Wieroniey. S. (2015).
Comments of the Adhesive and Sealant Council, the American Coatings Association, the
American Chemistry Council, the American Cleaning Institute, the Consumer Specialty Products
Association, and Waste Management on the 1,4-dioxane problem formulation and initial
assessment. (EPA-HQ-OPPT-2015-0078). Franz, C; Bennett, S; DeLeo, PC; Collatz, M; Kelly,
K; Nekoomaram, J; Wieroniey, S. https://arasp.americanchemistry.com/Media-Center/Industry-
Comm ents-June-2015 .pdf
Page 169 of 484
-------
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Gillespie. BW; Chen. Q; Reichca IS, 1'irmzblau. A; Hedeeman 1,1 epkowski > \f08
Godri Pollitt. KJ; Kim. J.H; Peccia. J; Elimelech. M; Zhang. t barkoftali u, Hodges. B; Zucker. I;
Huang. H; Deziel. NC: Murphy. K; Ishii. M; Johnson. CH; Boissevairi \ U'Keet'1 I \iustas.
PT; Orlickv. D; Thompson. DC; Vasiliou. ¥. (2019). 1,4-Dioxane as an emerging water
contaminant: State of the science and evaluation of research needs [Review], Sci Total Environ
690: 853-866. http://dx.doi.org/10.1016/i.scitotenv.20 r" 0 > i i;
Harlev. KG: Calderon. L; Nolan. JES; Maddalena. R; Russell. M; Roman. K; Mayo-Burgos. S: Cabrera.
J: Morga. N: Bradman. A. (2021). Changes in Latina women's exposure to cleaning chemicals
associated with switching from conventional to "green" household cleaning products: The
LUCIR intervention study. Environ Health Perspect 129: 97001.
http://dx.doi.orE 39/EHP8831
Harpi (2017). Shoshone-Bannock exposure scenario for use in risk assessment. Harper, B.
https://superfund.oregonstate.edu/sites/superfund.oregonstate.edu/files/shoshone bannock scena
rio 2017.pdf
Harptn Tn kanc<> 1 * (2009). Wabanaki traditional cultural lifeways exposure scenario. U.S.
Environmental Protection Agency and the Tribal Nations in Maine.
https://www.epa.gov/sites/default/files/2015-08/documents/ditca.pdf
Harptn r% 1 i _ H, I. Harris. S: Abevta. C: Kirschner. F. (2002). The Spokane Tribe's multipathway
subsistence exposure scenario and screening level RME. Risk Anal 22: 513-526.
http://dx.doi.orE >272-4332.00047
Harpi 'ding. AK: Waterhous. T: Harris. SG. (2007). Traditional tribal subsistence exposure
scenario and risk assessment guidance manual. (Grant Number EPA-STAR-J1-R831046).
Corvallis, OR: Oregon State University Department of Public Health and Confederated Tribes of
the Umatilla Indian Reservation.
https://health.oregonstate.edu/sites/health.oregonstate.edu/files/research/pdf/tribal-
grant/exposure scenario a : guidance manual v2.pdf
HHS. (2016). 1,4-Dioxane: CAS No. 123-91-1. In Report on Carcinogens (14th ed.). Research Triangle
Park, NC: National Toxicology Program, Department of Health and Human Services.
https://ntp.niehs.nih.gov/ntp/roc/content/profiles/dioxane.pdf
Hills. B: Ktincewicz. S: Blade. LM: Sack. I ) (1989). Health hazard evaluation report no. HETA-87-
367-1987, BMY Corporation, A Division of Harsco Corporation, York, Pennsylvania. (HETA-
87-367-1987). Cincinnati, OH: National Institue for Occupational Safety and Health.
https://www.cdc.gov/niosh/hhe/reports/pdfs/1987-0. 87.pdf
Jackson. LE: Lemkt (2019). Evidence for natural attenuation of 1,4-dioxane in a glacial aquifer
system. Hydrogeology Journal v. 27: 3009-3024. http://dx.doi.< 10 100 1 100 10 0 r"-02028-6
Kano. H: Umeda. Y: Kasai. T: Sasaki. T: Matsumoto. M: Yamazaki. K: Nagano. to. H:
Fukushima. S. (2009). Carcinogenicity studies of 1,4-dioxane administered in drinking-water to
rats and mice for 2 years. Food Chem Toxicol 47: 2776-2784.
http://dx.doi.orE 10 101 | ict.2009.0N 01 J
Kasai. T: Kano. H: Umeda. Y: Sasaki wa. N: Nishizawa. T: Nagano. K: Arito. H: Nagashima. H:
Fukushima. S. (2009). Two-year inhalation study of carcinogenicity and chronic toxicity of 1,4-
dioxane in male rats. Inhal Toxicol 21: 889-897. http://dx.doi.org/10 10SQ/0895837080262( \ h1
Kasti iller. MA. (2006). Kinetics of finite dose absorption through skin 2: Volatile
compounds. J Pharm Sci 95: 268-280. http://dx.doij >s.20497
Page 170 of 484
-------
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Lee. CS; Venkatesan. A.K; Walker. H )ler. CI. (2020). Impact of groundwater quality and
associated byproduct formation during UV/hydrogen peroxide treatment of 1,4-dioxane. Water
Res 173: 115534. http://dx.doi.ore/10.1016/j.watres.2020.115534
Leste inn an. EM; Sitterlev. oral ¦ inden. KG. (2015).
Characterization of hydraulic fracturing flowback water in Colorado: Implications for water
treatment. Sci Total Environ 512-513: 637-644.
http://dx.doi.org 10 101 i.scitotenv.20I 01 0 l'<
Li. M; Mathieu. J; Yarn \ t torenza. S: Dem \ lie. Z; Zhou j,_ \lvare (2013). Widespread
distribution of soluble di-iron monooxygenase (SDIMO) genes in Arctic groundwater impacted
by 1,4-dioxane. Environ Sci Technol 47: 9950-9958. http://dx.doi.org/10.102 1 /es402228x
Li. M; Van Orden. ET; Devries. DJ; Xione. Z: Hinchee. R; Ah o I (2015). Bench-scale
biodegradation tests to assess natural attenuation potential of 1,4-dioxane at three sites in
California. Biodegradation 26: 39-50. http://dx.doi.c
Lin. WT; Chen. WL; Cheng. V ang. H.C; Tsai. SW. (2017). Determining the residual
characteristics of alkylphenols, arsenic, and lead as well as assessing the exposures of 1,4-
dioxane from household food detergents. J AOAC Int 100: 1086-1093.
http://dx.doi.ore/10.5740/iaoacint. 16-0380
Makino. R; Kawasaki. H; Kishimoto. A; Gamo. M; Nakanishi. J. (2006). Estimating health risk from
exposure to 1,4-dioxane in Japan. Environ Sci 13: 43-58.
Marquart. H; Franken. R; Goei isman. W; Schinkel. J. (2017). Validation of the dermal
exposure model in ECETOC TRA. Ann Work Expo Health 61: 854-871.
http://dx.doi.ore/10.1093/annweh/wxx059
Mattie. PR; Bucher. TW; Carter. AL; Stoffreeen. DE; Reboulet. IE. (2012). Acute inhalation toxicity
study of 1, 4-dioxane in rats (Rattus norvegicus). GRA and I: 29.
McDaniel. MP; DesLauriei (2015). Kirk-Othmer encyclopedia of chemical technology
Ethylene polymers, HDPE. Hoboken, NJ: Wiley-Interscience.
http://dx.doi.ore 10 1002/0 i i. 3961.08090708 I I0"l"l" jQl wb3
Millard. SP. (2013). EnvStats: An R package for environmental statistics. New York, NY: Springer.
http://dx.doi.ore/10J007/9 • I I ti - h I
Minitab. (2022). A comparison of the Pearson and Spearman correlation methods [Website],
https://support.minitab.com/en-us/minitab/21/help-and~how-to/statistics/basic-
statistics/supporting-topics/correlation-and-covariance/a-comparison-of-the-pearson-and-
spearman-correlation-methods/
Mohi seppi. WH. (2010). Case Studies of 1,4-Dioxane Releases, Treatment, and Drinking
Water Contamination. In T Mohr; B Diguiseppi; JK Anderson; JW Hatton (Eds.), (2nd ed., pp.
369-420). Boca Raton, FL: CRC Press. huf; 'ivdoi.oa s < 06629-c8
N einer. E. (2020). Kirk-Othmer Encyclopedia of Chemical Technology
Detergency and detergents [Encyclopedia], New York, NY: John Wiley & Sons.
http://dx.doi.ore 10 1002/0 i i. 3961.04052005122 M II UOt |':iib3
NIOSH. (1987). NIOSH Guide to Industrial Respiratory Protection (pp. 87-1 16). (NIOSH/00177545).
Anonymous.
NTP. (1986). NTP Toxicology and carcinogenesis studies of dichloromethane (methylene chloride)
(CAS No. 75-09-2) in F344/N rats and B6C3F1 mice (inhalation studies). 306: 1-208.
https://heronet.epa.gov/heronet/index.cfm?action=search.view&reference id=l 03686800berle,
W. (2015). Monte Carlo simulations: Number of iterations and accuracy. (ARL-TN-0684).
Aberdeen Proving Ground, MD: US Army Research Laboratory.
https://apps.dtic.mil/sti/pdl0 \Vi \ _| |.i!f
OECD. (2009). Emission scenario documents on coating industry (paints, lacquers and varnishes).
(JT03267833). Paris, France.
Page 171 of 484
-------
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
http://www.oecd. org/officialdocuments/publicdisplavdocumentpdf/?cote=env%20/im/mono(200
9)24& docl an guage=en
OECD. (201 la). Emission scenario document on coating application via spray-painting in the
automotive refinishing industry. In OECD Series on Emission Scenario Documents No 11.
(ENV/JM/MONO(2004)22/REV1). Paris, France: Organization for Economic Co-operation and
Development.
http://www.oecd. org/officialdocuments/publicdisplaydocumentpdf/?cote=env/im/mono(2004)22/
rev 1& docl an guage=en
OECD. (201 lb). Emission scenario document on the chemicals used in water-based washing operations
at industrial and institutional laundries. (JT03304937). Paris, France, https://www.oecd-
ilibrary.org/docserver/9y892642209y3~
en.pdf?e: 50&id=id&accname=guest&checksum=B8Cyi4E510603EB565B89A
8AE6A24C14
OECD. (2015). Emission scenario document (ESD) on industrial use of industrial cleaners.
(JT03373556). Paris, France.
http://www.oecd. org/officialdocuments/publicdisplaydocumentpdf/?cote=ENV/JM/MQNQ(2015
)3 & docl an guage=en
OECD. (2017). Emission scenario document (ESD) on the use of textile dyes. In Emission Scenario
Document, Series No 36. (ENV/JM/MONO(2015)50). Paris, France: OECD, Inter-Organization
Programme for the Sound Management of Chemicals (IOMC).
http://www.oecd.org/chemicalsafety/risk-assessment/emissionscenariodocuments.htm
OECD. (2020). Emission scenario document on chemical additives used in automotive lubricants. Paris,
France.
Oracle. (2017). Oracle® Crystal Ball User's Guide; 2017 (Release 11.1.2.4)
Selecting probability distributions. Austin, TX.
https ://docs. oracl e. com/cd/ ^pasOS.html
Otto. M; Nagaraia. S. (2007). Treatment technologies for 1,4-Dioxane: Fundamentals and field
applications. Remediation Journal 17: 81-88. http://dx.doi.org/10.1002/rem.201 '< ?
Palisade. (2015a). Palisade Knowledge Base: 2.6. How many iterations do I need? [Website],
https://kb.palisade.com/index.php?pg=kb.page&id=125
Palisade. (2015b). Palisade Knowledge Base: 2.19. Random number generators [Website],
https://kb.palisade.com/index.php?pg=kb.page&id=25
Palisade. (2018). Palisade Knowledge Base: 2.18. Latin Hypercube versus Monte Carlo sampling
[Website], https://kb.palisade.com/index.php?pg=kb.page&id=28
Palisade. (2022a). Define correlation matrix [Website], https://help.palisade.eom/v8/en/@RISK/l-
Define/5-Correlation/Define-Correlation-Matrix.htm
Putz. VR; Johnson. BL; Setzer. JV. (1979). A comparative study of the effects of carbon monoxide and
methylene chloride on human performance. J Environ Pathol Toxicol 2: 97-112.
https://heronet.epa.gov/heronet/index.cfm?action=search.view&reference id=l 0626648 Saraii.
M; Shirvani. N. (2017). Determination of residual 1,4-dioxane in surfactants and cleaning agents
using headspace single-drop microextraction followed by gas chromatography-flame ionization
detection. Int. J Cosmet Sci 39: 36-41. http://dx.doi.org/10.1 1 1 1/ics. 12345
Stefl. ge. KF. (2014). Kirk-Othmer Encyclopedia of Chemical Technology
Antifreezes and deicing fluids [Encyclopedia], New York, NY: John Wiley & Sons.
http://dx.doi.org/iOJ002/0 i l _ • x mm l >200919200506 jQl |':iib2
US JitSl (2016). May 2016 Occupational Employment and Wage Estimates: National Industry-
Specific Estimates [Website], http://www.bls.gov/oes/tables.htm
Page 172 of 484
-------
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
IJ.S JitSl (2022). Labor Force Statistics from the Current Population Survey: Household data, Annual
averages, 11. Employed persons by detailed occupation, sex, race, and Hispanic or Latino
ethnicity [Numbers in thousands] [Website], https://www.bls.eov/cps/cpsaatl 1 .htm
sus Bureau. (2015). Statistics of U.S. Businesses (SUSB).
https://www.censiis.eov/data/tables/2015/econ/susb/2015-siisb-aimiial.html
(1988). National survey of solid waste (municipal) landfill facilities [EPA Report],
(EPA/530-SW88-034). Washington, DC: Office of Solid Waste and Emergency Response, U.S.
Environmental Protection Agency.
(1991). Chemical engineering branch manual for the preparation of engineering assessments.
Volume I. Ceb Engineering Manual. Washington, DC: Office of Pollution Prevention and
Toxics, US Environmental Protection Agency.
(1994a). Guidelines for Statistical Analysis of Occupational Exposure Data: Final. United
States Environmental Protection Agency :: U.S. EPA.
(1994b). Methods for derivation of inhalation reference concentrations and application of
inhalation dosimetry [EPA Report], (EPA600890066F). Research Triangle Park, NC.
https://crpub.epa.eov/ncea/risk/fecofdisplay.crm?deid=yi993«S 329&CFTOKEN=2
300b3 1 7
U.S. EPA. (2006a). 2006 community water system survey - Volume I: Overview [EPA Report], (EPA
815-R-09-001). Washington, DC: U.S. Environmental Protection Agency, Office of Water.
https://nepis.epa. eov/Exe/ZyPDF.cei?Dockey=P 1009JJI.txt
U.S. EPA. (2006b). Treatment technologies for 1,4-dioxane: Fundamentals and field applications. (EPA-
542-R-06-009). Washington, DC: Office of Solid Waste and Emergency Response.
http://costperformance.Org/remediation/pdf/I'<1 \ IV ^ ument of 1.4~Dioxane.pdf
U.S. EPA. (201 1). Exposure factors handbook: 201 1 edition [EPA Report], (EPA/600/R-090/052F).
Washington, DC: U.S. Environmental Protection Agency, Office of Research and Development,
National Center for Environmental Assessment.
https://nepis.epa. gov/Exe/ZyPURL.cgi?Dockey=P 100F2QS.txt
U.S. EPA. (2012). Advances in inhalation gas dosimetry for derivation of a reference concentration
(RfC) and use in risk assessment (pp. 1-140). (EPA/600/R-12/044). Washington, DC.
https://cfpub.epa.gov/ncea/risk/recordisplav.cfm?deid=244650 > , CFTOKEN=
U.S. EPA. (2015a). ChemSTEER user guide - Chemical screening tool for exposures and environmental
releases. Washington, D C. https://www.epa.gov/sites/production/files/2Q15-
05/documents/user guide.pdf
U.S. EPA. (2015b). TSCA. work plan chemical problem formulation and initial assessment. 1,4-
Dioxane. (740-R1-5003). Washington, DC: Environmental Protection Agency, Office of
Chemical Safety and Pollution Prevention.
http://nepis.epa. eov/Exe/ZyPURL. cei?Dockev=P 1OOMDC1. TXT
U.S. EPA. (2016a). Hydraulic fracturing for oil and gas: Impacts from the hydraulic fracturing water
cycle on drinking water resources in the United States [EPA Report], (EPA/600/R-16/236F).
Washington, DC. https://cfpub.epa.eov/ncea/hfstudy/recordisplay.cfm?deid=332990
U.S. EPA. (2016b). Public database 2016 chemical data reporting (May 2017 release). Washington, DC:
US Environmental Protection Agency, Office of Pollution Prevention and Toxics.
https://www.epa.gov/chemical-data-reporting
U.S. EPA. (2017a). Information on the various spray polyurethane foam products. U.S. Environmental
Protection Agency, Design for the Environment.
https://www.epa.eov/sites/prodiiction/files/2015-08/documents/spfprodi: ;s.pdf
Page 173 of 484
-------
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
U.S. EPA. (2017b). Preliminary information on manufacturing, processing, distribution, use, and
disposal: 1,4-dioxane. https://www.regulations.gov/document?D=] " \ ^ iiQ~QPPT~201 0 23-
0003
U.S. EPA. (2017c). Technical fact sheet 1,4-dioxane [EPA Report], (EPA-505/F-17/01 1). Office of
Land and Emergency Management, U.S. EPA.
https://nepis.epa. gov/Exe/ZyPURL.cgi?Dockev=P 100THWG.txt
U.S. EPA. (2018a). Application of spray polyurethane foam insulation - Generic scenario for estimating
occupational exposures and environmental releases -Draft. Washington, DC.
https://www.epa.gov/tsca-screening-tools/using-predictive-methods-assess-exposure-and-fate-
under-tsca
U.S. EPA. (2018b). Application of spray polyurethane foam insulation - Generic scenario for estimating
occupational exposures and environmental releases -Methodology review draft, revised.
https://www.epa.gov/tsca-screening-tools/chemsteer-chemical-screening-tool-exposures-and-
environmental-releases#genericscenarios
U.S. EPA. (2018c). Application of systematic review in TSCA risk evaluations. (740-P1-8001).
Washington, DC: U.S. Environmental Protection Agency, Office of Chemical Safety and
Pollution Prevention. https://www.epa.gov/sites/production/files/2Q18-
06/documents/final application of sr in tsc if
U.S. EPA. (2018d). User's Guide for the AMS/EPA Regulatory Model (AERMOD). (EPA Document
Number: EPA-454/B-18-001). U.S. EPA.
(2019a). Exposure factors handbook chapter 3 (update): Ingestion of water and other select
liquids [EPA Report], (EPA/600/R-18/259F). Washington, DC.
https://cfpub. epa.gov/ncea/efp/recordisplav. cfm?deid=343661
U.S. EPA. (2019b). User's Guide: Integrated Indoor-Outdoor Air Calculator (IIOAC). Washington, DC:
U.S. EPA.
https://heronet.epa.gov/heronet/index.cfm?action=search.view&refere
(2020b). Final scope of the risk evaluation for 4,4'-(l-Methylethylidene)bis[2, 6-dibromophenol]
(TBBPA); CASRN 79-94-7. (EPA 740-R-20-008). Washington, DC: Office of Chemical Safety
and Pollution Prevention. https://www.regulations.gov/document/EPA-HQ-OPPT-2018-Q462-
0047
U.S. EPA. (2020c). Risk evaluation for 1,4-dioxane (CASRN: 123-91-1) [EPA Report], (EPA-740-R1-
8007). Washington, DC: Office of Chemical Safety and Pollution Prevention.
https://www.regiilations.gov/dociiment/EPA-H.Q-0] 0238-0092
U.S. EPA. (202 la). Draft systematic review protocol supporting TSCA risk evaluations for chemical
substances, Version 1.0: A generic TSCA systematic review protocol with chemical-specific
methodologies. (EPA Document #EPA-D-20-031). Washington, DC: Office of Chemical Safety
and Pollution Prevention. https://www.regiilations.gov/dociiment/EPA-HQ-OPPT-2'
0005
U.S. EPA. (202 lb). Use of additives in plastic compounding - Generic scenario for estimating
occupational exposures and environmental releases (Revised draft) [EPA Report], Washington,
DC: Office of Pollution Prevention and Toxics, Risk Assessment Division.
U.S. EPA. (2022a). Chemicals used in furnishing cleaning products - Generic scenario for estimating
occupational exposures and environmental releases (revised draft). Washington, DC: Office of
Pollution Prevention and Toxics.
U.S. EPA. (2022b). Commercial use of automotive detailing products - Generic scenario for estimating
occupational exposures and environmental releases (Methodology review draft) [EPA Report],
Washington, DC: U.S. Environmental Protection Agency, Office of Pollution Prevention and
Toxics, Risk Assessment Division.
https://heronet.epa.gov/heronet/index.cfm?action=search.view&reference id=l 048047211.8.
Page 174 of 484
-------
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
EPA. (2022d). Emission scenario document on chemicals used in hydraulic fracturing (draft). In
OECD Environmental Health and Safety Publications, Series on Emission Scenario Documents.
Paris, France: Organization for Economic Co-operation and Development.
https://heronet.epa.gov/heronet/index.cfm?action=search.view&reference id=10480474U.S.
EPA. (2022h). Use of laboratory chemicals - Generic scenario for estimating occupational
exposures and environmental releases (Revised draft generic scenario) [EPA Report],
Washington, DC: U.S. Environmental Protection Agency, Office of Pollution Prevention and
Toxics, Existing Chemicals Risk Assessment Division.
U.S. EPA. (2023a). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: 1,4-Dioxane Supplemental Information Files EWISRD-XL-R probabilistic
model code. Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical
Safety and Pollution Prevention.
U.S. EPA. (2023b). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Air Exposure and Risk Estimates for 1,4-Dioxane Emissions from Hydraulic
Fracturing Operations. Washington, DC: Office of Pollution Prevention and Toxics, Office of
Chemical Safety and Pollution Prevention.
U.S. EPA. (2023c). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Air Exposures and Risk Estimates for Industrial Laundry. Washington, DC:
Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023d). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Air Exposures and Risk Estimates for Multi-Year Analysis. Washington, DC:
Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023e). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Air Exposures and Risk Estimates for Single Year Analysis. Washington, DC:
Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023f). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Land Releases
to Landfills. Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical
Safety and Pollution Prevention.
U.S. EPA. (2023g). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Land Releases
to Surface Impoundments. Washington, DC: Office of Pollution Prevention and Toxics, Office
of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023h). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Release to
Surface Water from Individual Facilities. Washington, DC: Office of Pollution Prevention and
Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023i). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Surface Water
Concentrations Predicted with Probabilistic Modeling. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023j). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Environmental Releases to Air. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023k). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Environmental Releases to Land for all OES Except Disposal. Washington,
DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution
Prevention.
Page 175 of 484
-------
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
U.S. EPA. (20231). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Environmental Releases to Land for the Disposal OES. Washington, DC:
Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023m). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Environmental Releases to Water for OES with TRI and DMR. Washington,
DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution
Prevention.
U.S. EPA. (2023n). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Environmental Releases to Water for OES without TRI or DMR data.
Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and
Pollution Prevention.
U.S. EPA. (2023o). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: EWISRDXL BrunswickCountyNC Case Study. Washington, DC: Office of
Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023p). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: EWISRDXL ColumbiaTN Case Study. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023q). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: EWISRDXL Liverpool OH Case Study. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023r). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: Occupational Exposure and Risk Estimates. Washington, DC: Office of
Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023s). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental
Information File: WQP Processed Surface Water Data. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
U.S. EPA. (2023t). Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Systematic Review
Supplemental File: Data Quality Evaluation and Data Extraction Information for Environmental
Release and Occupational Exposure. Washington, DC: Office of Pollution Prevention and
Toxics, Office of Chemical Safety and Pollution Prevention.
https://heronet.epa. gov/heronet/index.cfm?action=search.view&referenceid=3419938W al a-
Jerzykiewicz, A; Szymanowski, J. (1998). Headspace gas chromatography analysis of toxic
contaminants in ethoxylated alcohols and alkylamines. Chromatographia 48: 299-304.
http://dx.doi.ore 10 100 n 0._ I S6
Youtiv dt i tin. WH; Gehriin P < Ucrvath. BS; Daniel. RL. (1976). 1,4-Dioxane and beta-
hydroxyethoxyacetic acid excretion in urine of humans exposed to dioxane vapors. Toxicol Appl
Pharmacol 38: 643-646. http://dx.doi.org/i0 J0 i6/00 II 008X(76)90195-2
Yotffly 11 * Braun. WH; Ramp\ 1 \\ t lienowetli > 'Lm. Gl". (1977). Pharmacokinetics of 1,4-
dioxane in humans. J Toxicol Environ Health 3: 507-520.
http://dx.doi.orE 30/15287397709529583
Page 176 of 484
-------
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
APPENDICES
Appendix A ABBREVIATIONS AND ACRONYMS
30Q5
Lowest 30-day average flow that occurs in a 5-year period
7Q10
Lowest 7-day average flow that occurs in a 10-year period
AC
Acute concentrations
ACA
American Coatings Association
ACGM
American Conference of Governmental Industrial Hygienists
ADC
Average daily concentration
ADD
Average daily dose
ADR
Acute Dose Rate
AEC
Acute Exposure Concentration
APF
Assigned protection factor
ASTDR
Agency for Toxic Substances and Disease Registry
BHET
Bishydroxyethyl terephthalate
BLS
Bureau of Labor Statistics
BMD
Benchmark dose
BMDL
Benchmark dose level
CASRN
Chemical Abstracts Service Registry Number
CDR
Chemical Data Reporting
CEB
Chemical Engineering Branch
CEHD
Chemical Exposure Health Data
CERCLA
Comprehensive Environmental Response, Compensation and Liability Act
CFR
Code of Federal Regulations
COU
Condition of use
CSF
Cancer slope factor
CT
Central tendency
CWA
Clean Water Act
DAF
Dilution attenuation factor
DHHR
Department of Health and Human Services
DIY
Do-it-yourself
DMR
Discharge monitoring report
DRAS
Delisting Risk Assessment Software
DTD
Down-the-drain
DWI
Drinking water intake
DWT
Drinking water treatment
ECHA
European Chemicals Agency
ECHO
Environmental Compliance History Online database
EPA
Environmental Protection Agency
EPACMTP
Environmental Protection Agency Composite Model for Leachate Migration with
Transformation Products
EPCRA
Emergency Planning and Community Right-to-Know Act
EROM
Enhanced Runoff Method (database)
ESD
Emission Scenario Document
FRS
Facility Registry Service
FT
Full-text (screening)
GS
Generic Scenario
HAWC
Health Assessment Workplace Collaborative (tool)
Page 177 of 484
-------
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
5442
5443
5444
5445
5446
5447
5448
5449
5450
5451
5452
5453
5454
5455
5456
5457
5458
5459
5460
5461
5462
5463
5464
5465
5466
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
HE
High-end
HEC
Human equivalent concentration
HED
Human equivalent dose
HERO
Health and Environmental Research Online (EPA Database)
HHE
Health hazard evaluation
HSDB
Hazardous Substances Data Bank
ICIS
Integrated Compliance Information System
IFC
Industrial Function Category
IIOAC
Integrated Indoor/Outdoor Air Calculator (EPA)
IRIS
Integrated Risk Information System
IUR
Inhalation unit risk
Koc
Soil organic carbon: water partitioning coefficient
Kow
Octanol: water partition coefficient
LADC
Lifetime Average Daily Concentration
LADD
Lifetime Average Daily Dose
LOAEC
Lowest-observed-adverse-effect-concentration
LOD
Limit of detection
Log Koc
Logarithmic organic carbon: water partition coefficient
Log Kow
Logarithmic octanol: water partition coefficient
LOQ
Limit of quantitation
MLD
Million liters per day
MOE
Margin of exposure
MRD
Methodology Review Draft (EPA)
MW
Molecular weight
NAICS
North American Industry Classification System
ND
Non-detect
NEI
National Emissions Inventory
NIOSH
National Institute for Occupational Safety and Health
NOAEL
No-observed-adverse-effect-level
NPDES
National Pollutant Discharge Elimination System
OAQPS
Office of Air Quality Planning and Standards
OCF
One-component foam
OCSPP
Office of Chemical Safety and Pollution Prevention
OD
Operating days
OECD
Organisation for Economic Co-operation and Development
OEHHA
Office of Environmental Health Hazard Assessment
OES
Occupational exposure scenario
ONU
Occupational non-user
OPPT
Office of Pollution Prevention and Toxics
OSHA
Occupational Safety and Health Administration
PBZ
Personal breathing zone
PECO
Population, exposure, comparator, and outcome
PEL
Permissible exposure limit
PESS
Potentially exposed or susceptible subpopulations
PET
Polyethylene terephthalate
PF
Protection factor
PNOR
Particulates not otherwise regulated
POD
Point of departure
POTW
Publicly owned treatment works (wastewater)
Page 178 of 484
-------
5467
5468
5469
5470
5471
5472
5473
5474
5475
5476
5477
5478
5479
5480
5481
5482
5483
5484
5485
5486
5487
5488
5489
5490
5491
5492
5493
5494
5495
5496
5497
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
PPE
Personal protective equipment
PV
Production volume
PWS
Public water system
QA/QC
Quality assurance/quality control
QE
NHDPlus V2.1 flow values representing "the best EROM estimate of actual mean flow"
RE
(2020 RE) Risk Evaluation
RCRA
Resource Conservation and Recovery Act
REACH
Registration, Evaluation, Authorisation and Restriction of Chemicals (European Union)
RESO
Receptors, exposure, setting or scenario, and outcomes
SACC
Science Advisory Committee on Chemicals
SDS
Safety data sheet
SDWA
Safe Drinking Water Act
SHEDS-HT
Stochastic Human Exposure and Dose Simulation-High Throughput
SIC
Standard Industrial Classification
SOC
Standard Occupational Classification
SpERC
Specific Environmental Release Categories
SPF
Spray polyurethane foam
STORET
STOrage and RETrieval and Water Quality exchange
SUSB
Statistics of United States Businesses
SWIFT
Sciome Workbench for Interactive Computer-Facilitated Text-mining
TIAB
Title/abstract (screening)
TRI
Toxics Release Inventory
TSCA
Toxic Substances Control Act
TWA
Time-weighted average
UCMR
Unregulated Contaminant Monitoring Rule
U.S.
United States
USGS
U.S. Geological Survey
VOC
Volatile organic compound
VP
Vapor pressure
WQP
Water Quality Portal
WWT
Wastewater treatment
Page 179 of 484
-------
5498
5499
5500
5501
5502
5503
5504
5505
5506
5507
5508
5509
5510
5511
5512
5513
5514
5515
5516
5517
5518
5519
5520
5521
5522
5523
5524
5525
5526
5527
5528
5529
5530
5531
5532
5533
5534
5535
5536
5537
5538
5539
5540
5541
5542
5543
5544
5545
5546
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Appendix B LIST OF SUPPLEMENTAL DOCUMENTS
Associated Systematic Review Data Quality Evaluation and Data Extraction Documents - Provide
additional detail and information on individual study evaluations and data extractions including criteria
and data quality results.
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Systematic Review Supplemental
File: Data Quality Evaluation and Data Extraction Information for Environmental Release and
Occupational Exposure - Provides a compilation of tables for the data extraction and data
quality evaluation information for 1,4-dioxane. Each table shows the data point, set, or
information element that was extracted and evaluated from a data source that has information
relevant for the evaluation of environmental release and occupational exposure (
2023t). This supplemental file may also be referred to as the 1,4-Dioxane Supplement to the Risk
Evaluation Data Quality Evaluation and Data Extraction Information for Environmental Release
and Occupational Exposure.
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Systematic Review Supplemental
File: Data Quality Evaluation Information for General Population, Consumer, and
Environmental Exposure - Provides a compilation of tables for the data quality evaluation
information for 1,4-dioxane. Each table shows the data point, set, or information element that
was evaluated from a data source that has information relevant for the evaluation of general
population, consumer, and environmental exposure ( 23r). This supplemental file
may also be referred to as the 1,4-Dioxane Supplement to the Risk Evaluation Data Quality
Evaluation Information for General Population, Consumer, and Environmental Exposure.
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Systematic Review Supplemental
File: Data Extraction Information for General Population, Consumer, and Environmental
Exposure - Provides a compilation of tables for the data extraction for 1,4-dioxane. Each table
shows the data point, set, or information element that was extracted from a data source that has
information relevant for the evaluation of general population, consumer, and environmental
exposure ( 023h). This supplemental file may also be referred to as the 1,4-Dioxane
Supplement to the Risk Evaluation Data Extraction Information for General Population,
Consumer, and Environmental Exposure.
Associated Supplemental Information Files - Provide additional details and information on exposure,
hazard, and risk assessments.
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Environmental Releases to Air - Provides a summary of stack and fugitive air emissions for each
occupational exposure scenario (OES) in the 1,4-Dioxane Supplemental Risk Evaluation (
23j).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Environmental Releases to Landfor all OES Except Disposal - Provides a summary of land
releases for each 1,4-dioxane OES except for the Disposal OES ( :023k).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Environmental Releases to Landfor the Disposal OES - This spreadsheet contains a summary of
land releases for the 1,4-dioxane occupational Disposal OES (U.S. EPA. 20231).
Page 180 of 484
-------
5547
5548
5549
5550
5551
5552
5553
5554
5555
5556
5557
5558
5559
5560
5561
5562
5563
5564
5565
5566
5567
5568
5569
5570
5571
5572
5573
5574
5575
5576
5577
5578
5579
5580
5581
5582
5583
5584
5585
5586
5587
5588
5589
5590
5591
5592
5593
5594
5595
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Environmental Releases to Water for OES without TRI or DMR data - Provides a summary of
direct and indirect water releases for each 1,4-dioxane OES for which Toxics Release Inventory
(TRI) and Discharge Monitoring Report (DMR) data were not available ( 2023m).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Environmental Releases to Water for OES with TRI and DMR - Provides a summary of direct
and indirect water releases for each 1,4-dioxane OES for which TRI or DMR data were available
( >023m).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Occupational Exposure and Risk Estimates - Provides a summary of occupational exposures and
risks estimated for all conditions of use (COUs; (including those evaluated in this supplemental
evaluation as well as those previously evaluated in the 2020 RE) (U.S. EPA. 2023r).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Release to Surface Water from
Individual Facilities - Provides water concentrations estimated from individual facility releases
reported to TRI and calculates corresponding drinking water exposures and risks (
2023h).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Surface Water Concentrations
Predicted with Probabilistic Modeling - Provides water concentrations estimated by
probabilistic modeling for DTD releases, disposal of hydraulic fracturing waste to surface water,
and for aggregate concentrations estimated downstream of industrial release sites; calculates
corresponding drinking water exposures and risks. This file also calculates drinking water
exposures and risks estimated from drinking water monitoring data (U.S. EPA. 20231).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
EWISRDXL BrunswickCountyNC Case Study - Provides the Excel workbook file for the
Brunswick County, NC surface water case study ( )23o).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
EWISRDXL ColumbiaTN Case Study - Provides the Excel workbook file for the Columbia, TN
surface water case study (U.S. EPA. 2023p).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
EWISRDXL LiverpoolOH Case Study - Provides the Excel workbook file for the Liverpool, OH
surface water case study ( A. 2023 q).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane Supplemental Information File:
EWISRD-XL-Rprobabilistic model code - Provides the R script used to perform the probabilistic
surface water modeling by OES/COU ( 323a).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File: WQP
Processed Surface Water Data - Provides the processed monitoring data in surface water
retrieved from the Water Quality Portal ( 2023s).
Page 181 of 484
-------
5596
5597
5598
5599
5600
5601
5602
5603
5604
5605
5606
5607
5608
5609
5610
5611
5612
5613
5614
5615
5616
5617
5618
5619
5620
5621
5622
5623
5624
5625
5626
5627
5628
5629
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Land Releases to Landfills -
Provides calculations of groundwater concentration derived from the waste adjusted dilution
attenuation factor extracted from Delisting Risk Assessment Software (DRAS) for Landfills and
the corresponding risk calculations ( ,023f).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File:
Drinking Water Exposure and Risk Estimates for 1,4-Dioxane Land Releases to Surface
Impoundments - Provides calculations of groundwater concentration derived from the waste
adjusted dilution attenuation factor extracted from DRAS for release of hydraulic fluid produced
water to surface impoundments and the corresponding risk calculations ( ).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File: Air
Exposures and Risk Estimates for Single Year Analysis - Provides air concentrations estimated
by American Meteorological Society/Environmental Protection Agency Regulatory Model
(AERMOD) for air releases reported to TRI in 2019 and calculates corresponding exposure
concentrations and risk estimates ( 23e).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File: Air
Exposure and Risk Estimates for 1,4-Dioxane Emissions from Hydraulic Fracturing Operations
- Provides air concentrations estimated by Integrated Indoor/Outdoor Air Calculator (IIOAC)
based on Monte Carlo modeling of air releases from hydraulic fracturing operations and
calculates corresponding exposure concentrations and risks ( )23b).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File: Air
Exposures and Risk Estimates for Industrial Laundry - Provides air concentrations estimated by
IIO AC based on Monte Carlo modeling of air releases from hydraulic fracturing operations and
calculates corresponding exposure concentrations and risks ( )23c).
Draft Supplement to the Risk Evaluation for 1,4-Dioxane - Supplemental Information File: Air
Exposures and Risk Estimates for Multi-Year Analysis - Provides air concentrations estimated by
IIO AC for 6 years (2015 to 2020) of air releases reported to TRI and calculates the
corresponding exposure concentrations and risk estimates ( 3d).
Page 182 of 484
-------
5630
5631
5632
5633
5634
5635
5636
5637
5638
5639
5640
5641
5642
5643
5644
5645
5646
5647
5648
5649
5650
5651
5652
5653
5654
5655
5656
Appendix C
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
SYSTEMATIC REVIEW PROTOCOL FOR THE
DRAFT SUPPLEMENT TO THE RISK EVALUATION
FOR 1,4-DIOXANE
The U.S. EPA's Office of Pollution Prevention and Toxics (EPA/OPPT) applies systematic review
principles in the development of risk evaluations under the amended TSCA. TSCA section 26(h)
requires EPA to use scientific information, technical procedures, measures, methods, protocols,
methodologies, and models consistent with the best available science 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 CFR 702.33).
To meet the TSCA section 26(h) science standards, EPA used the TSCA systematic review process
described in the Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for Chemical
Substances (U.S. EPA. 2021a) (hereinafter referred to as "2021 Draft Systematic Review Protocol").
Table Apx C-l (Section 3 of the 2021 Draft Systematic Review Protocol) depicts the steps in which
information is identified and whether it undergoes the formal systematic review process (U.S. EPA.
2021a). Information attained via the systematic review process is integrated with information attained
from sources of information that do not undergo systematic review (e.g., EPA-generated model outputs)
to support a weight of the scientific evidence analysis.
Scope
Risk Evaluation
Systematic Review
1
Literature Searching
and Screening
Test
Order/
Rule Data
Data Evaluation
Data Extraction
(in scope chemicals)
Data Gap filling from sources
outside of the Systematic Review
process
(i.e., systematic approaches using
model outputs, analogue, qualitative
information on a COU)
Weight of the Scientific Evidence
Analysis'/or each discipline and across
disciplines
Evidence Integration of
Systematically Reviewed
data
Evidence Integration of data
obtained outside of
Systematic Review
Conclusions from
the Weight of the
Scientific Evidence
Analysis
Risk
Characterization
across disciplines
Legend
TSCA Process/Product
Systematic Review Step
Non-Systematic Review
Step (may encompass
systematic approaches)
Step utilizing Systematic
Review and non-
Systematic Review
Results
'Weight of the
Scientific Evidence
(WoSE) considers
the results of the
Systematic Review
method and
additional evidence
integrated from non-
systematic review
methods. Analysis of
the WoSE may
include evidence
integration across
disciplines.
FigureApx C-l. Overview of the TSCA Risk Evaluation Process with Identified Systematic
Review Steps
Page 183 of 484
-------
5657
5658
5659
5660
5661
5662
5663
5664
5665
5666
5667
5668
5669
5670
5671
5672
5673
5674
5675
5676
5677
5678
5679
5680
5681
5682
5683
5684
5685
5686
5687
5688
5689
5690
5691
5692
5693
5694
5695
5696
5697
5698
5699
5700
5701
5702
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
The process complements the risk evaluation process in that 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 (40 CFR 702.33).
C.l Clarifications and Updates to the 2021 Draft Systematic Review
Protocol
In 2021, EPA released the Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for
Chemical Substances (U.S. EPA. 2021a). a framework of systematic review approaches under TSCA, to
address comments received on a precursor systematic review approaches framework, the Application of
Systematic Review in TSCA Risk Evaluations ( 2018c). In April 2022, the SACC provided
comments on the 2021 Draft Systematic Review Protocol while additional comments on OPPT's
systematic review approaches were garnered during the public comment period. In lieu of an update to
the 2021 Draft Systematic Review Protocol, this systematic review protocol for the Draft Supplement to
the Risk Evaluation for 1,4-Dioxane describes some clarifications and different approaches that were
implemented than those described in the 2021 Draft Systematic Review Protocol in response to (1)
SACC comments, (2) public comments, or (3) to reflect chemical-specific risk evaluation needs.
C.l.l Clarifications and Updates
Throughout the 2021 Draft Systematic Review Protocol, there were some terms used that were not
explicitly defined, resulting in their different uses within the document (U.S. EPA. 2021a). Table Apx
C-l lists the terms that were updated to resolve some of the confusion expressed by the public and
SACC comments regarding the implementation of the respective systematic review-related step. One
main clarification is that all references that undergo systematic review are consideredfor use in the risk
evaluation, even those that do not meet the various discipline and sub-discipline screening criteria {i.e.,
RESO, PESO, PECO) or that are categorized as supplemental information at title and abstract (TIAB) or
full-text screening.
Section 4.2.5 of the 2021 Draft Systematic Review Protocol describes how data sources {e.g., individual
references, databases) may be tagged and linked in epidemiological cohort studies when information is
present in multiple studies ( 21a). References will generally undergo data quality evaluation
and extraction if there are data that pass screening criteria; however, to prevent the same data from being
represented multiple times and conflating the amount of available information on a subject area, EPA
selects the reference(s) that most appropriately describes the extractable results (indicated as the parent
reference in DistillerSR). For example, if two references portray the same information from the same
dataset, only one is counted in the overall dataset {i.e., deduplication). If two references contain
information about the same dataset, but only one provides additional contextual information or summary
statistics {e.g., mean), both data sources are linked but the extractable information from both may be
combined in DistillerSR. This allows the capture of key information while avoiding double counting the
data of interest, which may be the case whether or not one reference contains original or extractable data
that passes screening criteria. The linked reference containing the majority of the data, which are
evaluated and extracted, is identified in DistillerSR as the parent reference; the "complementary child
reference" in DistillerSR does not undergo data evaluation and extraction. Linking the references in
DistillerSR allows the reference with more limited information or only contextual information to be
tracked and utilized to evaluate the extracted data in the other related studies. The child reference may
undergo data quality evaluation and extraction if there are additional unique and original data that pass
screening criteria. One clarification is that this procedure of identifying potential duplicative information
applies to all information that is considered in a risk evaluation under TSCA (not just epidemiological
Page 184 of 484
-------
5703
5704
5705
5706
5707
5708
5709
5710
5711
5712
5713
5714
5715
5716
5717
5718
5719
5720
5721
5722
5723
5724
5725
5726
5727
5728
5729
5730
5731
5732
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
cohort studies). Also, this procedure may apply when there is duplicative information in two references,
even if it is more than just "contextual."
Section 5 of the 2021 Draft Systematic Review Protocol describes how EPA conducts data quality
evaluation of data/information sources considered for a respective chemical risk evaluation, with Section
5.2 specifically explaining the terminology used to describe both metric and overall data/information
source quality determinations ( 21a). To respond to both SACC and public comments
regarding the inappropriate use of quantitative methodologies to calculate both "Metric Rankings" and
"Overall Study Rankings", EPA decided to not implement quantitative methodologies to attain either
metric and overall data/information source quality determinations and therefore updated the
terminology used for both metric ("Metric Ranking") and overall data/information source ("Overall
Study Ranking") quality determinations (TableApx C-l). Specifically, metric and overall
data/information source quality determination terminology have been updated to "Metric Rating" and
"Overall Quality Determination", respectively. The word "level" was also often used synonymously and
inconsistently with the word "ranking" in the 2021 Draft Systematic Review Protocol; that inconsistency
has been rectified, resulting in the word "level" no longer being used to indicate either metric or overall
data/information source quality determinations ( 2021a).
Sections 4.3.2.1.3 and 6 of the 2021 Draft Systematic Review Protocol describes when EPA may reach
out to authors of data/information sources to obtain raw data or missing elements that are important to
support the data evaluation and data integration steps ( :021a). In such cases, the request(s)
for additional data/information, number of contact attempts, and responses from the authors are
documented. EPA's outreach is considered unsuccessful if those contacted do not respond to email or
phone requests within 1 month of initial attempt(s) of contact. One important clarification to this
guidance is that EPA may reach out to authors anytime during the systematic review process for a given
data/information source or reference, and that contacting authors does not explicitly happen during the
data quality evaluation or extraction steps.
Table Apx C-l. Terminology Clarifications between the 2021 Draft Systematic Review Protocol
and the 1,4-Dioxane Systematic Review Protocol
2021 Draft Systematic
Review Protocol Term
1,4-Dioxane Systematic
Review Protocol Term Update
Clarification
"Title and abstract" or
"Title/abstract"
Title and abstract (TIAB)
To increase consistency, the term "title and abstract"
will be used to refer to information specific to "title
and abstract" screening.
Variations of how
"include," "on topic" or
"PECO/PESO/RESO
relevant" implied a
reference was
considered for use in the
risk evaluation, whereas
"exclude," "off topic" or
"not PECO/PESO/RESO
relevant" implied a
reference was not
considered for use in the
risk evaluation.
Meets/does not meet
PECO/PESO/RESO screening
criteria
The term "include" or "exclude" falsely suggests
that a reference was or was not, respectively,
considered in the risk evaluation. There was also
confusion regarding whether "on topic" and
"PECO/PESO/RESO relevant" were synonymous
and suggested those references were explicitly
considered for use in the risk evaluation (and by
default, "off topic" and "not PECO/PESO/RESO
relevant" references were not). References that meet
the screening criteria (e.g., PECO, PESO, RESO)
proceed to the next systematic review step;
however, all references that undergo systematic
review at any time are considered in the risk
evaluation. Information that is categorized as
Page 185 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2021 Draft Systematic
Review Protocol Term
1,4-Dioxane Systematic
Review Protocol Term Update
Clarification
supplemental or does not meet screening criteria are
generally less relevant for quantitative use in the
risk evaluation but may be considered if there is a
data need identified. For instance, mechanistic
studies are generally categorized as supplemental
information at either title and abstract or full-text
screening steps but may undergo the remaining
systematic review steps if there is a relevant data
need for the risk evaluation (e.g., dose response,
mode of action).
Database source not
unique to a chemical
Database
Updated term and definition of "Database": Data
obtained from databases that collate information for
the chemical of interest using methods that are
reasonable and consistent with sound scientific
theory and/or accepted approaches and are from
sources generally using sound methods and/or
approaches (e.g., state or federal governments,
academia). Example databases include STORET
and the Massachusetts Energy and Environmental
Affairs Data Portal.
The term in the 2021 Draft Systematic Review
Protocol (Table_ApxN-l) incorrectly suggested
that databases that contain information on a singular
chemical are not considered (U.S. EPA, 2021a).
Furthermore, the wording "large" was removed to
prevent confusion and the incorrect suggestion that
there is a data size requirement for databases that
contain information that may be considered for
systematic review.
Metric Ranking or Level
Metric Rating
As explained above, EPA is not implementing
quantitative methodologies to indicate metric
quality determinations, therefore the term "ranking"
is inappropriate. The term "level" was inconsistently
used to indicate metric quality determinations
previously, therefore EPA is removing the use of
this term to reduce confusion when referring to
metric quality determinations. The term "Rating" is
more appropriate to indicate the use of professional
judgement to determine a quality level for
individual metrics.
Overall Study Ranking
or Level
Overall Quality Determination
(OQD)
As explained above, EPA is not implementing
quantitative methodologies to indicate overall
data/information source quality determinations,
therefore the term "ranking" is inappropriate. The
term "level" was inconsistently used to indicate
overall data/information source quality
determinations previously, therefore EPA is
removing the use of this term to reduce confusion
Page 186 of 484
-------
5733
5734
5735
5736
5737
5738
5739
5740
5741
5742
5743
5744
5745
5746
5747
5748
5749
5750
5751
5752
5753
5754
5755
5756
5757
5758
5759
5760
5761
5762
5763
5764
5765
5766
5767
5768
5769
5770
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
2021 Draft Systematic
Review Protocol Term
1,4-Dioxane Systematic
Review Protocol Term Update
Clarification
when referring to overall data/information source
quality determinations. The term "Rating" is more
appropriate to indicate the use of professional
judgement to determine a quality level for the
overall data/information source quality
determination.
C.2 Data Search
To expand upon the previous analysis conducted in the 2020 RE, this draft Supplement to the Risk
Evaluation for 1,4-Dioxane addresses additional COUs in which 1,4-dioxane is present as a byproduct of
the manufacturing process and evaluates risks from general population exposures to 1,4-dioxane
released to water, air, and land. This draft supplement focuses on evaluating additional exposure
pathways that were not addressed in the Final Risk Evaluation for 1,4-Dioxane ( 2020c).
Therefore, the data search focused on prioritizing updated literature search results to characterize
environmental releases and occupational exposure, and general population, consumer, and
environmental exposure information to evaluate the exposure pathways in scope for this draft
supplement. Data sources may also contain information that may be used to evaluate exposure pathways
already addressed in the 2020 RE ( 20c) {i.e., consumer exposure). Below are the four
additional exposure pathways being assessed in this Draft Supplement to the Risk Evaluation for 1,4-
Dioxane (Section 1.2).
• Occupational exposure to
o 1,4-dioxane present as a byproduct in commercial products during ethoxylation
processing or polyethylene terephthalate (PET) manufacturing and in hydraulic fracturing
waste (Sections 3.1, 5.2.1)
• General population exposures to
o 1,4-dioxane present in drinking water sourced from surface water as a result of direct and
indirect industrial releases and DTD releases of consumer and commercial products
(Sections 2.3.1, 3.2.2 and 5.2.2.1);
o 1,4-dioxane present in drinking water sourced from groundwater contaminated as a result
of disposals (Sections 2.3.2, 3.2.2.2 and 5.2.2.1.6); and,
o 1,4-dioxane released to air from industrial and commercial sources (Sections 2.3.3, 3.2.3,
and 5.2.2.3).
C.2.1 Multi-Disciplinary Updates to the Data Search
For this Draft Supplement to the Risk Evaluation for 1,4-Dioxane, the updated literature search was
conducted as described in Section 4 of the 2021 Draft Systematic Review Protocol (1 2021a).
where the peer-reviewed and gray literature updated search followed the approach outlined in Sections
4.2 and 4.3 of the 2021 Draft Systematic Review Protocol, respectively ( ). The updated
search for peer-reviewed and gray literature relevant references was completed in October 2021 and
January 2022, respectively, which also considered information found for the Final Risk Evaluation for
1,4-Dioxane ( )20c). Occasionally additional data sources relevant for the risk evaluation
may be identified after the initial search for peer-reviewed and gray literature; these data sources will
then undergo systematic review for the relevant discipline(s). Additionally, each discipline utilizes
different strategies (e.g., search strings) to attain their discipline-specific pools of data sources that
undergo systematic review
Page 187 of 484
-------
5771
5772
5773
5774
5775
5776
5777
5778
5779
5780
5781
5782
5783
5784
5785
5786
5787
5788
5789
5790
5791
5792
5793
5794
5795
5796
5797
5798
5799
5800
5801
5802
5803
5804
5805
5806
5807
5808
5809
5810
5811
5812
5813
5814
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
As mentioned in Section 4.2.2 of the 2021 Draft Systematic Review Protocol, a supplemental literature
search is conducted to fill data gaps, but in this draft supplement, the supplemental search was
conducted to update the literature search conducted to identify any potentially relevant environmental
release and occupation exposure and general population, consumer, and environmental exposure
information (U.S. EPA. 2021a). Rather than utilizing positive and negative seed references as described
in Section 4.2.4.2 of the 2021 Draft Systematic Review Protocol, search strings were used in SWIFT16-
Review to better identify relevant references to evaluate exposure pathways addressed in this draft
supplement ( 21a). The language describing the new exposure pathways and COUs that are
in scope for this draft supplement was used to derive the search strings listed below in Sections C.2.3.1
and C.2.3.2. When the search strings are identified in the title, abstract, keyword, or Medical Subject
Heading (MeSH) fields of a given reference in SWIFT-Review, those references proceeded with TIAB
screening.
The evaluation of physical and chemical properties, fate properties and environmental and human health
hazard information did not differ from the respective information provided in the Final Risk Evaluation
for 1,4-Dioxane ( ?20c) to address the additional exposure pathways in this draft
supplement, therefore no additional references were identified for these respective topics or underwent
systematic review for these disciplines. One minor clarification to what was described in the 2021 Draft
Systematic Review Protocol is that the PECO statement used to screen general population, consumer,
and environmental exposure information considered for this draft supplement, currently resides in
Appendix Section H.5 (which was intended to encompass PECO statements regarding environmental
and human health hazard information), rather than in Appendix Section H.4 ( ). Please
see Appendix C.3.2 below for additional updates specific to the implementation of the PECO statement.
C.2.2 Additional Data Sources Identified
As mentioned above in Appendix C.2, additional data sources containing potentially relevant
information for a respective risk evaluation may be identified. For this draft supplement, additional gray
literature data sources were identified for the characterization of environmental release and occupational
exposure and general population, consumer, and environmental exposure, as explained below in
Sections C.2.2.1 and C.2.2.2, respectively.
C.2.2.1 Additional Data Sources Identified for Environmental Release and
Occupational Exposure
As explained in Appendix E of the 2021 Draft Systematic Review Protocol ( ), generic
scenarios and emission scenarios documents are listed as part of the initial gray literature sources. Some
generic scenarios and a draft emission scenario document became available after the gray literature
search was completed in January 2022 and were considered for the environmental release and
occupational exposure assessment. This includes the Draft OECD ESD on Hydraulic Fracturing (U.S.
22d), Draft GS on Furnishing Cleaning Products ( 2a), EPA Methodology Review
Draft (MRD) on Commercial Use of Automotive Detailing Products ( 'A. 2022b). and Draft GS
on Use of Laboratory Chemicals ( !2h). The updated sources were added to EPA's Health
and Environmental Research Online (HERO) database in 2022 as well as the systematic review process.
In addition to the gray literature sources listed above, an online database called FracFocus 3.0 (GWPC
and IOGCC. 2022) was included in the pool of references EPA considered for environmental release and
occupational exposure through backward searching. These are described in Section 4.4 of the 2021 Draft
16 SWIFT is an acronym for "Sciome Workbench for Interactive Computer-Facilitated Text-mining." SWIFT-Active
Screeneruses machine learning approaches to save screeners' time and effort.
Page 188 of 484
-------
5815
5816
5817
5818
5819
5820
5821
5822
5823
5824
5825
5826
5827
5828
5829
5830
5831
5832
5833
5834
5835
5836
5837
5838
5839
5840
5841
5842
5843
5844
5845
5846
5847
5848
5849
5850
5851
5852
5853
5854
5855
5856
5857
5858
5859
5860
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Systematic Review Protocol ( 2021a). Backward searching from the Draft OECD ESD on
Hydraulic Fracturing (U.S. EPA. 2022d) led to EPA's identification of the FracFocus data. The Agency
gathered the data directly from the source and only pulled data specific to sites that reported using 1,4-
dioxane in fracturing fluids. This source was added to the HERO database as well as the systematic
review process.
C.2.2.2 Additional Data Sources Identified for General Population, Consumer, and
Environmental Exposure
In addition to the gray literature sources listed in Appendix E of the 2021 Draft Systematic Review
Protocol ( '2la), several other gray literature sources were considered for inclusion the
general population, consumer, and environmental exposure assessment and added to the HERO database
in 2022. The Water Quality Portal (WQP) database, the successor of EPA's STORET (STOrage and
RETrieval) database, was incorporated because it includes a large variety of chemical-specific data.
Also, WQP is a portal that combines data from multiple databases—not just STORET—such as the U.S.
Geological Survey's National Water Information System. Information from WQP was collected in July
2022.
A few additional gray literature sources (databases) were included in the pool of references EPA
considered on general population, consumer, and environmental exposure through backwards searching,
which is described in Section 4.3.3 of the 2021 Draft Systematic Review Protocol (U.S. EPA. 2021a).
Backwards searching from the Third Unregulated Contaminant Monitoring Rule (UCMR3) database
(\ c. < ^ \ 201 (!) led to EPA's identification of data from a few states collecting data on 1,4-dioxane
for longer periods of time than reported in UCMR3. EPA was able to secure data from three state
databases, which as added. In addition, elevated levels of 1,4-dioxane in samples from UCMR3 and a
reference found in the pool of peer-reviewed articles led to addition to databases with data on 1,4-
dioxane levels measured in drinking water and surface water in the Cape Fear Watershed in North
Carolina. Six of the highest concentrations of 1,4-dioxane in the UCMR3 database were reported in
water from this watershed, which were also discussed in one of the references found in the literature
search for peer-reviewed sources. All gray literature database sources added to the search, including the
datasets from North Carolina, were added to HERO and the systematic review process (see Section G. 1
for further information).
C.2.3 Search Strings
As explained above in Section C.2.1, the search strings below were used to identify references relevant
to evaluating environmental releases and occupational exposure, as well as general population,
consumer, and environmental exposure.
C.2.3.1 Environmental Release and Occupational Exposure Search Strings
Life Cycle:
TIAB: ("1,4-dioxane" AND ("MFG" OR "import" OR "processing" OR "manufactur*" OR "releases"
OR "waste disposal" OR "reaction product" OR "repackaging" OR "recycling" OR "throughput" OR
"operating days" OR "batch" OR "production speed"))
Treatment Efficiencies:
TIAB: ("1,4-dioxane" AND ("GAC" OR "granular activated carbon" OR "reverse osmosis" OR
"advanced oxidation" OR "hydrogen peroxide with ultraviolet" OR ("hydrogen peroxide" AND "UV")
OR "hydrogen peroxide with ozone" OR ("hydrogen peroxide" AND "ozone") OR "AOP" OR
"Fenton's reagent" OR "bioremediation"))
Page 189 of 484
-------
5861
5862
5863
5864
5865
5866
5867
5868
5869
5870
5871
5872
5873
5874
5875
5876
5877
5878
5879
5880
5881
5882
5883
5884
5885
5886
5887
5888
5889
5890
5891
5892
5893
5894
5895
5896
5897
5898
5899
5900
5901
5902
5903
5904
5905
5906
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Occupational Workers:
TIAB: ("1,4-dioxane" AND ("janitor*" OR "mechanic" OR "laborer" OR "custodia*" OR "painter*"
OR "laboratory technician" OR "laboratory employee*" OR ("pharmaceutical" AND ("employee" OR
"worker" OR "technician")) OR "residential construction" OR "industrial construction"))
General:
TIAB: ("1,4-dioxane" AND ("surfactant" OR "ethoxylat*" OR "nonylphenol ethoxylate" OR
"alkylphenol ethoxylate" OR "sulfated" OR "industrial laundr*" OR "commercial laundr*" OR
"institutional laundr*" OR "institutional laundr*" OR "advanced oxidation" OR "ozone-peroxide
advanced oxidation" OR "low dioxane" OR "low dioxane ether sulfates" OR "low dioxane ethoxylated
surfactants" OR "low 1,4 dioxane ether sulfates" OR "low 1,4 dioxane ethoxylated surfactants" OR
"safety data sheet" OR "material safety data sheet"))
Process Uses:
TIAB: ("1,4-dioxane" AND ("stabilizer" AND ("chlorinated solvents" OR "degreasing" OR
"electronics manufacturing" OR "metal finishing")) OR ("solvent" AND ("histology" OR "cellulose
acetate membrane" OR "microscopy" OR "organic chemical manufacturing" OR "organic chemical"))
OR ("textile" AND ("wetting" OR "dispersing")) OR ("esterification" AND ("by-product" OR
"byproduct")))
Product Uses:
TIAB: ("1,4-dioxane" AND ("solvent" AND ("paint*" OR "lacquer*" OR "varnish remover" OR
"stain" OR "printing" OR "scintillation" OR "resin*" OR "oil*" OR "rubber chemicals" OR "rubber"
OR "sealant*" OR "adhesive*" OR "wax*" OR "cement*")))
TIAB: ("1,4-dioxane" AND ("artificial leather" OR "purifying agent" OR "antifreeze" OR "de-icing"
OR "pesticide*" OR "fumigant*"))
CASRNs of Ethoxylated Chemicals:
TIAB: ("9005-65-6" OR "3088-31-1" OR "68081-98-1" OR "68439-50-9" OR "68551-12-2" OR
"68439-49-6" OR "9043-30-5" OR "26183-52-8" OR "9002-92-0" OR "9004-82-4" OR "9005-64-5"
OR "68131-40-8" OR "68991-48-0" OR "37251-67-5" OR "5274-68-0" OR "864529-51-1" OR
"84133-50-6" OR "68439-45-2" OR "68987-81-5" OR "9003-11-6" OR "61791-29-5" OR "9005-08-7"
OR "61791-13-7" OR "166736-08-9" OR "3055-99-0" OR "66455-14-9" OR "68131-39-5" OR
"68213-23-0" OR "68951-67-7" OR "66455-15-0" OR "61791-26-2" OR "9004-95-9" OR "9005-00-9"
OR "61827-42-7" OR "68081-91-4" OR "68585-40-0" OR "68815-56-5" OR "61788-85-0" OR "3055-
97-8" OR "120313-48-6" OR "68439-46-3" OR "69227-22-1" OR "68002-97-1")
C.2.3.2 General Population, Consumer, and Environmental Exposure Search Strings
Population:
TIAB: ("general population" OR "bystanders" OR "near-facility" OR "industrial facilit*" OR
"commercial facilit*" OR "employee" OR "employees" OR "worker*" OR "manufacturer" OR "near-
disposal" OR "near surface disposal" OR "child*" OR "teenage*" OR "susceptible population" OR
"immunocompromised" OR "preschool" OR "senior*" OR "older adults" OR "elderly" OR "pregnant
women" OR "preexisting condition*" OR "lactating women" OR "childbearing" OR "prenatal" OR
"infant*" OR "adolescen*")
Landfills:
Page 190 of 484
-------
5907
5908
5909
5910
5911
5912
5913
5914
5915
5916
5917
5918
5919
5920
5921
5922
5923
5924
5925
5926
5927
5928
5929
5930
5931
5932
5933
5934
5935
5936
5937
5938
5939
5940
5941
5942
5943
5944
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TIAB: ("dioxane" AND ("landfill" OR "leach*" OR "incineration" OR "wastewater" OR"GAC" OR
"granular activated carbon" OR "reverse osmosis" OR "waste site" OR "land disposal" OR "waste
disposal" OR "landfill leach*"))
Indoor Air and Water:
TIAB: ("dioxane" AND ("inhal*" OR "tap water" OR "water well" OR "indoor air" OR "surface water"
OR "groundwater" OR "outdoor air" OR "ambient air" OR "drinking water" OR ("biomonitoring" OR
"monitoring" AND ("air" OR "water")) OR "drinking" OR "aquifer" OR "leach*" OR "municipal
water")) NOT ("spill")
Consumer and Industrial Use:
TIAB: ("1,4-dioxane" AND ("ingest*" OR "swallow*" OR "showering" OR "bathing" OR "swimming"
OR "wading" OR "inhal*" OR "paint*" OR "industrial manufactur*" OR "residential construction" OR
"commercial construction" OR "cleaning" OR "dishwasher" OR "printing" OR "food supplement*" OR
"packaging" OR "breast milk" OR "human milk" OR "intake rates" OR "launder*" OR "surface
cleaner" OR "automotive"))
Concentration and Dose:
TIAB: ("reference concentration" OR "RfC" OR "NOAEL" OR "LOAEL" OR "benchmark
concentration" OR "reference dose" OR "RfD" OR "chronic oral" OR "chronic inhalation" OR "oral
slope factor" OR "soil screening level" OR "PEL" OR "permissible exposure limit" OR "weighted
average" OR "weight fraction" OR "emission rate*" OR "inhalation unit risk" OR "IUR" OR "dose-
response" OR "reverse dosimetry" OR "biomonitoring" OR "media concentration*" OR ("estimate*"
AND ("acute" OR "subchronic" OR "chronic")) OR "single-dose" OR "repeated-dose" OR "daily
intake")
C.3 Data Screening
Sections 4.2.5 and 4.3.2 of the 2021 Draft Systematic Review Protocol describe how title and abstract
(TIAB) and full-text (FT) screening, respectively, are conducted to identify references that may contain
relevant information for use in risk evaluations under TSCA using discipline-specific screening criteria
(defined below in Sections C.3.1.1 and C.3.2.1 ( 2021a). Specifically, TIAB screening efforts
may be conducted using the specialized web-based software programs DistillerSR17 and SWIFT-Active-
Screener18; however, for this draft supplement, EPA used SWIFT-Active-Screener exclusively.
Additional details on how SWIFT Active-Screener utilizes a machine-learning algorithm to
automatically compute which unscreened documents are most likely to be relevant19 are available in
Section 4.2.5 of the 2021 Draft Systematic Review Protocol (U.S. EPA. 2021a). When TIAB screening
is completed, references that meet screening criteria {i.e., PECO/RESO/PESO statements) then undergo
FT screening if the full reference is able to be retrieved and generated into a Portable Document Format
(PDF).
17 As noted on the DistillerSR web page, this systematic review software "automates the management of literature collection,
triage, and assessment using AI and intelligent workflows...to produce transparent, audit ready, and compliant literature
reviews." EPA uses DistillerSR to manage the workflow related to screening and evaluating references; the literature search
is conducted external to DistillerSR.
18 SWIFT-Active Screener is another systematic review software that EPA is adopting in the TSCA systematic review
process. From Sciome's SWIFT-Active Screener web page: "As screening proceeds, reviewers include or exclude articles
while an underlying statistical model in SWIFT-Active Screener automatically computes which of the remaining unscreened
documents are most likely to be relevant. This 'Active Learning' model is continuously updated during screening, improving
its performance with each reference reviewed. Meanwhile, a separate statistical model estimates the number of relevant
articles remaining in the unscreened document list."
19 Description comes from the SWIFT-Active Screener web page.
Page 191 of 484
-------
5945
5946
5947
5948
5949
5950
5951
5952
5953
5954
5955
5956
5957
5958
5959
5960
5961
5962
5963
5964
5965
5966
5967
5968
5969
5970
5971
5972
5973
5974
5975
5976
5977
5978
5979
5980
5981
5982
5983
5984
5985
5986
5987
5988
5989
5990
5991
5992
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Literature inventory trees were introduced in the scoping process for the risk evaluations that began
systematic review in 2019 in response to comments received from the SACC and public to better
illustrate how references underwent various systematic review steps (e.g., TIAB and FT screening). As
explained in various final scope documents (e.g., Section 2.1.2 in the Final Scope of the Risk Evaluation
for 4,4'-(I-Me/ln letin liilene) bis/2, 6-dibromophenol] ( :0b)). literature inventory trees
demonstrate how references that meet screening criteria progress to the next systematic review step.
EPA used the Health Assessment Workplace Collaborative (HAWC) tool to develop web-based
literature inventory trees to enhance the transparency of the decisions resulting from the screening
processes. Updates made to the available literature considered for the draft supplement that are made
between publishing the draft and final Supplement to the Risk Evaluation for 1,4-Dioxane (e.g.,
additional references may be provided to the EPA through public comment) will be reflected in HAWC
(see also hyperlinks to HAWC in the figure captions below for each respective literature inventory tree).
The web-based literature inventory trees in HAWC also allow users to directly access the references in
the HERO database (more details available in Section 1 of the 2021 Draft Systematic Review Protocol)
by selecting appropriate nodes, which indicate whether a reference has met screening criteria at different
screening steps and/or types of content that may be discerned at that respective systematic review step
( 2021a). Furthermore, as mentioned in the various final scope documents, the sum of the
numbers for the various nodes in the literature inventory trees may be smaller or larger than the
preceding node because some studies may have unclear relevance or be relevant for many categories of
information. The screening process for each discipline varies and the nodes in the literature tree indicate
the screening decisions determined for each reference and whether specific content could be determined;
if no references had a specific screening decision and/or contained specific content relevant for a
respective discipline, a node will not be present on the literature tree to depict this.
In the literature inventory trees below, which depict systematic review search results used to evaluate the
new exposure pathways in this draft supplement, some references were unattainable for FT screening.
The "PDF not available" node refers to references or sources of information for which EPA was unable
to obtain the entire reference or source of data/information but were identified in the literature search
because of the availability of the title and abstract. For the references considered to evaluate
environmental release and occupational exposure, all references that passed screening criteria were
found and underwent FT screening. For the references considered to evaluate general population,
consumer, and environmental exposure, one PDF could not be obtained through interlibrary loan or
through other channels available to EPA to obtain reprints of published sources.
As mentioned in Section C.l, although all information contained in references that enter systematic
review are considered for use in the risk evaluation, the references that satisfy the screening criteria are
generally deemed to contain the most relevant and useful information for characterizing the uses,
exposure, and hazard of a chemical of interest and are generally utilized in the risk evaluation (and can
be used later on to identify further data needs). On the other hand, data or information sources that do
not satisfy the screening criteria outlined below may undergo data quality evaluation and extraction
should a data need arise for the risk evaluation.
C.3.1 Environmental Release and Occupational Exposure
During data screening, EPA followed the process described in Appendix H, Section H-3 of the 2021
Draft Systematic Review Protocol ( 021a) to conduct TIAB and FT screening for 1,4-
dioxane literature search results guided by the RESO statement. RESO stands for Receptors, Exposure,
Setting or Scenario, and Outcomes. Data or information sources that comply with the screening criteria
Page 192 of 484
-------
5993
5994
5995
5996
5997
5998
5999
6000
6001
6002
6003
6004
6005
6006
6007
6008
6009
6010
6011
6012
6013
6014
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
specified in the RESO statement then undergo data quality evaluation and extraction. Figure_Apx C-2
presents the number of references that report general engineering data, environmental release, and
occupational exposure data that passed RESO screening criteria at TIAB and FT screening.
C.3.1.1 Environmental Release and Occupational Exposure Literature Tree
46
Meets RESO Criteria at FT
Screening
©
General Engineering
Assessment
®
Peer-Reviewed Literature
Search
0
Occupational Exposure
®
Environmental Release
©
Meets RESO Criteria at TIAB
Screening
0 0
TSCA Engineering 1,4-Dioxane - Does Not Meet RESO Criteria at
Part 2 (2021) TIAB Screening
Ooes Not Meet RESO Criteria at
FT Screening
©
Supplemental From FT
Screening
©
Gray Literature Search
©
General Engineering
Assessment
©
Occupational Exposure
©
Supplemental From TIAB
Screening
Environmental Release
FigureApx C-2. Literature Inventory Tree - Environmental Releases and Occupational
Exposure Search Results for 1,4-Dioxane
View the interactive literature inventory tree in HAWC. Data in this figure represent all references obtained from
the publicly available databases and gray literature references searches that were included in systematic review as
of March 30, 2023. Additional data may be added to the interactive version as they become available.
C.3.2 General Population, Consumer, and Environmental Exposure
The TIAB and FT screening process was consistent with what EPA previously outlined in Sections 4.2.5
and 4.3.2 of the 2021 Draft Systematic Review Protocol ( J.S. EPA. 2021a). PECO stands for
Population, Exposure, Comparator or Scenario, and Outcomes for Exposure Concentration or Dose. The
PECO statement, as depicted in Appendix H.5.14 of the 2021 Draft Systematic Review Protocol (U.S.
EPA. 202la), was refined to better identify references that may contain information relevant for this
draft supplement. Specifically, data that are relevant for characterizing exposure to 1,4-dioxane in food,
including biota that humans consume, was not evaluated and extracted because 1,4-dioxane is not
expected to bioaccumulate in organi sms likely to be consumed by humans. During TIAB screening, if it
is unclear if a reference will meet the PECO screening criteria without having the full reference to
review, or if a reference is determined to meet the PECO screening criteria, that reference will advance
to FT screening. Studies containing potentially relevant supplemental material were also tracked and
Page 193 of 484
-------
6015
6016
6017
6018
6019
6020
6021
6022
6023
6024
6025
6026
6027
6028
6029
6030
6031
6032
6033
6034
6035
6036
6037
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
categorized during the literature screening process. Relevant supplemental material may be reviewed,
evaluated for data quality, and incorporated into risk evaluations, as needed. For example, references
were considered supplemental if they contained data from countries outside of North America on 1,4-
dioxane levels associated with landfills because different countries have very different waste
management policies (including requirements for landfills), and local hydrogeology in other regions
may not be relevant to sites in the United States. FigureApx C-3 presents the number of references that
report general population, consumer, and environmental exposure data that passed PECO screening
criteria at TIAB and FT screening.
C.3.2.1 General Population, Consumer, and Environmental Exposure Literature Tree
©
Monitoring Study
(-)
©
Peer-Reviewed Literature Modeling Study
Search
©
Experimental Study
Meets PECO Criteria at FT
Screening
9 @
Gray Literature Search Database
0
1,4-Dioxane Supplement: Does PECO Criteria at
_ _ TIAB Screening
General Population, Consumer, "
Does Not Meet PECO Criteria at
FT Screening
©
and Environmental Exposure Supplemental From FT
Screening
©
Supplemental From TIAB
Screening
Figure Apx C-3. Literature Inventory Tree - General Population, Consumer, and Environmental
Exposure Search Results for 1,4-Dioxane
View the interactive literature inventory tree in HAWC. Data in this figure represent all references obtained from
the publicly available databases and gray literature references searches that were included in systematic review as
of November 28, 2022. Additional data may be added to the interactive version as they become available.
C.4 Data Evaluation and Data Extraction
Data evaluation and extraction for this draft supplement are as described in Sections 5 and 6 of the 2021
Draft Systematic Review Protocol (U.S. EPA. 2021a). Data evaluation is the systematic review step in
which EPA assesses quality of the individual data sources using the evaluation strategies and criteria for
each topic area (e.g., physical and chemical property data, fate and transport data, occupational exposure
and environmental release data, general population, consumer, and environmental exposure data). The
evaluation method uses a structured framework with predefined criteria for each type of
Page 194 of 484
-------
6038
6039
6040
6041
6042
6043
6044
6045
6046
6047
6048
6049
6050
6051
6052
6053
6054
6055
6056
6057
6058
6059
6060
6061
6062
6063
6064
6065
6066
6067
6068
6069
6070
6071
6072
6073
6074
6075
6076
6077
6078
6079
6080
6081
6082
6083
6084
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
data/information source. The goal of the method used by EPA is to provide transparency, consistency,
and as much objectivity as possible to the evaluation process along with meeting the TSCA science
standards. Data extraction is the systematic review step in which EPA identifies quantitative and
qualitative data/information from data/information sources that meet screening criteria and extract the
data/information using structured forms or templates.
As explained above in Section C.l, terminology updates were made regarding the description of both
metric and overall data/information source quality determinations from what was originally described in
the 2021 Draft Systematic Review Protocol (U.S. EPA. 2021a). Specifically, metric and overall
data/information source quality determination terminology have been updated to "Metric Rating" and
"Overall Quality Determination", respectively. For additional clarifications regarding these updates,
please see TableApx C-l.
Although references that meet screening criteria following FT screening will generally proceed to data
quality evaluation and extraction steps, one clarification to the procedures outlined in Section 6 of the
2021 Draft Systematic Review Protocol is that in situations where EPA is unable to extract
data/information from sources that meet screening criteria (e.g., formatting prohibits accurate
extraction), such sources may not have extracted data to present in the risk evaluation or the respective
supplemental documents. Systematic Review Supplemental Files for the draft supplement contain results
from the data quality evaluation and extraction systematic review steps. Also, the template used to
display the data may be modified from those that were provided in the 2021 Draft Systematic Review
Protocol ( 21a) because the purpose of these supplemental documents is to accommodate
the data needs for each respective risk evaluation. The following sections provide specific information
about the data quality and extraction process followed to address the exposure pathways in scope for this
draft supplement and any clarifications or updates regarding these systematic review steps as described
in the 2021 Draft Systematic Review Protocol ( 1021a).
C.4.1 Environmental Release and Occupational Exposure
As described in the 2021 Draft Systematic Review Protocol, evaluation and extraction followed the
steps outlined in Sections 5, 6, and 6.2 ( 021a). The data extraction and data quality results
are summarized in Table Apx E-8 for air, Table Apx E-4 for water, Table Apx E-6 for land, and
Table Apx F-33 for occupational exposure. The 1,4-Dioxane Supplement to the Risk Evaluation Data
Quality Evaluation and Data Extraction Information for Environmental Release and Occupational
Exposure provides the results from the data extraction and quality evaluation, including metric rating
and the overall quality determination for each data source (U.S. EPA. 2023t).
C.4.2 General Population, Consumer, and Environmental Exposure
As described in the 2021 Draft Systematic Review Protocol, evaluation and extraction generally
followed the steps outlined in Section 5 and 6 ( 021a). However, a few updates were made to
the data quality evaluation metrics for a few evidence streams since the metrics were published in the
2021 Draft Systematic Review Protocol. Most of the changes were editorial or minor clarifications,
including the standardization of some metrics that apply to multiple evidence streams, where
appropriate. For example, in the Quality Assurance/Quality Control (QA/QC) metric for evaluating
monitoring and experimental evidence streams, the acronym QA/QC was defined and replaced all
references to quality assurance and quality control when occurring separately or together, and the term
QA/QC techniques was changed to QA/QC measures, which already appeared in the metrics.
A few metrics applicable to multiple evidence streams were modified slightly to better fit some of the
unique situations that frequently arise for a certain type of evidence stream (e.g., databases). For
Page 195 of 484
-------
6085
6086
6087
6088
6089
6090
6091
6092
6093
6094
6095
6096
6097
6098
6099
6100
6101
6102
6103
6104
6105
6106
6107
6108
6109
6110
6111
6112
6113
6114
6115
6116
6117
6118
6119
6120
6121
6122
6123
6124
6125
6126
6127
6128
6129
6130
6131
6132
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
example, some metrics were updated to clarify the intent of the metric and better account for variation in
types of evidence included in one grouping (e.g., experiments involving chamber studies vs. product
concentration assessments). The domains did not change; however, see below for the changes and
updates made to the data evaluation metrics for the respective evidence types (i.e., monitoring,
experimental studies and databases) as presented in Sections C.4.2.2, C.4.2.3 and C.4.2.4. No changes
were made to the data evaluation metrics for modeling data, as described in Appendix N Section N.6.2
in the 2021 Draft Systematic Review Protocol. The 1,4-Dioxane Supplement to the Risk Evaluation
Data Quality Evaluation Information for General Population, Consumer, and Environmental Exposure
provides details of the data quality evaluation results, including metric rating and the overall quality
determination for each data source ( 023 r).
Data extraction is the process in which quantitative and qualitative data/information are identified from
each relevant data/information source and extracted using structured forms or templates. Data extraction
was conducted as described in Section 6 of the 2021 Draft Systematic Review Protocol for all evidence
streams relevant for this supplement. However, with respect to information stored within databases, EPA
does not conduct a separate data extraction because the data are more accessible and have additional
context in the original database format. Both the date and data present in the database when the database
underwent FT screening are available in the HERO database (HERO IDs: 10365582, 10365609,
10365665, 10365667, 10365696, 10365698, 10368680, 10410586, and 10501014). If a reference or
data/information source (e.g., a peer-reviewed reference) presents data from a database that did not
undergo systematic review (e.g., a foreign database that is not publicly accessible), the data would be
extracted from the reference or data/information source to the extent possible; this did not apply to
references or sources of data or information that underwent systematic review for this supplement.
As mentioned above in Section C.4, references may not undergo data extraction, regardless of data
quality rating, if they contain no extractable data points (e.g., values are contained in a non-digitizable
figure or are representative of unspecified media or treatment processes). This constitutes an update to
Section 6 of the 2021 Draft Systematic Review Protocol ( 021a). Extraction forms and
templates are tailored to fit the data extraction needs for each risk evaluation.
The types of fields extracted vary by evidence stream and generally followed Section 6.3 of the 2021
Draft Systematic Review Protocol with regard to the data characteristics captured ( 021a).
Examples of types of data extracted and the extraction formats for the four evidence streams identified
through systematic review to evaluate environmental, general population, and consumer exposure data
are listed in the extraction tables provided in the 1,4-Dioxane Supplement to the Risk Evaluation Data
Extraction Information for General Population, Consumer, and Environmental Exposure (
2023h).
C.4.2.1 Data Quality Evaluation Metric Updates
Shown below are the data evaluation metrics for three evidence streams, presenting which data
evaluation metrics changed since the publication of the 2021 Draft Systematic Review Protocol (]j _S
21a). For evidence streams not listed below, there were no changes to the data evaluation
metrics since the 2021 Draft Systematic Review Protocol was published. Other data quality criteria for
studies on consumer, general population, and environmental exposure appear in Appendix N of the 2021
Draft Systematic Review Protocol ( )21a). For example, the criteria for modeling studies
appear in Table Apx N9. Data quality criteria for other types of studies (e.g., environmental release and
occupational exposure assessment) are published in other appendices to the 2021 Draft Systematic
Review Protocol ( 021a).
Page 196 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
6133 For the below tables in Sections C.4.2.2, C.4.2.3 and C.4.2.4, in order to make it easier for the reader to
6134 see what the changes were to the data evaluation metrics, the following convention is used: text inserted
6135 is underlined, and text deleted is in strikeout.
6136
6137
6138
C.4.2.2 Data Evaluation Criteria for Monitoring Data, as Revised
Table Apx C-2. Evaluation Criteria for Sources of Monitoring Data
Quality Rating
Description
Domain I Rcliabilil\
Mclnc I Sam pi i ny nvlhoi.loloij\
High
Samples were collected according to publicly available SOPs that are scientifically sound and
widely accepted (i.e., from a source generally using known to use sound methods and/or
approaches) for the chemical and media of interest. Example SOPs include U.S. Geological
Survey (USGS') "National Field Manual for the Collection of Water-Quality Data," EPA's
"Ambient Air Sampling" (SESDPROC-303-R5), etc.
OR
The sampling protocol used was not a publicly available SOP from a source generally known to
use using sound methods and/or approaches, but the sampling methodology is clear, appropriate
(i.e., scientifically sound), and similar to widely accepted protocols for the chemical and media
of interest. All pertinent sampling information is provided in the data source or companion
source. Examples include:
• Sampling equipment
• Sampling procedures/regime
• Sample storage conditions/duration
• Performance/calibration of sampler
• Study site characteristics
• Matrix characteristics
Medium
Sampling methodology is discussed in the data source or companion source and is generally
appropriate (i.e., scientifically sound) for the chemical and media of interest; however, one or
more pieces of sampling information is not described. The missing information is unlikely to
have a substantial impact on results.
OR
Standards, methods, protocols, or test guidelines may not be widely accepted, but a successful
validation study for the new/unconventional procedure was conducted prior to the sampling
event and is consistent with sound scientific theory and/or accepted approaches. Or a review of
information indicates the methodology is acceptable and differences in methods are not
expected to lead to lower quality data.
Low
Sampling methodology is only briefly discussed; therefore, most sampling information is
missing and likely to have a substantial impact on results.
AND/OR
The sampling methodology does not represent best sampling methods, protocols, or
guidelines for the chemical and media of interest (e.g., outdated [but still valid] sampling
equipment or procedures, long storage durations).
AND/OR
There are some inconsistencies in the reporting of sampling information (e.g., differences
between text and tables in data source, differences between standard method and actual
procedures reported to have been used, etc.) that led to a low confidence in the sampling
methodology used.
Page 197 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Critically
Deficient
The sampling methodology is not discussed in the data source or companion source.
AND/OR
Sampling methodology is not scientifically sound or is not consistent with widely accepted
methods/approaches for the chemical and media being analyzed (e.g., inappropriate sampling
equipment, improper storage conditions).
AND/OR
There are numerous inconsistencies in the reporting of sampling information, resulting in high
uncertainty in the sampling methods used.
Not Rated/Not
Applicable
Reviewer's
communis
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Mel lie 2 AikiK lical mclhodolo-jN
High
Samples were analyzed according to publicly available analytical methods that are scientifically
sound and widelv accepted (i.e., from a source generally using known to use sound methods
and/or approaches) and are appropriate for the chemical and media of interest. Examples
include EPA SW-846 Methods, NIOSH Manual of Analytical Methods 5th Edition, etc.
OR
The analytical method used was not a publicly available method from a source generally using
known to use sound methods and/or approaches, but the methodoloav is clear and appropriate
(i.e., scientifically sound) and similar to widely accepted protocols for the chemical and media
of interest. All pertinent sampling information is provided in the data source or companion
source. Examples include:
• Extraction method
• Analytical instrumentation (required)
• Instrument calibration
• Limit of quantitation (LOQ), LOD, detection limits, and/or reporting limits
• Recovery samples
• Biomarker used (if applicable)
• Matrix-adjustment method (i.e., creatinine, lipid, moisture)
Medium
Analytical methodology is discussed in detail and is clear and appropriate (i.e., scientifically
sound) for the chemical and media of interest; however, one or more pieces of analytical
information is not described. The missing information is unlikely to have a substantial impact
on results.
AND/OR
The analytical method may not be standard/widely accepted, but a method validation
study was conducted prior to sample analysis and is expected to be consistent with sound
scientific theory and/or accepted approaches.
AND/OR
Samples were collected at a site and immediately analyzed using an on-site mobile laboratory,
rather than shipped to a stationary laboratory.
Low
Analytical methodology is only briefly discussed. Analytical instrumentation is provided and
consistent with accepted analytical instrumentation/methods. However, most analytical
information is missing and likely to have a substantial impact on results.
AND/OR
Analytical method is not standard/widely accepted, and method validation is limited or not
available.
AND/OR
Samples were analyzed using field screening techniques.
AND/OR
Page 198 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
LOQ, LOD, detection limits, and/or reporting limits not reported.
AND/OR
There are some inconsistencies or possible errors in the reporting of analytical information
(e.g., differences between text and tables in data source, differences between standard method
and actual procedures reported to have been used, etc.) which leads to a lower confidence in the
method used.
Critically
Deficient
Analytical methodology is not described, including analytical instrumentation (i.e., HPLC,
GC).
AND/OR
Analytical methodology is not scientifically appropriate for the chemical and media being
analyzed (e.g., method not sensitive enough, not specific to the chemical, out of date).
AND/OR
There are numerous inconsistencies in the reporting of analytical information, resulting in high
uncertainty in the analytical methods used.
Not Rated/
Not Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
\lelrie 3 Seleelion of biomarker of exposure
High
Biomarker in a specified matrix is known to have an accurate and precise quantitative
relationship with external exposure, internal dose, or target dose (e.g., previous studies (or the
current study) have indicated the biomarker of interest reflects external exposures).
AND
Biomarker (parent chemical or metabolite) is derived from exposure to the chemical of interest.
Medium
Biomarker in a specified matrix has accurate and precise quantitative relationship with external
exposure, internal dose, or target dose.
AND
Biomarker is derived from multiple parent chemicals, not only the chemical of interest, but
there is a stated method to apportion the estimate to only the chemical of interest
Low
Biomarker in a specified matrix has accurate and precise quantitative relationship with external
exposure, internal dose, or target dose.
AND
Biomarker is derived from multiple parent chemicals, not only the chemical of interest, and
there is NOT an accurate method to apportion the estimate to only the chemical of interest.
OR
Biomarker in a specified matrix is a poor surrogate (low accuracy and precision) for
exposure/dose.
Critically
Deficient
Not applicable. A study will not be deemed critically deficient based on the use of biomarker of
exposure.
Not rated/
applicable
Metric is not applicable to the data source.
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 2 Represenlali\e
Melric 4 ( ieouranhic area
Page 199 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
High
Geographic location(s) is reported, discussed, or referenced.
Medium
Not applicable. This metric is dichotomous (i.e., high vs. critically deficient).
Low
Not applicable. This metric is dichotomous (i.e., high vs. critically deficient).
Critically
Deficient
Geographic location is not reported, discussed, or referenced.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
\lclnc 5 Tcmnoralih
High
Timing of sample collection for monitoring data is consistent with current or recent exposures
(within 5 years) may be expected.
Medium
Timing of sample collection for monitoring data is less consistent with current or recent
exposures (>5 to 15 years) may be expected.
Low
Timing of sample collection for monitoring data is not consistent with when current exposures
(>15 years old) may be expected and likely to have a substantial impact on results.
Critically
Deficient
Timing of sample collection for monitoring data is not reported, discussed, or referenced.
Not Rated/
Not Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Melnc 10 or more samples for a single scenario).
• Use of replicate samples.
• Use of systematic or continuous monitoring methods.
• Sampling over a sufficient period of time to characterize trends.
• For urine, 24-hour samples are collected (vs. first morning voids or spot).
• For biomonitoring studies, the timing of sample collected is appropriate based on
chemical properties (e.g., half-life), the pharmacokinetics of the chemical (e.g., rate of
uptake and elimination), and when the exposure event occurred.
Medium
Sampling approach likely captures variability of environmental contamination in
population/scenario/media of interest based on the heterogeneity/homogeneity and
dynamic/static state of the environmental system. Some uncertainty may exist, but it is unlikely
to have a substantial impact on results. For example:
• Moderate sample size (i.e., 5-10 samples for a single scenario), or
• Use of judgmental (non-statistical) sampling approach, or
• No replicate samples.
• For urine, first morning voids or pooled spot samples.
Low
Sampling approach poorly captures variability of environmental contamination in
population/scenario/media of interest. For example:
Page 200 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• Small sample size (i.e., <5 samples), or
• Use of haphazard sampling approach, or
• No replicate samples, or
• Grab or spot samples in single space or time, or
• Random sampling that does not include all periods of time or locations, or
• For urine, un-pooled spot samples.
Critically
Deficient
Sample size is not reported.
Single sample collected per data set.
For biomonitoring studies, the timing of sample collected is not appropriate based on chemical
properties (e.g., half-life), the pharmacokinetics of the chemical (e.g., rate of uptake and
elimination), and when the exposure event occurred.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Mcli'ic 7 IAf>osuiv scenario
High
The data closely represent relevant exposure scenario (i.e., the population/scenario/media of
interest). Examples include:
• Amount and type of chemical/product used
• Source of exposure
• Method of application or by-stander exposure
• Use of exposure controls
• Microenvironment (location, time, climate)
Medium
The data likely represent the relevant exposure scenario (i.e., population/scenario/media of
interest). One or more key pieces of information may not be described but the deficiencies
are unlikely to have a substantial impact on the characterization of the exposure scenario.
AND/OR
If surrogate data, activities seem similar to the activities within scope.
Low
The data lack multiple key pieces of information, and the deficiencies are likely to have a
substantial impact on the characterization of the exposure scenario.
AND/OR
There are some inconsistencies or possible errors in the reporting of scenario information
(e.g., differences between text and tables in data source, differences between standard method
and actual procedures reported to have been used, etc.) which leads to a lower confidence in the
scenario assessed.
AND/OR
If surrogate data, activities have lesser similarity but are still potentially applicable to the
activities within scope.
Critically
Deficient
If reported, the exposure scenario discussed in the monitored study does not represent the
exposure scenario of interest for the chemical.
Not rated/
Not applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 3 Acccssilnlih clanl\
\Iclric N. Renoilintj ol'ivsulls
Page 201 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
High
Supplementary or raw data (i.e., individual data points) are reported, allowing summary
statistics to be calculated or reproduced.
AND
• Description of data set summarized (i.e., location, population, dates, etc.)
• Range of concentrations or percentiles
• Number of samples in data set
• Frequency of detection
• Measure of variation (coefficient of variation [CV], standard deviation)
• Measure of central tendency (mean, geometric mean, median)
• Test for outliers (if applicable)
Summary statistics are detailed and complete. Example parameters include:
AND
Both adjusted and unadjusted results are provided (i.e., correction for void completeness in
urine biomonitoring, whole-volume or lipid adjusted for blood biomonitoring, wet or dry
weight for environmental tissue samples or soil samples) [only if applicable].
Medium
Supplementary or raw data (i.e., individual data points) are not reported, and therefore
summary statistics cannot be reproduced.
AND/OR
Summary statistics are reported but are missing one or more parameters (see description for
high).
AND/OR
Only adjusted or unadjusted results are provided, but not both [only if applicable].
Low
Supplementary data are not provided, and summary statistics are missing most parameters (see
description for high).
AND/OR
There are some inconsistencies or errors in the results reported, resulting in low confidence in
the results reported (e.g., differences between text and tables in data source, less appropriate
statistical methods).
Critically
Deficient
There are numerous inconsistencies or errors in the calculation and/or reporting of results,
resulting in highly uncertain reported results.
Not Rated/
Not Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Mel l ie ^ Oualil\ assurance
High
The study quality assurance/quality control (QA/QC) measures and all pertinent QA/QC
information is provided in the data source or companion source. Examples include:
• Field, laboratory, and/or storage recoveries
• Field and laboratory control samples
• Baseline (pre-exposure) samples
• Biomarker stability
• Completeness of sample (i.e., creatinine, specific gravity, osmolality for urine
samples)
AND
No QA/QC issues were identified, or any identified issues were minor and adequately
addressed (i.e., correction for low recoveries, correction for completeness).
Medium
The study applied and documented QA/QC measures; however, one or more pieces of QA/QC
information is not described. Missing information is unlikely to have a substantial impact on
results.
Page 202 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
AND
No QA/QC issues were identified, or any identified issues were minor and addressed (i.e.,
correction for low recoveries, correction for completeness).
Low
QA/QC measures and results were not directly discussed but are implied through the study's
use of standard field and laboratory protocols.
AND/OR
Deficiencies were noted in QA/QC control measures that are likely to have a substantial impact
on results.
AND/OR
There are some inconsistencies in the QA/QC measures reported, resulting in low confidence in
the QA/QC measures taken and results (e.g., differences between text and tables in data
source).
Critically
Deficient
QA/QC issues have been identified which significantly interfere with the overall reliability of
the study.
Not Rated/
Not Applicable
Reviewer's
commcnls
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 4 Variabilis and unccrlains
\lclric In Variabilis and unccrlains
High
The slud) characterizes \ ariabi hL\ in Lhc population, media sludicd.
AND
Key uncertainties, limitations, and data gaps have been identified.
AND
The uncertainties are minimal and have been characterized.
Medium
The study has limited characterization of variability in the population/media studied.
AND/OR
The study has limited discussion of key uncertainties, limitations, and data gaps.
AND/OR
Multiple uncertainties have been identified but are unlikely to have a substantial impact on
results.
Low
The characterization of variability is absent.
AND/OR
Key uncertainties, limitations, and data gaps are not discussed.
AND/OR
Uncertainties identified may have a substantial impact on the exposure the exposure assessment
Critically
Deficient
Estimates are highly uncertain based on characterization of variability and uncertainty.
Not Rated/
Not Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Page 203 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
6139 C.4.2.3 Data Evaluation Criteria for Experimental Data, as Revised
6140
6141 Table Apx C-3. Evaluation Criteria for Sources of Experimental Data
Data Quality
Rating
Metric Description
Domain 1 Rcliabilil\
\lclnc 1 Sam pi i nu \lclhodoloij\ and ( oiidilions
High
Samples were collected according to publicly available SOPs, methods, protocols, or test
guidelines that are scientifically sound and widely accepted from a source generally known to
use sound methods and/or approaches such as EPA, NIST, American Society for Testing and
Materials, ISO, and ACGIH.
OR
The sampling protocol used was not a publicly available SOP from a source generally known
to use sound methods and/or approaches, but the sampling methodology is clear, appropriate
(i.e.. scientifically sound), and similar to widely accepted protocols for the chemical and
media of interest. All pertinent sampling information is provided in the data source or
companion source. Examples include:
• Sampling conditions (e.g., temperature, humidity)
• Sampling equipment and procedures
• Sample storage conditions/duration
• Performance/calibration of sampler
Medium
Sampling methodology is discussed in the data source or companion source and is generally
appropriate (i.e.. scientifically sound) for the chemical and media of interest, however, one or
more pieces of sampling information is not described. The missing information is unlikely to
have a substantial impact on results.
OR
Standards, methods, protocols, or test guidelines may not be widely accepted, but a successful
validation study for the new/unconventional procedure was conducted prior to the sampling
event and is consistent with sound scientific theory and/or accepted approaches.
Low
Sampling methodology is only briefly discussed. Therefore, most sampling information is
missing and likely to have a substantial impact on results.
AND/OR
The sampling methodology does not represent best sampling methods, protocols, or guidelines
for the chemical and media of interest (e.g., outdated (but still valid) sampling equipment or
procedures, long storage durations).
AND/OR
There are some inconsistencies in the reporting of sampling information (e.g., differences
between text and tables in data source, differences between standard method and actual
procedures reported to have been used, etc.) which lead to a low confidence in the sampling
methodology used.
Critically
Deficient
The sampling methodology is not discussed in the data source or companion source.
AND/OR
Sampling methodology is not scientifically sound or is not consistent with widely accepted
methods/approaches for the chemical and media being analyzed (e.g., inappropriate sampling
equipment, improper storage conditions).
AND/OR
There are numerous inconsistencies in the reporting of sampling information, resulting in high
uncertainty in the sampling methods used.
Page 204 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
High
Samples were analyzed according to publicly available analytical methods that are
scientifically sound and widely accepted (i.e., from a source generally using sound methods
and/or approaches) and are appropriate for the chemical and media of interest. Examples
include EPA SW-846 Methods, NIOSH Manual of Analytical Methods 5th Edition, etc.
OR
The analytical method used was not a publicly available method from a source generally
known to use sound methods and/or approaches, but the methodology is clear and appropriate
(/. e., scientifically sound) and similar to widely accepted protocols for the chemical and media
of interest. All pertinent analytical sampling information is provided in the data source or
companion source. Examples include:
• Extraction method
• Analytical instrumentation (required)
• Instrument calibration
• LOQ, LOD, detection limits, and/or reporting limits
• Recovery samples
• Biomarker used (if applicable)
• Matrix-adjustment method (B creatinine, lipid, moisture)
Medium
Analytical methodology is discussed in detail and is clear and appropriate (i.e.. scientifically
sound) for the chemical and media of interest; however, one or more pieces of analytical
information is not described. The missing information is unlikely to have a substantial impact
on results.
AND/OR
The analytical method may not be standard/widely accepted, but a method validation study
was conducted prior to sample analysis and is expected to be consistent with sound scientific
theory and/or accepted approaches.
AND/OR
Samples were collected at a site and immediately analyzed using an on-site mobile laboratory,
rather than shipped to a stationary laboratory.
Low
Analytical methodology is only briefly discussed. Analytical instrumentation is provided and
consistent with accepted analytical instrumentation/methods. However, most analytical
information is missing and likely to have a substantial impact on results.
AND/OR
Analytical method is not standard/widely accepted, and method validation is limited or not
available.
AND/OR
Samples were analyzed using field screening techniques.
AND/OR
LOQ, LOD, detection limits, and/or reporting limits not reported.
AND/OR
There are some inconsistencies or possible errors in the reporting of analytical information
(e.g., differences between text and tables in data source, differences between standard method
and actual procedures reported to have been used, etc.) which leads to a lower confidence in
the method used.
Critically
Deficient
Analytical methodology is not described, including analytical instrumentation (/. e., HPLC,
GC).
AND/OR
Page 205 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Analytical methodology is not scientifically appropriate for the chemical and media being
analyzed (e.g., method not sensitive enough, not specific to the chemical, out of date).
AND/OR
There are numerous inconsistencies in the reporting of analytical information, resulting in high
uncertainty in the analytical methods used.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Melnc 3 Selection ol'luomaiker ol'e\nosure
High
Biomarker in a specified maln\ is known lo ha\ e an accurate and precise quantilaln e
relationship with external exposure, internal dose, or target dose (e.g., previous studies (or the
current study) have indicated the biomarker of interest reflects external exposures).
AND
Biomarker (parent chemical or metabolite) is derived from exposure to the chemical of
interest.
Medium
Biomarker in a specified matrix has accurate and precise quantitative relationship with
external exposure, internal dose, or target dose.
AND
Biomarker is derived from multiple parent chemicals, not only the chemical of interest, but
there is a stated method to apportion the estimate to only the chemical of interest
Low
Biomarker in a specified matrix has accurate and precise quantitative relationship with
external exposure, internal dose, or target dose.
AND
Biomarker is derived from multiple parent chemicals, not only the chemical of interest, and
there is NOT a stated method to apportion the estimate to only the chemical of interest.
OR
Biomarker in a specified matrix is a poor surrogate (low accuracy and precision) for
exposure/dose.
Critically
Deficient
Not applicable. A study will not be deemed critically deficient based on the use of biomarker
of exposure.
Not Rated/Not
Applicable
Metric is not applicable to the data source.
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Metric 4 Testing
scenario
High
Testing conditions closely represent relevant exposure scenarios (i.e.,
population/scenario/media of interest). Examples include:
• Amount and type of chemical/product used
• Source of exposure/test substance
• Method of application or by-stander exposure
• Use of exposure controls
• Microenvironment (location, time, climate, temperature, humidity, pressure, airflow)
AND
Testing conducted under a broad range of conditions for factors such as temperature,
humidity, pressure, airflow, and chemical mass/weight fraction (if appropriate).
Page 206 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Medium
The data likely represent the relevant exposure scenario (i.e., population/scenario/media of
interest). One or more key pieces of information may not be described but the deficiencies are
unlikely to have a substantial impact on the characterization of the exposure scenario.
AND/OR
If surrogate data, activities seem similar to the activities within scope.
Low
The data lack multiple key pieces of information and the deficiencies are likely to have a
substantial impact on the characterization of the exposure scenario.
AND/OR
There are some inconsistencies or possible errors in the reporting of scenario information (e.g.,
differences between text and tables in data source, differences between standard method and
actual procedures reported to have been used, etc.) which leads to a lower confidence in the
scenario assessed.
AND/OR
If surrogate data, activities have lesser similarity but are still potentially applicable to the
activities within scope.
AND/OR
Testing conducted under a single set of conditions^except for experiments to determine a
weight fraction or concentration in a product.
Critically
Deficient
Testing conditions are not relevant to the exposure scenario of interest for the chemical.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Melnc 5 Saninle si/.e and \analnlil\
High
Sample size is reported and large enough (i.e., >10 samples) to be reasonably assured that the
samples represent the scenario of interest.
AND
Replicate tests performed and variability across tests is characterized (if appropriate).
Medium
Sample size is moderate (i.e., 5 to <10 samples), thus the data are likely to represent the
scenario of interest.
AND
Replicate tests performed and variability across tests is characterized (if appropriate).
Low
Sample size is small (i.e., <5 samples for most types of experiments or 1 per product for
experiments to determine a weight fraction or concentration in a product), thus the data are
likely to poorly represent the scenario of interest.
AND/OR
Replicate tests were not performed.
Critically
Deficient
Sample size is not reported.
AND/OR
Single sample collected per data set,.except for experiments to determine a weight fraction or
concentration in a product.
AND/OR
For biomonitoring studies, the timing of sample collected is not appropriate based on chemical
properties (e.g., half-life), the pharmacokinetics of the chemical (e.g., rate of uptake and
elimination), and when the exposure event occurred.
Not Rated/Not
Applicable
Page 207 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
\lclnc (¦> Taii|iomlii\
High
Source(s) of tested items appears to be current (within 5 years).
Medium
Source(s) of tested items is less consistent with when current or recent exposures (>5 to 15
years) are expected.
Low
Source(s) of tested items is not consistent with when current or recent exposures (>15 years)
are expected or is not identified.
Critically
Deficient
Temporality of tested items is not reported, discussed, or referenced.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 3 Acccssilnlil\ clanl\
Mcli'ic 7. Kqioriinu nl'ivsulls
High
Supplementary or raw data (i.e.. individual data points) are reported, allowing summary
statistics to be calculated or reproduced.
AND
Summary statistics are detailed and complete. Example parameters include:
• Description of data set summarized (i.e., location, population, dates, etc.)
• Range of concentrations or percentiles
• Number of samples in data set
• Frequency of detection
• Measure of variation (CV, standard deviation)
• Measure of central tendency (mean, geometric mean, median)
• Test for outliers (if applicable)
AND
Both adjusted and unadjusted results are provided (i.e., correction for void completeness in
urine biomonitoring, whole-volume or lipid adjusted for blood biomonitoring) [only if
applicable].
Medium
Supplementary or raw data (i.e.. individual data points) are not reported, and therefore
summary statistics cannot be reproduced.
AND/OR
Summary statistics are reported but are missing one or more parameters (see description for
high).
AND/OR
Only adjusted or unadjusted results are provided, but not both [only if applicable].
Low
Supplementary data are not provided, and summary statistics are missing most parameters (see
description for high).
AND/OR
There are some inconsistencies or errors in the results reported, resulting in low confidence in
the results reported (e.g., differences between text and tables in data source, less appropriate
statistical methods).
Critically
Deficient
There are numerous inconsistencies or errors in the calculation and/or reporting of results,
resulting in highly uncertain reported results.
Page 208 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Not Rated/Not
Applicable
Reviewer's
commenls
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Mel lie S OualiS assurance
High
The study applied quality assurance/quality control (QA/QC) measures and all pertinent
QA/QC information is provided in the data source or companion source. Examples include:
• Laboratory, and/or storage recoveries.
• Laboratory control samples.
• Baseline (pre-exposure) samples.
• Biomarker stability
• Completeness of sample (i.e., creatinine, specific gravity, osmolality for urine
samples)
AND
No QA/QC issues were identified, or any identified issues were minor and adequately
addressed (i.e., correction for low recoveries, correction for completeness).
Medium
The study applied and documented QA/QC measures; however, one or more pieces of QA/QC
information is not described. Missing information is unlikely to have a substantial impact on
results.
AND
No QA/QC issues were identified, or any identified issues were minor and addressed (i.e.,
correction for low recoveries, correction for completeness).
Low
QA/QC measures and results were not directly discussed but are implied through the study's
use of standard field and laboratory protocols.
AND/OR
Deficiencies were noted in QA/QC measures that are likely to have a substantial impact on
results.
AND/OR
There are some inconsistencies in the QA/QC measures reported, resulting in low confidence
in the QA/QC measures taken and results (e.g., differences between text and tables in data
source).
Critically
Deficient
QA/QC issues have been identified which significantly interfere with the overall reliability of
the study.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 4 Variabilis and unceilains
\Iclnc Variabilis and unceilaum
High
The study characterizes variability in the population/media studied.
AND
Key uncertainties, limitations, and data gaps have been identified.
AND
The uncertainties are minimal and have been characterized.
Medium
The study has limited characterization of variability in the population/media studied.
AND/OR
Page 209 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
The study has limited discussion of key uncertainties, limitations, and data gaps.
AND/OR
Multiple uncertainties have been identified but are unlikely to have a substantial impact on
results.
Low
The characterization of variability is absent.
AND/OR
Key uncertainties, limitations, and data gaps are not discussed.
AND/OR
Uncertainties identified may have a substantial impact on the exposure the exposure
assessment
Critically
Deficient
Estimates are highly uncertain based on characterization of variability and uncertainty.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
6142 C.4.2.4 Data Evaluation Criteria for Databases, as Revised
6143
Table Apx C-4
. Evaluation Criteria for Sources of Database Data
Data Quality
Rating
Description
Domain 1 Rcliubilih
Mclnc 1. Samnlinu nvlhodolou\
High
Widelv accepted sampling methodologies (i.e., from a source generally known to use using
sound methods and/or approaches) were used to generate the data presented in the database.
Example SOPs include USGS's "National Field Manual for the Collection of Water-Quality
Data," EPA's "Ambient Air Sampling" (SESDPROC-303-R5), etc.
Medium
One or more pieces of sampling methodology information is not described, but missing
information is unlikely to have a substantial impact on results.
OR
The sampling methodologies were consistent with sound scientific theory and/or accepted
approaches based on the reported sampling information but may not have followed published
procedures from a source generally known to use sound methods and/or approaches.
Low
The sampling methodologv was not reported in data source or readilv available companion data
source.
Critically
Deficient
The sampling methodologies used were not appropriate for the chemical/media of interest in
the database (e.g., inappropriate sampling equipment, improper storage conditions).
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
\lclnc 2 AnaK Heal mclhodolou\
Page 210 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
High
Widely accepted analytical methodologies (i.e., from a source generally using sound methods
and/or approaches) were used to generate the data presented in the database. Example SOPs
include EPA SW-846 Methods, NIOSH Manual of Analytical Methods 5th Edition, etc.
Medium
The analytical methodologies were consistent with sound scientific theory and/or accepted
approaches based on the reported analytical information but may not have followed published
procedures from a source generally known to use sound methods and/or approaches.
Low
The analytical methodology was not reported in data source or companion data source.
Critically
Deficient
The analytical methodologies used were not appropriate for the chemical/media of interest in
the database (e.g., method not sensitive enough, not specific to the chemical, out of date).
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 2 Rqnvsailali\c
\lclnc 3 ( ieouianhic aiva
High
Geographic location(s) is reported, discussed, or referenced.
Medium
Not applicable. This metric is dichotomous (i.e., high vs. critically deficient).
Low
Not applicable. This metric is dichotomous (i.e., high vs. critically deficient).
Critically
Deficient
Geographic location is not reported, discussed, or referenced.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
\lclnc4 Temporal
High
The data reflect current conditions (within 5 years)
AND/OR
Database contains robust historical data for spatial and temporal analyses (if applicable).
Medium
The data are less consistent with current or recent exposures (>5 to 15 years)
AND/OR
Database contains sufficient historical data for spatial and temporal analyses (if applicable).
Low
Data are not consistent with when current exposures (>15 years old) may be expected
AND/OR
Database does not contain enough historical data for spatial and temporal analyses (if
applicable).
Critically
Deficient
Timing of sample data is not reported, discussed, or referenced.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Page 211 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Mclnc 5 1 \nnsuiv scenario
High
The data closely represent relevant exposure scenario (i.e., the population/scenario/media of
interest). Examples include:
• Amount and type of chemical/product used
• Source of exposure
• Method of application or by-stander exposure
• Use of exposure controls
• Microenvironment (location, time, climate)
Medium
The data likely represent the relevant exposure scenario (i.e., population/scenario/media of
interest). One or more key pieces of information may not be described but the deficiencies
are unlikely to have a substantial impact on the characterization of the exposure scenario.
AND/OR
If surrogate data, activities seem similar to the activities within scope.
Low
The data lack multiple key pieces of information and the deficiencies are likely to have a
substantial impact on the characterization of the exposure scenario.
AND/OR
There are some inconsistencies or possible errors in the reporting of scenario information
(e.g., differences between text and tables in data source, differences between standard method
and actual procedures reported to have been used, etc.) which leads to a lower confidence in the
scenario assessed.
AND/OR
If surrogate data, activities have lesser similarity but are still potentially applicable to the
activities within scope.
Critically
Deficient
If reported, the exposure scenario discussed in the monitored study does not represent the
exposure scenario of interest for the chemical.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 3 \ccessilnlil\ clanl\
\lclnc ^ A\ailaliilil\ ol'dalalxise and sunnorlinu documenls
High
Database is widely accepted and/or from a source generally known to use sound methods
and/or approaches (e.g., raw data from NHANES. STORET).
Medium
The database may not be widely known or accepted (e.g., state-maintained databases), but the
database is adequately documented with most or all of the following information:
• Within the database, metadata is present (sample identifiers, annotations, flags, units,
matrix descriptions, etc.) and-data fields are generally clear and defined.
• A user manual and other supporting documentation is available, or there is sufficient
documentation in the data source or companion source.
• Database quality assurance and data quality control measures are defined and/or a
QA/QC protocol was followed.
Low
The database may not be widely known or accepted, and only limited database documentation
is available (see the medium rating).
Critically
Deficient
No information is provided on the database source or availability to the public.
Page 212 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Not Rated/
Applicable
Reviewer's
communis
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Mclnc 7 Rcinu'linu of ivsulls
High
The database or information source reporting the analvsis of the database data is well organized
and understandable by the target audience.
AND
Summary statistics in the data source are detailed and complete. Example parameters include:
• Description of data set summarized (i.e., location, population, dates, etc.)
• Range of concentrations or percentiles
• Number of samples in data set
• Frequency of detection
• Measure of variation (CV, standard deviation)
• Measure of central tendency (mean, geometric mean, median)
• Test for outliers (if applicable)
Medium
The database or information source reporting the analysis of the database data is well organized
and understandable by the target audience.
AND/OR
Summary statistics are missing one or more parameters (see description for high).
Low
The database or information source reporting the analysis of the database data is unclear or not
well organized.
AND/OR
Summary statistics are missing most parameters (see description for high)
AND/OR
There are some inconsistencies or errors in the results reported, resulting in low confidence in
the results reported (e.g., differences between text and tables in data source, less appropriate
statistical methods).
Critically
Deficient
There are numerous inconsistencies or errors in the calculation and/or reporting of results,
resulting in highly uncertain reported results.
AND/OR
The information source reporting the analysis of the database data is missing key sections or
lacks enough organization and clarity to locate and extract necessary information.
Not Rated/Not
Applicable
Reviewer's
Comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
Domain 4 \ anabilih and uiKvrlainl\
\lclnc S Yanalnlih and uiKvi'lanm
High
Variabilitv. kev uncertainties, limitations, and/or data saps have been identified.
AND/OR
The uncertainties are minimal and have been characterized.
Medium
The studv has limited discussion of variabilitv. kev uncertainties, limitations, and/or data saps.
AND/OR
Multiple uncertainties have been identified but are unlikely to have a substantial impact on
results.
Page 213 of 484
-------
6145
6146
6147
6148
6149
6150
6151
6152
6153
6154
6155
6156
6157
6158
6159
6160
6161
6162
6163
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Low
Variabilitv. kev uncertainties, limitations, and data saps are not discussed.
AND/OR
Uncertainties identified may have a substantial impact on the exposure the exposure assessment
Critically
Deficient
Estimates are highly uncertain based on characterization of variability and uncertainty.
Not Rated/Not
Applicable
Reviewer's
comments
[Document concerns, uncertainties, limitations, and deficiencies and any additional comments
that may highlight study strengths or important elements such as relevance]
C.5 Evidence Integration
As described in Section 7 of the 2021 Draft Systematic Review Protocol (U.S. EPA.. 2021a). evidence
integration refers to the consideration of evidence obtained from systematic review and scientific
information obtained from sources that did not undergo systematic review to implement a weight of the
scientific evidence approach. 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 CFR
702.33).The consideration of the quality and relevance of the data, while taking into account the
strengths and limitations of the data, to appropriately evaluate the evidence for this supplement, is
described in Section 7 of the 2021 Draft Systematic Review Protocol (U.S. EPA. 2021a).
TableApx C-5 and TableApx C-6, originally from Section 7.3 of the 2021 Draft Systematic Review
Protocol, provide general considerations and examples of factors that contribute to the strength of
evidence for each evidence stream and example weight of the scientific evidence judgments based on
these general considerations, respectively, when evaluating potentially relevant exposure data for this
draft supplement ( 021a).
Table Apx C-5. Considerations that Inform Evaluations of the Strength of the Evidence
Considerations
Factors that
Increase Strength
Factors that
Decrease Strength
The overall weight of the scientific evidence judgment considers the general considerations below as well as
chemical-specific considerations to designate each exposure scenario as robust, moderate, slight, or
indeterminate. The designation is a measure of the weight of the evidence supporting the representativeness of
the exposure estimates toward the true distribution of exposure (and releases) for the scenario.
Relevance to
exposure scenario
• Directly relevant to evaluated
exposure scenario
• Data used is for an alternative or surrogate
scenario
For modeled
estimates
• Model used has been peer-
reviewed and is being applied in a
manner appropriate to its design
and objective
• Evidence demonstrating implausibility
• Model has not been peer-reviewed and no
ground-truthing has been performed
• Parameterization is not well described,
documented or is not appropriate to the
evaluated scenario
Data quality
• Medium or high data quality
rating (via Data Evaluation)
• Low data quality rating (via Data Evaluation)
• Imprecision or inaccuracy
Page 214 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Considerations
Factors that
Increase Strength
Factors that
Decrease Strength
Data points
• High number of data points
• Low number of data points
• High proportion of data sampled prior to
changes in industry or other relevant
conditions (e.g., OSHA PEL)
Representative of
the whole industry
(for occupational
scenarios)
• Large proportion of sites included
within the exposure scenario were
measured
• Evidence may not be sufficiently
representative of all of the sites for the
exposure scenario
Representative of
the sub-population
• Applicable to most or all of the
different population groups
included within the exposure
scenario
• Information was not available to sufficiently
cover most or all of population groups
included within the exposure scenarios
Consistency
• Consistency and replication
within a study and across studies
• Inexplicable contradictory findings across
studies
Variability
• Variability is accounted in
estimates
• Full distributions of input
parameters
• Variability unaccounted in estimates
Uncertainties
• Uncertainties are low and the
uncertainties are unlikely to
significantly impact exposure
estimates
• Uncertainties that are likely to over- or under-
estimate exposure from the actual exposures
for the exposure scenario
6164
Page 215 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table Apx C-6. Evaluation of the Weight of the Scientific Evidence for
Exposure Assessments
Category
Robust
Moderate
Slight
Indeterminate
Overall Weight of
the Scientific
Evidence
Exposure Scenario
Factors
(e.g., habits, worker
activities, exposure
factors)
• Directly relevant to
evaluated exposure
scenario
• Applicable to most or all of
the different population
groups included within the
exposure scenario
• Full distributions of input
parameters
• High or medium quality
data ratings
• The habits, worker
activities, and/or use
patterns are accounted for,
are current
• Uncertainties are low and
the uncertainties are
unlikely to significantly
impact exposure estimates
• Surrogate scenarios from
similar chemicals are used
to infer similar exposures
or emissions.
• Some distribution of input
parameters
• High or medium quality
data ratings
• There is some, but not
complete, documentation
or description of
assumptions, limitations
and uncertainties
• Surrogate scenarios from
similar uses are used to
infer similar use patterns or
habits and practices
• Medium or low quality
data ratings
• Partially supported by
assumptions
• Uncertainties are not
fully known or
documented
• Habits and practices are
not fully known and
there is a high degree of
uncertainty in defining
use patterns
• Qualitative
descriptions of
exposure without
additional context.
• No supporting data
on habits and
practices are
available
The consideration
factors and the
categories to the left
result in an overall
weight of the
scientific evidence
judgment as one of
the following:
• Robust
• Moderate
• Slight
• Indeterminate
Measured/Monitored
Data
There is measured information
and the temporal and spatial
aspect of the measurements
are well described, relevant
and reflect current conditions
• Medium or high data
quality rating (via Data
Evaluation)
• High number of data points
• Multiple studies or a large
number of data points
which indicate similar
findings
• Large proportion of sites
included within the
exposure scenario were
measured
There is measured information
which does not reflect current
environmental conditions or
does not correspond to current
activities but provides
evidence of exposure
• Limited number of studies
or limited number of data
points which indicate similar
findings
• Information was not
available to sufficiently
cover most or all of
population groups included
within the exposure
scenarios
• There is some, but not
complete, documentation or
There is limited measured
information and
information and does not
reflect exposure conditions
and does not correspond to
known activities
• Information was not
available to sufficiently
cover most or all of
population groups
included within the
exposure scenarios
• Assumptions and
uncertainties are not
known or documented
No measured or
monitored data are
available
Page 216 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Category
Robust
Moderate
Slight
Indeterminate
Overall Weight of
the Seientifie
Evidcnee
• Consistency and
replication within a study
and across studies
• Uncertainties are low and
the uncertainties are
unlikely to significantly
impact exposure estimates
• Sensitivity of the exposure
estimates has been
described and quantified
incorporating assumptions,
limitations, and
uncertainties
description of assumptions,
limitations, and uncertainties
The consideration
factors and the
categories to the left
result in an overall
weight of the
scientific evidence
judgment as one of
the following:
• Robust
• Moderate
• Slight
• Indeterminate
The consideration
factors and the
categories to the left
results in an overall
weight of the
scientific evidence
judgment as one of
the following:
• Robust
• Moderate
• Slight
• Indeterminate
Estimation
Methodology/Data
• The methodology for
deriving the estimate is
well described and the
underlying computational
and/or scientific basis is
robust, has an empirical
basis or well documented
mathematical basis and
considers chemical
specificity (e.g., physical
and chemical properties
and fate)
• Applicable to most or all
of the different population
groups included within the
exposure scenario
(representative)
• Sensitivity of the exposure
estimates has been
described and quantified
incorporating assumptions,
limitations, and
uncertainties
• The methodology for
deriving the estimate is
well described and the
underlying computational
and/or scientific basis is
robust, however there is
uncertainty in the
parameterization or
applicability
• There is some, but not
complete, documentation
or description of
assumptions, limitations
and uncertainties.
• Modeling approach used
to estimate exposures is
not rooted in scientific
rigor or does not
mathematically
represent the exposure
scenario;
parameterization is not
complete or does not
utilize the best available
science.
• Assumptions and
uncertainties are not
known or documented
• Modeling approach
is not available for
the scenario or lack
of information on
parameters
prohibits use of
available models.
Comparison of
Estimated and
Measured Exposures (if
• There are comparable
estimates using alternate
approaches
• Modeled estimates and
measured exposure values
are comparable, however
• There is a lack of
correspondence between
measured exposures and
• Category does not
have indeterminate
criterion.
Page 217 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Category
Robust
Moderate
Slight
Indeterminate
Overall Weight of
the Seientifie
Evidcnee
both estimated and
measured estimates are
used)
• There is concordance
between measured and/or
reported and modeled
estimates/predictions for
the same exposure scenario
• Sensitivity of the exposure
estimates has been
described and quantified
incorporating assumptions,
limitations, and
uncertainties
differences in
methodology, collection,
or context make it difficult
to arrive at full
concordance
• There is some, but not
complete, documentation
or description of
assumptions, limitations
and uncertainties
modeled exposure
estimates even when
uncertainty and
variability are accounted
for.
• Assumptions and
uncertainties are not
known or documented
6166
Page 218 of 484
-------
6167
6168
6169
6170
6171
6172
6173
6174
6175
6176
6177
6178
6179
6180
6181
6182
6183
6184
6185
6186
6187
6188
6189
6190
6191
6192
6193
6194
6195
6196
6197
6198
6199
6200
6201
6202
6203
6204
6205
6206
6207
6208
6209
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
C.5.1 Environmental Release and Occupational Exposure
EPA evaluated environmental releases based on reported release data, modeling approaches, and
industry sector information from standard engineering sources such as TRI and DMR. As described in
Appendix E, EPA estimated COU-specific releases where supporting data existed and documented
uncertainties where an absence of such data required a broader application of release estimates.
EPA evaluated occupational exposures based on monitoring data, modeling approaches, and worker
activity information from standard engineering sources and systematic review as described in Appendix
F. EPA used COU-specific assessment approaches where supporting data existed and documented
uncertainties where supporting data were only applicable for broader assessment approaches.
C.5.2 General Population
General population exposures were evaluated for each exposure pathway based on environmental
release data identified as described above in Section C.4.1, environmental monitoring data identified
through available databases or as described in Section C.4.2, and any other relevant information
identified through systematic review. As described in Section 1, all physical and chemical and fate
properties evaluated in the 2020 RE were used to evaluate the in-scope exposure pathways of the
supplement.
C.5.2.1 General Population: Surface Water
To evaluate the surface water pathway, EPA relied on modeled surface water concentrations based on
environmental release data reported to TRI and DMR (Appendix E.3.1) and releases modeled for other
release types, including DTD and hydraulic fracturing (Appendix E.3.2).
EPA identified ambient surface water monitoring data through the WQP, drinking water monitoring
from PWSs through the UCMR3 database and three state-specific databases (Section 2.3.1.1). EPA used
available surface water monitoring data to confirm the accuracy of model predictions in location-
specific case-studies (Appendix G.2.3.2). In addition, available drinking water monitoring data (see 1,4-
Dioxane Supplement to the Risk Evaluation Data Extraction Information for General Population,
Consumer, and Environmental Exposure) were used to provide context and a point of reference for
modeled drinking water concentrations and risk estimates (Section 5.2.2.1.5) ( 23h).
C.5.2.2 General Population: Groundwater
To evaluate the land pathway (groundwater) releases, EPA relied on environmental release data reported
to TRI (Section 2.2.1.1 and Appendix E.4.1) and releases modeled for hydraulic fracturing operations
(Appendix E.5.2).
EPA identified groundwater monitoring data for 1,4-dioxane through the WQP as presented in Section
2.3.2.1 and described in Appendix H.l. Furthermore, EPA contextualized potential groundwater
concentrations identified in the literature through systematic review (see 1,4-Dioxane Supplement to the
Risk Evaluation Data Extraction Information for General Population, Consumer, and Environmental
Exposure) using search terms identified in Appendix C.2 ( IK).
C.5.2.3 General Population Exposure: Ambient Air
EPA did not identify quantitative outdoor air monitoring data for 1,4-dioxane through systematic
review. To evaluate the air pathway, EPA relied on modeled air concentrations based on industrial
releases reported to TRI (Section 2.3.3.2.2 and Appendix E.5.1), releases modeled for laundry facilities
Page 219 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
6210 (Section 2.3.3.2.4 and Appendix E.5.2), and releases modeled for hydraulic fracturing operations
6211 (Section 2.3.3.2.4 and Appendix E.5.2).
Page 220 of 484
-------
6212
6213
6214
6215
6216
6217
6218
6219
6220
6221
6222
6223
6224
6225
6226
6227
6228
6229
6230
6231
6232
6233
6234
6235
6236
6237
6238
6239
6240
6241
6242
6243
6244
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Appendix D COU-OES MAPPING AND CROSSWALK
This appendix contains additional information about the relationship between the COUs and OESs
determined for 1,4-dioxane.
Condition of Use (CPU): TSCA § 3(4) defines COUs 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'. COUs included in the scope of
EPA's risk evaluations are typically tabulated in scope documents and risk evaluation documents as
summaries of life cycle stages, categories, and subcategories of use. Therefore, a COU is composed of a
combination of life cycle stage, category, and subcategory. COU development may include Chemical
Data Reporting (CDR) information, market profile information, and literature sources. Early in the risk
evaluation process, EPA maps each COU to an occupational exposure scenario for the environmental
release and occupational exposure assessment.
Occupational Exposure Scenario (PES): This term is intended to describe the grouping or segmenting
of COUs for assessment of releases and exposures. For example, EPA may assess a group of multiple
COUs together as one OES due to similarities in release and exposure sources, worker activities, and use
patterns. Alternatively, EPA may assess multiple OES for one COU because there are different release
and exposure potentials for a given COU. OES determinations are also largely driven by the availability
of data and modeling approaches to assess occupational releases and exposures. For example, even if
there are similarities between multiple COUs, if there is sufficient data to separately assess releases and
exposures for each COU, EPA would not group them into the same OES.
D.l COU-OES Mapping
The details of an identified COU will determine the number of associated OES(s). Mapping OES to
COUs may come in many forms, including a direct one-to-one mapping of a single OES to a single
COU, mapping of one OES to multiple COUs, or mapping of multiple OES to a single COU, as shown
in Figure Apx D-l. The OES mapping is driven by similarities and differences in the expected
occupational exposures and releases for a COU and the reasonably available data to estimates such
exposures and releases, as discussed in Section 2.1.1. Further, there may be differences in the name of
an OES from the name of the COU to which it is mapped. This is because OES names are intended to be
succinct, capture all COUs where one OES is mapped to multiple COUs, and distinct enough to
represent the specific occupational exposure and release scenario.
Page 221 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
COU1
OES1
One COU may map to one OES
There may be differences in the COU and OES names because the OES name is intended to be
succinct and specific to the assessed occupational releases and exposures
For example, the 1,4-dioxane COU for "Byproduct produced during the production of
polyethylene terephthalate" maps only to the OES named "PET byproduct" (see excerpt from
crosswalk Table 2-1 andApxD-1 below)
6245
Condition of Use
OES
Life Cycle
Stage
Category
Subcategory
Processing
Byproduct
Byproduct produced during
production of PET
PET Byproduct
Multiple COUs may be mapped to the same OES
Multiple COUs may be mapped to one OES when the COUs have similar activities and exposure
potentials, and exposures and releases can be assessed for the COUs using a single approach
There may be differences between the COU and OES names because the OES is name is intended to
be succinct and encompass all COUs grouped therein under a general name
For example, the 1,4-dioxane COUs for "Industrial wastewater treatment", "underground injection",
"municipal landfill", and others were assessed together under the OES named "disposal" (see excerpt
from crosswalk Table 2-1 and Apx D-l below)
COU liCOU2lCOU3
OES1
6246
Condition of Use
Life Cycle
Stage
Category
Subcategory
OES
Industrial pre-treatment
Industrial wastewater treatment
Publicly owned treatment
works (POTW)
Underground injection
Disposal
Disposal
Municipal landfill
Disposal
Hazardous landfill
Other land disposal
Municipal waste incinerator
Hazardous waste incinerator
Off-site waste transfer
COU1
OESliOES2iOES3
One COU may be mapped to multiple OES
Mapping a COU to multiple OES allows for the assessment of distinct scenarios that are not
expected to result in similar releases and exposures
There may be differences between the COU and OES names because the OES capture more distinct
scenarios of occupational release and exposure than the COU
For example, the 1,4-dioxane COU for "dish soap, dishwasher detergent, laundry detergent" (which
is a single COU) was assessed as separate OES named "dish soap", "dishwasher detergent",
"laundry detergent (industrial)", and "laundry detergent (institutional)" (see excerpt from crosswalk
Table 2-1 and Apx D-l below)
Condition of Use
OES
Life Cycle
Stage
Category
Subcategory
Consumer use,
commercial
use
Laundry and
Dishwashing
Products
Dish soap
Dishwasher detergent
Laundry detergent
Dish Soap
Dishwasher Detergent
Laundry Detergent
(Industrial)
Laundry Detergent
(Institutional)
6248 Figure Apx D-1. COU and OES Mapping
6249
Page 222 of 484
-------
6250
6251
6252
6253
6254
6255
6256
6257
6258
6259
6260
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
D.2 CQU-QES Crosswalk
A crosswalk of the COU with the OES assessed is provided in TableApx D-l. As discussed in Section
2.1.1, a COU is a combination of life cycle stage, category, and subcategory and EPA mapped each
COU to an OES. The purpose of an OES is to group, where appropriate, COUs based on similarity of
the operations and data availability for each COU. EPA assessed environmental releases (air, water, and
land) and occupational exposures (inhalation and dermal) to 1,4-dioxane for each of the OES listed in
TableApx D-l. As noted in this table, some of these OESs were in scope of the Final Risk Evaluation
for 1,4-Dioxane ( 320c) while others were in scope of this supplemental risk evaluation.
Table Apx D-l. Categories and Subcategories of Conditions of Use Included in the Scope of the
Risk Evaluation
Condition of Use
OES
Risk Evaluation in Which
Occupational Exposures
Were Assessed
Life Cycle
Stage
Category"
Subcategory''
Manufacturing
Domestic
Manufacture
Domestic Manufacture
Manufacturing
2020 RE
Import
Import
Repackaging
Import and Repackaging
2020 RE
Processing
Processing as a
Reactant
Polymerization catalyst
Industrial Uses
2020 RE
Non-
incorporative
Basic organic chemical
manufacturing (process
solvent)
Byproduct
Byproduct produced
during processes
Ethoxylation Process
Byproduct
Supplemental RE
Byproduct produced
during production of PET
PET Byproduct
Supplemental RE
Recycling
Recycling
Disposal
2020 RE
Distribution in
commerce
Distribution
Distribution
Distribution activities
(e.g., loading) considered
throughout life cycle,
rather than using a single
distribution scenario
N/A
Industrial use
Intermediate use
Plasticizer intermediate
Catalysts and reagents for
anhydrous acid reactions,
brominations, and
sulfonations
Industrial Uses
2020 RE
Processing aids,
not otherwise
listed
Wood pulpingc
Extraction of animal and
vegetable oilsc
Wetting and dispersing
agent in textile
processing17
Polymerization catalyst
Purification of process
intermediates
Etching of fluoropolymers
Page 223 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Condition of Use
OES
Risk Evaluation in Which
Occupational Exposures
Were Assessed
Life Cycle
Stage
Category"
Subcategory''
Functional
fluids (open and
closed system)
Polyalkylene glycol
lubricant
Functional Fluids (Open-
System)
2020 RE
Synthetic metalworking
fluid
Cutting and tapping fluid
Hydraulic fluid
Functional Fluids
(Closed-System)
2020 RE
Industrial use,
commercial
use
Laboratory
Chemicals
Chemical reagent
Laboratory Chemicals
2020 RE
Reference material
Spectroscopic and
photometric measurement
Liquid scintillation
counting medium
Stable reaction medium
Cryoscopic solvent for
molecular mass
determinations
Preparation of histological
sections for microscopic
examination
Adhesives and
Sealants
Film cement
Film Cement
2020 RE
Other Uses
Spray polyurethane foam;
Printing and printing
compositions, including
3D printing; dry film
lubricant; Hydraulic
fracturing
Spray Foam Application
2020 RE
Printing Inks (3D)
2020 RE
Dry Film Lubricant
2020 RE
Hydraulic Fracturing
Supplemental RE
Consumer
use,
commercial
use
Paints and
Coatings
Latex wall paint or floor
lacquer
Paint and Floor Lacquer
Supplemental RE
Cleaning and
Furniture Care
Products
Surface cleaner
Surface Cleaner
Supplemental RE
Laundry and
Dishwashing
Products
Dish soap
Dishwasher detergent
Laundry detergent
Dish Soap
Dishwasher Detergent
Laundry Detergent
(Industrial)
Laundry Detergent
(Institutional)
Supplemental RE
Arts, Crafts, and
Hobby
Materials
Textile dye
Textile Dye
Supplemental RE
Automotive
Care Products
Antifreeze
Antifreeze
Supplemental RE
Other Consumer
Uses
Spray polyurethane foam
Spray Foam Application
2020 RE
Disposal
Disposal
Industrial pre-treatment
Disposal
2020 RE
Page 224 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Condition of Use
OES
Risk Evaluation in Which
Occupational Exposures
Were Assessed
Life Cycle
Stage
Category"
Subcategory''
Industrial wastewater
treatment
Publicly owned treatment
works (POTW)
Underground injection
Municipal landfill
Hazardous landfill
Other land disposal
Municipal waste
incinerator
Hazardous waste
incinerator
Off-site waste transfer
" These categories of conditions of use reflect Chemical Data Reporting (CDR) rule codes and broadly represent conditions
of use for 1,4-dioxane in industrial and/or commercial settings.
b These subcategories reflect more specific uses of 1,4-dioxane.
6261
Page 225 of 484
-------
6262
6263
6264
6265
6266
6267
6268
6269
6270
6271
6272
6273
6274
6275
6276
6277
6278
6279
6280
6281
6282
6283
6284
6285
6286
6287
6288
6289
6290
6291
6292
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Appendix E INDUSTRIAL AND COMMERCIAL
ENVIRONMENTAL RELEASES
This appendix contains additional information relevant to the assessment of industrial and commercial
environmental releases.
E.l Estimates of the Number of Industrial and Commercial Facilities with
Environmental Releases
As a part of the assessment of industrial and commercial environmental releases, EPA estimated the
number of facilities with releases for each OES. Where available, EPA used 2013 to 2019 TRI (
12e) and 2013-2019 DMR ( :022c) data to provide a basis to estimate the number of
sites using 1,4-dioxane within an OES. Additional information on how EPA utilized TRI and DMR to
estimate the number of sites using 1,4-dioxane within a COU can be found in Section 2.2.1.2.2 of the
December 2020 Final Risk Evaluation for 1,4-Dioxane ( 20c).
Where the number of sites could not be determined using TRI or DMR or where these data were
determined to not capture the entirety of sites within an OES, EPA supplemented the available data with
U.S. economic data using the following methods:
• Identify the North American Industry Classification System (NAICS) codes for the industry
sectors associated with these uses.
• Estimate total number of sites using the U.S. Census" Statistics of US Businesses (SUSB) (U.S.
Census Bureau. ) data on total establishments by 6-digit NAICS.
• Review available ESDs and GSs for established facility estimates for each occupational exposure
scenario.
• Combine the data generated in bullets 1 through 3 to produce an estimate of the number of sites
using 1,4-dioxane in each 6-digit NAICS code and sum across all applicable NAICS codes for
the COU, augmenting as needed with data from the ESDs and GSs, to arrive at a total estimate of
the number of sites within the COU.
A summary of the number of facilities EPA determined for each OES and each type of release is shown
in Table Apx E-l. The number of facilities may be different for each type of release within the same
OES if sufficient data were available to make this differentiation.
Table Apx E-l. Summary of EPA's Estimates for the Number of Facilities for Each OES
OES
Type of Release
Number of
Facilities
Notes
Manufacturing
Air, Land
1
Based on 2019 TRI reDortina (U.S. EPA. 2022s).
Surface Water
1
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA. 2022c. e).
POTW or Non-
POTWWWT
1
Based on 2013-2019 TRI reDortina (U.S. EPA. 2022s).
Import and
Repackaging
Air, Land
1
Based on 2019 TRI reDortina (U.S. EPA. 2022s).
Surface Water
6
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA. 2022c. s).
POTW or Non-
POTWWWT
6
Based on 2013-2019 TRI reDortina (U.S. EPA. 2022s).
Page 226 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Release
Number of
Facilities
Notes
Industrial Uses
Air, Land
12
Based on 2019 TRI reporting (U.S. EPA. 2022s).
Surface Water
24
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA. 2022c. a).
POTW or Non-
POTWWWT
17
Based on 2013-2019 TRI reoortina (U.S. EPA. 2022s).
Functional
Fluids (Open-
System)
Air, Land
2
Based on 2019 TRI reDortina (U.S. EPA. 2022s).
Surface Water
6
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA. 2022c. a).
POTW or Non-
POTWWWT
1
Based on 2013-2019 TRI reDortina (U.S. EPA. 2022a).
Functional
Fluids (Closed-
System)
All
N/A
Assessed as a part of Industrial Uses OES.
Laboratory
Chemical
All
132
Calculated using the GS on Use of Laboratory
Chemicals (U.S. EPA. 2022h) and the amount of 1.4-
dioxane used in laboratory uses per the December 2020
Final Risk Evaluation for 1,4-Dioxane (
2020c).
Film Cement
All
211
Based on the number of sites for this OES in the
December 2020 Final Risk Evaluation for 1,4-Dioxane
(U.S. EPA, 2020c). which is a bounding estimate based
on U.S. Census Bureau data forNAICS code 512199,
Other Motion Picture and Video Industries.
Spray Foam
Application
All
1,553,559
Based on the number of sites for this OES in the
December 2020 Final Risk Evaluation for 1,4-Dioxane
(U.S. EPA. 2020c). which is a bounding estimate based
on U.S. Census Bureau data forNAICS code 238310,
Dry wall and Insulation Contractors.
Printing Inks
(3D)
Air, Land
N/A
Assessed as a part of Industrial Uses OES.
Surface Water,
POTW, non-POTW
WWT
1
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA. 2022c. a).
Dry Film
Lubricant
All
8
Based on the number of sites for this OES in the
December 2020 Final Risk Evaluation for 1,4-Dioxane
(U.S. EPA, 2020c). which is based on conversations
with the Kansas City National Security Campus
(manufacturer and uses of dry film lubricants).
Disposal
Air
15
Based on 2019 TRI reporting (U.S. EPA, 2022s).
Surface Water
24
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA, 2022c. e).
POTW or Non-
POTW WWT,
Land
4
Based on 2013-2019 TRI reDortina (U.S. EPA. 2022s).
Page 227 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Release
Number of
Facilities
Notes
Textile Dye
All
783
Bounding estimate based on U.S. Census Bureau data
for NAICS code 313310, Textiles and Fabric Finishing
Mills.
Antifreeze
All
84,383
Bounding estimate based on U.S. Census Bureau data
for NAICS codes 811111, General Automotive Repair,
and 811198, All Other Automotive Repair and
Maintenance.
Surface Cleaner
All
Unknown
within
Liverpool OH;
55,998
(industry
bounding
estimate)
Land release estimates for this OES were developed for
the Liverpool, OH case study and the number of sites
within this locality is unknown (e.g., the release
estimates are not per site but for the entire locality).
Bounding estimate for the industry is based on U.S.
Census Bureau data for NAICS code 561720, Janitorial
Services.
Dish Soap
All
Unknown
within
Liverpool OH;
773,851
(industry
bounding
estimate)
Land release estimates for this OES were developed for
the Liverpool, OH, case study and the number of sites
within this locality is unknown (e.g., the release
estimates are not per site but for the entire locality).
Bounding estimate for the industry is based on U.S.
Census Bureau data for NAICS codes 623300, 713900,
721100, 721300, 722300, 722400, and 722500.
Dishwasher
Detergent
All
Unknown
within
Liverpool OH;
773,851
(industry
bounding
estimate)
Land release estimates for this OES were developed for
the Liverpool, OH, case study and the number of sites
within this locality is unknown (e.g., the release
estimates are not per site but for the entire locality).
Bounding estimate for the industry is based on U.S.
Census Bureau data for NAICS codes 623300, 713900,
721100, 721300, 722300, 722400, and 722500.
Laundry
Detergent
(Institutional)
All
95,533
Bounding estimate based on industry information as
described in the ESD on Water Based Washing
operations at Industrial and Institutional Laundries
(OECD.: ).
Laundry
Detergent
(Industrial)
All
2,453
Bounding estimate based on U.S. Census Bureau data
for NAICS code 812330, Linen and Uniform Supply.
Paints and Floor
Lacquer
All
33,648
Bounding estimate based on U.S. Census Bureau data
for NAICS code 811121, Automotive Body, Paint, and
Interior Repair and Maintenance.
Polyethylene
Terephthalate
(PET)
Byproduct
Air, Land
13
Based on 2019 TRI reporting (IIS. EPA. 2022s).
Surface Water
19
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA. 2022c. a).
POTW or Non-
POTWWWT
14
Based on 2013-2019 TRI reDortina (U.S. EPA. 2022s).
Ethoxylation
Process
Byproduct
Air, Land
8
Based on 2019 TRI reDortine (U.S. EPA. 2022s).
Surface Water
7
Based on 2013-2019 DMR and TRI reporting (U.S.
EPA. 2022c. s).
Page 228 of 484
-------
6293
6294
6295
6296
6297
6298
6299
6300
6301
6302
6303
6304
6305
6306
6307
6308
6309
6310
6311
6312
6313
6314
6315
6316
6317
6318
6319
6320
6321
6322
6323
6324
6325
6326
6327
6328
6329
6330
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Release
Number of
Facilities
Notes
POTW orNon-
POTWWWT
6
Based on 2013-2019 TRI reDortine (TJ.S. EPA. 2022a).
Hydraulic
Fracturing
All
411
Based on the number of sites that reported using 1,4-
dioxane to FracFocus 3.0 (GWPC and IOGCC. 2022).
E.2 Estimates of Number of Release Days for Industrial and Commercial
Releases
As a part of the assessment of industrial and commercial environmental releases, EPA also estimated the
number of release days for each OES. EPA referenced the December 2020 Final Risk Evaluation for
1,4-Dioxane (U.S. EPA. 2020c). Generic Scenarios (GS), Emission Scenario Documents (ESDs), or
made assumptions when estimating release days for each OES. In summary, EPA estimated the number
of operating days using the below sources of data:
1. Facility-Specific Data: Use facility-specific data if available. If facility-specific data is not
available, estimate the days/year using one of the following approaches:
a. If facilities have known or estimated average daily use rates, calculate the days/year as:
Days/year = Estimated Annual Use Rate for the Site (kg/year) / average daily use rate
from sites with available data (kg/day).
b. If sites with days/year data do not have known or estimate average daily use rates, use the
average number of days/year from the sites with such data.
2. Industry-Specific Data: Industry-specific data may be available in the form of GSs, ESDs, trade
publications, or other relevant literature. In such cases, these estimates should take precedent
over other approaches, unless facility-specific data are available.
3. Manufacture of Lower-PV Specialty Chemicals: For the manufacture of lower-PV specialty
chemicals like 1,4-dioxane, the chemical is not expected to be manufactured continuously
throughout the year. Therefore, a value of 250 days/year should be used. This assumes the plant
manufactures the chemical 5 days/week and 50 weeks/year (with 2 weeks down for turnaround).
For the manufacture of 1,4-dioxane as a byproduct (e.g., ethoxylation process, PET
manufacturing), 250 days/year is also used, assuming these industrial manufacturing facilities
have a similar operating schedule of 5 days/week and 50 weeks/year.
4. Processing as Reactant (Intermediate Use) in the Manufacture of Specialty Chemicals:
Similar to #3, the manufacture of specialty chemicals is not expected to occur continuously
throughout the year. Therefore, a value of 250 days/year can be used.
5. Other Chemical Plant OES (e.2.. Industrial Uses): For these OESs, it is reasonable to assume
that 1,4-dioxane is not always in use at the facility, even if the facility operates 24/7. Therefore,
in general, a value of 300 days/year can be used based on the "SpERC [Specific Environmental
Release Categories] fact sheet - Formulation & (re)packing of substances and mixtures -
Industrial (Solvent-borne)" which uses a default of 300 days/year for the chemical industry.
However, in instances where the OES uses a low volume of the chemical of interest, 250
days/year can be used as a lower estimate for the days/year.
6. POTWs: Although POTWs are expected to operate continuously over 365 days/year, the
discharge frequency of 1,4-dioxane from a POTW will be dependent on the discharge patterns of
the chemical from the upstream facilities discharging to the POTW. However, there can be
multiple upstream facilities (possibly with different OES) discharging to the same POTW and
Page 229 of 484
-------
6331
6332
6333
6334
6335
6336
6337
6338
6339
6340
6341
6342
6343
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
information to determine when the discharges from each facility occur on the same day or
separate days is typically not available. Therefore, an exact number of days/year the 1,4-dioxane
is discharged from the POTW cannot be determined and a value of 365 days/year should be
used.
7. All Other OESs: Regardless of the facility operating schedule, other OESs are unlikely to use
1,4-dioxane every day. Therefore, a value of 250 days/year should be used for these OESs.
A summary along with a brief explanation is presented in TableApx E-2 below. These estimates of
release days are applicable to the air and water release estimates for each OES; however, there is a high
level of variability and uncertainty associated with the number of days of release associated with land
releases. Therefore, EPA could not estimate the number of days of release for land releases.
Table Apx E-2. Summary of EPA's Estimates for Air and Water Release Days Expected for Each
OES
OES
Release
Days
Notes
Manufacturing
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities, consistent with the 2020 RE (U.S. EPA. 2020c).
Import and
Repackaging
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities, consistent with the December 2020 Final Risk Evaluation
for 1.4-Dioxane ("U.S. EPA. 2020c).
Industrial Uses
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities, consistent with the December 2020 Final Risk Evaluation
for 1.4-Dioxane (U.S. EPA. 2020c).
Functional
Fluids (Open-
System)
247
Per the 2011 OECD Emission Scenario Document on the Use of Metalworking
Fluids, consistent with the December 2020 Final Risk Evaluation for 1,4-
Dioxane (U.S. EPA. 2020c).
Functional
Fluids (Closed-
System)
N/A
Assessed as a part of Industrial Uses OES.
Laboratory
Chemical
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities, consistent with the December 2020 Final Risk Evaluation
for 1.4-Dioxane (U.S. EPA. 2020c).
Film Cement
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities, consistent with the December 2020 Final Risk Evaluation
for 1.4-Dioxane (U.S. EPA. 2020c).
Spray Foam
Application
3
Per the 2018 EPA generic scenario Application of Spray Polyurethane Foam
Insulation, consistent with the December 2020 Final Risk Evaluation for 1,4-
Dioxane (U.S. EPA. 2020c).
Printing Inks
(3D)
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities, consistent with the December 2020 Final Risk Evaluation
for 1.4-Dioxane (U.S. EPA. 2020c).
Dry Film
Lubricant
48
Per process description information provided in the December 2020 Final Risk
Evaluation for 1.4-Dioxane (U.S. EPA, 2020c).
Page 230 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Release
Days
Notes
Disposal
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities, consistent with the December 2020 Final Risk Evaluation
for 1.4-Dioxane (TJ.S. EPA. 2020c).
Textile Dye
31 to 295
Based on the 2015 OECD on Textile Dves (OECD. 2017) and Monte Carlo
Modeling.
Antifreeze
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities.
Surface Cleaner
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities.
Dish Soap
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities.
Dishwasher
Detergent
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities.
Laundry
Detergent
(Institutional)
250 to
365
Based on the 2011 OECD ESD on Industrial and Institutional Laundries
( M CD. JO I 1and Monte Carlo Modeling.
Laundry
Detergent
(Industrial)
20 to 365
Based on the 2011 OECD ESD on Industrial and Institutional Laundries
(¦ M CD, .'01 11») and Monte Carlo Modeling.
Paints and Floor
Lacquer
250
Based on the 2011 OECD ESD on Coating Application via Spray Painting in
the Automotive Refinishine Industry (OECD, 201 la).
Polyethylene
Terephthalate
(PET)
Byproduct
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities.
Ethoxylation
Process
Byproduct
250
Assumed 5 days per week and 50 weeks per year with 2 weeks per year for
shutdown activities.
Hydraulic
Fracturing
1 to 72
Based on the reported number of days for sites that use 1,4-dioxane in
FracFocus 3.0 (GWPC and IOGCC. 2022). This ranae of release davs refers to
only the hydraulic fracturing and not post-fracturing production stages. EPA's
estimates for flowback and produced water releases during production stages
occur over 350 davs/vear ("U.S. EPA. 2022d).
6344 E.3 Water Release Assessment
6345 This section describes EPA's methodology for estimating daily wastewater discharges from industrial
6346 and commercial facilities manufacturing, processing, or using 1,4-dioxane. Facilities report wastewater
6347 discharges either via Discharge Monitoring Reports (DMRs) under the NPDES or TRI. EPA used 2013
6348 to 2019 DMR ( 22c) and 2013 to 2019 TRI ( 2g) data to estimate daily
6349 wastewater discharges for the OES where available; however, EPA did not have these data for every
6350 OES. For OES without DMR and TRI data, EPA used alternate assessment approaches to estimate
6351 wastewater discharges. Both approaches, that for OESs with DMR and TRI data and that for OESs
6352 without these data, are described below.
Page 231 of 484
-------
6353
6354
6355
6356
6357
6358
6359
6360
6361
6362
6363
6364
6365
6366
6367
6368
6369
6370
6371
6372
6373
6374
6375
6376
6377
6378
6379
6380
6381
6382
6383
6384
6385
6386
6387
6388
6389
6390
6391
6392
6393
6394
6395
6396
6397
6398
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.3.1 Assessment Using TRI and DMR
EPA found 2013 to 2019 DMR and/or 2013 to 2019 TRI data for facilities within the following OESs:
• Manufacturing,
• Import and repackaging,
• Industrial uses,
• Functional fluids (open-system),
• 3D printing,
• Disposal,
• PET byproduct, and
• Ethoxylation byproduct.
The 2013 to 2019 TRI data were rated "medium" in EPA's systematic review process and the 2013 to
2019 DMR were rated "medium." EPA estimated daily discharges using TRI and DMR data for these
OESs, with the following general stages as described in the rest of this section:
1. Collect wastewater discharge data from 2013 to 2019 DMR and TRI data,
2. Map wastewater discharge data to occupational exposure scenarios,
3. Estimate the number of facility operating days per year, and
4. Estimate daily wastewater discharges and summarize wastewater discharges for each OES.
Note that EPA compared the TRI and DMR data used to estimate water releases for the PET byproduct
OES in this risk evaluation to information from a life cycle analysis on the PET manufacturing process
in Appendix E.6.
Step 1: Collect Wastewater Discharge Data from DMR and TRI
The first step in estimating daily releases was to obtain 2013 through 2019 DMR and TRI data. Under
the CWA, EPA regulates the discharge of pollutants into receiving waters through NPDES. A NPDES
permit authorizes discharging facilities to discharge pollutants to specified limits. NPDES permits apply
pollutant discharge limits to each outfall at a facility. For risk evaluation purpose, EPA is interested only
in the outfalls to surface water body. NPDES permits also include internal outfalls, but they aren't
included in this analysis. This is because these outfalls are internal monitoring points within the facility
wastewater collection or treatment system, so they do not represent discharges from the facility. The
permits require facilities to monitor their discharges and report the results to EPA and the state
regulatory agency. Facilities report these results in DMRs. EPA makes these reported data publicly
available via EPA's ECHO system and EPA's Water Pollutant Loading Tool (Loading Tool). The
Loading Tool is a web-based tool that obtains DMR data through ECHO, presents data summaries and
calculates pollutant loading (mass of pollutant discharged). EPA queried the ECHO Loading Tool to
pull data for each of years 2013 through 2019. EPA removed facilities reporting zero discharges for 1,4-
dioxane in DMR from the analysis because EPA cannot confirm if the pollutant is present at the facility.
Each facility subject to the TRI reporting rule must report annually the volume of chemical released to
the environment and/or managed through recycling, energy recovery, and treatment. Unlike DMR, TRI
includes both reports of annual direct discharges to surface water and annual indirect discharges to off-
site publicly owned treatment works (POTW) and wastewater treatment (WWT) facilities (non-POTW
WWT). Similar to the air release assessment, EPA included both TRI reporting Form R and TRI
reporting Form A submissions in the water release assessment. Where sites reported to TRI with Form
A, EPA used the Form A threshold for total releases of 500 lb/year. EPA used the entire 500 lb/year for
both direct and indirect wastewater discharges; however, since this threshold is for total site releases,
these 500 lb/year are attributed either to direct discharges or indirect discharges for this analysis, not
Page 232 of 484
-------
6399
6400
6401
6402
6403
6404
6405
6406
6407
6408
6409
6410
6411
6412
6413
6414
6415
6416
6417
6418
6419
6420
6421
6422
6423
6424
6425
6426
6427
6428
6429
6430
6431
6432
6433
6434
6435
6436
6437
6438
6439
6440
6441
6442
6443
6444
6445
6446
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
both (since that would double count the releases and exceed the total release threshold for Form A). EPA
pulled the TRI Basic Plus Data Files for each of years 2013 through 2019.
In summary, wastewater discharges reported to DMR and TRI include the following:
• DMR:
o On-site releases to surface water (direct discharges).
• TRI:
o On-site releases to surface water (direct discharges),
o Off-site transfers to POTWs (indirect discharges), and
o Off-site transfers to non-POTW WWT (indirect discharges).
Note that the two datasets are not updated concurrently. The Loading Tool automatically and
continuously checks ICIS-NPDES for newly submitted DMRs. The Loading Tool processes the data
weekly and calculates pollutant loading estimates; therefore, water discharge data (DMR data) are
available on a continual basis. Although the Loading Tool process data weekly, each permitted
discharging facility is only required to report their monitoring results for each pollutant at a frequency
specified in the permit (e.g., monthly, every two months, quarterly). TRI data is reported annually for
the previous calendar year and is typically released in October (i.e., 2020 TRI data is released in October
2021).
Step 2: Map Wastewater Discharge Data to Occupational Exposure Scenarios
The next step in estimating daily releases was to map 2013 through 2019 DMR and TRI data to the 1,4-
dioxane OES. EPA used the same mapping methodology for the water assessment as that described in
Appendix E.5.1. EPA ensured consistency in the OES mapping for sites that reported to both TRI and
DMR. EPA also ensured consistency in the OES mapping between the air, water, and land assessments.
Step 3: Estimate the Number of Facility Operating Days per Year
EPA then estimated the number of operating days (days/year) for each facility reporting wastewater
discharges to DMR and TRI. EPA generally used the same number of operating days for the same OES
for both the air and water analysis, which is based on the general methodology described previously in
Appendix E.2.
Step 4: Estimate Daily Wastewater Discharges and Summarize Wastewater Discharges for each OES
After the initial steps of selecting and mapping of the water discharge data and estimating the number of
facility operating days/year were completed, the next step was to summarize annual and daily
wastewater discharges for each OES. EPA summarized annual wastewater discharges reported in DMR
and TRI for each facility. EPA estimated daily wastewater discharges separately for direct and indirect
discharges, as discussed below.
EPA estimated the median and maximum daily direct wastewater discharges at each facility, using the
steps below. EPA presented the calculated median and maximum daily direct wastewater discharged
separately for the DMR and TRI datasets because these data do not always agree/match.
1. Obtained total annual loads calculated from the Loading Tool and reported annual surface water
discharges in TRI for years 2013 through 2019.
2. Divided the annual direct discharge over the number of estimated operating days for the OES to
which the facility has been mapped. The number of operating days differ for each OES, as
summarized in Appendix E.2.
3. Calculated the median daily direct wastewater discharge across all years of data for each facility,
separately for both DMR and TRI data.
Page 233 of 484
-------
6447
6448
6449
6450
6451
6452
6453
6454
6455
6456
6457
6458
6459
6460
6461
6462
6463
6464
6465
6466
6467
6468
6469
6470
6471
6472
6473
6474
6475
6476
6477
6478
6479
6480
6481
6482
6483
6484
6485
6486
6487
6488
6489
6490
6491
6492
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4. Identified the maximum daily direct wastewater discharge across all years of data for each
facility. EPA also noted which reporting year had this maximum daily direct wastewater
discharge, separately for both DMR and TRI data.
For indirect discharges to POTW or non-POTW WWT, EPA estimated the average daily indirect
discharges for each facility and each reporting year (2013 through 2019) in TRI using steps #1 and #2
above. DMR data do not include indirect discharges. EPA did not estimate the median or maximum
daily indirect discharges across all years.
A summary of the estimated daily discharges using 2013 to 2019 DMR and TRI is included in 1,4-
Dioxane Supplemental Information File: Environmental Releases to Water for OES with TRI and DMR
(U.S. EPA. 2023ml
E.3.2 Assessment for OES without TRI and DMR
EPA did not find DMR or TRI data for any of the years included in this analysis for the following OESs:
• Functional Fluids (Closed-Systems)
• Laboratory Chemicals
• Film Cement
• Spray Polyurethane Foam
• Dry Film Lubricant
• Textile Dye
• Antifreeze
• Surface Cleaner
• Dish Soap
• Dishwasher Detergent
• Laundry Detergent,
• Paints and Floor Lacquer
• Hydraulic Fracturing
For these OESs, EPA estimated daily wastewater discharges by using various modeling approaches—
including the use of surrogate TRI and DMR data and modeling using data from literature, GSs, and
ESDs. EPA's assessment of daily wastewater discharges for each of these OESs is described below.
Functional Fluids (Closed-Systems)
Wastewater discharge data were not available for this OES and EPA did not find any information to
model wastewater discharges for this OES using literature, GSs, or ESDs. EPA expects that the sources
of release for this OES to be similar to those for the Industrial Uses OES, based on the process
information in the Final Risk Evaluation for 1,4-Dioxane ( 2020c). Therefore, EPA grouped
the water release assessment for Functional Fluids (Closed-Systems) into that for Industrial Uses.
However, there is uncertainty in this assumption of similar release sources between these OESs.
Laboratory Chemicals
EPA estimated daily wastewater discharges for facilities within the laboratory chemicals OESs using the
Draft GS on Use of Laboratory Chemicals (U.S. EPA. 2022h). The GS on Use of Laboratory Chemicals
was rated "high" during EPA's systematic review process.
Per the GS on Use of Laboratory Chemicals, water releases are not expected for hazardous chemicals.
Because 1,4-dioxane is considered a hazardous substance under CERCLA (40 CFR Part 302.4) and the
PubChem Hazardous Substances Data Bank (HSDB), there are no water releases for this OES. This is
Page 234 of 484
-------
6493
6494
6495
6496
6497
6498
6499
6500
6501
6502
6503
6504
6505
6506
6507
6508
6509
6510
6511
6512
6513
6514
6515
6516
6517
6518
6519
6520
6521
6522
6523
6524
6525
6526
6527
6528
6529
6530
6531
6532
6533
6534
6535
6536
6537
6538
6539
6540
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
consistent with the water release assessment for this OES in the Final Risk Evaluation for 1,4-Dioxane
( 1020c). which indicates that water releases are not expected for laboratory uses of 1,4-
dioxane.
Film Cement
EPA estimated daily wastewater discharges for facilities within the Film Cement OES using process
information from the Final Risk Evaluation for 1,4-Dioxane (U.S. EPA. 2020c). The underlying process
information for this assessment was rated "high" during EPA's systematic review process.
Per the risk evaluation, EPA does not expect water releases of 1,4-dioxane for this OES. EPA expects
the glue bottles to be disposed of as solid waste without. There is some uncertainty as to whether and
how much 1,4-dioxane may remain in the glue bottles when disposed. However, due to the small
quantities of the glue and high volatility of the 1,4-dioxane, EPA expects any residual 1,4-dioxane to
evaporate to the air or remain in the solid waste stream ( 20c).
Spray Polyurethane Foam
EPA estimated daily wastewater discharges for facilities within the Spray Polyurethane Foam OES
using the same approach described for this OES in Appendix E.5.2, which is the use of the GS on
Application of Spray Polyurethane Foam Insulation ( »20c). The GS on the Application of
Spray Polyurethane Foam Insulation was rated "medium" during EPA's systematic review process.
The GS indicates that there are six release points:
1. Releases to fugitive air for volatile chemicals during unloading of raw materials from transport
containers;
2. Releases to water, incineration, or landfill from cleaning or disposal of transport containers;
3. Releases to fugitive air for volatile chemicals during transport container cleaning;
4. Releases to incineration or landfill from spray polyurethane foam application equipment
cleaning;
5. Releases to fugitive air for volatile chemicals during equipment cleaning; and
6. Releases to landfill of scrap foam from trimming applied foam.
Based on the GS, only release point #2 has the potential for wastewater discharges. To estimate this
release, EPA used the equations specified in the GS ( »20c). Apart from weight fraction in
spray polyurethan foam, EPA did not find any data specific to 1,4-dioxane in this OES. Therefore, the
calculation of releases using this GS are for a "generic site," using the default input parameter values
from the GS. Specifically, EPA used the same input parameter values that were used in the original risk
evaluation for estimates of occupational exposure; see Appendix G of the Final Risk Evaluation for 1,4-
Dioxane (U.S. EPA. 2020c).
Using this methodology, EPA calculated a range of wastewater releases for this OES. For the low-end,
EPA assumed there are no water releases, which is consistent with the GS explanation that containers
may be disposed of without rinsing. For the high-end, EPA assumed the containers may be rinsed /
poured down drains such that the entire release point #2 is to POTW. Direct water discharges are not
likely given the setting (construction/ renovation sites).
EPA's calculation of wastewater discharges for this OES, including all calculation inputs, can be found
in 1,4-Dioxane Supplemental Information File: Environmental Releases to Water for OES without TRI
or DMR data (11.S. EPA. 2023n).
Page 235 of 484
-------
6541
6542
6543
6544
6545
6546
6547
6548
6549
6550
6551
6552
6553
6554
6555
6556
6557
6558
6559
6560
6561
6562
6563
6564
6565
6566
6567
6568
6569
6570
6571
6572
6573
6574
6575
6576
6577
6578
6579
6580
6581
6582
6583
6584
6585
6586
6587
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Dry Film Lubricant
EPA estimated daily wastewater discharges for facilities within the Dry Film Lubricant OES using
process information from the Final Risk Evaluation for 1,4-Dioxane ( 2020c). The underlying
process information for this assessment was rated "high" during EPA's systematic review process.
Per the risk evaluation, EPA does not expect water releases of 1,4-dioxane for this OES. Based on
conversations the with only known user, EPA expects wastes to be drummed and sent to a waste handler
with residual wastes releasing to air or being disposed to landfill. ( 2020c).
Textile Dye
EPA estimated daily wastewater discharges for facilities within the Textile Dye OES using the OECD
ESD on Textile Dyes ( ) and Monte Carlo modeling. The ESD on Textile Dyes was rated
"medium" during EPA's systematic review process. The use of Montel Carlo modeling allows for
variation of calculation input parameters such that a distribution of environmental releases can be
calculated, from which EPA can estimate the 50th and 95th percentile releases. An explanation of this
modeling approach is included in Appendix E. 11.
Antifreeze
EPA did not find any information to model wastewater discharges for this OES using literature, GSs, or
ESDs, nor does EPA expect this OES to be similar to other OES such that surrogate data may be used.
EPA evaluated the potential for releases using the OECD ESD on Chemical Additives used in
Automotive Lubricants (OECD. 2020) and the EPA MRD on Commercial Use of Automotive Detailing
Products (U.S. EPA. 2022b). The ESD and MRD were both rated "high" during EPA's systematic
review process.
For the use of antifreeze, EPA expects releases may occur from volatilizations of 1,4-dioxane, disposal
or cleaning of empty antifreeze containers, and spent antifreeze. Both the ESD and MRD indicate that
containers of automotive maintenance fluids are typically small and are not rinsed, but rather disposed of
as solid waste ( v << \ 2022b; OEl O . '20). Additionally, the ESD on Chemical Additives used in
Automotive Lubricants indicates that spent lubricants are disposed of via incineration, which EPA
expects is similarly done for spent antifreeze (OECD. 2020). Therefore, based on this information, EPA
does not expect water releases of 1,4-dioxane for this OES.
Surface Cleaner
EPA estimated daily wastewater discharges for facilities within the Surface Cleaner OES using the
SHEDs-HT model, which is described in Section 2.1.1.2. This modeling was completed for one case
study location (Liverpool OH) and only estimates indirect wastewater discharges. EPA does not expect
direct wastewater discharges to surface water from the types of commercial facilities within this OES
(e.g., restaurants, office buildings, other locations with janitorial services).
Dish Soap
EPA estimated daily wastewater discharges for facilities within the Surface Cleaner OES using the
SHEDs-HT model, which is described in Section 2.1.1.2. This modeling was completed for one case
study location (Liverpool OH) and only estimates indirect wastewater discharges. EPA does not expect
direct wastewater discharges to surface water from the types of commercial facilities within this OES
(e.g., restaurants, assisted living facilities, amusement, and recreation facilities).
Page 236 of 484
-------
6588
6589
6590
6591
6592
6593
6594
6595
6596
6597
6598
6599
6600
6601
6602
6603
6604
6605
6606
6607
6608
6609
6610
6611
6612
6613
6614
6615
6616
6617
6618
6619
6620
6621
6622
6623
6624
6625
6626
6627
6628
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Dishwasher Detergent
EPA estimated daily wastewater discharges for facilities within the Surface Cleaner OES using the
SHEDs-HT model, which is described in Section 2.1.1.2. This modeling was completed for one case
study location (Liverpool OH) and only estimates indirect wastewater discharges. EPA does not expect
direct wastewater discharges to surface water from the types of commercial facilities within this OES
(e.g., restaurants, assisted living facilities, amusement, and recreation facilities).
Laundry Detergent
EPA estimated daily wastewater discharges for facilities within the Laundry Detergent OES using the
OECD ESD on Industrial and Institutional Laundries (OECD. ^ ) and Monte Carlo modeling. The
ESD on Industrial and Institutional Laundries was rated "medium" during EPA's systematic review
process. The use of Montel Carlo modeling allows for variation of calculation input parameters such that
a distribution of environmental releases can be calculated, from which EPA can estimate the 50th and
95th percentile releases. An explanation of this modeling approach is included in Appendix E.12.
Paints and Floor Lacquer
EPA estimated daily wastewater discharges for facilities within the Paints and Floor Lacquers OES
using the OECD ESD on Coating Application via Spray-Painting in the Automotive Refinishing
Industry (OECD. 201 la). The ESD was rated "medium" during EPA's systematic review process.
As described in the process description in Appendix F.4.7, 1,4-dioxane was identified by a public
comment as present in automotive refinishing products, thereby allowing EPA to identify the above
ESD as the most applicable GS/ESD available. This ESD indicates that releases are expected from
1. Releases to incineration or landfill from container cleaning/disposal,
2. Releases to incineration or landfill from equipment cleaning,
3. Releases to incineration or landfill from over sprayed coating that is captured by emission
controls, and
4. Releases to air from over sprayed coating that is not captured by emission controls.
None of these releases are expected to water (OECD. 201 la). Therefore, based on this ESD, EPA does
not expect water releases of 1,4-dioxane for this OES.
Hydraulic Fracturing
EPA estimated daily wastewater discharges for facilities within the Hydraulic Fracturing OES using the
Draft OECD ESD on Hydraulic Fracturing ( 2022d) and Monte Carlo modeling. The Draft
ESD on Hydraulic Fracturing was rated "high" during EPA's systematic review process. The use of
Montel Carlo modeling allows for variation of calculation input parameters such that a distribution of
environmental releases can be calculated, from which EPA can estimate the 50th and 95th percentile
releases. An explanation of this modeling approach is included in Appendix E.13.
E.3.3 Water Release Estimates Summary
A summary of industrial and commercial water releases estimated using the above methods is presented
in Table Apx E-3 below.
Page 237 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
6629 Table Apx E-3. Summary of Daily Industrial and Commercial Water Release Estimates for 1,4-Dioxane
OES
Type of Water
Discharge
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Manufacturing
Surface Water
1
1.21
21.4
250
Medium
TRI, DMR
POTW or
Industrial WWT
1
0
6.69
Medium
TRI
Import and
Repackaging
Surface Water
6
0.91™
250
Medium
TRI, DMR
POTW or
Industrial WWT
6
0
0.91
Medium
TRI
Industrial Uses
Surface Water
24
0
24.5
250
Medium
TRI, DMR
POTW or
Industrial WWT
17
0
105
Medium
TRI
Functional
Fluids (Open-
System)
Surface Water
6
0
0.67
247
Medium
TRI, DMR
POTW or
Industrial WWT
1
4.67
70.9
Medium
TRI
Functional
Fluids (Closed-
System)
All
Assessed as a part of Industrial Uses OES
N/A
N/A
Laboratory
Chemical
Surface Water,
POTW, or
Industrial WWT
132
0 (water releases not expected)
250
High
GSd
Film Cement
Surface Water,
POTW, or
Industrial WWT
211
0 (water releases not expected)
250
High
Process
information®
Spray Foam
Application
Surface Water
1,553,559
0 (surface water releases not
expected)
3
Medium
GS^
POTW
1,553,559
0
0.0036
Medium
GS^
Printing Inks
(3D)
Surface Water
1
0.018
0.022
250
Medium
TRI, DMR
POTW or
Industrial WWT
1
0 (no indirect releases per TRI)
Medium
Medium
Page 238 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Water
Discharge
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Dry Film
Lubricant
Surface Water,
POTW, or
Industrial WWT
8
0 (water releases not expected)
48
High
Process
information®
Disposal
Surface Water
24
0
31.8
250
Medium
TRI, DMR
POTW or
Industrial WWT
4
0
0.91
Medium
TRI
Textile Dye
POTW
783
1.50E-05
0.001
31 to 295
Medium
ESDg and
Modeling''
Land (unknown
landfill type) or
POTW (unknown
partitioning)
783
2.09E-07
9.72E-05
Medium
ESDg and
Modeling''
Antifreeze
Surface water,
POTW, or
Industrial WWT
84,383
0 (water releases not expected)
250
High
Process
information®
and
Modeling''
Surface Cleaner
POTW
Unknown
0.072 (single value for all sites in
Liverpool OH case study)
250
N/A
SHEDS-HT'
Land (unknown
landfill) or
POTW
Unknown
18"
High
SHEDS-HT,
Process
information®
Modeling''
Dish Soap
POTW
Unknown
0.064"
250
N/A
SHEDS-HT"
Dishwasher
Detergent
POTW
Unknown
0.00144"
250
N/A
SHEDS-HT'
Laundry
Detergent
Fugitive air, stack
air, or POTW
95,533
1.51 E— 10
0.00714
250 to 365
Medium
ESD! and
Modeling''
Page 239 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Water
Discharge
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
(Institutional) -
Liquid
Detergents
(unknown
partitioning)
Land (unknown
landfill),
incineration, or
POTW (unknown
partitioning)
95,533
4.05E-12
3.95E-05
Medium
ESD! and
Modeling''
Laundry
Detergent
(Institutional) -
Powder
Detergents
Fugitive air, stack
air, or POTW
(unknown
partitioning)
95,533
3.05E-08
2.10E-04
250 to 365
Medium
ESD! and
Modeling''
Land (unknown
landfill),
incineration, or
POTW (unknown
partitioning)
95,533
5.36E-08
0.0018
Medium
ESD! and
Modeling''
Laundry
Detergent
(Industrial) -
liquid detergents
Fugitive air, stack
air, or POTW
(unknown
partitioning)
2,453
5.48E-12
0.011
20 to 365
Medium
ESD! and
Modeling''
Land (unknown
landfill),
incineration, or
POTW (unknown
partitioning)
2,453
4.78E-12
1.46E-04
Medium
ESD! and
Modeling''
Page 240 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Water
Discharge
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Laundry
Detergent
(Industrial) -
powder
detergents
Fugitive air, stack
air, or POTW
(unknown
partitioning)
2,453
1.76E-09
0.0112
20 to 365
Medium
ESD! and
Modeling''
Land (unknown
landfill),
incineration, or
POTW (unknown
partitioning)
2,453
2.92E-11
3.92E-04
Medium
ESD1 and
Modeling''
Paints and Floor
Lacquer
Surface water,
POTW, or
Industrial WWT
33,648
0 (water releases not expected)
250
Medium
ESD> and
process
information®
PET Byproduct
Surface water
19
0
10,050
250
Medium
TRI, DMR
POTW or
Industrial WWT
14
0
682
Medium
TRI
Ethoxylation
Process
Byproduct
Surface water
7
0
0.25
250
Medium
TRI, DMR
POTW or
Industrial WWT
6
0
448
Medium
TRI
Hydraulic
Fracturing
Surface water,
incineration, or
landfill (unknown
partitioning)
411
3.61E-10
4.59
1 to 72
Medium
ESD' and
Modeling''
Recycle/Reuse
(48%),
underground
injection (43%),
Surface water
(6%), or land
(3%)
411
1.85E-10
1.12
Medium
ESD'' and
Modeling''
"Where available. EPA used 2013-2019 TRI (U.S. EPA, 2022g) and 2013-2019 DMR ( 2022c) data to provide a basis to estimate the
number of sites using 1,4-dioxane within a COU.
Page 241 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Water
Discharge
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Min
Max
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Where available. EPA used the December 2020 Final Risk Evaluation for 1.4-Dioxane (U.S. EPA. 20200). generic scenarios, and emission scenario
documents to provide a basis to estimate the number of release days of 1,4-dioxane within a COU.
c Narrative descriptions of all release estimate sources are provided in Appendix E.3.2.
d The generic scenario used for this COU is the GS on Use of Laboratory Chemicals (U.S. EPA. 202211).
'' For this COU, EPA used process information, which is further described in Appendix E.3.2.
' The generic scenario used for this COU is the GS on Application of Spray PoK urethane Foam Insulation (U.S. EPA. 2018b).
g The emission scenario document used for this COU is the ESD on Textile Dyes (OECD. 2017).
h For this COU, EPA used various models and literature for model input parameters as described in Appendix E.3.2.
1 The emission scenario document used for this COU is the ESD on Industrial and Institutional Laundries (OECD. 201 lb).
1 The emission scenario document used for this COU is the ESD on Coating Application via Spray Painting in the Automotive Refinishing Industry
(OECD. 2011a).
k The emission scenario document used for this COU is the Draft ESD on Hydraulic Fracturing (U.S. EPA. 2022d).
' This value is the Commercial Upstream POTW releases estimated from the SHEDS-HT Down the Drain Model for the Liverpool OH case study (see
Section 2.1.1.2).
m All sites for this OES reported under Form A in TRI.
" A single annual value was provided for all sites in the Liverpool OH case study.
6630
Page 242 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
6631 E.3.4 Summary of Weight of the Scientific Evidence Conclusions in Water Release Estimates
6632 TableApx E-4 provides a summary of EPA's weight of the scientific evidence conclusions in its water release estimates for each of the OES.
6633 Detailed descriptions of non-OES specific strengths, limitations, assumptions, and uncertainties (e.g., general limitations for TRI, DMR, etc.)
6634 are provided in Appendix E.6.
6635
6636 Table Apx E-4. Summary of Weight of the Scientific Evidence Conclusions in Water Release Estimates by OES
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Manufacturing
Wastewater discharges are assessed using reported discharges from 2013-2019 TRI and DMR. Factors that increase the
strength of evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI
and DMR have medium overall data quality determinations, and consistency within the dataset (all reporters use the same or
similar reporting forms). EPA included 7 years of TRI and DMR data in the analysis, which increases the variability of the
dataset. A strength of TRI data is that TRI compiles the best readily available release data for all reporting facilities.
Strengths of DMR data are that it is based on monitoring data collected by facilities and the annual pollutant load is
calculated by integrating release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course
of the year. Factors that decrease the strength of the evidence for this OES include the low number of data points, uncertainty
in the accuracy of reported releases, and the limitations in representativeness to all sites because TRI may not capture all
relevant sites. Additionally, EPA made assumptions on the number of operating days to estimate daily releases, which
introduces additional uncertainty. Based on this information, EPA has concluded that the weight of the scientific evidence
for this assessment is moderate to robust and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Import and Repackaging
Wastewater discharges are assessed using reported discharges from 2013-2019 TRI and DMR. Factors that increase the
strength of evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI
and DMR have medium overall data quality determinations, and consistency within the dataset (all reporters use the same or
similar reporting forms). EPA included seven years of TRI and DMR data in the analysis, which increases the variability of
the dataset. A strength of TRI data is that TRI compiles the best readily available release data for all reporting facilities.
Strengths of DMR data are that it is based on monitoring data collected by facilities and the annual pollutant load is
calculated by integrating release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course
of the year. Factors that decrease the strength of the evidence for this OES include the low number of data points, uncertainty
in the accuracy of reported releases, uncertainty in EPA's use of Form A submissions, and the limitations in
representativeness to all sites because TRI may not capture all relevant sites. Some facilities within this OES reported to TRI
using a Form A, which does not include any details on chemical release quantities. When a facility has submitted a Form A,
there is no way to discern the quantity released. Therefore, where facilities reported to TRI with a Form A, EPA used the
Form A threshold for total releases of 500 lb/year for each release media; however, there is uncertainty in this because the
actual release quantity is unknown. Furthermore, the threshold represents an upper limit on total releases from the facility;
therefore, assessing releases at the threshold value may overestimate actual releases from the facility. Additionally,
uncertainty is introduced from EPA's assumptions on the number of operating days to estimate daily releases and in the
Page 243 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
mapping of DMR-reporting facilities to this OES. Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is moderate to robust and provides a plausible estimate of releases in consideration of
the strengths and limitations of reasonably available data.
Industrial Uses
Wastewater discharges are assessed using reported discharges from 2013-2019 TRI and DMR. Factors that increase the
strength of evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI
and DMR have medium overall data quality determinations, and consistency within the dataset (all reporters use the same or
similar reporting forms). EPA included seven years of TRI and DMR data in the analysis, which increases the variability of
the dataset. A strength of TRI data is that TRI compiles the best readily available release data for all reporting facilities.
Strengths of DMR data are that it is based on monitoring data collected by facilities and the annual pollutant load is
calculated by integrating release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course
of the year. Factors that decrease the strength of the evidence for this OES include uncertainty in the accuracy of reported
releases, uncertainty in EPA's use of Form A submissions, and the limitations in representativeness to all sites because TRI
may not capture all relevant sites. Some facilities within this OES reported to TRI using a Form A, which does not include
any details on chemical release quantities. When a facility has submitted a Form A, there is no way to discern the quantity
released. Therefore, where facilities reported to TRI with a Form A, EPA used the Form A threshold for total releases of 500
lb/year for each release media; however, there is uncertainty in this because the actual release quantity is unknown.
Furthermore, the threshold represents an upper limit on total releases from the facility; therefore, assessing releases at the
threshold value may overestimate actual releases from the facility. Additionally, uncertainty is introduced from EPA's
assumptions on the number of operating days to estimate daily releases and in the mapping of DMR-reporting facilities to
this OES. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is
moderate to robust and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Functional Fluids (Open-
System)
Wastewater discharges are assessed using reported discharges from 2013-2019 TRI and DMR. Factors that increase the
strength of evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI
and DMR have medium overall data quality determinations, and consistency within the dataset (all reporters use the same or
similar reporting forms). EPA included seven years of TRI and DMR data in the analysis, which increases the variability of
the dataset. A strength of TRI data is that TRI compiles the best readily available release data for all reporting facilities.
Strengths of DMR data are that it is based on monitoring data collected by facilities and the annual pollutant load is
calculated by integrating release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course
of the year. Factors that decrease the strength of the evidence for this OES include the low number of data points, uncertainty
in the accuracy of reported releases, and the limitations in representativeness to all sites because TRI may not capture all
relevant sites. The assessment includes data from only two sites that reported to TRI (one of which reported zero water
releases) and four that reported to DMR. Additionally, uncertainty is introduced from EPA's assumptions on the number of
operating days to estimate daily releases and in the mapping of DMR-reporting facilities to this OES. Based on this
information, EPA has concluded that the weight of the scientific evidence for this assessment is moderate to robust and
provides a plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Page 244 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Functional Fluids (Closed-
System)
No data was available to estimate releases for this OES. For the water release assessment, EPA grouped this OES with the
Industrial Uses OES because the sources of release are expected to be similar between these OESs. Factors that increase the
strength of evidence for this OES are that TRI and DMR have medium overall data quality determinations and consistency
within the dataset (all reporters use the same or similar reporting forms). Additionally, EPA included seven years of TRI and
DMR data in the analysis, which increases the variability of the dataset. Factors that decrease the strength of evidence for
this OES are that the Industrial Releases OES release data are use as surrogate for this OES, uncertainty in the accuracy of
reported releases, and the limitations in representativeness to all sites because TRI may not capture all relevant sites.
Refer to the Industrial Uses OES discussion for additional discussion. Based on this information, EPA has concluded that the
weight of the scientific evidence for this assessment is slight and provides a plausible estimate of releases in consideration of
the strengths and limitations of reasonably available data.
Laboratory Chemicals
Wastewater discharges are assessed using the Draft GS on Use of Laboratory Chemicals. Per the GS, water releases are not
expected for hazardous chemicals. Because 1,4-dioxane is considered a hazardous chemical under CERCLA, no water
releases are expected for this OES according to the GS. Factors that increase the strength of evidence for this OES are that
the release estimates are directly relevant to the OES (as opposed to surrogate), the Draft GS on Use of Laboratory
Chemicals has a high overall data quality determination, and there is a low level of uncertainty in the data. Factors that
decrease the strength of the evidence for this OES include the that the GS has not been peer-reviewed, uncertainty in the
representativeness of the GS towards all sites in this OES, and a lack of variability in the analysis. Specifically, because the
default values in the ESD are generic, there is uncertainty in the representativeness of generic site estimates of actual releases
from real-world sites that use 1,4-dioxane. Another uncertainty is lack of consideration for release controls. The ESD
assumes that all activities occur without any release controls. Actual releases may be less than estimated if facilities utilize
pollution control methods, contributing to uncertainty. Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is slight to moderate and provides a plausible estimate of releases in consideration of
the strengths and limitations of reasonably available data.
Film Cement
Wastewater discharges are assessed using process information from the Final Risk Evaluation for 1,4-Dioxane. Per the
process information, EPA does not expect water releases of 1,4-dioxane for this OES because 1,4-dioxane volatilizes to air
after application of the film cement and empty cement bottles are disposed of as solid waste without rinsing. Factors that
increase the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to
surrogate), the underlying data sources for the process information have a high overall data quality determination, and there
is a low level of uncertainty in the data because the process information comes directly from actual users of 1,4-dioxane in
film cement. Factors that decrease the strength of the evidence for this OES include the uncertainty in the representativeness
of evidence to all sites in this OES and a lack of variability. Specifically, the process information for the production and use
of film cement is based on information from three use sites, one from Australia and two from the U.S. Based on this
information, EPA has concluded that the weight of the scientific evidence for this assessment is slight to moderate and
provides a plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Page 245 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Spray Foam Application
Wastewater discharges are assessed using the GS on Application of Spray Polyurethane Foam Insulation. Factors that
increase the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to
surrogate), the underlying data sources for the process information have a medium overall data quality determination, and
there is a low level of uncertainty in the data. Factors that decrease the strength of the evidence for this OES include
uncertainty in the representativeness of the GS to all sites since it is generic and not specific to sites that use 1,4-dioxane and
a lack of variability. Based on this information, EPA has concluded that the weight of the scientific evidence for this
assessment is slight to moderate and provides a plausible estimate of releases in consideration of the strengths and limitations
of reasonably available data.
Printing Inks (3D)
Wastewater discharges are assessed using reported discharges from 2013-2019 DMR. Factors that increase the strength of
evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that DMR has a
medium overall data quality determination, and consistency within the dataset (all reporters use the same or similar reporting
forms). Additionally, EPA used DMR data for seven years, which increases the variability of the dataset. Strengths of DMR
data are that it is based on monitoring data collected by facilities and the annual pollutant load is calculated by integrating
release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course of the year. Factors that
decrease the strength of the evidence for this OES include the low number of data points, uncertainty in the accuracy of
reported releases, and the limitations in representativeness to all sites. Additionally, no TRI data is available for this OES,
EPA made assumptions on the number of operating days, and there is uncertainty in the mapping of DMR-reporting facilities
to this OES. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is
moderate to robust and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Dry Film Lubricant
Wastewater discharges are assessed using process information from the Final Risk Evaluation for 1,4-Dioxane. Based on
conversations the with only known user who supplied this process information, EPA expects wastes to be drummed and sent
to a waste handler with residual wastes releasing to air or being disposed to landfill, such that there are no water releases.
Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as
opposed to surrogate), the underlying data sources for the process information have a high overall data quality determination,
and there is a low level of uncertainty in the data. Additionally, the process information comes directly from an actual user of
1,4-dioxane in dry film lubricants. Factors that decrease the strength of the evidence for this OES include uncertainty in the
representativeness of evidence to all sites and a lack of variability. Based on this information, EPA has concluded that the
weight of the scientific evidence for this assessment is slight to moderate and provides a plausible estimate of releases in
consideration of the strengths and limitations of reasonably available data.
Disposal
Wastewater discharges are assessed using reported discharges from 2013-2019 TRI and DMR. Factors that increase the
strength of evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI
and DMR have medium overall data quality determinations, and consistency within the dataset (all reporters use the same or
similar reporting forms). EPA included seven years of TRI and DMR data in the analysis, which increases the variability of
the dataset. A strength of TRI data is that TRI compiles the best readily available release data for all reporting facilities.
Strengths of DMR data are that it is based on monitoring data collected by facilities and the annual pollutant load is
Page 246 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
calculated by integrating release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course
of the year. Factors that decrease the strength of the evidence for this OES include uncertainty in the accuracy of reported
releases, uncertainty in EPA's use of Form A submissions, and the limitations in representativeness to all sites because TRI
may not capture all relevant sites. Some facilities within this OES reported to TRI using a Form A, which does not include
any details on chemical release quantities. When a facility has submitted a Form A, there is no way to discern the quantity
released. Therefore, where facilities reported to TRI with a Form A, EPA used the Form A threshold for total releases of 500
lb/year for each release media; however, there is uncertainty in this because the actual release quantity is unknown.
Furthermore, the threshold represents an upper limit on total releases from the facility; therefore, assessing releases at the
threshold value may overestimate actual releases from the facility. Additionally, uncertainty is introduced from EPA's
assumptions on the number of operating days to estimate daily releases and in the mapping of DMR-reporting facilities to
this OES. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is
moderate to robust and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Textile Dye
Wastewater discharges are assessed using Monte Carlo modeling with information from the ESD on Textile Dyes. Factors
that increase the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed
to surrogate), that the ESD on Textile Dyes has a medium overall data quality determination and was peer reviewed, the high
number of data points (simulation runs), consistency within the dataset, and full distributions of input parameters. The Monte
Carlo modeling accounts for the entire distribution of input parameters, calculating a distribution of potential release values
that represents a larger proportion of sites than a discrete value. Factors that decrease the strength of the evidence for this
OES include uncertainties and limitations in the representativeness of the estimates for sites that specifically use 1,4-dioxane
because the default values in the ESD are generic. Another uncertainty is lack of consideration for release controls. The ESD
assumes that all activities occur without any release controls. Actual releases may be less than estimated if facilities utilize
pollution control methods, contributing to uncertainty. Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is moderate and provides a plausible estimate of releases in consideration of the
strengths and limitations of reasonably available data.
Antifreeze
Wastewater discharges are assessed using the OECD ESD on Chemical Additives used in Automotive Lubricants and the
EPA MRD on Commercial Use of Automotive Detailing Products. Factors that increase the strength of evidence for this
OES are that the ESD and MRD used have high overall data quality determinations, consistency within the sources used, and
a low number of uncertainties. Both sources indicate that containers of automotive maintenance fluids are not typically
rinsed, but rather disposed of as solid waste or incinerated, such that there are no water releases, contributing to consistency
and a low level of uncertainty in the data. Factors that decrease the strength of the evidence for this OES include that the
ESD and MRD are not directly applicable to antifreeze uses (used as surrogate), uncertainty in the representativeness of the
ESD and MRD to all sites and sites that specifically use 1,4-dioxane since these documents contain generic values, and a
lack of variability. Based on this information, EPA has concluded that the weight of the scientific evidence for this
assessment is slight to moderate and provides a plausible estimate of releases in consideration of the strengths and limitations
of reasonably available data.
Page 247 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Surface Cleaner
Wastewater discharges are assessed using the SHEDS-HT model. Factors that increase the strength of evidence for this OES
include that the release estimates are directly relevant to the OES (as opposed to surrogate) and variability in the model input
parameters. Factors that decrease the strength of the evidence for this OES include uncertainty in the representativeness to all
sites because the estimate is based on one case study for Liverpool, OH and because the estimate is not site-specific (the
release estimate is a total for all sites in Liverpool, OH). Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is slight and provides a plausible estimate of releases in consideration of the strengths
and limitations of reasonably available data.
Dish Soap
Wastewater discharges are assessed using the SHEDS-HT model. Factors that increase the strength of evidence for this OES
include that the release estimates are directly relevant to the OES (as opposed to surrogate) and variability in the model input
parameters. Factors that decrease the strength of the evidence for this OES include uncertainty in the representativeness to all
sites because the estimate is based on one case study for Liverpool, OH and because the estimate is not site-specific (the
release estimate is a total for all sites in Liverpool, OH). Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is slight and provides a plausible estimate of releases in consideration of the strengths
and limitations of reasonably available data.
Dishwasher Detergent
Wastewater discharges are assessed using the SHEDS-HT model. Factors that increase the strength of evidence for this OES
include that the release estimates are directly relevant to the OES (as opposed to surrogate) and variability in the model input
parameters. Factors that decrease the strength of the evidence for this OES include uncertainty in the representativeness to all
sites because the estimate is based on one case study for Liverpool, OH and because the estimate is not site-specific (the
release estimate is a total for all sites in Liverpool, OH). Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is slight and provides a plausible estimate of releases in consideration of the strengths
and limitations of reasonably available data.
Laundry Detergent
Wastewater discharges are assessed using Monte Carlo modeling with information from the ESD on Industrial and
Institutional Laundries. Factors that increase the strength of evidence for this OES are that the release estimates are directly
relevant to the OES (as opposed to surrogate), that the ESD on Industrial and Institutional Laundries has a medium overall
data quality determination and was peer reviewed, there are a high number of data points (simulation runs), consistency
within the dataset, and full distributions of input parameters. Monte Carlo modeling accounts for the entire distribution of
input parameters, calculating a distribution of potential release values that represents a larger proportion of sites than a
discrete value. Additionally, EPA was able to separately estimate releases for industrial and institutional laundry settings.
Factors that decrease the strength of the evidence for this OES include uncertainties and limitations in the representativeness
of the estimates for sites that specifically use 1,4-dioxane because the default values in the ESD are generic. Another
uncertainty is lack of consideration for release controls. The ESD assumes that all activities occur without any release
controls. Actual releases may be less than estimated if facilities utilize pollution control methods. Based on this information,
EPA has concluded that the weight of the scientific evidence for this assessment is moderate and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Page 248 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Paint and Floor Lacquer
Wastewater discharges are assessed using OECD ESD on Coating Application via Spray-Painting in the Automotive
Refinishing Industry. According to the ESD, no releases are expected to water. Factors that increase the strength of evidence
for this OES are that the release estimates are directly relevant to the OES (as opposed to surrogate), the ESD has a medium
overall data quality determination, and a low number of uncertainties. F Factors that decrease the strength of the evidence for
this OES include a lack of variability and uncertainty in the representativeness of the ESD to all sites and sites that
specifically use 1,4-dioxane since the ESD is generic. Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is slight to moderate and provides a plausible estimate of releases in consideration of
the strengths and limitations of reasonably available data.
Polyethylene
Terephthalate (PET)
Byproduct
Wastewater discharges are assessed using reported discharges from 2013-2019 TRI and DMR. Factors that increase the
strength of evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI
and DMR have medium overall data quality determinations, consistency within the dataset (all reporters use the same or
similar reporting forms), and consistency with the emission data from the related life cycle analysis discussed in Appendix
E.6. EPA included seven years of TRI and DMR data in the analysis, which increases the variability of the dataset. A
strength of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Strengths of DMR
data are that it is based on monitoring data collected by facilities and the annual pollutant load is calculated by integrating
release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course of the year. Factors that
decrease the strength of the evidence for this OES include the uncertainty in the accuracy of reported releases and the
limitations in representativeness to all sites because TRI may not capture all relevant sites. Additionally, EPA made
assumptions on the number of operating days to estimate daily releases, which introduces additional uncertainty. Based on
this information, EPA has concluded that the weight of the scientific evidence for this assessment is moderate to robust and
provides a plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Ethoxylation Process
Byproduct
Wastewater discharges are assessed using reported discharges from 2013-2019 TRI and DMR. Factors that increase the
strength of evidence for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI
and DMR have medium overall data quality determinations, and consistency within the dataset (all reporters use the same or
similar reporting forms). EPA included seven years of TRI and DMR data in the analysis, which increases the variability of
the dataset. A strength of TRI data is that TRI compiles the best readily available release data for all reporting facilities.
Strengths of DMR data are that it is based on monitoring data collected by facilities and the annual pollutant load is
calculated by integrating release reports over shorter timeframes (e.g., monthly, quarterly) and extrapolating over the course
of the year. Factors that decrease the strength of the evidence for this OES include uncertainty in the accuracy of reported
releases and the limitations in representativeness to all sites because TRI may not capture all relevant sites. Additionally,
EPA made assumptions on the number of operating days to estimate daily releases, which introduces additional uncertainty.
Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is moderate to
robust and provides a plausible estimate of releases in consideration of the strengths and limitations of reasonably available
data.
Hydraulic Fracturing
Wastewater discharges are assessed using Monte Carlo modeling with information from the Draft ESD on Hydraulic
Fracturing and FracFocus 3.0. Factors that increase the strength of evidence for this OES are that the release estimates are
Page 249 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
directly relevant to the OES (as opposed to surrogate), that the Draft ESD on Hydraulic Fracturing and FracFocus 3.0 have
medium overall data quality determinations, the high number of data points (simulation runs), and full distributions of input
parameters. Monte Carlo modeling accounts for the entire distribution of input parameters, calculating a distribution of
potential release values that represents a larger proportion of sites than a discrete value. Factors that decrease the strength of
the evidence for this OES include that the Draft ESD has not been peer reviewed, uncertainties that may result in over-
estimates of releases, and limitations in the representativeness of the estimates for all sites. Specifically, EPA used some
input values from the Draft ESD; because the default values in the ESD are generic, there is uncertainty in the
representativeness of the generic site estimates of real-world sites that use 1,4-dioxane. Another uncertainty is lack of
consideration for release controls. The ESD assumes that all activities occur without any release controls. Actual releases
may be less than estimated if facilities utilize pollution control methods, contributing to uncertainty. Based on this
information, EPA has concluded that the weight of the scientific evidence for this assessment is moderate to robust and
provides a plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
6637
Page 250 of 484
-------
6638
6639
6640
6641
6642
6643
6644
6645
6646
6647
6648
6649
6650
6651
6652
6653
6654
6655
6656
6657
6658
6659
6660
6661
6662
6663
6664
6665
6666
6667
6668
6669
6670
6671
6672
6673
6674
6675
6676
6677
6678
6679
6680
6681
6682
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.4 Land Release Assessment
This section describes EPA's methodology for estimating annual land releases from industrial and
commercial facilities manufacturing, processing, or using 1,4-dioxane. EPA did not estimate daily land
releases due to the high level of uncertainty in the number of release days associated with land releases.
Facilities report annual land releases to the Toxics Release Inventory (TRI), which include a variety of
release mechanisms, including but not limited to underground injection, RCRA Subtitle C landfills,
other landfills, surface impoundments, and land treatment. EPA used 2019 TRI ( 2022e) data
to estimate annual land releases for the OES where available; however, EPA did not have these data for
every OES. For OES without TRI data, EPA used alternate assessment approaches to estimate annual
land releases.
In addition, EPA did a more in-depth analysis of TRI for sites within the disposal OES. Specifically,
EPA did an analysis of 2013 to 2019 TRI data for this OES. Operations at disposal sites are expected to
be more complex than those at sites in other OES, which typically generate waste for land disposal off
site. Additionally, the disposal OES includes the sites of ultimate disposal {i.e., they are the landfills
themselves) and EPA considered the impact of continuous years of land releases of 1,4-dioxane at these
sites on general population and ecological exposures.
E.4.1 Assessment Using TRI
EPA found 2019 TRI data for facilities within the following OESs:
• Manufacturing
• Import and Repackaging
• Industrial Uses
• Functional Fluids (Open-System)
• Disposal
• PET Byproduct
• Ethoxylation Byproduct
The TRI data were rated "medium" in EPA's systematic review process. EPA estimated annual land
releases using TRI for these OESs, with the following general stages as described in the rest of this
section.
1. Collect land release data from the 2013 to 2019 TRI for the Disposal OES and 2019 TRI data for
all other OES,
2. Map land release data to occupational exposure scenarios,
3. Analyze 2013 to 2019 TRI data for the disposal OES, and
4. Summarize 2019 annual land releases for the other OES.
Step 1: Collect Land Release Data from TRI
The first step in estimating land releases was to obtain TRI data. As previously discussed in Appendix
E.3.1, each facility subject to the TRI reporting rule must report annually the volume of chemical
released to the environment and/or managed through recycling, energy recovery, and treatment. Similar
to the air release assessment, EPA included both TRI reporting Form R and TRI reporting Form A
submissions in the land release assessment. Where sites reported to TRI with Form A, EPA used the
Form A threshold for total releases of 500 lb/year. EPA used the entire 500 lb/year for each type of land
release; however, since this threshold is for total site releases, these 500 lb/year are attributed one type
of land release at a time (since assessing it for more than one land release media at a time would double
count the releases and exceed the total release threshold for Form A). EPA pulled the TRI Basic Plus
Data Files for each of years 2013 through 2019.
Page 251 of 484
-------
6683
6684
6685
6686
6687
6688
6689
6690
6691
6692
6693
6694
6695
6696
6697
6698
6699
6700
6701
6702
6703
6704
6705
6706
6707
6708
6709
6710
6711
6712
6713
6714
6715
6716
6717
6718
6719
6720
6721
6722
6723
6724
6725
6726
6727
6728
6729
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI data include both on- and off-site land releases. In summary, TRI includes the following land
release media:
• On-site releases:
o Underground injection
o RCRA subtitle C landfills
o Other landfills
o Land treatment
o RCRA surface impoundments
o Other surface impoundments
o Other land disposal
o Waste rock
• Off-site releases:
o Underground injection
o RCRA subtitle C landfills
o Other landfills
o Land treatment
o RCRA surface impoundments
o Other surface impoundments
o Other land disposal
o Transfer to waste broker for disposal
o Solidification/stabilization
Step 2: Map Land Release Data to Occupational Exposure Scenarios
The next step in estimating land releases was to map the 2013 to 2019 TRI data to the 1,4-dioxane OES.
EPA used the same mapping methodology as that used for both the air and water assessments, which is
described in Appendix E.5.1. EPA ensured consistency in the OES mapping between the air, water, and
land assessments.
Step 3: Analyze and Summarize 2013 to 2019 TRI Data for the Disposal OES
For the sites that EPA mapped to the disposal OES in the 2013 to 2019 TRI data, EPA analyzed and
summarized the land release data as follows:
• EPA summarized which of the reporting years that each disposal facility submitted data to TRI.
This summary allows for visualization of which sites report recurring land disposal of 1,4-
dioxane between 2013 and 2019.
• EPA differentiated between disposal sites that transferred 1,4-dioxane to other sites for disposal
and the receiving sites that disposed of 1,4-dioxane on site. For the receiving sites at which 1,4-
dioxane was disposed of to land, EPA summarized the number of unique sites from which the
receiving sites received 1,4-dioxane for land disposal and the total amount of 1,4-dioxane
received for land disposal between 2013 and 2019.
• EPA summarized the total amount of 1,4-dioxane released to each land release media between
2013 and 2019. In summary, 1,4-dioxane was disposed of from disposal OES sites via on-site
and off-site RCRA subtitle C landfills, on-site and off-site underground injection, and off-site
other landfills between 2013 and 2019.
EPA's analysis and summary of land releases for 2013 to 2019 TRI sites in the disposal OES can be
found in 1,4-Dioxane Supplemental Information File: Environmental Releases to landfor the Disposal
OES ( 231).
Page 252 of 484
-------
6730
6731
6732
6733
6734
6735
6736
6737
6738
6739
6740
6741
6742
6743
6744
6745
6746
6747
6748
6749
6750
6751
6752
6753
6754
6755
6756
6757
6758
6759
6760
6761
6762
6763
6764
6765
6766
6767
6768
6769
6770
6771
6772
6773
6774
6775
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Step 4: Summarize Annual Land Releases for Other OES with 2019 TRI data
For the remaining OES for which 2019 TRI data were available, EPA summarized the annual land
releases by media type (e.g., underground injection, RCRA subtitle C landfills, other landfills, land
treatment) and site information, including site identity, city, state, zip code, TRI facility ID, and FRS ID.
EPA did not estimate daily land releases due to the high level of uncertainty in the number of release
days associated with land releases.
EPA's summary of land release for these OESs is included in 1,4-Dioxane Supplemental Information
File: Environmental Releases to Landfor all OES Except Disposal ( 023k).
E.4.2 Assessment for OES without TRI
EPA did not find 2019 TRI data for the following OES:
• Functional Fluids (Closed-Systems)
• Laboratory Chemicals
• Film Cement
• Spray Polyurethane Foam
• 3D Printing
• Dry Film Lubricant
• Textile Dye
• Antifreeze
• Surface Cleaner
• Dish Soap
• Dishwasher Detergent
• Laundry Detergent
• Paints and Floor Lacquer
• Hydraulic Fracturing
For these OESs, EPA estimated land releases by using various modeling approaches, including the use
of surrogate TRI data and modeling using data from literature, GSs, and ESDs. EPA's assessment of
land releases for each of these OESs is described below.
Functional Fluids (Closed-Systems)
Land release data were not available for this OES and EPA did not find any information to model land
release for this OES using literature, GSs, or ESDs. EPA expects that the sources of release for this OES
to be similar to those for the Industrial Uses OES, based on the process information in the Final Risk
Evaluation for 1,4-Dioxane (U, S. EPA. 2020c). Therefore, EPA grouped the land release assessment for
Functional Fluids (Closed-Systems) into that for Industrial Uses. However, there is uncertainty in this
assumption of similar release sources between these OESs.
Laboratory Chemicals
EPA estimated land releases for facilities within the Laboratory Chemicals OES using the Draft GS on
Use of Laboratory Chemicals ( 22h). The GS on Use of Laboratory Chemicals was rated
"high" during EPA's systematic review process.
The GS indicates that there are eight release points:
1. Release to air from transferring volatile chemicals from transport containers.
2. Release to air, water, incineration, or landfill from transferring solid powders.
3. Release to water, incineration, or land from cleaning or disposal of transport containers.
Page 253 of 484
-------
6776
6777
6778
6779
6780
6781
6782
6783
6784
6785
6786
6787
6788
6789
6790
6791
6792
6793
6794
6795
6796
6797
6798
6799
6800
6801
6802
6803
6804
6805
6806
6807
6808
6809
6810
6811
6812
6813
6814
6815
6816
6817
6818
6819
6820
6821
6822
6823
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4. Release to air from cleaning containers used for volatile chemicals.
5. Labware equipment cleaning residuals released to water, incineration, or landfill.
6. Release to air during labware equipment cleaning for volatile chemicals.
7. Release to air from laboratory analyses for volatile chemicals.
8. Release to water, incineration, or landfill from laboratory waste disposal.
Based on the GS, release points #2, 3, 5, and 8 have the potential for land releases; however, release
point #2 is not applicable because 1,4-dioxane is not a solid powder. To estimate the remaining land
releases, EPA used the equations specified in the Draft GS ( |22h). EPA did not find any
data specific to 1,4-dioxane in this OES. Therefore, the calculation of releases using this GS are for a
"generic site," using the default input parameter values from the GS.
Using this methodology, EPA calculated high-end and low-end annual land releases for this OES. The
low- and high-end estimates are based on the low-end or typical and high-end or worst-case calculation
input parameter defaults from the GS. EPA's calculation of land releases for this OES, including all
calculation inputs, can be found in 1,4-Dioxane Supplemental Information File: Environmental Releases
to Landfor all OES Except Disposal (U.S. EPA. 2023k).
Film Cement
EPA estimated land releases for facilities within the Film Cement OES using process information from
the Final Risk Evaluation for 1,4-Dioxane ( 2020c). The underlying process information for
this assessment was rated "high" during EPA's systematic review process.
The process of using film cement involves applying the cement onto edges of photographic films by
hand using a small brush, then joining the pieces of film by pressing and heating to dry the cement.
Based on this process information, EPA expects land releases may result from disposal of empty film
cement bottles that contain residual amounts of film cement containing 1,4-dioxane. EPA estimated this
land release as a range, using a film cement use rate of 2.5 to 12 L/site-year and a concentration of 1,4-
dioxane in the film cement of 45 to 50 percent from the process information in the Final Risk Evaluation
for 1,4-Dioxane (U.S. EPA. 2020c). and the EPA/OPPT Small Container Residual Model central
tendency loss fraction of 0.3 percent and high-end loss fraction of 0.6 percent. EPA is uncertain of the
specific type of land disposal for the empty film cement bottles.
EPA's calculation of land releases for this OES, including all calculation inputs and assumptions, can be
found in 1,4-Dioxane Supplemental Information File: Environmental Releases to Landfor all OES
Except Disposal (U.S.. 023k).
Spray Polyurethane Foam
EPA estimated land releases for facilities within the Spray Polyurethane Foam OES using the GS on
Application of Spray Polyurethane Foam Insulation (U.S. EPA. 2020c). The GS on the Application of
Spray Polyurethane Foam Insulation was rated "medium" during EPA's systematic review process.
The GS indicates that there are six release points:
1. Releases to fugitive air for volatile chemicals during unloading of raw materials from transport
containers;
2. Releases to water, incineration, or landfill from cleaning or disposal of transport containers;
3. Releases to fugitive air for volatile chemicals during transport container cleaning;
4. Releases to incineration or landfill from spray polyurethane foam application equipment
cleaning;
Page 254 of 484
-------
6824
6825
6826
6827
6828
6829
6830
6831
6832
6833
6834
6835
6836
6837
6838
6839
6840
6841
6842
6843
6844
6845
6846
6847
6848
6849
6850
6851
6852
6853
6854
6855
6856
6857
6858
6859
6860
6861
6862
6863
6864
6865
6866
6867
6868
6869
6870
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
5. Releases to fugitive air for volatile chemicals during equipment cleaning; and
6. Releases to landfill of scrap foam from trimming applied foam.
Based on the GS, release points #2, 4, and 6 have the potential for land releases. To estimate these
releases, EPA used the equations specified in the GS ( 1020c). Apart from weight fraction in
spray polyurethan foam, EPA did not find any data specific to 1,4-dioxane in this OES. Therefore, the
calculation of releases using this GS are for a "generic site," using the default input parameter values
from the GS. Specifically, EPA used the same input parameter values that were used in the original risk
evaluation for estimates of occupational exposure; see Appendix G of the Final Risk Evaluation for 1,4-
Dioxane (U.S. EPA. 2020c).
Using this methodology, EPA calculated high-end and low-end annual land releases for this OES. The
low- and high-end estimates are based on the low-end or typical and high-end or worst-case calculation
input parameter defaults from the GS. EPA's calculation of land releases for this OES, including all
calculation inputs, can be found 1,4-Dioxane Supplemental Information File: Environmental Releases to
Landfor all OES Except Disposal (U.S. EPA. 2023k).
3D Printing
Land release data were not available for this OES and EPA did not find any information to model land
releases for this OES using literature, GSs, or ESDs. EPA expects that industrial applications of this
OES to be accounted for in the Industrial Uses TRI data. Per the December 2020 Final Risk Evaluation
for 1,4-Dioxane ( 320c). 3D printing ink containing 1,4-dioxane is used in research labs to
print biomedical products. Because the 2019 TRI data for the Industrial Uses OES include medicinal and
pharmaceutical manufacturing NAICS codes, medical research labs that conduct 3D printing with 1,4-
dioxane inks may be captured in that OES. Therefore, EPA grouped the land release assessment for 3D
Printing into that for Industrial Uses. However, there is uncertainty in whether 3D printing sites are truly
captured in the Industrial Uses TRI data.
Dry Film Lubricant
EPA estimated land releases for facilities within the Dry Film Lubricant OES using process information
from the Final Risk Evaluation for 1,4-Dioxane ( '20c). The underlying process information
for this assessment was rated "high" during EPA's systematic review process.
The process for the production and use of dry film lubricant is described in the 2020 RE and is based on
information provided to EPA by the one known user. In summary, the process entails first producing the
concentrated dry film lubricant by mixing 1,4-dioxane with other additives, followed by the dilution of
the concentrated dry film lubricant with additional 1,4-dioxane and the use of the dry film lubricant. The
use involves spray application onto substrates in a vented paint booth and the subsequent curing in a
vented oven and cleaning of the dried parts in a 1,4-dioxane bath ( 10c). Based on this
process description, EPA expects land releases may result from
1. Residuals in empty containers of pure 1,4-dioxane used for mixing of the concentrated dry film
lubricant,
2. Cleaning residuals for equipment used for mixing of the concentrated dry film lubricant,
3. Residuals in empty containers of pure 1,4-dioxane used for diluting the concentrated dry film
lubricant,
4. Residuals in empty containers of concentrated dry film lubricant, and
5. Waste from cleaning spray application equipment and the parts onto which the dry film lubricant
was applied.
Page 255 of 484
-------
6871
6872
6873
6874
6875
6876
6877
6878
6879
6880
6881
6882
6883
6884
6885
6886
6887
6888
6889
6890
6891
6892
6893
6894
6895
6896
6897
6898
6899
6900
6901
6902
6903
6904
6905
6906
6907
6908
6909
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
EPA estimated land releases using 1,4-dioxane use rates derived from the process information and
standard EPA models. Specifically, EPA estimated land releases from release points #1, 3 and 4 using
the EPA/OPPT Small Container Residual Model, which has a central tendency loss fraction of 0.3
percent and a high-end loss fraction of 0.6 percent of the container volume. EPA used container volumes
specified in the process information, which are either 1-gallon or 0.5-pint containers ( ?20c).
EPA estimated releases from release point #2 using the EPA/OPPT Single Process Vessel Residual
Model, which has a central tendency loss fraction of 0.2 percent and a high-end loss fraction of 1 percent
of the 1,4-dioxane throughput. EPA estimated land releases from release point #5 by assuming the entire
volume of the cleaning bath used for equipment and parts is released to landfill. This is consistent with
the process information, which indicates that spent 1,4-dioxane is disposed of as chemical waste, which
EPA assumes may be to either incineration or RCRA subpart C landfills (U.S. EPA. 2020c).
EPA's calculation of land releases for this OES, including all calculation inputs and assumptions, can be
found in 1,4-Dioxane Supplemental Information File: Environmental Releases to Landfor all OES
Except Disposal ( BP A. 2023k).
Textile Dye
EPA estimated land releases for facilities within the Textile Dye OES using the OECD ESD on Textile
Dyes (OECD. 2017) and Monte Carlo modeling. The ESD on Textile Dyes was rated "medium" during
EPA's systematic review process. The use of Montel Carlo modeling allows for variation of calculation
input parameters such that a distribution of environmental releases can be calculated, from which EPA
can estimate the 50th and 95th percentile releases. An explanation of this modeling approach is included
in Appendix E. 11.
Antifreeze
EPA did not find any directly applicable GS/ESD or literature sources for this OES; however, EPA
evaluated the potential for releases using the OECD ESD on Chemical Additives used in Automotive
Lubricants (< 20) and the EPA MRD on Commercial Use of Automotive Detailing Products
(1 c. i i1 \ AV2b). The ESD and MRD were both rated "high" during EPA's systematic review
process.
For the use of antifreeze, EPA expects releases may occur from volatilizations of 1,4-dioxane during
unloading/ pouring antifreeze into vehicles, disposal, or cleaning of empty antifreeze containers, and
spent antifreeze. Both the ESD and MRD indicate that containers of automotive maintenance fluids are
typically small and are not rinsed, but rather disposed of as solid waste (I v < < \ .022 k * ^ CD.
2020). Additionally, the ESD on Chemical Additives used in Automotive Lubricants indicates that spent
lubricants are disposed of via incineration by blending with fuel oil (OECD. 2020). However, since
spent antifreeze is unlikely to be blended with fuel oil, EPA expects spent antifreeze may be disposed of
via incineration or landfills that take chemical waste. Therefore, EPA expects land releases result from
disposal of empty antifreeze containers and spent antifreeze.
To estimate the use rate of 1,4-dioxane for this OES, EPA used the consumer use rate of antifreeze (0.15
kg antifreeze/job) from the Final Risk Evaluation for 1,4-Dioxane ( '20c) and scaled this
value up to a commercial use rate based on a range of the number of cars serviced at auto shops from the
Near-Field/Far-Field Brake Model and Automotive Detailing MRD (1 to 9 jobs/day). EPA used a range
of concentration of 1,4-dioxane in antifreeze from the process description in Appendix F.4.2 and
assumed antifreeze container sizes ranging from 16 ounces to 5 gallons per the default container sizes in
the MRD and ESD, respectively (I v « « \ 202 J., * i n :020).
Page 256 of 484
-------
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
6931
6932
6933
6934
6935
6936
6937
6938
6939
6940
6941
6942
6943
6944
6945
6946
6947
6948
6949
6950
6951
6952
6953
6954
6955
6956
6957
6958
6959
6960
6961
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
To estimate the land release from container disposal, EPA used the calculated 1,4-dioxane throughput
based on the above batch parameters and the EPA/OPPT Small Container Residual Model, which has a
central tendency loss fraction of 0.3 percent and a high-end loss fraction of 0.6 percent. To estimate the
land release from spent antifreeze, EPA used the 1,4-dioxane throughput and a mass balance assuming
100 percent release minus upstream losses from container disposal and volatilizations during unloading
(estimated with the EPA/OAQPS AP-42 Loading Model).
EPA's calculation of land releases for this OES, including all calculation inputs and assumptions, can be
found in the supplemental attachment 1,4-Dioxane Supplemental Information File: Environmental
Releases to Landfor all OES Except Disposal (U.S. EPA. 2023k).
Surface Cleaner
EPA did not find any directly applicable GS/ESD or literature sources for this OES; however, EPA
estimated land releases using the SHEDs-HT modeling conducted for the one case study location
(Liverpool OH) and the assumptions described here. EPA expects that the main release points from the
use of surface cleaners are from
1. Disposal of empty containers containing residual cleaning solution,
2. Application of the cleaning solution, and
3. Disposal of cleaning solution by rinsing or wiping.
Because EPA did not find any directly applicable GSs or ESDs, EPA used the Draft GS on Furnishing
Cleaning ( 322a) to inform these releases due to the similarities in surface cleaning and
furnishing cleaning. The Draft GS on Furnishing Cleaning was rated "high" during EPA's systematic
review process. Per this Draft GS, empty containers may be rinsed out in sinks or disposed of without
rinsing, such that releases may be to wastewater or landfill; the GS uses the EPA/OPPT Small Container
Residual Model to estimate this release. Application losses are to fugitive air from spray application; the
GS uses literature data to estimate this release. Once applied, the cleaner may be rinsed off or wiped off
with rags or towels, such that releases may be to wastewater or landfill; the GS assumes 100 percent
release scenario, estimating this release by subtracting the upstream losses from the cleaner use rate
( >022a).
The SHEDs-HT modeling estimated wastewater discharges of 72 g of 1,4-dioxane per day for
commercial uses of surface cleaners containing 1,4-dioxane in Liverpool OH. As described previously,
because both release point #1 and #3 may also be to either wastewater or landfills, EPA assumes the
same quantity of 72 g of 1,4-dioxane per day from the SHEDs-HT model may be released to unknown
landfills for this OES. EPA notes that these 72 g is either entirely to wastewater or landfill or some split
between the two media. The 72 g is not to both wastewater and landfill because that would double count
the release,
EPA's calculation of land releases for this OES, including all calculation inputs and assumptions, can be
found in 1,4-Dioxane Supplemental Information File: Environmental Releases to Landfor all OES
Except Disposal ( BP A. 2023k).
Page 257 of 484
-------
6962
6963
6964
6965
6966
6967
6968
6969
6970
6971
6972
6973
6974
6975
6976
6911
6978
6979
6980
6981
6982
6983
6984
6985
6986
6987
6988
6989
6990
6991
6992
6993
6994
6995
6996
6997
6998
6999
7000
7001
7002
7003
7004
7005
7006
7007
7008
7009
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Dish Soap
EPA did not find any directly applicable GS/ESD or literature sources for this OES; however, EPA
estimated land releases using the SHEDs-HT modeling conducted for the one case study location
(Liverpool OH) and the assumptions described here. EPA expects that the main release points from the
use of dish soaps are from:
1. Disposal of empty containers containing residual dish soap, and
2. Cleaning and rinsing of dishes.
EPA expects that empty containers may be rinsed out in sinks or disposed of without rinsing, such that
releases may be to wastewater or landfill. Further, EPA expects that the entire amount of dish soap used
for cleaning dishes is rinsed down the drain of sinks during the cleaning and rinsing process. EPA uses
the SHEDs-HT modeled estimated of wastewater discharges for this OES (64 g 1,4-dioxane per day for
Liverpool OH) and back calculates a 1,4-dioxane use rate using the EPA/OPPT Small Container
Residual Model central tendency and high-end loss fractions and an assumption of 100 percent release.
Using this back-calculated 1,4-dioxane use rate, EPA then applied the EPA/OPPT Small Container
Residual Model to estimate land releases for the Liverpool OH case study. EPA expects that this land
release is to unknown landfills. EPA's calculation of land releases for this OES, including all calculation
inputs and assumptions, can be found in 1,4-Dioxane Supplemental Information File: Environmental
Releases to Landfor all OES Except Disposal ( A. 2023k).
Dishwasher Detergent
EPA did not find any directly applicable GS/ESD or literature sources for this OES; however, EPA
estimated land releases using the SHEDs-HT modeling conducted for the one case study location
(Liverpool OH) and the same assumptions as that described for the dish soap OES above. EPA's
calculation of land releases for this OES, including all calculation inputs and assumptions, can be found
in 1,4-Dioxane Supplemental Information File: Environmental Releases to Landfor all OES Except
Disposal ( )23k).
Laundry Detergent
EPA estimated land releases for facilities within the Laundry Detergent OES using the OECD ESD on
Industrial and Institutional Laundries (OI ) and Monte Carlo modeling. The ESD on
Industrial and Institutional Laundries was rated "medium" during EPA's systematic review process. The
use of Montel Carlo modeling allows for variation of calculation input parameters such that a
distribution of environmental releases can be calculated, from which EPA can estimate the 50th and 95th
percentile releases. An explanation of this modeling approach is included in Appendix E.12.
Paints and Floor Lacquer
EPA estimated land releases for facilities within the Paints and Floor Lacquers OES using the OECD
ESD on Coating Application via Spray-Painting in the Automotive Refinishing Industry (OECD.
201 la). The ESD was rated "medium" during EPA's systematic review process.
As described in the process description in Appendix F.4.7, 1,4-dioxane was identified by a public
comment as present in automotive refinishing products, thereby allowing EPA to identify the above
ESD as the most applicable GS/ESD available. This ESD indicates that releases are expected from
1. Releases to incineration or landfill from container cleaning/disposal,
2. Releases to incineration or landfill from equipment cleaning,
3. Releases to incineration or landfill from over sprayed coating that is captured by emission
controls, and
Page 258 of 484
-------
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
7026
7027
7028
7029
7030
7031
7032
7033
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
4. Releases to stack air from over sprayed coating that is not captured by emission controls.
Based on the GS, release points #1 through 3 have the potential for land releases. To estimate these
releases, EPA used the equations specified in the ESD (OB ). Apart from weight fraction in
coatings (see Appendix F.4.7), EPA did not find any data specific to 1,4-dioxane in this OES. Therefore,
the calculation of releases using this GS are for a "generic site," using the default input parameter values
from the ESD.
Using this methodology, EPA calculated the low-end and high-end land releases for this OES, which are
expected to be to unknown landfills per the ESD (OECD. 201 la). The low- and high-end estimates are
based on the low- and high-end calculation input parameter defaults from the ESD. EPA's calculation of
land releases for this OES, including all calculation inputs and assumptions, can be found in 1,4-
Dioxane Supplemental Information File: Environmental Releases to Landfor all OES Except Disposal
(U.S. EPA. 2023kY
Hydraulic Fracturing
EPA estimated land releases for facilities within the Hydraulic Fracturing OES using the Draft OECD
ESD on Hydraulic Fracturing ( 22d) and Monte Carlo modeling. The Draft ESD on
Hydraulic Fracturing was rated "high" during EPA's systematic review process. The use of Montel
Carlo modeling allows for variation of calculation input parameters such that a distribution of
environmental releases can be calculated, from which EPA can estimate the 50th and 95th percentile
releases. An explanation of this modeling approach is included in Appendix E.13.
E.4.3 Land Release Estimates Summary
A summary of industrial and commercial land releases estimated using the above methods is presented
in Table Apx E-5 below.
Page 259 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
7034 Table Apx E-5. Summary of Daily Industrial and Commercial Land Release Estimates for 1,4-Dioxane
Number of
Estimated Daily Release Range
across Sites
Estimated
Release
Overall Data
OES
Type of Land Release
Facilities with
Releases"
(kg/site-day)
Frequency
Range
(days)''
Quality
Determination
Sources'
Min
Max
Manufacturing
Land (all types)
1
0
250
Medium
TRI
Import and
Repackaging
Land (all types)
1
0
250
Medium
TRI
Industrial Uses
Land (all types)
12
0
227 (annually)
250
Medium
TRI
Functional Fluids
Land (all types)
2
0
0
247
Medium
TRI
(Open-System)
Functional Fluids
All
Assessed as a part of Industrial Uses OES
N/A
N/A
(Closed-System)
Laboratory
Chemical
Land (unknown type)
132
0
489 (annually)
250
High
GSd
Film Cement
Land (unknown type)
211
0.0035
(annually)
0.037
(annually)
250
High
Process
information®
Spray Foam
Application
Land (unknown type)
1,553,559
0.032 (annually)
0.047
(annually)
3
Medium
GS^
Printing Inks (3D)
Fugitive Air, Stack
Assessed as a part of Industrial Uses OES
250
N/A
N/A
Air, and Land (all
types)
Dry Film
Lubricant
Land (hazardous waste
landfill)
8
0
188 (annually)
48
High
Process
information®
Page 260 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Land Release
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Disposal
Land (RCRA Sub C
landfill)
18
0
7,307
(annually)
250
Medium
TRI
Land (Underground
injection)
18
0
331,980
(annually)
Medium
TRI
Land (Non-RCRA
landfills)
18
0
890 (annually)
Medium
TRI
Land (all other types)
18
0
0
Medium
TRI
Textile Dye
Land (unknown
landfill type) or
POTW (unknown
partitioning)
783
2.09E-07
9.72E-05
31 to 295
Medium
ESDg and
Modeling''
Antifreeze
Land (unknown
landfill)
84,383
3.75E-07
(annually)
0.029
(annually)
250
High
Process
information® and
Modeling''
Surface Cleaner
Land (unknown
landfill) or POTW
Unknown
18™
250
High
SHEDS-HT',
Process
information®
Modeling''
Dish Soap
Land (unknown
landfill)
Unknown
0.048 (annual
value for all
sites in
Liverpool OH
case study)
0.097 (annual
value for all
sites in
Liverpool OH
case study)
250
High
SHEDS-HT',
Process
information®
Modeling''
Dishwasher
Detergent
Land (unknown
landfill)
Unknown
1.08E-03
(annual value for
all sites in
Liverpool, OH
case study)
2.17E-03
(annual value
for all sites in
Liverpool, OH
case study)
250
High
SHEDS-HT',
Process
information®
Modeling''
Laundry Detergent
(Institutional) -
liquid detergents
Land (unknown
landfill), incineration,
95,533
4.05E-12
3.95E-05
250 to 365
Medium
ESD! and
Modeling''
Page 261 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Land Release
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
or POTW (unknown
partitioning)
Laundry Detergent
(Institutional) -
powder detergents
Land (unknown
landfill), incineration,
or POTW (unknown
partitioning)
95,533
5.36E-08
0.0018
250 to 365
Medium
ESD! and
Modeling''
Laundry Detergent
(Industrial) -
liquid detergents
Land (unknown
landfill), incineration,
or POTW (unknown
partitioning)
2,453
4.78E-12
1.46E-04
20 to 365
Medium
ESD! and
Modeling''
Laundry Detergent
(Industrial) -
powder detergents
Land (unknown
landfill), incineration,
or POTW (unknown
partitioning)
2,453
2.92E-11
3.92E-04
20 to 365
Medium
ESD! and
Modeling''
Paints and Floor
Lacquer
Land (unknown
landfill)
33,648
3.04E-06
(annually)
0.010
(annually)
250
Medium
ESD7
Polyethylene
Terephthalate
(PET) Byproduct
Land (Land treatment)
13
0
45.4 (annually)
Medium
TRI
Land (Non-RCRA
landfills)
13
0
0.10 (annually)
250
Medium
TRI
Land (all other types)
13
0
0
Medium
TRI
Ethoxylation
Process Byproduct
Land (underground
injection)
8
0
197,714
(annually)
250
Medium
TRI
Land (all other types)
8
0
0
Medium
TRI
Page 262 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Land Release
Number of
Facilities with
Releases"
Estimated Daily Release Range
across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Hydraulic
Fracturing
Surface water,
incineration, or landfill
(unknown
partitioning)
411
3.61E-10
4.59
1 to 72
Medium
ESD' and
Modeling11
Land (underground
injection)
411
5.35E-09
108
Medium
ESD' and
Modeling''
Recycle/Reuse (48%),
underground injection
(43%), Surface water
(6%), or land (3%)
411
1.85E-10
1.12
Medium
ESD'' and
Modeling''
"Where available. EPA used 2013-2019 TRI (U.S. EPA, 2022g) and 2013-2019 DMR (U.S. EPA, 2022c) data to provide a basis to estimate the number of sites
using 1,4-dioxane within a COU.
h Where available. EPA used the December 2020 Final Risk Evaluation for 1,4-Dioxane (U.S. EPA, 2020c). generic scenarios, and emission scenario documents
to provide a basis to estimate the number of release days of 1,4-dioxane within a COU.
c Narrative descriptions of all release estimate sources are provided in Appendix E.4.2.
d The generic scenario used for this COU is the GS on Use of Laboratory Chemicals (U.S. EPA. 2022h).
'' For this COU, EPA used process information, which is further described in Appendix E.4.2.
' The generic scenario used for this COU is the GS on Application of Sprav Polvurethane Foam Insulation (U.S. EPA. 2018b).
8 The emission scenario document used for this COU is the ESD on Textile Dves (OECD. 2017).
h For this COU, EPA used various models and literature for model input parameters as described in Appendix E.4.2.
1 The emission scenario document used for this COU is the ESD on Industrial and Institutional Laundries (OECD, 201 lb).
1 The emission scenario document used for this COU is the ESD on Coating Application via Sprav Painting in the Automotive Refinishing Industrv (OECD.
2011a).
k The emission scenario document used for this COU is the Draft ESD on Hvdraulic Fracturing (U.S. EPA, 2022d).
' EPA used the down the drain water release estimates from the SHEDs-HT model for the Liverpool OH case study (see Section 2.1.1.2) to estimate air and land
releases by back calculating 1,4-dioxane use rates and applying loss fractions for air and land releases using literature and standard models described in Appendix
E.4.2.
m A single annual value was provided for all sites in the Liverpool, OH case study.
7035 E.4.4 Summary of Weight of the Scientific Evidence Conclusions in Land Release Estimates
7036 TableApx E-6 provides a summary of EPA's weight of the scientific evidence conclusions in its land release estimates for each of the
7037 Occupational Exposure Scenarios assessed. Detailed descriptions of non-OES specific strengths, limitations, assumptions, and uncertainties
7038 {e.g., general limitations for TRI, DMR, etc.) are provided in Appendix E.6.
Page 263 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
7039 Table Apx E-6. Summary of Weight of the Scientific Evidence Conclusions in Land Release Estimates by PES
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Manufacturing
Land releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the low number of data points, lack of variability (only one year of data
used), uncertainty in the accuracy of reported releases, and the limitations in representativeness to all sites because TRI
may not capture all relevant sites. Additionally, EPA could not estimate the number of release days per year associated
with land releases. Based on this information, EPA has concluded that the weight of the scientific evidence for this
assessment is moderate to robust and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Import and Repackaging
Land releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the low number of data points, lack of variability (only one year of data
used), uncertainty in the accuracy of reported releases, and the limitations in representativeness to all sites because TRI
may not capture all relevant sites. Additionally, the land release assessment is based on one reporting site that reported no
land releases and EPA did not have additional sources to estimate land releases for other sites in this OES. Additionally,
EPA could not estimate the number of release days per year associated with land releases. Based on this information, EPA
has concluded that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Industrial Uses
Land releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the lack of variability (only one year of data used), uncertainty in the
accuracy of reported releases, uncertainty in EPA's use of Form A submissions, and the limitations in representativeness to
all sites because TRI may not capture all relevant sites. Some facilities within this OES reported to TRI using a Form A,
which does not include any details on chemical release quantities. When a facility has submitted a Form A, there is no way
to discern the quantity released. Therefore, where facilities reported to TRI with a Form A, EPA used the Form A threshold
for total releases of 500 lb/year for each release media; however, there is uncertainty in this because the actual release
quantity is unknown. Furthermore, the threshold represents an upper limit on total releases from the facility; therefore,
assessing releases at the threshold value may overestimate actual releases from the facility. Additionally, EPA could not
estimate the number of release days per year associated with land releases. Based on this information, EPA has concluded
Page 264 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible estimate of
releases in consideration of the strengths and limitations of reasonably available data.
Functional Fluids (Open-
System)
Land releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the low number of data points, lack of variability (only one year of data
used), uncertainty in the accuracy of reported releases, and the limitations in representativeness to all sites because TRI
may not capture all relevant sites. The land release assessment is based on two reporting sites that both reported no land
releases and EPA did not have additional sources to estimate land releases for sites in this OES. Additionally, EPA could
not estimate the number of release days per year associated with land releases. Based on this information, EPA has
concluded that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Functional Fluids (Closed-
System)
No data was available to estimate releases for this OES. For the land release assessment, EPA grouped this OES with the
Industrial Uses OES because the sources of release are expected to be similar between these OESs. Factors that increase
the strength of evidence for this OES are that TRI has a medium overall data quality determination and consistency within
the dataset (all reporters use the same or similar reporting forms). Factors that decrease the strength of evidence for this
OES are that the Industrial Releases OES release data are use as surrogate for this OES, uncertainty in the accuracy of
reported releases, limitations in representativeness to all sites because TRI may not capture all relevant sites, and lack of
variability (only one year of data used). Refer to the Industrial Uses OES discussion for additional discussion. Based on
this information, EPA has concluded that the weight of the scientific evidence for this assessment is slight and provides a
plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Laboratory Chemicals
Land releases are assessed using the Draft GS on Use of Laboratory Chemicals. Factors that increase the strength of
evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to surrogate), the Draft GS
on Use of Laboratory Chemicals has a high overall data quality determination, and the low level of uncertainty in the data.
Factors that decrease the strength of the evidence for this OES include the that the GS has not been peer-reviewed,
uncertainty in the representativeness of the GS towards all sites in this OES, and a lack of variability in the analysis.
Specifically, because the default values in the ESD are generic, there is uncertainty in the representativeness of generic site
estimates of actual releases from real-world sites that use 1,4-dioxane. Another uncertainty is lack of consideration for
release controls. The ESD assumes that all activities occur without any release controls. Actual releases may be less than
estimated if facilities utilize pollution control methods, contributing to uncertainty. Based on this information, EPA has
concluded that the weight of the scientific evidence for this assessment is slight to moderate and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Film Cement
Land releases are assessed using process information from the Final Risk Evaluation for 1,4-Dioxane and EPA/OPPT
models. Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant to the
Page 265 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
OES (as opposed to surrogate), the underlying data sources for the process information have a high overall data quality
determination, and the low level of uncertainty in the data because the process information comes directly from actual
users of 1,4-dioxane in film cement. Factors that decrease the strength of the evidence for this OES include uncertainty in
the representativeness of evidence to all sites in this OES and a lack of variability in the input parameters for the used
models. Specifically, the process information for the production and use of film cement is based on information from three
use sites, one from Australia and two from the U.S. Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is slight to moderate and provides a plausible estimate of releases in consideration
of the strengths and limitations of reasonably available data.
Spray Foam Application
Land releases are assessed using the GS on Application of Spray Polyurethane Foam Insulation. Factors that increase the
strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to surrogate),
the underlying data sources for the process information have a medium overall data quality determination, and a low level
of uncertainty in the data. Factors that decrease the strength of the evidence for this OES include uncertainty in the
representativeness of the GS to all sites since it is generic and not specific to sites that use 1,4-dioxane and a lack of
variability. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is
slight to moderate and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Printing Inks (3D)
No data was available to estimate releases for this OES. For the land release assessment, EPA grouped this OES with the
Industrial Uses OES because the sources of release are expected to be similar between these OESs. Factors that increase
the strength of evidence for this OES are that TRI has a medium overall data quality determination and consistency within
the dataset (all reporters use the same or similar reporting forms). Factors that decrease the strength of evidence for this
OES are that the Industrial Releases OES release data are use as surrogate for this OES, uncertainty in the accuracy of
reported releases, limitations in representativeness to all sites because TRI may not capture all relevant sites or smaller
commercial 3D printing uses, and lack of variability (only one year of data used). Refer to the Industrial Uses OES
discussion for additional discussion. Based on this information, EPA has concluded that the weight of the scientific
evidence for this assessment is slight and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Dry Film Lubricant
Land releases are assessed using process information from the Final Risk Evaluation for 1,4-Dioxane. Factors that increase
the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to
surrogate), that the underlying data sources for the process information have a high overall data quality determination, and
a low level of uncertainty in the data because the process information comes directly from an actual user of 1,4-dioxane in
dry film lubricants. Factors that decrease the strength of the evidence for this OES include uncertainty in the
representativeness of evidence to all sites and a lack of variability. Based on this information, EPA has concluded that the
weight of the scientific evidence for this assessment is slight to moderate and provides a plausible estimate of releases in
consideration of the strengths and limitations of reasonably available data.
Page 266 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Disposal
Land releases are assessed using reported discharges from 2013-2019 TRI. Factors that increase the strength of evidence
for this OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium
overall data quality determination, high number of data points, and consistency within the dataset (all reporters use the
same or similar reporting forms). Additionally, EPA included seven years of TRI data in the analysis, which increases the
variability of the dataset. A strength of TRI data is that TRI compiles the best readily available release data for all reporting
facilities. Factors that decrease the strength of the evidence for this OES include uncertainty in the accuracy of reported
releases, uncertainty in EPA's use of Form A submissions, and the limitations in representativeness to all sites because TRI
may not capture all relevant sites. Some facilities within this OES reported to TRI using a Form A, which does not include
any details on chemical release quantities. When a facility has submitted a Form A, there is no way to discern the quantity
released. Therefore, where facilities reported to TRI with a Form A, EPA used the Form A threshold for total releases of
500 lb/year for each release media; however, there is uncertainty in this because the actual release quantity is unknown.
Furthermore, the threshold represents an upper limit on total releases from the facility; therefore, assessing releases at the
threshold value may overestimate actual releases from the facility. Based on this information, EPA has concluded that the
weight of the scientific evidence for this assessment is moderate to robust and provides a plausible estimate of releases in
consideration of the strengths and limitations of reasonably available data.
Textile Dye
Land releases are assessed using Monte Carlo modeling with information from the ESD on Textile Dyes. Factors that
increase the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to
surrogate), the ESD on Textile Dyes has a medium overall data quality determination and was peer reviewed, the high
number of data points (simulation runs), consistency within the dataset, and full distributions of input parameters. The
Monte Carlo modeling accounts for the entire distribution of input parameters, calculating a distribution of potential
release values that represents a larger proportion of sites than a discrete value. Factors that decrease the strength of the
evidence for this OES include uncertainties and limitations in the representativeness of the estimates for sites that
specifically use 1,4-dioxane because the default values in the ESD are generic. Another uncertainty is lack of consideration
for release controls. The ESD assumes that all activities occur without any release controls. Actual releases may be less
than estimated if facilities utilize pollution control methods, contributing to uncertainty. Based on this information, EPA
has concluded that the weight of the scientific evidence for this assessment is moderate and provides a plausible estimate
of releases in consideration of the strengths and limitations of reasonably available data.
Antifreeze
Land releases are assessed using the OECD ESD on Chemical Additives used in Automotive Lubricants, the EPA MRD on
Commercial Use of Automotive Detailing Products, and EPA/OPPT models. Factors that increase the strength of evidence
for this OES are that the ESD and MRD used have high overall data quality determinations and consistency within the
sources used. Factors that decrease the strength of the evidence for this OES include that the ESD and MRD are not
directly applicable to antifreeze uses (used as surrogate), uncertainty in the representativeness of the ESD and MRD to all
sites and sites that specifically use 1,4-dioxane since these documents contain generic values, and a lack of variability.
Additionally, EPA scaled up a consumer antifreeze use rate to a commercial use rate based on information in the ESD and
MRD, which increases uncertainty. Based on this information, EPA has concluded that the weight of the scientific
Page 267 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
evidence for this assessment is slight to moderate and provides a plausible estimate of releases in consideration of the
strengths and limitations of reasonably available data.
Surface Cleaner
Land releases are assessed using SHEDS-HT modeled water releases in conjunction with the Draft GS on Furnishing
Cleaning. Factors that increase the strength of evidence for this OES include that the release estimates are directly relevant
to the OES (as opposed to surrogate), that the Draft GS used has a high overall data quality determination, and variability
in the model input parameters. Factors that decrease the strength of the evidence for this OES include uncertainty in the
representativeness to all sites because the SHEDS-HT estimate is based on one case study for Liverpool, OH and because
the estimate is not site-specific (the release estimate is a total for all sites in Liverpool, OH). Additionally, the Draft GS
describes potential release points for this OES, identifying releases that may be to either water or land depending on site
practices (e.g., surface cleaner may be rinsed down drains or wiped off with rags that are disposed of as trash). Because
there is no information to determine the quantity released specifically to land, EPA assumed that the entire quantity
modeled to water with the SHEDS-HT model may also be released to land, which introduces uncertainty. Based on this
information, EPA has concluded that the weight of the scientific evidence for this assessment is slight and provides a
plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Dish Soap
Land releases are assessed using the SHEDS-HT modeled water releases, the expected sources of release for this OES, and
EPA/OPPT models. Factors that increase the strength of evidence for this OES include that the release estimates are
directly relevant to the OES (as opposed to surrogate) and variability in the model input parameters. Factors that decrease
the strength of the evidence for this OES include uncertainty in the representativeness to all sites because the SHEDS-HT
estimate is based on one case study for Liverpool, OH and because the estimate is not site-specific (the release estimate is a
total for all sites in Liverpool, OH). Additionally, based on the quantity released to water and the expected loss fraction for
water releases, EPA back-calculated a 1,4-dioxane use rate and applied the expected land release loss fraction to estimate
land releases, which introduces uncertainty. Based on this information, EPA has concluded that the weight of the scientific
evidence for this assessment is slight and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Dishwasher Detergent
EPA used the same approach as described above for the dish soap OES to estimate land releases for the dishwasher
detergent OES.
Laundry Detergent
Land releases are assessed using Monte Carlo modeling with information from the ESD on Industrial and Institutional
Laundries. Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant to
the OES (as opposed to surrogate), that the ESD on Industrial and Institutional Laundries has a medium overall data
quality determination and was peer reviewed, there are high number of data points (simulation runs), consistency within
the dataset, and full distributions of input parameters. The Monte Carlo modeling accounts for the entire distribution of
input parameters, calculating a distribution of potential release values that represents a larger proportion of sites than a
discrete value. Additionally, EPA was able to separately estimate releases for industrial and institutional laundry settings.
Factors that decrease the strength of the evidence for this OES include uncertainties and limitations in the
representativeness of the estimates for sites that specifically use 1,4-dioxane because the default values in the ESD are
Page 268 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
generic. Another uncertainty is lack of consideration for release controls. The ESD assumes that all activities occur without
any release controls. Actual releases may be less than estimated if facilities utilize pollution control methods. Based on this
information, EPA has concluded that the weight of the scientific evidence for this assessment is moderate and provides a
plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Paint and Floor Lacquer
Land releases are assessed using OECD ESD on Coating Application via Spray-Painting in the Automotive Refinishing
Industry. Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant to
the OES (as opposed to surrogate), the ESD has a medium overall data quality determination and has been peer reviewed,
consistency within the sources used, and a low amount of uncertainties. Factors that decrease the strength of the evidence
for this OES include a lack of variability and uncertainty in the representativeness of the ESD to all sites and sites that
specifically use 1,4-dioxane since the ESD is generic. Based on this information, EPA has concluded that the weight of the
scientific evidence for this assessment is slight to moderate and provides a plausible estimate of releases in consideration
of the strengths and limitations of reasonably available data.
PET Byproduct
Land releases are assessed using reported discharges from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include uncertainty in the accuracy of reported releases, lack of variability (only one
year of data used), and the limitations in representativeness to all sites because TRI may not capture all relevant sites. The
land release assessment is based on 13 reporting sites, 11 of which reported no land releases. EPA did not have additional
sources to estimate land releases for site in this OES that may not report to TRI. Based on this information, EPA has
concluded that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Ethoxylation Process
Byproduct
Land releases are assessed using reported discharges from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include uncertainty in the accuracy of reported releases, lack of variability (only one
year of data used), and the limitations in representativeness to all sites because TRI may not capture all relevant sites. The
land release assessment is based on eight reporting sites, seven of which reported no land releases. EPA did not have
additional sources to estimate land releases for site in this OES that may not report to TRI. Based on this information, EPA
has concluded that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Hydraulic Fracturing
Land releases are assessed using Monte Carlo modeling with information from the Draft ESD on Hydraulic Fracturing and
FracFocus 3.0. Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant
to the OES (as opposed to surrogate), that the Draft ESD on Hydraulic Fracturing and FracFocus 3.0 have medium overall
Page 269 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
data quality determinations, high number of data points (simulation runs), consistency within the dataset, and full
distributions of input parameters. The Monte Carlo modeling accounts for the entire distribution of input parameters,
calculating a distribution of potential release values that represents a larger proportion of sites than a discrete value. Factors
that decrease the strength of the evidence for this OES include that the Draft ESD has not been peer reviewed and the
uncertainties and limitations in the representativeness of the estimates for sites that specifically use 1,4-dioxane because
the default values from the Draft ESD on Hydraulic Fracturing. Another uncertainty is lack of consideration for release
controls. The ESD assumes that all activities occur without any release controls. Actual releases may be less than estimated
if facilities utilize pollution control methods, contributing to uncertainty. Based on this information, EPA has concluded
that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible estimate of
releases in consideration of the strengths and limitations of reasonably available data.
7040
Page 270 of 484
-------
7041
7042
7043
7044
7045
7046
7047
7048
7049
7050
7051
7052
7053
7054
7055
7056
7057
7058
7059
7060
7061
7062
7063
7064
7065
7066
7067
7068
7069
7070
7071
7072
7073
7074
7075
7076
7077
7078
7079
7080
7081
7082
7083
7084
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.5 Air Release Assessment
This section describes EPA's methodology for estimating daily fugitive and stack air emissions from
industrial and commercial facilities manufacturing, processing, or using 1,4-dioxane. Facilities report air
emissions to the TRI. EPA used 2019 TRI (I v « « 4.. 2022e) data to estimate daily air emissions for
each OES where available; however, EPA did not have these data for every OES. For OES without TRI
data, EPA used alternate assessment approaches to estimate air emissions. These approaches are
described below.
E.5.1 Assessment Using TRI
EPA found 2019 TRI data for facilities within the following OESs:
• Manufacturing
• Import and repackaging
• Industrial uses
• Functional fluids (open-system)
• Disposal
• PET byproduct
• Ethoxylation byproduct
The 2019 TRI data were rated "medium" in EPA's systematic review process. EPA estimated daily air
emissions using TRI data for these OESs, with the following general steps as described in the rest of this
section.
1. Collect air emission data from 2019 TRI data,
2. Map air emission data to occupational exposure scenarios,
3. Estimate the number of facility operating days per year, and
4. Estimate daily air emissions and prepare a summary of the air emissions for each OES.
Note that EPA compared the TRI data used to estimate air releases for the PET byproduct OES in this
risk evaluation to information from a life cycle analysis on the PET manufacturing process in Appendix
E.6.
Step 1: Collect Air Emission Data TRI
The first step in the methodology for estimating air emissions was to obtain 2019 TRI data for the
chemical from EPA's Basic Plus Data Files. TRI requires U.S. facilities in various industry sectors to
report the annual release volumes to the environment through air emissions, water discharges, and land
disposal, and/or managed through recycling, energy recovery, and treatment, including by off-site
transfers. TRI reporters may report either with a Form R or a Form A. Facilities must report with a Form
R, which requires reporting of release quantities and uses/sub-uses of the chemical, among other
information, unless they meet the alternate threshold requirements for submitting a Form A.
Specifically, facilities may submit a Form A if the volume of chemical manufactured, processed, or
otherwise used does not exceed 1,000,000 lb per year (lb/year) and the total annual reportable releases
do not exceed 500 lb/year. Facilities do not need to report release quantities or uses/sub-uses on Form A.
EPA included both TRI reporting Form R and TRI reporting Form A submissions in the air release
assessment.
Air emissions in TRI are reported separately for stack air and fugitive air and always occur on-site at the
facility. Where sites reported to 2019 TRI with Form A, EPA used the Form A threshold for total
releases of 500 lb/year. EPA used the entire 500 lb/year for both the fugitive and stack air release
Page 271 of 484
-------
7085
7086
7087
7088
7089
7090
7091
7092
7093
7094
7095
7096
7097
7098
7099
7100
7101
7102
7103
7104
7105
7106
7107
7108
7109
7110
7111
7112
7113
7114
7115
7116
7117
7118
7119
7120
7121
7122
7123
7124
7125
7126
7127
7128
7129
7130
7131
7132
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
estimates; however, since this threshold is for total site releases, these 500 lb/year are attributed either to
fugitive air or stack air for this analysis—not both (to avoid double counting the releases and exceeding
the total release threshold for Form A).
Step 2: Map Air Emission Data to Occupational Exposure Scenarios
In the next step of air release assessment, EPA mapped the chemical's 2019 TRI data to the 1,4-dioxane
OES. EPA used the following procedure to map 2019 TRI data to OES:
1. Compile TRI uses/sub-uses: EPA first compiled all the reported TRI uses/sub-uses for each
facility into one column.
2. Map TRI uses/sub-uses to Chemical Data Reporting (CDR) IFC codes: EPA then mapped the
TRI uses/sub-uses for each facility to one or more 2016 CDR Industrial Function Category (IFC)
codes using the TRI-to-CDR Use Mapping crosswalk (see Appendix E.9).
3. Map OES to CDR IFC codes: EPA then mapped each COU/OES combination to a 2016 CDR
IFC code and description. The basis for this mapping was generally the COU category and
subcategory.
4. Map TRIfacilities to an OES: Using the CDR IFC codes from Step 2 and the COU-CDR
Mapping from Step 3, EPA mapped each TRI facility to an OES. EPA's rationale for the OES
determination is generally described below.
o In some cases, the facility mapped to only one OES and the mapping appeared to be
correct given the facility name and NAICS code. For these, the OES as mapped from
Steps 2 and 3 was used without adjustment,
o There were instances where a facility mapped to multiple OESs which required some
engineering judgement to identify a primary OES. EPA documented the rationale for
these determinations for each facility in 1,4-Dioxane Supplemental Information File:
Environmental Releases to Air ( 023]). In summary, these determinations
were made with the following considerations:
• Industry and NAICS codes reported in TRI (e.g., for a facility that reported TRI
uses for both waste treatment and ancillary use, EPA assigned the Disposal OES
if the NAICS code was 562211, Hazardous Waste Treatment and Disposal);
• Internet research of the types of products manufactured at the facility (e.g., if a
facility's website indicates the facility manufactures PET, the facility is likely to
produce 1,4-dioxane as a byproduct in PET manufacturing); and
• Grouping of similar OES (e.g., for facilities that reported the sub-use of chemical
processing aid, process solvent, or processing as a reactant), EPA assigned the
Industrial Uses OES because this includes multiple processes as described in the
2020 RE ( 020cV).
o In some cases, EPA identified that there were instances where the preliminary OES
mapping from the TRI use/sub-use - CDR IFC code required re-mapping. This re-
mapping is a result of limitations of the TRI-to-CDR Use Mapping crosswalk. For
example, the crosswalk maps the TRI use/sub-use of "Otherwise Use as Manufacturing
Aid (Other)" to only CDR IFC codes U013 (closed-system functional fluids) and U023
(plating agents and surface treating agents); however, this TRI use/sub-use may
encompass multiple other uses that are not captured in these CDR IFC codes. In these
cases, EPA reviewed the reported NAICS codes and conducted internet research on the
types of products manufactured at the facility to determine the likely OES.
o Additionally, EPA reviewed 2016 CDR (U.S. EPA. 2016b) for sites that reported
manufacturing (including importing) of 1,4-dioxane. If the sites that reported to 2016
CDR also reported in 2019 TRI, EPA assigned the OES according to 2016 CDR. Note
Page 272 of 484
-------
7133
7134
7135
7136
7137
7138
7139
7140
7141
7142
7143
7144
7145
7146
7147
7148
7149
7150
7151
7152
7153
7154
7155
7156
7157
7158
7159
7160
7161
7162
7163
7164
7165
7166
7167
7168
7169
7170
7171
7172
7173
7174
7175
7176
7177
7178
7179
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
that some sites that reported to 2019 TRI may not be in 2016 CDR (e.g., sites that
manufacture the chemical as a byproduct). In these cases, EPA determined the OES using
only the above bulleted steps.
5. Form A's: For Form A submissions, there were no reported TRI uses/sub-uses. To determine the
COU for these facilities, EPA used 2016 CDR as described above, the NAICS codes, and
internet searches to determine the type of products and operations at the facility.
The specific rationale for the OES mapping for each facility is described in 1,4-Dioxane Supplemental
Information File: Environmental Releases to Air (U.S. EPA. 2023D.
Step 3: Estimate the Number of Facility Operating Days per Year
EPA then estimated the number of operating days (days/year) for each facility reporting air emissions to
TRI. For the OES that were included in the Final Risk Evaluation for 1,4-Dioxane ( 2020c\
EPA used the number of operating days from that risk evaluation. For the additional OES included in
this supplemental risk evaluation, EPA estimated the number of operating days using the methodology
described in Appendix E.2.
Step 4: Estimate Daily Air Emissions and Summarize Air Emissions for each OES
The final step was to prepare a summary of the fugitive and stack releases. For each OES and facility
mapped to that OES, EPA summarized the annual fugitive and stack air emissions reported in 2019 TRI
and daily fugitive and stack air emissions that EPA estimated by dividing the annual emissions by the
number of operating days determined for the OES in Step 3. EPA also summarized site information,
including site identity, city, state, zip code, TRI facility ID, and Facility Registry Service (FRS) ID
because the subsequent exposure modeling is site and location specific. Latitude and longitude
coordinates are included in 1,4-Dioxane Supplemental Information File: Environmental Releases to Air
(1 c. i i1 \ 2023D but not in the summary tables.
To accompany the summary table for each OES, EPA also provided any reasonably available
information on the release duration and pattern, which are needed for the exposure modeling. Release
duration is the expected amount of time per day during which the air releases may occur. Release pattern
is the temporal variation of the air release, such as over consecutive days throughout the year, over
cycles that occur intermittently throughout the year, or in a puff/instantaneous release that occurs over a
short duration. The TRI dataset does not include release pattern or duration; therefore, EPA used
information from models or literature, where available. For release pattern, EPA provided the number of
release days with the associated basis as described in Step 3. However, for most OES, no information
was found on release duration and pattern. In such cases, EPA listed the release duration and pattern as
"unknown."
EPA's summary of air releases for each OES is included in 1,4-Dioxane Supplemental Information File:
Environmental Releases to Air ( 323i).
E.5.2 Assessment for OESs without TRI
EPA did not find TRI data for any of the years included in this analysis for the following OESs:
• Functional fluids (closed-systems)
• Laboratory chemicals
• Film cement
• Spray polyurethane foam
• 3D printing
• Dry film lubricant
Page 273 of 484
-------
7180
7181
7182
7183
7184
7185
7186
7187
7188
7189
7190
7191
7192
7193
7194
7195
7196
7197
7198
7199
7200
7201
7202
7203
7204
7205
7206
7207
7208
7209
7210
7211
7212
7213
7214
7215
7216
7217
7218
7219
7220
7221
7222
7223
7224
7225
7226
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• Textile dye
• Antifreeze
• Surface cleaner
• Dish soap
• Dishwasher detergent
• Laundry detergent
• Paints and floor lacquer
• Hydraulic fracturing
For these OESs, EPA estimated air emissions by using various modeling approaches, including the use
of surrogate TRI data and data from literature, GSs, and ESDs. EPA's assessment of air emissions for
each of these OESs is described below.
Functional Fluids (Closed-Systems)
Air emission data were not available for this OES and EPA did not find any information to model air
emissions for this OES using literature, GSs, or ESDs. EPA expects that the sources of release for this
OES to be similar to those for the Industrial Uses OES, based on the process information in the Final
Risk Evaluation for 1,4-Dioxane (U. 2020c). Therefore, EPA grouped the air release assessment
for Functional Fluids (Closed-Systems) into the OES for Industrial Uses. However, there is uncertainty
in the assumption of similar release sources between these OESs.
Laboratory Chemicals
EPA estimated air emissions for facilities within the Laboratory Chemicals OES using the Draft GS on
Use of Laboratory Chemicals ( 22h). The Draft GS on Use of Laboratory Chemicals was
rated "high" during EPA's systematic review process.
The GS indicates that there are eight release points:
1. Release to air from transferring volatile chemicals from transport containers
2. Release to air, water, incineration, or landfill from transferring solid powders
3. Release to water, incineration, or land from cleaning or disposal of transport containers
4. Release to air from cleaning containers used for volatile chemicals
5. Labware equipment cleaning residuals released to water, incineration, or landfill
6. Release to air during labware equipment cleaning for volatile chemicals
7. Release to air from laboratory analyses for volatile chemicals
8. Release to water, incineration, or landfill from laboratory waste disposal
Based on the GS, release points #1, 2, 4, 6, and 7 have the potential for air emissions; however, release
point #2 is not applicable because 1,4-dioxane is not a solid powder. To estimate the remaining air
releases, EPA used the equations specified in the Draft GS ( 22h). EPA did not find any
data specific to 1,4-dioxane in this OES. Therefore, the calculation of releases using this GS are for a
"generic site," using the default input parameter values from the GS.
Using this methodology, EPA calculated the "typical" and "worst-case" air emissions for this OES.
These characterizations are based on the GS, which provides default "typical" and "worst-case" input
parameters for the release calculations. EPA's calculation of air emissions for this OES, including all
calculation inputs, can be found in 1,4-Dioxane Supplemental Information File: Environmental Releases
to Air ( 023iY
Page 274 of 484
-------
7227
7228
7229
7230
7231
7232
7233
7234
7235
7236
7237
7238
7239
7240
7241
7242
7243
7244
7245
7246
7247
7248
7249
7250
7251
7252
7253
7254
7255
7256
7257
7258
7259
7260
7261
7262
7263
7264
7265
7266
7267
7268
7269
7270
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Film Cement
EPA estimated air emissions for facilities within the Film Cement OES using process information from
the Final Risk Evaluation for 1,4-Dioxane ( 1020c). The underlying process information for
this assessment was rated "high" during EPA's systematic review process.
The process of using film cement involves applying the cement onto edges of photographic films by
hand using a small brush, then joining the pieces of film by pressing and heating to dry the cement.
Based on this process information, EPA assumes that the 1,4-dioxane within film cement is volatilized
to air during the drying process and that 1,4-dioxane residual within empty film cement bottles may also
be volatilized to air. EPA estimated these air releases for this OES as a range, using a film cement use
rate of 2.5 to 12 L/site-year and a concentration of 1,4-dioxane in the film cement of 45 to 50 percent,
from the process information in the Final Risk Evaluation for 1,4-Dioxane ( 020c). These
releases may be to fugitive air or stack air depending on site-specific engineering controls.
EPA's calculation of air emissions for this OES, including all calculation inputs and assumptions, can be
found in 1,4-Dioxane Supplemental Information File: Environmental Releases to Air ( 023i).
Spray Polyurethane Foam
EPA estimated air emissions for facilities within the spray polyurethane foam OES using the GS on
Application of Spray Polyurethane Foam Insulation (U.S. EPA. 2020c). The GS on the Application of
Spray Polyurethane Foam Insulation was rated "medium" during EPA's systematic review process.
The GS indicates that there are six release points:
1. Releases to fugitive air for volatile chemicals during unloading of raw materials from transport
containers
2. Releases to water, incineration, or landfill from cleaning or disposal of transport containers
3. Releases to fugitive air for volatile chemicals during transport container cleaning
4. Releases to incineration or landfill from spray polyurethane foam application equipment cleaning
5. Releases to fugitive air for volatile chemicals during equipment cleaning
6. Releases to landfill of scrap foam from trimming applied foam
Based on the GS, release points #1, 3, and 5 have the potential for air emissions. To estimate these
releases, EPA used the equations specified in the GS ( 020c). Apart from weight fraction in
spray polyurethan foam, EPA did not find any data specific to 1,4-dioxane in this OES. Therefore, the
calculation of releases using this GS are for a "generic site," using the default input parameter values
from the GS. Specifically, EPA used the input parameter values that were presented in the original risk
evaluation for estimates of occupational exposure; see Appendix G of the Final Risk Evaluation for 1,4-
Dioxane (U.S. EPA. 2020c).
Using this methodology, EPA calculated the "typical" and "worst-case" air emissions for this OES.
These characterizations are based on the GS, which provides default "typical" and "worst-case" input
parameters for the release calculations. EPA's calculation of air emissions for this OES, including all
calculation inputs and assumptions, can be found in 1,4-Dioxane Supplemental Information File:
Environmental Releases to Air Q x \ \\ ,023i).
Page 275 of 484
-------
7271
7272
7273
7274
7275
7276
7277
7278
7279
7280
7281
7282
7283
7284
7285
7286
7287
7288
7289
7290
7291
7292
7293
7294
7295
7296
7297
7298
7299
7300
7301
7302
7303
7304
7305
7306
7307
7308
7309
7310
7311
7312
7313
7314
7315
7316
7317
7318
7319
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
3D Printing
Air emission data were not available for this OES and EPA did not find any information to model air
emissions for this OES using literature, GSs, or ESDs. EPA expects that industrial applications of this
OES to be accounted for in the Industrial Uses TRI data. Per the December 2020 Final Risk Evaluation
for 1,4-Dioxane ( 320c). 3D printing ink containing 1,4-dioxane is used in research labs to
print biomedical products. Because the 2019 TRI data for the Industrial Uses OES include medicinal and
pharmaceutical manufacturing NAICS codes, medical research labs that conduct 3D printing with 1,4-
dioxane inks may be captured in that OES. Therefore, EPA grouped the air release assessment for 3D
Printing into that for Industrial Uses. However, there is uncertainty in whether 3D printing sites are truly
captured in the Industrial Uses TRI data.
Dry Film Lubricant
EPA estimated air emissions for facilities within the Dry Film Lubricant OES using process information
from the Final Risk Evaluation for 1,4-Dioxane ( '20c). The underlying process information
for this assessment was rated "high" during EPA's systematic review process.
The process for the production and use of dry film lubricant is described in the 2020 RE and is based on
information provided to EPA by the one known user. In summary, the process entails producing the
concentrated dry film lubricant by mixing 1,4-dioxane with other additives, followed by the dilution of
the concentrated dry film lubricant with additional 1,4-dioxane and finally the use of the dry film
lubricant. The use involves spray application onto substrates in a vented paint booth and the subsequent
curing in a vented oven and cleaning of the dried parts in a 1,4-dioxane bath ( 20c). Based
on this process description, EPA assumes that 100 percent of the 1,4-dioxane in the applied dry film
lubricant is released to stack air from the paint booth and the oven. EPA estimated this release quantity
using batch parameters from the process description, including 5 percent 1,4-dioxane in the dry film
lubricant, 48 dry film lubricant applications per year, 0.5-pints of concentrated dry film lubricant, and
1.5-pints of pure 1,4-dioxane per application (U.S. EPA. 2020c).
EPA's calculation of air emissions for this OES, including all calculation inputs and assumptions, can be
found in 1,4-Dioxane Supplemental Information File: Environmental Releases to Air (]j_S IT \ 2023i).
Textile Dye
EPA used the OECD ESD on Textile Dyes (OE( ) to estimate water and land releases for this
OES; however, this ESD does not include approaches for estimating air releases. EPA did not find any
other GS/ESD, literature sources, or process information to model air releases for this OES. In addition,
EPA does not expect this OES to be sufficiently similar to other OES such that surrogate TRI data can
be used to estimate air emissions for this OES. Therefore, EPA was not able to estimate air releases for
these OESs.
Antifreeze
EPA did not find any directly applicable GS/ESD or literature sources for this OES; however, EPA
evaluated the potential for releases using the OECD ESD on Chemical Additives used in Automotive
Lubricants (OB 20) and the EPA MRD on Commercial Use of Automotive Detailing Products
( 'A. 2022b). The ESD and MRD were both rated "high" during EPA's systematic review
process.
For the use of antifreeze, EPA expects releases may occur from volatilizations of 1,4-dioxane during
unloading/ pouring antifreeze into vehicles, disposal, or cleaning of empty antifreeze containers, and
spent antifreeze. Both the ESD and MRD indicate that containers of automotive maintenance fluids are
Page 276 of 484
-------
7320
7321
7322
7323
7324
7325
7326
7327
7328
7329
7330
7331
7332
7333
7334
7335
7336
7337
7338
7339
7340
7341
7342
7343
7344
7345
7346
7347
7348
7349
7350
7351
7352
7353
7354
7355
7356
7357
7358
7359
7360
7361
7362
7363
7364
7365
7366
7367
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
typically small and are not rinsed, but rather disposed of as solid waste (1, c. < i1 \ J022U oU U
2020). Additionally, the ESD on Chemical Additives used in Automotive Lubricants indicates that spent
lubricants are disposed of via incineration, which EPA expects is similarly done for spent antifreeze
(OECD. 2020). Therefore, EPA expects the main source of air emissions to be from volatilizations of
1,4-dioxane during unloading/ pouring antifreeze into vehicles. EPA estimated this release using the
EPA/OAQPS AP-42 Loading Model and batch parameters from the ESD, MRD, and other sources.
Specifically, EPA used the consumer use rate of antifreeze (0.15 kg antifreeze/job) from the Final Risk
Evaluation for 1,4-Dioxane (U.S. EPA. 2020c) and scaled this value up to a commercial use rate based
on a range of the number of cars serviced at auto shops from the Near-Field/Far-Field Brake Model and
Automotive Detailing MRD (1 to 9 jobs/day). EPA used a range of concentrations of 1,4-dioxane in
antifreeze from the process description in Appendix F.4.2 and assumed antifreeze container sizes
ranging from 16 ounces to 5 gallons per the default container sizes in the MRD and ESD, respectively
(\ v i i \ t (i . '20). Using these batch parameters and the default parameters for the
EPA/OAQPS AP-42 Loading Model, EPA estimated low-end and high-end air emissions. EPA expects
these air emissions to be to fugitive air based on the use setting (e.g., outdoors, maintenance garages).
EPA's calculation of air emissions for this OES, including all calculation inputs and assumptions, can be
found in 1,4-Dioxane Supplemental Information File: Environmental Releases to Air ( 023i).
Surface Cleaner
EPA did not find any directly applicable GS/ESD or literature sources for this OES; however, EPA
estimated air releases using the SHEDs-HT modeling conducted for the one case study location
(Liverpool OH) and the assumptions described herein. EPA expects that the main release points from the
use of surface cleaners are from
1. Disposal of empty containers containing residual cleaning solution,
2. Application of the cleaning solution, and
3. Disposal of cleaning solution by rinsing or wiping.
Because EPA did not find any directly applicable GSs or ESDs, EPA used the Draft GS on Furnishing
Cleaning ( 322a) to inform these releases due to the similarities in surface cleaning and
furnishing cleaning. The Draft GS on Furnishing Cleaning was rated "high" during EPA's systematic
review process. Per this Draft GS, empty containers may be rinsed out in sinks or disposed of without
rinsing, such that releases may be to wastewater or landfill; the GS uses the EPA/OPPT Small Container
Residual Model to estimate this release. Application losses are to fugitive air from spray application; the
GS uses literature data to estimate this release. Once applied, the cleaner may be rinsed off or wiped off
with rags or towels, such that releases may be to wastewater or landfill; the GS assumes 100 percent
release scenario, estimating this release by subtracting the upstream losses from the cleaner use rate
( 2a).
The SHEDs-HT modeling estimated wastewater discharges of 72 g of 1,4-dioxane per day for
commercial uses of surface cleaners containing 1,4-dioxane in Liverpool OH. EPA used this quantity
and the above release information and models from the Draft GS on Furnishing Cleaning to back-
calculate a 1,4-dioxane use rate. EPA then applied the loss fraction to fugitive air from release point #2
to estimate air releases for this OES. EPA's calculation of air releases for this OES, including all
calculation inputs and assumptions, can be found in 1,4-Dioxane Supplemental Information File:
Environmental Releases to Air Q x \ \\ ,023i).
Page 277 of 484
-------
7368
7369
7370
7371
7372
7373
7374
7375
7376
7377
7378
7379
7380
7381
7382
7383
7384
7385
7386
7387
7388
7389
7390
7391
7392
7393
7394
7395
7396
7397
7398
7399
7400
7401
7402
7403
7404
7405
7406
7407
7408
7409
7410
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Dish Soap
EPA did not find relevant or surrogate TRI data, literature sources, sufficient process information, nor
ESD or GS with air release estimation approaches to estimate air releases for these OESs. Therefore,
EPA was not able to estimate air releases for these OESs.
Dishwasher Detergent
EPA did not find relevant or surrogate TRI data, literature sources, sufficient process information, nor
ESD or GS with air release estimation approaches to estimate air releases for these OESs. Therefore,
EPA was not able to estimate air releases for these OESs.
Laundry Detergent
EPA estimated air emissions for facilities within the Laundry Detergent OES using the OECD ESD on
Industrial and Institutional Laundries (OI ) and Monte Carlo modeling. The ESD on
Industrial and Institutional Laundries was rated "medium" during EPA's systematic review process. The
use of Monte Carlo modeling allows for variation of calculation input parameters such that a distribution
of environmental releases can be calculated, from which EPA can estimate the 50th and 95th percentile
releases. An explanation of this modeling approach is included in Appendix E.12.
Paints and Floor Lacquer
EPA estimated air emissions for facilities within the Paints and Floor Lacquers OES using the OECD
ESD on Coating Application via Spray-Painting in the Automotive Refinishing Industry (OECD.
201 la). The ESD was rated "medium" during EPA's systematic review process.
As described in the process description in Appendix F.4.7, 1,4-dioxane was identified by a public
comment as present in automotive refinishing products, thereby allowing EPA to identify the above
ESD as the most applicable GS/ESD available. This ESD indicates that releases are expected from
1. Releases to incineration or landfill from container cleaning/disposal,
2. Releases to incineration or landfill from equipment cleaning,
3. Releases to incineration or landfill from over sprayed coating that is captured by emission
controls, and
4. Releases to stack air from over sprayed coating that is not captured by emission controls.
Based on the GS, release point #4 has the potential for air emissions. To estimate this release, EPA used
the equations specified in the ESD (OECD. 201 la). Apart from weight fraction in coatings (see
Appendix F.4.7), EPA did not find any data specific to 1,4-dioxane in this OES. Therefore, the
calculation of releases using this GS are for a "generic site," using the default input parameter values
from the ESD.
Using this methodology, EPA calculated the low-end and high-end air emissions for this OES, which are
expected to be to stack air per the ESD ( ). The low- and high-end estimates are based on
the low- and high-end calculation input parameter defaults from the ESD. EPA's calculation of air
emissions for this OES, including all calculation inputs and assumptions, can be found in 1,4-Dioxane
Supplemental Information File: Environmental Releases to Air ( )23i).
Page 278 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
7411 Hydraulic Fracturing
7412 EPA estimated air emissions for facilities within the Hydraulic Fracturing OES using the Draft OECD
7413 ESD on Hydraulic Fracturing ( v «« \ . 022d) and Monte Carlo modeling. The Draft ESD on
7414 Hydraulic Fracturing was rated "high" during EPA's systematic review process. The use of Monte Carlo
7415 modeling allows for variation of calculation input parameters such that a distribution of environmental
7416 releases can be calculated, from which EPA can estimate the 50th and 95th percentile releases. An
7417 explanation of this modeling approach is included in Appendix E.13.
7418 E.5.3 Air Release Estimates Summary
7419 A summary of industrial and commercial air releases estimated using the above methods is presented in
7420 Table_Apx E-7 below.
Page 279 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
7421 Table Apx E-7 Summary of Daily Industrial and Commercial Air Release Estimates for 1,4-Dioxane
OES
Type of Air Release
Number of
Facilities with
Releases"
Estimated Daily Release
Range across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Manufacturing
Fugitive Air
1
2.62
250
Medium
TRI
Stack Air
1
0.0018
Medium
TRI
Import and
Repackaging
Fugitive Air
1
0
250
Medium
TRI
Stack Air
1
0.091
Medium
TRI
Industrial Uses
Fugitive Air
12
0
0.91
250
Medium
TRI
Stack Air
12
0
8.14
Medium
TRI
Functional Fluids
(Open-System)
Fugitive Air
2
0
0.009
247
Medium
TRI
Stack Air
2
0.19
1.38
Medium
TRI
Functional Fluids
(Closed-System)
All
Assessed as a part of Industrial Uses OES
N/A
N/A
Laboratory
Chemical
Fugitive Air or Stack
Air (-Unknown)
132
0.11 (typical)
0.41 (worst-
case)
250
High
GSd
Film Cement
Fugitive Air or Stack
Air (Unknown)
211
0.0046
0.025
250
High
Process
information®
Spray Foam
Application
Fugitive Air
1,553,559
0.0024
(typical)
0.012 (worst-
case)
3
Medium
GSf
Stack Air
1,553,559
0 (all air releases assessed to
fugitive)
Medium
GSf
Printing Inks (3D)
Fugitive Air, Stack
Air, and Land (all
types)
Assessed as a part of Industrial Uses OES
250
N/A
N/A
Dry Film Lubricant
Fugitive Air
8
0 (no fugitive releases per
process information)
48
High
Process
information®
Stack Air
8
0.75 (single value estimated
from process information)
High
Process
information®
Page 280 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Air Release
Number of
Facilities with
Releases"
Estimated Daily Release
Range across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Disposal
Fugitive Air
15
0
0.91
250
Medium
TRI
Stack Air
15
0
0.91
Medium
TRI
Textile Dye
Fugitive Air and
Stack Air
Not assessed
31 to 295
Medium
ESDg and
Modeling''
Antifreeze
Fugitive Air and
Stack Air
84,383
7.26E-16
1.80E-07
250
High
Process
information® and
Modeling''
Surface Cleaner
Fugitive Air
Unknown
0.0071
(typical; for
all sites in
Liverpool
OH, case
study)
0.013 (worst
case; for all
sites in
Liverpool, OH,
case study)
250
High
SHEDS-HT,'
Process
information®
Modeling''
Dish Soap
Fugitive air and
stack air
Not assessed
250
N/A
N/A
Dishwasher
Fugitive air and
Not assessed
250
N/A
N/A
Detergent
stack air
Fugitive air
95,533
1.83E-10
6.52E-07
Medium
ESD! and
Modeling''
Fugitive air, stack
95,533
1.51 E— 10
0.00714
Medium
ESD! and
Laundry Detergent
(Institutional) -
air, or POTW
(unknown
partitioning)
250 to 365
Modeling''
liquid detergents
Land (unknown
landfill),
incineration, or
POTW (unknown
partitioning)
95,533
4.05E-12
3.95E-05
Medium
ESD! and
Modeling''
Page 281 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Air Release
Number of
Facilities with
Releases"
Estimated Daily Release
Range across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Fugitive air
95,533
3.42E-12
2.77E-07
Medium
ESD! and
Modeling''
Stack air
95,533
1.40E-11
3.75E-06
Medium
ESD! and
Modeling''
Laundry Detergent
(Institutional) -
powder detergents
Fugitive air, stack
air, or POTW
(unknown
partitioning)
95,533
3.05E-08
2.10E-04
250 to 365
Medium
ESD! and
Modeling''
Land (unknown
landfill),
incineration, or
POTW (unknown
partitioning)
95,533
5.36E-08
0.0018
Medium
ESD! and
Modeling''
Fugitive air
2,453
6.25E-10
1.93E-06
Medium
ESD! and
Modeling''
Laundry Detergent
(Industrial) - liquid
Fugitive air, stack
air, or POTW
(unknown
partitioning)
2,453
5.48E-12
0.011
20 to 365
Medium
ESD! and
Modeling''
detergents
Land (unknown
landfill),
incineration, or
POTW (unknown
partitioning)
2,453
4.78E-12
1.46E-04
Medium
ESD! and
Modeling''
Page 282 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Air Release
Number of
Facilities with
Releases"
Estimated Daily Release
Range across Sites
(kg/site-day)
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
Min
Max
Fugitive air
2,453
3.13E-13
1.47E-05
Medium
ESD! and
Modeling''
Stack air
2,453
1.68E-12
1.82E-04
Medium
ESD! and
Modeling''
Laundry Detergent
(Industrial) -
powder detergents
Fugitive air, stack
air, or POTW
(unknown
partitioning)
2,453
1.76E-09
0.0112
20 to 365
Medium
ESD! and
Modeling''
Land (unknown
2,453
2.92E-11
3.92E-04
Medium
ESD! and
landfill),
Modeling''
incineration, or
POTW (unknown
partitioning)
Paints and Floor
Stack air
33,648
4.68E-10
1.60E-06
250
Medium
ESD7
Lacquer
Polyethylene
Fugitive Air
13
0
1.57
Medium
TRI
Terephthalate
(PET) Byproduct
Stack Air
13
0.0049
13.8
250
Medium
TRI
Ethoxylation
Fugitive Air
8
0
7.4
250
Medium
TRI
Process Byproduct
Stack Air
8
0
32
Medium
TRI
Fugitive air
411
1.99E-07
5482
Medium
ESD^ and
Modeling''
Stack air
411
0 (all air releases assessed to
Medium
ESD" and
Hydraulic
fugitive)
1 to 72
Modeling''
Fracturing
Surface water,
incineration, or
landfill (unknown
partitioning)
411
3.61E-10
4.59
Medium
ESD'' and
Modeling''
Page 283 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Type of Air Release
Number of
Facilities with
Releases"
Estimated Daily Release
Range across Sites
(kg/site-day)
Min
Max
Estimated
Release
Frequency
Range
(days)''
Overall Data
Quality
Determination
Sources'
"Where available, EPA used 2013-2019 TRI (U.S. EPA. 2022g) and 2013-2019 DMR (U.S. EPA. 2022c) data to provide a basis to estimate the number of sites
using 1,4-dioxane within a COU.
* Where available. EPA used the December 2020 Final Risk Evaluation for 1.4-Dioxane (U.S. EPA. 20200). generic scenarios, and emission scenario documents
to provide a basis to estimate the number of release days of 1,4-dioxane within a COU.
c Narrative descriptions of all release estimate sources are provided in Appendix E.5.
d The generic scenario used for this COU is the GS on Use of Laboratory Chemicals (U.S. EPA. 202211).
e For this COU, EPA used process information, which is further described in Appendix E.5.2.
' The generic scenario used for this COU is the GS on Application of Spray PoK urethane Foam Insulation (U.S. EPA. 2018b).
g The emission scenario document used for this COU is the ESD on Textile Dyes (OECD. 2017).
h For this COU, EPA used various models and literature for model input parameters as described in Appendix E.5.2.
1 The emission scenario document used for this COU is the ESD on Industrial and Institutional Laundries (OECD. 201 lb).
1 The emission scenario document used for this COU is the ESD on Coating Application via Spray Painting in the Automotive Refinishing Industry (OECD.
2011a).
The emission scenario document used for this COU is the Draft ESD on Hydraulic Fracturing (U.S. EPA. 2022d).
' EPA used the down the drain water release estimates from the SHEDs-HT model for the Liverpool OH case study (see Section 2.1.1.2) to estimate air releases
by back calculating 1,4-dioxane use rates and applying loss fractions for air releases using literature and standard models described in Appendix E.5.2.
7422
Page 284 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
7423 E.5.4 Summary of Weight of the Scientific Evidence Conclusions in Air Release Estimates
7424 Table Apx E-8 provides a summary of EPA's weight of the scientific evidence conclusions in its air release estimates for each of the
7425 Occupational Exposure Scenarios assessed. Detailed descriptions of non-OES specific strengths, limitations, assumptions, and uncertainties
7426 {e.g., general limitations for TRI, DMR, etc.) are provided in Appendix E.6.
7427
7428 Table Apx E-8 Summary of Weight of the Scientific Evidence Conclusions in Air Release Estimates by PES
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Manufacturing
Air releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this OES
are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the low number of data points, lack of variability (only one year of data
used), uncertainty in the accuracy of reported releases, and the limitations in representativeness to all sites because TRI
may not capture all relevant sites. Additionally, EPA made assumptions on the number of operating days to estimate daily
releases. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is
moderate to robust and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Import and Repackaging
Air releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this OES
are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the low number of data points, lack of variability (only one year of data
used), uncertainty in the accuracy of reported releases, uncertainty in EPA's use of Form A submissions, and the
limitations in representativeness to all sites because TRI may not capture all relevant sites. Some facilities within this OES
reported to TRI using a Form A, which does not include any details on chemical release quantities. When a facility has
submitted a Form A, there is no way to discern the quantity released. Therefore, where facilities reported to TRI with a
Form A, EPA used the Form A threshold for total releases of 500 lb/year for each release media; however, there is
uncertainty in this because the actual release quantity is unknown. Furthermore, the threshold represents an upper limit on
total releases from the facility; therefore, assessing releases at the threshold value may overestimate actual releases from
the facility. Additionally, EPA made assumptions on the number of operating days to estimate daily releases. Based on this
information, EPA has concluded that the weight of the scientific evidence for this assessment is moderate to robust and
provides a plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Industrial Uses
Air releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this OES
are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
Page 285 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
strength of the evidence for this OES include a lack of variability (only one year of data used), uncertainty in the accuracy
of reported releases, uncertainty in EPA's use of Form A submissions, and the limitations in representativeness to all sites
because TRI may not capture all relevant sites. Some facilities within this OES reported to TRI using a Form A, which
does not include any details on chemical release quantities. When a facility has submitted a Form A, there is no way to
discern the quantity released. Therefore, where facilities reported to TRI with a Form A, EPA used the Form A threshold
for total releases of 500 lb/year for each release media; however, there is uncertainty in this because the actual release
quantity is unknown. Furthermore, the threshold represents an upper limit on total releases from the facility; therefore,
assessing releases at the threshold value may overestimate actual releases from the facility. Based on this information, EPA
has concluded that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Functional Fluids (Open-
System)
Air releases are assessed using reported releases from 2019 TRI. Factors that increase the strength of evidence for this OES
are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the low number of data points (only two reporting sites), lack of variability
(only one year of data used), uncertainty in the accuracy of reported releases, and the limitations in representativeness to all
sites because TRI may not capture all relevant sites. Additionally, EPA made assumptions on the number of operating days
to estimate daily releases. Based on this information, EPA has concluded that the weight of the scientific evidence for this
assessment is moderate to robust and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Functional Fluids (Closed-
System)
No data was available to estimate releases for this OES. For the air release assessment, EPA grouped this OES with the
Industrial Uses OES because the sources of release are expected to be similar between these OESs. Factors that increase
the strength of evidence for this OES are that TRI has a medium overall data quality determination and consistency within
the dataset (all reporters use the same or similar reporting forms). Factors that decrease the strength of evidence for this
OES are that the Industrial Releases OES release data are use as surrogate for this OES, uncertainty in the accuracy of
reported releases, limitations in representativeness to all sites because TRI may not capture all relevant sites, and lack of
variability (only one year of data used). Refer to the Industrial Uses OES discussion for additional discussion. Based on
this information, EPA has concluded that the weight of the scientific evidence for this assessment is slight and provides a
plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Laboratory Chemicals
Air releases are assessed using the Draft GS on Use of Laboratory Chemicals. Factors that increase the strength of
evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to surrogate), the Draft GS
on Use of Laboratory Chemicals has a high overall data quality determination, and the low level of uncertainty in the data.
Factors that decrease the strength of the evidence for this OES include the that the GS has not been peer-reviewed,
uncertainty in the representativeness of the GS towards all sites in this OES, and a lack of variability in the analysis.
Specifically, because the default values in the ESD are generic, there is uncertainty in the representativeness of generic site
estimates of actual releases from real-world sites that use 1,4-dioxane. Another uncertainty is lack of consideration for
Page 286 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
release controls. The ESD assumes that all activities occur without any release controls. Actual releases may be less than
estimated if facilities utilize pollution control methods, contributing to uncertainty. Based on this information, EPA has
concluded that the weight of the scientific evidence for this assessment is slight to moderate and provides a plausible
estimate of releases in consideration of the strengths and limitations of reasonably available data.
Film Cement
Air releases are assessed using process information from the Final Risk Evaluation for 1,4-Dioxane. Factors that increase
the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to
surrogate), the underlying data sources for the process information have a high overall data quality determination, and the
low level of uncertainty in the data because the process information comes directly from actual users of 1,4-dioxane in film
cement. Factors that decrease the strength of the evidence for this OES include uncertainty in the representativeness of
evidence to all sites in this OES and a lack of variability. Specifically, the process information for the production and use
of film cement is based on information from three use sites, one from Australia and two from the U.S. Based on this
information, EPA has concluded that the weight of the scientific evidence for this assessment is slight to moderate and
provides a plausible estimate of releases in consideration of the strengths and limitations of reasonably available data.
Spray Foam Application
Air releases are assessed using the GS on Application of Spray Polyurethane Foam Insulation. Factors that increase the
strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to surrogate),
the underlying data sources for the process information have a medium overall data quality determination, and the low
level of uncertainty in the data. Factors that decrease the strength of the evidence for this OES include uncertainty in the
representativeness of the GS to all sites since it is generic and not specific to sites that use 1,4-dioxane and a lack of
variability. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is
slight to moderate and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Printing Inks (3D)
No data was available to estimate releases for this OES. For the air release assessment, EPA grouped this OES with the
Industrial Uses OES because the sources of release are expected to be similar between these OESs. Factors that increase
the strength of evidence for this OES are that TRI has a medium overall data quality determination and consistency within
the dataset (all reporters use the same or similar reporting forms). Factors that decrease the strength of evidence for this
OES are that the Industrial Releases OES release data are use as surrogate for this OES, uncertainty in the accuracy of
reported releases, limitations in representativeness to all sites because TRI may not capture all relevant sites or smaller
commercial 3D printing uses, and lack of variability (only one year of data used). Refer to the Industrial Uses OES
discussion for additional discussion. Based on this information, EPA has concluded that the weight of the scientific
evidence for this assessment is slight and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Dry Film Lubricant
Air releases are assessed using process information from the Final Risk Evaluation for 1,4-Dioxane. Factors that increase
the strength of evidence for this OES are that the release estimates are directly relevant to the OES (as opposed to
surrogate), that the underlying data sources for the process information have a high overall data quality determination, and
a low level of uncertainty in the data because the process information comes directly from an actual user of 1,4-dioxane in
Page 287 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
dry film lubricants. Factors that decrease the strength of the evidence for this OES include uncertainty in the
representativeness of evidence to all sites and a lack of variability. Based on this information, EPA has concluded that the
weight of the scientific evidence for this assessment is slight to moderate and provides a plausible estimate of releases in
consideration of the strengths and limitations of reasonably available data.
Disposal
Air releases are assessed using reported discharges from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, high number of data points, and consistency within the dataset (all reporters use the same or similar
reporting forms). A strength of TRI data is that TRI compiles the best readily available release data for all reporting
facilities. Factors that decrease the strength of the evidence for this OES include lack of variability (only one year of data
used), uncertainty in the accuracy of reported releases, uncertainty in EPA's use of Form A submissions, and the
limitations in representativeness to all sites because TRI may not capture all relevant sites. Some facilities within this OES
reported to TRI using a Form A, which does not include any details on chemical release quantities. When a facility has
submitted a Form A, there is no way to discern the quantity released. Therefore, where facilities reported to TRI with a
Form A, EPA used the Form A threshold for total releases of 500 lb/year for each release media; however, there is
uncertainty in this because the actual release quantity is unknown. Furthermore, the threshold represents an upper limit on
total releases from the facility; therefore, assessing releases at the threshold value may overestimate actual releases from
the facility. Additionally, uncertainty is introduced from EPA's assumptions on the number of operating days to estimate
daily releases. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment
is moderate to robust and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Textile Dye
EPA used the ESD on Textile Dyes to estimate land and water releases; however, this ESD does not include approaches for
estimating air releases. EPA did not find any other GS/ESD, literature sources, or process information to model air releases
for this OES. Furthermore, EPA does not expect this OES to be sufficiently similar to other OES such that surrogate TRI
data can be used to estimate air emissions for this OES. Therefore, EPA was not able to estimate air releases for this OES
and concluded that the weight of the scientific evidence is indeterminant.
Antifreeze
Air releases are assessed using the OECD ESD on Chemical Additives used in Automotive Lubricants, the EPA MRD on
Commercial Use of Automotive Detailing Products, and EPA/OPPT models. Factors that increase the strength of evidence
for this OES are that the ESD and MRD used have high overall data quality determinations and consistency within the
sources used. Factors that decrease the strength of the evidence for this OES include that the ESD and MRD are not
directly applicable to antifreeze uses (used as surrogate), uncertainty in the representativeness of the ESD and MRD to all
sites and sites that specifically use 1,4-dioxane since these documents contain generic values, and a lack of variability.
Additionally, EPA scaled up a consumer antifreeze use rate to a commercial use rate based on information in the ESD and
MRD, which increases uncertainty. Based on this information, EPA has concluded that the weight of the scientific
evidence for this assessment is slight to moderate and provides a plausible estimate of releases in consideration of the
strengths and limitations of reasonably available data.
Page 288 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
Surface Cleaner
Air releases are assessed using the SHEDS-HT model and the Draft GS on Furnishing Cleaning. To estimate air releases,
EPA used loss fractions for water releases from the GS and the modeled water release from SHEDS-HT to back-calculate a
1,4-dioxane use rate. EPA then applied loss fractions for air releases from the GS to estimate air releases for this OES.
Factors that increase the strength of evidence for this OES include that the release estimates are directly relevant to the
OES (as opposed to surrogate), that the Draft GS used has a high overall data quality determination, and variability in the
model input parameters. Factors that decrease the strength of the evidence for this OES include uncertainty in the
representativeness to all sites because the SHEDS-HT estimate is based on one case study for Liverpool, OH and because
the estimate is not site-specific (the release estimate is a total for all sites in Liverpool, OH). Based on this information,
EPA has concluded that the weight of the scientific evidence for this assessment is slight and provides a plausible estimate
of releases in consideration of the strengths and limitations of reasonably available data.
Dish Soap
EPA did not find relevant or surrogate TRI data, literature sources, sufficient process information, nor ESD or GS with air
release estimation approaches to estimate air releases for these OESs. Therefore, EPA was not able to estimate air releases
for this OES and concluded that the weight of the scientific evidence is indeterminant.
Dishwasher Detergent
EPA did not find relevant or surrogate TRI data, literature sources, sufficient process information, nor ESD or GS with air
release estimation approaches to estimate air releases for these OESs. Therefore, EPA was not able to estimate air releases
for this OES and concluded that the weight of the scientific evidence is indeterminant.
Laundry Detergent
Air releases are assessed using Monte Carlo modeling with information from the ESD on Industrial and Institutional
Laundries. Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant to
the OES (as opposed to surrogate), that the ESD on Industrial and Institutional Laundries has a medium overall data
quality determination and was peer reviewed, high number of data points (simulation runs), consistency within the dataset,
and full distributions of input parameters. The Monte Carlo modeling accounts for the entire distribution of input
parameters, calculating a distribution of potential release values that represents a larger proportion of sites than a discrete
value. Additionally, EPA was able to separately estimate releases for industrial and institutional laundry settings. Factors
that decrease the strength of the evidence for this OES include uncertainties and limitations in the representativeness of the
estimates for sites that specifically use 1,4-dioxane because the default values in the ESD are generic. Another uncertainty
is lack of consideration for release controls. The ESD assumes that all activities occur without any release controls. Actual
releases may be less than estimated if facilities utilize pollution control methods. Based on this information, EPA has
concluded that the weight of the scientific evidence for this assessment is moderate and provides a plausible estimate of
releases in consideration of the strengths and limitations of reasonably available data.
Paint and Floor Lacquer
Air releases are assessed using OECD ESD on Coating Application via Spray-Painting in the Automotive Refinishing
Industry. Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant to
the OES (as opposed to surrogate), the ESD has a medium overall data quality determination, consistency within the
sources used, and a low amount of uncertainties. Factors that decrease the strength of the evidence for this OES include a
lack of variability and uncertainty in the representativeness of the ESD to all sites and sites that specifically use 1,4-
dioxane since the ESD is generic. Based on this information, EPA has concluded that the weight of the scientific evidence
Page 289 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Weight of the Scientific Evidence Conclusion in Release Estimates
for this assessment is slight to moderate and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Polyethylene Terephthalate
(PET) Byproduct
Air releases are assessed using reported discharges from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, consistency within the dataset (all reporters use the same or similar reporting forms), and
consistency with the emission data from the related life cycle analysis discussed in Appendix E.6. A strength of TRI data is
that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the strength of the
evidence for this OES include uncertainty in the accuracy of reported releases, lack of variability (only one year of data
used), and the limitations in representativeness to all sites because TRI may not capture all relevant sites. Additionally,
EPA made assumptions on the number of operating days to estimate daily releases, which introduces additional
uncertainty. Based on this information, EPA has concluded that the weight of the scientific evidence for this assessment is
moderate to robust and provides a plausible estimate of releases in consideration of the strengths and limitations of
reasonably available data.
Ethoxylation Process
Byproduct
Air releases are assessed using reported discharges from 2019 TRI. Factors that increase the strength of evidence for this
OES are that the release data are directly relevant to the OES (as opposed to surrogate), that TRI has a medium overall data
quality determination, and consistency within the dataset (all reporters use the same or similar reporting forms). A strength
of TRI data is that TRI compiles the best readily available release data for all reporting facilities. Factors that decrease the
strength of the evidence for this OES include the uncertainty in the accuracy of reported releases, lack of variability (only
one year of data used), and the limitations in representativeness to all sites because TRI may not capture all relevant sites.
Additionally, EPA made assumptions on the number of operating days to estimate daily releases, which introduces
additional uncertainty. Based on this information, EPA has concluded that the weight of the scientific evidence for this
assessment is moderate to robust and provides a plausible estimate of releases in consideration of the strengths and
limitations of reasonably available data.
Hydraulic Fracturing
Air releases are assessed using Monte Carlo modeling with information from the Draft ESD on Hydraulic Fracturing and
FracFocus 3.0. Factors that increase the strength of evidence for this OES are that the release estimates are directly relevant
to the OES (as opposed to surrogate), that the Draft ESD on Hydraulic Fracturing and FracFocus 3.0 have medium overall
data quality determinations, high number of data points (simulation runs), consistency within the dataset, and full
distributions of input parameters. The Monte Carlo modeling accounts for the entire distribution of input parameters,
calculating a distribution of potential release values that represents a larger proportion of sites than a discrete value. Factors
that decrease the strength of the evidence for this OES include that the Draft ESD has not been peer reviewed and the
uncertainties and limitations in the representativeness of the estimates for sites that specifically use 1,4-dioxane because
the default values from the Draft ESD on Hydraulic Fracturing. Another uncertainty is lack of consideration for release
controls. The ESD assumes that all activities occur without any release controls. Actual releases may be less than estimated
if facilities utilize pollution control methods, contributing to uncertainty. Based on this information, EPA has concluded
that the weight of the scientific evidence for this assessment is moderate to robust and provides a plausible estimate of
releases in consideration of the strengths and limitations of reasonably available data.
7429
Page 290 of 484
-------
7430
7431
7432
7433
7434
7435
7436
7437
7438
7439
7440
7441
7442
7443
7444
7445
7446
7447
7448
7449
7450
7451
7452
7453
7454
7455
7456
7457
7458
7459
7460
7461
7462
7463
7464
7465
7466
7467
7468
7469
7470
7471
7472
7473
7474
7475
7476
7477
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E,6 Comparison to PET Life Cycle Analysis
EPA identified a relevant life cycle analysis (LCA) published by the National Association for PET
Container Resources (NAPCOR) that included 1,4-dioxane emission factors from PET resin production
(Franklin Associates. 2020). EPA did not use these emission factors to estimate releases in this Risk
Evaluation because there were site-specific releases reported in TRI and DMR. This LCA only provided
generic emission factors for air and surface water releases aggregated across seven unspecified sites; the
LCA did not provide 1,4-dioxane emission factors for land releases. The emission factors in the LCA
were reported by three producers (seven sites) that account for 50% of the 2015 U.S. PET production in
a survey, and the basis of the emission factors is not provided. However, the survey states that the
release data is primary data (i.e., the data were provided by directly by the surveyed PET producers). As
opposed to conventional emission factors, the report only provides the order of magnitude of the average
amount of 1,4-dioxane released per amount of PET produced. Discrete, site-specific emission factors are
not provided. As a result, the variability of 1,4-dioxane releases from site to site is unknown. EPA
prefers the use of site-specific release data as opposed to generic emission factors. Therefore, a
comparison between total annual air and water releases from the LCA and from the TRI and DMR data
used in this Risk Evaluation is provided below for context.
The LCA estimated that 4.7 million tons of PET capacity was available in 2015 in North America
(Franklin Associates. 2020). To obtain total annual air and water release estimates from the LCA, EPA
multiplied this production volume by the reported 1,4-dioxane emission factors of 0.001 kg 1,4-dioxane
emitted per 1000 kg PET for air releases, and 0.01 kg 1,4-dioxane emitted per 1000 kg PET for surface
water releases. To obtain the total annual air and water releases from the TRI and DMR used in this Risk
Evaluation, EPA summed all reported annual site-specific air emissions and surface water discharges
that were mapped to the "PET Manufacturing" OES (see Appendix E.3 and E.5 for additional
information on the use of TRI and DMR). The total annual releases from the LCA and from TRI and
DMR is compiled in Table Apx E-9. EPA did this comparison with 2019 TRI/DMR because EPA's
Risk Evaluation largely uses 2019 data, as well as 2015 TRI/DMR data because the releases estimated
with the LCA data are based on 2015 PET manufacturing data.
For air emissions, the LCA estimate and EPA's estimates from the 2019 and 2015 TRI are comparable,
being within an order of magnitude. Differences in the estimates likely arise since EPA's analysis
accounted for emissions from 13 PET manufacturing facilities compared to the seven facilities in the
LCA. Additionally, the LCA is an aggregate of releases across sites whereas EPA's analysis accounts
for variability by using data from individual sites.
For surface water discharges, the LCA estimate and EPA's estimates from the 2019 TRI and DMR show
a larger discrepancy, with EPA's estimate being two orders of magnitude larger than the LCA estimate.
However, over 2.51 million kg of the approximately 2.53 million kg (99.2%) of surface water discharges
in EPA's estimate comes from a single facility's 2019 DMR report (DAK Americas LLC, Columbia).
The other 12 facilities with non-zero surface water discharges in EPA's analysis had site-specific annual
discharges ranging between 5.37 and 8,922 kgs for 2019. EPA reviewed the DAK Americas LLC,
Columbia DMR reports from other years for comparison, which indicated approximately 14,000 kg of
1,4-dioxane were discharged in 2022, 8,800 kg in 2021, 6.8 million kg in 2020, and 2,300 kg in 2018.
DAK Americas LLC, Columbia did not include 1,4-dioxane in their DMRs in 2016 or 2017 (the two
earliest reporting years EPA looked at for this analysis). It is unclear why DAK Americas LLC,
Columbia's discharges were significantly higher in 2019 and 2020 or why these discharges were
different than other PET manufacturers in EPA's analysis. However, it is more likely that the facilities
analyzed in the LCA were more similar to the other PET manufacturing facilities in EPA's analysis,
Page 291 of 484
-------
7478
7479
7480
7481
7482
7483
7484
7485
7486
7487
7488
7489
7490
7491
7492
7493
7494
7495
7496
7497
7498
7499
7500
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
with the total surface water emissions estimated from the LCA data being within one order of magnitude
of EPA's estimate when excluding the discharges from the DAK Americas LLC, Columbia facility, as
shown in Table Apx E-9. The LCA estimate and EPA's estimates from the 2015 TRI are comparable.
As indicated above, estimates from the LCA analysis and EPA's analysis with TRI and DMR data may
differ for several reasons. First, the LCA provides a single emission factor aggregated from data across
seven sites, reducing the impact of site-specific variability in releases. Whereas EPA's analysis uses site-
specific release data from 13 sites (for air emissions) and 19 sites (for surface water discharges, six of
which reported zero surface water discharges). EPA also does not have access to site identities, or the
underlying data/methodologies used to estimate emission factors in the LCA, which limits EPA's ability
to do a direct site-to-site comparison of results between the two analyses. Additionally, the LCA study
states that some emissions are reported only by the order of magnitude of the average to protect the
confidentiality of individual companies, introducing further uncertainty in the emission factors presented
in the study. Lastly, the LCA data is from 2015 whereas EPA used data from 2019.
Table Apx E-9. Comparison of TRI/DMR Release Data to LCA Study for PET Byproduct
Data Source
Total Release for All Sites
(kg/vr)
Air emissions
EPA Estimate in this Risk Evaluation -
Based on 2019 TRI (including DAK Americas LLC, Columbia)
10,695
EPA Estimate in this Risk Evaluation -
Based on 2015 TRI (including DAK Americas LLC, Columbia)
12,407
LCA Estimate (Franklin Associates. 2020) - Based on 2015 data
4,264
Surface water discharges
EPA Estimate in this Risk Evaluation -
Based on 2019 DMR and TRI (including DAK Americas LLC,
Columbia)
2,531,730
EPA Estimate in this Risk Evaluation -
Based on 2019 DMR and TRI (excluding DAK Americas LLC,
Columbia)
19,296
EPA Estimate in this Risk Evaluation -
Based on 2015 DMR and TRI
20,511
LCA Estimate (Franklin Associates. 2020) - Based on 2015 data
42,638
E.7 Detailed Strengths, Limitations, Assumptions and Key Sources of
Uncertainties for the Environmental Release Assessment
This section includes detailed strengths, limitations, assumptions, and uncertainties associated with
EPA's approaches for estimating air, water, and land releases in this supplemental risk evaluation. This
section is intended to supplement the summary of strengths, limitations, assumptions, and uncertainties
discussed in Section 2.2.1.3 with additional details.
Page 292 of 484
-------
7501
7502
7503
7504
7505
7506
7507
7508
7509
7510
7511
7512
7513
7514
7515
7516
7517
7518
7519
7520
7521
7522
7523
7524
7525
7526
7527
7528
7529
7530
7531
7532
7533
7534
7535
7536
7537
7538
7539
7540
7541
7542
7543
7544
7545
7546
7547
7548
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Facilities Not Captured by TRI or DMR Reporting
When using TRI data to analyze chemical releases, it is important to acknowledge that TRI reporting
does not include all releases of the chemical and therefore, the number of sites for a given OES may be
underestimated. For each OES that had TRI or DMR data, the analysis of releases for those OES was
limited to the facilities that reported releases to TRI and DMR. Therefore, it is uncertain the extent to
which sites not captured in these databases have air, water, or land releases of 1,4-dioxane and what the
exact media of release for those releases would be (e.g., stack vs. fugitive air, surface water vs POTW,
RCRA or another type of landfill). To the extent additional sites are not captured, releases may be
underestimated; however, the magnitude of this underestimation is unknown. TRI data do not include
• Releases from any facility that used the chemical in quantities below the applicable annual
chemical activity threshold (e.g., 25,000 lb manufactured or processed, or 10,000 lb otherwise
used, for most chemicals);
• Releases from any facility that is not in a TRI covered sector; and
• Releases from any facility that does not meet the TRI employment threshold of greater than 10
full-time employee equivalents (20,000 labor hours) for the year.
DMR Release Data
For facilities that reported having zero pollutant loads to DMR, the ECHO Pollutant Loading Tool
Advanced Search applies a hybrid method to analyze nondetects. The EZ Search Load Module uses a
combination of setting nondetects equal to zero and as one half the detection limit to calculate the annual
pollutant loadings. Specifically, if the pollutant was measured as nondetect for all monitoring periods in
a reporting year, then the EZ Search Load Module sets the annual pollutant load to zero. If the pollutant
was detected for at least one monitoring period in a reporting year, then the EZ Search Load Module
calculates the annual pollutant load by setting the nondetects equal to one half the detection limit. This
method could cause overestimation or underestimation of annual and daily pollutant loads.
TRI Release Data
EPCRA section 313 states that facilities may estimate their release quantities using "readily available
data," including monitoring data, collected for other purposes. When data are not readily available,
EPCRA section 313 states that "reasonable estimates" may be used. The facility is not required to
monitor or measure the quantities, concentration, or frequency of any toxic chemical release for TRI
reporting. TRI guidance states that not using readily available information, such as relevant monitoring
data collected for compliance with other regulations, could result in enforcement and penalties.
For each release quantity reported, TRI facilities select a "Basis of estimate" code indicating the
principal method used to determine the amount of the release. TRI provides six basis of estimate codes
to choose from: continuous monitoring, periodic monitoring, mass balance, published emissions factors,
site-specific emissions factors, or engineering calculations/best engineering judgment. In facilities where
a chemical is used in multiple operations, the facility may use a combination of methods to calculate the
release reported. In such cases, TRI instructs the facility to enter the basis of estimate code of the
method that applies to the largest portion of the release quantity. Additional details on the basis of
estimate, such as any calculations and underlying assumptions, are not reported. Depending on the
inputs and/or monitoring methods used by each facility, any of the methods used to estimate releases
may over or underestimate releases. The magnitude of this uncertainty is unknown.
For any release quantity that is less than 1,000 lb, facilities may report either the estimated quantity or a
range code. The 1,000-pound limit for range code reporting applies to each type of release reported to
TRI - fugitive air emissions, stack air emissions, water discharges, each type of land disposal, and each
Page 293 of 484
-------
7549
7550
7551
7552
7553
7554
7555
7556
7557
7558
7559
7560
7561
7562
7563
7564
7565
7566
7567
7568
7569
7570
7571
7572
7573
7574
7575
7576
7577
7578
7579
7580
7581
7582
7583
7584
7585
7586
7587
7588
7589
7590
7591
7592
7593
7594
7595
7596
7597
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
type of off-site transfer. There are three TRI range codes: 1 to 10; 11 to 499; and 500 to 999 lb. TRI data
tools display the approximate midpoint of the range {i.e., 5, 250, or 750 lb). Using this midpoint value
may be either an over or an underestimate of the true value, depending where on the range the true value
lies. Although analyses using data that was reported as a range code may add uncertainty, it is not clear
that the uncertainty associated with a range code is greater than that associated with any other estimated
release value. Range code reporting is not permitted for chemicals of special concern.
TRI guidance states that release estimates need not be reported to more than two significant figures.
However, the guidance also states that facilities should report release quantities at a level of precision
supported by the accuracy of the underlying data and the estimation techniques on which the estimate
was based. If a facility's release calculations support reporting an amount that is more precise than two
significant digits, then the facility should report that more precise amount. The facility makes the
determination of the accuracy of their estimate and the appropriate significant digits to use.
For chemicals that meet certain criteria, facilities have the option of submitting a TRI Form A
Certification Statement instead of a TRI Form R. The Form A does not include any details on the
chemical release or waste management quantities. The criteria for a Form A are that during the reporting
year, the chemical (1) did not exceed 500 lb for the total annual reportable amount (including the sum of
on- and off-site quantities released, treated, recycled, and used for energy recovery); (2) amounts
manufactured, processed, or otherwise used do not exceed 1 million lb; and (3) the chemical is not a
chemical of special concern. When conducting analyses of chemical releases and a facility has submitted
a Form A for the chemical, there is no way to discern the quantity released to each medium or even if
there were any releases. Where facilities reported to TRI with a Form A, EPA used the Form A
threshold for total releases of 500 lb/year for each release media {e.g., fugitive air, stack air, surface
water, POTW, non-POTW WWT, RCRA landfill, other landfill). EPA used the entire 500 lb/year for
each release media; however, since this threshold is for total site releases, these 500 lb/year are to only
one of these media at a time (since assessing 500 lb/year for all media at once would double count the
releases and exceed the total release threshold for Form A). Furthermore, the threshold represents an
upper limit on total releases to all environmental media from the facility; therefore, assessing releases at
the threshold value likely overestimates actual releases from the facility.
Mapping TRI and DMR Facilities to OES
EPA used a crosswalk between TRI uses/sub-uses and CDR Industrial Function Category (IFC) codes
(see Appendix E.9) along with a mapping of CDR IFC codes to OES to assign the OES for each facility
that reported to TRI. However, there are limitations to this approach. For example, this approach may
result in the mapping of multiple OES for one facility. Additionally, there are limitations to the TRI -
CDR crosswalk. For example, a TRI use/sub-use may encompass multiple uses that are not captured in
the crosswalked CDR IFC codes. In these instances, EPA determined the primary OES using the NAICS
codes that the facility reported in TRI, information on types of products that the facility manufactures
found from internet searches, and which OES is most likely to result in release. OES are related to
COUs as defined in the crosswalk in Table 2-1 and Appendix D.2.This approach requires some level of
engineering judgment to determine which OES is the most applicable to the facility, which introduces
uncertainty in the OES mapping. Additionally, this approach assumes only one OES is applicable to the
facility, which may be incorrect if the facility uses 1,4-dioxane for multiple purposes. If facilities were
categorized under a different OES, the annual releases for each site would remain unchanged; however,
average daily releases may change depending on the release days expected for the different OES.
Additional uncertainty is present in the OES mapping for TRI sites that reported using a Form A and
DMR sites because there is no reported use/sub-use information. EPA used a similar procedure as
Page 294 of 484
-------
7598
7599
7600
7601
7602
7603
7604
7605
7606
7607
7608
7609
7610
7611
7612
7613
7614
7615
7616
7617
7618
7619
7620
7621
7622
7623
7624
7625
7626
7627
7628
7629
7630
7631
7632
7633
7634
7635
7636
7637
7638
7639
7640
7641
7642
7643
7644
7645
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
described above to map these sites to an OES, involving the use of NAICS and Standard Industrial
Classification (SIC) codes reported to TRI and DMR, internet searches on the types of products made at
the facility, and which OES is most likely to result in release. Since this approach involves engineering
judgment to determine which OES is the most applicable to the facility, there is uncertainty in the OES
mapping.
There is also uncertainty in the NAICS codes and SIC codes reported in TRI and DMR. TRI facilities
enter the facility's primary NAICS code indicating the primary economic activity at the facility.
Facilities can also enter secondary NAICS codes. When using TRI chemical release data for a facility
that also reported secondary NAICS codes, there may be uncertainty as to which NAICS is associated
with the use of the chemical. Additionally, NAICS codes and SIC codes are reported for the facility as a
whole and are not chemical specific.
Estimating Daily Releases from Annual TRI and DMR Release Data
Facilities reporting to TRI and DMR only report annual releases; to assess daily air and water releases,
EPA estimated the release days and averaged the annual releases over these days. There is some
uncertainty that all facilities for a given OES operate for the assumed duration; therefore, the average
daily release may be higher if sites have fewer release days or lower if they have greater release days.
Furthermore, chemical concentrations in air emissions and wastewater streams at each facility may vary
from day to day such that on any given day the actual daily releases may be higher or lower than the
estimated average daily discharge. Thus, this approach minimizes spikes and drops in emissions and
discharges from day to day.
Pollutant Load values of "0" in the ECHO Pollutant Loading Tool Advanced Search output file are by
default reported as below the detection limit in the monitoring reports used by the Loading Calculator
Module.
EPA did not estimate daily land releases due to the high level of uncertainty in the number of release
days associated with land releases; instead, EPA estimated annual land releases.
Representativeness of TRI and DMR for an OES as a Whole
The representativeness of TRI and DMR data for an OES as a whole is dependent on (1) the extent to
which these reporting mechanisms capture all potential sites within the OES and, (2) the extent to which
the release quantities provided by reporting sites reflect releases from non-reporting sites.
For some OES, the total number of sites was determined from TRI. For these OESs, there is uncertainty
in if there may be additional sites using the chemical within the OES that did not report to TRI (e.g., due
to being below reporting thresholds). For some OES, such as manufacturing and other OES involving
larger industrial sites, TRI is more likely to capture the majority of potential sites because these sites
typically meet the reporting threshold. For other OES, such as functional fluids (open-systems), 3D
printing, and other OES that may be performed at a range of different scales, the extent to which TRI
captures all potential sites is more uncertain because not all sites may meet the reporting threshold. This
uncertainty may result in an underestimate in the total number of sites using the chemical within the
OES. To the extent additional sites are not captured, releases may be underestimated; however, the
magnitude of this underestimation is unknown for each OES. In some cases, the total number of
facilities for a given OES was estimated using data from the U.S. Census. This may result in an
overestimate of the actual number of sites that use the chemical for that OES.
Page 295 of 484
-------
7646
7647
7648
7649
7650
7651
7652
7653
7654
7655
7656
7657
7658
7659
7660
7661
7662
7663
7664
7665
7666
7667
7668
7669
7670
7671
7672
7673
7674
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Additionally, it is unknown how representative release estimates from TRI and DMR reporting sites
accurately reflect all releases from within an OES since releases from non-reporting sites cannot be
quantified. Specifically, where the number of sites was estimated from U.S. Census data, the average
daily release calculated from sites reporting to TRI or DMR was applied to the total number of sites
reported in (U.S. Census Bureau. ^ ). It is uncertain how accurate this average release is to actual
releases at these sites; therefore, releases may be higher or lower than the calculated amount.
Estimating Emissions for OES without TRI Data
For release estimates developed for an OES when directly applicable TRI or DMR data were not
available, there are uncertainties related to the use of surrogate TRI or DMR data or, in their absence,
the use of modeling.
Use of surrogate TRI or DMR data may introduce uncertainties related to the extent to which the
surrogate OES and the OES being assessed are similar. Thus, the representativeness of the surrogate
release data towards the actual releases for the OES being assessed is uncertain.
Although no new models were developed for this release assessment, the adaptations made to and uses
of these models as part of the analysis (e.g., varying input parameters, Monte Carlo simulation) may
result in release estimates higher or lower than the actual amount. EPA presented central tendency (50th
percentile) and high-end (95th percentile) modeled release values to capture a range of potential releases
and reduce the uncertainty associated with using a single release estimate. Additionally, the approaches
used for estimating releases based on modeling or literature differs from the facility-specific approach
used for OES for which TRI or DMR data were available (as described previously, the modeled
estimates are for a "generic site"). This may introduce uncertainties that differ from those of the
scenarios using TRI or DMR data, described above.
E.8 Weight of the Scientific Evidence Conclusions for Environmental
Releases
As discussed in Section 2.2.1.2, Table_Apx E-10 presents a summary of EPA's overall weight of the
scientific evidence conclusions for its release estimates for each of the assessed OES.
Page 296 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
7675 Table Apx E-10. Summary of Overall Weight of the Scientific Evidence Conclusions for Environmental Release Estimates by PES
OES
Monitoring"
Modeling
Weight of the Seientitle Evidence
Conclusion
Notes
Air
Water
Land
Data
Quality
Rating
Air
Water
Land
Data
Quality
Rating
Air
Water
Land
Manufacturing
/
•
M/H
X
X
X
N/A
Moderate
to Robust
Moderate to
Robust
Moderate
to Robust
Based on TRI and DMR which have
"medium" data quality ratings.
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited, but uncertainties are known.
Import and
Repackaging
•
M/H
X
X
X
N/A
Moderate
to Robust
Moderate to
Robust
Moderate
to Robust
Based on TRI and DMR which have
"medium" data quality ratings.
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited, but uncertainties are known.
Industrial Uses
/
/
•
M/H
X
X
X
N/A
Moderate
to Robust
Moderate to
Robust
Moderate
to Robust
Based on TRI and DMR which have
"medium" data quality ratings.
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited, but uncertainties are known.
Functional
Fluids (Open-
System)
•
M/H
X
X
X
N/A
Moderate
to Robust
Moderate to
Robust
Moderate
to Robust
Based on TRI and DMR which have
"medium" data quality ratings.
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited, but uncertainties are known.
Functional
Fluids (Closed-
System)
Assessed as a part of Industrial Uses OES
Slight
Slight
Slight
No data were available to estimate
releases for this OES, so it was grouped
with Industrial Uses OES. There is
uncertainty in the representativeness of
the Industrial Uses data for this OES.
Laboratory
Chemicals
X
X
X
N/A
H
Slight to
Moderate
Slight to
Moderate
Slight to
Moderate
Assessed using Laboratory Chemicals GS
which has a "high" data quality rating.
Activities could vary drastically on a site-
by-site basis due to uncertainties and
limitations in the model.
Film Cement
X
X
X
N/A
Not
expected
H
Slight to
Moderate
Slight to
Moderate
Slight to
Moderate
The underlying data sources for model
parameters have a "high" data quality
rating. Modeling may not be sufficiently
representative of all the sites for this OES.
Spray Foam
Application
X
X
X
N/A
M
Slight to
Moderate
Slight to
Moderate
Slight to
Moderate
Assessed using SPF GS which has a
"medium" data quality rating. Activities
could vary drastically on a site-by-site
basis due to uncertainties and limitations
in the model.
Page 297 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Monitoring"
Modeling
Weight of the Seientitle Evidence
Conclusion
Notes
Air
Water
Land
Data
Quality
Rating
Air
Water
Land
Data
Quality
Rating
Air
Water
Land
Printing Inks
(3D)
Assumed
included
in
Industria
1 Uses
OES
/
Assumed
included
in
Industrial
Uses OES
M/H
X
X
X
N/A
Slight
Moderate to
Robust
Slight
Based on TRI and DMR which have
"medium" data quality ratings.
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited. There is uncertainty in the
representativeness of the Industrial Uses
data for this OES.
Dry Film
Lubricant
X
X
X
N/A
/
Not
expected
/
H
Slight to
Moderate
Slight to
Moderate
Slight to
Moderate
The underlying data sources for model
parameters have a "high" data quality
rating. Modeling may not be sufficiently
representative of all the sites for this OES.
Disposal
/
/
•
M/H
X
X
X
N/A
Moderate
to Robust
Moderate to
Robust
Moderate
to Robust
Based on TRI and DMR which have
"medium" data quality ratings.
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited, but uncertainties are known.
Textile Dye
X
X
X
N/A
Not
assess
ed
/
/
M
Indetermi
nate
Moderate
Moderate
Assessed using ESD on Textile Dyes,
which has a "medium" data quality rating.
Monte Carlo modeling allows for
parameter variation; however, the
modeling may not be sufficiently
representative of all the sites for this OES.
Antifreeze
X
X
X
N/A
Not
expected
/
H
Slight to
Moderate
Slight to
Moderate
Slight to
Moderate
Assessed using process information from
GSs with "high" data quality ratings.
Activities could vary drastically on a site-
by-site basis due to uncertainties and
limitations in the model.
Surface Cleaner
X
X
X
N/A
/
/
H
Slight
Slight
Slight
Assessed using SHEDs-HT data for the
Liverpool OH case study and the
Furnishing Cleaning GS, which has a data
quality rating of "high." There is
uncertainty in the application of this
modeling for a commercial setting, and
this case study does not represent all sites
in this OES.
Dish Soap
X
X
X
N/A
Not
assess
ed
N/A
Indetermi
nate
Slight
Slight
Assessed using SHEDs-HT model for the
Liverpool OH case study and standard
EPA/OPPT models. There is uncertainty
in the application of this modeling for a
Page 298 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Monitoring"
Modeling
Weight of the Seientitie Evidence
Conclusion
OES
Data
Data
Notes
Air
Water
Land
Quality
Rating
Air
Water
Land
Quality
Rating
Air
Water
Land
commercial setting, and this case study
does not represent all sites in this OES.
Dishwasher
X
X
X
N/A
Not
/
/
N/A
Indetermi
Slight
Slight
Assessed using SHEDs-HT data for the
Detergent
assess
ed
nate
Liverpool OH case study and standard
EPA/OPPT models. There is uncertainty
in the application of this modeling for a
commercial setting, and this case study
does not represent all sites in this OES.
Laundry
X
X
X
N/A
/
/
M
Moderate
Moderate
Moderate
Assessed using Laundries ESD, which has
Detergent
(Industrial)
a "medium" data quality rating. Monte
Carlo modeling allows for parameter
variation; however, the modeling may not
be sufficiently representative of all the
sites for this OES.
Laundry
X
X
X
N/A
/
/
/
M
Moderate
Moderate
Moderate
Assessed using Laundries ESD, which has
Detergent
(Institutional)
a "medium" data quality rating. Monte
Carlo modeling allows for parameter
variation; however, the modeling may not
be sufficiently representative of all the
sites for this OES.
Paints and Floor
Lacquer
X
X
X
N/A
/
Not
expected
/
M
Slight to
Moderate
Slight to
Moderate
Slight to
Moderate
Assessed using Automotive Spray
Painting ESD, which has a "medium" data
quality rating. Modeling may not be
sufficiently representative of all the sites
for this exposure scenario.
PET Byproduct
/
/
/
M/H
X
X
X
N/A
Moderate
to Robust
Moderate to
Robust
Moderate
to Robust
Based on TRI and DMR which have
"medium" data quality ratings.
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited, but uncertainties are known.
Ethoxylation
/
/
/
M/H
X
X
X
N/A
Moderate
Moderate to
Moderate
Based on TRI and DMR which have
Process
to Robust
Robust
to Robust
"medium" data quality ratings.
Byproduct
Information on the conditions of use of
1,4-dioxane at facilities in TRI and DMR
is limited, but uncertainties are known.
Hydraulic
M
M
Moderate
Moderate to
Moderate
Based on FracFocus 3.0 and the Hydraulic
Fracturing
to Robust
Robust
to Robust
Fracturing ESD, which has a "medium"
data quality rating. Monte Carlo modeling
allows for parameter variation; however,
Page 299 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
OES
Monitoring"
Modeling
Weight of the Seientitle Evidence
Conclusion
Notes
Air
Water
Land
Data
Quality
Rating
Air
Water
Land
Data
Quality
Rating
Air
Water
Land
the modeling may not be sufficiently
representative of all the sites for this OES.
7676
7677
7678
7679
7680
E.9 TRI to CDR Crosswalk
Table Apx E-l 1 presents the TRI-CDR Crosswalk used to map facilities to the OES for each chemical. Blanks in the 2016 CDR code column
indicate there is no corresponding CDR code that matches the TRI code.
Table Apx E-ll. TRI-CDR Use Code Crosswalk
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.1.a
Manufacture:
Produce
N/A
N/A
N/A
N/A
N/A
3.1.b
Manufacture:
Import
N/A
N/A
N/A
N/A
N/A
3.1.c
Manufacture: For
on-site
use/processing
N/A
N/A
N/A
N/A
N/A
3.1.d
Manufacture: For
sale/distribution
N/A
N/A
N/A
N/A
N/A
3.1.e
Manufacture: As
a byproduct
N/A
N/A
N/A
N/A
N/A
3.1.f
Manufacture: As
an impurity
N/A
N/A
N/A
N/A
N/A
3.2.a
Processing: As a
reactant
N/A
N/A
PC
Processing as a
reactant
Chemical substance is used in chemical reactions for the
manufacturing of another chemical substance or product.
3.2.a
Processing: As a
reactant
P101
Feedstocks
N/A
N/A
N/A
3.2.a
Processing: As a
reactant
P102
Raw Materials
N/A
N/A
N/A
Page 300 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.2.a
Processing: As a
reactant
P103
Intermediates
U015
Intermediates
Chemical substances consumed in a reaction to produce
other chemical substances for commercial advantage. A
residual of the intermediate chemical substance which has
no separate function may remain in the reaction product.
3.2.a
Processing: As a
reactant
P104
Initiators
U024
Process regulators
Chemical substances used to change the rate of a
chemical reaction, start or stop the reaction, or otherwise
influence the course of the reaction. Process regulators
may be consumed or become part of the reaction product.
3.2.a
Processing: As a
reactant
P199
Other
U016
Ion exchange agents
Chemical substances, usually in the form of a solid
matrix, that are used to selectively remove targeted ions
from a solution. Examples generally consist of an inert
hydrophobic matrix such as styrenedivinylbenzene or
phenol-formaldehyde, cross-linking polymer such as
divinylbenzene, and ionic functional groups including
sulfonic, carboxylic or phosphonic acids. This code also
includes aluminosilicate zeolites.
3.2.a
Processing: As a
reactant
P199
Other
U019
Oxidizing/
reducing agent
Chemical substances used to alter the valence state of
another substance by donating or accepting electrons or
by the addition or removal of hydrogen to a substance.
Examples of oxidizing agents include nitric acid,
perchlorates, hexavalent chromium compounds, and
peroxydisulfuric acid salts. Examples of reducing agents
include hydrazine, sodium thiosulfate, and coke produced
from coal.
3.2.a
Processing: As a
reactant
P199
Other
U999
Other (specify)
Chemical substances used in a way other than those
described by other codes.
3.2.b
Processing: As a
formulation
component
N/A
N/A
PF
Processing-
incorporation into
formulation, mixture,
or reaction product
Chemical substance is added to a product (or product
mixture) prior to further distribution of the product.
3.2.b
Processing: As a
formulation
component
P201
Additives
U007
Corrosion inhibitors
and anti-scaling
agents
Chemical substances used to prevent or retard corrosion
or the formation of scale. Examples include
phenylenediamine, chromates, nitrates, phosphates, and
hydrazine.
Page 301 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.2.b
Processing: As a
formulation
component
P201
Additives
U009
Fillers
Chemical substances used to provide bulk, increase
strength, increase hardness, or improve resistance to
impact. Fillers incorporated in a matrix reduce production
costs by minimizing the amount of more expensive
substances used in the production of articles. Examples
include calcium carbonate, barium sulfate, silicates, clays,
zinc oxide and aluminum oxide.
3.2.b
Processing: As a
formulation
component
P201
Additives
U010
Finishing agents
Chemical substances used to impart such functions as
softening, static proofing, wrinkle resistance, and water
repellence. Substances may be applied to textiles, paper,
and leather. Examples include quaternary ammonium
compounds, ethoxylated amines, and silicone compounds.
3.2.b
Processing: As a
formulation
component
P201
Additives
U017
Lubricants and
lubricant additives
Chemical substances used to reduce friction, heat, or wear
between moving parts or adjacent solid surfaces, or that
enhance the lubricity of other substances. Examples of
lubricants include mineral oils, silicate and phosphate
esters, silicone oil, greases, and solid film lubricants such
as graphite and PTFE. Examples of lubricant additives
include molybdenum disulphide and tungsten disulphide.
3.2.b
Processing: As a
formulation
component
P201
Additives
U034
Paint additives and
coating additives not
described by other
codes
Chemical substances used in a paint or coating
formulation to enhance properties such as water
repellence, increased gloss, improved fade resistance,
ease of application, foam prevention, etc. Examples of
paint additives and coating additives include polyols,
amines, vinyl acetate ethylene emulsions, and aliphatic
polyisocyanates.
3.2.b
Processing: As a
formulation
component
P202
Dyes
U008
Dyes
Chemical substances used to impart color to other
materials or mixtures (i.e., substrates) by penetrating the
surface of the substrate. Example types include azo,
anthraquinone, amino azo, aniline, eosin, stilbene, acid,
basic or cationic, reactive, dispersive, and natural dyes.
3.2.b
Processing: As a
formulation
component
P202
Dyes
U021
Pigments
Chemical substances used to impart color to other
materials or mixtures (i.e., substrates) by attaching
themselves to the surface of the substrate through binding
Page 302 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
or adhesion. This code includes fluorescent agents,
luminescent agents, whitening agents, pearlizing agents,
and opacifiers. Examples include metallic oxides of iron,
titanium, zinc, cobalt, and chromium; metal powder
suspensions; lead chromates; vegetable and animal
products; and synthetic organic pigments.
3.2.b
Processing: As a
formulation
component
P203
Reaction
Diluents
U030
Solvents (which
become part of
product formulation or
mixture)
Chemical substances used to dissolve another substance
(solute) to form a uniformly dispersed mixture (solution)
at the molecular level. Examples include diluents used to
reduce the concentration of an active material to achieve a
specified effect and low gravity materials added to reduce
cost.
3.2.b
Processing: As a
formulation
component
P203
Reaction
Diluents
U032
Viscosity adjustors
Chemical substances used to alter the viscosity of another
substance. Examples include viscosity index (VI)
improvers, pour point depressants, and thickeners.
3.2.b
Processing: As a
formulation
component
P204
Initiators
U024
Process regulators
Chemical substances used to change the rate of a
chemical reaction, start or stop the reaction, or otherwise
influence the course of the reaction. Process regulators
may be consumed or become part of the reaction product.
3.2.b
Processing: As a
formulation
component
P205
Solvents
U030
Solvents (which
become part of
product formulation or
mixture)
Chemical substances used to dissolve another substance
(solute) to form a uniformly dispersed mixture (solution)
at the molecular level. Examples include diluents used to
reduce the concentration of an active material to achieve a
specified effect and low gravity materials added to reduce
cost.
3.2.b
Processing: As a
formulation
component
P206
Inhibitors
U024
Process regulators
Chemical substances used to change the rate of a
chemical reaction, start or stop the reaction, or otherwise
influence the course of the reaction. Process regulators
may be consumed or become part of the reaction product.
3.2.b
Processing: As a
formulation
component
P207
Emulsifiers
U003
Adsorbents and
absorbents
Chemical substances used to retain other substances by
accumulation on their surface or by assimilation.
Examples of adsorbents include silica gel, activated
alumina, and activated carbon. Examples of absorbents
include straw oil, alkaline solutions, and kerosene.
Page 303 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.2.b
Processing: As a
formulation
component
P208
Surfactants
U002
Adhesives and sealant
chemicals
Chemical substances used to promote bonding between
other substances, promote adhesion of surfaces, or
prevent seepage of moisture or air. Examples include
epoxides, isocyanates, acrylamides, phenol, urea,
melamine, and formaldehyde.
3.2.b
Processing: As a
formulation
component
P208
Surfactants
U023
Plating agents and
surface treating agents
Chemical substances applied to metal, plastic, or other
surfaces to alter physical or chemical properties of the
surface. Examples include metal surface treating agents,
strippers, etchants, rust and tarnish removers, and
descaling agents.
3.2.b
Processing: As a
formulation
component
P208
Surfactants
U031
Surface active agents
Chemical substances used to modify surface tension when
dissolved in water or water solutions or reduce interfacial
tension between two liquids or between a liquid and a
solid or between liquid and air. Examples include
carboxylates, sulfonates, phosphates, carboxylic acid,
esters, and quaternary ammonium salts.
3.2.b
Processing: As a
formulation
component
P209
Lubricants
U017
Lubricants and
lubricant additives
Chemical substances used to reduce friction, heat, or wear
between moving parts or adjacent solid surfaces, or that
enhance the lubricity of other substances. Examples of
lubricants include mineral oils, silicate and phosphate
esters, silicone oil, greases, and solid film lubricants such
as graphite and PTFE. Examples of lubricant additives
include molybdenum disulphide and tungsten disulphide.
3.2.b
Processing: As a
formulation
component
P210
Flame
Retardants
U011
Flame retardants
Chemical substances used on the surface of or
incorporated into combustible materials to reduce or
eliminate their tendency to ignite when exposed to heat or
a flame for a short period of time. Examples include
inorganic salts, chlorinated or brominated organic
compounds, and organic phosphates/phosphonates.
3.2.b
Processing: As a
formulation
component
P211
Rheological
Modifiers
U022
Plasticizers
Chemical substances used in plastics, cement, concrete,
wallboard, clay bodies, or other materials to increase their
plasticity or fluidity. Examples include phthalates,
trimellitates, adipates, maleates, and lignosulphonates.
Page 304 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.2.b
Processing: As a
formulation
component
P211
Rheological
Modifiers
U032
Viscosity adjustors
Chemical substances used to alter the viscosity of another
substance. Examples include viscosity index (VI)
improvers, pour point depressants, and thickeners.
3.2.b
Processing: As a
formulation
component
P299
Other
U003
Adsorbents and
absorbents
Chemical substances used to retain other substances by
accumulation on their surface or by assimilation.
Examples of adsorbents include silica gel, activated
alumina, and activated carbon. Examples of absorbents
include straw oil, alkaline solutions, and kerosene.
3.2.b
Processing: As a
formulation
component
P299
Other
U016
Ion exchange agents
Chemical substances, usually in the form of a solid
matrix, that are used to selectively remove targeted ions
from a solution. Examples generally consist of an inert
hydrophobic matrix such as styrenedivinylbenzene or
phenol-formaldehyde, cross-linking polymer such as
divinylbenzene, and ionic functional groups including
sulfonic, carboxylic or phosphonic acids. This code also
includes aluminosilicate zeolites.
3.2.b
Processing: As a
formulation
component
P299
Other
U018
Odor agents
Chemical substances used to control odors, remove odors,
mask odors, or impart odors. Examples include
benzenoids, terpenes and terpenoids, musk chemicals,
aliphatic aldehydes, aliphatic cyanides, and mercaptans.
3.2.b
Processing: As a
formulation
component
P299
Other
U019
Oxidizing/
reducing agent
Chemical substances used to alter the valence state of
another substance by donating or accepting electrons or
by the addition or removal of hydrogen to a substance.
Examples of oxidizing agents include nitric acid,
perchlorates, hexavalent chromium compounds, and
peroxydisulfuric acid salts. Examples of reducing agents
include hydrazine, sodium thiosulfate, and coke produced
from coal.
3.2.b
Processing: As a
formulation
component
P299
Other
U020
Photosensitive
chemicals
Chemical substances used for their ability to alter their
physical or chemical structure through absorption of light,
resulting in the emission of light, dissociation,
discoloration, or other chemical reaction. Examples
include sensitizers, fluorescents, photovoltaic agents,
ultraviolet absorbers, and ultraviolet stabilizers.
Page 305 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.2.b
Processing: As a
formulation
component
P299
Other
U027
Propellants and
blowing agents
Chemical substances used to dissolve or suspend other
substances and either to expel those substances from a
container in the form of an aerosol or to impart a cellular
structure to plastics, rubber, or thermo set resins.
Examples include compressed gasses and liquids and
substances which release ammonia, carbon dioxide, or
nitrogen.
3.2.b
Processing: As a
formulation
component
P299
Other
U028
Solid separation
agents
Chemical substances used to promote the separation of
suspended solids from a liquid. Examples include
flotation aids, flocculants, coagulants, dewatering aids,
and drainage aids.
3.2.b
Processing: As a
formulation
component
P299
Other
U999
Other (specify)
Chemical substances used in a way other than those
described by other codes.
3.2.c
Processing: As
an article
component
N/A
N/A
PA
Processing-
incorporation into
article
Chemical substance becomes an integral component of an
article distributed for industrial, trade, or consumer use.
3.2.c
Processing: As
an article
component
N/A
N/A
U008
Dyes
Chemical substances used to impart color to other
materials or mixtures (i.e., substrates) by penetrating into
the surface of the substrate. Examples types include azo,
anthraquinone, amino azo, aniline, eosin, stilbene, acid,
basic or cationic, reactive, dispersive, and natural dyes.
3.2.c
Processing: As
an article
component
N/A
N/A
U009
Fillers
Chemical substances used to provide bulk, increase
strength, increase hardness, or improve resistance to
impact. Fillers incorporated in a matrix reduce production
costs by minimizing the amount of more expensive
substances used in the production of articles. Examples
include calcium carbonate, barium sulfate, silicates, clays,
zinc oxide and aluminum oxide.
3.2.c
Processing: As
an article
component
N/A
N/A
U021
Pigments
Chemical substances used to impart color to other
materials or mixtures (i.e., substrates) by attaching
themselves to the surface of the substrate through binding
or adhesion. This code includes fluorescent agents,
luminescent agents, whitening agents, pearlizing agents,
Page 306 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
and opacifiers. Examples include metallic oxides of iron,
titanium, zinc, cobalt, and chromium; metal powder
suspensions; lead chromates; vegetable and animal
products; and synthetic organic pigments.
3.2.c
Processing: As
an article
component
N/A
N/A
U034
Paint additives and
coating additives not
described by other
codes
Chemical substances used in a paint or coating
formulation to enhance properties such as water
repellence, increased gloss, improved fade resistance,
ease of application, foam prevention, etc. Examples of
paint additives and coating additives include polyols,
amines, vinyl acetate ethylene emulsions, and aliphatic
polyisocyanates.
3.2.c
Processing: As
an article
component
N/A
N/A
U999
Other (specify)
Chemical substances used in a way other than those
described by other codes.
3.2.d
Processing:
Repackaging
N/A
N/A
PK
Processing-
repackaging
Preparation of a chemical substance for distribution in
commerce in a different form, state, or quantity. This
includes transferring the chemical substance from a bulk
container into smaller containers. This definition does not
apply to sites that only relabel or redistribute the
reportable chemical substance without removing the
chemical substance from the container in which it is
received or purchased.
3.2.e
Processing: As
an impurity
N/A
N/A
N/A
N/A
N/A
3.2.f
Processing:
Recycling
N/A
N/A
N/A
N/A
N/A
3.3.a
Otherwise Use:
As a chemical
processing aid
N/A
N/A
U
Use-non incorporative
Activities
Chemical substance is otherwise used (e.g., as a chemical
processing or manufacturing aid).
3.3.a
Otherwise Use:
As a chemical
processing aid
Z101
Process Solvents
U029
Solvents (for cleaning
or degreasing)
Chemical substances used to dissolve oils, greases, and
similar materials from textiles, glassware, metal surfaces,
and other articles. Examples include trichloroethylene,
perchloroethylene, methylene chloride, liquid carbon
dioxide, and n-propyl bromide.
Page 307 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.3.a
Otherwise Use:
As a chemical
processing aid
Z102
Catalysts
U020
Photosensitive
chemicals
Chemical substances used for their ability to alter their
physical or chemical structure through absorption of light,
resulting in the emission of light, dissociation,
discoloration, or other chemical reaction. Examples
include sensitizers, fluorescents, photovoltaic agents,
ultraviolet absorbers, and ultraviolet stabilizers.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z102
Catalysts
U025
Processing aids,
specific to petroleum
production
Chemical substances added to water-, oil-, or synthetic
drilling muds or other petroleum production fluids to
control viscosity, foaming, corrosion, alkalinity and pH,
microbiological growth, hydrate formation, etc., during
the production of oil, gas, and other products from
beneath the earth's surface.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z102
Catalysts
U026
Processing aids, not
otherwise listed
Chemical substances used to improve the processing
characteristics or the operation of process equipment or to
alter or buffer the pH of the substance or mixture, when
added to a process or to a substance or mixture to be
processed. Processing agents do not become a part of the
reaction product and are not intended to affect the
function of a substance or article created. Examples
include buffers, dehumidifiers, dehydrating agents,
sequestering agents, and chelators.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z103
Inhibitors
U024
Process regulators
Chemical substances used to change the rate of a
chemical reaction, start or stop the reaction, or otherwise
influence the course of the reaction. Process regulators
may be consumed or become part of the reaction product.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z103
Inhibitors
U025
Processing aids,
specific to petroleum
production
Chemical substances added to water-, oil-, or synthetic
drilling muds or other petroleum production fluids to
control viscosity, foaming, corrosion, alkalinity and pH,
microbiological growth, hydrate formation, etc., during
the production of oil, gas, and other products from
beneath the earth's surface.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z103
Inhibitors
U026
Processing aids, not
otherwise listed
Chemical substances used to improve the processing
characteristics or the operation of process equipment or to
alter or buffer the pH of the substance or mixture, when
Page 308 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
added to a process or to a substance or mixture to be
processed. Processing agents do not become a part of the
reaction product and are not intended to affect the
function of a substance or article created. Examples
include buffers, dehumidifiers, dehydrating agents,
sequestering agents, and chelators.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z104
Initiators
U024
Process regulators
Chemical substances used to change the rate of a
chemical reaction, start or stop the reaction, or otherwise
influence the course of the reaction. Process regulators
may be consumed or become part of the reaction product.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z104
Initiators
U025
Processing aids,
specific to petroleum
production
Chemical substances added to water-, oil-, or synthetic
drilling muds or other petroleum production fluids to
control viscosity, foaming, corrosion, alkalinity and pH,
microbiological growth, hydrate formation, etc., during
the production of oil, gas, and other products from
beneath the earth's surface.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z104
Initiators
U026
Processing aids, not
otherwise listed
Chemical substances used to improve the processing
characteristics or the operation of process equipment or to
alter or buffer the pH of the substance or mixture, when
added to a process or to a substance or mixture to be
processed. Processing agents do not become a part of the
reaction product and are not intended to affect the
function of a substance or article created. Examples
include buffers, dehumidifiers, dehydrating agents,
sequestering agents, and chelators.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z105
Reaction
Terminators
U024
Process regulators
Chemical substances used to change the rate of a
chemical reaction, start or stop the reaction, or otherwise
influence the course of the reaction. Process regulators
may be consumed or become part of the reaction product.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z105
Reaction
Terminators
U025
Processing aids,
specific to petroleum
production
Chemical substances added to water-, oil-, or synthetic
drilling muds or other petroleum production fluids to
control viscosity, foaming, corrosion, alkalinity and pH,
microbiological growth, hydrate formation, etc., during
Page 309 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
the production of oil, gas, and other products from
beneath the earth's surface.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z105
Reaction
Terminators
U026
Processing aids, not
otherwise listed
Chemical substances used to improve the processing
characteristics or the operation of process equipment or to
alter or buffer the pH of the substance or mixture, when
added to a process or to a substance or mixture to be
processed. Processing agents do not become a part of the
reaction product and are not intended to affect the
function of a substance or article created. Examples
include buffers, dehumidifiers, dehydrating agents,
sequestering agents, and chelators.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z106
Solution Buffers
U026
Processing aids, not
otherwise listed
Chemical substances used to improve the processing
characteristics or the operation of process equipment or to
alter or buffer the pH of the substance or mixture, when
added to a process or to a substance or mixture to be
processed. Processing agents do not become a part of the
reaction product and are not intended to affect the
function of a substance or article created. Examples
include buffers, dehumidifiers, dehydrating agents,
sequestering agents, and chelators.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z199
Other
U002
Adhesives and sealant
chemicals
Chemical substances used to promote bonding between
other substances, promote adhesion of surfaces, or
prevent seepage of moisture or air. Examples include
epoxides, isocyanates, acrylamides, phenol, urea,
melamine, and formaldehyde.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z199
Other
U006
Bleaching agents
Chemical substances used to lighten or whiten a substrate
through chemical reaction, usually an oxidative process
which degrades the color system. Examples generally fall
into one of two groups: chlorine containing bleaching
agents (e.g., chlorine, hypochlorites, N-chloro compounds
and chlorine dioxide); and peroxygen bleaching agents
(e.g., hydrogen peroxide, potassium permanganate, and
sodium perborate).
Page 310 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.3.a
Otherwise Use:
As a chemical
processing aid
Z199
Other
U018
Odor agents
Chemical substances used to control odors, remove odors,
mask odors, or impart odors. Examples include
benzenoids, terpenes and terpenoids, musk chemicals,
aliphatic aldehydes, aliphatic cyanides, and mercaptans.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z199
Other
U023
Plating agents and
surface treating agents
Chemical substances applied to metal, plastic, or other
surfaces to alter physical or chemical properties of the
surface. Examples include metal surface treating agents,
strippers, etchants, rust and tarnish removers, and
descaling agents.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z199
Other
U025
Processing aids,
specific to petroleum
production
Chemical substances added to water-, oil-, or synthetic
drilling muds or other petroleum production fluids to
control viscosity, foaming, corrosion, alkalinity and pH,
microbiological growth, hydrate formation, etc., during
the production of oil, gas, and other products from
beneath the earth's surface.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z199
Other
U026
Processing aids, not
otherwise listed
Chemical substances used to improve the processing
characteristics or the operation of process equipment or to
alter or buffer the pH of the substance or mixture, when
added to a process or to a substance or mixture to be
processed. Processing agents do not become a part of the
reaction product and are not intended to affect the
function of a substance or article created. Examples
include buffers, dehumidifiers, dehydrating agents,
sequestering agents, and chelators.
3.3.a
Otherwise Use:
As a chemical
processing aid
Z199
Other
U028
Solid separation
agents
Chemical substances used to promote the separation of
suspended solids from a liquid. Examples include
flotation aids, flocculants, coagulants, dewatering aids,
and drainage aids.
3.3.b
Otherwise Use:
As a
manufacturing
aid
N/A
N/A
U
Use-non
incorporative
Activities
Chemical substance is otherwise used (e.g., as a chemical
processing or manufacturing aid).
3.3.b
Otherwise Use:
As a
Z201
Process
Lubricants
U017
Lubricants and
lubricant additives
Chemical substances used to reduce friction, heat, or wear
between moving parts or adjacent solid surfaces, or that
Page 311 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
manufacturing
aid
enhance the lubricity of other substances. Examples of
lubricants include mineral oils, silicate and phosphate
esters, silicone oil, greases, and solid film lubricants such
as graphite and PTFE. Examples of lubricant additives
include molybdenum disulphide and tungsten disulphide.
3.3.b
Otherwise Use:
As a
manufacturing
aid
Z202
Metalworking
Fluids
U007
Corrosion inhibitors
and antiscaling agents
Chemical substances used to prevent or retard corrosion
or the formation of scale. Examples include
phenylenediamine, chromates, nitrates, phosphates, and
hydrazine.
3.3.b
Otherwise Use:
As a
manufacturing
aid
Z202
Metalworking
Fluids
U014
Functional fluids
(open systems)
Liquid or gaseous chemical substances used for one or
more operational properties in an open system. Examples
include antifreezes and de-icing fluids such as ethylene
and propylene glycol, sodium formate, potassium acetate,
and sodium acetate. This code also includes substances
incorporated into metal working fluids.
3.3.b
Otherwise Use:
As a
manufacturing
aid
Z203
Coolants
U013
Functional fluids
(closed systems)
Liquid or gaseous chemical substances used for one or
more operational properties in a closed system. Examples
include: heat transfer agents (e.g., coolants and
refrigerants) such as polyalkylene glycols, silicone oils,
liquified propane, and carbon dioxide;
hydraulic/transmission fluids such as mineral oils,
organophosphate esters, silicone, and propylene glycol;
and dielectric fluids such as mineral insulating oil and
high flash point kerosene. This code does not include
fluids used as lubricants.
3.3.b
Otherwise Use:
As a
manufacturing
aid
Z204
Refrigerants
U013
Functional fluids
(closed systems)
Liquid or gaseous chemical substances used for one or
more operational properties in a closed system. Examples
include: heat transfer agents (e.g., coolants and
refrigerants) such as polyalkylene glycols, silicone oils,
liquified propane, and carbon dioxide;
hydraulic/transmission fluids such as mineral oils,
organophosphate esters, silicone, and propylene glycol;
and dielectric fluids such as mineral insulating oil and
Page 312 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
high flash point kerosene. This code does not include
fluids used as lubricants.
3.3.b
Otherwise Use:
As a
manufacturing
aid
Z205
Hydraulic Fluids
U013
Functional fluids
(closed systems)
Liquid or gaseous chemical substances used for one or
more operational properties in a closed system. Examples
include: heat transfer agents (e.g., coolants and
refrigerants) such as polyalkylene glycols, silicone oils,
liquified propane, and carbon dioxide;
hydraulic/transmission fluids such as mineral oils,
organophosphate esters, silicone, and propylene glycol;
and dielectric fluids such as mineral insulating oil and
high flash point kerosene. This code does not include
fluids used as lubricants.
3.3.b
Otherwise Use:
As a
manufacturing
aid
Z299
Other
U013
Functional fluids
(closed systems)
Liquid or gaseous chemical substances used for one or
more operational properties in a closed system. Examples
include: heat transfer agents (e.g., coolants and
refrigerants) such as polyalkylene glycols, silicone oils,
liquified propane, and carbon dioxide;
hydraulic/transmission fluids such as mineral oils,
organophosphate esters, silicone, and propylene glycol;
and dielectric fluids such as mineral insulating oil and
high flash point kerosene. This code does not include
fluids used as lubricants.
3.3.b
Otherwise Use:
As a
manufacturing
aid
Z299
Other
U023
Plating agents and
surface treating agents
Chemical substances applied to metal, plastic, or other
surfaces to alter physical or chemical properties of the
surface. Examples include metal surface treating agents,
strippers, etchants, rust and tarnish removers, and
descaling agents.
3.3.c
Otherwise Use:
Ancillary or
other use
N/A
N/A
U
Use-non
incorporative
Activities
Chemical substance is otherwise used (e.g., as a chemical
processing or manufacturing aid).
3.3.c
Otherwise Use:
Ancillary or
other use
Z301
Cleaner
U007
Corrosion inhibitors
and antiscaling agents
Chemical substances used to prevent or retard corrosion
or the formation of scale. Examples include
phenylenediamine, chromates, nitrates, phosphates, and
hydrazine.
Page 313 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.3.c
Otherwise Use:
Ancillary or
other use
Z301
Cleaner
U029
Solvents (for cleaning
or degreasing)
Chemical substances used to dissolve oils, greases, and
similar materials from textiles, glassware, metal surfaces,
and other articles. Examples include trichloroethylene,
perchloroethylene, methylene chloride, liquid carbon
dioxide, and n-propyl bromide.
3.3.c
Otherwise Use:
Ancillary or
other use
Z302
Degreaser
U003
Adsorbents and
Absorbents
Chemical substances used to retain other substances by
accumulation on their surface or by assimilation.
Examples of adsorbents include silica gel, activated
alumina, and activated carbon. Examples of absorbents
include straw oil, alkaline solutions, and kerosene.
3.3.c
Otherwise Use:
Ancillary or
other use
Z302
Degreaser
U029
Solvents (for cleaning
or degreasing)
Chemical substances used to dissolve oils, greases, and
similar materials from textiles, glassware, metal surfaces,
and other articles. Examples include trichloroethylene,
perchloroethylene, methylene chloride, liquid carbon
dioxide, and n-propyl bromide.
3.3.c
Otherwise Use:
Ancillary or
other use
Z303
Lubricant
U017
Lubricants and
lubricant additives
Chemical substances used to reduce friction, heat, or wear
between moving parts or adjacent solid surfaces, or that
enhance the lubricity of other substances. Examples of
lubricants include mineral oils, silicate and phosphate
esters, silicone oil, greases, and solid film lubricants such
as graphite and PTFE. Examples of lubricant additives
include molybdenum disulphide and tungsten disulphide.
3.3.c
Otherwise Use:
Ancillary or
other use
Z304
Fuel
U012
Fuels and fuel
additives
Chemical substances used to create mechanical or thermal
energy through chemical reactions, or which are added to
a fuel for the purpose of controlling the rate of reaction or
limiting the production of undesirable combustion
products, or which provide other benefits such as
corrosion inhibition, lubrication, or detergency. Examples
of fuels include coal, oil, gasoline, and various grades of
diesel fuel. Examples of fuel additives include
oxygenated compound such as ethers and alcohols,
antioxidants such as phenylenediamines and hindered
phenols, corrosion inhibitors such as carboxylic acids,
Page 314 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
amines, and amine salts, and blending agents such as
ethanol.
3.3.c
Otherwise Use:
Ancillary or
other use
Z305
Flame Retardant
U011
Flame retardants
Chemical substances used on the surface of or
incorporated into combustible materials to reduce or
eliminate their tendency to ignite when exposed to heat or
a flame for a short period of time. Examples include
inorganic salts, chlorinated or brominated organic
compounds, and organic phosphates/phosphonates.
3.3.c
Otherwise Use:
Ancillary or
other use
Z306
Waste
Treatment
U006
Bleaching agents
Chemical substances used to lighten or whiten a substrate
through chemical reaction, usually an oxidative process
which degrades the color system. Examples generally fall
into one of two groups: chlorine containing bleaching
agents (e.g., chlorine, hypochlorites, N-chloro compounds
and chlorine dioxide); and peroxygen bleaching agents
(e.g., hydrogen peroxide, potassium permanganate, and
sodium perborate).
3.3.c
Otherwise Use:
Ancillary or
other use
Z306
Waste
Treatment
U018
Odor agents
Chemical substances used to control odors, remove odors,
mask odors, or impart odors. Examples include
benzenoids, terpenes and terpenoids, musk chemicals,
aliphatic aldehydes, aliphatic cyanides, and mercaptans.
3.3.c
Otherwise Use:
Ancillary or
other use
Z306
Waste
Treatment
U019
Oxidizing/reducing
agent
Chemical substances used to alter the valence state of
another substance by donating or accepting electrons or
by the addition or removal of hydrogen to a substance.
Examples of oxidizing agents include nitric acid,
perchlorates, hexavalent chromium compounds, and
peroxydisulfuric acid salts. Examples of reducing agents
include hydrazine, sodium thiosulfate, and coke produced
from coal.
3.3.c
Otherwise Use:
Ancillary or
other use
Z306
Waste
Treatment
U028
Solid separation
agents
Chemical substances used to promote the separation of
suspended solids from a liquid. Examples include
flotation aids, flocculants, coagulants, dewatering aids,
and drainage aids.
Page 315 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
3.3.c
Otherwise Use:
Ancillary or
other use
Z307
Water Treatment
U006
Bleaching agents
Chemical substances used to lighten or whiten a substrate
through chemical reaction, usually an oxidative process
which degrades the color system. Examples generally fall
into one of two groups: chlorine containing bleaching
agents (e.g., chlorine, hypochlorites, N-chloro compounds
and chlorine dioxide); and, peroxygen bleaching agents
(e.g., hydrogen peroxide, potassium permanganate, and
sodium perborate).
3.3.c
Otherwise Use:
Ancillary or
other use
Z307
Water Treatment
U018
Odor agents
Chemical substances used to control odors, remove odors,
mask odors, or impart odors. Examples include
benzenoids, terpenes and terpenoids, musk chemicals,
aliphatic aldehydes, aliphatic cyanides, and mercaptans.
3.3.c
Otherwise Use:
Ancillary or
other use
Z307
Water Treatment
U019
Oxidizing/reducing
agent
Chemical substances used to alter the valence state of
another substance by donating or accepting electrons or
by the addition or removal of hydrogen to a substance.
Examples of oxidizing agents include nitric acid,
perchlorates, hexavalent chromium compounds, and
peroxydisulfuric acid salts. Examples of reducing agents
include hydrazine, sodium thiosulfate, and coke produced
from coal.
3.3.c
Otherwise Use:
Ancillary or
other use
Z307
Water Treatment
U028
Solid separation
agents
Chemical substances used to promote the separation of
suspended solids from a liquid. Examples include
flotation aids, flocculants, coagulants, dewatering aids,
and drainage aids.
3.3.c
Otherwise Use:
Ancillary or
other use
Z308
Construction
Materials
N/A
N/A
N/A
3.3.c
Otherwise Use:
Ancillary or
other use
Z399
Other
U001
Abrasives
Chemical substances used to wear down or polish
surfaces by rubbing against the surface. Examples include
sandstones, pumice, silex, quartz, silicates, aluminum
oxides, and glass.
3.3.c
Otherwise Use:
Ancillary or
other use
Z399
Other
U013
Functional fluids
(closed systems)
Liquid or gaseous chemical substances used for one or
more operational properties in a closed system. Examples
include: heat transfer agents (e.g., coolants and
Page 316 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
TRI
Section
TRI Description
TRI Sub-
use Code
TRI Sub-use
Code Name
2016
CDR
Code
2016 CDR Code
Name
2016 CDR Functional Use Definition
refrigerants) such as polyalkylene glycols, silicone oils,
liquified propane, and carbon dioxide;
hydraulic/transmission fluids such as mineral oils,
organophosphate esters, silicone, and propylene glycol;
and dielectric fluids such as mineral insulating oil and
high flash point kerosene. This code does not include
fluids used as lubricants.
3.3.c
Otherwise Use:
Ancillary or
other use
Z399
Other
U014
Functional fluids
(open systems)
Liquid or gaseous chemical substances used for one or
more operational properties in an open system. Examples
include antifreezes and de-icing fluids such as ethylene
and propylene glycol, sodium formate, potassium acetate,
and, sodium acetate. This code also includes substances
incorporated into metal working fluids.
3.3.c
Otherwise Use:
Ancillary or
other use
Z399
Other
U018
Odor agents
Chemical substances used to control odors, remove odors,
mask odors, or impart odors. Examples include
benzenoids, terpenes and terpenoids, musk chemicals,
aliphatic aldehydes, aliphatic cyanides, and mercaptans.
3.3.c
Otherwise Use:
Ancillary or
other use
Z399
Other
U020
Photosensitive
chemicals
Chemical substances used for their ability to alter their
physical or chemical structure through absorption of light,
resulting in the emission of light, dissociation,
discoloration, or other chemical reaction. Examples
include sensitizers, fluorescents, photovoltaic agents,
ultraviolet absorbers, and ultraviolet stabilizers.
3.3.c
Otherwise Use:
Ancillary or
other use
Z399
Other
U023
Plating agents and
surface treating agents
Chemical substances applied to metal, plastic, or other
surfaces to alter physical or chemical properties of the
surface. Examples include metal surface treating agents,
strippers, etchants, rust and tarnish removers, and
descaling agents.
7681
7682
Page 317 of 484
-------
7683
7684
7685
7686
7687
7688
7689
7690
7691
7692
7693
7694
7695
7696
7697
7698
7699
7700
7701
7702
7703
7704
7705
7706
7707
7708
7709
7710
7711
7712
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.10 Developing Models that Use Monte Carlo Methods
This appendix provides background information on Monte Carlo methods, including an overview of
deterministic and stochastic processes, an overview of the implementation of Monte Carlo methods, and
a discussion of EPA's approach for building models that utilized Monte Carlo methods.
This appendix is only intended to provide general background information; information related to the
specific models for which EPA implemented Monte Carlo methods is included in Appendices E. 11
through E.13 and Appendices F.5 through F.9.
E.10.1 Background on Monte Carlo Methods
A deterministic process has a single output (or set of outputs) for a given input (or set of inputs). The
process does not involve randomness and the direction of the process is known.
In contrast, stochastic processes are non-deterministic. The output is based on random trials and can
proceed via multiple, or even infinite, directions.
Monte Carlo methods fall under the umbrella of stochastic modeling. Monte Carlo methods are a
replication technique for propagating uncertainty through a model. The model is run multiple times, and
each run uses different input values and generates different output values: each run is independent of
each other. The sample of output values is used to estimate the properties of the actual probability
distribution of the outputs.
E.10.2 Implementation of Monte Carlo Methods
The implementation of Monte Carlo methods generally follows the following steps:
1. Define probability distributions for input parameters.
2. Generate a set of input values by randomly drawing a sample from each probability distribution.
3. Execute the deterministic model calculations.
4. Save the output results.
5. Repeat steps 2 through 4 through the desired number of iterations.
6. Aggregate the saved output results and calculate statistics.
Figure Apx E-l illustrates a flowchart of a Monte Carlo method implemented in a Microsoft Excel-
based model using a Monte Carlo add-in tool, such as the Palisade @Risk software.
Page 318 of 484
-------
7713
7714
7715
7716
7717
7718
7719
7720
7721
7722
7723
7724
7725
7726
7727
7728
7729
7730
7731
7732
7733
7734
7735
7736
7737
7738
7739
7740
7741
7742
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Results are Stored until
Desired Iterations are
Met
Iterations
Met
Results Summary
and Descriptive
Statistics
Define
Inputs
Monte Carlo
Add-In Tool
Randomly
Selected
Inputs
Deterministic Model
Outputs
FigureApx E-l. Flowchart of a Monte Carlo Method Implemented in a Microsoft Excel-Based
Model Using a Monte Carlo Add-In Tool
E.10.3 Building the Model
The steps for building a release or exposure model that incorporates Monte Carlo methods are as
follows:
1. Build the deterministic model.
2. Define probability distributions for input parameters.
3. Select model outputs for aggregation of simulation results.
4. Select simulation settings and run model.
5. Aggregate the simulation results and calculate output statistics.
Each of these steps is discussed in the subsections below.
E. 10.3.1 Build the Deterministic Model
First, the model is built as a deterministic model. EPA uses Microsoft Excel in order to use Palisade's
@Risk software that is used for probabilistic analyses in Excel. The model parameters and equations are
programmed into the spreadsheet. Model parameters are programmed in a summary table format for
transparency and to aid in the assignment of probability di stributions. Such summary tables are included
in the model-specific write-ups in Appendices E, 11 through E. 13 and Appendices F.5 through F.9.
E. 10.3.2 Define Probability Distributions for Input Parameters
Defining a probability distribution for an input parameter generally involves three steps:
1. Select the model input parameters for which probability distributions will be developed.
2. Determine a probability distribution from the available data.
3. Investigate if any parameters are statistically correlated. Define a statistical correlation among
parameters if a correlation is desired.
Step 1: Select Input Parameters for Probability Distribution Development
When selecting parameters for which probability distributions will be developed, the following factors
are considered:
• The availability of data to inform a distribution.
• The dependency of the input parameters on one another.
The sensitivity of the model results to each input parameter.
Page 319 of 484
-------
7743
7744
7745
7746
7747
7748
7749
7750
7751
7752
7753
7754
7755
7756
7757
7758
7759
7760
7761
7762
7763
7764
7765
7766
7767
7768
7769
7770
7771
7772
7773
7774
7775
7776
7777
7778
7779
7780
7781
7782
7783
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Availability of Data to Inform a Distribution: Data sources to investigate for available data to inform
probability distributions of model inputs include but are not limited to the following:
• EPA Generic Scenarios,
• OECD Emission Scenario Documents,
• Peer reviewed literature,
• Published chemical assessments, and
• Other gray literature.20
Model parameters may vary greatly in their available data. There may be a single study that provides
detailed measurements or observation data. There may be multiple studies that provide limited
measurements or observations. There may be only overall statistics available for a parameter. For a
given model development, the available data goes through a systematic review process to evaluate the
data quality, integrate the data, and decide how to use the data.
Dependency of Input Parameters on One Another: The model parameters are evaluated for any
dependency on each other. When each varied parameter is sampled according to its defined probability
distribution, they are sampled independently of each other. Therefore, the value of a sampled parameter
should be independent of the other sampled parameters. An exception is if a statistical correlation is
desired among two or more parameters. Correlating sampled parameters is discussed below in Step 3.
An example of dependency is the relationship between a facility's number of operating days, annual
production volume (PV), and daily PV. These three parameters are not all independent of each other.
The annual PV may be calculated from the daily PV and the operating days. Alternatively, the daily PV
may be calculated from the annual PV and the operating days. Additionally, operating days may be
calculated from the annual PV and daily PV. It is necessary to first understand the mathematical
relationship among these parameters before selecting parameters for which probability distributions will
be developed.
Sensitivity of the Model Results to Each Input Parameter: One consideration in selecting model
parameters for probability distribution development is the sensitivity of the model outputs to each
parameter. A sensitivity analysis can inform how sensitive each model output is to each model input
parameter. EPA may choose to prioritize probability distribution development for parameters to which
model outputs are more sensitive. Since the model outputs are more sensitive to these parameters, it
would be more important to capture variability and/or uncertainty for these parameters compared to
parameters to which model outputs are less sensitive.
A sensitivity analysis is conducted by varying each desired parameter and performing a Monte Carlo
simulation. The varied range for each parameter should be consistent with the expected range in values
for the parameter. The @Risk software (Palisade. 2022b) can perform sensitivity analyses. The statistic
of the outputs for which sensitivity is measured, such as mean, mode, or a percentile, can be selected. As
the simulation is run, the software tracks how each output changes with respect to each varied input.
20 Gray literature is defined as the broad category of data/information sources not found in standard, peer-reviewed literature
databases. Gray literature includes data/information sources such as white papers, conference proceedings, technical reports,
reference books, dissertations, information on various stakeholder websites, and various databases.
Page 320 of 484
-------
7784
7785
7786
7787
7788
7789
7790
7791
7792
7793
7794
7795
7796
7797
7798
7799
7800
7801
7802
7803
7804
7805
7806
7807
7808
7809
7810
7811
7812
7813
7814
7815
7816
7817
7818
7819
7820
7821
7822
7823
7824
7825
7826
7827
7828
7829
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Step 2: Determine a Probability Distribution
To determine a probability distribution, first, all the information known about the parameter is evaluated
(Oracle.: ). The following considerations can help guide summarizing important information about
the parameter (Analvtica. 2015):
• Discrete or continuous
o Consider whether the parameter is discrete or continuous. Does the parameter have a
finite or countable number of possible values? Is the parameter logical or Boolean such as
having possible values of "yes or no" or "true or false"? Can the parameter be
represented by all real numbers within a domain?
• Bounds
o Consider whether the parameter has bounds. A parameter may have a lower bound and/or
an upper bound. Alternatively, a parameter may be unbounded and can range to negative
and/or positive infinity.
• Modes
o Consider whether the parameter has one or more modes. Does the parameter have no
mode (such as represented by a uniform distribution)? If it has a mode, is it unimodal or
multimodal? If multimodal, is the parameter a combination of two or more populations?
In which case, the parameter may be best separated into its separate components and then
develop probability distributions for the individual components.
• Symmetric or skewed
o Consider whether the parameter is symmetric or skewed. If skewed, consider whether the
parameter is positively skewed (thicker upper tail) or negatively skewed (thicker lower
tail).
Second, review standard probability distributions and identify possible candidates that meet the
considerations identified in the first step (Oracle. 2017). The following are common probability
distributions:
• Uniform distribution
o A uniform distribution has finite upper and lower bounds and all values between the
bounds have equal probability.
• Triangular distribution
o A triangular distribution has finite upper and lower bounds and a modal value. The modal
value is the value that occurs most frequently. If the most frequent value is not known
another statistic, such as the mean or a percentile, could be used to define the triangular
distribution.
• Normal distribution
o The parameters of a normal distribution are its mean and standard deviation. A normal
distribution is unbounded, and values range from negative to positive infinity. If desired,
the range of values of a normal distribution may be truncated to finite bounds to prevent
unrealistic values from being sampled.
• Lognormal distribution
o If a variable is lognormally distributed, it means that the logarithm of that variable is
normally distributed. The parameters of a lognormal distribution are its mean and
standard deviation. A lognormal distribution is bounded from zero to positive infinity. A
lognormal distribution may be shifted and its upper bound truncated to fit the observed
data and prevent unrealistic values from being sampled.
Page 321 of 484
-------
7830
7831
7832
7833
7834
7835
7836
7837
7838
7839
7840
7841
7842
7843
7844
7845
7846
7847
7848
7849
7850
7851
7852
7853
7854
7855
7856
7857
7858
7859
7860
7861
7862
7863
7864
7865
7866
7867
7868
7869
7870
7871
7872
7873
7874
7875
7876
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Lastly, select the best suited probability distribution (Oracle. 2017). Review the available data for the
parameter to determine how to define the distribution's parameters. For example, if the only available
data are an overall range (with a minimum and a maximum), then a uniform distribution is the
appropriate distribution to use. If the only available data are an overall range and a mode, then a
triangular distribution is the appropriate distribution to use. If historical data for the parameter are
available, consider data fitting to determine the appropriate distribution and regress the distribution
parameter values.
Step 3: Check for and Define Statistical Correlations
When developing a Monte Carlo model and setting statistical distributions for parameters, EPA
evaluates possible correlations among parameters. When distributions are defined for the parameters,
each parameter is independently sampled on each iteration of the model. This may result in
combinations of parameter values that are not logical for the scenario. In the example of a model that
uses annual PV, daily PV, and operating days as parameters, there are set distributions for annual PV
and operating days, with the daily production volume calculated from the other two parameters. But
annual PV and operating days may be correlated. For example, if a site has a fixed manufacturing
capacity (as determined by the equipment size and production lines), then annual PV is a function of the
number of operating days. A facility is more likely to scale-up or scale-down their annual PV by varying
the operating days rather than varying their daily PV. Varying annual PV and operating days
independently in the model may arrive at value combinations that are not logical. For example, one
iteration may sample a high annual PV value with a low number of operating days that may result in a
high daily production rate that is not logical. In this example, a different probability distribution strategy
may be appropriate, such as defining probability distributions for daily PV and operating days since
those two parameters are likely more independent of each other than annual PV and operating days.
When developing distributions from observed data, there are statistical tests that can be performed to
indicate a statistical correlation. Two common ones are: 1) the Pearson product-moment correlation
coefficient, which measures the linear correlation between two data sets; and 2) Spearman's rank
correlation coefficient, which is a measure of rank correlation and how well a relationship between two
data sets can be described using a monotonic function. A monotonic relationship is one where the two
variables change together but not necessarily at a constant rate (Minitab. 2022). A linear correlation is
necessarily monotonic. But a monotonic correlation is not necessarily linear.
Both the Pearson and Spearman coefficients range from -1 to +1. A value close to +/-1 indicates a
strong correlation (either positive or negative). A positive correlation means as one variable increases,
the other also increases. A negative correlation means as one variable increases, the other decreases. A
value close to 0 means a weak or no correlation exists between the variables. The Pearson correlation
only measures linear relationships, and the Spearman correlation only measures monotonic
relationships. If two variables are correlated by a relationship that is neither linear nor monotonic, then
the Pearson and Spearman coefficients would not be informative of the nature of the correlation
(Minitab. 2022).
After testing for statistical correlations, statistical correlations can be defined for input parameters using
@Risk. @Risk only uses Spearman coefficients to define statistical correlations among input
parameters. Spearman coefficients to correlate two or more input parameters are defined through a
correlation matrix. The correlation matrix allows the Spearman coefficient to be defined for each pair of
correlated input parameters (Palisade. 2022a).
Page 322 of 484
-------
7877
7878
7879
7880
7881
7882
7883
7884
7885
7886
7887
7888
7889
7890
7891
7892
7893
7894
7895
7896
7897
7898
7899
7900
7901
7902
7903
7904
7905
7906
7907
7908
7909
7910
7911
7912
7913
7914
7915
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.10.3.3 Select Model Outputs for Aggregation of Simulation Results
The last step before running the model is to select the model outputs for which statistical results are
desired. Defining these outputs in @Risk will allow the software to save the output results from each
iteration and aggregate the simulation results over all iterations together.
E.10.3.4 Select Simulation Settings and Run Model
Simulation settings must be defined before running the model. Important simulation settings include the
number of iterations, the sampling method, and the random number generator.
• Number of Iterations. Generally speaking, a larger number of iterations is desired to ensure
adequate sampling and representation of lower probability events. The number of iterations to
achieve a desired margin of error for a given confidence interval for an output can be calculated
using the Central Limit Theorem (OJ>n I'1 JO I \ <\tlisade. 2015a). The equation shows that the
margin of error is inversely proportional to the square root of the number of iterations. Therefore,
the greater the number of iterations, the smaller the margin of error. Calculating the number of
iterations can be difficult as the sample standard deviation is not known beforehand. EPA
typically uses 100,000 iterations to ensure convergence and have minimal cost to the simulation
time.
• Sampling Method. The sampling method is the method used to draw random samples from the
input parameter probability distributions. @Risk uses two methods: Latin Hypercube (the
default) and Monte Carlo. Monte Carlo sampling is a purely random sampling method. This can
lead to clustering and under-representing low probability events. Latin Hypercube sampling is a
stratified sampling method. This ensures the sampled input parameter distribution matches the
assigned probability distribution closely. EPA typically uses Latin Hypercube sampling because
it is efficient and can achieve convergence with fewer iterations than Monte Carlo sampling
(Palisade. 2018).
• Random Number Generator. The random number generator is used to generate pseudorandom
numbers that are used in an algorithm to draw random samples from the probability distributions.
The @Risk default is Mersenne Twister, which is a robust and efficient random number
generator (Palisade. 2015b).
E.10.3.5 Aggregate the Simulation Results and Produce Output Statistics
During the simulation, @Risk will save the defined model outputs for aggregation on each iteration.
After the simulation is completed, EPA can generate desired statistical results and distributions of the
defined outputs. EPA typically uses the 50th percentile and 95th percentile of the output as the central
tendency and high-end estimates, respectively.
E.ll Textile Dye Modeling Approach and Parameters for Estimating
Environmental Releases
This appendix presents the modeling approach and equations used to estimate environmental releases of
1,4-dioxane during the commercial use of textile dyes. This approach utilizes the OECD ESD on Textile
Dyes (OECD. 2017) combined with Monte Carlo simulation (a type of stochastic simulation). This ESD
includes a diagram of release and exposure points during textile dying, as shown in Figure Apx E-2.
Page 323 of 484
-------
7916
7917
7918
7919
7920
7921
7922
7923
7924
7925
7926
7927
7928
7929
7930
7931
7932
7933
7934
7935
7936
7937
7938
7939
7940
7941
7942
7943
7944
7945
7946
7947
7948
7949
7950
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Dve Formulation
© Dust Emissions
during Unloading
/Tn Disposal of Spent
Dyebath
Equipment Cleaning
©<§> Container
Residue
Cleaning and/or
Disposal
<£> Worker Exposure
During Dyeing
Operation
FigureApx E-2. Environmental Release and Occupational Exposure Points during Textile Dying
Based on Figure Apx E-2, EPA identified the following release points (note that diagram points 3 and 4
were combined for ease of analysis):
• Release point 1 (RP1): Dust emissions of during unloading of solid powders to air, landfill,
POTW, or incineration;
• Release point 2 (RP2): Container residual losses to POTW, landfill, or incineration; and
• Release point 3 (RP3): Disposal of spent dye bath and equipment cleaning losses to POTW.
Environmental releases of textile dyes are a function of the chemical's physical properties, container
size, mass fractions, and other model parameters. Although physical properties are fixed, some model
parameters are expected to vary from one facility to another. An individual model input parameter could
either have a discrete value or a distribution of values. EPA assigned statistical distributions based on
available literature data or engineering judgment to address the variability in mass fraction of dye
formulation in the bath (Fdye dyebath). container size (Vcontainer), textile production rate (Vfabric), operating
days (OD), and container residue fractions (Fcontainer residue).
A Monte Carlo simulation was then conducted to capture variability in the model input parameters
described above. The simulation was conducted using the Latin hypercube sampling method in @Risk
(Palisade, Ithaca, NY). The Latin hypercube sampling method is a statistical method for generating a
sample of possible values from a multi-dimensional distribution. Latin hypercube sampling is a stratified
method, meaning it guarantees that its generated samples are representative of the probability density
function (variability) defined in the model. EPA performed 100,000 iterations of the model to capture
the range of possible input values, including values with low probability of occurrence.
From the distribution resulting from the Monte Carlo simulation, EPA selected the 95th and 50th
percentile values to represent a high-end release and central tendency release level, respectively. The
statistics were calculated directly in @Risk. The following subsections detail the model design equations
and parameters used for environmental release estimates.
E.ll.l Model Equations
Daily use rate of dye formulation is calculated using the following equation:
Equation Apx E-l
Qdye_formulation_day ~ ^fabric * ^fabric * ^dye_fabric
Page 324 of 484
-------
7951
7952
7953
7954
7955
7956
7957
7958
7959
7960
7961
7962
7963
7964
7965
7966
7967
7968
7969
7970
7971
7972
7973
7974
7975
7976
7977
7978
7979
7980
7981
7982
7983
7984
7985
7986
7987
7988
7989
7990
7991
7992
7993
7994
7995
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Where:
\ye_formulation_day
Qd,
^fabric
Pfabric
Pdye_fabric
Daily use rate of dye formulation [kg/site-day]
Textile production rate [kg/site-day]
Mass fraction of textiles treated with dye [kg/kg]
Mass fraction of dye used per mass of textile dyed [kg/kg]
Daily use rate of 1,4-dioxane formulation is calculated using the following equation:
EquationApx E-2
Qd
ioxane_site_day
Qdye_formulation_day * Pdioxane_dye * Pdye
Where:
Q dioxane_site_day
Q dye_formulation_day
Pdioxane_dye
Pdye
Daily use rate of 1,4-dioxane [kg/site-day]
Daily use rate of dye formulation [kg/site-day]
Mass fraction of 1,4-dioxane in dye formulation [kg/kg]
Fraction of dye containing 1,4-dioxane [kg/kg]
Containers emptied per facility is calculated using the following equation:
Equation Apx E-3
JV,
Qd
container _unload_site_yr
ioxane_site_day
* OD
Pdioxane_dye * ^•
container * 3.79 ^ * RHOform
Where:
Ncontainer _unload_site_yr
Qdioxane_site_day ~
OD
Pdioxane_dye ~
V,
container
RHO
form
Containers emptied per facility [containers/site-year]
Daily use rate of 1,4-dioxane [kg/site-day]
Operating days [days/year]
Mass fraction of 1,4-dioxane in dye formulation [kg/kg]
Container size [gal/container]
Dye density [kg/L]
Container residual fraction is calculated using the below equations. To make the simulation more
realistic, EPA assessed container size based on the dye formulation use rate. This avoids situations
where a small container size is associated with a large use rate, such that an unrealistic number of
containers are used each year, and vice-versa.
Equation Apx E-4
If Qdye_formulation_day ^700 kg/site-day.
Pcontainer_residue ~ Pcontainer_residue_tote
If Qdye_formulation_day 'S 200-700 kg/site-day.
Pcontainerjresidue ~ Pcontainer_residue_drum
if Qd
\ye_formulation_day
< 200 kg/site-day
container _residue
= P,
container _residue_pail
Page 325 of 484
-------
7996
7997
7998
7999
8000
8001
8002
8003
8004
8005
8006
8007
8008
8009
8010
8011
8012
8013
8014
8015
8016
8017
8018
8019
8020
8021
8022
8023
8024
8025
8026
8027
8028
8029
8030
8031
8032
8033
8034
8035
8036
8037
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Where:
Qdye_formulation_day ~
Fcontainer residue ~
Fcontainer _residue_tote ~
Fcontainer _residue_drum~
Fcontainer_residue_pail ~
Daily use rate of dye formulation [kg/site-day]
Container residual fraction [kg/kg]
Container residual fraction for totes [kg/kg]
Container residual fraction for drums [kg/kg]
Container residual fraction for pails [kg/kg]
Mass fraction of 1,4-dioxane in dye bath is calculated using the following equation:
EquationApx E-5
Where:
Fdioxane_dyebath
Fdioxane_dye
Fdye_dyebath
Fdioxane_dyebath ~ Fdioxane ciye * Fdye_dyebath
Mass fraction of 1,4-dioxane in dye bath [kg/kg]
Mass fraction of 1,4-dioxane in dye formulation [kg/kg]
Mass fraction of the dye formulation in the dyebath [kg/kg]
Release point 2 (container residual) release per day is calculated using the following equation:
Equation Apx E-6
Release_perDayRP2 = Qd
Where:
Release_perDayRP2 =
Qdioxane_site_day ~
Fcontainer _residue ~
ioxane_site_day * ^container_reSidiie
Container residual release from release point 2 [kg/site-day]
Daily use rate of 1,4-dioxane [kg/site-day]
Container residual fraction [kg/kg]
Release point 3 (spent dye bath and equipment cleaning) release per release day is calculated using the
following equation:
Equation Apx E-7
Release _perD ayRP3 = Qd
ioxane_site_day
(1 ^fixation)
Where:
Release_perDayRP 3 =
Qd
ioxane_site_day
Ffixation
Dye bath and equipment cleaning release from point 3 [kg/site-day]
Daily use rate of 1,4-dioxane [kg/site-day]
Fraction of dye affixed to textile during dye process [kg/kg]
E.11.2 Model Input Parameters
Table Apx E-12 summarizes the model parameters and their values for the Monte Carlo simulation.
Additional explanations of EPA's selection of the distributions for each parameter are provided after this
table. High-end and central tendency releases are estimated by selecting the 50th and 95th percentile
values from the output distribution.
Page 326 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
8038 Table Apx E-12. Summary of Parameter Values and Distributions Used in the Textile Release Model
Input Parameter
Symbol
Unit
Constant Model
Parameter Values
Variable Model Parameter Values
Rational/Basis
Value
Lower
Bound
Upper
Bound
Mode
Distribution
Type
Number of Sites
Ns
sites
783
-
-
-
-
See Section E.11.3
Fraction of dye
containing 1,4-dioxane
Fdye
kg/kg
1
-
-
-
-
See Section E.11.4
1,4-dioxane vapor
pressure
VP
Torr
40
-
-
-
-
Physical property
1,4-dioxane molecular
weight
MW
g/mol
88.1
-
-
-
-
Physical property
Operating days
OD
days/year
157
30
296
157
Triangular
See Section E.11.5
Mass fraction of 1,4-
dioxane in dye
formulation
Fdioxane dye
kg/kg
0.0000047
See Section E.11.6
Textile production rate
V fabric
kg/day
9,100
3,250
50,000
9,100
Triangular
See Section E.11.7
Mass fraction of textiles
treated with dye
F fabric
kg/kg
0.3
-
-
-
-
See Section E.11.8
Mass fraction of dye used
per mass of textile dyed
Fdye fabric
kg/kg
0.1
-
-
-
-
See Section E.11.9
Mass fraction of the dye
formulation in the
dyebath
Fdye dyebath
kg/kg
0.06
0.002
0.06
0.02
Triangular
See Section E.11.10
Container size for dye
formulation
V container
gal
35
7
264
35
Triangular
See Section E.ll.ll
Dye density
RHOform
kg/L
1
ESD assumes a
density equal to that
of water
Container residual
fraction for totes
Fcontainer residue totes
kg/kg
0.002
0.0007
0.002
0.0007
Triangular
See Section E.11.12
Container residual
fraction for drums
Fcontainer residue drums
kg/kg
0.03
0.0003
0.03
0.025
Triangular
See Section E.11.13
Page 327 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Input Parameter
Symbol
Unit
Constant Model
Parameter Values
Variable Model Parameter Values
Rational/Basis
Value
Lower
Bound
Upper
Bound
Mode
Distribution
Type
Container residual
fraction for pails
Fcontainer residue_pails
kg/kg
0.006
0.0003
0.006
0.003
Triangular
See Section E.11.14
Fraction of dye product
affixed to textile during
dyeing process substrate
F fixation
kg/kg
Multiple Triangular Distributions
See Section E.11.15
8039
Page 328 of 484
-------
8040
8041
8042
8043
8044
8045
8046
8047
8048
8049
8050
8051
8052
8053
8054
8055
8056
8057
8058
8059
8060
8061
8062
8063
8064
8065
8066
8067
8068
8069
8070
8071
8072
8073
8074
8075
8076
8077
8078
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.11.3 Number of Sites
EPA did not find data on the number of textile dye sites that specifically use dyes containing 1,4-
dioxane. As a bounding estimate, EPA used U.S. Census and BLS data for the NAICS code 313310,
Textiles and Fabric Finishing Mills, to estimate a total of 783 sites within the industry (U.S. BLS. 2016).
E.11.4 Mass Fraction of Dye Containing 1,4-Dioxane
The ESD on the Use of Textile Dyes provided a single value for the mass fraction of dyes containing the
chemical of interest, which is 1,4-dioxane. The ESD assumes that 100 percent of dyes contain the
chemical of interest (OECD. 2017). Therefore, EPA could not develop a distribution of values for this
parameter and used the single value of 1 kg dye with 1,4-dioxane/kg dye used from the ESD.
E.11.5 Operating Days
EPA modeled the number of operating days per year using a triangular distribution with a lower bound
of 31 days per year, and upper bound of 295 days per year, and a mode of 157 days per year. This is
based on the ESD on the Use of Textile Dyes (OECD. 2017). The ESD cites the basis of these data as
past new chemical submissions that were submitted to EPA from 2006 through 2014. EPA used the 5th
percentile, average, and 95th percentile in the ESD as the lower bound, mode, and upper bound of this
distribution, respectively.
E.11.6 Mass Fraction of 1,4-Dioxane in Dye Formulation
The December 2020 Final Risk Evaluation for 1,4-Dioxane (U.S. EPA. 2020c) indicates that 1,4-
dioxane is a byproduct in dye formulations and provided a single value of 0.0000047 kg 1,4-dioxane/kg
dye. Therefore, EPA could not develop a distribution of values for this parameter and used the single
value of 0.0000047 kg 1,4-dioxane/kg dye from the risk evaluation.
E.11.7 Textile Production Rate
EPA modeled textile production rate using a triangular distribution with a lower bound of 3,520 kg/site-
day, and upper bound of 50,000 kg/site-day, and a mode of 9,100 kg/site-day. This is based on the ESD
on the Use of Textile Dyes (OECD. 2017). The ESD cites data provided in fabric finishing new
chemical submissions during 1993 and 1994. Note that the ESD uses a "typical" value as default and
does not say what the typical is based on (e.g., average, median). EPA used the range of textile
production rates and the default typical value provided in the ESD as the lower and upper bounds and
mode of the triangular distribution for this parameter, respectively.
E.11.8 Mass Fraction of Textiles Treated with Dye
The ESD on the Use of Textile Dyes provided a single value for the mass fraction of all textiles treated
with dyes. The ESD states that the median share of textiles processed per day using the primary dyestuff
is 30 percent (OECD. 2017). Therefore, EPA could not develop a distribution of values for this
parameter and used the single value of 30 percent from the ESD.
E.11.9 Mass Fraction of Dye Used per Mass of Textile Dyed
The ESD on the Use of Textile Dyes provided a single value for the mass fraction of dye used per mass
of textile dyed. The ESD states that as a "realistic worst case," liquid dye formulations are used in an
amount of 10 percent ( ). Therefore, EPA could not develop a distribution of values for this
parameter and used the single value of 0.10 kg dye/kg textiles from the ESD.
Page 329 of 484
-------
8079
8080
8081
8082
8083
8084
8085
8086
8087
8088
8089
8090
8091
8092
8093
8094
8095
8096
8097
8098
8099
8100
8101
8102
8103
8104
8105
8106
8107
8108
8109
8110
8111
8112
8113
8114
8115
8116
8117
8118
8119
8120
8121
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.11.10 Mass Fraction of the Dye Formulation in the Dyebath
EPA modeled mass fraction of dye formulation in the dyebath using a triangular distribution with a
lower bound of 0.002 kg dye/kg bath, an upper bound of 0.06 kg dye/kg bath, and a mode of 0.02 kg
dye/kg bath. This is based on the ESD on the Use of Textile Dyes. The ESD states that typical dye
concentrations may range from 1.5 to 2.5 percent, lighter shades may be as low as 0.2 to 0.3 percent, and
heavier shades may be between 4 to 6 percent (OECD. ^ ). EPA used the overall range of dye
concentrations (0.2 to 6 percent) and the mid-range of the typical concentration (2 percent) provided in
the ESD as the lower and upper bounds and mode of the triangular distribution for this parameter,
respectively.
E.ll.ll Container Size for Dye Formulation
EPA modeled container size using a triangular distribution with a lower bound of 7 gallons, an upper
bound of 264 gallons, and a mode of 35 gallons. This is based on the ESD on the Use of Textile Dyes.
The ESD states that dyes can be transported in containers ranging from 25 kg through 1,000 kg, but
most are shipped in 35-galIon drums (OE ). EPA converted this range from kilograms to
gallons using an assumed dye density of 1 kg/L and a conversion factor of 3.785 L/gal.
E.11.12 Container Residual Fraction for Totes
EPA modeled container residual fraction for totes using a triangular distribution with a lower bound of
0.0007 kg residual/kg dye, and upper bound of 0.002 kg residual/kg dye, and a mode of 0.0007 kg
residual/kg dye. The lower and upper bounds of this distribution are based on the central tendency and
high-end values listed in the EPA/OPPT Bulk Transport Residual Model from the ChemSTEER User
Guide ( ). EPA used the central tendency value as the mode of the triangular
distribution.
E.11.13 Container Residual Fraction for Drums
EPA modeled container residual fraction for drums using a triangular distribution with a lower bound of
0.0003 kg residual/kg dye, an upper bound of 0.03 kg residual/kg dye, and a mode of 0.025 kg
residual/kg dye. The lower bound is based on the minimum value for pouring and the upper bound is
based on the default high-end value in the EPA/OPPT Drum Residual Model from the ChemSTEER
User Guide ( ). EPA used the central tendency value for pumping as the mode of the
triangular distribution.
E.11.14 Container Residual Fraction for Pails
EPA modeled container residual fraction for small containers using a triangular distribution with a lower
bound of 0.0003 kg residual/kg dye, an upper bound of 0.006 kg residual/kg dye, and a mode of 0.003
kg residual/kg dye. The lower bound is based on the minimum value for pouring and the upper bound is
based on the default high-end value listed in the EPA/OPPT Small Container Residual Model from the
ChemSTEER User Guide ( ). EPA used the central tendency value for pouring as the
mode of the triangular distribution.
E.11.15 Fraction of Dye Product Affixed to Textile during Dyeing Process Substrate
EPA modeled the fraction of dye product affixed to textiles during dyeing process substrate using
multiple triangular distributions. The ESD on the Use of Textile Dyes provides a table containing ranges
and averages for dye fixation percentages based on the nine different classes of dyes (OECD. 2017).
EPA used the low-end of the range of dye fixation as the lower bound, the high-end of the range of dye
fixation as the upper bound, and the average dye fixation as the mode for each of the nine triangular
distributions. In the Monte Carlo simulation, each of the nine triangular distributions from the ESD has
Page 330 of 484
-------
8122
8123
8124
8125
8126
8127
8128
8129
8130
8131
8132
8133
8134
8135
8136
8137
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
an equal probability of being selected and used for the parameter's output. The distribution selection
probabilities and values are shown in Table Apx E-13.
Table Apx E-13. Triangular Distributions Ffixation
Dye Type
Dye Fixation (%) Triangular Distribution
Lower Bound
Upper Bound
Mode
Acid
85
98
93
Azoic
76
95
84
Basic
95
100
99
Direct
64
96
88
Disperse
80
100
96
Metal-
Containing
82
98
94
Reactive
50
97
85
Sulfur
60
95
70
Vat
70
95
85
Source: ("OECD. 2017)
E.12 Laundry Detergent Modeling Approach and Parameters for
Estimating Environmental Releases
This appendix presents the modeling approach and equations used to estimate environmental releases of
1,4-dioxane during the industrial and institutional use of laundry detergents. This approach utilizes the
OECD ESD on the Chemicals Used in Water Based Washing Operations at Industrial and Institutional
Laundries (OECD. i ) combined with Monte Carlo simulation (a type of stochastic simulation).
This ESD categorizes laundry facilities into either industrial or institutional facilities and includes
different process parameters for each. Therefore, EPA modeled the two types of laundry facilities
separately. In addition, laundry detergents can be in liquid or powder physical forms. Because the
difference in physical form results in different parameter distributions, EPA modeled liquid and powder
detergents separately. This ESD includes a diagram of release and exposure points during textile dying,
as shown in Figure Apx E-3.
Page 331 of 484
-------
8138
8139
8140
8141
8142
8143
8144
8145
8146
8147
8148
8149
8150
8151
8152
8153
8154
8155
8156
8157
8158
8159
8160
8161
8162
8163
8164
8165
8166
8167
8168
8169
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
©©©Container Residue and Cleaning ©©©Container Residue and Cleaning
©@ Fugitive Air Release and Dust ©@ Fugitive Air Release and Dust
Emissions During Transfers Emissions During Transfers
©@© Releases During
Operations
FigureApx E-3. Environmental Release and Occupational Exposure Points during
Industrial/Institutional Laundering Operations
Based on Figure Apx E-3, EPA identified the following release points:
• Release point 1 (RP1): Container residual losses to POTW, landfill, or incineration;
• Release point 2 (RP2): Fugitive air releases during container cleaning;
• Release point 3 (RP3): Fugitive air releases during container unloading;
• Release point 4 (RP4): Dust releases during container unloading;
o 4a: Uncaptured dust releases;
o 4b: Captured, uncontrolled dust releases;
o 4c: Captured and controlled dust releases;
• Release point 5 (RP5): Fugitive air releases during washing; and
• Release point 6 (RP6): Release from washing and drying operations to fugitive air, stack air, or
POTW.
Environmental releases of laundry detergent are a function of the chemical's physical properties,
container size, mass fractions, and other model parameters. Although physical properties are fixed, some
model parameters are expected to vary from one facility to another. An individual model input
parameter could either have a discrete value or a distribution of values. EPA assigned statistical
distributions based on available literature data or engineering judgment to address the variability in mass
fraction of 1,4-dioxane in the detergent (Fdioxane laundry), container size (Vcontainer), daily use rate of
detergent (Qfaciiity day), air speed (RATEair speed), duration of release (OHCOnt unload), operating days (OD),
container residue fractions (Fcontainer residue), and dust capture/control efficiency (Fdust).
A Monte Carlo simulation was conducted to capture variability in the model input parameters described
above. The simulation was conducted using the Latin hypercube sampling method in @Risk (Palisade,
Ithaca, New York). The Latin hypercube sampling method is a statistical method for generating a
sample of possible values from a multi-dimensional distribution. Latin hypercube sampling is a stratified
method, meaning it guarantees that its generated samples are representative of the probability density
function (variability) defined in the model. EPA performed 100,000 iterations of the model to capture
the range of possible input values, including values with low probability of occurrence.
Page 332 of 484
-------
8170
8171
8172
8173
8174
8175
8176
8177
8178
8179
8180
8181
8182
8183
8184
8185
8186
8187
8188
8189
8190
8191
8192
8193
8194
8195
8196
8197
8198
8199
8200
8201
8202
8203
8204
8205
8206
8207
8208
8209
8210
8211
8212
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
From the distribution resulting from the Monte Carlo simulation, EPA selected the 95th and 50th
percentile values to represent a high-end release and central tendency release level respectively. The
statistics were calculated directly in @Risk. The following subsections detail the model design equations
and parameters used for environmental release estimates.
E.12.1 Model Equations
Daily use rate selection based on physical form of detergent is based on the following two equations, the
first being for liquid detergent and the second being for powder detergent:
EquationApx E-8
Qfacility_day ~ Qfacility_day_liquid
or
Qfacility_day ~ Qfacility_day_powder
Where:
Qf acility_day = Daily use rate based on physical form of detergent [kg/site-day]
Qf acility dayUquid = Daily use for liquid form detergent [kg/site-day]
Qf acility_day_powder = Daily use for powder form detergent [kg/site-day]
Daily use rate of laundry detergents containing 1,4-dioxane is calculated using the equation below:
Equation Apx E-9
Q facility _day_adjusted ~ Qf acility_day * Fformulations_dioxane
Where:
Qfacility_day_adjusted = Daily use rate of detergent containing 1,4-dioxane selected based
on the physical form of the detergent [kg/site-day]
Qf acility _day = Daily use rate based on physical form of detergent [kg/site-day]
Pformuiations_dioxane = Fraction of laundry detergents containing 1,4-dioxane [kg/kg]
Daily use rate of 1,4-dioxane is calculated using the equation below:
Equation Apx E-10
Qdioxane_day ~ Qfacility_day_adjusted * Pdioxanejaundry
Where:
Qdioxane_day = Daily usage rate of 1,4-dioxane [kg/site-day]
Qf acility _day_ad justed = Daily use rate of detergent with 1,4-dioxane [kg/site-day]
Fdioxanejaundry = Mass fraction of 1,4-dioxane in laundry detergent [kg/kg]
Number of containers used per year is calculated using the equation below:
Page 333 of 484
-------
8213
8214
8215
8216
8217
8218
8219
8220
8221
8222
8223
8224
8225
8226
8227
8228
8229
8230
8231
8232
8233
8234
8235
8236
8237
8238
8239
8240
8241
8242
8243
8244
8245
8246
8247
8248
8249
8250
8251
8252
8253
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
EquationApx E-ll
N,
Q facility j
day_ad justed
* OD
cont_site_yr
V,
container
* gal * ^^^detergent
Where:
Ncont_site_yr
Q facility _day _ad justed
OD
^container
RHO
detergent
Number of containers used per site per year [containers/site-year]
Daily use rate of detergent with 1,4-dioxane [kg/site-day]
Operating days [days/year]
Container volume [gal/container]
Detergent density [kg/L]
Vapor pressure correction factor for release points 2 and 3 is calculated using the equation below:
Equation Apx E-12
Fdioxane
X,
e laundry/
/,
MW
cleanjinload
Fdioxane Jaundry , 1 FdioxaneJaundry
MW + 18
Where:
X,
clean unload
Fd ioxane Jaundry
MW
Vapor pressure correction factor for release points 2 and 3
[mol 1,4-dioxane/mol water]
Mass fraction of 1,4-dioxane in detergent [kg/kg]
1,4-dioxane molecular weight [g/mol]
Fraction of 1,4-dioxane in wash water is calculated using the equation below:
Equation Apx E-13
Fni
dioxanejwash
~ Fdilution * Fdioxanejaunciry
Where:
Fd ioxane _w ash
Fdilution
Fd ioxane Jaundry
Fraction of 1,4-dioxane in wash water [kg 1,4-dioxane/kg water]
Dilution factor for detergent in the wash [unitless]
Mass fraction of 1,4-dioxane in detergent [kg/kg]
Vapor pressure correction factor for release point 5 is calculated using the equation below:
Equation Apx E-14
Fdioxanejwash /
'MW
X,
washing
Fdioxanejwash _j_ ^ ^df'o:mr!e_was/i
MW
18
Where:
Xwashing
Fd ioxane_wash
MW
Vapor pressure correction factor for release point 5
[mol 1,4-dioxane/mol water]
Fraction of 1,4-dioxane in wash water [kg 1,4-dioxane/kg water]
1,4-dioxane molecular weight [g/mol]
Page 334 of 484
-------
8254
8255
8256
8257
8258
8259
8260
8261
8262
8263
8264
8265
8266
8267
8268
8269
8270
8271
8272
8273
8274
8275
8276
8277
8278
8279
8280
8281
8282
8283
8284
8285
8286
8287
8288
8289
8290
8291
8292
8293
8294
8295
8296
8297
8298
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Container residual fraction is calculated using the following equations. To make the simulation more
realistic, EPA assessed container size based on the detergent use rate. This avoids situations where a
small container size is associated with a large use rate, such that an unrealistic number of containers are
used each year, and vice-versa:
EquationApx E-15
If Qfacility_day > 600 ~ '
site—day
Pcontainer residue ~ Pcontainer residue tote
KQ facility .day = 200 - 600^^'
site—day
Pcontainer residue ~ Pcontainer residue drum
If <3facility_day < 200 sitJday-
Pcontainer_residue ~ Pcontainer_residue_pail
If physical form of detergent is powder:
Pcontainer_residue ~ Pcontainer_residue_powder
Where:
Qfacility_day = Daily use rate based on physical form of detergent [kg/site-day]
pcontainer residue = Container residual fraction [kg/kg]
p'container residuejote = Container residual fraction for totes [kg/kg]
pcontainer residue_drum= Container residual fraction for drums [kg/kg]
pcontainer residue_Vaii = Container residual fraction for pails [kg/kg]
p'container residue_Powder = Container residual fraction for solid detergents [kg/kg]
Release Point 1 site release per day is calculated using the equation below:
Equation Apx E-16
Release_perDayRp1 — Qdioxane_day * Pcontainerjresidue
Where:
Release_perDayRP1 = Daily 1,4-dioxane release at release point 1 [kg/site-day]
Qdioxane_day = Daily usage rate of 1,4-dioxane [kg/site-day]
pcontainer residue = Container residual fraction [kg/kg]
Release Point 2 fugitive emissions from container cleaning for pails and drums per day is calculated using
the Penetration Model equation below (air speed <100 ft/min):
Equation Apx E-17
Release jper DayRP2 =
s ka x ) * (MW ) * Xcleanun[oad * VP * * (0-257lDcontainer_opening)
3600— * 0.001— * y
fir a t0-05 * ff) .[p
a i <\j ^container_o~pening *1
Where:
Page 335 of 484
-------
8299
8300
8301
8302
8303
8304
8305
8306
8307
8308
8309
8310
8311
8312
8313
8314
8315
8316
8317
8318
8319
8320
8321
8322
8323
8324
8325
8326
8327
8328
8329
8330
8331
8332
8333
8334
8335
8336
8337
8338
8339
8340
8341
8342
8343
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Release_perDayRP2 =
MW
X clean_unload ~
VP
T
R(Xt6airspeed ~
Dcontainer _opening ~
Release point 2 fugitive emissions from pail/drum cleaning
per day [kg/site-day]
1,4-dioxane molecular weight [g/mol]
Vapor pressure correction factor release point 2
[mol 1,4-dioxane/mol water]
1,4-dioxane vapor pressure [torr]
Ambient temperature [K]
Air speed [cm/s]
Diameter of container opening [cm]
Atmospheric pressure [atm]
Release Point 2 fugitive emissions from container cleaning per day for totes is calculated using the Mass
Transfer Coefficient Model equation below (air speed >100 ft/min):
EquationApx E-18
Release jper DayRP2 =
3600—*0.001
hr g
kg (1-93 x 10 7) * (MVK0'78) * Xcleanun[oad * VP * RO-tSair_speed * containerjopenlng) J29 ^ MW
Where:
Release_perDayRP2 =
X clean_unload ~
MW
VP
T
R(Xt6airspeed ~
Dcontainer _opening ~
j10.4 n 0.11
container _opening
(VT - 5.87)2/3
Release point 2 fugitive emissions from tote cleaning per day
[kg/site-day]
Vapor pressure correction factor release point 5
[mol 1,4-dioxane/mol water]
1,4-dioxane molecular weight [g/mol]
Vapor Pressure [torr]
Ambient Temperature [K]
Air speed [cm/s]
Diameter of container opening [cm]
Atmospheric pressure [atm]
Release Point 3 fugitive emissions from unloading of pails and drums during the day is calculated using
the Penetration Model equation below (air speed <100 ft/min):
Equation Apx E-19
Release jper DayRP3 =
OHcont_unload * 3600 — * 0.001— *
kg (8.24 x lO"8) * (WJS) * Xcleanunload * VP * jRateairspeea * (0.25*#,
)1- + —
container _opemng J ..I 29 MW
'VA
container_opening
VP
Where:
Release_perDayRP3
X clean_unload
MW
VP
T
Point 3 fugitive emissions from unloading during the day
[kg/site-day]
Vapor pressure correction factor release point 5
[mol 1,4-dioxane/mol water]
1,4-dioxane molecular weight [g/mol]
Vapor pressure [torr]
Ambient temperature [K]
Page 336 of 484
-------
8344
8345
8346
8347
8348
8349
8350
8351
8352
8353
8354
8355
8356
8357
8358
8359
8360
8361
8362
8363
8364
8365
8366
8367
8368
8369
8370
8371
8372
8373
8374
8375
8376
8377
8378
8379
8380
8381
8382
8383
8384
8385
8386
8387
8388
8389
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Rateair_Speed = Air speed from EPA model [cm/s]
Dcontainer opening = Diameter of the opening for containers [cm]
P = Atmospheric pressure [atm]
OHcont unload = Duration of container unloading [hrs/day]
Release Point 3 fugitive emissions from unloading totes during the day is calculated using the Mass
Transfer Coefficient Model equation below (air speed >100 ft/min):
EquationApx E-20
Release jper DayRP3 =
s kq (193 x 10 7) * (MW0,78) * Xcieanunload * VP * Rate°Jr8speed * (0.25?lD2container opening)\js +
OHcont_unload3600—* O.OOl^* 3
hr ' g 7-o.40o.il (^/T-5.87)2^
1 ^container_opemng\v 1 j
Where:
Release_perDayRP3 = Point 3 fugitive emissions from unloading during the day
[kg/site-day]
Xdeanunioad = Vapor pressure correction factor release point 5
[mol 1,4-dioxane/mol water]
MW = 1,4-dioxane molecular weight [g/mol]
VP = Vapor pressure [torr]
T = Ambient temperature [K]
Rateair Speed = Air speed from EPA model [cm/s]
Dcontainer opening = Diameter of the opening for containers [cm]
P = Atmospheric pressure [atm]
OHcont unload = Duration of container unloading [hours/day]
Release Point 4a dust not captured to fugitive air, water, incineration, or landfill is calculated using the
following equation:
Equation Apx E-21
Release_perDayRPAa — Qdioxane_day * Fdustgenerati0n * Fdustcapture)
Where:
Release_perDayRP4a= Dust not captured to fugitive air, water, incineration, or landfill
[kg/site-day]
Fdustgeneration = Fraction of chemical lost during transfer of solid powders [kg/kg]
Fdustcapture = Capture efficiency for dust capture methods [kg/kg]
Release Point 4b dust captured but not controlled to stack air is calculated using the following equation:
Equation Apx E-22
Release_perDayRP4b = Qdioxane_day * Fdustgeneration * Fdustcapture * (1 — Fdustcontrol)
Where:
Release_perDayRP4b= Dust captured but not controlled to stack air [kg/site-day]
Fdustgeneration = Fraction of chemical lost during transfer of solid powders [kg/kg]
Fdustcavture = Capture efficiency for dust capture methods [kg/kg]
Page 337 of 484
-------
8390
8391
8392
8393
8394
8395
8396
8397
8398
8399
8400
8401
8402
8403
8404
8405
8406
8407
8408
8409
8410
8411
8412
8413
8414
8415
8416
8417
8418
8419
8420
8421
8422
8423
8424
8425
8426
8427
8428
8429
8430
8431
8432
8433
8434
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Fdustcontroi = Control efficiency for dust control methods [kg/kg]
Release Point 4c dust captured and controlled to incineration of landfill is calculated using the following
equation:
EquationApx E-23
Release_perDayRP4b = Qdioxane_day * Fdustgeneration * Fdustcavture * Fdustcontroi
Where:
Release_perDayRP4b= Dust captured but not controlled to stack air [kg/site-day]
Fdustgeneration = Fraction of chemical lost during transfer of solid powders [kg/kg]
Fdustcapture = Capture efficiency for dust capture methods [kg/kg]
Fdustcontroi = Control efficiency for dust control methods [kg/kg]
Release Point 5 fugitive emissions during washing per day is calculated when air speed <100 ft/min using
the Penetration Model in the equation shown below:
Equation Apx E-24
Release jper DayRPS =
(8.24 x 10~8) * (Miy0,835) * Xcleanunload *VP* lRateairspeed * (025uDtontainer_opening)% + ^
OH * 3600— * 0.001— * N
hr q T005 * fn ¦ \fP
u V container_opemng v 1
Where:
Release_perDayRP5 = Point 5 fugitive emissions from washing [kg/site-day]
Xdean unload = Vapor pressure correction factor release point 5
[mol 1,4-dioxane/mol water]
MW = 1,4-dioxane molecular weight [g/mol]
VP = Vapor pressure [torr]
T = Ambient temperature [K]
Rateair speed = Air speed [cm/s]
Dcontainer opening = Diameter of the opening for containers [cm]
P = Atmospheric pressure [atm]
OH = Operating hours [hours/day]
Release Point 5 fugitive emissions during washing per day is calculated when air speed >100 ft/min using
the Mass Transfer Coefficient Model shown below:
Equation Apx E-25
Release jper DayRP5 =
s kq (193 x 10 7) * (MW0,78) * Xcieanunload * VP * Rate0Jr8speed * (0.25?lD2container opening)\js +
OHcont_unload3600—* 0.001^* 3
hr ' g ro.40O.ii (Vf- 5.87)2/3
1 ^container_opemng\v 1 j
Where:
Release_perDayRP5 = Point 5 fugitive emissions from washing [kg/site-day]
Xdeanunioad = Vapor pressure correction factor release point 5
[mol 1,4-dioxane/mol water]
MW = 1,4-dioxane molecular weight [g/mol]
Page 338 of 484
-------
8435
8436
8437
8438
8439
8440
8441
8442
8443
8444
8445
8446
8447
8448
8449
8450
8451
8452
8453
8454
8455
8456
8457
8458
8459
8460
8461
8462
8463
8464
8465
8466
8467
8468
8469
8470
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
VP = Vapor pressure [torr]
T = Ambient temperature [K]
Rateair speed = Air speed [cm/s]
Dcontainer opening = Diameter of the opening for containers [cm]
P = Atmospheric pressure [atm]
OH = Operating hours [hours/day]
Release Point 6 site release per day (washing and drying) is calculated using the equations and criteria
below:
EquationApx E-26
If £f=i Release_perDayRPi < Qdioxane_day:
5
Release _per Day RP6 Qdioxane_day I Release_perDayRPi
i=1
If £f=i Release_perDayRPi > Qdioxane_day:
Liquid detergent:
Release.jperDayRP6 = Qdioxane_day ~ Release jperD ayRP1
Powder detergent:
Release _perD ay RPb Qdioxane_day ReleasepgrDay RP^ Release _per Day RPA
Point 1 container residual releases [kg/site-day]
Point 2 fugitive emissions from container cleaning [kg/site-day]
Point 3 fugitive emissions from unloading [kg/site-day]
Point 4 fugitive dust emissions [kg/site-day]
Point 5 fugitive emissions from washing [kg/site-day]
Point 6 daily site releases (washing and drying) [kg/site-day]
Daily usage rate of 1,4-dioxane [kg/site-day]
The sum of release points 1-5 emissions [kg/site-day]
Where:
Release_perDayRP1 =
Release_perDayRP2 =
Release_perDayRP3 =
Release_perDayRP4 =
Release_perDayRPS =
Release_perDayRP6 =
Qdioxane_day ~
£f=1 Release _perD ay RPi =
E.12.2 Model Input Parameters
Table Apx E-14 summarizes the model parameters and their values for the Monte Carlo simulation.
Additional explanations of EPA's selection of the distributions for each parameter are provided after this
table. High-end and central tendency releases are estimated by selecting the 50th and 95th percentile
values from the output distribution.
Page 339 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Table Apx E-14. Summary of Paramel
ter Values and Distributions Used in the Industrial and Institutional Laundry E
please Model
Input Parameter
Symbol
Unit
Constant Model
Parameter Values
Variable Model Parameter Values
Rational/ Basis
Value
Lower
Bound
Upper
Bound
Mode
Distribution
Type
Operating Days
OD
days/year
260
Industrial:
19
Institutional:
249
366
260
Triangular
See Section
E.12.3
Mass Fraction of
1,4-dioxane in
Laundry Detergent
Fdioxane laundry
kg/kg
0.000014
5.00E-08
0.000014
Uniform
See Section
E.12.4.
Daily Use Rate of
Liquid Laundry
Detergents
Qfacility day liquid
kg/day
Industrial:
35.7
Institutional:
16
Industrial:
0.116
Institutional:
0.124
Industrial:
814
Institutional:
513
Discrete
See Section
E.12.5
Daily Use Rate of
Powder Laundry
Detergents
Qfacility day_powder
kg/day
Industrial:
110.45
Institutional:
8.63
Industrial:
1.33
Institutional:
3.71
Industrial:
1,917.44
Institutional:
15
Discrete
See Section
E.12.5
Container Size
V container
gal
55
5
550
55
Triangular
See Section
E.12.6
Air Speed
RATEair speed
cm/s
10
1.3
202.2
—
Lognormal
See Section
E.12.7
Container residual
fraction for Totes
Fcontainer residue totes
kg/kg
0.002
0.0002
0.002
0.0007
Triangular
See Section
E.12.8
Container residual
fraction for Drums
Fcontainer residue drums
kg/kg
0.03
0.017
0.03
0.025
Triangular
See Section
E.12.9
Container residual
fraction for Pails
Fcontainer residue_pails
kg/kg
0.006
0.0003
0.006
0.003
Triangular
See Section
E.12.10
Container residual
fraction for Powders
Fcontainer residue_powders
kg/kg
0.01
—
—
—
—
See Section
E.12.11
Fraction of Laundry
Detergents
Containing 1,4-
dioxane
Fformulations dioxane
unitless
0.5
0.111
1
Industrial:
Discrete
Institutional:
See Section
E.12.12
Page 340 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Input Parameter
Symbol
Unit
Constant Model
Parameter Values
Variable Model Parameter Values
Rational/ Basis
Value
Lower
Bound
Upper
Bound
Mode
Distribution
Type
Uniform
Duration of Release
for Container
Unloading
OUcont unload
h/day
Industrial:
0.0043
Institutional:
0.0114
Industrial:
0.0043
Institutional:
0.0114
Industrial:
12
Institutional:
8
Uniform
See Section
E. 12.13
Fraction of chemical
lost during transfer
of solid powders
F dust_generation
kg/kg
0.0050
0.0010
0.03
0.005
Triangular
See Section
E. 12.14
Control efficiency
for dust control
methods
Fdust control
kg/kg
0.7900
0.0000
1
0.79
Triangular
See Section
E. 12.15
Capture efficiency
for dust capture
methods
Fdust capture
kg/kg
0.9633
0.9310
1
0.9633
Triangular
See Section
E. 12.16
Number of Sites
Ns
sites
Industrial:
2,453
Institutional:
95,533
See Section
E. 12.17
Vapor Pressure of
1,4-dioxane
VP
Torr
40
—
—
—
—
Physical
property
Molecular Weight of
1,4-dioxane
MW
g/mol
88.1
—
—
—
—
Physical
property
Diameter of
Container Opening
Dcontainer opening
cm
5.08
—
—
—
—
See Section
E. 12.18
Diameter of Wash
Opening
Dwash opening
cm
73
—
—
—
—
See Section
E. 12.19
Ambient
Temperature
T
K
298
—
—
—
—
Process
parameter
Ambient Pressure
P
atm
1
—
—
—
—
Process
parameter
Page 341 of 484
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
Input Parameter
Symbol
Unit
Constant Model
Parameter Values
Variable Model Parameter Values
Rational/ Basis
Value
Lower
Bound
Upper
Bound
Mode
Distribution
Type
Dilution Factor
F dilution
unitless
0.016
—
—
—
—
See Section
E. 12.20
Density of Laundry
Detergent
RHOform
kg/L
1
ESD assumes a
density equal to
that of water
Container Fill Rate
RATEfin
containers
/ hour
20
—
—
—
—
See Section
E.12.21
8472
Page 342 of 484
-------
8473
8474
8475
8476
8477
8478
8479
8480
8481
8482
8483
8484
8485
8486
8487
8488
8489
8490
8491
8492
8493
8494
8495
8496
8497
8498
8499
8500
8501
8502
8503
8504
8505
8506
8507
8508
8509
8510
8511
8512
8513
8514
8515
8516
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.12.3 Operating Days
EPA modeled the operating days per year using a triangular distribution with a lower bound of 20 days
per year, an upper bound of 365 days per year, and a mode of 260 days per year for industrial laundries.
EPA used a triangular distribution with a lower bound of 250 days per, an upper bound of 365 days per
year, and a mode of 260 days per year for institutional laundries. This is based on the ESD on the
Chemicals Used in Water Based Washing Operations at Industrial and Institutional Laundries (
2i ). The ESD provides the range and average of operating days for six separate years, which EPA
took the minimum, maximum, and average of the 6 years to form the distributions.
E.12.4 Mass Fraction of 1,4-Dioxane in Laundry Detergent
EPA modeled the mass fraction of 1,4-dioxane in laundry detergent using a uniform distribution with a
lower bound of 5.00x 10~8 kg of 1,4-dioxane/kg detergent and an upper bound of 1.4/ 10 5 kg of 1,4-
dioxane/kg detergent for both industrial and institutional laundries. This is based on the December 2020
Final Risk Evaluation for 1,4-Dioxane ( ,020c). The risk evaluation indicates that 1,4-dioxane
is a byproduct in the laundry detergents and provides a concentration range of 0.05 to 14 ppm of 1,4-
dioxane in the detergent.
E.12.5 Daily Use Rate of Detergent
EPA modeled the daily use rate of detergent using a discrete distribution. For industrial laundries, the
distribution ranged from 0.116 kg/day to 814 kg/day for liquid detergents and 1.33 kg/day to 1917.44
kg/day for powder detergents. For institutional laundries, the distribution ranged from 0.124 kg/day to
513 kg/day for liquid detergents and 3.71 kg/day to 15 kg/day for powder detergents. This discrete data
was pulled from survey data from laundries sites used in the ESD on the Chemicals Used in Water
Based Washing Operations at Industrial and Institutional Laundries (OECD. ^ ). Equal probability
was given to each discrete survey value.
E.12.6 Container Size
EPA modeled container size using a triangular distribution with a lower bound of 5 gallons, an upper
bound of 550 gallons, and a mode of 55 gallons for industrial laundries. Because EPA expects industrial
laundries to have variation in the sizes of containers, EPA used values of 5, 55, and 550 gallons for the
triangular distribution based on the default values from the EPA/OPPT Small Container Residual Model,
Drum Residual Model, and Bulk Transport Residual Model, respectively.
EPA used a single value of 5 gallons for institutional laundries. This is based on the ESD on the default
value for institutional laundries from the Chemicals Used in Water Based Washing Operations at
Industrial and Institutional Laundries ESD (OEC ).
E.12.7 Indoor Air Speed
Baldwin (1998) measured indoor air speeds across a variety of occupational settings in the United
Kingdom. Fifty-five work areas were surveyed across a variety of workplaces. EPA analyzed the air
speed data from Baldwin (1998) and categorized the air speed surveys into settings representative of
industrial facilities and representative of commercial facilities. EPA fit separate distributions for these
industrial and commercial settings and used the industrial distribution for laundry facilities.
EPA fit a lognormal distribution for both data sets as consistent with the authors observations that the air
speed measurements within a surveyed location were lognormally distributed and the population of the
mean air speeds among all surveys were lognormally distributed. Since lognormal distributions are
bound by zero and positive infinity, EPA truncated the distribution at the largest observed value among
Page 343 of 484
-------
8517
8518
8519
8520
8521
8522
8523
8524
8525
8526
8527
8528
8529
8530
8531
8532
8533
8534
8535
8536
8537
8538
8539
8540
8541
8542
8543
8544
8545
8546
8547
8548
8549
8550
8551
8552
8553
8554
8555
8556
8557
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
all of the survey mean air speeds from Baldwin (1998). EPA fit the air speed surveys representative of
industrial facilities to a lognormal distribution with the following parameter values: mean of 22.414
cm/s and standard deviation of 19.958 cm/s. In the model, the lognormal distribution is truncated at a
maximum allowed value of 202.2 cm/s (largest surveyed mean air speed observed in Baldwin (1998)) to
prevent the model from sampling values that approach infinity or are otherwise unrealistically large.
Baldwin (1998) only presented the mean air speed of each survey. The authors did not present the
individual measurements within each survey. Therefore, these distributions represent a distribution of
mean air speeds and not a distribution of spatially variable air speeds within a single workplace setting.
E.12.8 Container Residual Fraction for Totes
EPA modeled container residual fraction for totes using a triangular distribution with a lower bound of
0.0007 kg residual/kg detergent, and upper bound of 0.002 kg residual/kg detergent, and a mode of
0.0007 kg residual/kg detergent. The lower and upper bounds of this distribution are based on the central
tendency and high-end values listed in the EPA/OPPT Bulk Transport Residual Model from the
ChemSTEER User Guide ( ). EPA used the central tendency value as the mode of the
triangular distribution.
E.12.9 Container Residual Fraction for Drums
EPA modeled container residual fraction for drums using a triangular distribution with a lower bound of
0.0003 kg residual/kg detergent, an upper bound of 0.03 kg residual/kg detergent, and a mode of 0.025
kg residual/kg detergent. The lower bound is based on the minimum value for pouring and the upper
bound is based on the default high-end value in the EPA/OPPT Drum Residual Model from the
ChemSTEER User Guide ( 15a). EPA used the central tendency value for pumping as the
mode of the triangular distribution.
E.12.10 Container Residual Fraction for Pails
EPA modeled container residual fraction for small containers using a triangular distribution with a lower
bound of 0.0003 kg residual/kg detergent, an upper bound of 0.006 kg residual/kg detergent, and a mode
of 0.003 kg residual/kg detergent. The lower bound is based on the minimum value for pouring and the
upper bound is based on the default high-end value listed in the EPA/OPPT Small Container Residual
Model from the ChemSTEER User Guide ( ) EPA used the central tendency value for
pouring as the mode of the triangular distribution.
E.12.11 Container Residual Fraction for Powders
The EPA/OPPT Solid Residuals in Transport Containers Model provides a loss fraction 0.01 kg of solid
chemicals remaining in a container per kg transported. Therefore, EPA could not develop a distribution
of values for this parameter and used the single value 0.01 kg/kg from the model ( |).
E.12.12 Fraction of Laundry Detergents Containing 1,4-Dioxane
EPA modeled the fraction of laundry detergents containing 1,4-dioxane using a discrete distribution. For
industrial and institutional laundries, the distribution ranged from 0.111 to 1 kg detergents containing
1,4-dioxane/kg all detergents. This discrete data was pulled from survey data from laundries sites used
in the ESD on the Chemicals Used in Water Based Washing Operations at Industrial and Institutional
Laundries (OEC ). Equal probability was given to each discrete survey value.
Page 344 of 484
-------
8558
8559
8560
8561
8562
8563
8564
8565
8566
8567
8568
8569
8570
8571
8572
8573
8574
8575
8576
8577
8578
8579
8580
8581
8582
8583
8584
8585
8586
8587
8588
8589
8590
8591
8592
8593
8594
8595
8596
8597
8598
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.12.13 Duration of Release for Container Unloading
EPA modeled the duration of release for container unloading using a uniform distribution. For industrial
and institutional laundries, EPA assumed the distribution had a maximum of 12 hours/day. The lower
bound was based on the length of time to unload detergent containers each day, calculated using the
number of containers used per day and the container fill rate (see Section E. 12.21). This means that each
iteration of the simulation would calculate a new lower bound based on the parameters for that iteration.
E.12.14 Fraction of Chemical Lost during Transfer of Solid Powders
EPA modeled the fraction of chemical lost during transfer of solid powders using a triangular
distribution with a lower bound of 0.001 kg dust lost/kg transferred, an upper bound of 0.03 kg dust
lost/kg transferred, and a mode of 0.005 kg dust lost/kg transferred for both industrial and institutional
laundries. These values were taken from the EPA/OPPT Dust Emissions from Transferring Solids Model
from the ChemSTEER User Guide ( 015a).
E.12.15 Control Efficiency for Dust Control Methods
EPA modeled the control efficiency for dust control methods using a triangular distribution with a lower
bound of 0 kg controlled/kg transferred, an upper bound of 1 kg controlled/kg transferred, and a mode of
0.79 kg controlled/kg transferred for both industrial and institutional laundries. These values were taken
from the EPA/OPPT Dust Emissions from Transferring Solids Model from the ChemSTEER User Guide
(I >015a).
E.12.16 Capture Efficiency for Dust Capture Methods
EPA modeled the capture efficiency for dust capture methods using a triangular distribution with a
lower bound of 0.9310 kg captured/kg transferred, an upper bound of 1 kg captured/kg transferred, and a
mode of 0.9633 kg captured/kg transferred for both industrial and institutional laundries. These values
were taken from the EPA/OPPT Dust Emissions from Transferring Solids Model from the ChemSTEER
User Guide ( ).
E.12.17 Number of Sites
EPA did not find data on the number of laundry sites that specifically use detergents containing 1,4-
dioxane. As a bounding estimate for the number of industrial laundries, EPA used U.S. Census and BLS
data for the NAICS code 812330, Linen and Uniform Supply, to estimate a total of 2,453 industrial
laundry sites within the industry (\' S HI S 101 * -). As a bounding estimate for the number of
institutional sites, EPA used industry information as described in the ESD to estimate a total of 95,533
institutional laundries (OECD. 201 lb).
E.12.18 Diameter of Container Opening
The ChemSTEER User Guide (U.S. EPA. 2015a) provides a typical diameter of container openings as
5.08 cm. Therefore, EPA could not develop a distribution of values for this parameter and used the
single value 5.08 cm from the ChemSTEER User Guide.
E.12.19 Diameter of Wash Opening
The ESD on the Chemicals Used in Water Based Washing Operations at Industrial and Institutional
Laundries provided a single value for the diameter of washer openings to estimate air releases during
operation ( ). The ESD states that the wash opening is 73 cm (OECD. 2017). Therefore,
EPA could not develop a distribution of values for this parameter and used the single value of 73 cm
from the ESD.
Page 345 of 484
-------
8599
8600
8601
8602
8603
8604
8605
8606
8607
8608
8609
8610
8611
8612
8613
8614
8615
8616
8617
8618
8619
8620
8621
8622
8623
8624
8625
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
E.12.20 Dilution Factor
The December 2020 Final Risk Evaluation for 1,4-Dioxane (U.S. EPA. 2020c) provided a single value
for the dilution factor of 1,4-dioxane in laundry detergents. The risk evaluation states that a dilution
factor of 0.016 was estimated assuming a high-end mass of product used (60g) in one gallon of water
(U.S. EPA. 2020c). Therefore, EPA could not develop a distribution of values for this parameter and
used the single value of 0.016 from the 2020 RE.
E.12.21 Container Fill Rate
The ChemSTEER User Guide (U.S. EPA. 2015a) provides a typical fill rate of 20 containers per hour
for containers with 20 to 100 gallons of liquid. Therefore, EPA could not develop a distribution of
values for this parameter and used the single value 20 containers/hour from the ChemSTEER User
Guide.
E.13 Hydraulic Fracturing Modeling Approach and Parameters for
Estimating Environmental Releases
This appendix presents the modeling approach and equations used to estimate environmental releases of
1,4-dioxane during hydraulic fracturing. This approach utilizes the Draft ESD on Chemicals Used in
Hydraulic Fracturing (U.S. EPA. 2022d) combined with Monte Carlo simulation (a type of stochastic
simulation). This ESD includes a diagram of release and exposure points during textile dying, as shown
in FigureApx E-4.
FigureApx E-4. Environmental Release and Occupational Exposure Points during Hydraulic
Fracturing
Based on Figure Apx E-4, EPA identified the following release points:
• Release point 1 (RP1): Fugitive air releases during container unloading;
• Release point 2 (RP2): Container residue losses to surface water, incineration, or landfill;
• Release point 3 (RP3): Fugitive air releases during container cleaning;
Page 346 of 484
-------
8626
8627
8628
8629
8630
8631
8632
8633
8634
8635
8636
8637
8638
8639
8640
8641
8642
8643
8644
8645
8646
8647
8648
8649
8650
8651
8652
8653
8654
8655
8656
8657
8658
8659
8660
8661
8662
8663
8664
8665
8666
8667
8668
8669
8670
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
July 2023
• Release point 4 (RP4): Equipment and storage tank cleaning losses to surface water, incineration,
or landfill;
• Release point 5 (RP5): Fugitive air releases during equipment and storage tank cleaning;
• Release point 6 (RP6): Release of hydraulic fracturing fluid that remains underground to deep
well injection; and
• Release point 7 (RP7): Hydraulic fracturing fluid flowback and produced wastewater to
recycle/reuse, deep well injection, surface water, or land.
Environmental releases of hydraulic fracturing are a function of the chemical's physical properties,
container size, mass fractions, and other model parameters. Although physical properties are fixed for a
chemical, some model parameters are expected to vary from one facility to another. An individual model
input parameter could either have a discrete value or a distribution of values. EPA assigned statistical
distributions based on available literature data or engineering judgment to address the variability in
operating days (OD), mass fraction of 1,4-dioxane in fracturing fluid (Fdioxane fracturing fluid), mass fraction
of 1,4-dioxane in additive (Fdioxane additive), container container size (Vcont), annual use rate of fracturing
fluids (Qsite yr), saturation factor (Fsaturation), container cleaning losses (FCOnt cleaning), and fraction of
injected fracturing fluid that returns to the surface (Frecovered).
A Monte Carlo simulation (a type of stochastic simulation) was conducted to capture variability in the
model input parameters. The simulation was conducted using the Latin hypercube sampling method in
@Risk (Palisade, Ithaca, NY). The Latin hypercube sampling method is a statistical method for
generating a sample of possible values from a multi-dimensional distribution. Latin hypercube sampling
is a stratified method, meanin |