United Stales
Environmental Protection
Agency
Office of Toxic Substances
TS-792
EPA 560/5-90-014
July 1990
Background Document
to the Integrated Risk
Assessment for Dioxins
and Furans from Chlorine
Bleaching in Pulp and
Paper Mills
-------
EPA 560/5-90-014
July 1990
BACKGROUND DOCUMENT TO THE INTEGRATED RISK ASSESSMENT
FOR DIOXINS AND FURANS FROM CHLORINE BLEACHING IN
PULP AND PAPER MILLS
by:
Greg Schweer, Bentley Gregg, Lee Schultz
Patricia Wood, Timothy Leighton, Carl D'Ruiz
Robert Fares, Geoffrey Huse, Clay Carpenter,
James Konz and Daniel Arrenholz
EPA Contract No. 68-D9-0166
Project Officer:
Thomas M. Murray
Exposure Evaluation Division
Exposure Assessment Branch
401 M Street, SW
Washington, D.C. 20460
U.S. Environmental Protection Agency
Office of Pesticides and Toxic Substances
Washington, DC 20460
July 1990
-------
DISCLAIMER
This document has been reviewed and approved for publication by the
Office of Toxic Substances, Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency. The use of trade names or
commercial products does not constitute Agency endorsement or
recommendation for use.
i 11
-------
IV
-------
1.
2.
3.
TABLE OF CONTENTS
Page No,
1.1
1.2
1.3
1.4
1.5
Background
Purpose and Scope
Industry Profile
1.3.1 Overview of the Industry
1.3.2 Processes/Operation
1.3.3 Potential Source of Dioxin/Furans and Their
Concentrations in Pulp and Paper
1.3.4 End-Uses of Products from Pulp and Paper .....
Organization of This Report
References
CHEMISTRY AND FATE OF DIOXINS AND FURANS
2.1
2.2
2.3
2.4
Chemical Identity of PCDDs and PCDFs
Chemistry and Fate of 2,3,7,8-TCDD
2.2,1 Chemical Identity..
2.2.2 Chemical and Physical Properties
2.2.3 Environmental Fate and Transport
Chemistry and Fate of 2,3,7,8-TCDF
2.3.1 Chemical Identity
2.3.2 Chemical and Physical Properties
2.3.3 Environmental Fate and Transport
References
1-1
1-7
1-16
1-16
1-16
1-18
1-18
1-20
1-21
2-1
2-1
2-5
2-5
2-5
2-5
2-24
2-24
2-24
2-25
2-28
DIOXIN AND FURAN HAZARD ASSESSMENT: HUMANS, TERRESTRIAL
AND
3.1
3.2
3.3
AVIAN WILDLIFE, AND AQUATIC LIFE
Introduction
Human Hazard Assessment of PCDDs and PCDFs
Human Health Hazard of 2,3,7,8-TCDD
3.3.1 Cancer Effects
3.3.2 Non-Cancer Effects
3.3.3 Toxicity Equivalence Factors
3-1
3-1
3-1
3-3
3-3
3-7
3-9
-------
TABLE OF CONTENTS (continued)
Pace No,
3.4 Ecological Hazard of 2,3,7,8-TCDD and 2,3,7,8-TCDF ... 3-15
3.4.1 Aquatic Toxicity of Dioxins 3-16
3.4.2 Aquatic Toxicity of PCDFs 3-26
3.4.3 Conclusions Concerning Aquatic Toxicity 3-26
3.5 Toxicity of PCDDs and PCDFs to Wildlife 3-27
3.5.1 Toxicity Assessment for Birds 3-28
3.5.2 Toxicity Assessment for Bird Eggs 3-28
3.5.3 Toxicity Assessment for Wildlife Mammals 3-31
3.6 Analysis of Uncertainties 3-31
3.7 Conclusions 3-33
3.8 References . 3-35
4. ASSESSMENT OF RISKS TO WORKERS FROM EXPOSURE TO DIOXINS
AND FURANS FROM MANUFACTURE, PROCESSING, AND COMMERCIAL USE
OF PULP, PAPER, AND PAPER PRODUCTS AND FROM PROCESSING AND
COMMERCIAL USE OF PULP AND PAPER MILL SLUDGE 4-1
4.1 Introduction 4-1
4.2 Worker Exposure to Dioxins in the Manufacture,
Processing, and Commercial Use of Pulp, Paper, and
Paper Products 4-1
4.2.1 Pulp and Paper Industry Workforce
Characterization 4-2
4.2.2 Worker Exposure Estimating Methodologies 4-7
4.2.3 Summary of Worker Inhalation and Dermal
Exposure, Individual Cancer Risks, and
Population Cancer Risks 4-19
4.3 Worker Exposure from Processing and Commercial Use of
Pulp and Paper Mill Sludge 4-20
4.3.1 Sludge Formation, Processing, and Disposal
Operations 4-20
4.3.2 Dioxins in Pulp and Paper Mill Sludges 4-25
4.3.3 Sludge Handling/Disposal Workforce
Characterization 4-26
-------
TABLE OF CONTENTS (continued)
Page No.
4.3.4 Worker Exposure Estimating Methodologies 4-28
4.3.5 Summary of Worker Exposure, Individual Cancer
Risks, and Population Cancer Risks from
Processing and Commercial Use of Sludge 4-35
4.4 Analysis of Uncertainties 4-35
4.4.1 Worker Exposure from Manufacture, Processing,
and Commercial Use of Pulp, Paper, and Paper
Products 4-35
4.4.2 Worker Exposure from Sludge Processing and
Commercial Use 4-38
4.5 References 4-39
5. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM
EXPOSURE TO DIOXINS AND FURANS DURING USE AND DISPOSAL OF
PULP AND PAPER MILL SLUDGE AND DISPOSAL OF PAPER WASTES .... 5-1
5.1 Introduction 5-1
5.2 Estimates of Exposure and Risks to the General
Populations from Disposal and Use of Sludge from the
Pulp and Paper Industry and Disposal of Paper Products 5-3
5.2.1 Exposure and Risks from Disposal of Pulp and
Paper Sludge in Landfills 5-5
5.2.2 Exposure and Risks from Disposal of Paper
Products in Municipal Landfills 5-18
5.2.3 Exposures and Risks from Disposal of Pulp and
Paper Sludge in Surface Impoundments 5-26
5.2.4 Exposures and Risks from Land Application of
Pulp and Paper Mill Sludge 5-39
5.2.5 Exposures and Risks from Distribution and
Marketing of Pulp and Paper Sludge 5-74
5.3 Analysis of Uncertainty 5-96
5.4 Conclusions 5-99
5.5 References 5-101
6. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM THE
DISCHARGE OF EFFLUENTS FROM THE PULP AND PAPER INDUSTRY 6-1
6.1 Introduction 6-1
6.1.1 Purpose 6-1
6.1.2 Scope 6-1
vii
-------
TABLE OF CONTENTS (continued)
Page No.
6.2 Exposure and Risk Assessment Methodology Requirements 6-2
6.3 Exposure Assessment Methodology 6-3
6.4 Risk Assessment Methodology 6-8
6.4.1 Bioavailable Dose from Ingestion of
Contaminated Fish Tissue and Drinking Water . 6-8
6.5 Results of the Assessment 6-10
6.5.1 Exposure Assessment Results 6-11
6.5.2 Risk Assessment Results 6-14
6.6 Discussion of Results 6-23
6.6.1 Assumptions, Limitations, and Uncertainties . 6-23
6.6.2 Conclusions 6-29
6.7 References 6-29
7. ASSESSMENTS OF RISKS TO THE GENERAL POPULATION FROM EXPOSURE
TO DIOXINS AND FURANS RESULTING FROM PULP/PAPER WASTEWATER
SLUDGE INCINERATION 7-1
7.1 Introduction 7-1
7.2 Methodology 7-1
7.2.1 Unit Risk Estimate 7-1
7.2.2 EPA Human Exposure Model (HEM) (Background) .. 7-1
7.2.3 Pulp and Paper Mill Source Data 7-2
7.2.4 Risk Calculations 7-8
7.3 Results 7-9
7.4 Analytical Uncertainties 7-9
7.4.1 The Unit Risk Estimate 7-9
7.4.2 Emission Estimates 7-9
7.4.3 Sludge Burning at Municipal Incinerators 7-15
7.4.4 Exposure Assumptions 7-15
7.4.5 Conclusions 7-16
7.5 References 7-17
-------
TABLE OF CONTENTS (continued)
Page No.
8. ASSESSMENT OF CANCER RISK FROM EXPOSURE TO PCODs AND
PCDFs IN CONSUMER PRODUCTS 8-1
8.1 Introduction 8-1
8.2 Methodology 8-1
8.2.1 Exposure Assessment 8-1
8.2.2 Risk Assessment 8-11
8.3 Results 8-14
8.3.1 Individual Cancer Risk 8-14
8.3.2 Population Cancer Risk 8-17
8.3.3 Non-Cancer Endpoints 8-17
8.4 Uncertainty Analysis 8-17
8.4.1 Liquid Mediated Extraction 8-17
8.4.2 Unmediated Migration 8-19
8.4.3 Use of in vitro Percutaneous Absorption Data . 8-19
8.4.4 Factors Affecting Percutaneous Absorption .... 8-19
8.5 References 8-20
9. ASSESSMENT OF RISKS TO THE GENERAL POPULATION EXPOSURE TO
DIOXINS AND FURANS RESULTING FROM THE USE OF PULP-CONTAINING
MEDICAL DEVICES 9-1
9.1 Introduction 9-1
9.2 Estimates of Exposures and Risks from Dermal Contact
with Pulp-Containing Medical Devices 9-1
9.2.1 Exposure Parameters 9-1
9.2.2 Exposure/Risk Assessment for Medical Devices . 9-8
9.3 Uncertainty Analysis 9-11
9.4 Conclusions 9-12
9.5 References 9-13
ix
-------
TABLE OF CONTENTS (continued)
Page No.
10. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM DIOXINS
AND FURANS IN FOODS PACKAGED IN OR CONTACTING BLEACHED
PAPER PRODUCTS 10-1
10.1 Introduction 10-1
10.2 Dioxin Concentrations In Bleached Wood Pulp and Paper
Food-Contact Articles 10-2
10.3 Food-Paper Migration Studies 10-3
10.3.1 Types of Articles Investigated 10-3
10.3.2 Conclusions from the Migration Studies 10-5
10.4 Food Intake Information 10-6
10.5 Estimated Exposures 10-7
10.5.1 Estimated Total Dioxin TEQ Intake: All Foods
Contacting Bleached Paper 10-7
10.5.2 Estimated Dioxin TEQ Intake: Food-by-Food
Basis for Individual Paper Products 10-10
10.6 Risk Assessment 10-10
10.6.1 Cancer Risks 10-10
10.6.2 Non-Cancer Risks 10-13
10.6.3 Uncertainties 10-13
10.7 References 10-14
11. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM DIOXINS
AND FURANS IN CELLULOSE DERIVATIVES USED IN FOOD, DRUG, AND
COSMETIC FORMULATIONS 11-1
11.1 Introduction 11-1
11.2 Exposure to Dioxins and Furans from Use of Cosmetic
Products 11-1
11.2.1 Identity and Use of Cellulose Derivatives 11-1
11.2.2 Dioxin Concentration in Cellulose Derivatives 11-1
11.2.3 Cosmetic Product Use 11-2
11.2.4 Dermal Absorption of Dioxin from Cosmetic
Products Applied to the Skin 11-3
11.2.5 Oioxin Exposure from Cosmetic Products Applied
to the Skin 11-3
-------
TABLE OF CONTENTS (continued)
Page No.
11.3 Exposure to Dloxins and Furans from Use of Cosmetic
Wet Wipes 11-3
11.3.1 Identity and Use of Cosmetic Wet Wipes 11-3
11.3.2 Dioxin Concentration In and Extraction From
Wet Wipes 11-5
11.3.3 Wet Wipe Use Information 11-5
11.3.4 Dermal Absorption of Dioxin Congeners 11-6
11.3.5 Dioxin Exposure from Wet Wipes 11-6
11.4 Exposure to Dioxins and Furans in Cellulose Derivatives
Used in Foods and Drug Products 11-6
11.4.1 Identity and Use of Cellulose and Cellulose
Derivatives — 11-6
11.4.2 Dioxin Concentration in Cellulose Derivatives 11-8
11.4.3 Dioxin Exposure Estimates 11-8
11.5 Cancer Risk Estimates 11-14
11.6 Analysis of Uncertainties 11-14
11.7 References 11-17
12. FDA ASSESSMENT OF RISKS FROM EXPOSURE TO DIOXIN AND FURANS
IN FISH CONTAMINATED BY BLEACHED KRAFT PULP AND PAPER MILLS 12-1
12.1 Introduction 12-1
12.2 Levels of Dioxin Congeners in Fish 12-1
12.3 Sources of Information on Fish Intake 12-2
12.4 Estimation of Fish Intake by Subsistence and Sports
Fishers and Their Families 12-5
12.4.1 Subsistence Fishers 12-5
12.4.2 Sports Fishers 12-6
12.5 Risk Assessment for Cancer 12-7
12.6 Non-Cancer Toxicological Effects of Dioxins 12-9
12.7 Risk of Non-Cancer Toxicological Effects to
Subsistence and Sports Fishers 12-10
12.8 References 12-11
-------
TABLE OF CONTENTS (continued)
Pace No.
13. ESTIMATES OF RISKS TO TERRESTRIAL AND AVIAN WILDLIFE FROM
LAND APPLICATION OF PULP AND PAPER MILL SLUDGE AND TO
AQUATIC LIFE FROM DISCHARGE OF EFFLUENTS 13-1
13.1 Introduction 13-1
13.2 Terrestrial and Avian Wildlife Risk Assessment 13-2
13.2.1 Development of Benchmark Doses to Which
Terrestrial Wildlife Exposures (Adjusted for
Adsorption) Are Compared 13-2
13.2.2 Estimating Exposures to Terrestrial Wildlife . 13-4
13.2.3 Summary of Results: Terrestrial Wildlife 13-21
13.3 Aquatic Life Risk Assessment 13-25
13.3.1 Risk to Fish 13-25
13.3.2 Risks to Aquatic Plants and Herbivores 13-27
13.3.3 Risks to Benthic Organisms 13-28
13.3.4 Risks to Fish-Eating Birds and Mammals 13-29
13.4 References 13-29
APPENDIX A - Bioavailability A-l
APPENDIX B - Summary Information on Chlorinated Chemicals Other
than PCDDs and PCDFs (OCOs) Identified in Pulp
Mill Effluents, Sludges, and Pulps B-l
-------
LIST OF TABLES
Page No.
Table 1-1. Concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
Unbleached and Bleached Kraft Pulps from the Five-
Mill Study (Cooperative Dioxin Screening Study)
(ppt) 1-3
Table 1-2. Concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
Filtrates from Various Stages of Bleaching Process
in the Five-Mill Study (Cooperative Dioxin
Screening Study) (ppt) 1-5
Table 1-3. Summary Results of the 104-Mill Study 1-6
Table 1-4. Concentration (ppt) of 2,3,7,8-TCDD and 2,3,7,8-TCDF
Consumer Products 1-8
Table 1-5. Exposure Routes and Pathways Examined for Each
Exposure/Risk Assessment Performed for the Dioxin-
in Paper Project 1-10
Table 1-6. Effect of Defoamer Addition to Brownstock Kraft
Pulp on Subsequent Dioxin Formation During the
Bleaching Process 1-19
Table 2-1. Possible Number of Positional PCDD and PCDF Isomers 2-2
Table 2-2. Summary of Estimated Physical/Chemical Properties
of PCDFs and PCDDs 2-4
Table 2-3. Physical and Chemical Properties of 2,3,7,8-TCDD .. 2-6
Table 2-4. Summary of Environmental Fate of Dibenzo-p-
Dioxins 2-12
Table 2-5. Physical and Chemical Properties of 2,3,7,8-TCDF .. 2-20
Table 2-6. Summary of Environmental Fate of Dibenzofurans — 2-26
Table 3-1. Factors Used by EPA, FDA, and CPSC in Calculating
Their Risk Estimates for 2,3,7,8-TCDD Using
Multistage Models 3-4
Table 3-2. Terms Associated with Dose-Response Modeling 3-5
Table 3-3. I-TEFs/89 and TEFs Developed by Other Groups 3-11
xm
-------
LIST OF TABLES (continued)
Page No,
Table 3-4. International Toxicity Equivalency Factors/89
(I-TEFs/89): Comparison of Relative Potency Data
for the 2,3,7,8-Substituted CDDs and CDFs 3-13
Table 3-5. Aquatic Toxicity Data for Chlorinated
Dibenzo-p-Dioxins 3-17
Table 3-6. Aquatic Toxicity Data for Chlorinated
Dibenzofurans 3-23
Table 3-7. Summary of Studies on the Toxicity of 2,3,7,8-TCDD
to Wildlife Mammals and Birds 3-29
Table 4-1. Total Employees and Production Workers in the
Entire Paper and Allied Products Industry, 1985 ... 4-3
Table 4-2. Number of Workers in the Pulp and Pulp Products
Job Categories 4-6
Table 4-3. Summary of Individual and Population Cancer Risks
for Workers Involved in the Manufacturing,
Processing, and Commercial Usage of Pulp, Paper,
and Paper Products 4-21
Table 4-4. Summary of Outer Bounds of Individual and Population
Cancer Risks for Workers Involved in Processing, and
Commercial Usage of Pulp, and Paper Mill Sludge ... 4-36
Table 5-1. Distribution of 2,3,7,8-TCDD and 2,3,7,8-TCDF Sludge
Concentrations for All Plants in 104 Mill Study ... 5-2
Table 5-2. Exposure Pathways Evaluated for Each Pulp and Paper
Mill .Sludge Disposal or Use Practice 5-4
Table 5-3. Physical/Chemical Properties and Fate/Transport
Assumptions: All Exposure Pathways 5-6
Table 5-4. Assumptions and Parameter Values - Landfills:
All Exposure Pathways 5-7
Table 5-5. Assumptions and Parameter Values - Landfills:
Volatilization Pathway 5-10
xi v
-------
LIST OF TABLES (continued)
Page No.
Table 5-6. Assumptions and Parameter Values - Landfills:
Ground-Water Pathways 5-12
Table 5-7. Assumptions and Parameter Values - Landfills:
Surface Water Pathways 5-16
Table 5-8. Estimates of Health Risks to the General Public
fromLandfi11 Disposal of Pulp and Paper Sludges
Contaminated with 2,3,7,8-TCDD and 2,3,7,8-TCOF ... 5-19
Table 5-9. Site and Waste Characteristics for Municipal
Landfills Receiving Waste Paper Contaminated with
TCDD and TCDF 5-20
Table 5.-10. Assumptions and Parameter Values - Paper Wastes in
Municipal Landfills: Ground-Water Pathway 5-24
Table 5-11. Estimates of Health Risks to the General Population
from Landfill Disposal of Paper Contaminated with
2,3,7,8-TCDD and 2,3,7,8-TCDF 5-27
Table 5-12. Assumptions and Parameter Values - Surface
Impoundments: Al 1 Exposure Pathways 5-29
Table 5-13. Assumptions and Parameter Values - Surface
Impoundments: Volatilization Pathway 5-30
Table 5-14. Assumptions and Parameter Values - Surface
Impoundments: Ground-Water Pathway 5-33
Table 5-15. Assumptions and Parameter Values - Surface
Impoundments: Surface Water Pathway 5-37
Table 5-16. Estimates of Health Risks to the General Population
from Surface Impoundment of Pulp and Paper Sludges
Contaminated with 2,3,7,8-TCDD and 2,3,7,8-TCDF ... 5-40
Table 5-17. Assumptions and Parameter Values - Land
Application: All Exposure Pathways 5-42
Table 5-18. Assumptions and Parameter Values - Land
Application: Dermal Pathway 5-44
xv
-------
LIST OF TABLES (continued)
Table 5-19.
Table 5-20.
Table 5-21.
Table 5-22.
Table 5-23.
Table 5-24.
Table 5-25.
Table 5-26.
Table 5-27.
Table 5-28.
Table 5-29.
Table 5-30.
Page No.
Assumptions and Parameter Values - Agricultural
Application: Dietary Pathway ..................... 5-51
Assumptions and Parameter Values - Land
Application: Soil Ingesti on Pathway .............. 5-61
Assumptions and Parameter Values - Land
Application: Vapor and Particulate Inhalation
Pathways .......................................... 5-63
Assumption and Parameter Values - Land
Application: Surface Water Pathways .............. 5-70
Estimates of Health Risks to the General Population
from Land Application of Pulp and Paper Sludges
Contaminated with 2,3,7,8-TCDD and 2,3,7,8-TCDF ... 5-75
Distribution and Marketing Sludge and Soil
Contaminant Concentration ......................... 5-78
Parameters and Model Inputs Used to Model:
Distribution and Marketing ........................ 5-79
Assumptions and Parameter Values - Distribution and
Marketing: Dermal Pathway ........................ 5-82
Assumptions and Parameter Values - Distribution and
Marketing: Dietary Pathway ....................... 5-88
Assumptions and Parameter Values - Distribution and
Marketing: Soil Ingestion Pathway ................ 5-90
Assumptions and Parameter Values - Distribution and
Marketing: Vapor and Particulate Inhalation
Pathways .......................................... 5-94
Estimates of Health Risks to the General Population
from Distribution and Marketing of Pulp and Paper
Sludges Contaminated with 2,3,7,8-TCDD and
2,3,7,8-TCDF ...................................... 5-97
xvi
-------
LIST OF TABLES (continued)
Page No.
Table 5-31. Maximum General Population Health Risks from
Exposure to TCDD and TCDF, by Waste Management
Method 5-100
Table 7-1. Locations of Pulp and Paper Mill Wastewater
Sludge Incinerators 7-3
Table 7-2. Pulp and Paper Mill Incinerator Stack Parameters .. 7-4
Table 7-3. Dioxin TEQ and Furan TEQ Emission Rates Based on
Reported Stack Gas Concentrations 7-5
Table 7-4. Dioxin/Furan Emissions and Stack Flow Rate Data
Used to Calculate Emission Rates Based on Stack Gas
Concentrations 1 7-6
Table 7-5. 2,3,7,8-TCDD and 2,3,7,8-TCDF Emission Rates Based
on Site-Specific Concentrations in Sludge 7-7
Table 7-6. Site-Specific Exposure Analysis Based on 2,3,7,8-TCDD
Concentrations in Sludge (Maximum Radius « 50 km) . 7-10
Table 7-7. Site-Specific Exposure Analysis Based on 2,3,7,8-TCDF
Concentration in Sludge (Maximum Radius « 50 km) .. 7-11
Table 7-8. Site-Specific Exposure Analysis Based on Dioxin
Concentrations in Stack Gas (Expressed as dioxin
TEQs) - Maximum Values (Maximum Radius « 50 km) ... 7-12
xvii
-------
LIST OF TABLES (continued)
Table 7-9.
Table 7-10.
Table 8-1.
Table 8-2.
Table 8-3.
Table 8-4.
Table 8-5.
Table 8-6.
Table 8-7.
Table 8-8.
Table 8-9.
Table 9-1.
Table 9-2.
Table 9-3.
Page No.
Site-Specific Exposure Analysis Based on Furan
Concentrations in Stack Gas (expressed as dioxin
TEQs) - Maximum Values (Maximum Radius = 50 km) .... 7-13
Combined Dioxin/Furan Risks and Annual Incidence .. 7-14
Concentrations (ppt) of 2,3,7,8-TCDD and 2,3,7,8-TCDF
in Consumer Products 8-3
Parameters for Estimating Exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF in Disposable Infant Diapers 8-6
Parameters for Estimating Exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF in Paper Towels 8-8
Parameters for Estimating Exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF in Facial Tissue 8-9
Parameters for Estimating Exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF in Toilet Tissue 8-10
Parameters for Estimating Exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF in Communications Paper 8-12
Parameters for Estimating Exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF in Paper Napkins 8-13
Individual Lifetime and Population Cancer Risks from
2,3,7,8-TCDD and 2,3,7,8-TCDF in Consumer Products . 8-15
Risks for Non-Cancer Adverse Effects from 2,3,7,8-
TCDD and 2,3,7,8-TCDF in Consumer Paper Products... 8-18
Medical Devices for Which Exposures and Risks Were
Estimated and Their Corresponding Uses 9-2
Exposure/Risk Parameters for Medical Devices 9-3
Average Concentrations of 2,3,7,8-TCDD and
2,3,7,8-TCDF in Pulp Calculated Based on Results
from the 104-Mill Data Base 9-5
xvm
-------
LIST OF TABLES (continued)
Table 9-4.
Table 9-5.
Table 10-1.
Table 10-2.
Table 10-3.
Table 10-4.
Table 11-1.
Table 11-2.
Table 11-3.
Table 11-4.
Table 11-5.
Table 12-1.
Table 12-2.
Paoe No.
Concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
Pulp at Pulp Mills that Produce Pulp Dissolving
Cellulose 9-6
Estimates of Risks to the General Population from
the Use of Pulp-Containing Medical Devices 9-9
Representative Dioxin Congener Levels in Pulp and
Paper Matrices 10-4
Carcinogenic Risk for Consumers Resulting from Total
Dioxin TEQ Intake from All Foods Contacting Bleached
Paper ("mean consumer - total sample basis") 10-8
Industry-Based Per Capita Dioxin TEQ Exposures 10-9 •
Upper Bound Carcinogenic Risk for Consumers of
Foods Containing Bleached Paper Contaminated with
Dioxin ("eaters only - food-by-food basis") 10-11
Dioxin Exposure from Cosmetic Products Applied to
the Skin 11-4
Dioxin Exposure from Met Wipes 11-7
Total Sample Basis Dioxin TEQ Exposure from Foods . 11-11
Upper Bound Intake of Dioxin TEQ from Drugs 11-15
Upper Bound Carcinogenic Risk for Users of Food,
Drug, and Cosmetic Products Containing Cellulose
Derivatives 11-16
Dioxin TEQ Intake by Subsistence and Sports
Fishers 12-3
Cancer Risk for Subsistence and Sports Fishers 12-8
xix
-------
LIST OF TABLES (continued)
Page No.
Table 13-1. Assumptions and Parameter Values - Land Application:
Wildlife Exposures 13-8
Table 13-2. Characteristics of Land Application Sites and Soil
Concentrations Used in Wildlife Analysis 13-13
Table 13-3. Body Weights and Daily Food Consumption of
Animals Affected by Land Application of Sludge .... 13-14
Table 13-4. Mix of Food Sources for Birds and Mammals 13-15
Table 13-5. Estimates of Exposure and Risk to Adult and Hatch-
ing Birds from 2,3,7,8-TCDO in Sludge-Treated Land. 13-
Table 13-6. Estimates of Exposure and Risk to Mammals from
2,3,7,8-TCDD in Sludge-Treated Land 13-
Table 13-7. Summary of Risks to Birds ("Best Estimate") 13-22
Table 13-8. Summary of Risks to Bird Eggs ("Best Estimate") 13-23
Table 13-9. Summary of Risks to Mammals ("Best Estimate") 13-24
Table 13-10. Results of Preliminary Search for Endangered
and Threatened Species Found in the Counties Where
Pulp and Paper Mills Are Located that Apply
Dioxin- and Furan-Contaminated Pulp and Paper Mill
Sludge to Land 13-26
Table 13-11. Distribution of Dioxin Concentrations in Whole
Fish Sampled in the National Bioaccumulation
Study 13-30
xx
-------
LIST OF FIGURES
Figure 2-1.
Figure 6-1.
Figure 6-2.
Figure 6-3.
Figure 6-4.
Figure 6-5.
Figure 6-6.
Paoe No,
Destruction Curves of 2,3,7,8-TCDD 2-14
Distribution of the number of mills for which
discharges would result in a given range of water
column contaminant concentrations as estimated by
the simple dilution method
Distribution of the number of mills for which
discharges would result in a given range of water
column contaminant concentrations as estimated by
the EXAMS II water column method
6-12
6-13
Distribution of the number of mills for which
discharges would result in a given range of lifetime
cancer risk due to the consumption of contaminated
fish tissue as estimated by the simple dilution
method (6.5 g/day consumption rate and BCF
of 5,000 for 2,3,7,8-TCDD)
Distribution of the number of mills for which
discharges would result in a given range of lifetime
cancer risk due to the consumption of contaminated
fish tissue as estimated by the simple dilution
method (30 and 140 g/day consumption rates and BCF
of 50,000 for 2,3,7,8-TCDD)
Distribution of the number of mills for which
discharges would result in a given range of lifetime
cancer risk due to the consumption of contaminated
fish tissue as estimated by the EXAMS II method
(6.5 g/day consumption rate and BCF of 5,000 for
2,3,7,8-TCDD) ;....
Distribution of the number of mills for which
discharges would result in a given range of lifetime
cancer risk due to the consumption of contaminated
fish tissues as estimated by the EXAMS II method
(30 and 140 g/day consumption rates and BCF of
50,000 for 2,3,7,8-TCDD)
6-16
6-17
6-18
6-19
xxi
-------
LIST OF FIGURES (continued)
Page No.
Figure 6-7. Distribution of the number of mills for which
discharges would result in a given range of lifetime
cancer risk due to the ingestion of contaminated
drinking water as estimated by the simple dilution
method 6-21
Figure 6-8. Distribution of the number of mills for which
discharges would result in a given range of lifetime
cancer risk due to the ingestion of contaminated
drinking water as estimated by the EXAMS II method . 6-22
Figure 6-9. Distribution of the number of mills for which
discharges would result in a given range of human
doses from a one-time exposure to contaminated fish
tissue as estimated by the simple dilution method .. 6-24
Figure 6-JO. Distribution of the number of mills for which
discharges would result in a given range of human
doses from a one-time exposure to contaminated fish
tissue as estimated by the EXAMS II method 6-25
xxii
-------
ACKNOWLEDGMENTS
This risk assessment was a cooperative Federal agency effort
involving the United States Environmental Protection Agency (EPA), the
Food and Drug Administration (FDA), and the Consumer Product Safety
Commission (CPSC). Dwain Winters of the EPA Office of Toxic Substances
(OTS) served as coordinator of the Interagency Working Group on Dioxins
in Bleached Wood Pulp. Dr. Robert Scheuplein and Dr. Melvin Stratmeyer
served as the FDA representatives to this Working Group; Sandra Eberle
and Dr. Andrew Ulsamer served as the CPSC representatives. The support
and management guidance provided by these individuals is gratefully
acknowledged.
The Interagency Working Group risk assessment activities were
coordinated by EPA-OTS. Lois Dicker served as Chair of the Interagency
Risk Assessment Workgroup and James Kwiat served as the Assistant Chair.
Versar Inc. of Springfield, Virginia, served as the prime contractor
through EPA Contract No. 68-D9-0166 (Tasks 1 and 34). The EPA-OTS Task
Manager for this effort was Patricia Jennings and the EPA Program Manager
was Thomas Murray.
A number of Versar Inc. personnel have contributed to this task over
the period of performance, as shown below:
Program Management Gayaneh Contos
Task Management Greg Schweer
Technical Support Bentley Gregg
Lee Schultz
Patricia Wood
Timothy Leighton
Carl D'Ruiz
Robert Fares
Geoffrey Huse
Clay Carpenter
James Konz
Daniel Arrenholz
Editing Martha Martin
Secretarial/Clerical Sally Gravely
Lynn Maxfield
Susan Perry
Kammi Johannsen
The contribution provided by each of these individuals is gratefully
acknowledged.
xxiii
-------
ACKNOWLEDGMENTS (continued)
A number of individuals on the Interagency Risk Assessment Workgroup
for Dioxin in Bleached Wood Pulp have served in key roles in completing
or managing efforts on specific sections of this Background Document to
the Integrated Risk Assessment. For each section of this Background
Document, the key individuals and the organization to which they belong
are identified below:
Section 1: Patricia Szarek, EPA Office of Toxic Substances
Section 2: Christina Cinalli, EPA Office of Toxic Substances
Section 3: Cheng-Chun Lee, EPA Office of Toxic Substances
Section 4: Nhan Nguyen, EPA Office of Toxic Substances
George Heath, EPA Office of Toxic Substances
Section 5: Patricia Jennings, EPA Office of Toxic Substances
Priscilla Halloran, EPA Office of Solid Waste
Section 6: Stephen Kroner, EPA Office of Water Regulations and Standards
Section 7: Michael Dusetzina, EPA Office of Air Quality Planning and
Standards
Section 8: Michael Babich, CPSC Directorate for Health Sciences
Section 9: Christina Cinalli, EPA Office of Toxic Substances
Donald Galloway, FDA Center for Devices and Radiological Health
Section 10: Gregory Cramer, FDA Center for Food Safety and Applied Nutrition
Michael Bolger, FDA Center for Food Safety and Applied Nutrition
Section 11: Gregory Cramer, FDA Center for Food Safety and Applied Nutrition
Michael Bolger, FDA Center for Food Safety and Applied Nutrition
Section 12: Gregory Cramer, FDA Center for Food Safety and Applied Nutrition
Michael Bolger, FDA Center for Food Safety and Applied Nutrition
Section 13: Robert Morcock, EPA Office of Toxic Substances
Appendix A: Michael Babich, CPSC Directorate for Health Sciences
Ivan Boyer, FDA Center for Food Safety and Applied Nutrition
Susan Griffin, EPA Office of Health and Environmental Assessment
Frederick DiCarlo, EPA Office of Toxic Substances
Anne Sergeant, EPA Office of Health and Environmental Assessment
Kim Hoang, EPA Office of Health and Environmental Assessment
xxiv
-------
ACKNOWLEDGMENTS (continued)
Appendix B: Christina Cinalli, EPA Office of Toxic Substances
Robert Morcock, EPA Office of Toxic Substances
The support and guidance provided by these individuals is greatly
appreciated.
The efforts on behalf of the Integrated Risk Assessment by the
Quantitative Risk Assessment Committee (QRAC) of FDA's Center for Food
Safety and Applied Nutrition are also acknowledged. The members of the
QRAC are listed below:
Sara Hale Henry (Executive Secretary) Linda R. Tollefson
Ronald J. Lorentzen (Co-Chair) Benjamin A. Jackson
Janet A. Springer (Co-Chair) Patricia S. Schwartz
Robert N. Brown Christine J. Lewis
Robert J. Scheuplein Paul M. Kuznesof
Many members of the Interagency Risk Assessment Workgroup on Dioxins
in Bleached Wood Pulp whose names have not been mentioned previously
provided helpful comments and suggestions on the Integrated Risk
Assessment. The efforts of these individuals, listed below, are greatly
appreciated:
Ernest Falke, EPA Office of Toxic Substances
Patrick Kennedy, EPA Office of Toxic Substances
Julie Lyddon, EPA Office of Toxic Substances
Maurice Zeeman, EPA Office of Toxic Substances
David Cleverly, EPA Office of Technology Transfer and Regulatory Support
Susan Norton, EPA Office of Health and Environmental Assessment
Jacqueline Moya, EPA Office of Health and Environmental Assessment
Tom Hale, EPA Office of Policy Analysis
Jennie Helms, EPA Office of Water Regulations and Standards
Harold Podall, EPA Office of Toxic Substances
Gary Grindstaff, EPA Office of Toxic Substances
Alexander McBride, EPA Office of Solid Waste
Wardner Penberthy, EPA Office of Toxic Substances
Alan Rubin, EPA Office of Water Regulations and Standards
William Rabert, EPA Office of Toxic Substances
Robert Lipnick, EPA Office of Toxic Substances
Murray Cohn, CPSC Directorate for Health Sciences
John Rigby, EPA Office of Toxic Substances
Paul White, EPA Office of Health and Environmental Assessment
Gary Foureman, EPA Office of Toxic Substances
Ann Clevenger, EPA Office of Toxic Substances
Jennifer Seed, EPA Office of Toxic Substances
Michael Adams, FDA Center for Food Safety and Applied Nutrition
Janet Remmers, EPA Office of Toxic Substances
Daljit Sawhney, EPA Office of Toxic Substances
xxv
-------
xx vi
-------
1. INTRODUCTION
1.1 Background
Various isomers of polychlorinated dibenzodioxins (PCDDs) and
polychlorinated dibenzofurans (PCDFs) have been found to be formed during
bleaching of wood pulp with chlorine or chlorine-based bleaching
chemicals. Quantitative studies conducted by the U.S. Environmental
Protection Agency (EPA) and the paper industry have shown that PCDDs and
PCDFs may be retained in low levels in bleached pulp, crude paper
products (e.g., newsprint, paperboard, fibers), and finished commercial
and consumer grade pulp/paper-based products. Furthermore, PCDDs and
PCDFs may also be retained in wastewater and sludge generated during the
manufacture of these products.
On July 27, 1988, the EPA entered into a Consent Agreement with the
Environmental Defense Fund and National Wildlife Federation concerning
the regulation of PCDDs and PCDFs formed in pulp and paper processing.
In response to the Consent Decree, EPA is conducting risk assessments for
PCDDs and PCDFs from pulp and paper mill products and wastes.' The Office-
of Toxic Substances (OTS) has the responsibility to fulfill the
commitments as required by the Consent Decree. The assessment is being
performed as an inter-Agency, inter-Office effort. Other Agencies
involved are the Food and Drug Administration (FDA) and the Consumer
Product Safety Commission (CPSC). Exposure and risk resulting from the
direct use of bleached pulp and paper in food, drugs, cosmetics, and
medical devices is being assessed by FDA. CPSC is evaluating exposure
and risk from body contact papers that do not fall into the medical
devices category and from other related consumer products. EPA is
responsible for the assessment of risks to humans and wildlife resulting
from the discharge of effluents from pulp and paper mills and the
disposal of wastewater sludge generated at pulp and paper mills. EPA is
also assessing risks to workers in the pulp and paper industry as well as
processors and commercial users of pulp and paper products.
Upon completion of a multiple exposure pathway risk assessment, EPA
must publish a proposed schedule for the regulation of dioxin
(2,3,7,8-TCDD) from bleached chemical pulp mills, or publish the Agency's
decision not to regulate. The Agency is under a legal consent agreement
to consider risks to humans from disposal of sludges and water effluents
from these mills as well as those associated with the use and disposal of
products made from bleached paper pulp. These risks are to be considered
from an occupational and non-occupational exposure perspective. Risks
posed to aquatic organisms and wildlife are also to be evaluated.
In an attempt to quantify concentrations of dioxins produced during
pulp and paper manufacturing operations, several studies were undertaken
cooperatively by the EPA and the paper industry. These studies include
the Five-Mill Study (USEPA 1988) and the 104-Mill Study (USEPA 1989).
1-1
1584q
-------
As a result of finding 2,3,7,8-tetrachlorodibenzo-p-dioxin
(2,3,7,8-TCDD) in native fish collected downstream from a number of pulp
and paper mills in the EPA National Dioxin Study (USEPA 1987a) and
subsequent findings of 2,3,7,8-TCDD in bleached kraft pulp and paper mill
wastewater sludges, the EPA planned a detailed process evaluation at one
mill. Through subsequent discussions with the paper industry, EPA and
the industry agreed in June 1986 to conduct a cooperative screening study
of five bleached kraft pulp and paper mills on a shared-resource basis.
Three mills were selected on the basis of known 2,3,7,8-TCDD levels in
sludges and two mills were volunteered by their parent companies to
attain the geographical diversity desired for the study. The selection
of the five mills, which represent about six percent of the bleached
kraft mills in the United States, was not intended to characterize the
entire industry. The principal objectives of the study, commonly
referred to as the "Five-Mill Study," were to:
• Determine, if present, the source or sources of 2,3,7,8-TCDD and
other polychlorinated dibenzo-p-dioxins (PCDDs) and
polychlorinated dibenzofurans (PCDFs) at five bleached kraft pulp
and paper mills; and
• Quantify the untreated wastewater discharge loadings, final
effluent discharge loadings, sludge concentrations, and wastewater
treatment system efficiency for 2,3,7,8-TCDD and other PCDDs and
PCDFs.
The Five-Mill Study was conducted during the period June 1986-January
1987 through the combined efforts of four EPA regional offices, five
state environmental control agencies, the National Council of the Paper
Industry for Air and Stream Improvement, Inc. (NCASI), and the
participating paper companies. Samples of wastewater, sludge, and pulp
were analyzed for PCDDs and PCDFs. The results uniformly showed that
2,3,7,8-TCDD and 2,3,7,8-TCDF were the principal PCDDs and PCDFs found.
This was particularly evident when the data were considered in light of
EPA's 2,3,7,8-TCDD toxicity equivalency factor (TEF) method of risk
assessment for complex mixtures of PCDDs and PCDFs. (These findings were
more recently confirmed in the 104-Mill Study—see below.)
The analysis further showed that 2,3,7,8-TCDD and 2,3,7,8-TCDF are
formed during the bleaching of kraft hardwood and softwood pulps with
chlorine and chlorine derivatives at the mills in the study. Table 1-1
presents concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in unbleached
and bleached Kraft pulps at the five mills. Although no 2,3,7,8-TCDD was
detected in unbleached pulp, three of the seven samples tested above
detection levels for 2,3,7,8-TCDF. The positive analysis for 2,3,7,8-
TCDF in unbleached pulp may be due to reuse of dioxin-contaminated paper
machine wastewater for brown stock pulping or dilution at the mill where
the samples were taken. Most of the bleached pulp samples tested
1-2
1584q
-------
8898H
Table 1-1. Concentration of 2,3.7,8-TCDD and 2.3.7,8-TCDF in Unbleached and
Bleached Kraft Pulps from the Five-Mill Study (Cooperative Oioxin
Screening Study) (ppt)
Pulp
2.3.7.8-TCDD 2.3.7.8-TCDF
No. of No. of
samples Range Mean Median samples Range Mean Median
Unbleached
Bleached
NO (0.3-1) NO NO 7
<0.6-51 13 5 9
<0.16-2.3 1.5 NO
<1.2-330 93 50
Source: USEPA (1988).
1-3
-------
positively for 2,3,7,8-TCDD and 2,3,7,8-TCDF, which led to the conclusion
that their formation was caused by the bleaching of Kraft pulp by
chlorination.
Table 1-2 presents concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
untreated filtrates from the various stages of the bleaching processes
used at the five mills. All of the samples tested above detection levels
for 2,3,7,8-TCDF. In five samples, 2,3,7,8-TCDD was not detected.
Table 1-2 shows that filtrate wastewater from the caustic extraction
stage generally contained the highest concentrations of 2,3,7,8-TCDD and
2,3,7,8-TCDF, followed by the hypochlorite, chlorination, and chlorine
dioxide stages. The bleach plant filtrate data do not clearly define the
point of dioxin formation but do indicate formation in the chlorination
(C) stage and possibly in the .extraction (E) stage. It is not possible
from the data to determine whether 2,3,7,8-TCDD and 2,3,7,8-TCDF are
formed in the C stage and extracted in the E stage or if there is
additional formation in the E stage. The data also suggest formation of
these compounds in subsequent bleaching stages.
The EPA/Paper Industry Cooperative Dioxin Study, commonly referred to
as the "104-Mill Study," was conducted from April 1988 to August 1989 at
104 domestic mills manufacturing chemical pulp. In this study,
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF were measured at 87 kraft
(sulfate) and 17 sulfite pulp mills that use chlorine-based bleaching
processes (chlorine, chlorine dioxide, or hypochlorite). Samples were
taken of the following:
• Bleached pulp after the final stage of bleaching,
• Wastewater sludge, and
• Treated wastewater effluent.
The study also collected data on waste treatment operations, waste
discharge characteristics, and sludge disposal information.
The 2,3,7,8-TCDD concentrations in sulfite pulp were much lower than
those for the kraft pulp concentrations, ranging from 1 ppt to 15 ppt
with a median of no detection corresponding to a detection level of 1
ppt. This observation confirmed the Five-Mill Study which showed the
highest level of dioxin concentrations to be associated with the kraft
process. The analytical results of the 104-Mill Study are presented in
Table 1-3. The results indicate that the concentrations of 2,3,7,8-TCDF
measured in the 104-Mill Study are greater than those of 2,3,7,8-TCDD,
almost by an order of magnitude. Also, the results of this study confirm
the finding of the Five-Mill Study that 2,3,7,8-TCDD and 2,3,7,8-TCDF are
the principal PCDDs and PCDFs present. In this study, pulp, effluent,
and sludge samples from 9 mills were analyzed for total non-2,3,7,8-
substituted isomers and for 18 specific 2,3,7,8-substituted isomers.
1-4
15B4q
-------
8898H
Table 1-2. Concentration of 2.3.7,8-TCDD and 2.3.7.8-TCDF in Filtrates from
Various Stages of Bleaching Process in the Five-Mill Study
(Cooperative Dioxin Screening Study) (ppt)
Pulp
C Stages
(C. CD.C/cl)a
E Stages
(E.ED)b
H Stages
(H,H/D)C
2.3.7.8-TCDD
No. of
samples Range Mean Median
8 0.02-0.24 0.07 0.05
8 0.045-3.6 1.1 0.26
10 0.025-1.9 0.40 0.19
2.3.7,8-TCDF
No. of
samples Range Mean
8 0.068-3.8 0.65
8 0.056-14 3.3
10 0.086-9.2 2.3
Median
0.24
0.51
0.59
D Stages (D)c
N0e 0.015 NAf
(0.003J-0.03
2 0.014-0.13 0.072 NA
f
aC Stage represents chlorination stage of the mills. C represents use of chlorine only.
Cp represents use of a mixture of chlorine dioxide and chlorine, which is predominantly
chlorine. C/0 represents use of a mixture of chlorine and chlorine dioxide. 2.3,7,8-TCDD
was not detected In one sample (Cp) with a detection level of 0.006 ppt.
E Stage represents caustic extraction stage following the bleaching stages at the mills. E
represents use of sodium hydroxide only. Er, represents use of a mixture of sodium
hydroxide and oxygen, which is predominantly sodium hydroxide. 2.3.7.8-TCDD was not detected
in two samples (both EO) with detection levels of 0.011 and 0.033 ppt.
CM Stage represents hypochlorfte stage of the mills. H represents use of sodium or calcium
hypochlorlte only. H/0 represents use of a mixture of sodium or calcium hypochlorite and
chlorine dioxide. 2.3.7,8-TCDD was not detected in one sample (H) with a detection limit of
0.017 ppt.
D Stage represents chlorine dioxide stage at the mills. D represents use of chlorine
dioxide only. 2,3.7,8-TCDD was not detected In one sample with a detection level of
0.003 ppt.
eND - Not detected.
fNA = Not applicable.
Source: USEPA (1988).
1-5
-------
8898H
Table 1-3. Summary Results of the 104 Mill Study3
Pulp
2.3.7.8-TCDD
Z.3.7.8-TCOF
Effluent
2.3,7,8-TCDD
2.3,7.8-TCDF
Sludge
2,3.7.8-TCDD
2,3,7.8-TCDF
Range
(ng/kg)
ND-116
ND-2.620
ND-0.640
ND-8.400
ND-3,800
2.4-17.100
Mean
(ng/kg)
8.8
94.9
0.068
1.033
77.5
749.6
Median
(ng/kg)
4.9
19.0
0.023
0.094
18.0
89.0
No. of mills
Std. Dev. with no detected
(ng/kg) values
11.8 21
283.7 6
0.106 20
2,358 7
163.9 2
2,079 0
Based on final results obtained from EPA's Office of Water Regulations
and Standards 1n October 1989.
The analytical objectives for detection limits of both compounds were
0.01 ng/kg (or ppt) for effluents and 1 ppt for pulps and sludges.
1-6
-------
Based on the 1987 TEF method (USEPA 1987b), the results show that
2,3,7,8-TCDD and 2,3,7,8-TCDF generally account for more than 90 percent
of the dioxin toxic equivalents (TEQ) found in pulp, sludge, and effluent
(Helms 1989).
In addition to the EPA studies, the National Council of the Paper
Industry for Air and Stream Improvement (NCASI) and others have performed
some analyses for 2,3,7,8-TCDD and 2,3,7,8-TCDF in composites of several
bleached pulp-based products. Table 1-4 summarizes the results of these
analyses.
1.2 Purpose and Scope
The purpose of this background document is to present an integrated,
multiple exposure pathway assessment of risks resulting from exposure to
PCDDs and PCDFs formed during the production of bleached wood pulp. The
assessment evaluates and estimates exposure and risk to humans, aquatic
organisms, and wildlife, and is based on existing data that have been
presented in reports generated for the EPA, CPSC, and FDA. The
individual assessments from which this background document was developed
encompass the following exposure and risk categories:
• An analysis of the chemistry and environmental fate of 2,3,7,8-
TCDD and 2,3,7,8-TCDF;
• An assessment of the hazard/toxicity of 2,3,7,8-TCDD and
2,3,7,8-TCDF to humans, aquatic organisms, and avian and
terrestrial wildlife;
• An assessment of exposures and risks to workers in the pulp and
paper industry;
• Assessments of exposures and risks to the general population
from:
- use and disposal of pulp and paper mill wastewater sludge and
land disposal of paper waste
- discharge of effluents from the pulp and paper industry
- incineration of pulp and paper mill wastewater sludge
- use of pulp-containing medical devices
- use of paper consumer products
- ingestion of foods packaged in or contacting bleached paper
products
1-7
1584q
-------
8898H
Table 1-4. Concentrations (ppt) of 2.3.7.8-TCDD and 2.3.7.8-TCDF in
Consumer Products
Product
2,3,7.8-TCOO 2,3,7,8-TCDF
Reference
Disposable diapers
Paper towels
Bond paper
Facial tissue
Scrap paper
Newsprint
Tissue
NO (2.6)
NO (2.1)
NO
3.7
13
13
1.1
0.6
ND
1.3
8.8
7.2
3.7
32
290
240
13
13
NO
31.1
NCASI (1987); Blosser (1987)
NCASI (1987); Blosser (1987)
WJLA TVa
NCASI (1987)
NCASI (1987)
NCASI (1987)
Beck et al. (1988)
Beck et al. (1988)
Beck et al. (1983)
LeBel et al. (1989)
a Assays for WJLA TV were performed by Triangle Laboratories, Research Triangle
Park. NC.
NO » Not detected; the number in parentheses is the detection limit.
1-8
-------
- use of food, drug, and cosmetic products containing cellulose
derivatives;
• An assessment of exposures and risks to avian and terrestrial
wildlife from land application of sludge and to aquatic organisms
from the discharge of pulp and paper mill effluents; and
• An analysis of information on the properties and toxicity of
other chlorinated chemicals (OCOs), those organic chemicals other
than PCDDs and PCDFs, that are present in pulp and paper mill
effluents, sludges, and pulps.
Each pertinent report(s) is cited as the source(s) for the chapters for
which they have been used. The reader should refer to the original
source documents for specific citations to the many information sources
used in each assessment.
As mentioned above, this assessment addressed the range of exposure
routes resulting from the production and use of bleached paper products.
In certain cases, many specific exposure scenarios were evaluated within
each exposure route. Table 1-5 provides a summary of each exposure
scenario addressed within each exposure route.
Although many scenarios using different methodologies were assessed
by the various agencies and offices participating in the Dioxin-in-Paper
Project, several common assumptions were agreed to by the participants.
1. The Toxicity Equivalency Factor (TEF) method states that
2,3,7,8-TCDF is assumed to have one-tenth the potency of
2,3,7,8-TCDD. EPA and FDA agreed to employ this policy for this
assessment. Because CPSC does not place similar emphasis on
risks calculated by the TEF method as it does for 2,3,7,8-TCDD,
it was agreed that, to the extent possible, CPSC risk estimates
for each scenario would be based on the contribution to risk of
2,3,7,8-TCDD alone.
2. The assessments focused on exposures and risks to 2,3,7,8-TCDD
and 2,3,7,8-TCDF. Based on the TEF values formally adopted by
EPA in 1987, the results of the "5-Mill Study" indicated and the
results of the "104-Mill Study" confirmed that these two dioxin
congeners generally account for more than 90 percent of the
dioxin toxic equivalents (TEQ) found in pulps, sludges, and
effluents from the pulp and paper mill samples analyzed.
3. EPA, FDA, and CPSC have each derived an estimated slope factor
(qj* or q^ for 2,3,7,8-TCDD based on linear-at-low-
dose extrapolation procedures. However, because the Agencies
differ with respect to selection of animal data and details of
1-9
1584q
-------
8898H
Table 1-5. Exposure Scenarios and Pathways Examined for Each Exposure/Risk Assessment Performed for the Dioxin-in-Paper Project
Assessment
Scenarios
Pathways/Products
Exposures and risks to workers in the pulp and
paper industry
i
t-»
o
Exposure to bleach operators, pulp testers, and
utility operators during manufacture
Exposure to operators during pulp drying
Exposure to wet-end operators during paper
manufacture
Exposure to dry-end and utility operators during
paper manufacture
Exposure to workers during paper converting
operations
Exposure to workers during nonwovens manufacture
Exposure to commercial users of paper
Exposure to waste treatment plant operators
and sludge haulers/front-end loader operators
during sludge handling/processing
Exposure to equipment operators during land-
filling and land applications operations
Exposure to equipment operators, compost
haulers, and screen operators during composting
operations
• Inhalation of volatilized TCDD and TCDF
• Dermal contact with pulp
• Inhalation of volatilized TCDO and TCDF
• Dermal contact with pulp
• Inhalation of volatilized TCDD and TCDF
• Dermal contact with paper
• Inhalation of volatilized TCDO and TCDF
• Inhalation of contaminated dust
• Dermal contact with paper
• Inhalation of contaminated dust
• Dermal contact with paper
• Inhalation of contaminated dust
• Dermal contact with paper
• Dermal contact with paper
• Inhalation of volatilized TCDD and TCDF
• Inhalation of contaminated participates
» Dermal contact with sludge
• Inhalation of volatilized TCDO and TCDF
• Inhalation of contaminated particulates
• Dermal contact with sludge
• Inhalation of volatilized TCDO and TCDF
• Inhalation of contaminated particulates
• Dermal contact with sludge
-------
Table 1-5. (Continued)
Assessment
Scenarios
Pathways/Products
Exposures and risks resulting from the use and
disposal of sludge front the pulp and paper industry
and land disposal of paper products
Exposures to individuals residing near or
utilizing surface or ground waters near
landfills in which pulp and paper mill sludge is
disposed
Inhalation of volatilized TCDO and TCDF
Ingest ion of drinking water from ground
water sources
Ingest ion of drinking water from runoff-
contaminated surface water sources
Ingestion of fish from runoff-contaminated
surface water
Exposures to individuals residing near or
utilizing ground water near municipal landfills
in which paper wastes are disposed
Exposures to individuals residing near or
utilizing surface or ground waters near
surface impoundments in which pulp and paper
mill sludge is disposed.
Exposures to individuals residing near or
utilizing surface or ground waters near sites
at which sludge is land-applied
Inhalation of volatilized TCDO and TCOF
Inhalation of drinking water from
contaminated ground water sources
Inhalation of volatilized TCOD and TCDF
Ingestion of drinking water from contaminated
ground water sources
Ingestion of drinking water from runoff-
contaminated surface water sources
Ingestion of fish from runoff-contaminated
surface water
Inhalation of volatilized TCDD And TCDF
Inhalation of contaminated particulates
Ingestion of drinking water from contaminated
ground water sources
Ingestion of drinking water from runoff -
contaminated surface water sources
Ingestion of fish from runoff-contaminated
surface water
Ingestion of contaminated soil
Ingestion of foods produced with contaminated
soi 1
Dermal contact with contaminated soil
-------
3898H
Table 1-5. (Continued)
Assessment
Scenarios
Pathways/Products
Exposures to individuals utilizing
commercially distributed pulp and paper mill
sludge as a soil amendment
Exposures and risks to the general population from
the discharge of pulp and paper mill effluents
Exposures and risks to aquatic organisms from the
discharge of pulp and paper mill effluents
Exposures and risks to the general population
from the incineration of pulp and paper mill
sludge
Exposures and risks to the general population from
the use of pulp-containing medical devices
Exposures to individuals utilizing surface
waters downstream from pulp and paper mill
effluents
Exposures to aquatic organisms directly down-
steam from pulp and paper mill effluent
discharge
Exposures to individuals residing near
incinerators in which pulp and paper mill
sludge is disposed
Exposures from direct contact with various
medical devices under FDA jurisdiction
Inhalation of volatilized TCOO and TCDF
Inhalation of contaminated particulates
Ingest ion of contaminated soil
Ingest ion of foods produced with contaminated
soil
Dermal contact with contaminated soil
Ingest ion of drinking water
Ingest ion of fish
• Exposure under 7Q10 flow conditions
Inhalation of TCDO and TCDF in incinerator
emissions
Dermal contact with:
- menstrual pads and tampons
- alcohol pads
- skin preparation wipes
- absorbable hemostatic agents
- surgical apparel
- adult diapers
- medical disposable bedding
- medical absorbent fiber
- absorbent-tipped applicators
- examination gowns
- opthalmic sponges
- hydroxypropymethyl cellulose
- cottonoid paddies
- electro-conductive media
- cutaneous electrodes
- isolation gowns
-------
Table 1-5. [Continued)
Assessment
Scenarios
Pathways/Products
i
i-*
CO
Exposures and risks to the general population from
the use of pulp and paper products
Exposures from direct contact with a variety
of consumer products under CPSC jurisdiction
Exposures and risks to the general population from
the use of food and drug products containing
cellulose derivatives and from the use of paper
food-contact articles
Expoures and risks to the general population from
the use of cosmetic products containing cellulose
derivatives
Exposures and risks to terrestrial and avian
wildlife from land application of sludge
Exposures from ingest ion of food products and
drugs that contain cellulose derivatives
Exposures from ingest ion of foods packaged in
or contacting bleached paper
Exposures from application of cosmetic products-
to the skin and use of dentifrice
Exposures and risks to birds, bird eggs, and
manuals
Dermal contact with:
- disposable infant diapers
- paper towels
- facial tissue
- toilet tissue
- conmunications paper
- paper napk ins
Ingest ion of foods and drugs containing:
- powdered cellulose
- microcrystalline cellulose
- metnyIcellulose
- carboxymethylcellulose
- other cellulose derivatives
Ingestion of foods in contact with the
following articles:
- dairy cartons
- juice cartons
- bakery cartons
- Dvariable board
- paper cups, plates, arid tra>s
- coffee fiIters
- tea bags
- microwave popcorn bags
- butter/margarine wraps
Derma! use of the following products:
- lotions
- shampoo
- wet wipes
Ingestion of dentrifice
Ingest ion of contaminated soil
Ingestion of contaminated prey items
-------
extrapolation, the risk estimates differ by as much as a factor
of 10. The Agencies agreed that this Integrated Assessment would
report cancer risk estimates calculated by each Agency.
4. The analytical results of the 104-Mill Study (i.e., 2,3,7,8-TCDD
and 2,3,7,8-TCDF concentrations in pulp, effluents, and sludge)
were to be used in all assessments unless use of alternate data
(e.g., product-specific concentrations) could be justified.
5. With the exception of risks calculated by EPA/OSW, EPA/OW, and
FDA/CFSAN, all estimated human cancer risks were calculated by
multiplying the estimated lifetime average daily doses (LADD) by
the slope factor (qj or qt*) for 2,3,7,8-TCDD and
dividing by the fraction of TCDD absorbed (A) during the animal
bioassay from which the slope factor was derived:
Risk = (LADD x
The value of A depends on the fraction of 2,3,7,8-TCDD absorbed
by the test animals during the bioassay used to estimate qt
or qj*. For the EPA and FDA slope factors, which are based on
a dietary bioassay, A is assumed to equal 0.55 since it was
estimated that 55 percent of the 2,3,7,8-TCDD was absorbed by the
test animals; similarly, for the CPSC slope factor, which is
based on a gavage bioassay, A is assumed to equal 0.75. The
total or population risk was estimated by multiplying the average
lifetime risk by the number of persons exposed and dividing by
the average life expectancy.
6. With the exception of the bioavailablity values used by EPA/OSW
for each ingest ion pathway applicable to sludge disposal and
reuse, standard values developed for the bioavailability, or
fraction of 2,3,7,8-TCDD and 2,3,7,8-TCDF absorbed, were
developed and used for each exposure route and pathway. For the
inhalation exposure route, the bioavailability was assumed to be
100 percent for 2,3,7,8-TCDD and 2,3,7,8-TCDF vapors and 100
percent for particulate-bound 2,3,7,8-TCDD and 2,3,7,8-TCDF that
reach the alveoli. Standard values for the bioavailability of
2,3,7,8-TCDD and 2,3,7,8-TCDF were assumed to be 100 percent from
ingestion of drinking water, 85 to 95 percent for ingestion of
fatty or oily foods (e.g., milk, fish, meats), 60 to 70 percent
for ingestion of paper dust and sludge, and 45 to 55 percent for
ingestion of soil. EPA/OSW, however, assumed 100 percent
absorption for each exposure pathway applicable to sludge
disposal and use except for dermal exposures. For dermal
exposures, a dermal transfer coefficient of 0.012/hour is assumed
for 2,3,7,8-TCDD and 2,3,7,8-TCDF that are not bound up within a
matrix (e.g., soil or paper products).
1-14
1584q
-------
A variety of terms were used in the source documents for the
Integrated Assessment to describe the exposure case or exposed individual
for which risks were estimated. Many of these terms are presented in the
Background Document to the Integrated Assessment and in Section 2 of this
report as they appear in the source document. These exposure case
descriptions include: low, high, average, typical, reasonable worst
case, extreme worst case, and maximum exposed individual (MEI). Such
exposure case descriptions are used by exposure/risk assessors to
describe where in the range of exposures possible for a given scenario,
they either know (from a statistical array of exposures or exposure
parameter values) or judge the calculated exposure to reside.
Historically there has been no generally accepted convention or guidance
specifying what defines a given exposure description. Therefore, the use
of these terms has not been consistent within this report.
In order to provide some consistency in summarizing the results of
the Integrated Assessment, the various exposure case descriptions used in
the source documents have been collapsed in this section into the
following two classifications: typical and reasonable worst case. The
majority of exposure cases developed in the source documents fit into one-
of these two categories if the following somewhat broad definitions are
used:
Typical exposure - exposure parameter values selected are:
(a) values conventionally used for certain exposure parameters; or
(b) average or most probable values when distribution data for the
parameter are available; or (c) values considered "typical" or
frequently observed based on best professional judgment.
Examples of the types of exposures mentioned in this report that
would be characterized as typical include inhalation of volatilized
2,3,7,8-TCDD/TCDF; inhalation of particulate matter (i.e., paper dust
or sludge) containing 2,3,7,8-TCDD/TCDF; dermal contact with pulp,
paper, or sludge containing 2,3,7,8-TCDD/TCDF; and ingestion of food,
water, and drugs containing 2,3,7,8-TCDD/TCDF.
Reasonable worst case exposure - similar to typical exposure with the
exception that values for one or more significant exposure parameters
are selected within the upper portion of the range of actual or
expected values so that the resulting exposure calculated represents
a relatively high but possible exposure.
Examples of the types of exposures mentioned in this report that
would be characgerized as reasonable worst-case include those cases
in which an individual is exposed to the highest possible
concentration (i.e., 90th percentile and above); or instances where
an individual is exposed at a frequency or duration higher than what
is typically observed.
1-15
1584q
-------
1.3 Industry Profile
This section provides a brief overview of the pulp and paper industry
and its products, the pulping and bleaching processes, and potential
sources of dioxins and furans in pulp and paper. This section is
compiled, in large part, from:
Clement Associates, Inc. 1989. Oioxin production in the pulp/paper
industry. Revised draft report. U.S. Environmental Protection
Agency, Office of Toxic Substances. EPA Contract No. 68-08-0116.
September 29, 1989.
1.3.1 Overview of the Industry
There are 391 pulp and paper mill companies in the United States with
802 establishments. An establishment is defined as the location of
manufacturing or non-manufacturing activities. An establishment can be a
single pulp mill, a single paper mill, an integrated mill which has one
or more pulp mills as well as one or more paper mills; or a central
headquarters and/or a Research and Development facility when these are
located separately from the manufacturing activity. There are 605 mills
classified as paper mills (this is most likely the majority of pulp
buyers; it is unknown whether textile and pharmaceutical companies that
buy pulp are included in this figure) and 358 mills (some may be
inactive) that produce pulp which is used in 605 paper mills.
The 358 mills fall into two categories—chemical and mechanical.
These categories differ in the method used to produce pulp from the
wood. Currently in the industry, 49.7 percent of the pulp mills use
chemical pulping methods; 30.4 percent use mechanical methods; and the
remaining 19.8 percent involve miscellaneous methods (these include
deinking, rag, soda, rope, flax, bagasse, and cotton 1 inters pulp
mills). There are three types of chemical pulping methods within the
industry—kraft (sulphate), sulphite, and semi-chemical. These methods
differ in the chemicals, wood types, and techniques used, as well as in
the end products produced. In the United States, 104 chemical pulping
mills using chlorine in the bleaching process were identified by the EPA
Office of Water. Chlorine has been identified as a necessary precursor
to dioxin formation. Some of these mills are involved in more than one
pulping process, but the dominant process is chemical pulping.
1.3.2 Processes/Operation
(1) Pulping process. The (wood) raw material that enters the
pulping process is composed of cellulose fibers, lignin, semicellulose,
and other compounds. Lignin is a complex polymer that is believed to
contain dioxin precursors. It is responsible for cementing and
strengthening the wood fibers, and for the discoloration of wood. In the
1-16
1584q
-------
production of paper, some lignin is removed to allow flexibility of the
fibers. The amount of lignin remaining determines the length of time the
paper remains flexible. The remaining portion, about 10 percent in the
kraft pulping process, is bleached to produce a desired whiteness and to
deter the discoloration produced with age. The amount of lignin
destroyed/bleached is specific to the grade of paper product, and ranges
from paper sacks and newsprint to writing paper.
The pulping process utilizes mechanical and chemical techniques to
produce the desired pulp type. The mechanical process is used primarily
with softwoods that are easier to tear and grind, and the resultant pulp
is generally used in the manufacture of newsprint, catalogues, and
toweli ng.
Chemical pulping removes the lignin to enhance fibers flexibility,
resulting in a stronger paper product but lower fiber yields (40 to
50 percent). The three chemical pulping categories in increasing order
of use are sulphite, semi-chemical, and kraft (sulphate). Sulphite pulp,
often blended with mechanical pulps as a strengthener, is used for the
production of viscose rayon, acetate fibers and films, plastic fillers,
and cellophanes. Semi-chemical pulping is generally used for newsprint,
containers, and computer cards.
The kraft pulping dominates the U.S. pulp and paper industry,
accounting for about 75 percent of the pulp produced for paper and paper-
board (OTA 1989). The advantages of this process are (1) the versatility
in the types of wood it can process and (2) the released extracts (e.g.,
turpentine, tall oil, and resin) that can be separated for sale as
commodity chemicals. In addition, process chemicals are recovered and
recycled. Within the chemical pulping process, kraft pulping is used at
over 70 percent of U.S. mills.
(2) Bleaching process. After the wood goes through the pulping
process, the degree of bleaching it receives is dependent upon the
whiteness that is desired. The bleaching process for chemical pulp
proceeds in two phases—del ignification and brightening. They are
carried out in alternating acid and alkaline stages. The conventional
bleaching sequence involves five-stages, and chlorine is used as the
dominant chemical. The length of bleaching time and the chemical used
depend on the type/condition of the pulp and the desired characteristics
of the final products. Generally, sodium hydroxide is used 1n the
alkaline extraction stage. More recently, oxygen and peroxide have been
added to improve the extraction efficiency.
There are a wide variety of sequences used to ensure a high level of
delignification and whiteness. In 1989, chlorine was used as the initial
bleaching chemical by the vast majority of U.S. chemical pulping mills
(USEPA 1989). It 1s the most effective chemical for the delignification
1-17
1584q
-------
and produces a high quality range of white paper products. Chlorine
removes a number of contaminants from the pulp, and makes a strong board
without attacking the cellulose.
1.3.3 Potential Source of Dioxin/Furans in Pulp and Paper
The chemical reactions and conditions under which dioxin is formed in
pulp and papermaking operations are not yet completely understood.
Dioxin is formed as a byproduct during these manufacturing operations and
is considered a contaminant. Chlorination of a dioxin precursors in the
pulp during bleaching stages may result in the production of 2,3,7,8-TCDD
and 2,3,7,8-TCDF and other isomers.
Possible sources of dioxin precursors include natural constituents in
wood, contaminants in plant pipes and machinery, and additives. Dioxin
formation may result from condensation of chlorophenols formed by
chlorination of naturally occurring phenolic compounds such as lignin,
which comprises 25 percent of wood, and other plant constituents. To
date, the only substances found to contain precursors that have been
studied in depth are the oil-based defoamers.
A recent Canadian study has identified potential dioxin precursors in
oil-based defoamers. The oil-based defoamers are additives that improve
unbleached pulp washing properties. The study positively identified the
presence of one dioxin precursor, dibenzofuran (DBF), in oil-based
defoamers. DBF is a contaminant in the oil-based defoamers, and the
study revealed the potential for DBF contamination in tap water used for
process makeup water, air, and raw wood. The presence of another
precursor, dibenzo-p-dioxin (DBP), was suspected in oil-based defoamers;
however, it could not be confirmed because of excessive interference from
other substances during analysis. Tests revealed that the addition of
oil-based defoamers resulted in elevated levels of 2,3,7,8-TCDD and
2,3,7,8-TCDF in the final chlorinated pulp. Table 1-6 presents data on
the effect of defoamer addition to laboratory-prepared brown stock Kraft
pulp on subsequent dioxin formation during the bleaching process. The
data were collected from studies using laboratory-prepared western
hemlock Kraft pulp. Oil-based defoamer was added at one percent the
weight of the pulp, which is 10 times greater than normal industry
practice. This was done so that oil-based defoamers could be Identified
as being the source of the precursors.
1.3.4 End-Uses of Products from Pulp and Paper
The pulp and paper industry is comprised of those establishments that
(1) produce pulp for market only; (2) produce pulp for in-house only; (3)
buy pulp for use in paper product manufacture; and (4) produce pulp for
in-house use in their paper mill(s) and for sale in the market. A pulp
mill is usually a separate facility from the paper mill, even at combined
pulp and paper manufacturing plants.
1-18
1584q
-------
8898H
Table 1-6. Effect of Defoamer Addition to Brownstock Kraft Pulp on Subsequent
Dioxin Formation During the Bleaching Process3
2,3.7,8-TCDD 2.3.7.8-TCDF
concentration, concentration.
Pulp additive ppt ppt
None 11 160
IX- of oil-based defoamer A (virgin oil base) 110 910
1% of oil-based defoamer B from Canadian kraft mill 81 280
Vi- of oil-based defoamer C (recycled oil base) 140 1.200
1% of recycled oil base used in defoamer C 170 1,400
3Voss (1988).
1-19
-------
The majority of bleached pulp produced in the U.S. is converted to
paper and paperboard products such as printing/writing paper, tissue
paper, packaging paper and paperboard, and non-packaging paperboard. The
category, tissue, usually includes commonly used household items such as
disposable diapers, facial and toilet tissue, wadding, coffee filters,
napkins, toweling, and hygiene pads. Food-related paper products include
food wrapping and bags. Food-related packaging paperboard includes milk
carton stock and ovenable board, and non-packaging paperboard includes
cup and plate stock.
1.4 Organization of This Report
This report is organized into 12 sections and appendices. Following
is a brief description of each section:
• Section 2 provides an analysis of the chemistry and fate of
dioxins and furans.
• Section 3 describes the human, terrestrial and avian wildlife,
and aquatic life hazards of 2,3,7,8-TCDD and 2,3,7,8-TCDF.
• Section 4 provides an assessment of exposures and risks to
workers in the pulp and paper industry.
• Section 5 provides an assessment of exposures and risks to
humans from use and disposal of pulp and paper mill sludge and
land disposal of paper waste. Exposures were evaluated Based on
sludge disposal in landfills, surface impoundments, land
application/1andfarming, and from marketing the sludge.
• Section 6 provides an assessment of exposures and risks to
humans associated with the discharge of effluents from bleached
kraft and sulfite pulp and paper mills.
• Section 7 provides an assessment of exposures and risks to
humans resulting from incineration of pulp and paper mill
wastewater sludges.
• Section 8 provides an assessment of exposures and risks to
consumers from use from pulp and paper products.
• Section 9 provides an assessment of exposures and risks to the
general population from dermal contact with pulp-containing
medical devices.
• Section 10 provides an assessment of exposures and risks to the
general population from use of food packaged in or containing
bleached paper products.
1-20
1584q
-------
• Section 11 provides an assessment of exposures and risks to the
general population from the use of food, drug, and cosmetic
formulations containing cellulose derivatives.
• Section 12 presents FDA's assessment of human exposures and
risks resulting from ingestion of fish from pulp and paper mill
receiving waters.
• Section 13 provides an assessment of risks to terrestrial and
avian wildlife resulting from land application of sludges and an
assessment of risks to aquatic life resulting from discharge of
pulp and paper mill effluents.
Appendix A provides PCDD and PCDF bioavailability information.
Appendix B describes the environmental effects of other chlorinated
organics.
Each exposure section evaluates an exposure pathway, identifies and
enumerates exposed populations, estimates risk associated with that
exposure and presents an analysis of uncertainty in the assessment.
1.5 References
Beck H, Eckart K, Mathar W, Wittkowski R. 1988. Occurrence of PCDD and
PCDF in different kinds of paper. Chemosphere, 17: 51-57.
Blosser RO. 1987. Communication to U.S. Consumer Product Safety
Commission (CPSC). September 25, 1987.
Helms J. 1989. U.S. EPA/Paper Industry cooperative dioxin study field
congener analyses. Memorandum from J. Helms (USEPA/PWRS) to G. Schweer
(USEPA/OTS). August 11, 1989.
LeBel GL, Williams DT, Benoit FM. 1989. Determination of chlorinated
dibenzodioxins and dibenzofurans in selected paper products. Ninth
international symposium on chlorinated dioxins and related compounds,
Toronto Ontario, September 17-22, 1989. Abstract PLP23.
NCASI. 1987. National Council of the Paper Industry for Air and Stream
Improvement. Assessment of potential health risks from dermal exposure
to dioxin in paper products. Technical Bulletin No. 534. November 1987.
OTA. 1989. Office of Technology Assessment. Technologies for reducing
dioxin in the manufacture of bleached wood pulp. OTA-BP-0-54. May 1989.
USEPA. 1987a. U.S. Environmental Protection Agency. National dioxin
study. Washington, D.C.: U.S. Environmental Protection Agency.
EPA 530/SW-87-025.
1-21
1584q
-------
USEPA. 1987b. U.S. Environmental Protection Agency. Interim procedures
for estimating risks associated with exposures for mixtures of
chlorinated dibenzo-p-dioxins and -dibenzofurans (CDDs and CDFs). Risk
Assessment Forum. EPA 625/3-87-012. National Technical Information
Service, Springfield, VA. PB89-125041.
USEPA. 1988. U.S. Environmental Protection Agency. U.S. EPA/Paper
Industry cooperative dioxin screening study. Washington, D.C.: Office
of Water Regulations and Standards. EPA 440/1-88-025.
USEPA. 1989. U.S. Environmental Protection Agency. U.S. EPA/Paper
Industry cooperative dioxin study. Data submittals during 1988 and 1989
to EPA, Office of Water Regulations and Standards.
Voss RH, et al. 1988. Some new insights into the origins of dioxins
formed during chemical pulp bleaching. Pulp and Paper Canada, 89(12):
151-161.
1-22
1584q
-------
2.
CHEMISTRY AND FATE OF DIOXINS AND FURANS
This section reviews the available information regarding the chemistry
and fate of the general group of compounds termed polychlorinated dibenzo-
p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs), as well as
the specific compounds 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD)
and 2,3,7,8-tetrachlorodibenzofuran (2,3,7,8-TCDF). This section is
compiled from:
Versar Inc. 1989. Chemistry and fate of dioxins and furans.
Washington, DC: U.S. Environmental Protection Agency, Office of
Toxic Substances. Contract No. 68-02-4254, Task No. 231.
2.1
Chemical Identity of PCDDs and PCDFs
Very little published information is available on the physical and
chemical properties of most PCDDs and PCDFs, other than recently reported
values for some congeners for such important properties as water
solubility and vapor pressure. In general, structure-activity principles
hold with higher chlorinated congeners being less soluble and showing
lower vapor pressures, thus being less volatile.
PCDDs and PCDFs are tricyclic aromatic compounds having three-ring
structures consisting of two benzene rings connected by a third ring.
For PCDDs, the connection of the benzene rings is through a pair of
oxygen atoms as opposed to one oxygen atom for the PCDFs. In general,
both compound classes have similar physical, chemical, biological, and
toxicological properties (Rappe et al. 1987; EPRI 1983). The basic
structure and numbering of each chemical class is shown below.
PotychlortMMd dlb«nzo-p-dloiln« (PCDDt)
PolychlorlMUd diunwlufint (PCOFs)
There are 75 possible different positional isomers of PCDDs and 135
different isomers of PCDFs (see Table 2-1). The literature reviewed
reports that neither chemical class reacts with weak acids and bases, nor
with most redox agents, and that each is only slightly soluble in water
and in many organic solvents. The chemical stability increases with
increasing halogen content (EPRI 1983).
2-1
1583q
-------
8992H
Table 2-1. Possible Number of Positional PCDO and PCOF Isaners
Number of isaners
Chlorine
substitution PCODs PCDFs
Mono 2 4
Di 10 16
Tri 14 28
Tetra 22 38
Penta 14 28
Hexa 10 16
Hepta 2 4
Octa 1 1
2-2
-------
According to EPRI (1983) the chemistry of these compounds also
suggests that PCDFs can form directly from PCBs but the PCDDs cannot.
However, both compound classes can be formed from chlorophenols or
chlorinated benzenes. In addition, Svenson et al. (1989) reported the
formation of PCDDs and PCDFs in an enzyme-catalyzed oxidation of
2,4,5-trichlorophenol. Treatment of the chlorophenol with hydrogen
peroxide/peroxidase resulted in formation of small amounts of
2,3,7,8-tetra-substituted congeners as well as other derivatives.
Little, if any, research or analyses have been performed to determine
physical and chemical properties of TCDFs and TCDDs, with the exception
of the 2,3,7,8-TCDD isomer. Few of the possible isomers are available
commercially and preparation and synthesis can be both time consuming and
difficult. In addition, these isomers prepared in pure form may not be
available in sufficient quantities for testing, and the high toxicity of
these compounds necessitates extreme precautions to prevent adverse
effects in the workers in the toxicology laboratories (EPRI 1983).
Due to the lack of experimental data, the physical/chemical properties
of the various PCDF and PCDD congeners were estimated (Table 2-2). Esti-'
mates were derived from the limited experimental data for 2,3,7,8-TCDD
and via the methods described below. Water solubilities, soil adsorption
coefficients (Koc), and bioconcentration factors (BCF) were estimated
from the equations in Lyman et al. (1982). Vapor pressures were estimated
via the Meissner method as described in Lyman et al. (1982). The log
Kow values were estimated by estimating the value of a fragment constant
representing the substitution of a chlorine atom for a hydrogen atom of a
PCB. The log Kow values for PCDDs were estimated by starting with the
reported log Kow value for 2,3,7,8-TCDD, and for PCDFs by starting with
the reported value for dibenzofuran, 4.10 (Leo and Hansch 1979).
The measured bioconcentration factor (BCF) values for PCDDs and PCDFs
are much lower than the predicted BCF values using SAR Analysis and the
calculated octanol/water partition coefficient (Kow). The differences
between predicted versus measured values have been attributed to
metabolism. PCDDs were bioconcentrated to greater extent than PCDFs, as
predicted by SAR Analysis. There was also a relationship between BCF and
exposure concentration for PCDDs and PCDFs, whether exposures were from
water or diet. As exposure concentration increased, the BCF decreased.
2-3
1583q
-------
8992H
Table 2-2. Summary of Estimated Physical/Chenical Properties of PCOFs and PCDOs
Congener
PCOFs
Tetrachloro
Pentachloro
Hexachloro
Heptachloro
Octachloro
PCDOs
Tetrachloro
Pentachloro
Hexachloro
Heptachloro
Octachloro
Average M^
306
340
375
409
444
322
356
391
425
460
Physical
State3
(«25*C)
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Boiling
Point
CC)
340
357
373
389
405
341
358
374
390
406
Melting
Point
CC)
306
320
335
349
363
306b
321
336
350
364
LogKow
5.78
6.20
6.62
7.04
7.46
6.15b
6.57
6.99
7.41
7.83
Sol. in
Water
(•9/1)
5.4 x 10"5
1.2 x 10~5
2.7 x 10"6
6.2 x 10"7
1.4 x 10~7
2.0 x 10"4
4.8 x 10~6
1.1 x 10"6
2.4 x 10"7
5.4 x 10"8
1
3.3
5.6
9.5
1.6
2.7
5.3
8.9
1.5
2.6
4.3
x 104
xlO4
x 104
xlO5
xlO5
x 104
xlO4
xlO5
xlO5
xlO5
Vapor
Pressure
(mn-Hg 9 25'C)
1.1 x 10"B
2.4 x 10~7
5.6 x 10~8
1.3 x 10~8
3.2 x 10~9
1.7 x lO"6*1
2.1 x 10~7
5.5 x 10"8
1.1 x 10"8
2.8 x 10"9
Henry's
Constant
8.2 x 10"3
8.9 x 10"3
1.1 x 10"2
1.1 x 10~2
1.3 x 10 2
3.6 x 10"3
2.0 x 10"2
2.6 x 10"2
2.6 x 10"2
2.7 x 10"2
Bioconcentrat ion
Factor
1.5 x 104
3.0 x 104
6.3 x 104
1.3 x 105
2.8 x 105
2.8 x 104
5.8 x 104
1.2 x 105
2.5 x 105
5.2 x 105
a AssuKd to be a solid at room to^erature (parent confound is a solid).
for other PCODs and PCOFs
Measured values for 2.3.7.8-TCOO. Log
ne su
listed were estimated using «ethods from Lywan et al. (1982).
ere estimated from measured values of log k for 2.3.7.8-TCDD or 2.3.7.8-TCDF
and a fragment constant for chlorine substitution on PCBs estinated from Measured log k(jw values for 27 PCB isomers.
Values for all other properties
-------
2.2 Chemistry and Fate of 2.3.7.8-TCDD
2.2.1 Chemical Identity
2,3,7,8-Tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) is a member of the
75 compounds known as polychlorinated dibenzo-p-dioxins (PCDDs). The
structure of 2,3,7,8-TCDD is:
2.2.2 Chemical and Physical Properties
The measured properties of 2,3,7,8-TCDD obtained from the literature
are listed in Table 2-3. Estimated values might be reported if no
measured values are available for an important physical or chemical
property. In general, 2,3,7,8-TCDD is very sparingly soluble in water,
has a high octanol-water partition coefficient and, therefore, sorbs to
organic matter, and has a low vapor pressure but will volatilize into the
air under favorable conditions.
2.2.3 Environmental Fate and Transport
2,3,7,8-TCDD is persistent in soils. Upon deposition of 2,3,7,8-TCDD
onto surfaces, there is a high initial loss due to photodegradation and
perhaps volatilization. Once 2,3,7,8-TCDD moves into soils or sediments,
however, it is apparently strongly sorbed unless there are chlorinated
organic co-contaminants present. Some recent studies, however, have
shown that there may be slow rates of vapor phase transport out of soils,
although other recent studies have shown very low mobility.
The only environmentally significant path for destruction of 2,3,7,8-
TCDD appears to be photodechlorination. This process, however, requires
the presence of another organic material to donate hydrogen atoms. Obser-
vations on bioaccumulation indicate that 2,3,7,8-TCDD is readily biocon-
centrated in fish, but the data for humans are inconclusive, and the
dioxins in soil and sediments are considered to be essentially nonbiode-
gradable. Erosion and aquatic transport of sediment appear to be the
main transport mechanism of sorbed dioxins. Table 2-4 is a summary of
the environmental fate of chlorinated dibenzo-p-dioxins.
2-5
1583q
-------
8992H
Table 2-3. Physical and Chemical Properties of 2.3.7.8-TCDO
Property
Value
Conner) ts
References
Analytical Method
ro
Chemical name
Synonym
Empirical formula
CAS number
Molecular weight
Pure Fare
Melting Point
Decomposition temperature
Critical temperature
Critical pressure
Critical volume
Water solubility
2 . 3 . 7 . 8-Tet rach lorod i benzo-
p-dioxin
TCDD; 2.3.7.8 TCDO
1746-01-6
321.96
White crystalline solid
302-305'C
>700'C
66PC
2.372.829 kPa
763 c»3/g mole
19.3 i 3.7* opt (ng/L)
7.91 i 2.7 ppt (ng/L)
200 ppt
12.9 ± 1.2 ppt
483 *. 94 ppt
25'C
22'C
25'C
25-C
4.3 t 0.2'C
17.3 i O.PC
USEPA (198/b)
USEPA (1987b)
Reid et al. (1977)
Reid et al. (1977)
Reid et al. (1977)
Marple et al.
(1986a)
Adams and Blaine
(1986)
Kearney, et al.
(19/3)
Lodge (1989)
Lodge (1989)
Thin film equilibrium. 6C/LRHS
with Hi electron capture
detector.
3H-labelled TCDO. scintillation
detector.
Generator column followed by HPLC
Generator colimn followed by HPLC
-------
Table 2-3. (continued)
Property
Cownents
References
Analytical Method
Solubility in Organic
Solvents at 25*C
o-dichlorobenzene
chlorobenzene
anisole
xylene
benzene
chloroforM
n-octanol
•etnano1
acetone
dioxane
Solubility (wg/L)
1400
720
1730
3580
570
370
48
10
110
380
Eposito et al. (1980)
Log Octanol-water partition 6.64* (Average)
coefficient. (Log K^ 6.54 - 6.95 (Range)
or Log P)
7.02 * 0.5
6.15 to 7.28 (Range)
Apx. 8.5
(7.4 ± 0.04) x 10~10*
Marple et al. (1986b) Diffusion. GC/LRHS
Vapor pressure (•nHg)
(at 25*C. unless
specified)
30. PC
30-C
50*C
(3.49 + 0.05) x 10
1.5 x 10"9
3.4 x 10"9
7.2 x 10"8
9
Values extrapolated
Reverse-phase HPLC and LRNS
Burknard and Kuehl
(1986)
USEPA (1984)
Sarna et al. (1984)
Podoll et al. (1986) 14C-labelled 2.3.4.8 TCDO and
saturation technique with combustion
to 14C02.
Schroy et al. (1985) Gas saturation with GC/HS
Schroy et al. (1985)
gas
* Reconnended values at stated temperature.
-------
8992H
Table 2-3. (continued)
Property
Value
Contents
References
Analytical Method
Vapor pressure (continued)
i
00
305'C
421-C
Henry's Lav Constant
49.4
760
1.5 x 10
,-9
4.65 x 10
1.6 x 10"5 a
2.12 x 10
Volatilization fro* water 32 days
16 days
-6
'/•»!*
Measured
Predicted
Rordorf (1387)
Rordorf (1987)
Calculated from the water Podoll et al. (1986)
solubility and vapor pressure
listed above.
Schroy et al. (1985)
Half life for ponds and lakes Podoll et al. (1S86)
Half life for rivers Podoll et al. (1986)
Soil/sedi*ent adsorption
coefficient 7.15-7.34
6.S5
7.39
„ 6.6+ 0.7
Bioconcentration factor
in fish:
Exposure fro* water 20.600
Experimental
Predicted
Calculated
Jackson et al. (1986)
Walters and Guiseppi-
Elie (1988)
Nabholz (1989)
Geometric Mean for all fresh
Hater fish. N=7. Most Reliable
in concentration range of
0.03B to 107 ppt.
-------
Table 2-3. (continued)
Property
Value
Conments
References
Analytical Method
Bioconcentration factor
in fish (continued)
Exposure from Mater
(cant inued)
24.100
ro
i
10
7.900
Exposure fro* dietary 0.09697
sources
0.1119
0.08016
Geometric Means for cold water
species. N=6. Most Reliable
in concentration range of
0.038 to 107 ppt.
Kara water species. N=l, con-
centration was 0.87 ppt.
Geometric wean for all fresh
water fish. H=7. Host Reliable
in concentration range of 39
to 2000 ppt.
Geometric Mean for cold water
species. 11=4. at a concentra-
tion of 494 ppt.
Geometric Mean for warn water
species, N-3. Host Reliable
in concentration range 39
to 2000 ppt.
Bioconcentration factors
in humans:
Geyer et al. (1987)
Weight weight basis
104 to 206 (155 average)
* Reccomended values at stated temperature.
Vet weight basis. Calculated
assuming estimated daily
intakes of 0.0325-0.065'ng of
TCDD/day and measured concen-
trations of TCOO in adipose
tissue of humans.
-------
8992H
Table 2-3. (continued)
Property
Value
Commits
References
Analytical Method
I
!-•
O
Bioconcentratitti factors
in humans (continued)
Lipid basis
Wet weight basis
115-229 (172 average)
153
Lipid basis
170
Calculated on a lipid basis. Geyer et al. (1987)
FrcM Measured half-life and
Measured concentrations in
body fat at steady state
using a linear, one-com-
partment phamacokinetic
Model.
From Measured haIf-life and
Measured concentrations in
body fat at steady state
using a linear. one-com-
partment phamacokinetic
Model.
Bioconcentration factors
in animals:
Rats
Beef cattle
BCF,
Liver
BCF,
Fat
24.5
24.3
10.9
0.7
24.5
8.1
3.7
3.5
24.8
Duration of
Feeding
2 years
2 years
2 years
28 days
499 days
Concentration
in diet
(nq/kq)
22
210
2200
24
24
Geyer et al. (1987)
Rhesus
24 - 40
4 years
-------
Table 2-3. (continued)
Property
Value
Contents
References
Analytical Method
Bioconcentration factor:
Vegetation 0.013
Milk 0.010
Beef O.OS
Calculated
Calculated
Calculated
Travis and Anns (1988)
Travis and Anns (1988)
Travis and Anns (1988)
ro
RecoMended values at stated temperature.
-------
8992H
Table 2-4. Suimary of Environmental Fate of Dibenzo-p-dioxins
Environmental
process
Suwnary
statement
Confidence
in data
Photolysis
Oxidation
Hydrolysis
Volatilization
Sorpt ion
Bioaccunulation
Biodegradation
May be only natural mechanism
leading to destruction of
dioxins.
Dioxins are stable to oxidation.
Dioxins are stable to
hydrolysis.
Possible important mechanism
for transport from water.
Volatility depressed by
presence of organic solids.
Dioxins strongly sorted by
solids, especially with high
organic content.
Available data indicates process
may be important. The data
show high degree of confidence
for bioconcentration in fish.
but low confidence in the limited
data concerning bioaccumulation
in humans.
Considered essentially non-
biodegradable.
High
Low
High
Medium
High
Medium
Medium
2-12
-------
(1) Photolysis. Dioxins (i.e., dibenzodioxin and its chlorinated
derivatives) absorb electromagnetic radiation above 290 nm, and can be
expected, therefore, to be subject to photolysis by sunlight (EPRI 1983).
Solar radiation of wavelengths lower than 290 nm is filtered out in the
troposphere, so absorption by chemicals of radiation at lower wavelengths
is moot in considering environmental photodegradation processes. All
chlorinated dioxins can be dechlorinated photolytically in the presence of
a suitable hydrogen donor (Crosby et al. 1973, Crosby 1978). Dechlorina-
tion is fast in organic media, slow in water, but practically nonexistent
on dry inorganic surfaces or in the crystalline state (Crosby 1978).
(a) Solid surfaces. Substantial research on the environmental
persistence of 2,3,7,8-TCDD has been done as part of the decontamination
of the area around the ICMESA chemical plant in Seveso, Italy, which was
contaminated when a trichlorophenol reaction vessel overheated in 1976,
blowing out the safety devices and spraying dioxin contaminated material
into the environment. The levels of dioxin in the soil decreased substan-
tially during the first six months following the accident, reaching a
steady state of 1/5 to 1/11 of the initial levels (DiDomenico et al.
1982). An experiment was conducted at this site to determine' the effec- •
tiveness of photolysis in decontaminating surface deposits of dioxin on
foliage. Test plots were sprayed with olive oil to act as an H-donor, and
the levels of dioxin on grass were found to be reduced by over 80% within
nine days (Crosby 1981). 2,3,7,8-TCDD in contaminated soil was also found
to be photolabile in sunlight when the soil was suspended in an aqueous
solution of a surfactant. The destruction of 8 ug/ml of 2,3,7,8-TCDD in
0.02 M hexadecylpyridinium chloride could be accomplished in 4 hours
(Botre et al. 1978). Figure 2-1 illustrates the photodegradation of
2,3,7,8-TCDD in the presence of hydrogen donors (Bertoni et al. 1978).
(b) Vapor phase. Photolysis in the gas phase also appears to
be a degradative mechanism for 2,3,7,8-TCDD (EPRI 1983). In studies con-
ducted at Stanford Research Institute, the half life of gaseous 2,3,7,8-
TCDD in sunlight was found to be 5 to 24 days, and may be the result of
indirect photolysis through the attack of hydroxyl radicals on the dioxin.
This mode of destruction would be similar to the attack of hydroxyl radi-
cals on PCBs (Leifer et al. 1983). In general, however, it appears that
a hydrogen donor is necessary for photodegradation of dioxins (Crosby
1978), although Gebefugi et al. (1977) concluded that in sunlight photo-
decomposition of TCDD present on soils is likely to occur even in the
absence of hydrogen donors (organic solvents).
Orth et al. (1989) conducted photolysis experiments with vapor-phase
2,3,7,8-TCDD under illumination in the UV region from 250 nm to 340 nm.
The rate constants in helium and air were very similar, 5.4 x 10"3 sec"1
and 5.9 x 10"3 sec"1, respectively, which corresponds to a quantum yield
in air of 0.033 ± 0.046. No products could be observed in the mass spec-
trometer, so. Orth et al. (1989) postulated that the product might be
sorbing to the surface of the photolysis cell and lost from potential
analysis.
2-13
1583q
-------
100
o
cc
UJ
a.
25-
DAYS
Figure 2-1. Destruction curves of 2,3,7.8-TCOO. (1) Seveso soil spread
with mixture of ethyl oleate and xylene exposed to natural
solar radiation. (2) TCOO sample placed in a room spread
with the ethyl oleate-xylene mixture and irradiated with a
Philips HLU 300-W lamp. Irradiation power on the sample:
ZOuW/an2.
2-14
-------
Podoll et al. (1986) estimated the photolysis rate of 2,3,7,8-TCDD
vapors in the atmosphere based on the quantum yield data for photolysis in
hexane. The half-life in summer sunlight at 40° latitude was calcu-
lated to be 58 minutes, but Podoll et al. (1986) stated this estimate is
an upper limit. The rate constant was also estimated for oxidation of
2,3,7,8-TCDD vapors by OH radical, the dominant transformation pathway in
the atmosphere. At an average concentration of OH radicals in the atmos-
phere of 3 x 10"15H, the half-life of 2,3,7,8-TCDD vapor via OH radical
oxidation was estimated to be 200 hours.
Atkinson (1987) also estimated reaction rate constants and atmospheric
half-lives for 2,3,7,8-TCDD and similar haloaromatics. The atmospheric
lifetime of 2,3,7,8-TCDD due to the OH radical reaction was calculated to
be 7 to 8 hours; based on the room temperature OH radical reaction rate
constant and OH radical concentration for the average 12-h daytime, the
atmospheric lifetime for 2,3,7,8-TCDD due to this gas-phase reaction was
calculated to be an estimated 3 days.
Mill et al. (1987) reported preliminary photolysis experiments with
atmospheric 2,3,7,8-TCDD. The half-life for vapor phase 2,3,7,8-TCDD in •
simulated sun was reported to be several hours. 2,3,7,8-TCDD was also
sorbed to fly ash particulates, and the particulate 2,3,7,8-TCDD was
suspended in recirculating air. The 2,3,7,8-TCDD sorbed to particulate
fly ash showed half-life values of several hundred hours; in fact, for
2,3,7,8-TCDD in particulate form, there was virtually no photolytic reac-
tion after 30 hours of illumination.
(c) Liquid phase. Dulin et al. (1986) studied the photolysis of
2,3,7,8-TCDD in aqueous solutions and reported a quantum yield of 0.05,
similar to that reported by Orth et al. (1989) for vapor-phase-2,3,7,8-
TCDD. In addition, Dulin et al. (1986) reported that the calculated sun-
light photolysis rate constant for 2,3,7,8-TCDD in surface water in summer
with sunlight levels found at 40° latitude would yield an estimated
half-life value of 4.6 days. The quantum yield for photodegradation of
2,3,7,8-TCDD in water was three times greater under artificial light at
313 nm than under sunlight, and the artificial light photolysis quantum
yield in hexane was 20 times greater than in water-acetonitrile as the
solvent. Based on their analyses of photodegradation products, Dulin et
al. (1986) concluded that cleavage of the C-0 bond rather than the C-C1
bond may be the major pathway for 2,3,7,8-TCDD photolytic loss.
Podoll et al. (1986) used the Dulin et al. (1986) quantum yield data
for 313 nm, 2.2 x 10'3, and calculated half-life values for dissolved
2,3,7,8-TCDD under sunlight at 40° latitude in clear near-surface
water under clear skies (24-h days). The seasonal values for half-lives
were calculated to be 21 hours in summer, 51 hours in fall, 118 hours in
winter, and 27 hours in spring. Sorption of 2,3,7,8-TCDD to sediments,
even suspended solids, would have the effect of slowing the photolysis
rates.
2-15
1583q
-------
Buser (1988) studied the photolytic decomposition rates of 2,3,7,8-
TCDD in a mixture of other chloro- and bromo-dibenzodioxins and dibenzo-
furans. Studies were performed in dilute hydrocarbon solutions as well as
solid phases on quartz surfaces, under both sunlight and artificial labor-
atory illumination (fluorescent lights). Photochemical decomposition pro-
ceeds much faster in hydrocarbon solutions under sunlight compared with
decomposition of 2,3,7,8-TCDD dispersed as thin solid films. 2,3,7,8-TCDD
in i-octane solutions in quartz vials showed first-order kinetics with an
estimated half-life ranging from 14 min to 200 min in two solutions. In
tests under ordinary fluorescent laboratory lights, the same solutions
were exposed in glass vials (rather than quartz), and half-lives of great-
er than 28 days were reported. 2,3,7,8-TCDD dispersed as a thin film in
quartz vials and exposed to sunlight showed an estimated half-life of
120 hours. Buser (1988) stated the reported half-life values have limited
accuracy because of the few data and because of the formation of 2,3,7,8-
TCDD during photodecomposition of the higher halogenated forms, so the
estimated accuracy was reported to be ±50%. From the data, Buser
(1988) concluded that in hydrocarbon solutions, the major photochemical
reaction was reductive dehalogenation, leading to lower chlorinated
dibenzodioxins which then likely are further degraded to unsiibstituted
dibenzo-p-dioxin.
(2) Oxidation. Stehl (1973) has suggested that 2,3,7,8-TCDD is
probably stable to oxidation in the ambient environment. In general,
2,3,7,8-TCDD is highly persistent in soils below the level that can be
reached by sunlight (Miyata and Kashimoto 1979). Although the previously
noted gas-phase photolysis of 2,3,7,8-TCDD could mean that dioxins are
oxidized by hydroxyl radicals, there is no direct evidence for this envi-
ronmental process in the absence of sunlight.
(3) Hvdrolvsls. There is no available evidence indicating that
hydrolysis would be an operative environmental process for degradation of
dioxins (EPRI 1983).
(4) Volatilization. The vapor pressure values of 2,3,7,8-TCDD at
25°C (in mm Hg) have been reported to range from (7.4 ± 0.4) x 10"10 by
Podoll et al. (1986) to 1.5 x 10'9 by Schroy et al. (1985) to 4.65 x 10'9
by Rordorf (1987). The reported log KQC values are 6.6 ± 0.7 (Walters and
Guiseppi-Elie 1988) and 7.15 to 7.34 (Jackson et al. 1986). Based on
these values, 2,3,7,8-TCDD should not volatilize rapidly from organic
soils and sediments; however, due to its stability and persistence via
other transformation and transport pathways, volatilization should not be
ignored as a mechanism for loss from soils.
The water solubility of 2,3,7,8-TCDD has recently been reported to be
7.91 ± 2.7 ng/1 (ppt) at 25'C by Adams and Blaine (1986), 19.3 ± 3.7 ng/1
at 22°C by Marple et al. (1986a), and 483 ± 94 ng/1 at 17.3eC by Lodge
(1989). This latter value by Lodge is even higher than values from the
1970's, such as 200 ng/1 reported by Kearney et al. (1973) with less re-
fined analytical methods. Lodge (1989) also reported a water solubility
2-16
1583q
-------
value at 4.3°C of 12.9 ± 1.2 ng/1 which is more in line with the solubili-
ty values at 25°C and 22°C reported by Adams and Blaine (1986) and Marple
et al. (1986a), respectively.
Podoll et al. (1986) utilized their reported vapor pressure value and
the Marple et al. (1986a) water solubility to compute a Henry's law con-
stant of 1.6 x 10"5 atm»nr/mol. Estimates of the Henry's law constant
lack precision for vapor pressures and water solubility measurements,
because of the ranges of reported values, but the values reported do
suggest that 2,3,7,8-TCDD may volatilize from water and moist surfaces.
Observations from the Seveso incident indicate that when 2,3,7,8-TCDD
is deposited on the soil surface, both volatilization and photodegradation
are initially rapid (DiDomenico et al. 1982). Once it has been washed
into the soil, however, 2,3,7,8-TCDD is unaffected by either volatiliza-
tion or photolysis (Plimmer 1978, Kriebel 1981). Podoll et al. (1986)
reported that when 2,3,7,8-TCDD has been mixed to a depth of only a few
millimeters in dry soil, the volatility will be extremely slow because of
the very low unsteady-state diffusion of the vapor in the dry soil air
phase. .However, Freeman and Schroy (1985) point out that low volatility
chemicals may bind strongly with very dry soil but, once a molecular
microlayer of water covers the soil particles, the chemical should become
more volatile; in addition, water vaporization may enhance the rate of
chemical vaporization from a soil column.
Freeman and Schroy (1985) reported that a simple half-life model is
totally inadequate to describe volatilization and environmental persis-
tence, and used two coupled partial differential equations both coupled to
a second-order heat transfer equation to describe 2,3,7,8-TCDD transport
in soils. At the surface, the apparent half-life for 2,3,7,8-TCDD is
measured in weeks, but mixed with soils below a depth of 5cm, the apparent
half-life is measured in years and the "half life concept" is invalid.
At Times Beach, MO, 2,3,7,8-TCDD volatilized from soils most rapidly in
summer and did not volatilize to any appreciable extent during the winter.
At the end of the first summer, over 90% of the applied 2,3,7,8-TCDD con-
tained in the top one centimeter of soil was lost, and by the end of the
first year, over 95% of the applied 2,3,7,8-TCDD had been lost from the
top 1 cm. Further, of the total applied 2,3,7,8-TCDD, over 50% volatil-
ized during the first summer after the initial application at Times Beach
in 1972, based on the estimates of losses provided by Freeman and Schroy
(1985).
Subsequently, Freeman and Schroy (1989) utilized another mathematical
model to evaluate the transport of 2,3,7,8-TCDD in soil via vapor trans-
port and other mechanisms. Results of the simulation predicted that only
0.01 percent of the 2,3,7,8-TCDD present in the top 1 cm of soil should be
lost due to vaporization over the 16 months of the experiment, and that
0.1 percent of the 2,3,7,8-TCDD in the top 1 mm should have volatilized.
An observed reduction of approximately 50 percent in the 2,3,7,8-TCDD
2-17
1583q
-------
concentration in the top 3 mm of soil was not explainable by the model
results, so another mechanism such as surface photodegradation was postu-
lated. In addition to clarifying that the rate of transport in soil is
very slow via vaporization, Freeman and Schroy (1989) reported that
analyses indicated that over 99 percent of the 2,3,7,8-TCDD applied to the
roads at Times Beach, Missouri is still in the soil.
Eduljee (1987) estimated the volatilization fluxes for 2,3,7,8-TCDD
from the Jury et al. mathematical model for transport of chemicals through
soil and loss at the soil-air interface. The diffusion of 2,3,7,8-TCDD
in soil was estimated to be vapor-dominated up to a volumetric water
content of 0.3 nr/ro > and then liquid-dominated to saturation; i.e.,
initially decreasing in transport as pore spaces are filled, then increas-
ing in transport as liquid diffusion predominates. In addition, the model
predicted that 2,3,7,8-TCDD is subject to wicking, and Eduljee (1987)
pointed out this was contrary to the views of Podoll et al. (1985). In-
creasing the organic content of the soil retards the movement of 2,3,7,8-
TCDD and depresses the volatilization flux.
Palausky et al. (1986) injected 2,3,7,8-TCDD in various organic sol-
vents into soil columns to determine the extent of vapor phase diffusion
of 2,3,7,8-TCDD in soil and the effects of carrier medium on the degree
of migration. The solvents used included acetone/water, xylene, chloro-
form, toluene, methanol, iso-octane, and decane. No noticeable changes in
2,3,7,8-TCDD concentration profiles were observed after 30 days incubation
at temperatures ranging from 0° to 20°; however, a measurable change was
observed at 40°C, indicating volatilization may have been causing transport,
Little or no migration was observed with the acetone/water mixture. With
organic solvents, the slowest migration was with saturated hydrocarbons,
somewhat greater with methanol, and migration was highest with aromatic
solvents and chloroform. The extent of migration was related to solubili-
ty of 2,3,7,8-TCDD in the solvent, but a direct correlation between solu-
bility and extent of migration was not found. Thus, the data at 40°C
reflect not only the possible 2,3,7,8-TCDD vapor transport but also the
volatility, viscosity, and interaction with soil organic matter for each
of the solvents.
Concerning the volatility of 2,3,7,8-TCDD from water, Podoll et al.
(1986) utilized data for the molecular diffusivity in air and in water to
calculate volatilization half-lives. The volatilization half-life was
about 32 days for ponds and lakes and about 15 days for rivers.
(5) Sorotion. If 2,3,7,8-TCDD is washed into a water body, it
will become strongly sorbed to bottom sediments (NRC Canada 1981). Its
lipophilicity (log KQW * 6.6 to 7) implies that 2,3,7,8-TCDD can be
part of the oily surface scum of a water body as well as part of the sedi-
ment. 2,3,7,8-TCDD is not easily washed out of soils, and it is consid-
ered to be virtually immobile to aqueous phase transport in all but very
sandy soils (Young 1981).
2-18
1583q
-------
Palausky et al. (1986) conducted studies with 2,3,7,8-TCDD in various
organic solvents and found that transport of 2,3,7,8-TCDD does occur. By
inference, these data suggest that 2,3,7,8-TCDD may be highly sorptive to
soils, especially with high levels of organic matter, but that organic
solvents may play a role in the sorption/transport of 2,3,7,8-TCDD by
acting as dispersing media within soils. Puri et al. (1989) reported
research which indicated that 2,3,7,8-TCDD is highly sorptive, but that
organic co-contaminants such as waste oil and surfactants may act to
enhance the translocation of 2,3,7,8-TCDD. Also, the distribution
coefficient of 2,3,7,8-TCDD in soils is dependent on the total organic
matter, with higher sorption at higher soil organic levels, but the
degree of sorption is also dependent on the types of organic matter
present and their stability and solubility in the presence of the various
co-contaminants.
Kapila et al. (1989) also tested the soil migration of 2,3,7,8-TCDD
in waste crankcase oil. Field experiments showed that 2,3,7,8-TCDD moved
downward in the soil column, and this movement was attributed to applica-
tion of water to simulate rainfall. The water application resulted in
displacement of the waste oil components and 2,3,7,8-TCDD from the macro--
pore spaces in the soil column. The formation of colloidal suspensions
and the presence of 2,3,7,8-TCDD in these suspensions may play a signifi-
cant role in the downward movement of 2,3,7,8-TCDD. Since the amount of
2,3,7,8-TCDD recovered from each column after 12 months was "approximately
the same as the amount initially applied," it was inferred by Kapila et
al. (1989) that there was little loss of 2,3,7,8-TCDD from the soil sur-
face via volatilization or photolysis during the one-year period.
Walters and Guiseppi-Elie (1988) evaluated the sorption of labeled
2,3,7,8-TCDD to soils from methanol/water mixtures in batch shake testing.
The log KQC, (the logarithm of the aqueous phase partition coefficient
normalized for the organic content of soils) was reported to be 6.6 ± 0.7,
as extrapolated from a linear regression analysis of various data. Data
were presented which indicated that methanol/water mixtures of 2,3,7,8-
TCDD did not result in solvent-soil interactions to increase 2,3,7,8-TCDD
accessibility to soil organic matter to the same degree as that for var-
ious other organic compounds which are less hydrophobic. At the methanol
concentrations tested, there was no significant increase in 2,3,7,8-TCDD
mobility over that expected in the absence of solvents in the liquid
phase. This relationship might apply to other water-miscible solvents,
but Walters and Guiseppi-Elie (1988) stated it was unclear whether
immiscible solvents would affect 2,3,7,8-TCDD mobility.
(6) Bioaccumulat1on/bloconcentrat1on. Bioconcentration factors
(BCF) for exposures from water and sediment are presented for 2,3,7,8-
TCDD in Table 2-3. Nabholz (1989) reported that the data base for
2,3,7,8-TCDD was much better than the information for 2,3,7,8-TCDF
(Table 2-5) (I.e., there were 14 measured factors for 2,3,7,8-TCDD and
only 3 for the TCDF). In addition, the data for freshwater fish from
2-19
1583q
-------
8992H
Table Z-5. Physical and Chemical Properties of 2.3.7.8-TCDF
Property
Value
Convents
References
Analytical Hethod
ro
i
ro
O
Chemical name
Synony*
Empirical formula
CAS number
Molecular weight
Pure form
Melting Point
logic
Vapor density
(above solid)
Evaporation rate
(fro* absorbed state)
Water solubility
Henry's law constant
2.3.7.8-Tetrachlorodibenzo-
furan
TCOF; 2.3.7.8-TCDF
51207-31-9
305.98
White crystalline solid
227-228
5.82
Vapor pressure at 25*C 9.21 x 10~* torr (mm Hg)
3.0 x 10 U g/cm3
2.1 x 10'12 g/c«2/s
4.33 ug/1
8.6 x 10~5 atm.M3/mol
3.49 x 10
25'C
USEPA (1978)
Burkhard and Kuehl
(1986)
Eitzer and Kites
(1988)
USEPA (1978)
USEPA (1978)
Chenest. Versar (1989)
Chenest. Versar (1989)
Chemest. Versar (1989)
Reverse-phase HPLC and LRMS
detection.
Gas saturation and effusion Method.
-------
Table 2-5. (continued)
Property
Value
Convents
References
Analytical Method
Volatilization from Water
14 hrs.
12 hrs.
Bioconcentratlon factors
in fish:
Exposure fro* water only 3.900
ro
i
ro
Exposure fro* dietary
sources 0.1538
Half-life for rivers
Ha If-life for lakes * ponds
Versar (1989)
Versar (1989)
Nabholz (1989)
Geometric Means for fresh
Mater fish, equilibrium tine
is 7 days. n=2. Most Reliable
in concentration range of
0.410 to 3.93 ppt. There is
no data on ware water species
exposure to 2.3.7.8-TCDF;
therefore, the geometric
•ean for fresh water fish
is calculated froa cold-
water species data.
Based on one study on
Harm water species (H-l),
dietary concentration of
162.0 ppt.
* Recommended values at stated temperature.
-------
water exposures of 2,3,7,8-TCDD were better than the data from exposure
from dietary sources (i.e., food, sediment, or fly ash). While there
were an equal number of studies (seven for water exposure and seven for
dietary exposure), the 2,3,7,8-TCDD concentrations used in the water expo-
sure studies had a much broader range than the range of exposure concen-
trations used in the studies with contaminated dietary sources.
There was a relationship between BCF and exposure concentration,
whether the exposures were from water or dietary sources. As exposure
concentration increased, the BCF decreased. This relationship was ob-
served both for cold water species of freshwater fish exposed to 2,3,7,8-
TCDD in water and for warm water species of freshwater fish exposed to
2,3,7,8-TCDD in diet.
For freshwater fish exposed to 2,3,7,8-TCDD via water only, the geo-
metric mean BCF is 20,600. This mean BCF was based on 7 measurements.
This BCF is most reliable for water concentrations from 0.038 ppt to
107 ppt; outside this range (especially the upper bound), this BCF is
less reliable.
Nabholz (1989) reported that for cold water species of freshwater
fish, there was a significant correlation between water concentration (in
ppt) and fish BCF (on a whole fish, wet weight basis): the lower the con-
centration of 2,3,7,8-TCDD in the water, the higher the BCF in fish. The
regression equation describing the relationship between exposure concen-
tration in water and fish BCF is:
log Y * 4.44397 - 0.27923 (log X)
where Y « fish BCF on a wet weight, whole body basis; X • water concentra-
tion in ppt; N - 6; rz - 0.76; and r = 0.87 (which is statistically sig-
nificant at JD < 0.05). This regression was calculated based on six
measurements. This equation is most reliable for water concentrations
from 0.038 to 107 ppt; outside this range (especially the upper bound),
this equation is less reliable. The geometric mean fish BCF from this
data set is 24,100.
Nabholz (1989) reported that for warm water species of freshwater fish
there was only one data point. The water concentration was 0.870 ppt and
the fish BCF was 7,900.
There were no data available for bioconcentration factors for
primarily water exposures of 2,3,7,8-TCDD in freshwater benthic fish
species, marine pelagic fish, or marine benthic species.
When the exposure of 2,3,7,8-TCDD to fish is mostly through dietary
sources (i.e., food, sediment or fly ash) rather than water, then a BCF
based on dietary exposure must be computed. Bioconcentration factors
(BCF) based on dietary exposures are probably more realistic than those
2-22
1583q
-------
based on water concentrations because of the strong potential for 2,3,7,8-
2,3,7,8-TCDD to partition to suspended solids and sediments in the
aquatic environment.
Nabholz (1989) reported that for freshwater fish, the geometric mean
BCF for 2,3,7,8-TCDD from contaminated dietary sources was 0.09697 and the
mean equilibrium time was 84 days. This BCF was based on 7 measurements,
and is most reliable for dietary concentrations of 2,3,7,8-TCDD from
39.0 ng/kg (ppt) to 2,000.0 ng/kg (ppt); outside this range {especially
the upper bound), this BCF is less reliable.
For cold water fish species exposed primarily via dietary sources,
the fish BCF was 0.1119 (geometric mean, N - 4), and the time to
equilibrium was 91 days. This BCF was based on 4 fishes, all collected
from the same contaminated area, which had a sediment 2,3,7,8-TCDD
concentration of 494 ng/kg (ppt).
Nabholz (1989) reported that for warm water species, there was a sig-
nificant correlation between dietary concentration (ng/kg; ppt) and fish
BCF (on a whole fish, wet weight basis): the lower the water concentra-
tion, the higher the BCF. The regression equation describing the rela-
tionship between exposure concentration in diet and fish BCF is:
log Y = 1.544 - 1.11199 (log X)
where f * fish BCF on a wet weight, whole body basis; X = concentration of
2,3,7,8-TCDD in dietary sources in ng/kg (ppt); N - 3; rz = 0.56; and
r = 0.81, This regression was calculated from 3 measurements. This equa-
tion is most reliable for dietary concentrations from 39.0 ng/kg (ppt) to
2,000.0 ng/kg (ppt); outside this range (especially the upper bound), this
equation is less reliable. The geometric mean fish BCF from this data set
is 0.08016.
There are apparently no data available for fish BCF values primarily
via dietary sources for either freshwater benthic species, marine pelagic
species, or marine benthic species.
(7) Biotransformation and biodeqradation. Investigations on
dioxins have focused on the microbial degradation of 2,3,7,8-TCDD.
Matsumura and Benezet (1973) tested approximately 100 strains of micro-
organisms which had previously shown the ability to degrade persistent
pesticides, and only five strains showed any ability to degrade 2,3,7,8-
TCDD, based on autoradiographs of thin-layer chromatograms. Although it
is possible that the less chlorinated dioxins are more susceptible to bio-
degradation, microbial action on 2,3,7,8-TCDD is very slow under optimum
conditions (Mutter and Philippi 1982). Long-term incubations of radio-
labeled 2,3,7,8-TCDD yielded no radioactivity in carbon dioxide traps
after one year, and analyses of the cultures showed that at most, 1 to
2 percent of a potential metabolite (assumed to be an hydroxylated deriva-
2-23
1583q
-------
tive of 2,3,7,8-TCDD) could be detected. Camoni et al . (1982) added
organic compost to soil in an attempt to enrich the soil and enhance the
2,3,7,8-TCDD biodegradation rate, but the soil amendment with compost was
shown to have had no clear effect on degradation of 2,3,7,8-TCDD. It has
been suggested that losses of dioxins from soil are due to processes other
than microbial degradation.
Bumpus et al . (1985) tested the white rot fungus, Phanerochaete
chrvsosporium. which secretes a unique H^Og -dependent extracellular
lignin-degrading enzyme system capable or generating carbon-centered free
radicals. Radiolabeled 2,3,7,8-TCDD was oxidized to labeled C02 by
nitrogen-deficient, ligninolytic cultures of P. chrvsosporium. Since the
label was restricted to the ring, it was concluded that the strain was
able to degrade halogenated aromatic rings. In 10 ml cultures containing
1,250 pmol of substrate, 27.9 pmol of 2,3,7,8-TCDD were converted to
labeled-C02 during the 30-day incubation period, thus only about 2 per-
cent of the starting material was converted.
Arthur and Frea (1988) studied the microbial activity in Times Beach,
Missouri soils contaminated at levels from 0.008 to 2.4 ug/g (±10 percent)
2,3,7,8-TCDD. Although biodegradation of 2,3,7,8-TCDD per se was not
tested, there were no adverse effects on numbers of microorganisms
(aerobic eutrophic or oligotrophic bacteria, fungi, or actinomycetes), on
enzymatic assays (including dehydrogenase, arylsulfatase, rhodanase, and
acid and alkaline phosphatase), or on soil respiration (CO? evolution).
The data suggest a lack of observed toxicity to microorganisms, and may
indicate that 2,3,7,8-TCDD is not bioavailable to microorganisms because
of soil binding or other factors causing low mobility.
2.3 Chemistry and Fate of 2,3,7,8-TCDF
2.3.1 Chemical Identity
2,3,7,8-Tetrachlorodibenzofuran (2,3,7,8-TCDF) is a member of the
135 compounds known as polychlorinated dibenzofurans (PCDFs). The struc-
ture of 2,3,7,8-TCDF is:
2.3.2 Chemical and Physical Properties
The few measured properties of 2,3,7,8-TCDF obtained from the litera
ture are listed in Table 2-5. In general, fewer measured properties are
available than for 2,3,7,8-TCDD, and the chemicals show somewhat similar
2-24
1583q
-------
values for those properties where data are available or have been
estimated.
2.3.3 Environmental Fate and Transport
Polychlorinated dibenzofurans can be photodechlorinated by sunlight in
the presence of organic substances that can serve as donors of hydrogen
atoms. This process of photodechlorination is similar to what occurs in
the degradation of dioxins, and it is probably the only degradative fate
pathway for dibenzofurans in the environment.
Since there is little or no information on dibenzofurans for other
environmentally relevant processes, fate and transport pathways must be
derived from behavior of structurally similar dioxins. Dibenzofurans can
be expected then to be sorbed strongly to soils and sediments, to be
bioconcentrated in fish, and to be essentially nonbiodegradable in the
environment. Erosion and aquatic transport of sediment will be the main
transport pathway of the sorbed dibenzofurans. Table 2-6 is a summary of
the environmental fate of dibenzofurans.
(1) Photolysis. Polychlorinated dibenzofurans (PCDFs) absorb
electromagnetic radiation at wavelengths above 290 nm and should be
capable, therefore, of undergoing photolysis when subjected to sunlight
(EPRI 1983). Crosby et al. (1973) report that polychlorinated dibenzo-
furans undergo photolytic dechlorinate in the presence of a hydrogen
donor, with more highly chlorinated congeners being more stable. In
contrast, Hutzinger (1973) and Buser (1976) report that the more highly
chlorinated congeners undergo photodegradation more rapidly. Hutzinger
(1973) found that both 2,8-dichloro- and octachlorodibenzofuran photode-
chlorinate rapidly in methanol or hexane. The potential for photodegrada-
tion of dibenzofurans in the environment appears, therefore, to be similar
to the photodegradation potential of dioxins; in the presence of a hydro-
gen donor and sunlight, polychlorinated dibenzofurans will dechlorinate.
Buser (1988) studied the photolytic decomposition rates of 2,3,7,8-
TCDF in a mixture of other chloro- and bromo-dibenzofurans and dioxins.
Studies were performed in dilute hydrocarbon solutions and as solid phases
on quartz surfaces under sunlight and artificial laboratory illumination
(fluorescent lights). 2,3,7,8-TCDF undergoes rapid photolytic decomposi-
tion in dilute hydrocarbon solutions of i-octane. When the solutions were
illuminated with sunlight in quartz vials, the estimated half-life was
220 min and 90 min in each of the two solutions. For the same solutions
illuminated with artificial light in glass vials, the half lives were
greater than 28 days. When 2,3,7,8-TCDF was dispersed as a thin film in
quartz vials and exposed to sunlight, the estimated half-life was reported
to be 35 hours. While these data do indicate that photolytic decomposi-
tion of 2,3,7,8-TCDF does occur, Buser (1988) stated the half-life values
have limited accuracy because of the few data and because of the formation
2-25
15B3q
-------
B992H
Table 2-6. Suimary of Environmental Fate of Oibenzofurans
Environmental
process
Sumary
statement
Confidence
in data
Photolysis
Oxidation
Hydrolysis
Volatilization
Sorption
Bioaccmulation
Biodegradation
May be only natural mechanism
leading to destruction of
dibenzofurans.
No information found.
Dibenzofurans are stable to
hydrolysis.
No information found.
Oibenzofurans strongly sorbed by
solids, especially with high
organic content.
No specific information found;
potential is inconclusive.
Probably nondegradable in the
environment.
Medium
Low
High
Low
Medium
Low
Low
2-26
-------
of 2,3,7,8-TCDF during photodegradation of higher halogenated forms, so
the estimated accuracy of all the half-life data was reported to be ±50%.
From the data, Buser (1988) concluded that the major photochemical
reaction in hydrocarbon solution was reductive dehalogenation, leading to
lower chlorinated dibenzofurans which would then likely be further
degraded to unsubstituted dibenzofuran.
(2) Oxidation. There is no information available on the oxidation
of dibenzofurans under environmentally relevant conditions (EPRI 1983).
(3) Hvdrolvsls. No information is available indicating that
hydrolysis would be an operative environmental process for degradation of
polychlorinated dibenzofurans (EPRI 1983).
(4) Volatilization. The vapor pressure of 2,3,7,8-TCDF 1s 9.21 x
10~7 mm Hg at 25°C (Eitzer and Hites 1988). The reported log Kow
value is 5.82 (Burkhard and Kuehl 1986), suggesting the soil-sorption co-
efficient would be low. Based on these data, TCOF should not volatilize
rapidly from organic soils or sediments; however, due to the stability
and persistence of TCDF via other transformation and transport pathways,
volatilization should not be ignored as a mechanism of loss from soils.
Experimental values have not yet been reported for the water solubili-
ty or Henry's law constant for 2,3,7,8-TCDF, but similarities to 2,3,7,8-
TCDD suggest that TCDF might volatilize rapidly from water or moist
surfaces. Estimations from Chemest (Versar 1989) indicate volatilization
half-lives from water bodies on the order of 12 to 14 hours.
(5) Sorotlon. Sorption of PCDFs has not been studied (EPRI 1983).
However, the log KQW value of 5.82 (Burkhard and Kuehl 1986) and the
structural similarity of PCDFs to PCDDs suggest that after TCDF has been
washed into a soil, it will readily sorb to organic materials and other
components of soils.
(6) B1oaccumulat1on/b1oconcentrat1on. The data available for bio-
concentration factors (BCF) of 2,3,7,8-TCDF (Table 2-5) are more limited
than the data for 2,3,7,8-TCDD (Table 2-3). The data suggest that BCF
values for 2,3,7,8-TCDF are lower than BCF values for 2,3,7,8-TCDD.
Nabholz (1989) reported that for freshwater fish exposed to 2,3,7,8-
TCDF via water only (not dietary sources), the geometric mean BCF value
was 3,900 and the time to equilibrium is 7 days. This mean BCF was based
on 2 measurements. This BCF 1s most reliable for water concentrations
from 0.410 to 3.93 ppt; outside this range (especially the upper bound)
this BCF is less reliable.
Nabholz (1989) also reported that for cold water species of freshwater
fish, there was a significant correlation between water concentration (in
ppt) and fish BCF (on a whole fish, wet weight basis): the lower the
2-27
1583q
-------
water concentration, the higher the BCF. The regression equation describ-
ing the relationship between exposure concentration in water and fish BCF
is:
log Y = 3.6272 - 0.39897 (log X)
where Y = fish BCF on a wet weight, whole body basis; X = water concentra-
tion in ppt; N = 2; rz » 1.0; and r - 1.0. This regression was calcu-
lated based on 2 measurements. This equation is most reliable for water
concentrations from 0.410 to 3.93 ppt; outside this range (especially the
upper bound), this equation is less reliable. The geometric mean fish
BCF from this data set is 3,900.
There are apparently no data available for BCF values for 2,3,7,8-TCDF
for warm water pelagic or benthic freshwater species or for marine fishes.
Nabholz (1989) reported that there was only one study available which
measured the bioconcentration of 2,3,7,8-TCDF from contaminated dietary
sources, and that was for a warm water species of freshwater fish. The
BCF was 0.1538 on a whole body, wet or fresh weight basis, based on a die-
tary concentration of 182.0 ng/kg (ppt). There were apparently no data
available for cold water species of freshwater pelagic fish, or for fresh-
water benthic species, or for marine fishes.
(7) Blotransformation and biodeqradation. No information was
found relating to the biodegradation of chlorinated dibenzofurans.
Structurally similar chlorinated dioxins are considered to be essentially
nonbiodegradable in the environment (EPRI 1983), and it is, therefore,
possible that chlorinated dibenzofurans will behave in a like manner and
remain persistent to attack by microorganisms and other types of biotrans-
formation processes.
2.4 References
Adams WJ, Blaine KM. 1986. A water solubility determination of
2,3,7,8-TCDD. Chemosphere 15:1397-1400.
Arthur MF, Frea JI. 1988. Microbial activity in soils contaminated with
2,3,7,8-TCDD. Environ. Toxicol. Chem. 7:5-13.
Atkinson R. 1987. Estimation of OH radical reaction rate constants and
atmospheric lifetimes for polychlorobiphenyls, dibenzo-p-dioxins, and
dibenzofurans. Environ. Sci. Technol. 21:305-307.
Bertoni G, Brocco D, DiPalo V, Liberti A, Possanzini M, Bruner F. 1978.
Gas chromatographic determination of 2,3,7,8-tetra-chlorodibenzodioxin in
the experimental decontamination of Seveso soil by ultraviolet radiation.
Anal. Chem. 50:732-735.
2-28
1583q
-------
Botre C. Memoli A, Al-Haique F. 1978. TCDD solubilization and
photodecomposition in aqueous solutions. Environ. Sci. Techno!.
12:335-336 as cited in Vuceta et al. (1983).
Bumpus JA, Tien M, Wright D, Aust SD. 1985. Oxidation of persistent
environmental pollutants by a white rot fungus. Science 228:1434-1436.
Burkhard LP, Kuehl DW. 1986. n-Octanol/water partition coefficients by
reverse phase liquid chromatography/mass spectrometry for eight
tetrachlorinated planar molecules. Chemosphere 15:163-167.
Buser HR. 1976. Preparation of qualitative standard mixtures of
polychlorinated dibenzo-p-dioxins and dibenzofurans by ultraviolet and
gamma irradiation of the octachlorocompounds. J. Chromatogr. 192:303-307
as cited in Vuceta et al. (1983).
Buser HR. 1988. Rapid photolytic decomposition of brominated and
brominated/chlorinated dibenzodioxins and dibenzofurans. Chemosphere
17:889-903.
Camoni I, Dimuccio A, Pontecorvo D, Taggi F, Vergori I. 1982. Laboratory
investigation for the microbial degradation of 2,3,7,8-tetrachlorodibenzo-
p-dioxin in soil by addition of organic compost. Pergammon Ser. Environ.
Sci. 5:95-103.
Crosby DG, Moilanen KW, Wong AS. 1973. Environmental generation and
degradation of dibenzodioxins and dibenzofurans. Environ. Health
Perspect., Exp. Issue. 5:259-266.
Crosby DG. 1978. Conquering the monster - the photochemical destruction
of chlorodioxins. In: Disposal and decontamination of pesticides. MV
Kennedy, ed. ACS Symposium Series 73:1-12 as cited in Vuceta et al.
(1983).
Crosby DG. 1981. Methods of photochemical degradation of halogenated
dioxins in view of environmental reclamation. Paper presented on "Human
health aspects of accidental exposure to dioxins. Strategy for
environmental reclamation and community protection," Bethesda, MD,
October 5-7, 1982. Davis, CA: University of California, Department of
Environmental Toxicology.
DiDomenico A, Viviano G, Zapponi G. 1982. Environmental persistence of
2,3,7,8-TCDD at Seveso. In: Chlorinated dioxins and related compounds,
impact on the environment. Hutzinger et al., eds. Elmsford, NY:
Pergamon Press, pp. 105-113 as cited in Vuceta et al. (1983).
Dulin D, Grossman H, Mill T. 1986. Products and quantum yields for
photolysis of chloroaromatics in water. Environ. Sci. Technol. 20:72-77.
2-29
1583q
-------
EPRI. 1983. Electric Power Research Institute. State-of-the-art
review: PCDDs and PCDFs in utility fluid. Palo Alto, CA: EPRI-CS-3308.
Eduljee G. 1987. Volatility of TCDD and PCB from soil. Chemosphere
16:907-920.
Eitzer BD, Hites RA. 1988. Vapor pressures of chlorinated dioxins and
dibenzofurans. Environ. Sci. Techno!. 22:1362-1364.
Esposito MP, Teirnan TD, Dryden FE. 1980. Dioxins. U.S. EPA, IERL,
Office of Research and Development, Cincinatti, OH. EPA-600/2-80-197 as
cited in Marple et al (1986a).
Freeman RA, Schroy JM. 1985. Environmental mobility of TCDD.
Chenosphere 14:873-876.
Freeman RA, Schroy JM. 1989. Comparison of the rate of TCDD transport
at Times Bleach and at Eg!in AFB. Chemosphere 18:1305-1312.
Gebefugi I, Baumann R, Korte F. 1977. Naturwissenschaften 64:1337 as
cited in Choudhry and Webster (1987).
Geyer HF, Scheunert I, Korte F. 1987. Correlation between the
bioconcentration potential of organic environmental chemicals in humans
and their n-octanol/water partition coefficients. Chemosphere 16:239-252.
Mutter R, Philippi M. 1982. Studies on microbial metabolism of TCDD
under laboratory conditions, in: Chlorinated dioxins and related
compounds, impact on the environment. 0. Hutzinger et al. eds. Elmsford,
NY: Pergammon Press, pp. 87-93 as cited in Vuceta et al. (1983).
Hutzinger 0. 1973. Photochemical degradation of di and
octachlorodibenzofuran. Environ. Health. Perspect. Exp. Issue.
5:253-256.
Jackson DR, Roulier MH, Grotta HM, Rust SW, Warner JS. 1986. Solubility
of 2,3,7,8-TCDD in contaminated soils. In: Chlorinated dioxins and
dibenzofurans in perspective. Rappe C, Choudhary G, Keith LH, eds.
Chelsea, MI: Lewis Pub. Inc.: pp. 185-200.
Kapila S, Yanders AF, Orazio CE, Meadows JE, Cerlesi S, Clevenger TE.
1989. Field and laboratory studies on the movement and fate of
tetrachlorodibenzo-p-dioxin in soil. Chemosphere 18:1297-1304.
Kearney PC, Isensee RA, Helling CS, Woolson EA, Plimmer JR. 1973.
Chlorodicxins - origin and fate. Blair EH, ed., Advances in Chemistry
Series 120; Washington, DC: American Chemical Society, p. 106 as cited
1n Marple et al. (1986a),
2-30
1583
-------
Kriebel D. 1981. The dioxins: toxic and still troublesome. Environment
23:6-13 as cited in Vuceta et al. (1983).
Leifer A, Brink RH, Thorn GC, Partymiller KG. 1983. Environmental
transport and transformation of polychlorinated biphenyls. Washington,
DC: U.S. Environmental Protection Agency, Office of Toxic Substances.
EPA-560/5-83-025.
Leo A, Hansch C. 1979. Substituent constants for correlation analysis
in chemistry and biology. New York: John Wiley and Sons.
Lodge KB. 1989. Solubility studies using a generator column for
2,3,7,8-Tetrachlorodibenzo-p-dioxin, Chemosphere 18:933-940.
Lyman V(J, Reehl WF, Rosenblatt DH. 1982. Handbook of chemical property
estimation methods. New York, NY: McGraw-Hill.
Marple L, Brunck R, Throop L. 1986a. Water solubility of
2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ. Sci. Technol. 20:180-182.
Marple L, Berridge B, Throop L. 1986b. Measurement of the water-
octanol partition coefficient of 2,3,7,8-tetrachlorodibenzo-p-dioxin.
Environ. Sci. Technol. 20:397-399.
Matsumura F, Benezet JH. 1973. Studies on the bioaccumulation and
microbial degradation of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ.
Health Perspect. Sept. 253-258.
Mill T, Rossi M, McMillen D, Coville M, Leung D, Spang J. 1987.
Photolysis of tetrachlorodioxin and PCBs under atmospheric conditions.
Internal report prepared by SRI International for USEPA, Office of Health
and Environmental Assessment, Washington, DC.
Miyata H, Kashimoto T. 1979. Investigation on organochlorinated
compounds formed in Kanemi rice oil that caused the "yusho." J. Food
Hyg. Soc. Japan. 20:1-9 as cited in Vuceta et al. (1983).
Nabholz V. 1989. Bioconcentration factors for 2,3,7,8-chlorinated
dibenzodioxin and 2,3,7,8-chlorinated dibenzofuran. Unpublished.
Washington, DC: U.S. Environmental Protection Agency, Office of Toxic
Substances.
NRC Canada. 1981. National Research Council Canada, Ottawa.
Polychlorinated dibenzo-p-diox1ns: criteria for their effects on man and
his environment. As cited in Vuceta et al. (1983).
Orth RG, Ritchie C, Hileman F. 1989. Measurement of the photoinduced
loss of vapor phase TCDD. Chemosphere 18:1275-1282.
2-31
1583q
-------
Palausky J, Kapila S, Manahan SE, Yanders AF, Malhotra RK, Clevenger TE.
1986. Studies on vapor phase transport and role of dispersing medium on
mobility of 2,3,7,8-TCDD in soil. Chemosphere 15:1387-1396.
Plimmer JR. 1978. Photolysis of TCDD and triflural n on silica and soil.
Bull. Environ. Contam. Toxicol. 20:87-92.
Podoll RT, Jaber HM, Mill T. 1986. Tetrachlorodibenzodioxin: Rates of
volatilization and photolysis in the environment. Environ. Sci. Techno!.
20:490-492.
Puri RK, Clevenger TE, Kapila S, Yanders AF, Malhotra RK. 1989. Studies
of parameters affecting translocation of tetrachlorodibenzo-p-dioxin in
soil. Chemosphere 18:1291-1296.
Rappe C, Anderson PA, Bergqvist C, Brohede M, Hansson M, Kjeller LO,
Lindstrom G, Marklund S, Nygren M, Swanson SE, Tysklind M, Wiberg K.
1987. Overview of environmental fate of chlorinated dioxins and
dibenzofurans, sources, levels, and isomeric pattern in various matrices.
Chemosphere 16:1603-1618.
Reid RC, Prausnitz JM, Sherwood TK. 1977. The Properties of gases and
liquids, 3rd Ed. New York: McGraw-Hill Book Company, as cited in Schroy
et al. (1985).
Rordorf BF. 1985. Thermodynamic properties of polychlorinated compounds:
The vapor pressures and enthalpies of sublimation of ten dibenzo-para-
dioxins. Thermochimica Acta. 85:435-438.
Sarna LP, Hodge PE, Webster GRB. 1984. Octanol-water partition
coefficients of chlorinated dioxins and dibenzofurans by reversed-phase
HPLC using several CIS columns. Chemosphere 13(9):975-983.
Schroy JM, Hileman FD, Cheng SC. 1985. Physical/chemical properties of
2,3,7,8-TCDD (2,3,7,8-tetrachlorodibenzo-p-dioxin). Chemosphere
14:877-80.
Stehl RH. 1973. The stability of pentachlorophenol and chlorinated
dioxins to sunlight, heat, and combustion, la: Chlorodioxins: origin
and fate. Blair EH, ed. Adv. Chem. Ser. 120:119-125 as cited in Vuceta
et al. (1983).
Svenson A, Kjeller L, Rappe C. 1989. Enzyme-mediated formation of
2,3,7,8-tetrasubstituted chlorinated dibenzodioxins and dibenzofurans.
Environ. Sci. Technol. 23:900-902.
Travis CC, Arms AD. 1988. Bioconcentration of organics in beef, milk,
and vegetation. Environ. Sci. Technol. 22:271-274.
2-32
1583q
-------
USEPA. 1978. U.S. Environmental Protection Agency. Report of the Ad
Hoc Study Group on Pentachlorophenol Contaminants. U.S. Environmental
Protection Agency, Environ. Health Advisory Comm., Sc. Advisory Bd. as
cited by Van Den Berg and 01ie (1985).
USEPA. 1984. U.S. Environmental Protection Agency. Ambient water
quality criteria for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Washington,
DC: U.S. Environmental Protection Agency. EPA 440/5-84-007.
USEPA. 1987b. Office of Solid Waste and Emergency Response. U.S.
Environmental Protection Agency. National dioxin study. Report to
Congress. Washington, DC: U.S. Environmental Protection Agency.
EPA 530/SW-87-025.
Versar, Inc. 1989. Chemistry and fate of dioxins and furans.
Washington, DC: U.S. Environmental Protection Agency, Office of Toxic
Substances. Contract No. 68-07-4254, Task No. 231.
Walters RW, Guiseppi-Elie A. 1988. Sorption of 2,3,7,8-tetrachloro-
dibenzo-p-dioxin from water/methanol mixtures. Environ Sci. Techno!.
22:819-825.
Young AL. 1981. The chlorinated dibenzo-p-dioxins. in: The science of
2,4,5-T and associated phenoxy herbicides. Bovey RW, Young AL, eds., New
York: Wiley and Sons, pp. 133-205 as cited in Vuceta et al. (1983).
2-33
1583q
-------
3. DIOXIN AND FURAN HAZARD ASSESSMENT: HUMANS, TERRESTRIAL AND
AVIAN WILDLIFE, AND AQUATIC LIFE
3.1 Introduction
The human health section of this chapter was compiled from:
Lee CC. 1989. Human health hazard assessment of dioxins/furans.
U.S. EPA, Office of Toxic Substances. Memorandum to L. Dicker, EPA,
Office of Toxic Substances, October 31, 1989.
The ecological effects section was compiled from:
Rabert WS. 1989. Update of aquatic toxicity and bioavailability
data of polychlorinated dioxins and polychlorinated difurans. U.S.
EPA, Office of Toxic Substances. Memorandum to S. Kroner, EPA,
Office of Water Regulations and Standards, July 28, 1989.
The wildlife effects section was compiled from:
USEPA. 1990. United States Environmental Protection Agency.
Assessment of risks from exposure of humans, terrestrial and avian
wildlife, and aquatic life to dioxins and furans from disposal and
use of sludge from bleached kraft and sulfite pulp and paper mills.
Washington, DC: Office of Toxic Substances and Office of Solid
Waste. EPA 560/5-90-013.
3.Z Human Hazard Assessment of PCDDs and PCDFs
The USEPA (1984a, 1985), the Ontario Ministry of the Environment
(MOE) (1984), and other organizations have compiled and evaluated the
existing toxicological data on PCDDs and PCDFs. Although there is
extensive literature on a few of these isomers, the toxicological
information on these families of more than 200 compounds (75 PCDD and 135
PCDF isomers) is far from complete. Nevertheless, a growing body of
information on mechanisms of action and structure-activity relationships
(SARs) within these families of compounds makes it possible, with
reasonable confidence, to infer information where data are missing.
Among the 210 congeners of PCDDs and PCDFs, the compound that appears
to be the most toxic and has generally raised the greatest health
concerns is 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD).
Experimental studies with 2,3,7,8-TCDD in animal systems have
demonstrated a variety of toxic effects resulting from exposure to this
compound (USEPA 1985). These effects include carcinogenesis, cancer
promotion, reduced fertility and postnatal survival, teratogenic effects,
immunotoxic effects, thymic atrophy, liver damage, effects on the
thyroid, and.chloracne and other effects on the skin. Acute exposures of
sensitive species of animals to 2,3,7,8-TCDD result in a characteristic
3-1
1580q
-------
"wasting syndrome," followed by death. Extensive experimental studies
have revealed marked variations among species in both the array of
effects caused by 2,3,7,8-TCDD and the dose levels at which these effects
are elicited (USEPA 1985, Pitot et al. 1986). Limited toxicological
testing of other PCDDs and PCDFs has demonstrated that several of these
compounds cause similar toxicological effects, but that higher doses of
these compounds are generally required to cause effects of comparable
magnitude to those induced by 2,3,7,8-TCDD.
The nature and extent of effects in humans exposed to 2,3,7,8-TCDD
are not nearly so well defined (Ontario MOE 1984, USEPA 1985, Pitot et
al. 1986). There is a consensus that exposure of humans to 2,3,7,8-TCDD
can result in a skin condition known as chloracne, an acne-like lesion
which, while not life-threatening, can be disfiguring, persistent, and
refractory to treatment.
The evidence for an association of 2,3,7,8-TCDD with human cancer is
equivocal. Several studies of human populations exposed to chemical
mixtures containing 2,3,7,8-TCDD have suggested increased frequencies of
certain cancers (Hardell and Sandstrom 1979, Hardell et al. 1981, Thiess
et al. 1982, MDPH 1983b), but inconsistencies among the studies and
incomplete characterization of exposure make the evidence inconclusive
(USEPA 1985, Blair 1986). Evidence for reproductive impairment in humans
exposed to 2,3,7,8-TCDD, including one study conducted in Midland County,
Michigan (MDPH 1983a), is inconclusive for similar reasons (USEPA 1985,
Kimbrough 1986). Other effects in humans that have been more clearly
associated with exposure to mixtures containing 2,3,7,8-TCDD include
disturbances in lipid metabolism (Moses et al. 1984, Suskind and
Hertzberg 1984) and increased frequency of gastric ulcers (Bond et al.
1983, Suskind and Hertzberg 1984).
For the PCDFs, compared with the PCDDs, there is more specific and
quantitative information available on toxic effects in humans as a result
of two large-scale poisoning incidents in Japan and Taiwan (Kuratsune and
Shapiro 1984). Over periods of weeks to months, the affected persons
ingested food contaminated with a mixture of PCDFs, polychlorinated
biphenyls (PCBs) and polychlorinated quaterphenyls (PCQs). Comparative
toxicological studies have indicated that PCDFs were the primary toxic
agents in these poisoning incidents and that 2,3,4,7,8-PCDF was probably
the most important single compound (Bandiera et al. 1982, Masuda and
Yoshimura 1984, Kunita et al. 1984, 1985, Masuda et al. 1985, Chen et al.
1985, Miyata et al. 1985). The most prominent toxic signs were skin
eruptions similar to those of chloracne, along with skin pigmentation and
eye abnormalities (Lu and Wong 1984, Urabe and Asahi 1985). Other
effects reported include impairments in lipid metabolism and immune
function (Okumura et al. 1974, Chang et al. 1982a,b) and persistent
respiratory symptoms (Nakanishi et al. 1985). A preliminary report by
Kuratsune et al. (1987) indicated a significant excess frequency of liver
cancer deaths (9 observed vs. 1.6 expected) and possibly lung cancer
3-2
1580q
-------
among male victims within 15 years after exposure. However, the excess
incidence of liver cancer deaths had an uneven geographic distribution,
and four of the deaths occurred relatively soon after the poisoning
incident. In addition, potent confounding variables, such as alcohol
consumption, were not assessed. Reported effects on reproduction include
menstrual disturbances (Kusuda 1971), skin hyperpigmentation in infants
(Yamashita and Hayashi 1985, Hsu et al. 1985), and perinatal mortality
(Hsu et al. 1985). These effects observed in humans are qualitatively
similar to those reported in animals exposed to PCDFs and PCOOs (McNulty
1985); this provides support for the use of animal data as the basis for
hazard assessment for other members of these families of compounds.
3.3 Human Health Hazard of 2.3.7.8-TCDD
3.3.1 Cancer Effects
2,3,7,8-TCDD is classified by the EPA as a "probable human
carcinogen" (Category B2, according to the EPA (1986) classification
scheme), based on "sufficient" evidence in animals but "inadequate"
evidence in humans (USEPA 1988a). 2,3,7,8-TCDD is the most potent animal
carcinogen which has been evaluated by EPA (USEPA 1985). Very low oral
doses of 2,3,7,8-TCDD have increased the incidence of tumors at a variety
of sites in rats, mice, and hamsters (USEPA 1985, Rao et al. 1988),
principally and most consistently in the liver. As discussed below, the
tumor data for the female rat (Kociba et al. 1978) and the male mouse
(NTP 1982) were used in estimating the increased cancer risk from
exposure to 2,3,7,8-TCDD. In a dietary study (Kociba et al. 1978),
2,3,7,8-TCDD caused significantly increased incidences of tumors in the
liver (neoplastic nodules/heptaocellular carcinoma), lung (keratinizing
cell carcinoma), and nasal turbinates/hard palate (squamous cell
carcinoma) in female Sprague-Dawley rats. In a National Toxicology
Program (NTP 1982) gavage study, the incidence of hepatocellular
carcinoma was significantly increased in male B6C3F1 mice.
USEPA, U.S. Food and Drug Administration (FDA) and U.S. Consumer
Product Safety Commission (CPSC) each derived a cancer risk estimate for
2,3,7,8-TCDD (USEPA 1985, 1988b, FDA 1983, Babich 1988). The risk
estimates differ by as much as a factor of 10 (EPA vs. FDA). This range
reflects the current lack of consensus among the agencies regarding
approaches to carcinogen risk assessment and dose-response modeling. All
three agencies based their estimates on a multistage model with
linear-at-low-dose extrapolation procedures, but differed with respect to
selection of animal data and details of extrapolation, Table 3-1 shows
the factors used by each agency and the resulting risk estimates. The
dose-response modeling terminology used in the following discussion of
the cancer risk estimates is defined in Table 3-2.
EPA employed the linearized multistage (LMS) procedure to estimate a
plausible upper bound slope factor, designated as qj* for carcinogens.
3-3
1580q
-------
Table 3-1. Factors Used by EPA. FDA. and CPSC In Calculating Their Risk Estimates for Z.3.7.8-TCDO Using Multistage Models
Factor
EPA
FDA
CPSC
Effect of difference on upper-bound
unit risk estimates (EPA vs FDA only)
LO
I
AfHnal study used
Pathologist (Kociba or Squire)
Adjustment for early Mortality
Selection of tuaor types
Kociba female rat feeding
study
Both
Tes
Liver, lung, hard palate/
nasal turbtnates
AniMal-to-«an dose equivalence Dose/surface area
Kociba female rat
feeding study
Kociba
No
Liver only
Dose/body weight
DTP.Male Mouse gavaoe
study
NA
HA
Liver only
Less than 10X
Adjustment changes estimates by a factor of
+1.7 or 1/2.6*
Less than 10X
Dose/surface area Dose/surface area Increases estimate by •
factor of 5.38
Dose used for curve fit
Extrapolation curve
Slope factor*1 (pg/kg-day)"1
Dose at risk of 10~6; units of
pg/kg/day
Administered
Upper-bound
I.6xHf4
0.006
Administered
Upper-bound
l.BxlO"5
0.060
Administered
Maximum likelihood
estimate
6.7jclO~5
0.015
* Adjustment Increases by a factor of about l.T compared with unadjusted Kociba anlaysis; adjustment decreases estimate by a factor of 2.6 compared with
unadjusted Squire analysis where high dose is excluded due to poor fit.
b The slope factors should not be used if the exposure (in pg/kg/day) exceeds 63 for EPA (556. FDA; 149. CPSC). since above these exposures the slope
factor may differ from that stated. These exposures are associated with a risk of 1 in 100 or greater.
Source: Adapted from USEPA (198on).
-------
8672h
Table 3-2. Terms Associated with Dose-Response Modeling
Modeling
Coefficient (or parameter)
Maximum likelihood estimate (MLE)
*»*-spectf ic dose (RsD)
Slope factor
unit risk
An estimation of the incremental or extra cancer risk to humans arrived at
by fitting a Mathematical function to animal response data.
A constant in the model, associated either vith the control response, or
the time or dose variable Inputs. For example, in the Crump linearized
multistage procedure, the coefficient associated with the variable that is
linear in dose is denoted as q..
The estimate from a statistical procedure by which the coefficients of
the model are estimated to fit the observed data. The NLE has many
properties, in a statistical sense, which allow it to be referred to as a
"statistical average" or "best" estimate.
In the Crump multistage procedures, qx is the coefficient associated
with the variable that is linear in dose. QJ can be defined as the
increase in cancer risk associated with an incremental increase per unit
of dose. For this reason. QJ is expressed In units of reciprocal dose
such as (ng/kg-day) .
In the Crump multistage procedure QJ* is the upper bound estimate
associated with the linear term. QJ* is expressed in units
of reciprocal dose such as (ng/kg/day)"1. The upper bound increased
lifetime cancer risk associated with a given exposure can be estimated by
multiplying the qj* (ng/kg/day)"1 times the exposure (ng/kg/day).
A dose (loosely, an exposure) associated with a specified cancer risk.
For example, suppose a linearized multistage procedure is used with some
data and the coefficient estimates are qt « 3.0 x 10"3 (ng/kg-day)"1
and q,* - 7.5 x 10"3 (ng/kg-day)'1. Then for an extra risk of 1 in
1,000.000. the dose (d) would be the solution to 10"6 - l-exp(-7.5 x
10~3 d). and RsD - 1.3 x 10~4 ng/kg-day would be called its risk-
specific dose. If the MLE is used Instead of the upper bound (as by
CPSC), then q, is used instead of qj* and the solution here would be
RsD - 3.3 x 10~4 ng/kg-day.
The slope of the upper bound dose extrapolation model at doses approaching
zero. Also called qj*.
The incremental upper bound lifetime risk estimated to result from life-
time exposure to an agent if it is in the air at a concentration of
1 microgram per cubic meter or In the water at a concentration of 1 micro-
gram per liter.
3-5
-------
8672H
Table 3-2. (continued)
Upper-confidence bounds (UBJ The estimates resulting from a statistical procedure in which the upper-
limit values of the coefficients still consistent with the data are esti-
mated. In the linearized suitistage procedure (Crunp). the upper lisnt
associated with the linear tem is designated qj*. It is selected to
be the 95X upper-li»it estimate, in a statistical sense, of the linear
ten* associated with the fitting of a Multistage Model to the inIMS! data.
In Making cross-species extrapolations to hununs, however,the "9HT label
is dropped, since the uncertainty associated with cross-species
extrapolations is considered far greater than the statistical uncertainty
associated with the Model-fitting procedure.
3-6
-------
FDA used a similar extrapolation procedure to estimate the upper bound
slope factor. CPSC used the maximum likelihood estimate from the
multistage model, since it, as well as the upper bound, is linear at low
dose for the data set used. CPSC also differed with EPA and FDA in the
selection of animal data. CPSC used the liver tumor data for male mice
from the NTP (1982) gavage study, whereas EPA and FDA used the tumor data
for female rats from a dietary study (Kociba et al. 1978). EPA and FDA
differed in their selection of tumor types in female rats. EPA combined
all tumor types of significantly increased incidence, whereas FDA
considered only liver tumor incidence. EPA made two adjustments in the
data not made by FDA. EPA adjusted for high early mortality in high-dose
animals by excluding animals dying during the first year. EPA also
incorporated two pathologists' reviews of the liver slides by using the
geometric mean of the slope factors derived from each pathologist's
analysis, while FDA considered only the original pathologist's review of
the slides. While these differences in data selection account for some
of the difference in the EPA and FDA risk estimates (Table 3-1), the
major contributing factor was choice of interspecies dose scaling
factor. EPA used the surface area correction, whereas FDA used dose per
body weight. When extrapolating from rat to man, the use of dose per
surface area versus dose per body weight increases the risk estimate by a
factor of 5.4. CPSC, which used the male mouse data, applied the surface
area correction, but the CPSC selection of extrapolation procedures
cannot be compared quantitatively with that of the other agencies because
of the use of different data sets.
USEPA estimated a plausible upper bound slope factor (qi*) of 1.6 x
10'* (pg/kg-day)'1 for 2,3,7,8-TCDD. This qi* gives an exposure
(also called risk-specific dose, see Table 3-2) of 0.006 pg/kg/day at an
upper bound increased lifetime cancer risk.of one in a million (USEPA
1985). The FDA derived a qj* of 1.8 x 10"5 (pg/kg-day)'1, which
gives an exposure of 0.06 pg/kg/day for an increased upper bound lifetime
cancer risk of one in a million (USEPA 1988a). CPSC calls for the use of
the maximum likelihood estimate (MLE) of the unit risk; for humans this
is 6.7 x 10'D (pg/kg-day)'1. Thus, the exposure (here also an MLE)
at which the extra lifetime cancer risk is one in a million was estimated
by CPSC to be 0.015 pg/kg/day (Bablch 1988).
3.3.2 Non-Cancer Effects
In assessing the risk of non-cancer effects, USEPA is concerned about
two aspects of exposure, the dose level and the duration. Adverse
effects could result from long-term, low-level exposure, or from a
relatively brief exposure to a high dose. For evaluating long-term
exposures, USEPA has adopted the concept of the Reference Dose (RfD)
(USEPA 1987a). The RfD is defined as an estimate of the lifetime daily
exposure to the human population likely to be without any appreciable
risk of a deleterious effect. For assessing brief exposures to high
levels, USEPA has adopted procedures to establish exposure levels for
3-7
ISSOq
-------
issuing "health advisories" (HA) (USEPA 1987b). Thus, the HA dose level
is appropriate for comparison with single dose or single-day intakes, or
short-term exposure, whereas the RfD is appropriate for comparison with
long-term or lifetime exposures. Both values are operationally derived
from the "No Observed Adverse Effect Level (NOAEL)" determined in the
critical toxicological study, divided by an "Uncertainty Factor (UF)"
selected on the basis of specific attributes of the study.
In a previous risk assessment, EPA has determined that reproductive
effects and developmental toxicity in animals are the most critical or
sensitive noncarcinogenic effects to consider for risk assessment of
2,3,7,8-TCDO (USEPA 1988a). A number of studies have been conducted to
assess the possible.adverse effects of chronic 2,3,7,8-TCDD exposure on
reproductive capacity. These studies have shown that exposure to
2,3,7,8-TCDO prior to mating, as well as during pregnancy, results in
reduced fertility, fetotoxicity, and/or reduced survival of the
offspring. A three-generation rat reproductive study (Murray et al.
1979) was selected as the "critical" study for determining the RfD for
low-level, long-term exposure to 2,3,7,8-TCDD. A combined UF of 1000 was
selected, which includes three subfactors of 10: one subfactor of 10
because the lowest administered dose of 1 ng/kg/day was not a NOAEL, the
second subfactor of 10 to account for possible interspecies differences
in sensitivity, and the third subfactor to account for possible
intraspecies differences in sensitivity. Thus, for purposes of this risk
assessment an estimated RfD of 1 pg/kg/day was calculated by dividing the
lowest dose administered (1 ng/kg/day) by the UF of 1000.
A number of animal studies have been conducted to assess the
potential developmental toxicity of short-term 2,3,7,8-TCDD exposure
during pregnancy (USEPA 1985). These studies clearly demonstrated that
exposure to 2,3,7,8-TCDD during pregnancy is teratogenic and/or fetotoxic
in all animal species tested. The most common malformations observed in
mice after 2,3,7,8-TCDD exposure are cleft palate and kidney anomalies;
while edema, hemorrhage, and kidney anomalies are most commonly observed
in rats. At higher doses, 2,3,7,8-TCDD has marked fetotoxic effects in
both rats and mice, as measured by decreased fetal body weight and
increased fetal death. Adverse effects have also been demonstrated in
nonhuman primates (Bowman et alI. 1989a, b).
For purposes of this risk assessment, a rat teratology study
(Sparschu et al. 1971) was selected as the "critical" study for
estimating a HA. The NOAEL in this study was 30 ng/kg/day and an UF of
100 was used, which includes two subfactors of 10, one each to account
for possible interspecies and intraspecies differences in sensitivity;
these data yield an estimated HA of 300 pg/kg/day.
Since the estimated RfD and HA for noncarcinogenic effects are
derived from reproductive and teratology studies, some uncertainty arises
as to whether these values are applicable to people of nonreproductive
age (e.g., children or post-menopausal women), or people who are not
reproducing for other reasons. The RfD is probably applicable to the
3-8
1580q
-------
general population since 2,3,7,8-TCDD is stored for periods of years in
the body fat, and hence may exert effects long after the exposure
occurred. To derive the HA which could be applied directly to any
population except pregnant women, the data selected were from an acute
toxicity study in which hepatic effects were the sensitive endpoint
(Turner and Collins 1983). The LOAEL for hepatic lesions in this study
was 100 ng/kg/day. For purposes of this risk assessment, a one-day HA of
100 pg/kg/day was calculated, using procedures for issuing HA (USEPA
1987b), by dividing the LOAEL by the UF of 1,000. A 10-day HA of 10
pg/kg/day was calculated by dividing the LOAEL by an additional UF of 10
(a total UF of 10,000).
3.3.3 Toxicity Equivalence Factors
The toxicity equivalency factor (TEF) method is an interim procedure
for assessing the risks associated with exposures to complex mixtures of
chlorinated dibenzo-p-dioxins and dibenzofurans (USEPA 1989). The method
relates the toxicity of the 210 structurally related chemicals and is
based on a limited data base of jn vivo and in vitro toxicity testing.
By relating the toxicity of the 209 PCDDs and PCDFs to the highly-studied
2,3,7,8-TCDD, the approach simplifies the assessment of risks involving
exposures, to mixtures of PCDDs and PCDFs.
Little is known about the toxicity of the various PCDDs or PCDFs. Of
the limited number of these congeners tested thus far, other than
2,3,7,8-TCDD, only a mixture of 1,2,3,6,7,8- and 1,2,3,7,8,9-HxCDD has
been tested for carcinogenicity and liver tumors in rats and mice when
administered at low doses for a lifetime (NCI 1980). A larger body of
data is available on short-term in vivo studies and a variety of In vitro
studies for a number of PCDDs and PCDFs (NATO-CCMS 1988b). These
experiments cover a wide variety of end points (e.g., LDsg, body weight
loss, thymic atrophy and dysfunction of the immune system, cell
transformation, the enzyme aryl hydrocarbon hydroxylase (AHH) Induction,
and reproductive and/or teratogenic effects). While the doses necessary
to elicit the toxic responses differ in each case, the relative potencies
of the different compounds (compared to 2,3,7,8-TCDD) are generally
consistent from one end point to another.
This toxicity information, developed by researchers in several labs
around the world, reveals a strong structure-activity relationship
between the chemical structure of a particular PCDD or PCDF congener and
its ability to elicit a biological/toxic response in various in vivo and
in vitro test systems (Bandiera et al. 1984, Bellin and Barnes 1987,
NATO-CCMS 1988a,b, Olson et al. 1989). Research has also revealed a
mechanistic basis for these observations. That 1s, a necessary (but not
sufficient) condition for expression of much of the toxicity of a given
PCDD or PCDF congener is its ability to bind with great specificity to a
particular protein receptor located in the cytoplasm of the cell. This
congener-receptor complex then migrates to the nucleus of the cell, where
3-9
1580q
-------
it initiates reactions leading to the expression of toxic effects (Poland
and Knutson 1982).
To assess the risks of mixtures of PCDDs and PCDFs, the TEF approach,
first suggested in 1977 (Grant 1977), was pursued by scientists and
jurisdictions during the early- and mid-1980s (Ontario MOE 1982, Swiss
Government 1982, Commoner et al. 1984, Eadon et al. 1986, California Air
Resources Board 1986). As a result, numerous slightly different TEFs
existed, thus complicating the communication among scientists and
agencies in addressing the toxicological significance of complex mixtures
of PCDDs and PCDFs.
An international effort was conducted under the auspices of the North
Atlantic Treaty Organization's Committee on Challenges of Modern Society
(NATO-CCMS) with participation by scientists from United States and other
countries. The International group adopted a common set of TEFs, the
International TEFs/89 (I-TEFs/89) (NATO-CCMS 1988a,b) to promote
consistency in addressing contamination involving PCDDs and PCDFs. USEPA
has adopted the I-TEFs/89 as a revision to the EPA-TEFs/87 currently in
use for assessing risks of PCDDs and PCDFs other than 2,3,7,8-TCDD (USEPA
1989). The I-TEF is used to estimate carcinogenic and noncarcinogenic
effects of PCDD/PCDF mixtures in terms of an equivalent amount of
2,3,7,8-TCDD, as if they were concentrations of 2,3,7,8-TCDD itself. The
I-TEFs and TEFs used by other groups are presented in Table 3-3.
Relative potency data for the PCDD and PCDF isomers substituted in the
2,3,7,8-positions are shown in Table 3-4.
The International Report by NATO-CCMS (1988a) identified limitations
to the TEF approach, e.g., the extrapolation from short-term to long-term
effects and the possible differences in metabolic effects among species.
For example, many of the short-term results seen in the metabolic systems
of mice have not been observed in similar systems in rats. Also, the
connection between the enzyme induction response (which supports several
of the TEF values) and several of the toxic end points manifested by
PCDDs/PCDFs is unclear. Other mechanisms of action (e.g., effect on
vitamin A synthesis and estrogen-like activity) have been suggested as
playing an important role in the toxicity of PCDDs/PCDFs. Thus, the TEF
approach remains interim in nature and will be revised as new data are
developed.
It should also be noted that the accuracy of an assessment of the
risks associated with exposure to a mixture of PCDDs and PCDFs depends
upon a number of factors, of which the uncertainty of the TEF approach is
only one and perhaps not the largest. The uncertainties relating to
estimated intakes, bioavailability, interspecies extrapolation, safety
factors or mathematical models, and possible antagonistic or synergistic
interactions are likely to carry as much or more uncertainty than the TEF
values themselves.
3-10
1580q
-------
Table 3-3. f-TFFs/89 and TEFs Developed by Other Groups
Basis
compound
(Basis)
Mono thru di COOS
Tri COOS
2.3.7,8-TCDO
Other TCDDS
2.3.7.8-PeCDOs
Other PeCOOs
2.3.7.8-HxCDOs
Other HxCDOs
2.3.7.8-HpCODs
Other HpCOOs
OCOD
2.3.7.8-TCDFs
Other TCDFs
2.3.7.8-PeCDFs
Other PeCDFs
2.3,7.8-HxCDFs
Other HxCOFs
Swiss
Government3
Enzyme
0
0
1
0.01
0.1
0.1
0.1
0.1
0.01
0.01
0
0.1
0.1
0.1
0.1
O.I
0.1
Grantb
Oliec
Commoner
0
0
1
1
0.1
0.1
0.1
0.1
0.1
0.1
0
0.1
0.1
0.1
0.1
0.1
0.1
New
York
State6
LD50
0
0
1
0
1
0
0.03
0
0
0
0
0.33
0
0.33
0
0.01
0
Ontario'
Various
effects
0
1
1
0.01
1
0.01
1
0.01
1
0.01
0
0.02
0.0002
0.02
0.0002
0.02
0.0002
FDA9
Various
effects
0
1
0
0
0
0.02
0.02
0.005
0.005
<0. 00001
0
0
0
0
0
0
CAh
0
0
1
0
1
0
1
0
1
0
1
1
0
1
0
1
a
EPA
1981*
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
EPA
current
recaimendat ion
Various
effects
0
0
1
0.01
0.5
0 005
0.04
0 0004
0.001
0.00001
0
0.1
0.001
0.1
0.001
0.01
0.002
I-TEFs/89
Various
effects
0
0
1
0
0.5
0
0.1
0
0.01
0
0.001
0.1
0
0.05.0.5^
0
0.1
0
-------
867 2H
Table 3-3. (continued)
Co
t
ro
6rantb
Basis Swiss Oliec
confound Government3 Conwoner
(Basis) Enzyme
2.3,7.8-HpCDFs 0.1 0.1
Other HpCDfs 0 0.1
OCOF 0 0
nGW
York
State6 Ontarif/ FDA9
Various Various
L050 effects effects
0 0.12 0
0 0.002 0
00 0
EPA
EPA current
CAh 1981 ' recomnendation I-TEFs/89
Various Various
effects effects
1 0 0.001 0.01
0 0 0.00001 0
000 0.001
aSwiss Government 1982.
Vant 1977.
C0lie et al. 1983.
et al. 1984.
eEadon et al. 1986.
f Ontario HOE 1982.
9U.S. DHHS 1983.
^ravitz et aV 1983.
^.S. EPA 1981.
j 0.05 for 1.2,3.7,8-PeCOF. 0.5 for 2.3.4.7,8-PeCDF
-------
TaS10 3-a_ Intel-national To»l«=tt» EQul
Fa€=tor»/aS |I-TEFa/B9|:
•Ison of Rel<
latlve Potency Data for the Z.3.7,B-Sutost Ituted COOs and COFs
Congener l-TEFs/89
2.3.7.8-TCOD 1
1.2.3.7.8-PCDO 0.5 Range
Data
1.2.3.4.7,8 HxCDO 0.1 Range
Data
1.2.3.6.7.B-HxCOO 0.1 Range
Data
^ 1.2.3.7.8.9-HxCOD 0.1 Range
^ Data
i__*
co
1.2.3.4.6.7.8-t*>CDD 0.01
OCOD 0.001
2.3.7.8-TCDF 0.1 Range
Data
2.3.4,7.8-PCOF 0.5 Range
Data
Observer
(In vivo toxicities)
1
= 0.053 - 0.59
= (0.59*1. 0.42s1. 0.081r. 0.053r)
= 0.018 - 0.24
= (0.24*1. 0.084r, 0.0189'. 0.13r)
= 0.015 - 0.16
= (0.16*1. 0.015gl)
= 0.016 - 0.14
= (0.01691. 0.14"1)
~
-
= 0.016 - 0.17
= (0.01791. 0.17*1. 0.05"1. 0.0254.
0.016r)
= 0.048 - 0.80
= (0.8*1. 0.479. 0.43r. 0.139\ 0.12*.
1 TEF ranges
Aryl hydrocarbon hydraxvlase (AHH) induction
. (in vivo) (in vitro)
1 1
0.13 Range = 0.0065 - 0.011
(0.13r) Data = (0.011/0. 0065h)
0.13 Range = 0.034 - 0.046
(0.13r) Data = (0. 034/0. M6h)
0.012
(0.012**)
0.008
(0.008**)
0.003
(0.003**)
0.0002 0.0006
(0.0006**)
0.006 Range = 0.018 - 0.09
(0.006r) Data = (0. 018/0. 09h)
0.11 - 0.23 Range = 0.28 - 1.41
(0.239. O.llr) Data = (0.28/1.41*1)
0.048r)
1.2.3.7.8-PCOF
0.05
Range = 0.019 - 0.95
Data = (0.959. 0.05r. O.OSl"1. 0.0l9r)
0.003 - 0.047
(0.0479. 0.003r)
Range = 0.028 - 0.06
Data = (Q.06/0.028h)
-------
ttbtZH
Table 3-4. (continued)
CO
Congener
l.2.3.6.7.B-H*COF
Observed TEF ranges
I-TEFs/89
(In vivo toxicities)
•1
•t
•I
b
hfi
0.1
Range = 0.016 - 0.097
Data •= (0.097r. 0 016r)
flryl hydrocarbon hydroxvlase (AHH) Induction
(In vivo) (in vitro)
1.2.3.4.7.R-HXCOF
0.1 Range = 0.013 - 0.18
Data = (0.16r. 0.03Br.
0.013"*)
0.014
(0.014r)
Range = 0.20 - 0.50
Data = (0.20/0.50h)
0.01Z
9 guinea pig data
r rat data
guinea pig lethalities
•nine lethalities
muse teratogenicity
•ouse iajMunotoxiclty
rat hepatow daU (ratio of activity of aryl hydrocarbon hydroxylase (AHH) to activity of ethoxyresorufin 0-deethylase)
rat hepatwa data (activity of AHH|
Range = 0.049 - 0.153
Data = (0.049/0.153h)
1.2.3.7.8.9-H*COF
2.3.4.6.7.8-lfeCDF
1 t 3 4 6 7 B'HCOF
I.2.3.4.7.8.9-HpCOF
OCDF
0.1
0.1 Range = 0.011 - 0.097
Data = (0.097rr O.OI8r. 0.01I9)
O.I
0.01
0.001
-
0.015 Range = 0.11 - 0.33
(0.0151") Data = (0.1I/0.33hJ
_ _
-
-
Source: MfCHXMS <19B8b).
-------
3.4 Ecological Hazard of 2.3.7.8-TCDD and 2.3.7.8-TCDF
Based on the available data, the most toxic congener of PCODs
appears to be 2,3,7,8-TCDD. Many aquatic tests have been conducted with
2,3,7,8-TCDD, but most of these studies were either bioavai lability tests
or short-term exposure studies with long post-exposure observation
periods. The available test data on 2,3,7,8-TCDD indicate that none of
the studies are adequate to define an acceptable, no-observed-effect
concentration (NOEC). Only one study exposed a fish to 2,3,7,8-TCDD in
water for a reasonable duration, 28 days. Even then, the exposure
duration in the rainbow trout study was only about half of the 48 days
estimated to be required to achieve 90 percent of the steady-state BCF
level in fish. The NOEC was less than 0.038 ng/1, a concentration which
produced 45 percent mortality in rainbow trout fry.
The mode of action of 2,3,7,8-TCDD is unknown, but it appears related
to neural toxicity and may affect the immune system, like many other
organochlorine chemicals. The pattern of TCDD- induced mortality
typically occurs between 30 to 80 days after the initial exposure, even
for exposure as short as 6 hours. Mortality in fish appears to be a
function of both exposure duration and the test concentration. None of
the bioconcentration tests were of sufficient duration to achieve a
steady-state bioconcentration factor (BCF). All BCF values have been
estimated using the uptake rate (kj) and depuration rate (k2) values.
Many sublethal effects have been reported for polychlorinated dioxins
and furans. 2,3,7,8-TCDD elicits a broad range of toxic effects which
include: reduced growth; weight loss; abnormal hatching; cellular
alterations; numerous types of lesions in a broad spectra of organs; and
behavioral effects on swimming, reduced feeding, and loss of response to
external stimuli.
In a risk assessment, the normal procedure is to identify the NOEC
for the most sensitive toxicological endpoint and apply a safety margin
(e.g., 100) to that NOEC. Unfortunately, the lowest tested concentration
(0.038 ng/1) among all known 2,3,7,8-TCDD studies produced 45 percent
mortality. Consequently, a definitive NOEC for 2,3,7,8-TCDD has not yet
been determined. All studies that have been completed are of insufficient
duration to determine the full extent of TCDD effects. For purposes of
this assessment, it was determined that the margin of safety for TCDD
should be 1000. This factor 1s justified for use in this assessment
because of the high level of lethality found at 0.038 ng/1, because of
the exposure duration which was too short for steady-state equilibrium,
and because of the absence of exposure to pre- spawn ing adults and
TCDD-deposition in egg yolk, which generally yields the most sensitivity
life stage for chlorinated chemicals.
3-15
1580q
-------
3.4.1 Aquatic Toxicity of PCDDs
The available studies on PCDDs and PCDFs published after 1984 were
collected and reviewed in this document to update the aquatic toxicity
literature published subsequent to the comprehensive EPA reviews in the
ambient water quality criteria document (USEPA 1984a) and the health and
environmental effects profile (USEPA 1984b). Thus, the information
contained herein does not include a detailed review of earlier aquatic
toxicity data. Test concentrations and effects are summarized for PCDDs
in Table 3-5 and for PCDFs in Table 3-6.
It appears that the most toxic PCDD is 2,3,7,8-TCDD. While several
methods for calculation of toxic equivalence have been developed for
2,3,7,8-TCDD and other dioxin congeners, the methods have substantially
different toxicity factors and therefore have considerably different
outcomes (Helder and Seinen 1985).
Moreover, PCDDs appear to be more toxic to fish than to other aquatic
species. However, aquatic invertebrates also show toxic effects,
including oligochaetes, pulmonate snails, and mosquito larvae exposed to
2,3,7,8-TCDD at 0.2 ppb for 55, 36, and 36 days, respectively (Miller
et al. 1973).
Aquatic studies with 2,3,7,8-TCDD have generally been conducted
either as continuous long-term exposures or short-term exposures (6 to 96
hours) with long-term observation periods. The long observation periods
are necessary to monitor the delayed mortality which is characteristic of
2,3,7,8-TCDD. Most of the lethal effects after short-term exposures are
reported to be delayed (e.g., for from 30 to 50 or more days following
exposure). The occurrence of delayed effects has some serious
implications for both hazard evaluations and risk assessments. For
example, the reported results of typical acute toxicity tests are suspect
because of the short observation periods. In risk scenarios, sporadic
acute releases may cause serious delayed effects. Of most concern is
that there are insufficient data to assess the effects from the many
complex combinations of exposure concentrations and the various durations
of the exposure that occur in nature.
Delayed mortality found in dioxin studies indicates that the most
appropriate method of reporting the effects of dioxin is not a simple
96-hour 1X50 value or a maximum acceptable toxicant concentration
(MATC) range. The toxicity of 2,3,7,8-TCDD is a function of the duration
of the exposure, as well as the dose level. At any given test
concentration, continuous exposures appear to produce toxic effects in a
shorter time than do short-term exposures. Table 3-5 summarizes dioxin
effects from various routes of exposure, including water, sediment, diet,
and intraperitoneal injection. Based on the data in Table 3-5, exposure
via water appears to be the most toxic, single route of exposure;
however, it is likely that the degree of an effect could be increased by
3-16
1580q
-------
Duration (days)
Cheaical fora Test species
Nonochloro- No data
Dichloro- No data
Trlchloro- No data
Tetrachloro-
1.2.3.7-TOJO Rainbow trout
(fry - 0.2 g)
Fathead Minnow
(0.2 g)
^ Rainbow trout
-1 (0.5 - 1.09)
Fathead Minnor
(1.0 g)
1.3.6.B-TCOD Rainbow trout
(0.1 - 0.5 g)
Fathead Minnow
(1.5 -2.5g)
2.3.7.8-TCDD Fathead Minnow
Fathead Minnow
Exposure
5
(water)
5
(water)
30
(diet)
30
(diet)
5
(water)
5
(water)
20
(water)
1.2.3.4
(water)
Concentration
/ Observation (ng/1)
/ +24 days 134
54
/ +24 days 28
23
/ +30 days 110 ng/g
/ +30 days 110 ng/g
/ +48 days 211
+ 48 days 74
+ 48 days 4
/ +29 days 41
+ 48 days 10
/ +20 days 1.0
/ +60 days 0.0
0.12
Toxic effects
No deaths
No deaths
No deaths
No deaths
No significant (P = 0.05)
adverse effects
No significant (P = 0.05)
adverse effects
No deaths
No deaths
No deaths
No deaths
No deaths
3 out of 30 dead
10X dead
<5X dead
Reference
Nuir et al. (1985)
Nuir et al. (1985)
Nuir et al. (1988)
Nuir et al. (1988)
Nuir et al. (1986)
Nuir et al. (1986)
Adams et al. (1986)
Adams et al. (1986)
-------
Table 3-5. (continued)
CA>
I—>
00
Choaical fora lest species
2.3.7.8-TCOO Fathead MinnoM
(continued) (28-day LCg,
1-7 ng/1)
Coho Gabon
Rafnfaow trout
(359)
Rainbow trout
(yolk sac fry)
Rainbow trout
(fry - 0.38 g)
Duration (days)
Exposure / Observat ion
28 / + 20 days
(water)
1 /
(Mater)
0.25 / +139 days
(water)
4 / * 16 days
(water)
28 1 * 28 days
(water)
Concentration
(ng/1)
1.7
6.7
63
0.056
107
54.6
33.1
20.1
7.4
4.5
2.7
1.6
0.789
0.382
0.176
Toxic effects
53X dead (28 days)
100X dead (22 days)
LOOK dead (12 days)
Deaths after several weeks
4 deaths, one each on
days 78. 136. 137. 139
and decreased growth
1001 dead (13 days)
100X dead (15 days)
IOOX dead (17 days)
95X dead (20 days)
63X dead (20 days)
20X dead (20 days)
n dead (20 days)
851 dead (28 days)
731 dead (28 days)
SOX dead (28 days)
Reference
Adans et al. (1986)
Miller et al. (1973)
Branson et al. (1985)
Helder and Seinen (1985)
Mehrle et al. (1988)
Daphnia
(48-hr ECM
>1.030 ng/1)
2
(water)
/ * 7 days
0.079
0.038
0.0011 (control)
0.2 - 1.030
95X dead (56 days)
18X dead (28 days)
83X dead (56 days)
6X dead (28 days)
45X dead (56 days)
5X dead (28 days)
71 dead (56 days)
No Mortality
Mms et al. (1986)
-------
Table 3-i». < con* 1 nu
CO
t—>
ID
Duration (days)
Chemical form Test species Exposure / Observation
2.3,7.8-TCDD Rainbow trout 91 / + 91 days
(continued) (diet)
Yellow perch 91 / + 91 days
(diet)
Rainbow trout 1 dose / +14 days
intraperi-
tonea 1
injection
(i.p.)
Rainbow trout 1 dose / +80 days
(LC50 - 10 ug/kg) (i.p.)
Rainbow trout 1 dose / +80 days
(LD50 - 3 ug/kg) (i.p.)
Carp 1 dose / +80 days
(LD50 - 3 ug/kg) (i.p.)
Bullhead 1 dose / +80 days
(LD50 - 5 ug/kg) (i-P-)
Concentration
(ng/1)
494
494
10 ng/kg
1.0
0.1
0.01
125
25
5
1
125
25
5
1
125
25
5
1
125
25
5
1
Toxic effects
No overt effects
No overt effects
Hypophagic, etc.
Normal
Normal
Normal
95% dead (80 days)
90% dead (80 days)
20% dead (80 days)
0% dead (80 days)
95% dead (22 days)
100% dead (40 days)
80% dead (75 days)
8% dead (80 days)
100% dead (45 days)
100% dead (55 days)
90% dead (75 days)
5% dead (80 days)
100% dead (22 days)
100% dead (22 days)
50% dead (65 days)
3% dead (80 days)
Reference
Kleeman et al. (1986a)
Kleeman et al. (1986b)
Spitsberg (1986)
Spitsberg et al. (1988a)
Kleeman et al. (1986a)
Kleeman et al. (1988)
Kleeman et al. (1988)
-------
8672H
Table 3-5. (continued)
Chemical form Test species
2.3.7,8-TCDO Largenouth bass
(continued) C~^50 ~ ^ "9/kg)
Bluegill
(LD5fl - 16 ug/kg)
Yellow perch
(LD50 - 10 ug/kg)
OJ
ro
Yellow perch
(LC50 - 3 ug/kg)
Pentachloro-
1.2.3.4.7-PCDO Rainbow trout
(fry - 0.2 g)
Fathead minnow
(0.2 g)
Rainbow trout
(0.5 - 1.0 g)
Fathead minnow
JU-Oq )
Duration (days)
Exposure / Observation
1 dose / +80 days
(i.p.)
1 dose / +80 days
(i.p.)
1 dose / +80 days
(i P-)
1 dose / +80 days
(i.p.)
5 / + 24 days
(water)
5 / + 24 days
(water) + 24 days
30 / + 30 days
(diet)
30 / + 30 days
(diet)
Concentration
(ng/1)
125
25
5
1
125
25
5
1
125
25
5
1
125
25
5
1
16
19
11
105 ng/g
105 ng/g
Toxic effects
100X dead (22 days)
95X dead (80 days)
5X dead (80 days)
5X dead (80 days)
100X dead (23 days)
SOX dead (76 days)
2X dead (80 days)
5X dead (80 days)
100X dead (35 days)
90X dead (40 days)
20X dead (60 days)
5X dead (30 days)
<95X dead (28 days)
<95X dead (28 days)
SOX dead (80 days)
OX dead (80 days)
No deaths
No deaths
No deaths
No significant (P = 0.05)
adverse effects
Ho significant (P = 0.05)
adverse effects.
Reference
Kleeman et al. (1988)
Kleeman et al. (1988)
Kleeman et al. (1986b)
Spitsberg et al. (1988b)
Huir et al. (1985)
Huir et al. (1985)
Huir et al. (1988)
Huir et al. (1988)
-------
able 3-5 , < cont 1 Fin
CO
I
Duration (days)
Chemical form Test species
Hexachloro-
1.2.3.4.7.8-HxCDD Rainbow trout
(fry - 0.2 g)
Fathead minnow
(0.2 g)
Rainbow trout
(0.5 - 1.0 g)
Fathead minnow
(1.0 g)
1,2,3.7.8.9-HxCDO Rainbow trout
(yolk sac fry)
Heptachloro-
1.2.3,4.6.7,8-HpCDD Rainbow trout
(fry - 0.2 g)
Fathead minnow
(0.2 g)
Rainbow trout
(0.5 - 1.0 g)
Fathead minnow
(1.0 g)
Exposure
5
(water)
5
(water)
5
(water)
30
(diet)
30
(diet)
4
(water)
5
(water)
5
(water)
30
(diet)
30
(diet)
/ Observation
/ +24 days
/ +48 days
/ +24 days
/ +48 days
/ +30 days
/ +30 days
/ +10 days
/ +24 days
/ +48 days
/ +24 days
/ +48 days
/ +30 days
/ +30 days
Concentration
(ng/1) Toxic effects
47 26% dead (12 days)
and weight loss
10 No effects
18 No deaths
7 No deaths
109 ng/g Significant (P = 0.05)
growth rate effects
109 ng/g No significant (P = 0.05)
adverse effects
Less than 10~5
times as toxic as
2.3,7.8-TCDO
55 No deaths
11 No deaths
39 No deaths
8 No deaths
109 ng/g No significant (P = 0.05)
adverse effects
109 ng/g No significant (P = 0.05)
adverse effects
Reference
Muir et al. (1985)
Muir et al. (1985)
Muir et al. (1988)
Muir et al. (1988)
Helder and Seinen (1986)
Muir et al. (1985)
Muir et al. (1985)
Muir et al. (1988)
Muir et al. (1988)
-------
Table 3-5. (continued)
Duration (days)
Chmical fom
Octach lore-
Test species
Rainbow trout
(0.1 - 0.5 g)
Fathead •irmow
(1.5 - 2.5 g)
Exposure
5
(water)
5
(water)
Concentration
/ Observation (ng/1)
/ +18 days 415
+ 3Z days 30
/ +48 days 9
Toxic effects
No deaths
No deaths
No deaths
Reference
Nuir et al. (1986)
Nuir et al. (1986)
CO
I
ro
-------
Duration (days)
Concentration
Chemical form Test species Exposure / Observation (ng/1)
Monochloro- No data
Dichloro- No data
Trichloro- No data
Tetrachloro-
1,3.7,8-TCDF Rainbow trout 4 / + 10 days 5.7 x 10"4
(yolk sac fry) (Mater) times as toxic as
2.3.7.8-TCDO
2,3.6,8-TCOF Rainbow trout 4 / + 10 days 9.5 x 10"5
(yolk sac fry) (water) times as toxic as
oo 2.3.7.8-TCDO
rv>
2.3.7,8-TCDF Rainbow trout 28 / + 28 days 8.78
(fry - 0.38 g) (water)
3.93
1.79
0.90
0.41
Controls
Rainbow trout / days
(same effects at same cone, as TCDD)
Rainbow trout 4 / + 10 days 8.9 x 10~Z
(yolk sac fry) times as toxic
Toxic effects
28% dead (28 days)
46% dead (56 days)
18% dead (28 days)
22% dead (56 days)
3% dead (28 days)
3% dead (56 days)
6X dead (28 days)
6% dead (56 days)
2% dead (28 days)
2% dead (56 days)
0% dead (56 days)
Lethal edema
Reference
Helder and Seinen (1986)
Nehrle et al. (1988)
Helder (1980)
Helder and Seinen (1986)
as 2,3,7,8-TCDD
-------
a combination of exposure routes, such as would be found in a natural
ecosystems, especially since the low water solubility and high
octanol/water partition coefficient of 2,3,7,8-TCDD suggest most of that
which would be found in the aquatic ecosystem would be sorbed to or
bioaccumulated in dietary sources.
The duration of an exposure is important when evaluating the inherent
chronic toxicity of 2,3,7,8-TCDD. Toxic effects were reported even for
the shortest 2,3,7,8-TCDD exposure period in water, 6 hours for rainbow
trout, following which rainbow trout mortalities occurred between 78 and
139 days after the exposure (Branson et al. 1985). Adams et al. (1986)
estimated that 90 percent of steady-state uptake of 2,3,7,8-TCDD in
rainbow trout would take longer than 48 days. Consequently, an adequate
chronic study would require that an exposure period be extended until the
maximum level of 2,3,7,8-TCDD has been reached, plus sufficient
additional time for effects to be produced. Review of the studies in
Table 3-5 indicates that none of the studies with water exposures are of
sufficient duration for steady-state concentration of 2,3,7,8-TCDD to
have been achieved in fish tissues.
The longest exposure to 2,3,7,8-TCDD via water was reported to be 28
days, with, an additional 28-days post-treatment for observation for
rainbow trout fry (Mehrle et al. 1988). In that study, a NOEC was not
determined. At the lowest test concentration (0.038 ng/1), 45 percent
mortality occurred before the test ended.
The 28-day trout study (Mehrle et al. 1988) did not include
deposition of 2,3,7,8-TCDD in the eggs by the female. Some evidence is
available which suggests that for trout the most sensitive life stage for
chlorinated chemicals occurs when the toxicant stored in the yolk is
sorbed into the developing embryo at the end of the yolk sac stage.
Similarly, Helder (1980) reported that following a 96-hour exposure to
2,3,7,8-TCDD, the highest mortality in freshly fertilized pike eggs
occurred during the resorption of the yolk. Mortality reached almost 100
percent at 2,3,7,8-TCDD exposure concentration of 10 ng/1. The NOEC was
reported as less than the lowest test concentration (0.1 ng/1).
Sublethal effects of PCDDs have been studied by several researchers,
but it appears that none of the studies have been conducted for
sufficient duration to determine the Inherent toxicity of PCDDs. The
effects of 2,3,7,8-TCDD on the development of medaka embryos have been
studied by Wisk and Cooper (1986). It was found that 20 to 40 ug/1
caused tube hearts, hemostasls, and liver necrosis. In another study
(Helder and Seinen 1986), 2,3,7,8-TCDD and 2,3,7,8-TCDF produced lethal
edematous conditions in rainbow trout. Similar effects of 2,3,7,8-TCDD
were reported by Helder (1980) during early life stages of the pike (Esox
luciush in which the sublethal effects included: reduced size at hatch,
3-24
ISflOq
-------
tail-first-hatching, alteration of blood vessel walls, pericardial edema,
enlarged nuclei and degradation in hepatocytes, and severe generalized
edemas preceeding death.
Sublethal effects on rainbow trout have been studied by Spitsberg et
al. (1988a) following single intraperitoneal injections of either 1, 5,
25, or 125 ug 2,3,7,8-TCDD/kg. At both 25 and 125 ug/kg, mortality
occurred before body weight loss could be detected. At 5 ug/kg, effects
included reductions in activity, feeding, and growth. Moderate to marked
leukopenia and thrombocytopenia were also found. The 2,3,7,8-TCDD-treated
trout also suffered higher stress from handling than controls. Numerous
types of lesions were reported in 1ymphomyeliod tissues of the thymus,
spleen, and kidneys, as well as epithelial lesions of the stomach mucosa,
liver, pancreas, gill, and skin. The types of lesions included the
following: thymic involution; splenic lymphoid depletion; decreased
kidney hematopoiesis; multifocal necrosis, atrophy, and hyperplasia of
the stomach mucosa; mild to severe hepatocyte vacuolation and ballooning
degeneration of the liver and hyperplasia of the bile-duct; vacuolation
of pancreatic exocrine cells; mild fusion of gill lamellae; and necrosis
of fin margins. Histologic changes in epithelial organs resembled the
lesions found in early stage studies of rainbow trout exposed to
water-borne 2,3,7,8-TCDD.
Sublethal effects in yellow perch following single intraperitoneal
injections of either 1, 5, 25, or 125 ug 2,3,7,8-TCBD/kg have also beer
reported (Spitsberg et al. 1988b). Many effects were, in general,
similar to those reported in rainbow trout. Yellow perch did not,
however, show lethality resulting from handling as seen in rainbow
trout. Lesions in the thymus, spleen, kidney, stomach, and skin are
similar to the effects observed in rainbow trout. Cardiac lesions not
seen in rainbow trout were found, including: necrosis of myocytes
subjacent to the epicardial surface of the ventricle; fibrinous
pericarditis; and hypertrophy and hyperplasis of the pericardial
mesothelium. Many of the histologic lesions, including fibrinous
pericarditis, have also been found in mammals, chickens, or turkeys
treated with 2,3,7,8-TCDD.
Spitsberg et al. (1986) measured immune responses in rainbow trout at
14 to 30 days after single intraperitoneal injections of either 0.1, 1.0,
or 10 ug/kg of 2,3,7,8-TCDD. Trout injected with 2,3,7,8-TCDD at either
0.1 or 1.0 ug/kg remained clinically normal. Trout treated with 10 ug/kg
of 2,3,7,8-TCDD became hypophagic and exhibited fin necrosis, ascites,
and suppression of hematopoiesis. Concanavalin A-induced blastogenesis
of thymic and splenic lymphocytes were not significantly changed,
however, suppression of the pokeweed mitogen-induced response of splenic
lymphocytes occurred. No statistically significant alterations occurred
in humoral iirtoune responses, and phagocytic activity of peritoneal
macrophages was not decreased. In rainbow trout, immunosuppression was
3-25
1580q
-------
evident only at 2,3,7,8-TCDD doses approaching 20 ug/kg, which is the
80-day, single-dose, parenteral LD50 value.
Spitsberg et al. (1988c) found no statistically significant effects
on rainbow trout mortality or mean time-until-death following a challenge
with infectious haematopoietic necrosis virus (IHNVJ. Rainbow trout were
injected intraperitoneally with single doses of either 0.01, 0.1, 1.0, or
10 ug/kg 2,3,7,8-TCDD. In virus-free trout, TCDD-induced effects first
appeared four to five weeks after injection of 10 ug/kg. The effects
included fin necrosis, as well as reduction in activity and food
consumption. No deaths occurred in the virus-free rainbow trout during
the five weeks after treatment.
Behavioral effects induced by 2,3,7,8-TCDD in rainbow trout have also
been reported by Merhle et al. (1988). Rainbow trout were severely
stressed at all test concentrations (i.e., 0.038 to 0.789 ng/1 of
2,3,7,8-TCDD}. Behavioral impairments increased over time and with
increasing test concentration. The behavioral changes included:
lethargic swimming, abnormal head-up swimming posture, feeding
inhibition, and lack of response to external stimuli. All of these
behaviors would increase the likelihood of predation in a natural
ecosystem.
3.4.2 Aquatic Toxicity of PCDFs
The longest exposure to 2,3,7,8-TCDF via water was 28 days, with an
additional 28-days .post-treatment for observation of the rainbow trout
fry (Table 3-6) (Merhle et al. 1988). The NOEC for 2,3,7,8-TCDF was
reported to be 0.41 ng/1. It appears that 2,3,7,8-TCDF levels in the
trout may have reached equilibrium at the higher concentration (3.93
ng/1), but it is not evident that equilibrium was achieved in 28 days at
the lower concentration (0.41 ng/1). Consequently, it is uncertain
whether the only study available concerning 2,3,7,8-TCDF was of
sufficient duration to produce maximum toxic effects.
The behavioral effects induced by 2,3,7,8-TCDF were similar to those
observed in 2,3,7,8-TCDD exposures; however, the observed responses were
of lesser magnitude (Merhle et al. 1988).
3.4.3 Conclusions Concerning Aquatic Toxicity
A definitive NOEC has not been reported for 2,3,7,8-TCDD. Even the
lowest test concentration (0.038 ng/1) produced 45 percent mortality in
rainbow trout exposed to 2,3,7,8-TCDD for 28 days. The reported NOEC
value for 2,3,7,8-TCDF is 0.41 ng/1, but this is also uncertain because
of the limited duration of observation. The toxicity data available on
2,3,7,8-TCDD, and possibly 2,3,7,8-TCDF, do not adequately define the
inherent toxicity of these substances for two reasons; (1) the exposure
periods are of insufficient duration for a steady-state equilibrium to be
3-26
1580q
-------
reached; and (2) the studies do not address toxic effects on developing
embryos resulting from deposition of either 2,3,7,8-TCDD or 2,3,7,8-TCDF
in the eggs by the female.
The most sensitive stage in the life of a rainbow trout for
chlorinated chemicals appears to occur during resorption of the yolk sac
in developing embryos. During this stage, the lipophilic substances
(e.g., PCDDs and PCDFs) deposited in yolk by the female become more
concentrated in the yolk on a per-unit-weight basis and they are
resorbed. This occurs as the fry metabolizes and incorporates nutrients
contained in the yolk. As the toxic substance is released during
metabolism of the yolk, the amount of substance entering the blood stream
reaches a higher concentration than would otherwise normally be found.
It is at this stage of development of organisms that the greatest
sensitivity has been observed for substances such as DDE in both birds
and fish.
While the adult-to-fry test results will not be available for some
time, an estimate is needed at this time to evaluate the aquatic toxicity
risks. Because the duration of the available studies are too short and
NOEC values have not been found, it is suggested that a larger-than-
normal margin of safety should be used, but how large a margin of safety
is uncertain. Margins of safety usually are 100 in the USEPA Offices of
Water, Pesticide Programs, and Toxic Substances, and the Office of Water
has used margins of safety as large as 1000. For the purpose of this
assessment, it was determined that the margin of safety for 2,3,7,8-TCDD
should be 1000, based on the level of lethality found at 0.038 ng/1 and
the short exposure period tested. Based on a margin of safety of 1000
and an NOEC for TCDD that is <0.038 ng/1, it was assumed for the purpose
of this assessment that concentrations of 2,3,7,8-TCDD in water greater
than 0.038 pg/1 will exhibit toxic effects to some aquatic species.
Applying this same approach to 2,3,7,8-TCDF, it was assumed that
concentrations of 2,3,7,8-TCDF greater than 0.41 pg/1 will exhibit toxic
effects to some aquatic species.
3.5 Toxicitv of PCDDs and PCDFs to Wildlife
The adverse effects to individual wildlife species from 2,3,7,8-TCDD
have been documented in laboratory studies. Using the results of these
studies to estimate effects on wild populations has some limitations.
The route and medium of adminstration and the duration of exposure to
2,3,7,8-TCDD for laboratory animals usually will differ from that of wild
animals. Using these studies to assess effects on wild species assumes
that the wild species are as sensitive or more sensitive to 2,3,7,8-TCDD
than the laboratory species. The methodologies for predicting the
effects of chemicals on terrestrial wildlife populations and ecosystems,
however, are'still in development. Thus, in the absence of sophisticated
predictive methods, measures of the effects of chemicals on reproduction
are currently the most useful indicators of possible effects on the
populations of the species in the wild.
3-27
1580q
-------
The estimated exposures to each animal listed in Table 3-7 are
compared to benchmark doses that have been identified as causing adverse
effects in laboratory species. Exposure levels approaching or exceeding
the selected benchmark suggest that the exposed animal is at risk for
experiencing adverse effects. Where possible, doses observed to cause
adverse reproductive effects were selected as benchmarks; the exposure of
a number of individual members of a species to a dose exceeding such a
reproductive effect benchmark may lead to adverse overall population
effects.
For birds, the estimated daily dose to adult birds is compared to the
dose that had no observable adverse effects level (the NOAEL) in
laboratory experiments. The concentration in bird eggs is compared to
the lowest observed adverse effects level (the LOAEL) in laboratory
studies. For mammals, the dose is compared to the lowest observable dose
to cause reproductive effects in laboratory animals. The studies upon
which the LOAELS or NOAELS were based are described below. The
discussion that follows also describes the adjustments made to these
doses in order to compare them to estimated wildlife exposures.
3.5.1 Toxicity Assessment for Birds
2,3,7,8-TCDD was administered at 100 ng/kg body weight/day in a corn
oil/acetone vehicle to 3-day old white leghorn chickens (Schwetz et al.
1973). This dose was administered for 21 days and produced no adverse
effects. It was assumed that the 2,3,7,8-TCDD was 100 percent absorbed
from the corn oil/acetone vehicle (see Appendix A.2). However, it is
possible that the absorption of 2,3,7,8-TCDD from laboratory feed or food
sources for wild animals would not be the same as the assumed 100 percent
absorption of 2,3,7,8-TCDD from a corn oil/acetone vehicle. Accordingly,
the estimated dose to wildlife species from the ingestion of prey items
is adjusted by the percent of 2,3,7,8-TCDD assumed to be absorbed from
the diet. Values for this percentage are found in a recent review of the
literature performed by Boyer (see Appendix A.2). In addition, the
laboratory dose must be converted to an equivalent dose over the length
of time that wild species of birds are exposed to 2,3,7,8-TCDD.
3.5.2 Toxicity Assessment for Bird Eggs
Bird eggs can contain 2,3,7,8-TCDD transferred from the mother's body
burden of 2,3,7,8-TCDD. Eggs are an important model to consider in
determining a toxicity endpoint because of the sensitivity of eggs to
2,3,7,8-TCDD. Sullivan et al. (1987) concluded that the LOAEL for
chicken embryos is 65 pg/g in the egg (65 ppt), based on a study that
found a 2-fold increase in cardiovascular malformations in chicken
embryos at an estimated egg concentration of 65 pg/g. Although effects
were found at lower concentrations of 2,3,7,8-TCDD, the study concluded
3-28
1580q
-------
Species
Exposure level
Synptoms/effects
Reference
CJ
I
rv>
vo
Chickens
Eastern Bluebird
Chickens
White Leghorn Chickens
Northern Bobwhite
Ringed Turtle-Dove
Mallards
Rhesus Monkeys
Monkeys
1 ng/kg/day
10.000 ppt
10"10 M concentration of TCDO
30—65 ppt
15 ug/kg body weight single oral dose
810 ug/kg body weight single oral dose
108 mg/kg body weight single oral dose
.0017 ug/kg body weight daily (7-29 months)
.011 ug/kg/day (9.3 months)
.0017 ug/kg/day (7-20 months)
.24 ug/kg three times weekly for 3 weeks of early gestation
.048 ug/kg three times weekly for 3 weeks of early gestation
.0095 ug/kg three tines weekly for 3 weeks of early gestation
Positive chick edema (lesions)
LOAEL for eggs.
60X reduction in tynphoid
cell numbers.
LOAEL for eggs
LD50
L050
L050
Abortion and weight loss.
Generalized toxicity: aneaia.
hair loss, death. 1 of 2
pregnancies aborted.
Slight loss of weight and hair.
4 of 7 pregnancies aborted.
Severe toxicity (death).
2/2 pregnancies aborted.
Slight toxicity.
3/4 pregnancies aborted.
No toxicity.
1/4 pregnancies aborted.
Scnwetz et al. (1973)
Thiel et al. (188)
Nikolaidis et al. (1988)
Sullivan et al. (1987)
Eisler (1986)
Eisler (1986)
Eisler (1986)
Eisler (1986)
Kociba and Scnwetz (1982)
Kociba and Scnwetz (1982)
Control
3/11 pregnancies aborted.
-------
867 2H
Table 3-7. (continued)
Species
Exposure level
Symptoms/effects
Reference
Bobwnite
Ha Hard
Turkey
Chicken
167 ppt oral exposure In a feeding test for 5 days with total
observation period of 8 day totals
3 ppt 18 Mk/18 wk test
0.3 ppt 18 Mk/18 «k test
27ft ppt oral exposure In a feeding test for 5 days with total
observation period of 8 days
>Z59 ppt 11 day/11 day test
>500 ng/kg/dy 21 day/21 day test
>200 ng/kg/dy 21 day/21 day test
>100 ng/kg/dy 21 day/21 day test
LC
50
Ho effect on reproduction.
LC
10
"ECQ" (no effect level)
LC90
Reduced feeding and growth.
"LCg" (no Mortality).
"LC0" (no Mortality)
Kenaga and Harris (1983)
Kenaga and Morris (1983)
Kenaga and Nor Ms (1983)
Kenaga and Morris (1983)
-------
that the evidence for effects at these lower levels was inconclusive;
thus, the 65 ppt value was used for comparison with predicted egg
concentrations for wild species.
3.5.3 Toxlcity Assessment for Wildlife Mammals
This analysis compares small mammals (i.e., mammals less than 1 kg)
to the lowest observed adverse reproductive effect level in laboratory
rats. Murray et al. (1979) administered rats at 100, 10, and 1 ng/kg/day
through the diet, and studied the effects on subsequent generations. At
the 10 ng/kg/day level, Murray et al. (1979) found decreased fertility in
the fj and ?2 generations.
For larger mammals, the expected dose for wild species is compared to
the lowest dose observed to produce adverse reproductive effects in
rhesus monkeys. Schwetz et al. (1973) reported that rhesus monkeys were
given 1.7 ng/kg body weight of 2,3,7,8-TCDD in the diet. Of the seven
pregnancies which occurred, it was observed that four had been terminated
because of chemical-induced abortions.
In these laboratory studies, it was assumed that absorption from a
laboratory diet is similar to the absorption from a wild diet, and that
these doses are directly comparable to the daily dose to wild species
from the ingestion of prey items.
3.6 Analysis of Uncertainties
The data base available concerning toxicological information on
2,3,7,8-TCDD is not complete, less Information is available on
2,3,7,8-TCDF, and there is practically no toxicological information on
the other congeners. Thus, SARs must be relied on to evaluate the
potential toxic effects and to develop I-TEFs. However, different animal
species exhibit different types of adverse effects and at different
levels of exposure to 2,3,7,8-TCDD, so predicting the exact type of
adverse effect at given exposure level for another congener has a great
deal of uncertainty.
Human health effects following exposure to 2,3,7,8-TCDD are not
nearly so well defined as effects on laboratory animals. There is
consensus concerning chloracne, but evidence regarding the association of
2,3,7,8-TCDD with human cancer 1s equivocal, as 1s the evidence
concerning reproductive effects.
Information concerning adverse human health effects of PCDFs is
available, but little specific information is available concerning
2,3,7,8-TCDF.- More studies are needed on past poisoning incidents
concerning the specific exposure chemicals and the follow-up health
histories to develop epidemiological information concerning both
2,3,7,8-TCDF and 2,3,7,8-TCDD.
3-31
1560q
-------
The cancer risk estimates developed by three U.S. regulatory agencies
(USEPA, FDA, and CPSC) differ by as much as a factor of 10, because of
differences in the models, details of extrapolation, and the selection of
animal data utilized. The approaches are technically valid, but the fact
the three agencies have each developed a different value may be the cause
of some confusion among the users of this information.
Based on the available data, the most sensitive noncarcinogenic
effects of 2,3,7,8-TCDD in animals are reproductive effects and
developmental toxicity. There is uncertainty as to whether these
endpoints are applicable to people of nonreproductive age or who are not
reproducing for other reasons. In addition, the noncancer effects data
for the other congeners are more limited, so sensitive endpoints and
Reference Doses are apparently, not available for the other PCDDs and
PCDFs.
The TEF approach is used to estimate carcinogenic and noncarcinogenic
effects of PCDD and PCDF mixtures in terms of an equivalent amount of
2,3,7,8-TCDD. Although a common set of TEF values was adopted in 1989
(I-TEFs/89), the approach remains interim in nature and will be revised
as new data are developed or until replaced (i.e., should a more direct
biological assay appear to be feasible in the future). In addition to an
accurate assessment of the congeners present, the uncertainties relating
to estimated intakes, bioavailability, interspecies extrapolation, safety
factors or mathematical models, and possible antagonistic or synergistic
interactions are likely to carry as much or more uncertainty than the TEF
values themselves.
Studies are available concerning the effects of 2,3,7,8-TCDD on
aquatic organisms, but most of the studies involved bioavailability tests
or short-term exposure studies with long post-exposure observation
periods. The available test data on 2,3,7,8-TCDD indicate that none of
the studies are adequate to define an acceptable, no-observed effect
concentration (NOEC). Only one study exposed a fish to TCDD in water for
a reasonable duration, 28 days. Even then, the exposure duration was
only about half of the 48 days estimated to be required to achieve 90
percent of the steady-state BCF level in that species of fish.
The reported results of typical acute toxicity tests are suspect
because of the short observation periods. In addition, intermittent,
sporadic releases may cause serious delayed effects.
In a risk assessment, the normal procedure is to identify the NOEC
for the most sensitive toxicological endpoint and apply a safety margin
(e.g., 100) to that NOEC. Unfortunately, the lowest tested concentration
(0.038 ng/1) among all known 2,3,7,8-TCDD studies produced 45 percent
mortality. Consequently, a definitive NOEC for 2,3,7,8-TCDD has not yet
been determined. All studies which have been completed are of
insufficient duration to determine the full extent of TCDD effects. For
3-32
ISSOq
-------
purposes of this assessment, it was determined that the margin of safety
for TCDD should be 1000. This factor is justified for use in this
assessment because of the high level of lethality (found to be 45 percent
at 0.038 ng/1, the lowest concentration tested), because of the exposure
duration (which was too short for steady-state equilibrium), and because
of the absence of exposure to pre-spawning adults and TCDD-deposition in
egg yolk (which generally yields the most sensitivity life stage for
chlorinated chemicals).
Sublethal effects have been documented for 2,3,7,8-TCDD, but none of
the studies conducted have been of sufficient duration to determine the
inherent toxicity of 2,3,7,8-TCDD. The most sensitive life stage appears
to be during yolk-sac resorption in developing embryos, but there are no
adult-to-fry test results available at present.
There are insufficient data on mixtures of PCDD/PCDFs to assess the
effects from the many complex combinations of exposure concentrations and
the various durations of exposure which occur in nature.
Laboratory studies of 2,3,7,8-TCDD to individual wildlife species are
well-documented, but there are uncertainties in extrapolating these
results to wild populations. The exposure route, medium of
administration, and duration of exposure in the laboratory will be
different from these experienced by wild animals.
Effects have been observed at very low levels, and the no observed
effects levels reported are only for the specific endpoints evaluated in
each study. In addition, the absorption from laboratory diet may be
different from the absorption from natural diets, such as ingestion of
prey items.
3.7 Conclusions
The most toxic congener among the PCDDs and PCDFs is 2,3,7,8-TCDD.
Among the carcinogens that have been evaluated by EPA, it is the most
potent animal carcinogen. Its reported effects in animal systems include
carcinogenesis, cancer promotion, reduced fertility and postnatal
survival, teratogenic effects, immunotoxic effects, thymic atrophy, liver
damage, effects on the thyroid, and chloracne and other effects on the
skin. There are marked variations among species in both the types of
observed effects and the dose levels of 2,3,7,8-TCDD causing these
effects. Higher doses of the congeners are required to elicit the same
type of adverse effect.
There is consensus that 2,3,7,8-TCDD causes chloracne in humans, but
the evidence concerning cancer and reproductive effects is equivocal.
There is stronger evidence that 2,3,7,8-TCDD causes disturbances in lipid
metabolism and increased frequency of gastric ulcers.
3-33
ISSOq
-------
PCDF exposure in humans has been shown to cause skin eruptions
similar to chloracne, skin pigmentation, and eye abnormalities, as well
as some evidence concerning persistent respiratory symptoms and
impairments of Tipid metabolism, and immune function. Further analyses
are needed for some preliminary reports of liver cancer deaths and lung
cancers among males following PCDF exposure. Some of these effects are
qualitatively similar to those observed in animals, supporting the use of
animal data as a basis for hazard assessment for other members of these
families of compounds.
The cancer risk estimate is represented as the upper bound slope
factor, qj (EPA and FDA) or as the maximum likelihood estimate of
the extra risk, qj (CPSC); these factors have been estimated by the
three U.S. regulatory agencies using somewhat different methods, with
estimated values for 2,3,7,8-TCDD of 1.6x10': (pg/kg/day)-l for EPA,
6.7xlO'5 (pg/kg/dayr1 for CPSC, and 1.8xlO"& (pg/kg/day)'J for
FDA. Based on these values, the estimated exposure (also called
risk-specific dose), giving an upper bound increased lifetime cancer risk
of one in a million, ranges from 0.006 pg/kg/day (EPA) to 0.06 pg/kg/day
(FDA).
Concerning noncancer effects of 2,3,7,8-TCDD, EPA has calculated, for
the purpose of this assessment, lifetime daily exposure to human
populations likely to be without appreciable risk of deleterious effect,
called a Reference Dose (RfD). Based on observed reproductive effects in
animal testing studies and the lack of a NOAEL from this testing, and
based on Uncertainty Factors to account for interspecies and intraspecies
differences, an RfD of 1 pg/kg/day was calculated for 2,3,7,8-TCDD.
Reproductive effects and developmental toxicity in animals are the most
sensitive noncarcinogenic effects for 2,3,7,8-TCDD.
International Toxicity Equivalency Factors (I-TEFs) have been
developed for 2,3,7,8-TCDD and the other PCDD and PCDF congeners. The
doses necessary to elicit the toxic responses differ in each case, but
the relative potencies of the different compounds (compared to
2,3,7,8-TCDD) are generally consistent from one endpoint to another. The
I-TEFs are based on ia vivo and & vitro test systems. Various systems
had been developed by various scientists and regulatory agencies
including international efforts. In 1989, a common set of TEFs were
adopted (I-TEFs/89) to assess the risks of PCDDs and PCDFs, including
mixtures of congeners, relative to an equivalent amount of 2,3,7,8-TCDD,
as if the components of the mixtures were concentrations of 2,3,7,8-TCDD
itself. There remain uncertainties in the TEF approach, however.
In aquatic toxicity studies, the most toxic congener of
polychlorinated dibenzo-p-dioxins and dibenzofurans appears to be
2,3,7,8-TCDD, probably due to its tendency to be taken up more readily
than other congeners. Dioxins appear to be mor« toxic to fish than to
other aquatic species.
3-34
ISBOq
-------
The mode of action of 2,3,7,8-TCDD is unknown, but it appears related
to neural toxicity and may affect the immune system, like many other
organochlorine chemicals. The pattern of TCDD-induced mortality
typically occurs between 30 to 80 days after the initial exposure, even
for exposure durations as short as 6 hours. Mortality in fish appears to
be a function of both exposure duration and the test concentration.
None of the bioconcentration tests were of sufficient duration to
achieve a steady-state bioconcentration factor (BCF). Thus, all the
reported BCF values have been estimated using the uptake rate (kj) and
depuration rate (k£) values.
Many sublethal effects have been reported for polychlorinated dioxins
and furans. 2,3,7,8-TCDD elicits a broad range of toxic effects which
include: reduced growth; weight loss; abnormal hatching; cellular
alterations; numerous types of lesions in a broad spectra of organs; and
behavioral effects on swimming, reduced feeding, and loss of response to
external stimuli.
Continuous exposures to fish appear to produce toxic effects more
rapidly than short-term exposure to the same test concentration. Based
on the available data, the most sensitive fish life stage at present
appears to be during yolk sac resorption in the developing embryo.
For wildlife species, the lowest NOEL for 2,3,7,8-TCDD exposure to
birds is 100 ng/kg/day in white leghorn chickens over a 21-day exposure
period. The LOAEL for chicken embryos is 65 pg/g (ppt) in the egg, but a
NOEL value is apparently not available. For wildlife mammals, a NOEL of
10 ng/kg/day in the rat has been reported in a multi-generation effects
study.
3.8 References
Adams WJ, Degraeve GM, Sanbourin TD, Cooney JD, Mosher GM. 1986.
Toxicity and bioconcentration of 2,3,7,8-tetrachlorodibenzo-p-dioxin to
fathead minnows (Pimeohales oromelasi. Chemosphere 15:1503-1511.
Babich MA, 1988. Unit risk estimate of the carcinogenicity of 2,3,7,8-
TCDD, U.S. Consumer Product Safety Commission, Directorate for Health
Sciences, November 21, 1988.
Bandiera S, Safe S, Okey AB. 1982. Binding of polychlorinated biphenyls
classified as either phenobarbitone-, 3-methylcholanthrene- or mixed-type
inducers to cytosolic Ah receptor. Chem.-Biol. Interact. 39:259-277.
Bandiera S, Sawyer T, Romkes M, Zmudzka B, Safe L, Mason G, Keys B,
Safe S. 1984. Polychlorinated dibenzofurans (PCDFs): Effects of
structure on binding to the 2,3,7,8-TCDD cytosolic receptor protein, AHH
induction and toxicity. Toxicology 32:131-144.
3-35
1580q
-------
Bell in JS, Barnes DG. 1987. U.S. Environmental Protection Agency.
Interim procedures for estimating risks associated with exposures to
mixtures of chlorinated dibenzo-p-dioxins and -dibenzofurans (CDDs and
CDFs). Risk Assessment Forum, EPA/625/3-87/012. National Technical
Information Service, Springfield, VA. PB89-125041.
Blair A. 1986. Review paper attached to Pitot et al. 1986.
Bond GG, Ott MG, Brenner FE, Cook RR. 1983. Medical and morbidity
surveillance findings among employees potentially exposed to TCDD. Br.
J. Ind. Med. 40:318-324.
Bowman RE, Schantz S, Gross ML, Ferguson S. 1989a. Behavioral effects
in monkeys exposed to 2,3,7,8-TCDD transmitted maternally during
gestation and four months of nursing. Chemosphere 18:235-242.
Bowman RE, Schentz SL, Weerasinghe NCA, Gross ML, Barsotti DA. 1989b.
Chronic dietary intake of 2,3,7,8-tetrachlorodibenzo-p-dioxin at 5 and 25
parts per trillion in the monkey: TCDD kinetics and dose-effect estimate
of reproductive toxicity. Chemosphere 18:243-252.
Branson OR, Takahashi IT, Parker WM, Blau GE. 1985. Bioconcentration
kinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rainbow trout.
Environ. Toxicol. Chem. 4:779-788.
California Air Resources Board. 1986. Staff Report: Public hearing to
consider adoption of a regulatory amendment identifying chlorinated
dioxins and dibenzofurans as a toxic air contaminant. June 1986.
Chang KJ, Hsieh KH, Lee TP, Tung TC. 1982a. Immunologic evaluation of
patients with polychlorinated biphenyl poisoning: Determination of
phagocyte Fc and complement receptors. Environ. Res. 28:329-334.
Chang KJ, Lu FJ, Tung TC, Lee TP. 1982b. Studies on patients with
polychlorinated biphenyl poisoning. 2. Determination of coproporphyrin,
uroporphyrin, delta-aminolevulinic urinary acid and porphobilinogen.
Res. Commun. Chem. Pathol. Pharmacol. 30:547-554.
Chen PH, Wong CK, Rappe C, Nygren M. 1985. Polychlorinated biphenyls,
dibenzofurans and quaterphenyls in toxic rice-bran oil and in the blood
and tissues of patients with PCB poisoning (Yu-Cheng) in Taiwan.
Environ. Health Perspect. 59:59-65.
Commoner B, Shapiro K, Webster T. 1984. Environmental and economic
analysis of alternative municipal solid waste disposal technologies. I.
An assessment of the risks due to emissions of chlorinated dioxins and
dibenzo-furans from proposed New York City incinerators.
3-36
-------
Couture LA, Elwell MR, Birnbaum LS. 1988. Dioxin-like effects observed
in male rats following exposure to octachlorodibenzo-p-dioxin (OCDD)
during a 13 week study. Toxicol. Appl. Pharmacol. 93:31-46.
Eadon G, Kaminsky L, Silkworth J, Aldous K, Hilder D, O'Keefe P, Smith R,
Gierthy J, Hawley J, Kim N, Decaprio A. 1986. Calculation of
2,3,7,8-TCDD equivalent concentrations of complex environmental
contaminant mixtures. Environ. Health Perspect. 70:221-227.
Eisler R. 1986. Dioxin hazards to fish, wildlife, and invertebrates: A
synoptic review. U.S. Department of the Interior, Fish and Wildlife
Service, Biological Report 85 (1.8).
FDA. 1983. U.S. Food and Drug Administration. Statement by S.A.
Miller, Director, Bureau of Food, before the Subcommittee on Natural
Resources, Agriculture, Research and the Environment, U.S. House of
Representatives.
June 30, 1983.
Grant DL. 1977. Proceedings of the 12th annual workshop on pesticide
residue analysis. Winnipeg, Canada.
Gravitz N, et al. 1983. Interim guidelines for acceptable exposure
levels in office settings contaminated with PCB and PCB combustion
products. Epidemiological Studies Section, California Department of
Health Services. Nov. 1.
Hardell L, Eriksson L, Lundgren E. 1981. Malignant lymphoma and
exposure to chemicals, especially organic solvents, chlorophenols and
phenoxy acids. A case-control study. Br. J. Cancer 43:169-176.
Hardell L, Sandstrom A. 1979. Case-control study: Soft-tissue sarcoma
and exposure to phenoxyacetic acids or chlorophenols. Br. J. Cancer
39:711-717.
Helder T. 1980. Effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)
on the early life stages of the pike (Esox lucius L.) Sci. Total
Environ. 14:225-264.
Helder T, Seinen W. 1985. Standardization and application of an
E.L.S.-bioassay for PCDDs and PCDFs. Chemosphere 14:183-193.
Helder T, Seinen W. 1986. Relative toxicities of some CDDs and CDFs and
toxic potentials of incineration products assessed by E.L.S.-bioassay.
Chemosphere 15(9-12):1165-1172.
Hoar SK, Blair A, Holmes FF, Boysen CD, Robe! RJ, Hoover R, Fraumeni JF.
1986. Agricultural herbicide use and risk of lymphoma and soft-tissue
sarcoma. JAMA. 256:1141-1147.
3-37
ISSOci
-------
Hsu ST, MA CI, Hsu SK, WU S S, Hsu NH M, Yeh C C, Wu SB. 1985.
Discovery and epidemiology of PCB poisoning in Taiwan: A four-year
followup. Environ. Health Perspect. 59:5-10.
Kenaga EE, Norris LA. 1983. Environmental toxicity of TCDD. In: Human
and environmental risks of chlorinated dioxins and related compounds.
Tucher RE, Young, AL, Gray AG, eds. New York: Plenum Press.
Kimbrough RD. 1986. Review paper attached to Pitot et al. 1986.
Kleeman JM, Olson JR, Chen SM, Peterson RE. 1986a. Metabolism and
deposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rainbow trout.
Toxicol. Appl. Pharmacol. 83: 391-401.
Kleeman JM, Olson JR, Chen SM, Peterson RE. 1986b. 2,3,7,8-Tetra-
chlorodibenzo-p-dioxin metabolism and deposition in yellow perch.
Toxicol. Appl. Pharmacol. 83:402-411.
Kleeman JM, Olson JR, Peterson RE. 1988. Species differences in
2,3,7,8-tetrachlorodibenzo-p-dioxin toxicity and biotransformation in
fish. Fundam. Appl. Toxicol. 10:206-213.
Kociba RJ, Keyes DG, Beyer JE, Carreon RM, Wade CE, Dittenber DA, Kalnins
RP, Frauson LE, Park CN, Barnard SD, Hummel RA, Humiston CG. 1978.
Results of a two-year chronic toxicity and oncogenicity study of
2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol. Appl. Pharmacol.
46:279-303.
Kociba RJ, Schwetz BA. 1982. Toxicity of 2,3,7,8-tetrachlorodibenzo-p-
dioxin (TCDD). Drug Metabolism Reviews. 13(3):387-406.
Kunita N, Hori S, Obana H, Otake T, Nishimura H, Kashimoto T, Ikegami N.
1985. Biological effect of PCBs, PCQs, and PCDFs present in the oil
causing Yusho and Yu-Cheng. Environ. Health Perspect. 59:79-84.
Kunita N, Kashimoto T, Miyata H, Fukushima S, Hori S, Obana H. 1984.
Causual agents of Yusho. Am. J. Ind. Med. 4:45-58.
Kuratsune M, Nakamura Y, Ikeda M, Hirohata. 1987. Analysis of deaths
seen among patients with Yusho-a prelminary report. Chemosphere
16:2085-2088.
Kuratsune M, Shapiro R. 1984. Preface: PCB poisoning in Japan and
Taiwan. Am. J. Ind. Med. 5:1-2.
Kusuda M. 1971. Yusho and women: A study on sexual functions of women
with the rice bran oil poisoning. Sanko to Fujinka 38:67-76. (in
Japanese).
3-38
1580q
-------
Lu YC, Wong PN. 1984. Dermatological, medical, and laboratory findings
of patients in Taiwan and their treatments. Am. J. Ind. Med. 5:81-115.
MDPH. 1983a. Michigan Department of Public Health. Evaluation of
congenital malformation rates for Midland and other selected Michigan
counties compared nationally and statewide: 1970-1981. (Unpublished).
MDPH. 1983b. Michigan Department of Public Health. Evaluation of soft
and connective tissue cancer mortality rates for Midland and other
selected Michigan counties compared nationally and statewide.
(Unpublished).
Masuda Y, Kuroki H, Haraguchi K, Nagayama J. 1985. PCB and PCDF
congeners in the blood and tissues of Yusho and Yu-Chang patients.
Environ. Health Perspect. 59:53-58.
Masuda Y, Yoshimura H. 1984. Polychlorinated biphenyls and dibenzofurans
in patients with Yusho and their toxicological significance: A review.
Am. J. Ind. Med. 5:31-44.
McNulty WP. 1985. Toxicity and fetotoxicity of TCDD, TCDF and PCB
isomers in- rhesus macaques (Macaca mulatta). Environ. Health Perspect.
60:77-88.
Merhle PM, Buckler DR, Little EE, Smith LM, Petty JD, Peterson PH,
Stalling DL, DeGaeve GM, Goyle JJ, Adams WL. 1988. Toxicity and
bioconcentration of 2,3,7,8-tetrachlorodibenzo-p-dioxin and
2,3,7,8-tetrachlorodibenzofuran in rainbow trout. Environ. Toxicol.
Chem. 7:47-62.
Miller RA, Morris LA, Hawes CL. 1973. Toxicity of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in aquatic organisms. Environ. Health
Perspect. 5:177-186.
Miyata H, Fukushima L, Kashimoto T, Kunita N. 1985. PCBs, PCQs and
PCDFs in tissues of Yusho and Yu-Chang patients. Environ. Health
Perspect. 39:67-72.
Moses M, Lilis R, Crow KD, Thornton J, Fischbein A, Anderson HA, Selikoff
IJ. 1984. Health status of workers with past exposure to 2,3,7,8-
tetrachlorodibenzo-p-dioxin in the manufacture of 2,4,5-trichlorophenoxy
acetic acid: Comparison of findings with and without chloracne. Am. J.
Ind. Med. 5:161-182.
Mtiir DCG, Yarechewski AL, Webster GRB. 1985. Bioconcentration of four
chlorinated dioxins by rainbow trout and fathead minnows. .In.: Bahner RC
and Hansen DJ (eds.). Aquatic toxicology and hazard assessment: Eighth
Symposium. Philadelphia: ASTM, Spec. Tech. Publ. 891. pp. 440-454.
3-39
1580q
-------
Muir DCG, Yarechewski AL, Knoll A. Webster GRB. 1986. Bioconcentration
and disposition of 1,3,6,8-tetrachlorodibenzo-p-dioxin and
octachlorodibenzo-p-dioxin by rainbow trout and fathead minnows. Environ.
Toxicol. Chem. 5:261-271.
Muir DCG, Yarechewski AL. 1988. Dietary accumulation of four chlorinated
dioxin congeners by rainbow trout and fathead minnows. Environ. Toxicol.
Chem. 7:227-236.
Murray FJ, Smith FA, Nitschke KD, Humiston CG, Kociba RJ, Schwetz BA.
1979. Three-generation reproduction study of rats given 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in the diet. Toxicol. Appl. Pharmacol
50:241-251.
NATO-CCMS (North Atlantic Treatment Organization, Committee on the
Challenges of Modern Society). 1988a. International toxicity
equivalency factor (I-TEF) method of risk assessment for complex mixtures
of dioxins and related compounds. Report No. 176.
NATO-CCMS (North Atlantic Treatment Organization, Committee on the
Challenges of Modern Society). 1988b. Scientific basis for the
development of international toxicity equivalency (I-TEF) factor method
of risk assessment for complex mixtures of dioxins and related
compounds. Report No. 178.
NCI. National Cancer Institute. 1980. Bioassay of a mixture of
1,2,3,6,7,8- hexachlordibenzo-p-dioxin and 1,2,3,7,8,9-hexachlordibenzo-
p-dioxin for possible carcinogenicity (Gavage Study). Bethesda, MD:
National Cancer Institute Carcinogenesis Technical Report Series No. 198.
NTP. National Toxicology Program. 1982. Carcinogenesis bioassay of
2,3,7,8-tetrachlorodibenzo-p-dioxin in Osborne-Mendel Rats and B6C3F1
Mice (Gavage Study). Research Triangle Park, NC: NTP Technical Report
Series No. 209.
Nakanishi Y, Shigematsu N, Kurita Y, Matsuba K, Kanegae H, Ishimaru S,
Kawazoe Y. 1985. Respiratory involvement and immune status in Yusho
patients. Environ. Health Perspect. 59:31-36.
Nikolaidis E, Brunstrom B, Dencker L. 1988. Effects of TCDD and its
congeners 3,3,4,4-tetrachlorazoxybenzene and 3,3,4,4-tetrachlorobiphenyl
on lymphoid development in the thymus of avian embryos. Pharmacol.
Toxicol. 63:333-336.
Okumura M, Masuda Y, Nakamura S. 1974. Correlation between blood PCB
and serum triglyceride levels in patients with PCB poisoning. Fukoka
Igaku Zasshi 65:84-87. (in Japanese, summary in English).
01 ie K et al. 1983. Formation and fate of PCDD and PCDF from combustion
processes. Chemosphere 12:627-737.
3-40
1580q
-------
Olson JR, Bell in JS, Barnes DG. 1989. Re-examination of data used for
establishing toxicity equivalency factors (TEFs) for chlorinated
dibenzo-p-dioxins and dibenzofurans (CDDs and CDFs). Presented at the
7th International Symposium on Chlorinated Dioxins and Related Compounds;
October 4-9, 1987; Las Vegas, Nevada. Chemosphere 18:371-381.
Ontario MOE. 1982. Ontario Ministry of the Environment. Chlorinated
dioxins and chlorinated dibenzofurans. Ambient air guideline. Health
Studies Service, Ministry of Labour. Province of Ontario, Canada.
Ontario MOE. 1984. Ontario Ministry of the Environment. Scientific
criteria document for standard development. Polychlorinated
dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs).
Ministry of the Environment, No. 4-84, December.
Pitot H, Blair A, Dean J, Gallo M, Poland A, Hoel D, et al. 1986.
Report of the Dioxin Update Committee. Submitted to USEPA, Office of
Pesticides and Toxic Substances, Washington, D.C. August 28.
Poland A, Knutson JC. 1982. 2,3,7,8-Tetrachlorodibenzo-p-dioxin and
related halogenated aromatic hydrocarbons: Examination of the mechanism
of toxicity. Ann. Rev. Pharmacol. Toxicol. 22:517-554.
Rao MS, Subbarao V, Prusad JD, Scarpelli DG. 1988. Carcinogenicity of
2,3,7,8-tetrachlorodibenzo-p-dioxin in the Syrian golden hamster.
Carcinogenesis 9:1677-1679.
Schwetz BA, Norris JM, Sparschu GL, Rowe VK, Gehring PO, Emerson JL,
Gergib CG. 1973. Toxicology of chlorinated dibenzo-p-dioxins. Environ.
Health Perspec. 5:87-99.
Sparschu GL Jr, Dunn FL Jr, Rowe VK Jr. 1971. Study of the
teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Food
Cosmet. Toxicol. 9:405-412.
Spitsberg JM, Schat KA, Kleeman JM, Peterson RE. 1986. Interactions of
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) with immune responses of
rainbow trout. Vet. Immunol. Immunopathol. 12:263-280.
Spitsberg JM, Kleeman JM, Peterson RE. 1988a. Morphological lesions and
acute toxicity in rainbow trout fSalmo qairdneri) treated with
2,3,7,8-tetrachlorodibenzo-p-dioxin. J. Toxicol. Environ. Health
23(3):333-358.
Spitsberg JM, Kleeman JM, Peterson RE. 1988b. 2,3,7,8-Tetrachloro-
dibenzo-p-dioxin toxicity in yellow perch (Perca flavescens). J.
Toxicol. Envir. Health 23:359-383.
3-41
1580q
-------
Spitsberg JM, Schat KA, Kleeman JM, and Peterson RE. 1988c. Effects of
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) or Aroclor 1254 on the
resistance of rainbow trout, Salmo qairdneri Richardson, to infectious
hematopoietic necrosis virus. J. Fish Dis. 11:73-83.
Sullivan JR, Kubiak TJ, Amundson TE, Martini RE, Olson LJ, Hill GA.
1987. A wildlife exposure assessment for landspread sludges which
contain dioxins and furans. In Proceed. Tenth Ann. Madison Internat.
Waste Conf: Municip. Indust. Waste. Sept. 29-30, 1987. Madison, VII:
Univ. of Wisconsin.
Suskind RR, Hertzberg VS. 1984. Human health effects of 2,4,5-T and its
toxic contaminants. JAMA 251:2372-2380.
Swiss Government (Bundesamt for Umweltschutz, Bern). 1982. Environmental
pollution due to dioxins and furans from chemical rubbish incineration
plants. Schriftenreighe Umweltschutz, No. 5.
Thiel DA, Martin SG, Duncan JW, Lemke MJ, Lance WR, Peterson R. 1988.
Evaluation of the effects of dioxin-contaminated sludges on wild birds.
In: Proceedings of the 1988 TAPPI Environmental Conference.
Thiess AM, Frentzel-Beyme R, Link R. 1982. Mortality study of persons
exposed to dioxin in a trichlorophenol-process accident that occurred in
the BASF AG on November 17, 1953. Am. J. Ind. Med. 3:179-189.
Turner ON, Collins DN. 1983. Liver morphology in guinea pigs
administered either pyrolysis products of a polychlorinated biphenyl
transformer fluid or 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Appl.
Pharmacol. 67:417-429.
USEPA. 1981. U.S. Environmental Protection Agency. Interim evaluation
of human health risks associated with emissions of tetrachlorinated
dioxins from municipal waste resource recovery facilities. Office of the
Administrator. Nov. 19.
USEPA. 1984a. U.S. Environmental Protection Agency. Ambient water
quality criteria for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Washington,
D.C.: EPA. Office of Water Regulations and Standards. EPA-440/5-84-007.
USEPA. 1984b. U.S. Environmental Protection Agency. Health effects
profile for tetra-, penta- and hexachlorodibenzo-p-dioxins. Final
draft. Cincinnati, OH. USEPA, Off. Res. Oevel., Environ. Crit. Assess.
Off. ECAD-CIN-P004.
USEPA. 1985. U.S. Environmental Protection Agency. Health Assessment
Document for polychlorinated dibenzo-p-dioxins. EPA 600/8-84-014F.
Office of Research and Development. Washington, DC: Office of Research
and Development. EPA 600/8-84-014F.
3-42
ISSOq
-------
USEPA. 1986. U.S. Environmental Protection Agency. Guidelines for
carcinogen risk assessment. Federal Register. Vol. 51, No. 185,
33992-34003. Septemer 24.
USEPA. 1987a. U.S. Environmental Protection Agency. Reference Dose:
Description and use in health risk assessments. Integrated Risk
Information System. Appendix A. Washington, DC: USEPA.
USEPA. 1987b. U.S. Environmental Protection Agency. Health advisory
for 2,3,7,8-TCDD. Washington, DC: Office of Drinking Water.
USEPA. 1988a. U.S. Environmental Protection Agency. Risk Assessment
for Dioxin Contamination in Midland, Michigan. EPA/905/4-88/005.
USEPA. 1988b. U.S. Environmental Protection Agency. A cancer
risk-specific dose estimate for 2,3,7,8-TCDD. Review draft.
EPA/600/6-88007Aa.
USEPA. 1989. U.S. Environmental Protection Agency. Interim procedures
for estimating risks associated with exposures to mixtures of chlorinated
dibenzo-p-dioxins and -dibenzofurans (CDDs and CDFs) and 1989 update.
EPA-625/3-89/March 1989.
USDHHS. 1983. U.S. Department of Health and Human Services. Levels of
concern for hexa- (HCDD), hepta- (HpCDD) and octachlorodibenzo-p-dioxin
(OCDD) in chicken and eggs. Memorandum. Apr. 29.
Wisk J, Cooper KR. 1986. Comparison of toxicity between 2,3,7,8-tetra-
chlorodibenzo-p-dioxin (TCDD) and several tetrachlorodibenzofuran isomers
(TCDF) in the Japanesa medaka embryo-larval bioassay. Proceedings of the
7th Annual Meeting of SETAC.
Urabe H, Asahi M. 1985. Past and current dermatological status of Yusho
patients. Environ. Health Perspect. 59:11-15.
Yamashita F, Hayashi M. 1985. Fetal PCS syndrome: Clinical features,
intrauterine growth retardation and possible alteration in calcium
metabolism. Environ. Health Perspect. 59:41-45.
3-43
1580q
-------
4. ASSESSMENT OF RISKS TO WORKERS FROM EXPOSURE TO DIOXINS AND
FURANS FROM MANUFACTURE, PROCESSING, AND COMMERCIAL USE OF PULP,
PAPER, AND PAPER PRODUCTS AND FROM PROCESSING AND COMMERCIAL USE
OF PULP AND PAPER MILL SLUDGE
4.1 Introduction
Workers may be exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF at various
stages of the paper production process. Exposures may occur during the
bleaching of wood pulp, and from contact with raw paper products (e.g.,
newsprint, paperboard, fibers), finished commercial paper products, and
wastewater and sludge generated during the manufacture of paper
products. Additional exposure may occur from the commercial use of pulp
and paper mill sludge.
This chapter summarizes data presented in the following two reports
prepared for EPA:
PEI (1990a). PEI Associates, Inc. Estimated worker exposure to
2,3,7,8-TCDD and 2,3,7,8-TCDF in the manufacture, processing, and
commercial use of pulp, paper, and paper products. Washington,
D.C.: USEPA, Office of Toxic Substances. Contract No. 68-D8-0112.
March 1990.
PEI (1990b). PEI Associates, Inc. Estimated worker exposure to
2,3,7,8-TCDD and 2,3,7,8-TCDF from processing and commercial use of
pulp and paper mill sludge. Washington, D.C.: USEPA, Office of
Toxic Substances. Contract No. 68-D8-0112. April 12, 1990.
Section 4.2 summarizes data from the PEI (1990a) report, Section 4.3
summarizes the PEI (1990b) report, and Section 4.4 addresses the issue of
uncertainties.
4.2 Worker Exposure to Dioxins In the Manufacture. Processing, and
Commercial Use of Pulp, Paper, and Paper Products
This section provides an overview of worker exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF during the manufacture, processing, and commercial use
of pulp, paper, and paper products. Section 4.2.1 characterizes the
worker population for different operations involving the manufacture,
processing, and commercial use of pulp, paper, and paper products.
Section 4.2.2 discusses the potential for exposure to 2,3,7,8-TCDD and
2,3,7,8-TCDF in pulp and paper processes/operations; it also includes a
description of the methodology used to estimate the levels of inhalation
and dermal exposure for workers involved in different types of
processes. Section 4.2.3 summarizes the estimated exposure, individual
cancer risks, and population cancer risks.
4-1
1581q
-------
4.2.1 Pulp and Paper Industry Workforce Characterization
In 1986. the paper and allied products industry employed over 674,000
people, of which approximately three-fourths were directly involved in
pulp and paper production (OTA 1989). Table 4-1 summarizes the number of
total employees and production workers in 1985 in the various paper and
allied product categories. Both integrated and non-integrated paper
mills are included in Table 4-1.
As Table 4-1 shows, over 75 percent of the employees in the industry
are production workers. Not all of these production workers, however,
are exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF; the exposure varies for
different job categories. It is important to note that job categories
and descriptions vary considerably throughout the industry depending on
equipment layout, automation, and complexity. Pulp and papermaking mills
generally operate 24 hours per day, 3 shifts per day, 7 days per week.
Paper converting operations usually shut down at night and on weekends
(Soklow 1984).
The workforce characterization in kraft pulp and paper mills has not
been fully studied. A limited amount of information was obtained from
such sources as EPA literature, NIOSH data bases, industry data, and
various industrial contacts. Additional assumptions were needed to
calculate the number of workers in each job category. It was assumed
that the number of workers in kraft pulp mills was proportional to the
ratio of the amount of bleached pulp to the total amount of pulp
produced. Other assumptions are stated in the respective sections.
(1) Brown stock washing. In brown stock washing operations, the
potential for 2,3,7,8-TCDD and 2,3,7,8-TCDF exposure exists only if paper
machine white waters, which have been used in the processing of bleached
pulp, are recycled. However, the recycling of white waters is a rare
practice throughout the pulp and paper industry. The job categories
which could have a potential for exposures for 2,3,7,8-TCDD and
2,3,7,8-TCDF are the wash and screen room operators, pulp testers, and
utility employees.
The wash and screen room operators monitor and control operations of
washer lines, screens, filtrates, and high-density storage tanks (Soklow
1984). The number of wash and screen operators in a plant may range from
0 to 4, averaging 1 to 2 per shift (NIOSH 1983).
Pulp testers are responsible for retrieving and analyzing production
area samples. Samples are usually analyzed in laboratory areas located
away from the production area such as a wash/screen control room (Soklow
1984). The number of testers per shift may range from 0 to 3, but is
usually 1 (NIOSH 1983).
4-2
-------
9068H
Table 4-1. Total Employees and Production Workers In the Entire
Paper and Allied Products Industry. 1985a
Total number Production
Industry group of employees workers
Pulp mills 16,000 12.000
Paper mills, except building paper 132,000 102,000
Paperboard mills 54.000 41,000
Miscellaneous converted paper products 211.000 161,000
Paperboard containers and boxes 188,000 143,000
Building paper and board mills 4.000 3,000
TOTAL 604,000 462.000
aUSOOC (1987)
4-3
-------
(2) Bleaching operations. In the bleaching operations, 2,3,7,8-
TCDD and 2,3,7,8-TCDF are formed through the use of chlorine or chlorine
derivatives, and thus are expected to be at its highest concentrations.
Four major job activities in such operations include: (1) process
monitoring and control; (2) manual operation, and adjustment of
equipment, or inspection; (3) process quality control sampling and
testing; and (4) housekeeping and spill cleanup. The total number of
production workers per shift ranges from 2 to 5 (NIOSH 1983). The bleach
plant population potentially exposed to 2,3,7,8-TCDO and 2,3,7,8-TCDF is
expected to be 1,300 workers total, based on the assumption that there
are three employees on each of the four shifts within the 104 mills.
(3) Paper-making and finishing. In this stage, additives such as
coatings, colorings, or sizing agents can be mixed with the pulp stock.
Job categories potentially exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF
include wet-end additives/stock preparation operators, coating
preparation operators, dry-end paper machine operators, off-machine
coating operators, assistants, and utility employees.
Wet-end additives/stock preparation operators are responsible for
operating and monitoring the equipment during the following types of
activities: dry pulp dissolving; beating, refining, thickening,
cleaning, and blending of pulps; preparing additives such as sizes,
fillers, and covering agents; and mixing and blending of ingredients.
Operators in the coating preparation category are responsible for the
preparation of the coatings, unloading the raw materials, and monitoring
the system (Soklow 1984).
Dry-end paper machine operators are responsible for web drying,
on-machine coating and drying, reeling, on-machine calendering, slitting/
rewinding, quality control testing, and packaging and shipping. Drying
operations utilize hot air circulated under a hooded ventilation system
for the final drying step (Soklow 1984). Off-machine coating operations
include application of the coating formulation, paper drying, and
rewinding of the paper into rolls. Operators are responsible for
monitoring the process, checking paper quality, assisting the utility
personnel with loading or unloading rolls of coated paper, and cleaning
of the dryer during upsets (Soklow 1984).
The number of workers per shift in the papermaking operations will
vary depending on (1) number and location of paper machines, coaters, and
slitters/rewinders; (2) the type and complexity of wet-end operations;
(3) the complexity, speed, and automation of papermaking equipment; and
(4) the type, number, and complexity of the dry-end operations (NIOSH
1983).
The paper mill population potentially exposed to PCDDs and PCDFs is
estimated to be 32,000 workers total. This is based on the assumption
that there are two machines per facility, with 18 workers per each of the
4-4
1581q
-------
four shifts working at each machine in the 221 facilities processing
bleached kraft and sulfite pulp. The number of facilities processing
bleached pulp was estimated to be proportional to the quantity of
bleached pulp production.
(4) Pulp drying. Operators in the drying stage of pulp production
are responsible for monitoring the equipment, weighing the pulp sheets,
off-loading and stacking them for shipment. The utility operators clean
up any spills and assist the operators in keeping the production area
clean (McCubbin 1989). The number of pulp drying operators potentially
exposed to PCDDs and PCDFs was estimated to be 240 people assuming four
workers in each shift of the 15 mills that use pulp drying (USDOC 1988).
(5) Converting operations. Converting operations transform paper
into end products such as paper towels, cardboard boxes, and typing
paper. One hundred twenty-nine thousand paper converting workers are
estimated to be potentially exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF
based on Bureau of Census data (USDOC 1988) and the assumption that the
number of workers is proportional to the amount of bleached paper
produced.
(6) Nonwoven operations. Operators in the nonwoven industry are
responsible for machine operation and monitoring the system, raw material
handling, quality control testing, and general housekeeping. Typically,
there are two to four operators per shift for each finished-product
machine (Nonwovens 1989). No precise information was available on the
total number of workers in the nonwoven industry; however, the number of
workers has been estimated to be more than 15,000 workers (Cunningham
1990).
(7) Commercial users. Commercial users of paper products include
almost all workers. Some occupational classifications where workers are
exposed for large portions of their work day include lawyers, computer
programmers, secretaries, accountants, librarians, teachers, architects,
postal workers, printers, and other government workers. According to
information from the Bureau of Census, over 50 million workers in the
United States have occupations which involve handling bleached paper
products.
Medical workers who use nonwoven products containing bleached pulp
include doctors, nurses, dentists, and other workers involved in health
maintenance and diagnosing. These workers may wear nonwoven garments or
breathe through nonwoven face masks for several hours each day.
Table 4-2 summarizes the overall number of workers (by industry
segment and job category) exposed to bleached pulp and paper products.
The following section presents the levels of exposure associated with
these various activities.
4-5
I581q
-------
9068H
Table 4-2 Number of Workers in the Pulp
and Pulp Products Job Categories
Job category No. of workers
Pulp mi 11
Pulp manufacture 1,300
- Bleach operators 434
- Pulp testers 433
- Utility operators 433
Pulp drying 240
- Operators 160
- Utility workers 80
Paper mill
Paper and paperboard manufacture 32,000
- Wet-end operator 10,667
- Dry-end operator 12,445
- Utility operator 8.888
Converting operations
Paper converting operations 68.000
- General worker
Paperboard converting operations 61,000
Nonwovens production (pulp converting) 15,000
- General worker
Commercial users
Paper and paperboard 48,671,000
Nonwovens 2.895.000
4-6
-------
4.2.2 Worker Exposure Estimating Methodologies
Although considerable data have been collected on concentrations of
2,3,7,8-TCDD and 2,3,7,8-TCDF for pulp, sludge, and wastewaters in the
pulp and paper industry, no inhalation or dermal exposure data for
2,3,7,8-TCDD and 2,3,7,8-TCDF are currently available. Furthermore,
little information is available on the effectiveness of engineering
controls or personal protective equipment in protecting workers in this
industry. Therefore, modeling techniques were used to estimate worker
exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF.
Pulp and paper mill workers may be exposed to 2,3,7,8-TCDD and
2,3,7,8-TCDF either through dermal contact with wet or dry pulp, paper,
or bleaching filtrates, or through inhalation of volatilized
2,3,7,8-TCDD/2,3,7,8-TCDF or particulate containing these chemicals. The
source of particulates is paper dust generated from paper cutting,
rolling, or packaging operations.
The estimates of inhalation exposures due to volatilized .
2,3,7,8-TCDD/2,3,7,8-TCDF are relatively low because of the low vapor
pressures of these chemicals. In addition, TCDDs/TCDFs have a tendency
to preferentially bind with organic matter. No attempt was made to
compensate for 2,3,7,8-TCDD/2,3,7,8-TCDF affinity for organic materials.
(1) Pulp manufacturing. In pulp manufacturing operations, the
potential for exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF exists primarily
in the bleaching operations. In brownstock washing operations, exposure
can occur if water used for these operations is recycled paper machine
white water containing 2,3,7,8-TCDD and 2,3,7,8-TCDF formed during bleach
operations. The most common wash method, however, is effluent from the
filters and fresh hot water which are not likely to contain PCDDs, PCDFs,
or their precursors. Worker activities in brownstock operations include
servicing of the brownstock washers and screens, sample collection and
testing, and general plant maintenance, which usually involves cleanup of
spills- Brownstock washers are well ventilated, with exhausts through
canopy hoods or full enclosures.
Most kraft pulping processes are highly automated and, consequently,
the operators spend considerable portions of each shift inside control
rooms {NIOSH 1983). In brownstock washing operations with no control
room in the washer area, operators may spend essentially the entire shift
in the production area, usually at operating consoles (Soklow 1984).
Average times spent in the control rooms for workers involved in
brownstocking were 65 percent, 75 percent, and 33 percent for the
operators, testers, and utility employees, respectively (NIOSH 1983).
The number of bleach plant workers potentially exposed to dioxin and
thei*" job descriptions depend upon the process size, degree of
automation, plant layout and equipment, and integration of bleach
4-7
158 la
-------
chemical handling and preparation operations. Various bleaching process
characteristics may inadvertently reduce the potential for dioxin
exposure. Ventilation of chlorine and chlorine dioxide towers, washers,
and washer filtrate tanks, with or without gas scrubbing, in bleaching
operations may inadvertently reduce worker exposure and thereby control
worker inhalation exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF. Chlorine
alarms are installed in the pulp bleaching building and usually set to
go off at 1 ppm, the current OSHA Short-Term Exposure Limit (STEL) (15
minutes maximum for chlorine). The revised OSHA PEL for chlorine is 0.5
ppm (8-hr TWA). Rinsing operations often occur in open washers. Workers
can often reach into the wash stream to gather samples of bleached or
partially bleached wet pulp. Sampling of the wet bleached pulp or spill
cleanup provides potential for worker dermal exposure unless proper
personal protective equipment is worn.
The pulp manufacturing population exposed to 2,3,7,8-TCDD and
2,3,7,8-TCDF is estimated to be 1,300 workers total. The following
sections discuss the various modeling techniques and assumptions used to
estimate dermal and inhalation exposures.
(a) Inhalation exposure. In the pulp mill, there are three
job categories of workers (bleach plant operators, pulp testers, and
utility operators) who are potentially exposed to PCODs and PCDFs through
inhalation. These workers spend approximately 75, 25, and 20 percent of
their shifts in the control room.
In the manufacture of pulp, only vapors generated from volatilization
of the PCDDs and PCDFs constitute a potential route for inhalation
exposure. No pulp dust is generated at this point in the production
process. Pulp dust may be generated during pulp drying operations, which
are discussed in the next subsection. There are no existing data
available to determine inhalation exposure to 2,3,7,8-TCDD/2,3,7,8-TCDF
vapors during pulp manufacturing. In the absence of exposure monitoring
data, worker exposure to 2,3,7,8-TCDD/2,3,7,8-TCDF was estimated using
two different approaches (based on the nature of the worker activities).
The first approach utilizes a mass balance model to estimate worker
exposure for specific activities {e.g., for pulp testers during
sampling). The second approach is applicable for workers in a general
area (e.g., bleach plant operators, utility operators) and is based on
estimating the maximum 2,3,7,8-TCDD/2,3,7,8-TCDF air concentration
available for inhalation based on their partial pressures. These partial
pressures are calculated by assuming that TCDD, TCDF, and pulp water
solutions behave as ideal mixtures. Specific details of the approaches
are discussed under each scenario for which worker exposures are
estimated.
There are many mass balance models available for estimating worker
exposure. The following equation, derived using one of the mass balance
models (Clement 1982) in combination with an equation describing the
4-8
-------
generation rate (Thibodeaux 1979), was used to estimate inhalation
exposure concentrations to 2,3,7,8-TCDD/2,3,7,8-TCDF for pulp testers:
r _ 6.3 x 10'b K A Pa
Lv " k Q T
(4-1)
where:
Cv
K
A
Pa
k
T
Q
concentration of 2,3,7,8-TCDD/2,3,7,8-TCDF in the vapor, ppm
gas mass transfer coefficient, cm/sec
surface area, cm2
partial pressure of a component, atm
mixing factor, unitless
temperature, K
ventilation rate, ft3/min
The partial pressures in Equation 4-1 are calculated based on
Raoult's law. The mass transfer coefficients for 2,3,7,8-TCDD/
2,3,7,8-TCDF are not available in the literature and are estimated by an
equation presented in Clement (1982).
The approach for estimating worker inhalation exposure concentrations
to 2,3,7,8-TCDD/2,3,7,8-TCDF around closed systems in the bleaching area
was based on comparison with the OSHA PEL for another chemical (chlorine)
handled in the same process area and a knowledge of the vapor pressures
of chlorine and 2,3,7,8-TCDD/2,3,7,8-TCDF. The airborne concentrations
of 2,3,7,8-TCDD and 2,3,7,8-TCDF were estimated based on a comparison
with the PEL for chlorine, since there are alarms in the bleach
processing area for chlorine to limit workers' exposure to this chemical
below its PEL. The use of a PEL to estimate inhalation exposures gives a
maximum or reasonable worst case workplace concentration. The exposures
for the bleach plant operator and utility operator were assumed to be
comparable since their job duties require them to be in the bleaching
area of the pulping process for a portion of the shift. Therefore, the
same approach was used to estimate inhalation exposure for these job
categories; and the empirical equation for this approach is presented in
Equation 4-2:
Cv
(4-2)
where:
Cv
Cvc
Po
PC
hourly concentration of 2,3,7,8-TCDD/2,3,7,8-TCDF, ppm
8-hour PEL for chlorine, ppm (0.5 ppm)
vapor pressure of the pure component at 258C, atm
vapor pressure of chlorine at 25"C, atm (7.9 atm).
I581q
4-9
-------
The inhalation exposure from volatilization for all job categories is
converted from ppm to mg/nr using the equation presented in
Equation 4-3. This value is substituted into Equation 4-4 resulting in a
daily inhalation exposure (Iv) in mg/day.
It is pointed out that all the calculated inhalation exposure levels
from volatilization are biased high because no consideration is given to
the 2,3,7,8-TCDD/2,3,7,8-TCDF binding with organic matter and the
presence of other chemicals in the matrices that could interfere with the
volatilization of the 2,3,7,8-TCDD/2,3,7,8-TCDF. No estimates could be
either found in the literature or provided by contacts in the field which
would allow for quantifying the impact of these interferences on
volatilization.
Cm = Cv M/Vm
Iv = Cm x 1.25 m3/n x ED
(4-3)
(4-4)
where:
Cm = concentration of 2,3,7,8-TCDD/2,3,7,8-TCDF in the vapor, mg/m3
M = molecular weight, g/mole
Vm = molar volume, liter/mole (24.45 liter/mole at T = 25°C and
P = 760 mm Hg)
Iv = daily inhalation exposure from volatilization, mg/day
ED = exposure duration, h/day
The relative toxicity of 2,3,7,8-TCDF with respect to 2,3,7,8-TCDD
can be determined by calculating toxicity equivalents (TEQ) for the daily
exposure. In addition, the percent exposure due to 2,3,7,8-TCDD can also
be calculated. Equation 4-5 presents the equation for calculating TEQ,
while Equation 4-6 is used for calculating the percent of the exposure
due to 2,3,7,8-TCDD. These equations are found in USEPA (1989b).
DTEQv = IvTCDD +0.1 IvTCDF
Iv-,
%TCDDv
'TCDD
IVTCDD + IVTCDF
x 100
(4-5)
(4-6)
where:
DTEQv
IvTCDD
IvTCDF
%TCDDv
daily toxicity equivalents from volatilization of 2,3,7,8-
TCDD and 2,3,7,8-TCDF
daily exposure to 2,3,7,8-TCDD from volatilization, mg/day
daily exposure to 2,3,7,8-TCDF from volatilization, mg/day
percent of the exposure from volatilization due to 2,3,7,8-
TCDD, %
1581q
4-10
-------
In addition to daily TEQ, the lifetime average daily TEQ (LTEQv) for
workers was also calculated. This is presented in Equation 4-7.
LTEQv = DTEQv x DY x LF/(BW x LE) (4-7)
where:
LTEQv = lifetime average daily TEQ from volatilization, mg/day-kg
DTEQv = daily TEQ from volatilization, mg/day
DY = number of days per year exposed, day/year
LF = number of years of exposure per lifetime, years/lifetime
BW = average body weight for a worker, kg
LE = lifetime expectancy, days/lifetime
The number of years of exposure per lifetime (LF) was assumed to be
40 years and the lifetime expectancy (LE) was assumed to be 25,550 days
(i.e., 70 years). The average body weight for male workers (BW) is 70 kg
and a female worker is 58 kg (NCASI 1988a). It was assumed that the
worker would be in the plant for 250 days per year.
(b) Dermal exposure. Dermal exposure levels to 2,3,7,8-TCDD
and 2,3,7,8-TCDF were computed based on the assumption that workers do
not wear any type of gloves that effectively limit exposure to PCDDs and
pCDFs. However, in actual practice, some operators may wear chemical-
resistant gloves during the handling of the pulp. Since the type of
qlove used, the extent of glove use, and the frequency of glove
replacement could not be determined, consideration in the estimation of
dermal exposure could not be given to the degree of protection provided
by personal protective equipment.
There are a few different approaches available for estimating dermal
exposure. The approach used was that agreed upon by EPA, the Food and
Orug Administration, and the Consumer Product Safety Commission (Babich
et al. 1989) for use in this project. This approach considers the
partitioning of PCDD/PCDF from the appropriate matrix (e.g., soil,
sludges, paper) to a liquid (i.e., water, skin oils, urine, blood) and
percutaneous absorption of PCDDs and PCDFs from the liquid. In this
reference, common assumptions for the assessment of dermal exposure are
presented; however, equations for estimating dermal exposures were not
provided. CPSC supplied three equations for estimating dermal exposure
(CPSC 1989); these equations are for estimating dermal exposure to pulp,
paper, and sludge/soil. The equation for handling wet pulp was selected
and is presented in Equation 4-8.
DEW - DC (ppt) x P (mg/cm3) x FT (cm) x j x AD (h'1)
x S (cm2) x ED (h/day) (4-8)
4-11
-------
where:
DEW = dermal exposure from handling wet pulp, mg/day
DC = 2,3,7,8-TCDD.2,3,7,8-TCDF concentration in the wet pulp, ppt
P = density of the wet pulp, 1,000 mg/cnr
FT = liquid film thickness, 0.025 cm
K = liquid equilibrium partition coefficient of 2,3,7,8- TCDD and
2,3,7,8-TCDF from water, unitless, 13,000 and 29,000,
respectively
AD = absorption coefficient of TCDD/TCDF through the skin,
0.012 hr'1
S = skin surface area, cm2
ED = exposure duration, h/day.
Most of the variables in Equation 4-8 are constants except the skin
surface area and exposure duration. Assumptions were made for these
variables depending on the individual job category.
The concentrations reported in the 104 Mill Study for 2,3.,7,8-TCDD
and 2,3,7,8-TCDF were based on dry weight fraction; however, a wet
weight fraction for 2,3,7,8-TCDD and 2,3,7,8-TCDF was needed for
Equation 4-8. Based on an average pulp composition of 11 weight percent
pulp (NIOSH 1983) and the assumption that the aqueous phase contains
2,3,7,8-TCDD/2,3,7,8-TCDF and water, the weight fraction in the aqueous
phase is 0.12 times the weight fraction on a dry basis.
(2) Pulp drying.
(a) Inhalation exposure. Inhalation is one route of exposure
to PCDDs and PCDFs in pulp-drying operations. During drying of the
pulp, vapors are released into the air. It is estimated that the
pulp-drying and utility operators may be exposed to PCDDs and PCDFs by
inhalation for two hours during their shift because their job activities
are conducted away from the drying machines for a majority of the
shift. The method for estimating inhalation exposure from volatiliza-
tion for this open operation is based on the maximum partial pressure of
2,3,7,8-TCDD and 2,3,7,8-TCDF computed using the ideal gas law. This is
a worst-case approach. In this approach, it is assumed that
2,3,7,8-TCDD, 2,3,7,8-TCDF, pulp and water mixtures have two phases—an
aqueous phase and a solid or pulp phase. Furthermore, TCDDs and TCDFs
are assumed to reside only in the aqueous portion of the mixture. The
aqueous phase is assumed to consist only of TCDDs, TCDFs, and water and
is assumed to behave as an ideal solution and obey Raoult's law. It was
assumed that TCDDs/TCDFs and water are removed from the pulp at rates
that ensure that relative concentrations of these components in the
ideal mixture remain constant throughout the drying phase. The ideal
gas law equation is presented in Equation 4-9. This equation transforms
into Equation 4-10 by solving the right-hand side of the equation for
4-12
ISSlq
-------
moles per volume (nr) and then multiplying both sides of the equation
by the molecular weight to obtain the inhalation exposure concentration
of 2,3,7,8-TCDD/2,3,7,8-TCDF.
PxV=nxRxT (4-9)
Cm = Mxy=Mx-j^rxlxl06 (4-10)
where:
P = partial pressure of 2,3,7,8-TCDD/2,3,7,8-TCDF, atm
V = volume of the gas, nr
n = number of moles, moles
R = ideal gas constant, atm. liter/mole-K (0.0821)
T = temperature of the gas, K
Cm = concentration of 2,3,7,8-TCDD/2,3,7,8-TCDF in the vapor, mg/m3
M = molecular weight, g/mole.
The inhalation exposure concentration in mg/m3 based on the ideal
gas law is then substituted into Equation 4-4 resulting in a daily
inhalation exposure (Iv) in mg/day.
The relative toxicity of 2,3,7,8-TCDF with respect to 2,3,7,8-TCDD
can be determined by calculating TEQ values. The percent exposure due
to 2,3,7,8-TCDD can also be calculated. Using Equations 4-5 and 4-6,
these two variables are calculated.
(b) Dermal exposure. Workers in the pulp-drying operations
are potentially exposed to PCDDs and PCDFs through dermal contact with
the wet or dry pulp. Dermal exposure may occur during pulp sheet
weighing when sheets are added or removed by hand to achieve a
predetermined weight. Fork-truck drivers usually do not handle the dry
pUlp. Dermal exposure to the wet or dry pulp is possible for the
utility workers if personal protective equipment is not worn.
Dermal exposures to PCDDs and PCDFs were based on the worst-case
assumption that pulp-drying workers do not wear any type of gloves that
effectively limits exposure to PCDDs and PCDFs. Pulp-drying operators
are potentially exposed to PCDDs and PCDFs while handling the dry pulp
sheets during weighting operations. This is similar to the handling of
dry paper; therefore, the equation for handling paper was selected for
the pulp-drying workers. Dermal exposure from handling dry paper or
pU1p is presented in Equation 4-11 (CPSC 1989).
DED (mg/day) = DC (ppt) x PW (g) / PS (cm2) x R (h"1) (4-11)
x %AD x S (cm2) x ED (h/day)
4-13
I581q
-------
where:
DED = dermal exposure from handling dry material, mg/day
DC = 2,3,7,8-TCDD/2,3,7,8-TCDF concentration in the dry pulp, ppt
PW = weight of the dry pulp sheets, g
PS = surface area of the dry pulp sheets, cnr
R = rate of transfer from pulp to the skin, h"1
%AD = percent of 2,3,7,8-TCDD/2,3,7,8-TCDF available for dermal
absorption; fractional value used in calculations
S = skin surface area, cnr
ED = exposure duration, h/day.
Utility workers who handle the wet and dry pulp during cleanup of
the production area were assumed to come into contact with the pulp two
hours per day. The worst-case assumption that the workers do not wear
any type of glove that effectively limits exposure to PCDDs and PCDFs
was also used in this exposure assessment. The equation for handling
wet pulp was selected for this operation since the workers have a higher
potential for contacting wet pulp rather than dry pulp. This, equation
was presented previously in Equation 4-8.
Equations 4-5 and 4-6 are used to calculate the TEQ and the percent
exposure due to 2,3,7,8-TCDD. In these two equations, the daily
inhalation exposure from volatilization (Iv) is replaced with the daily
dermal exposure (DEW or DED) from 2,3,7,8-TCDD and 2,3,7,8-TCDF.
(3) Paper-making. Papermaking operations include wet-end
additives/stock preparations, coating preparation, paper machine wet-
end, paper machine dry-end, and off-machine coating. There is potential
for pulp exposure in the papermaking stage of pulp and papermaking
operations. In this stage, bleached pulp containing PCDDs and PCDFs is
processed, and it is assumed that the concentrations of PCDDs and PCDFs
remain constant. PCDDs and PCDFs can also be introduced into this stage
by the use of recycled white water from the paper machine to dilute the
pulp slurry prior to its feeding into the wet end of the papermaking
machine. The additional PCDD/PCDF contribution, however, should be
small in comparison to the concentrations of PCDDs and PCDFs in the
bleached pulp. In the papermaking and finishing operations, a potential
for PCDD and PCDF exposure through dermal absorption exists if workers
come in contact with either dioxin/furan-contaminated wet pulp which is
to be fed into beaters or refining equipment or with the paper products.
The workers in dry-end operations are potentially exposed to PCDDs
and PCDFs through dermal contact with paper products, inhalation of
vapors during drying operations, and inhalation of paper dust during
normal process operations. For a particular plant evaluated in a NIOSH
study, engineering controls and work practices used to prevent or
decrease the amount of dust inhaled included exhaust systems, the use of
dust masks, and cleaning of equipment every shift to remove accumulated
4-14
1581q
-------
paper dust. A NCASI study to determine the particulate size
distribution of paper dust showed that total paper dust mass
concentration levels ranged from 0.2 to 4.90 mg/nr during normal
operations, and 31 to 109 mg/nr at the time of operation during
blowdown (operations requiring approximately 15 minutes per shift for
machinery cleaning) (NCASI 1988b).
(a) Inhalation exposure. Three job categories of workers in
the paper mill (i.e., wet-end operator, dry-end operator, and utility
operator) are potentially exposed to PCDDs and PCDFs.
There are two routes for inhalation exposure: (1) vapors from
volatilization of PCDDs and PCDFs and (2) particulates (dust). The
utility operator and dry-end operator may be exposed by both routes of
inhalation exposure, while the wet-end operator is estimated to be
exposed only from volatilization of PCDDs and PCDFs. It was estimated
that wet-end operators, dry-end operators, and utility operators spend
2, 4, and 6 hours of their shifts, respectively, in areas of the plant
in which there is a potential for inhalation exposure.
(1) Volatilization. The same calculation procedures
previously employed to estimate inhalation exposure to PCDDs and PCDFs
by vaporization (using Equations 4-4 and 4-10) for pulp drying workers
was used to estimate inhalation exposure to paper mill workers.
(11) PartIculate matter. Equation 4-12 was used to calculate
the inhalation exposure from PCDD and PCDF contained in particulate
matter generated during paper mill operations. This equation is similar
to Equation 4-4 except that the Cm in this equation is for the total
particulate concentration rather than the 2,3,7,8-TCDD/2,3,7,8-TCDF
concentration. In the absence of data on 2,3,7,8-TCDD/2,3,7,8-TCDF
concentrations in the paper dust, the fraction of
2,3,7,8-TCDD/2,3,7,8-TCDF in the dry pulp was used to allocate the
portion of the paper dust which is 2,3,7,8-TCDD/2,3,7,8-TCDF. The dry
I/eight fraction of 2,3,7,8-TCDD/2,3,7,8-TCDF in the pulp, as reported in
the 104-Mill Study, was used in the calculation because the paper dust
that is emitted is dry.
Ip = Cm (mg/m3) x 1.25 m3/h x ED (h/day) x W (4-12)
where:-
Ip = daily inhalation of particulate matter, mg/day
Cm = concentration of paper mill dust, mg/m3 (with existing
engineering controls)
ED = exposure duration, h/day
W = weight fraction 2,3,7,8-TCDD/2,3,7,8-TCDF in the paper dust.
4-15
-------
Equations 4-5 and 4-6 are used to calculate the TEQ and the percent
exposure due to 2,3,7,8-TCDD. In these two equations, the daily
inhalation exposure from volatilization (TEQv) for 2,3,7,8-TCDD and
2,3,7,8-TCDF are replaced with the daily inhalation exposure from
particulate matter (TEQp) for 2,3,7,8-TCDD and 2,3,7,8-TCDF.
(b) Dermal exposure. The same calculation procedures
previously employed to estimate dermal exposure levels for the pulp mill
operators and pulp-drying workers were used to estimate the dermal
exposure levels for the paper mill operators.
The wet-end operator responsibilities include pulping, bleaching,
beating, refining, application of additives, and blending
proportioning. Dermal exposure duration was based on engineering
judgment. The equation for handling wet material was selected for this
operation, since the worker will be handling the wet end of the process
line. This equation was presented in Equation 4-8.
The dry-end operator responsibilities include supercalendering,
rewinding, slitting, cutting and sheeting, trimming, packaging, and
shipping. Some shipping and packaging operations may be performed in a
separate part of the same plant. Dermal exposure duration was based on
engineering judgment. The equation for handling dry material was
selected for this operation, since the worker handles dry paper. This
equation was presented in Equation 4-11.
The utility operator responsibilities include machine cleaning
(blowdown), and assisting dry-end and wet-end operators. Dermal
exposure duration was based on engineering judgment. The equation for
handling dry material was selected for this operation, since the worker
has a higher potential for contacting dry paper than wet paper. This
equation was presented in Equation 4-8.
(4) Paper converting. There is potential for PCDD and PCDF
exposure in converting operations. The potential for dermal exposure
arises from manual handling of dry bleached paper-based stock and
finished product which contain PCDDs and PCDFs, whereas the potential
for inhalation exposure arises from dry paper dusts created during the
various converting operations. In the finishing stages, dermal contact
may occur through arm and other skin surfaces during the following
building operations: (1) building of the reel; (2) changing of the
roll; and (3) trimming, cutting, transporting, wrapping, or packaging of
the paper. For certain activities during converting operations (e.g.,
quality assurance), workers may also have skin contact with the final
paper product.
Data from a NIOSH study of worker exposures to paper dust from the
dry end of paper machines producing tissue paper, paper towels, and
newsprint were used. The following is an estimation of inhalation
exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF for pulp mill workers.
4-16
1581q
-------
(a) Inhalation exposure. The various job categories in the
converting operation include machine operators; 3rd, 4th, and 5th hands;
back tenders; slitters; and cutters. In the paper converting industry,
dust generated during cutting and trimming operations is the only
potential route for inhalation exposure. All operators in the
production area were assumed to be exposed to the paper dust for the
entire 8 hours of their shift.
The same calculation procedure previously employed to estimate
inhalation exposure from dust for the paper mill operators was used to
estimate the inhalation exposure levels from dust for the paper
converting workers. Values for the dust concentration were calculated
for an 8-hour exposure period from concentrations given in the NIOSH
study.
(b) Dermal exposure. Dermal exposure to 2,3,7,8-TCDD and
2,3,7,8-TCDF was based on the assumption that workers do not wear gloves
that effectively limit exposure. The same calculation procedure as that
used for the paper mill operators' dermal exposure was used to estimate
the dermal exposure levels for the paper converting operator. Dermal
exposure was based on engineering judgment. The equation for handling
paper was selected for this operation since the worker will be handling
dry paper rather than wet pulp. This equation was presented in Equation
4-11-
(5) Nonwoven Industry. Workers in the nonwoven industry are
potentially exposed to PCDDs and PCDFs through the handling of the
pulp. Inhalation of dust may be possible during the machining of pulp;
however, area dust samples for nonwoven operations are not available to
identify any potential for dust inhalation.
Inhalation exposure to all workers in a nonwovens manufacturing
facility may occur because of the mechanical processing of the dry pulp,
which may create pulp dust in the workplace. Plants which produce
personal and medical hygiene products such as diapers and surgical masks
are required to follow Food and Drug Administration regulations on the
amount of allowable area dust (Cunningham 1990). Typically, pulp dust
resulting from the hammer mill is reclaimed through a vacuum screen
formed by a vacuum filter. The filter has a 99+ percent efficiency and
can filter particles as low as 1 to 1-1/2 microns. These filters
recycle the pulp dust to the hammer mill (Lammers 1989). Dermal
exposure may occur to the operator feeding the pulp sheet manually into
the hammer mill, but in automated nonwoven facilities, the pulp is
machine-fed into the hammer mill, and therefore there is no skin contact
With the pulp. Non information was available on the number of manual
versus automated plants (Cunningham 1990).
4-17
1581P
-------
(a) Inhalation exposure. Inhalation exposure to PCDDs and
PCDFs was based on the assumption that workers do not wear respiratory
protection. The same calculation method as that used for paper
converting operators was used to estimate the exposure levels for
nonwoven workers. Exposure duration was assumed to be the entire 8-hour
shift since no data were available.
(b) Dermal exposure. The same calculation method as that
used for pulp mill operators' dermal exposure was used to estimate the
exposure levels for the nonwoven workers. Since worker activity data
were not available, it was assumed that the nonwoven worker would come
into contact with the pulp 6 hours per day. Dermal exposure duration
was based on engineering judgment, since no data were available.
(6) Commercial users. During the commercial use of paper
products, workers may be exposed to PCDDs and PCDFs through dermal
contact with bleached paper products. The skin surface area contacting
the bleached paper product and the amount of contact time varies with
each job category. Almost all workers contact paper at some point in
the work day. A wide variety of worker categories including
secretaries, librarians, teachers, and accountants use various types of
paper products for a large portion of the work day. Since many of the
variables for the calculation of dermal exposure are not known, some
assumptions were made. The number of workers in certain job categories
was obtained from the United States Census Bureau and was the most
complete and recent data. The concentration of 2,3,7,8-TCDD and
2,3,7,8-TCDF was estimated to be that of the pulp concentration because
the PCDD and PCDF levels in paper products measured in the NCASI study
was not representative of the entire paper industry. For medical
workers, the concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF was assumed
to be half since it was assumed that the nonwoven garments these workers
contact are 50 percent pulp/50 percent textile fibers. The surface
areas contacted by nonwoven garments and masks were assumed to be half
of the face, and the entire palm and finger surfaces. Nonwoven garments
worn by the medical workers were assumed to be used as a covering over
clothing or other cloth garments. No data were available for exposure
frequency and duration except for clerical workers and managers. All
other exposure durations were based on engineering judgment and general
knowledge of job related tasks involving the handling of paper and
nonwoven products. The skin contact area for workers who are required
to handle sheets of paper was assumed to be 20 percent of the total of
palm and finger surfaces of both hands. These values were derived from
studies done by NCASI on typical commercial users of paper products.
The equation for handling dry material was used for commercial users of
paper since they handle the dry paper. This equation is presented in
Equation 4-11.
4-18
1581q
-------
Several of the job categories were combined because exposures were
assumed to be similar. In Group 1, accountants, auditors, architects,
librarians, archivists, and duplicating and mail/message distribution
personnel were assumed to have similar exposures; Group 2 is made up of
lawyers, judges, computer programmers and operators, record processors,
management, and miscellaneous administrative support personnel.
Secretaries, stenographers, and typists make up Group 3; Group 4
includes teachers and sales workers; and Group 5 includes medical
workers.
4.2.3 Summary of Worker Inhalation and Dermal Exposure, Individual
Cancer Risks, and Population Cancer Risks
Dioxins are formed during'the production of bleached pulp, thereby
resulting in a potential for worker exposure to these chemicals in the
production, processing, and commercial use of pulp, paper, and paper
products. Studies have shown that the use of chlorine and chlorine-based
bleaching agents cauld result in the generation of dioxins such as
2,3,7,8-TCDO, 2,3,7,8-TCDF, and 1,2,7,8-TCDF. Concentrations of dioxin
isomers have been detected in the bleached pulp and in the paper
products themselves. Results from the 104 Mill Study showed that dioxin
concentrations in bleached pulp ranged from 0.10 to 49 ppt for
2,3,7,8-TCDD and 0.25 to 2620 ppt for 2,3,7,8-TCDF.
Pulp and paper mill workers may be exposed to 2,3,7,8-TCDD and
2,3,7,8-TCDF primarily via two routes: 1) dermal contact with the
bleaching wastewaters, bleached pulp or paper products; and 2) inhalation
of paper dusts which are created during converting, rewinding, sizing,
pulp-fluffing, cutting, or other operations. Inhalation of vapors is
not a major route of exposure since 2,3,7,8-TCDD and 2,3,7,8-TCDF have a
low vapor pressure and tend to bind with organic matter. The extent of
dermal exposure to workers varies depending on the worker job category.
Worker contact with the raw bleached pulp stock is minimal since many
processes are automated. Brown stock washing operators, pulp testers, or
utility employees can be exposed through dermal contact with either the
brown stock washed with recycled paper machine white waters containing
2,3,7,8-TCDD and 2,3,7,8-TCDF formed during bleaching operations or
directly with the recycled paper machine waters. Operators, tenders,
helpers, and utility employees in pulp manufacturing plants are also
potentially exposed through dermal contact with the bleached pulp and
filtrate during bleaching operations. The potential for inhalation
exposure during pulp manufacturing is low because the bleaching towers
ar-e often installed outdoors and the operations are well controlled.
£xposure through dermal contact with the bleached pulp may also occur
during paperrnaking and finishing operations. Specific worker job
categories with the potential for exposure include operators,
assistants, and utility employees. In the paperrnaking operations, the
wet-end operators, assistants, and utility employees are potentially
exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF through dermal contact with
4-19
-------
pulp. The dry-end paper machine operators, assistants, and utility
employees may be exposed through the inhalation of paper dust and dermal
contact with the paper products. Both of these routes of exposure are
also applicable to workers in converting operations.
Table 4-3 summarizes the individual risks and population risks for
workers involved in pulp manufacture, pulp-drying, paper manufacture,
paper converting, and nonwovens production.
4.3 Worker Exposure from Processing and Commercial Use of Pulp and
Paper Mill Sludge
This section assesses worker exposure to 2,3,7,8-TCDD and
2,3,7,8-TCDF during the processing and commercial use of sludge
generated from wastewater treatment operations in pulp and paper mills.
Section 4.3.1 provides a discussion of industrial processes and
operations that may result in exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF
during the processing and commercial use of pulp and paper mill
sludges. Section 4.3.2 presents the results of studies that have
measured contaminant concentration in sludges. Section 4.3.3 provides a
characterization of the worker population for different operations
involving the processing and commercial use of pulp and paper mill
sludge. Section 4.3.4 discusses the potential for dermal and inhalation
exposure to dioxins. Section 4.3.5 provides a summary of the estimates
of the levels of inhalation and dermal exposure for workers involved in
different operations, individual cancer risks, and population cancer
risks.
4.3.1 Sludge Formation, Processing, and Disposal Operations
This section describes sludge formation, processing, and disposal
operations which may be employed at pulp and paper mills.
(1) Sludge formation and processing. Two kinds of sludge are
generated by wastewater treatment at pulp and paper mills: primary and
secondary sludge. Primary sludges consist of solids which are lost from
the pulp and paper manufacturing process and are subsequently removed by
primary clarification (Ledbetter 1976). These solids are composed of
fibers, clay filler materials, coating clays, and other chemical
additives (Kirk-Othmer 1981a). In addition to fibers, other organic
components such as wood dust, fiber debris, starches, dextrine, resins,
and protein may be present. Typical water contents of nondewatered
primary sludges from pulp and paper manufacturing operations may range
from 90 to 98 percent (Ledbetter 1976). Secondary sludges are largely
biological in nature and are harder to handle and dewater (Kirk-Othmer
1981a). Secondary sludge water contents may range from 98 to
99.5 percent (Hammer 1975).
4-20
1581q
-------
Table 4-3. Summary of Individual and Population Cancer Risks for Workers Involved in Manufacturing,
Processing, and Commercial Usage of Pulp, Paper, and Paper Products
a
Estimated risk
No. of
Job category workers
>u Ip manufacturing operations
•Bleach plant operator 434
•Pulp testers 433
• Utility operator 433
Pulp drying operations
• Pulp drying operator 160
•Pulp drying utility operator 80
P^per manufacturing operations
•Wet-end operator 10,667
Individual risk
Exposure pathway0
Inhalation-volati 1 ization
Dermal
Inhalation- volatilization6
Dermal1
Inha lat ion- vo lat i 1 i zat ion 1
Dermal'
Inhalation-volatilization
Dermal
Inhalation-volati 1 ization
Dermal
Inha lation-volatil ization
Derma lf
Low
4xlO'7
(0.08)
2xlO"13
(47)
IxlO'19
(0.03)
ZxlO"12
(47)
IxlQ"6
(0.08)
4xlO~12
(47)
2xlQ"U
(0.03)
9xlQ-10
(29)
ZxlO"11
(0.03)
2xlO"13
(29)
ZxlO"11
(0.03)
ZxlO"13
(47)
High
SxlO"7
(0.08)
5xlO"10
(4)
2xlO"15
(0.002)
3xlO"9
(4)
IxlO"6
(0.08)
7xlO"9
(4)
2xlO"7
(0.002)
3xlO"6
(2)
2xlO"7
(0.002)
5xlO"10
(2)
2xlO"7
(0.002)
5xlO-10
(4)
5
Population risk
i cases/ vr
Low
4xlO"6
3xlO"12
IxlO"18
2xlO"U
IxlO"5
4xlO"H
7xlO"U
4xlO"9
SxlO"11
SxlO"13
SxlO"9
7x10" »
High
SxlO"5
SxlO"9
3X10'14
4xlO"8
2xlO"5
7xlO"8
9xlO"7
lxlO"S
4xlO"7
IxlO"9
6xlO"5
IxlO"7
4-21
-------
9058H
Table 4-3. (Continued)
a
Estimated risk
No. of
Job category workers
Paper manufacturing operations
(continued)
- Dry-end operator 12,445
- Utility operator 8,888
Paper converting operations
- General worker 129.000
Nonwoven operations
- General worker 15,000
Individual risk
Exposure pathway Low
Inhalation-volatilization^ 3x10" H
(0.03)
Inhalation-particulate matterh 2xlO"10
(29)
Dermalh 2xlO~10
(29)
Inhalation-volati lization' 5x1 0
(0.03)
p _g
Inhalation-particulate matter 8x10
(29)
Dermal1 3xlO"10
(29)
k - 10
Inhalation-particulate matter 4x10
(29)
Dermalk 4xlO"10
(29)
Inhalation-particulate matter 4x10
(29)
Dermal' 8x10" 13
(29)
High
4xlO"7
(0.002)
IxlO"5
(2)
7xlO~7
(2)
7xlO"7
(0.002)
8xlO~5
(2)
IxlO"6
(2)
5xlO"5
(2)
lx!0"6
(2)
3xlO'6
(2)
3xlO~9
(2)
b
Populat ion risk
# cases/vr
Low High
IxlO"8 IxlO"4
6xlO"8 5xlO"3
6xlO"8 2xlO"4
IxlO"8 IxlO"4
2xlO"6 2xlO"2
6xlO"8 2xlO"4
IxlO"6 2X10"1
IxlO"6 4xlO"3
IxlO"7 IxlO"3
3xlO"10 IxlO'6
4-22
-------
2HH
Table 4-3. (Continued)
Estimated risk
Job category
. , d
juercial users
•Group 1
•Group 2
•Sroup 3
•Group 4
•Group 5
No. of
workers Exposure pathway
2,639,000 Dermal1
26,933,000 Derma lh
5,004,000 Dermal9
14.095,000 Derma lf
793,000 Derma lf
Individual risk
Low
7X10'11
(29)
5X10"11
(29)
2xlO"n
(29)
ZxlO"11
(29)
3X10'10
(29)
High
3xlO"7
(2)
2xlO"7
(2)
8xlO"8
(2)
SxlO"8
(2)
8xlO"7
(2)
b
Population risk
# cases/vr
Low High
SxlO"6 2xlO~Z
3xlO"6 IxlO'2
IxlO"6 SxlO"3
2xlO"6 6xlO"3
ZxlO"5 5xlO~2
curators, and duplicating and mail/message
records processing, management, miscellaneous
"Values in parentheses represent percent exposure to 2,3.7,8-TCDD; risk estimates are based on EPA's slope factor..
Values represent cases per year.
''he frequency of exposure assumed to be 250 days per year.
Wip 1 includes accountants, auditors, architects, librarians, archivists
distribution occupations.
Group 2 includes lawyers, judges, computer programmers, computer operators
idninistrative support occupations.
Group 3 includes secretaries, stenographers, and typists.
Group 4 includes teachers and sales representatives.
Group 5 includes medical workers who may come in contact with nonwoven products such as garments and masks.
Vat ion of exposure assumed to be 1 hour per day.
Wation of exposure assumed to be 2 hours per day.
'duration of exposure assumed to be 3 hours per day.
Wat ion of exposure assumed to be 4 hours per day.
Wation of exposure assumed to be 6 hours per day.
'Duration of exposure assumed to be 7 hours per day.
Wation of exposure assumed to be 8 hours per day.
4-23
-------
Primary and secondary sludges must be processed for volume reduction
prior to disposal. Processing techniques which involve sludge volume
reduction include thickening, conditioning, and dewatering (Kirk-Othmer
1981a). Thickening is primarily accomplished using either gravity
settling, which is applied to mixtures of primary and secondary sludges,
or dissolved air flotation which is usually applied to secondary sludges
only. Conditioning is performed to improve the sludge dewatering
characteristics. Dewatering is accomplished using centrifugation, vacuum
filtration, or pressure filtration.
No data are available in the literature to characterize the typical
unit operations at the wastewater treatment facilities of pulp and paper
mills. The vast majority employ primary treatment and some form of
secondary treatment along with dewatering of sludges from these
operations.
The mechanism by which contamination from 2,3,7,8-TCDD and
2,3,7,8-TCDF occurs in pulp and paper mill sludges is currently being
investigated by the paper industry. 2,3,7,8-TCDD and 2,3,7,8-TCDF are
known to be formed during chlorine or chlorine derivative bleaching
operations employed in pulp mills, and may be present in bleach plant and
paper machine wastewater. 2,3,7,8-TCDD and 2,3,7,8-TCDF have a strong
affinity for organic matter, and binding with organic matter in primary
and secondary sludge may occur (Olson 1988).
(2) Sludge disposal operations. Disposal methods for sludge from
pulp and paper mills include landfilling, incineration, surface
impoundments, land application, and distribution as a salable product.
Landfilling is the predominant disposal operation of choice currently in
the industry, although a number of mills are also using incineration and
surface impoundments.
(a) Landfilling. Landfilling involves the disposal of sludge
in company-owned, publicly-owned, or privately-owned landfills. In this
disposal operation, sludge is almost universally dewatered to a solids
content of 20 to 25 percent to reduce its volume and lower transportation
costs (USEPA 1979). Typically, the dewatered sludge is transferred from
the dewatering operations to storage or directly to heavy duty dump
trucks. If storage operations are used, the dewatered sludge will be
loaded into dump trucks by a front-end loader (Ledbetter 1976). The
dewatered sludge is hauled to the landfill where the basic operations
involve spreading, compacting, and covering the sludge with excavated
soil daily (Hammer 1975).
(b) Incineration. Incineration is a two-step oxidation
process involving drying of the sludge followed by combustion. This
process may occur in separate pieces of equipment or successively in a
single unit (USEPA 1979). Incineration converts the sludge into an inert
ash which is handled in wet or dry form (Hammer 1975). In wet form, the
4-24
1581q
-------
ash is slurried with scrubber water and is pumped to an ash lagoon. In
dry form, the ash is conveyed mechanically by a bucket elevator to a
storage hopper for discharge into a dump truck for eventual disposal in a
landfill (Metcalf & Eddy 1979).
According to the 104-Mill Study data, all mills that practice sludge
incineration burn the sludge in waste fuel- or combination-power
boilers. Potential exposures and risks resulting from pulp and paper
mill sludge incineration are addressed in Section 7 of this report.
(c) Land application. Land application involves disposal of
sludge by surface spreading or subsurface injection, usually on
agricultural lands. Dewatered sludge is usually applied to land by a
conventional fertilizer or manure spreader. Liquid sludges are surface
spread by tank truck or injected below the surface (Metcalf & Eddy
1979). Surface spreading may require movement and burying of the sludge
below the soil by plows, graders, or bulldozers (USEPA 1979).
(d) Distribution of sludge as a salable product. Sludge is
processed for distribution as a salable product by heat-drying or
composting techniques. The sludge is usually dewatered prior to
processing to reduce the water content (USEPA 1979). Heat-drying is
employed to remove moisture from sludge so that it can be efficiently
processed into fertilizer. Heat-drying is also employed prior to
incineration to ensure proper combustion of the sludge. The drying of
the sludge is necessary in fertilizer manufacturing to permit grinding of
the dried sludge.
Composting is undertaken to biologically degrade the sludge into a
stable end product which is used as a soil conditioner. The composting
process involves three steps: (1) preparation of the wastes to be
composted, (2) decomposition of the prepared wastes, and (3) preparation
and marketing of the product.
4.3.2 Dioxins In Pulp and Paper Mill Sludges
Sampling of sludge from pulp and paper mills that produce and/or
process bleached pulp and bleached pulp-based products has shown that
2 3,7>8-TCDD and 2,3,7,8-TCDF are predominantly present in primary and
secondary sludges formed during wastewater treatment operations.
2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations are much higher in
secondary sludges as compared to primary and combined dewatered sludges.
Combined dewatered sludges, which are a mixture of secondary and primary
sludge, have higher 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations than
primary sludges. Concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF may be
higher in secondary sludges as compared to primary sludges because
secondary solids are comprised of organic biological solids for which
pCDDs and PCDFs have an affinity.
4-25
J58lq
-------
The concentrations of 2,3,7.8-TCDF are typically greater than those
of 2,3,7,8-TCDD in such sludges. Median concentrations of 2,3,7,8-TCDD
in combined dewatered sludges were 27.5 and 36.0 ppt for the 5-Mill Study
and the 104-Mill Study, respectively, whereas the corresponding median
concentrations of 2,3,7,8-TCDF were 359.5 and 149.0 ppt, respectively.
It must be noted, however, that concentrations of 2,3,7,8-TCDD and
2,3,7,8-TCDF measured in the various sludges do not necessarily represent
worker exposure to these compounds during sludge handling, processing,
and disposal operations. Worker exposure depends on a variety of
conditions (e.g., engineering controls, personal protective equipment,
work practices) which are addressed in Section 4.3.4.
4.3.3 Sludge Handling/Disposal Workforce Characterization
No previous studies have been conducted specifically on the workforce
population in sludge handling and disposal operations at pulp and paper
mills (Fisher 1989). The workforce for these operations were estimated
based on parallels with municipal sewage sludge handling and disposal
operations. Many sources were used to estimate the number of workers.
These sources included EPA literature, NIOSH data bases, industry data,
and discussions with NCASI and the United Paperworkers International
Union (UPIU).
(1) Sludge formation and processing operations. Three job
categories of workers are postulated for the sludge formation and
processing operations: waste treatment plant operators, sludge haulers,
and front-end loader operators.
Sludge processing operations will involve preparation of the waste
sludge for disposal. In most cases, the sludge is loaded into dump
trucks by conveyors or a front-end loader and is hauled away for
disposal. Typical activities of sludge haulers and front-end loader
operators will center around their respective equipment. Front-end
loader operators will operate, maintain, and clean their equipment, if
necessary. The front-end loader operator drives the front-end loader and
transfers sludge from storage piles into dump trucks. The sludge hauler
drives the dump truck and may maintain and clean it if necessary. The
sludge hauler will transport the sludge to the disposal site. The sludge
hauler and front-end loader operator population exposed to 2,3,7,8-TCDD
and 2,3,7,8-TCDF is estimated to be approximately 400 workers.
(2) Sludge disposal operations. The number of workers involved
with sludge disposal operations depends on the specific disposal
technique applied by the facility, the quantity of sludge disposed and
the location of final disposal.
4-26
1581q
-------
(a) Landfill ing operations. Landfill ing operations may
involve pulp and paper mill employees if sludge is landfilled on
company-owned land, or employees of waste disposal firms if the sludge is
landfilled at a site not affiliated with the pulp and paper mill. Only
one job category of workers is postulated for landfill ing operations:
equipment operators.
At the landfill site, the number of equipment operators depends on
the landfill technique used and quantity of sludge to be disposed. The
total pieces of equipment may range between one and five and may even be
greater in the case of large quantities of sludge (USEPA 1979). The
landfill equipment operator population exposed to 2,3,7,8-TCDD and
2,3,7,8-TCDF is estimated to be approximately 400 workers.
(b) Incineration. All of the mills involved in the 104-Mill
Study which incinerate sludge do so on-site, and thus sludge incineration
operations will involve only pulp and paper mill employees. Disposal of
residual bottom ash and fly ash, however, will also involve landfilling
personnel as previously described. Sludge is typically burned with wood
waste and fuel in power boilers for energy recovery. Thus, incineration
of sludge and subsequent disposal of bottom ash and fly ash will involve
power plant operators and maintenance staff. The number of workers will
depend on the size of the mill and the number of boilers operated.
(c) Land application. One job category of workers is
postulated for land application operations: equipment operators. Sludge
application will involve equipment operators to operate fertilizer
spreaders, tank trucks, plows, bulldozers, or other equipment depending
on the type of sludge application. The number of equipment operators
Wi11 depend on the quantity of sludge to be disposed, the sludge
application techniques, and whether one operator is exclusively dedicated
to one piece of equipment. The land application equipment operator
population exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF is estimated to be
approximately 20 workers.
(d) Distribution of sludge as a salable product. Sludge is
converted into a salable product by heat-drying or composting. Heat-
drying will require skilled operators for the dryers (USEPA 1979). Other
empl°yees rcay D6 used to aid in conveying and bagging of the dried sludge
e
nrior to shipment to fertilizer manufacturers. The fertilizer
manufacturing operations will employ various personnel in the processing
Of the dried sludge.
Composting operations may involve personnel not affiliated with the
niilP ana< PaPer miH wnen composting is done off-site, as is generally the
case- There may be three job categories of workers in composting
operations: equipment operators, screen operators, and compost haulers.
FauiPment operators perform unloading and placement of the sludge into
Windrows or aerated piles. Equipment operators remove the compost from
4-27
-------
the windrows and piles and load it into screens. They also load the
finished compost into dump trucks for distribution. If the compost is
screened, an equipment operator and screen operator are necessary (USEPA
1979). Other personnel may be required if the compost is ground, blended
or further processed in some manner prior to distribution. The screen
operator is responsible for proper maintenance and operation of the
screen, e.g., unplugging the screen if the compost is too wet. The
compost hauler transports the finished product to farms or other
facilities which may further process the compost.
The equipment operator population exposed to 2,3,7,8-TCDD and
2,3,7,8-TCDF is estimated to be approximately 150 workers. The screen
operator population exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF is estimated
to be approximately 20 workers. The compost hauler population exposed to
2,3,7,8-TCDD and 2,3,7,8-TCDF is estimated to be approximately 50 workers.
4.3.4 Worker Exposure Estimating Methodologies
Although considerable data have been collected on concentrations of
2,3,7,8-TCDD and 2,3,7,8-TCDF for pulp, sludge, and wastewaters in the
pulp and paper industry, no inhalation or dermal exposure data for
2,3,7,8-TCDD and 2,3,7,8-TCDF are currently available. Furthermore,
little information is available on the effectiveness of engineering
controls or the use of personal protective equipment in this industry.
Therefore, modeling techniques and assumptions were used to estimate
worker exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF.
Workers processing and commercially using pulp and paper sludge may
be exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF either through dermal contact
with the sludge or through inhalation exposure of either volatilized
2,3,7,8-TCDD/2,3,7,8-TCDF or particulates containing these chemicals.
The potential for volatilization exposure should be minimal because these
chemicals have a low vapor pressure and have a tendency to bind with
organic matter rather than volatilize freely. The inhalation exposure
levels computed in this report provide an estimate of the quantities of
2,3,7,8-TCDD and 2,3,7,8-TCDF that would freely volatilize; the values
hence represent worst-case estimates since the effects on volatilization
of 2,3,7,8-TCDD and 2,3,7,8-TCDF due to binding with organic matter and
from interferences due to other chemicals present in the sludge are not
considered. There is some potential for inhalation exposure from
particulates during sludge handling operations such as loading/unloading,
spreading, compacting, plowing into soil, and composting. There is some
potential for dermal exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF through
handling of contaminated sludge, although this would be minimal because
most operations are highly automated with generally little contact
between the sludge and employees except in cases of emergency maintenance
and cleanup. The modeling technique presented in this section for sludge
handling and processing were also used for all other sludge disposal
operations in the subsequent section.
4-28
-------
(1) Sludge formation and processing. In sludge formation and
processing, the potential for 2,3,7,8-TCDD and 2,3,7,8-TCDF exposure for
waste treatment plant operators exists through dermal contact with sludge
when taking samples and when performing maintenance. Dermal contact is
otherwise limited by the high degree of automation of sludge processing
operations. Contact with sludge can occur through cleanup of spills of
wet sludge or dewatered sludge that has fallen off the conveyors.
Cleanup is usually done with shovels, minimizing dermal contact. Repair
of dewatering equipment, pumps, and conveyors may also result in dermal
contact. Since the process is highly automated and cleanup is usually
done with a shovel, dermal exposure should be minimal (Hawks 1989).
Therefore, as a worst case, it was estimated that typical maintenance and
cleaning activities may result in 2 hours of dermal contact with the
sludge during a shift.
The potential for inhalation exposure to waste treatment operators
exists through volatilization of 2,3,7,8-TCDD and 2,3,7,8-TCDF in the
sludge and inhalation of particulate matter generated during sludge
handling operations. Waste treatment plant operators may be exposed to
2,3,7,8-TCDD and 2,3,7,8-TCDF through volatilization when inspecting
and/or repairing equipment, taking samples, and cleanup of spilled
sludge. Waste treatment plant operators may also be exposed to
2,3,7>8-TC9D and 2,3,7,8-TCDF through the generation of particulate
matter during sludge handling operations. The quantity of particulate
matter generated is estimated to be minimal because of the high moisture
content of the sludge (approximately 70 percent).
Sludge haulers and front-end loader operators may be exposed to
2 3,7,8-TCDD and 2,3,7,8-TCDF through handling and loading of the
sludge. Exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF may occur through
inhalation of particulate matter generated during sludge handling and
loading, volitilization from the wet sludge, and dermal contact during
maintenance activities.
(a) Inhalation exposure. In sludge handling/processing
operations, there are three job categories of workers (waste treatment
olant operators, sludge haulers and front-end loader operators) who are
potentially exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF through inhalation
of vapors and particulate matter containing dioxin and dibenzofuran.
(i) Volatilization. There are no existing data available to
determine inhalation exposure from volatilization of
? 3,7,8-TCDD/2,3,7,8-TCDF during sludge handling/processing. On-site
7 3*7,8-1000/2,3,7,8-TCDF concentrations in air were estimated based on a
"box model" approach, which was used to simulate the dilution impacts of
wind in an outdoor setting, by using Equation 4-13.
Cm " (L) (HH) (V) x 100° (4
4-29
1581P
-------
where:
Cm = 2,3,7,8-TCDD/2,3,7,8-TCDF vapor concentration in air, mg/m3
G = generation rate of 2,3,7,8-TCDD/2,3,7,8-TCDF, g/sec
L = equivalent side length of the area source perpendicular to the
direction of the winds, m
MH = mixing height of vapors before inhalation by an individual, m
V = average wind speed at the inhalation height, m/sec.
The average wind speed chosen was 2.2 m/s and the mixing height
chosen was 1.5 m (USEPA 1988b). The "box model" approach used assumes
that 2,3,7,8-TCDD/2,3,7,8-TCDF present at the site travels only a short
distance before it is inhaled by the worker.
There are many models which can be used to estimate generation
rates. The selected model is based on a set of equations presented in
USEPA (1986), and Hwang and Falco (1986) as described in EPA (1988b).
The daily inhalation exposure from volatilization is then calculated
using Equation 4-14. It should be noted that the calculated inhalation
exposure levels from volatilization are biased high because no percent
reductions are incorporated in the calculations for 2,3,7,8-TCDD/
2,3,7,8-TCDF binding with organic matter and the presence of other
chemicals in the matrices that could interfere with the volatilization of
the 2,3,7,8-TCDD/2,3,7,8-TCDF. No estimates could be found in the
literature or provided by knowledgeable contacts in the field regarding
the quantitative reductions in volatilization from such interferences.
Iv = Cm x 1.25 m3/h x ED (4-14)
where:
Cm = concentration of 2,3,7,8-TCDD/2,3,7,8-TCDF in the vapor,
mg/irr
Iv = daily inhalation exposure from volatilization, mg/day
ED = exposure duration, h/day
The relative toxicity of 2,3,7,8-TCDF with respect to 2,3,7,8-TCDD
can be determined by calculating toxicity equivalents (TEQ). In
addition, the percent exposure due to 2,3,7,8-TCDD can also be
calculated. Equation 4-15 presents the equation for calculating TEQ
while Equation 4-16 is used for calculating the percent of the exposure
due to 2,3,7,8-TCDD. These equations are found in USEPA (1989b).
DTE(JV= IvTCDD + (M IvTCDF (4'15)
%TCDDv = j- p^fi x 100 (4
1VTCDD + 1VTCDF
4-30
158iq
-------
In addition to daily TEQ, the lifetime average daily TEQ (LTEQv) for
workers was also calculated. This is presented in Equation 4-17:
LTEQv = DTEQv x DY x LF/(BW x LE) (4-17)
where:
LTEQv = lifetime average daily TEQ from volatilization, mg/day-kg
DTEQv = daily TEQ from volatilization, mg/day
DY = number of days per year exposed, day/year
LF = number of years of exposure per lifetime, years/lifetime
BW = average body weight for a worker, kg
LE = lifetime expectancy, days/lifetime
The number of years of exposure per lifetime (LF) was assumed to be
40 years and the lifetime expectancy (LE) was assumed to be 25,550 days
(i.e., 70 years). The average body weight for male workers (BW) is 70 kg
and a female worker is 58 kg (NCASI 1988c). It was assumed that the
worker would be in the plant for 250 days per year.
(ii) Inhalation exposure from particulate matter. Equation
4-18 was used to calculate the inhalation exposure from 2,3,7,8-TCDD and
2,3,7,8-TCDF contained in particulate matter generated during sludge
handling/processing operations. This equation is similar to Equation
4_14 except that Cm in this equation represents the total particulate
concentration rather than the 2,3,7,8-TCDD and 2,3,7,8-TCDF concentration.
Ip = Cm (mg/m3) x 1.25 m3/h x ED (h/day) x WF (4-18)
where:
Ip = daily inhalation from particulate matter, mg/day
Cm = concentration of particulate matter, mg/m3
WF = weight fraction of 2,3,7,8-TCDD/2,3,7,8-TCDF in the dry sludge
ED • exposure duration, h/day.
The concentration of 2,3,7,8-TCDD/2,3,7,8-TCDF in the particulate
matter was assumed to be equal to the concentration of
2 3,7,8-TCDD/2,3,7,8-TCDF in the sludge. The dry weight concentration of
2^3,7,8-TCDD/2,3,7,8-TCDF, as reported in the 104-Mill Study, was used in
tf,e calculations because the particulate matter which is emitted is
assumed to be dry.
The concentration of particulate matter was calculated using Equation
4-19:
Cm
(L)(MH)(V)
4-31
i58iq
-------
where:
o
Cm = participate concentration, mg/nr
Qp = total particulate matter emission rate, mg/s
L = equivalent side length of the site perpendicular to the
direction of the winds, m
MH = mixing height of the particulate matter before inhalation by an
individual (assumed to be 1.5 m)
V = average wind speed at the inhalation height (assumed to be 2.2
m/s).
Total particulate emission rates for a given disposal scenario, which
in this case is dewatered sludge loading and unloading, are used in the
equation. The length of the area from which the particulate matter is
being emitted corresponds to the length of an open area of the dump
truck, length of daily landfill and land application areas, and length of
the area of composting activities. Particulate matter emission rate for
sludge handling were calculated using AP-42 emission factors. These
AP-42 emission factors are widely used for estimating particulate
emission rates for similar applications.
(b) Dermal exposure. Dermal exposure levels to 2,3,7,8-TCDD
and 2,3,7,8-TCDF were computed based on the assumption that workers do
not wear any types of gloves that are effective in limiting exposure to
PCDDs and PCDFs. Dermal exposure levels for sludge haulers and front-end
loader operators were assumed to be equal because of their similar job
functions.
There are a few different approaches available for estimating dermal
exposure. The approach selected was agreed upon by EPA, Federal Drug
Administration, and the Consumer Product Safety Commission (Babich et al.
1989). This approach considers the partitioning of PCDD/PCDF from the
appropriate matrix (e.g., soil, sludges, pulp, paper) to a liquid (i.e.,
water, skin oils, urine, blood) and percutaneous absorption of PCDDs and
PCDFs from the liquid. In this reference, common assumptions for the
assessment of dermal exposure are presented. However, in this reference,
equations for estimating dermal exposures were not present. CPSC
supplied three equations for estimating dermal exposure (CPSC 1989).
These equations are for estimating dermal exposure to pulp, paper, and
sludge/soil. The equation for handling wet sludge/soil was selected for
the sludge haulers and front-end loader operator handling wet sludge and
is presented in Equation 4-20.
DEW = DC (ppt) x P (mg/cm3) x FT (cm) x B x AD (h'1) (4-20)
x S (cm2) x ED (hr/day)
4-32
1581q
-------
where:
DEW = dermal exposure from handling wet material, mg/day
DC = adjusted 2,3,7,8-TCDD/2,3,7,8-TCDF concentration to account
for handling of wet sludge, ppt
P = density of the dewatered sludge, mg/cm3
FT = liquid film thickness, cm
B = bioavailability factor for sludge, unitless
AD = absorption coefficient of TCDD/TCDF through the skin, h"1
S = skin surface area, cm2
ED = exposure duration, h/day.
No data were available on duration or extent of dermal exposure.
Therefore, dermal exposure duration was based on engineering judgment.
It was assumed that only the palms and fingers of both hands of the
hauler/front-end loader operator would be in contact with the wet sludge,
whereas it was assumed that both hands and forearms of the waste
treatment plant operator would be in contact with the wet sludge. The
density for dewatered sludge is 1058 mg/cm3 (USEPA 1979) and. a liquid
film thickness for the wet sludge was estimated to be 0.025 cm. It was
estimated that 2,3,7,8-TCDD was absorbed at an average rate of
approximately 0.012 h'1 (AD) over the time period from 0.5 to 17 hours
(Babich et al. 1989).
The relative toxicity of 2,3,7,8-TCDF in respect to 2,3,7,8-TCDD can
be determined by calculating TEQ values. The percent exposure due to
2,3,7,8-TCDD can also be calculated. Equations 4-15 and 4-16 are used to
calculate these two variables. In these two equations, the daily
inhalation exposure from volatilization (Iv) for 2,3,7,8-TCDD and
2 3,7,8-TCDF are replaced with the daily dermal exposure (DEW) for
2,3,7,8-TCDD And 2,3,7,8-TCDF.
(2) Sludge disposal operations. The route and amount of exposures
to 2,3,7,8-TCDD and 2,3,7,8-TCDF in sludge disposal operations will
depend on the specific sludge disposal technique, the amount of sludge
disposed, and the degree of processing of the sludge. In addition,
atmospheric events such as wind and rain will affect the degree of
exposure in sludge disposal operations conducted outdoors.
(a) Landfill ing operations. Inhalation is the primary route
Of exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF in landfilling operations.
During unloading, spreading, compacting, and burying of the sludge with
various pieces of equipment, particulate matter is generated which may be
inhaled by the equipment operators. In addition, 2,3,7,8-TCDD and
2,3,7,8-TCDF may volatilize from the sludge and be inhaled.
Dermal exposure to the sludge is minimal because workers are
qenerally not in contact with the sludge unless maintenance of the
equipment is required.
4-33
-------
(b) Land application. Exposure to 2,3,7,8-TCDD and
2,3,7,8-TCDF through land application is very similar to that of
landfilling operations. Inhalation is the primary route of exposure to
2,3,7,8-TCDD and 2,3,7,8-TCDF in land application operations.
Particulate matter generated during unloading, spreading, and plowing
operations may be inhaled by the equipment operators. In addition,
2,3,7,8-TCDD and 2,3,7,8-TCDF may volatilize from the sludge and
consequently be inhaled.
Dermal exposure to the sludge is minimal because equipment operators
are generally not in contact with the sludge unless maintenance of the
equipment is required.
(c) Distribution of sludge as a saleable product. In
heat-drying operations, which are used to dry the sludge so that it can
be further processed into fertilizer, exposure to 2,3,7,8-TCDD and
2,3,7,8-TCDF may occur primarily through inhalation of particles
entrained in the air during handling and conveying of the dried sludge to
storage and distribution. Inhalation exposure will also occur in
fertilizer manufacturing operations in which grinding, screening, and
bagging of the dried sludge will produce particulate matter emissions.
Dermal exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF may occur at the mills
through cleanup and emergency maintenance, but is more likely in
fertilizer manufacturing operations in which the sludge is much further
processed. Dermal and inhalation exposures for this type of operation
are not estimated because only one of the mills practices flash drying.
Equipment operators may be exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF
by inhalation of particulate matter generated during composting
operations. Particulate matter, which may be inhaled by the equipment
operator, is generated during unloading of sludge, its placement in a
windrow or pile, mixing bulking agents with sludge, turning and mixing of
the compost piles, removal of compost from piles and unloading at the
screening operations, loading compost into the screens, and loading
screened compost into piles and eventually into trucks for distribution.
In addition, the equipment operators may inhale 2,3,7,8-TCDD and
2,3,7,8-TCDF which has volatilized from the sludge.
Dermal exposure of equipment operators to the sludge/compost is
minimal because operators are generally not in contact with the
sludge/compost unless maintenance of the equipment is required.
The screen operators may be exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF
via inhalation of particulate matter generated during screening of the
compost. In addition, exposure may also occur from volatilization of
2,3,7,8-TCDD and 2,3,7,8-TCDF from the compost. Dermal exposure of
screen operators to the compost occurs during routine maintenance of the
screen, such as unplugging, which occurs when the compost is too wet, and
cleaning of any spilled material.
4-34
-------
Compost haulers may be exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF
through loading and unloading of the compost from the dump truck.
Exposure may occur through inhalation of participate matter generated
during compost loading and unloading, and from volatilization from the
compost. Dermal exposure occurs during maintenance activities.
4.3.5 Summary of Worker Exposure, Individual Cancer Risks, and
Population Cancer Risks from Processing and Commercial Use of
Sludge
2,3,7,8-TCDD and 2,3,7,8-TCDF are present in primary and secondary
sludges formed during wastewater treatment operations at pulp and paper
mills, thereby resulting in a potential for worker exposure to these
chemicals in the processing and commercial use of pulp and paper mill
sludge. Two studies have quantified their presence in sludge: the
5-Mill and 104-Mill studies. Results from the 5-Mill Study showed that
concentrations in primary, secondary, and combined dewatered sludge
ranged from 17 to 710 ppt for 2,3,7,8-TCDD and 32, to 10,900 ppt for
2,3,7,8-TCDF, with the largest concentrations reported in secondary
sludge samples. Results from the 104-Mill Study showed that concentra-
tions corresponding to the low and high TEQ in combined dewatered sludges
ranged from 0.7 to 1390 ppt for 2,3,7,8-TCDD and 3 to 17,100 ppt for
2,3,7,8-TCDF. It also showed that concentration in nondewatered sludges
ranged from 6 to 4500 ppq for 2,3,7,8-TCDD and 6 to 14,000 ppq for
2,3,7,8-TCDF.
Workers involved in pulp and paper mill sludge processing and
commercial use of the sludge be exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF
via three major routes: (1) dermal contact with wet or dry sludge and
compost during maintenance of equipment and performance of job functions;
(2) inhalation of particulate matter generated by sludge and compost
handling during sludge processing, landfill ing, land application, and
composting operations involving unloading, loading, spreading, burying,
Or screening of the materials; and (3) inhalation of 2,3,7,8-TCDD and
2,3,7,8-TCDF volatilized from the sludge and/or compost.
Estimated risks for workers exposed during the processing and
commercial use of pulp and paper mill sludge are summarized in Table 4-4.
44 Analysis of Uncertainties
4.4.1 Worker Exposure from Manufacture, Processing, and Commercial Use
of Pulp, Paper, and Paper Products
Considerable research has been conducted by EPA and API to understand
+he formation of and subsequently reduce the generation of dioxins in the
oulp anc' PaPer industry. Additional information, however, is needed in
the following areas: (1) potential for worker exposure in paper
converting operations; (2) the frequency and duration of potential dermal
4-35
-------
9068H
Table 4-4. Summary of Outer Bounds of Individual and Population Cancer Risks for Workers
Involved in Processing and Commercial Usage of Pulp and Paper Mill Sludge
No. of
Job category workers
Sludge handling/processing
- waste treatment plant operators 1300
- Sludge haulers/front-end loader
operators 400
Landfilling operations
- Equipment operators 400
Land application operations
- Equipment operators 20
3
Est i ma ted ri sk ._•• -
Population risk
Individual risk * cases/vr^,
Exposure type0
Inhalation-volatilization
Inhalation-particulate matter
Dermal6
Inhalation-volatilization
Inhalation-particulate matter
Dermal
Inhalat ion-volati 1 1zat1on9
Inhalation-particulate matter9
Derma ld
Inhalation-volatilization9
Inhalation-particulate matter9
Dermal
Low
2xlO"12
(0.6)
2xlO"U
(19)
IxlO"7
(19)
2xlO"U
(°-6),n
1X10'10
(19)
6xlO"9
(19)
SxlO"10
(0.6)
IxlO"8
(19)
6xlO"9
(19)
3xlO"8
(0.6)
IxlO"6
(19)
IxlO"7
(19)
High Low High
9xlO"9 2xlO"U 9xlO"8
(0.2)
7xlO"8 2xlO"l° 7x10
(8)
4xlO"4 IxlO"6 4x10"
(8)
2xlO"7 8xlO"l° 5xlO"6
(02), o 5
SxlO'7 4xlO"9 ZxlO
(8)
3xlO"5 2xlO"7 8x10"
(8)
2xlO~6 SxlO*9 2xlO"5
(0.2)
2xlO"5 IxlO"7 2x10
(6) t . -4
IxlO"5 6xlO'8 1x10
(6)
9xlO"7 IxlO"8 4x10"
(1) .5
6xlO"5 SxlO"7 3x10
(35) .6
7xlO"6 SxlO"8 4x10
(35)
4-36
-------
3068H
Table 4-4. (Continued)
Estimated risk
No. of
Job category workers Exposure pathway0
Composting operations
•Equipment operators 150 Inhalation-volatilization9
Inhalat ion-part icu late matter9
Dermald
•Compost haulers 50 Inhalation-volatilization^
Inhalation-particulate matter
Derma ld
•Screen operators 20 Inhalation-volatilization9
Inhalation-particulate matter9
Derma lf
Individual risk
Low
IxlO"7
(0.2)
3xlO"7
(8)
4xlO"9
(8)
3X10'10
(0.2)
3xlO"8
(8)
4xlO"9
(8)
IxlO"9
(0.20)
2xlO"6
(8)
2xlO~7
(8)
High
BxlO"6
(0.2)
IxlO"5
(6)
2xlO~7
(6)
IxlO"8'
(0.2)
IxlO"6
(6)
2xlO"7
(6)
7xlO"8
(0.2)
7xlO"5
(6)
8xlO"6
(6)
Population risk
# cases/vr
Low High
5xlO"7 2xlO"5
IxlO"6 4xlO"5
2xlO"8 9xlO"7
4xlO"10 2xlO"8
4xlO~8 2xlO"6
SxlO"9 3xlO"7
7xlO"10 3xlO"8
9xlO"7 4xlO"5
9xlO"8 4xlO"6
Values in parentheses are percent risk due to 2,3.7,8-TCDD; risk estimates are based on EPA's slope factor.
tallies represent cases per year.
clhe frequency of exposure assumed to be 250 days per year.
wation of exposure assumed to be 1 hour per day.
'duration of exposure assumed to be 2 hours per day.
Duration of exposure assumed to be 4 hours per day.
'duration of exposure assumed to be 8 hours per day.
4-37
-------
and inhalation worker exposure to dioxins in the pulp and paper industry;
(3) the extent of the use of personal protective equipment and
engineering controls; and (4) the number of workers in job categories
potentially exposed to dioxins. Additional information is also needed on
the potential for exposure to pulp and paper workers during infrequent
activities such as bi-yearly cleaning of grinding pit and paper roll
residuals which may contain high dioxin concentrations.
There are some ongoing as well as planned studies which may clarify
some of the uncertainties found in this report. A NIOSH study to
characterize worker exposure to dioxins at a pulp/paper manufacturing
plant is currently underway. The 104 Mill Study has been completed;
however, the data that were collected need to be analyzed with respect to
plant operating parameters such as production rate, type of wood used
(e.g., softwood, hardwood), and quantity of bleaching chemical used. The
25 Bleach Line Study conducted by NCASI will provide dioxin concentration
data for 25 bleach lines in 31 mills. Additional information will also
be available soon on workforce characterization, converting operations,
and production data from the American Paper Institute. That data should
assist in filling some of the data gaps. Site visits to pulp
manufacturing, papermaking, and paper converting operations would also
provide additional insight on the potential for exposure to workers when
performing different activities.
4.4.2 Worker Exposure from Sludge Processing and Commercial Use
No previous studies have been conducted regarding workforce
characterization and worker exposure to PCDDs and PCDFs for pulp and
paper mill sludge processing and commercial use. Information is needed
in the following areas: (1) worker job category descriptions; (2) the
number of workers in different job categories; (3) the potential for
worker exposure in the various sludge processing and disposal operations;
(4) the frequency and duration of potential dermal and inhalation
exposure to PCDDs and PCDFs from pulp and paper mill sludge processing
and use; and (5) the extent of use of personal protective equipment and
engineering controls. This information is also needed for operations
which may not be affiliated with pulp and paper mills such as composting
operations and processing of dried sludge for fertilizer manufacture.
Most of the estimates regarding types and numbers of workers and duration
of exposure used in this report were based upon parallels with municipal
sludge-handling operations and engineering judgment.
There are some ongoing studies which may clarify some of the
uncertainties and fill some data gaps. The 104-Mill Study has been
completed; however, the data that was collected needs to be analyzed with
respect to plant operating parameters such as sludge and pulp production
rates, type of wood used (e.g., softwood, hardwood), pulping technology
used (e.g., Kraft, sulfite), and quantity of bleaching chemical used.
The 25 Bleach Line Study by NCASI will provide TCDD and TCDF
4-38
1581q
-------
concentration data for sludges in addition to effluent, bleach plant
filtrates, and intermediate and final pulps. NCASI may be able to
provide additional information, if any is available, regarding workforce
characterization and worker exposure from sludge processing and
commercial use which should assist in filling some data gaps. Site
visits to pulp and paper manufacturing operations as well as to
affiliated or nonaffiliated disposal or commercial sludge processing
sites would also provide additional insight on the potential for exposure
to workers when performing different activities.
4.5 References
Abt Associates, Inc. 1989. Multimedia exposure assessment for re-use
and disposal of sludge from pulp and paper industry and disposal of paper
products (final draft). Prepared for the Office of Pesticides and Toxic
Substances of the U.S. Environmental Protection Agency. September 28,
1989.
Babich M, Adams M, Cinalli C, Galloway D, Hoang K, Huang S, Rogers P.
1989. Common assumptions for the assessment of human dermal exposure to
polychlorinated dibenzo-p-dioxins and dibenzofurans. Interagency
Dioxin-in-Paper Workgroup, Dermal Unavailability Workgroup.
December 12, 1989.
Beck H, Eckart K, Mathar W, Wittkowski R. 1988. Occurrence of PCDD and
PCDF in different kinds of paper. Chemosphere, 17: 51-57.
Bond G. 1989. Personal communication between Gary Bond, NCASI and PEI
Associates, Inc. July 1989.
CPSC. 1989. Personal communications between Mike Babich, Consumer
Products Safety Commission and PEI Associates, Inc. on equations for
estimating dermal exposures from wet and dry pulp, paper, and sludges.
October 1989.
Clement Associates, Inc. 1981. Mathematical models for estimating
workplace concentration levels: A literature review. Prepared for the
Economics and Technology Division of the U.S. Environmental Protection
Agency.
Clement Associates, Inc. 198Z. Methods for estimating workplace
exposure to PMN substances. Prepared for the Economics and Technology
Oivision of the U.S. Environmental Protection Agency.
Cunningham N. 1990. Personal communication with Neil Cunningham, James
River Corporation. January 1990.
Orivas PS, Simmonds PG, Shair FH. 1981. Experimental characteristics of
ventilation systems in buildings. Current Research 6:609-614.
4-39
-------
Eitzer BD, Hites RA. 1988. Vapor pressures of chlorinated dioxins and
dibenzofurans. Environ. Sci. Techno!. 22:1362-1364.
Fisher R. 1989. Personal communications between Robert Fisher, NCASI,
and PEI Associates.
Hammer MJ. 1975. Water and waste-water technology. New York: John
Wiley & Sons.
Hanmer RW. 1988. Environmental protection in the United States pulp,
paper, and paperboard industry: An overview of regulation of wastewater
under the U.S. Clean Water Act. Water Science Technology 20(1): 1-7.
Hawks R. 1989. Personal communication between Ron Hawks and PEI
Associates, Inc. 1989.
Hawley. 1987. Hawley's condensed chemical dictionary, llth ed. New
York: Van Nostrand Reinhold Company.
Hornung RW, Reed LD. 1987. Estimation of average concentrations in the
presence of nondetectable values. Cincinnati, OH: National Institute for
Occupational Safety and Health.
Hwang J., Falco R. 1986. Estimation of multimedia exposures related to
hazardous waste facilities. In: Cohen Y., ed., Pollutants in a
multimedia environment. New York: Plenum Publishing Co.
Kimbrough R, et al. 1984. Health implications of 2,3,7,8-tetra-
chlorodibenzo-p-dioxin (TCDD) contamination of residential soil. Journal
of Toxicology and Environmental Health, 14:47-93.
Kirk-Othmer. 1981a. Encyclopedia of Chemical Technology. 3rd. ed.,
Vol. 16. New York: John Wiley & Sons.
Kirk-Othmer. 1981b. Encyclopedia of Chemical Technology. Srded.,
Vol. 19. New York: John Wiley & Sons.
lammers 0. 1989. Personal communication with Don Lammers, Curt G. OOA,
Inc. December 1989.
ledbetter RH. 1976. Design considerations for pulp and paper-mill
sludge landfills. EPA-600/3-76-111.
McCubbin N. 1989. Personal communication with Neil McCubbin,
Independent Consultant. July and September 1989.
Metcalf & Eddy, Inc. 1979. Wastewater engineering: treatment,
disposal, reuse. New York: McGraw-Hill Book Company.
4-40
I581q
-------
NCASI. 1987. National Council of the Paper Industry for Air and Stream
Improvement, Inc. Assessment of potential health risks from dermal
exposure to dioxin in paper products. Technical Bulletin No. 534. New
York.
NCASI. 1988a. National Council of the Paper Industry for Air and Stream
Improvement, Inc. Risk associated with dioxin exposure through
inhalation of paper dust in the work place. Technical Bulletin No. 537.
New York.
NCASI. 1988b. National Council of the Paper Industry for Air and Stream
Improvement, Inc. Results of measurements of airborne particle size
distributions in paper converting areas. Technical Bulletin No. 554.
New York.
NCASI. 1988c. National Council of the Paper Industry for Air and Stream
Improvement, Inc. Assessment of potential health risks to pulp mill
workers from dermal exposure to dioxin in bleached pulp, paper, and
pulp-based products. Technical Bulletin No. 549. New York.
NIOSH. 1983. National Institute for Occupational Safety and Health.
Control technology assessment in the pulp and paper industry.
Anderson-Nichols and Co., Inc. Prepared for the U.S. Department of
Health and Human Services under Contract No. 210-79-0008. April 1983.
NIOSH. 1985. National Institute for Occupational Safety and Health.
NIOSH pocket guide to chemical hazards. U.S. Department of Health and
Human Services. NIOSH Publication No. 85-114.
Nonwovens Industry. 1989. Rodman Publications, Inc. 20(5). May 1989.
OTA 1989- u>St Con9>"ess, Office of Technology Assessment.
Tprhnologies for reducing dioxin in the manufacture of bleached wood
pulp- OTA-BP-0-54. May 1989.
Olson LJ, et al. 1988. Landspreading dioxin-contaminated papermill
sludge: A complex problem. Archiv. Environ. Health 43(2): 186-189.
podoll RT, Jaber HM, Mill T. 1986. Tetrachlorodibenzodioxin: Rates of
volatilization and photolysis in the environment. Environ. Sci. Technol
20: 490-492.
popendorf WJ, Leffingwell JT. 1982. Regulating of pesticide residues
for farmworker protection. Residue Review. 82:156-157.
Schroy JM, Hileman FD, Cheng SC. 1986. Physical/chemical properties of
2 3 7,8-tetrachlorodibenzo-p-dioxin. In: ASTM Spec. Publ. 891 (Aquat.
Toxicol. Haz. Assess. 8th Symp.): 409-421.
4-41
-------
Soklow R. 1984. Paper production and processing occupational exposure
and environmental release study. EPA-600/2-84-120.
Stockton MB, Stelling JH. 1987. Criteria pollutant emission factors for
the 1985 NAPAP emissions inventory. EPA-600/7-87-015.
Sullivan C. 1989. Personal communication with Claire Sullivan, United
Paperworkers International Union and PEI Associates, Inc. July 1989.
Thibodeaux LJ. 1979. Chemodynamics. New York: John Wiley and Sons.
UPIU. 1989. United Paperworkers International Union. AFL-CIO/CLC. The
dioxin data: What does it mean? The Paperworker, 17(7): 18-21.
USDOC. 1987. U.S. Department of Commerce, Bureau of the Census.
Statistical Abstract of the United States, 1988. 108th Edition.
Washington, D.C. December 1987.
USDOC. 1988. U.S. Department of Commerce, Bureau of the Census.
Current Business Reports. Pulp, Paper, and Board 1987. Washington,
D.C. November 1988.
USEPA. 1979. U.S. Environmental Protection Agency. Process design
manual for sludge treatment and disposal. EPA-625/1-79-011.
USEPA. 1984. U.S. Environmental Protection Agency. Health assessment
document for polychlorinated dibenzo-p-dioxins. Draft document.
Washington, D.C.: Office of Health and Environmental Assessment.
USEPA. 1985. U.S. Environmental Protection Agency. Compilation of air
pollutant emission factors. Fourth edition. AP-42.
USEPA. 1986. U.S. Environmental Protection Agency. Development of
advisory levels for polychlorinated biphenyls (PCBs) cleanup.
Washington, D.C.: Office of Health and Environmental Assessments.
EPA 600/6-86-002.
USEPA. 1988a. U.S. Environmental Protection Agency. U.S. EPA/Paper
Industry cooperative dioxin screening study. EPA-440/1-88-025.
USEPA. 1988b. U.S. Environmental Protection Agency. Estimating
exposures to 2,3,7,8-TCDD. Washington, D.C.: Office of Health and
Environmental Assessment. EPA/600/6-88/005A.
USEPA. 1988c. U.S. Environmental Protection Agency. NEDS source
classification codes and emission factor listing-PMlO. Second edition.
4-42
1581q
-------
USEPA. 1988d. U.S. Enviromental Protection Agency. Supplement B to
compilation of air pollution emission factors. Volume I: Stationary
point and area sources. AP-42.
USEPA. 1989a. Personal communication between Jennie Helms, U.S. EPA,
and PEI Associates, Inc. 1989.
USEPA. 1989b. U.S. Environmental Protection Agency. Correspondence
between Christina Cinalli, Environmental Protection Agency Exposure
Assessment Branch, and George Heath, EPA Engineering Technology Division
on calculation procedure for lifetime average daily exposures, percent
exposure due to 2,3,7,8-TCDD and risk.
Versar. 1984. Exposure assessment for retention of chemical liquids on
hands. Prepared for the U.S. EPA under Contract No. 68-01-6271.
Wong K. 1983. Unpublished EPA report.
4-43
-------
5. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM EXPOSURE TO
DIOXINS AND FURANS DURING USE AND DISPOSAL OF PULP AND PAPER
MILL SLUDGE AND DISPOSAL OF PAPER WASTES
5.1 Introduction
Sludge from pulp and paper mills that use chlorine in the bleaching
process contains measureable amounts of PCDDs and PCDFs as evidenced by
data submitted pursuant to the 104-Mill Study (USEPA 1989a) (see
Table 5-1). Other information submitted pursuant to the 104 Mill Study
indicates that approximately 2.5 million metric tons of such sludge is
generated annually. Most is landfilled (44 percent) or placed in surface
impoundments (24 percent); the remainder is incinerated (12 percent),
land-applied (12 percent), or distributed and marketed (8 percent). Each
of these five use or disposal practices presents potential risks of human
health impacts resulting from exposure to dioxins in the sludge.
Exposure to dioxins may also occur during the disposal of paper
wastes. EPA estimates that about 45 million metric tons of pulp and
paper products are disposed annually (USEPA 1988a). When discarded paper
products containing PCDDs/PCDFs are buried in municipal landfills or
burned in municipal incinerators, PCDDs and PCDFs may be released into
the environment, resulting in potential exposure.
This chapter estimates human exposure and risks associated with the
use and disposal of sludge from pulp and paper mills, as well as with the
disposal of paper wastes in municipal landfills. Exposures and risks
associated with the potential release into the environment of PCDDs/PCDFs
contained in sludge during incineration are assessed in Section 7 of this
report. The information presented here was compiled from:
USEPA. 1990. U.S. Environmental Protection Agency. Assessment of
risks from exposure of humans, terrestrial and avian wildlife, and
aquatic life to dioxins and furans from disposal and use of sludge
from bleached kraft and sulfite pulp and paper mills. Washington,
DC: Office of Toxic Substances and Office of Solid Waste. EPA
560/5-90-013.
The following five waste disposal practices are addressed in this chapter:
• . Landfill ing of pulp and paper sludge (Section 5.2.1),
• Landfill ing of paper wastes (Section 5.2.2),
, Surface impoundment of pulp and paper sludge (Section 5.2.3),
5-1
1592H
-------
Table 5-1. Distribution of 2,3,7,8-TCDO and 2,3.7,8-TCDF Sludge
Concentrations for All Plants in 104-Mill Study
Distribution
descriptor
Number
Mean
Std. Dev.
2.3,7,8-TCDD
concentration
(ng/kg or ppt)
79
162.9
464.7
2,3,7,8-TCDF
concentration
(ng/kg or ppt)
79
865.4
2.303
Percent lies:
100% (maximum)
99%
95%
90%
75%
50%
25%
10%
5%
1%
0% (minimum)
3,600
3.800
680
293
119
51
12
3
1.9
0
0
17,100
17,100
2.940
1,760
799
158
34
6
2.4
0
0
Source: USEPA (1990)
9035H
5-2
-------
• Land application of pulp and paper sludge (Section 5.2.4), and
• Distribution and marketing of pulp and paper sludge
(Section 5.2.5).
Exposure pathways evaluated for each sludge/waste use and disposal
practice considered in this assessment are presented in Table 5-2.
Section 5.3 provides a discussion on uncertainties inherent in the
analyses described in Sections 5.2.1 through 5.2.5, and a summary of the
risk estimates developed in this chapter is presented in Section 5.4.
Conclusions to this report are applicable only to pulp and paper mill
sludges. At this time, the Agency has not assessed risks to human health
and the environment from the use and disposal of sludges generated at
publicly and privately owned treatment plants that treat domestic
sewage. This evaluation with potential subsequent regulation of sewage
sludges will be performed in the next two to three years in the second
round of sewage sludge regulations under 40 CFR Part 503.
g.2 Estimates of Exposure and Risks to the General Populations from
Disposal and Use of Sludoe from the Pulp and Paper industry and
Disposal of Paper Products
In the assessment, exposures and risks to a hypothetical maximum
exposed individual (MEI) as well as exposures and risks to an individual
considered to be typical were addressed. The "typical" individual risk
estimates were used in conjunction with estimates of potentially exposed
populations to predict potential population risks.
Two approaches were used for estimating exposures and risks. These
differ primarily in the selection of 2,3,7,8-TCDD/2,3,7,8-TCDF sludge
concentration data used to estimate exposures and risks. One approach
used the mean and highest concentration data reported in the 104-Mill
Study (USEPA 1989a) for those mills employing the particular disposal
oractice being assessed (i.e., when assessing risks from sludge
P-po-filling, the mean and highest concentrations reported for any of the
mills using this disposal method were used). The other approach,
hereinafter referred to as the "generic" approach, used values
representative of concentrations in sludge from all mills, regardless of
rurrent disposal practices. By assuming a constant level of sludge
rontamination regardless of the method of disposal, pathways and disposal
methods that are intrinsically more risky could be pointed out. For
stimating typical exposure, this generic assessment assumed a
93 7,8-TCDD/2,3,7,8-TCDF concentration equal to the mean concentration
poorted over all plants 1n the 104-M111 Study. For the assessment of
Ell exposure, the 90th percentile concentrations were used. The
Keneric" approach further did not divide the slope factor for dloxin by
hich served to account for absorption In the animal studies which
5-3
1592H
-------
Table 5-2. Exposure Pathways Evaluated for Each Pulp and Paper Mill Sludge
Disposal or Use Practice
Landfill ing Landfill ing Surface Land Distribution
of of impoundment application and marketing
sludge paper of sludge of sludge of sludge
Ingest ion exposure
From drinking
contaminated ground water
From drinking surface water
contaminated by runoff
From foods produced with
contaminated soil
From consumption of fish
caught in contaminated
surface water
From direct ingest ion of
contaminated soil
Inhalation exposure
To volatilized contaminants
To particulate from
contaminated soils
Dermal exposure
From contact with
contaminated soil
9035H
5-4
-------
were used to derive the slope factor. For the purpose of this integrated
report, results based on the generic approach used are presented.
Physical/chemical property and fate/transport information on
2,3,7,8-TCDD and 2,3,7,8-TCDF that were used in USEPA (1990) assessment
are presented in Table 5-3..
5.2.1 Exposure and Risks from Disposal of Pulp and Paper Sludge in
Landfills
Landfill ing of sludge from the pulp and paper industry is defined as
the burial of sludge on land, usually accompanied by the regular
application of soil cover. Fifty-nine of the pulp and paper mills in the
104-Mill Study (USEPA 1989a) reported using landfilling to dispose of
their sludge. Of these, at least 15 dispose of their sludge in municipal
landfills. Because specific information at each site was not available,
all landfill sites were assumed to be of similar size, to contain sludge
with the same concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF, to use the
same management practices, and to emit contaminants at similar rates.
Table 5-4 summarizes the characteristics of these hypothetical
landfills and the landfilled sludge. The landfill design is intended to
be representative of industrial landfills that receive pulp and paper
sludge. For this analysis, it was assumed that risks from these
hypothetical industrial landfills can be generalized to municipal
landfills.
This analysis estimated exposures and risks to 2,3,7,8-TCDD and
2,3,7,8-TCDF through four exposure pathways associated with sludge
landfills:
• Contaminants volatilize from the landfill and are transported by
wind to neighboring areas. Humans inhale contaminated air and are
exposed.
• Storm runoff carries contaminant-laden particles of soil from
the surface of the landfill to nearby surface water bodies.
Contaminants are then released from stream or lake sediments Into
surface water, which is withdrawn for drinking water supplies.
Humans ingest the contaminated water and are exposed.
, Storm runoff carries 2,3,7,8-TCDD and 2,3,7,8-TCDF to surface
water bodies, as described above. Fish accumulate the contaminants
from the water or sediment. Humans Ingest fish and are exposed.
• Rain water or sludge moisture carry dissolved contaminants from
the bottom of the landfill to an aquifer underneath a landfill.
Dissolved contaminants are then transported by the aquifer to
5-5
1592H
-------
8897H
Table 5-3. Physical/Chemical Properties and Fate/Transport Assumptions: All Exposure Pathways
en
Input
parameter
TCDD diffusivity in air (cmZ/sec)
TCDD diffusivity in water (on2/ sec)
TCDD solubility in water (ug/g)
TCDD photolysis half-life (min)
TCDO molecular weight
TCDD K^. (cm3/g)
TCDO Henry's Law constant (atm m /ml)
TCDF diffusivity in air (cmZ/sec)
TCDF diffusivity in water (cm2/ sec)
TCDF solubility in water (ug/g)
TCDF photolysis
TCDF molecular weight
TCDFK^
TCOF Henry's Law constant (atm m3/mol)
Estimate for
typical individual
0.05
5.5 x IO"6
0.02
4.320
322
1 x IO7
1.6 x IO"5
0.05
5.6 x IO"6
4.3
4.320
306
3.5 x IO4
8.6 x IO"5
Estimate for
HE I individual
0.05
5.5 x IO"6
0.02
4.320
322
1 x IO7
1.6 x IO"5
0.05
5.6 x IO"6
4.3
4.320
306
3.5 x 10*
8.6 x IO"5
Notes/exp lanat ion/ ( reference )
Eduljee (1987); USEPA (1988b)
Estimated w/Hayduk and Laudie method (Lyman et al. 1982).
TCDO concentrations do not reach the solubility limit
model simulations (see Section 2 of this report).
(see Section 2 of this report).
(see Section 2 of this report).
(see Section 2 of this report).
Estimated w/Uilke and Lee method (Lyman et al. 1982).
in
Estimated w/Hayduk and Laudie method (Lyman et al. 1982).
TCDF concentrations do not reach the solubility limit
model simulations [see Section 2 of this report).
Assured same as TCDD.
(see Section 2 of this report).
(see Section 2 of this report).
in
-------
Table S-4. AssuB«>t1ons and Parameter Values - Land!" t 1 Is-. All Exposure Pathways
Input
parameter
Estimate for
typical individual
Estimate for
ME I individual
Notes/explanat ion/(reference)
in
i
Area of landfill (cm')
Quantity of sludge received (DMT/yr)
Landfill dimensions (m x m)
Facility lifetime (yrs)
Depth of landfill (m)
Site locations
Depth of cover
Fraction organic carbon in cover
Fraction organic carbon in sludge
Fraction organic carbon in landfill soil
Soil porosity
Soil disconnectiveness index
1.2 x 109
19.500
350 x 350
14
6
2.4 x 109
76.700
500 x 500
14
6
Location of town containing each landfill.
0 0
IX IX
25X
25X
20X
0.25
3.7
20X
0.25
3.7
30 acres is thought to be typical. 60 acres is assumed
for high estimate (USEPA 1985. 1989a).
Average and highest values reported in 104 Mill Study
(USEPA 1989a).
Assumes square site for 30 and 60 acre landfills.
(USEPA 1985a).
One landffll reported depth of 20 feet (USEPA 1989a).
Assumed to be representative due to lack of additional
information.
(USEPA 1988b).
(USEPA 1988b).
Median nitrogen content in combined sludges is 0.85X.
Typical organic carbon-to- nitrogen ratios range from
17:1 to 45:1 (HCASI 1984). Thus, fraction of organic
carbon ranges from 14 to 38X (midpoint of 25X).
Appropriate for sludge.
Typical values.
Typical values.
-------
Table 5-4. (continued)
Input
parameter
Soil intrinsic permeability (cm2)
Estimate for
typical individual
1 x 10~B
Estimate for
ME I Individual
1 x 10~8
Notes/exp lana t ion/ ( reference )
Within range of permeabilities repc
>rted for "silty sand"
True density of sludge/soil
Bulk density of soil (kg/*3)
TCDO concentration in sludge (ng/kg)
TCDF concentration in sludge (ng/kg)
2.65
1.400
163
2.65
1.400
293
885
1.750
and "clean sand" (Freeze and Cherry 1979).
Density of quartz.
Typical value far sand (Teh 1981).
Average and 90th percentile reported values for all mills
(USEPA 1989a).
Average and 90th percentile reported values for all mills
(USEPA 19B9a).
cn
i
00
-------
nearby drinking water wells. Humans ingest contaminated water
withdrawn from the wells and are exposed.
Each of these potential pathways of exposure is discussed below.
(1) Estimates of exposures and risks from Inhalation of vapors.
Estimating human exposure and risk through the volatilization pathway
involved the three steps described below. Information about models used
to estimate volatile emissions and a few key assumptions used are
presented in Table 5-5.
(a) Estimating volatile emissions. Criteria for design and
operating characteristics for industrial landfills are not currently
available. Consequently, the extent to which daily and final cover are
used at landfills is not known. For this analysis, it was assumed that a
typical landfill applies no cover to its sludge, the Hwang and Falco
(1986) equations were used to estimate the emission of dioxins and furans
from the soil for both the MEI and typical exposure scenarios.
(b) Estimating wind transport of volatile emissions. This
analysis used the Industrial Source Complex, Long Term (ISCLT) model
(a Gaussian plume dispersion model) to estimate air concentrations of
volatilized 2,3,7,8-TCDD and 2,3,7,8-TCDF to which populations downwind
may be exposed. The ISCLT model estimated ambient air concentrations
extending 50 km in all directions from the landfill sites. Site-specific
meteorological data were used.
(c) Methods for estimating human exposure and risk. Individual
exposure and cancer risk were calculated by:
E , NNNH w
D (BW)(LE)
where:
BW - average body weight (assumed to be 70 kg)
CD - estimated air concentration at distance D (pg/m3)
En - average lifetime individual dose for person residing at
u distance D (pg/kg/day)
p. » fraction of contaminant absorbed from Inhaled air by humans
(unitless) (assumed to be 1)
Iu * volume of air Inhaled daily (assumed to be 23 nr/day)
Lp - life expectancy 1n days/lifetime (assumed to be 70 years)
Ic « number of days of exposure in a lifetime (assumed to be
70 years x 365 days/year)
and:
5-9
159ZH
-------
3B97H
Table 5-5. Assumptions and Parameter Values - Landfills: Volatilization Pathway
en
Input
parameter
Estimate for
typical individual
Estimate for
MEI individual
Notes/exp lanat ion/ ( reference)
Model used to estimate emissions
Model used to estimate wind transport.
air concentrations, and human exposure
Hetereological data for ISCLT
Eiiission period (yrs.)
' Landfill status at beginning of
o emission period
USEPA (1988b) USEPA (19886)
Industrial ISCLT
Source
Complex. Long
Term (ISCLT)
model (area
source)
Stability array (STAR) sumary data accessed through GENS (site specific)
70 70
Hwang and Falco (1986) and USEPA (1986) in USEPA (I986b)
Location - specific modeling for each site
(Bowers et a1.1980].
Full
Full
Corresponds to length of exposure for cancer risk
calculations.
Will tend to overestimate emissions.
-------
IRD = (ED)^I*) (5~2)
where:
IRD = Lifetime individual cancer risk for person residing at
distance D
qj* = Cancer potency for 2,3,7,8-TCDD or 2,3,7,8-TCDF (pg/kg/day)'1
ISCLT also provided estimates of population-weighted average concentra-
tions of the 2,3,7,8-TCDD or 2,3,7,8-TCDF in ambient air surrounding all
landfill facilities. Aggregate cancer risks were calculated with the
following expression:
EAVE =
AVt (BW)(LE)
where:
* average air concentration of contaminant, computed by
weighting each level of contaminant concentration by the number
of persons exposed to that level (pg/nr)
= population-weighted average exposure for all persons living
within 50 km of a pulp and paper sludge landfill (pg/kg/day)
and:
RT - [(Eave)(qi*){POP)l/LE (5-4)
where:
POP - Total exposed population
Rj - Aggregate cancer risk for exposed population (Incremental
cancer cases/year)
(d) Estimates of populations exposed. The ambient air concen-
trations estimated by ISCLT were overlayed onto actual human populations
within 50 km of the sites examined. Population data were drawn from the
1980 Census, mapped to the level of Census block group and enumeration
district.
(2) Estimates of exposures and risks from inoestlon of around water.
for lack of sufficient empirical data, this analysis relied on mathemati-
cal models to estimate the extent of possible human exposure and risk.
Table 5-6 lists key assumptions and input parameters used to estimate
potential ground-water contamination. The estimation of exposure and
risk Involved three steps as described below.
5-11
1592H
-------
8897H
Table 5-6. Assumptions and Parameter Values - Landfills: Ground-Water Pathway
Input
parameter
Estimate for
typical individual
Estimate for
HE I individual
Hotes/explanat ion/(reference)
Model used to estimate loading to
aquifer
SESOIL
SESOIL
(Bonanzountas and Wagner 1984).
in
i—i
ro
Result used for exposure estimates
Model used for transport through
saturated zone
Result used for exposure estimates
Distance to aquifer (•)
Aquifer medium
Depth of aquifer (m)
Width of aquifer (m)
Distance to model wells (m)
Distances represented
Steady state
AT123D
Steady state
3
Sand and gravel
100
infinite
200
1.200
3.000
0-400
400-2.000
2.000-4.000
Steady state
AT123D
Steady state
3
Sand and gravel
100
infinite
200
200
(Yen 1981).
Assumes 6 meter depth to water table. For best estimate
scenario, sludge layer extends to 3m below landfill
surface, but contaminants and organic carbon are
distributed throughout 6m landfill depth (DRASTIC data
base).
Most cannon aquifer medium reported in DRASTIC data base
for counties containing landfills.
Midpoints of ranges listed below.
-------
Table 5-6. (continued)
Input
parameter
Estimate for
typical individual
Estimate for
NCI individual
(totes/explanation/(reference)
en
i
Exposed populations
0-400 m
400-2000 m
2000-4000 m
Decay
Effective porosity of aquifer medium
Hydraulic conductivity (m/hr)
Hydraulic gradient
Assumes density of 68 persons per square mile.
co Longitudinal dispersivity (m)
Transverse dispersivity (m)
Vertical dispersivity (m)
TCDD KQ in aquifer
TCOF KQ in aquifer
Annual recharge rate (cm)
195
4.675
14.600
0
0.25
10
IX
20
1
10,000
35
43
0
0.25
10
IX
20
1
10.000
35
43
No decay rate available for TCDD or TCOF in grounoVater.
Appropriate for sand and gravel (Freeze and Cherry 1979).
Appropriate for productive sand and gravel aquifer
(Freeze and Cherry 1979).
Typical value.
Within range of "typical" values reported for sand
or silt. Values equal approximately 1/10 distance
to receptor well (Teh 1981).
Within range of "typical" values reported for sand
or silt. 1/20 and 1/10 of longitudinal
dispersivity (Yen 1981).
Within range of "typical" values reported for sand.
Does not affect steady state results.
Does not affect steady state results.
Based on meteorological data for a landfill site in
Wisconsin.
-------
(a) Estimating contaminant loadings to ground water. The
SESOIL model was used to simulate contaminant transport through soil
layers. The model considered monthly climate data, and maintained a mass
balance for contaminant transport through multiple soil layers.
Estimates of potential human exposure and risks were derived once steady
state conditions had been reached in the unsaturated zone. The
simulation assumed that the upper three meters of the landfill contain
pure sludge, and the lower three meters contain soil into which
2,3,7,8-TCDD, 2,3,7,8-TCDF, and organic carbon from the sludge have
migrated. At steady state, the top three meters were assumed to contain
20 percent organic carbon; the lower three meters contain 10 percent.
(b) Use of AT123D to predict contaminant transport through the
aquifer. From the estimated loadings to ground water, the AT123D model
(Analytical Transport One-, Two-, and Three-Dimensional Model) was used
to predict contaminant concentrations at wells downgradient of the site.
Exposure estimates in this analysis were based on "steady state" results
from AT123D.
(c) Estimates of exposures and risks. Individual exposure and
cancer risk were calculated by:
ED - (W)(FA){C0)(LF)/{BW}(LE) (5-5)
where:
W - amount of water consumed daily (assumed to be 2 liters)
FA • fraction of contaminant absorbed from ingested water (unitless)
(assumed to be 1)
Eg - average lifetime individual dose (pg/kg/day)
Cn - estimated water concentration (pg/1)
BW » average body weight (assumed to be 70 kg)
LE = life expectancy in days/lifetime (assumed to be 70 years)
Lp - number of days of exposure in a lifetime (assumed to be
70 years/lifetime)
and:
where:
IRQ - lifetime individual cancer risk .
qj* - cancer potency for 2,3,7,8-TCDD or 2,3,7,8-TCDF (pg/kg/day)"1
Exposure and risk were estimated separately for persons taking drinking
water at each of the three model distances from a surface landfill site.
5-14
1592H
-------
Maximum exposure and risk were assumed to occur at the nearest well
location.
(d) Estimates of populations exposed. It was assumed that the
density of persons drinking groundwater is constant near each landfill,
and that this density corresponds to the average for the United States as
a whole, or approximately 68 persons per square mile (0.26 per hectare)
(USDOC 1987). This value is used to estimate population sizes in
Table 5-6.
(3) Estimates of exposure and risks from ingestion of drinking water
from surface water sources. Where pulp and paper sludge is deposited in
uncovered landfills, particles of sludge or soil from the landfill
surface can be transported by erosion to nearby lakes or streams. If
humans consume water from these lakes or streams, they may be exposed to
2,3,7,8-TCDD and 2,3,7,8-TCDF from the landfilled sludge. Methods used
to estimate the extent of this potential exposure and its associated
risks to human health are discussed in this section. Table 5-7 presents
the assumptions and input parameters used.
(a) Estimating 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations in
surface water. This analysis used the Universal Soil Loss Equation,
together with estimates of sediment delivery ratios, to estimate the
fraction of a lake or stream's sediment that originates from the
landfill- By multiplying this fraction by the original concentration of
2,3,7,8-TCDD and 2,3,7,8-TCDF in sludge or soil particles on the landfill
surface, the methodology derived estimates of the concentration of
contaminants in the sediment. This contaminant load was then partitioned
Between adsorbed and dissolved phases, based on the assumption of
equilibrium partitioning between the two phases.
(b) Estimates of exposures and risks. Individual exposure and
cancer risk were calculated using Equations 5-5 and 5-6, respectively.
(c) Estimates of populations exposed. It was assumed that
the density of persons potentially exposed corresponds to the average for
the United States as a whole, or approximately 68 persons per square mile
(0.26 per hectare) (USDOC 1987).
(4) Estimates of exposure and risks from incestion of fish from
j-nrfage water sources. Where pulp and paper sludge is deposited in
Uncovered landfills, particles of sludge or soil from the landfill
surface can be transported by erosion to nearby lakes or streams. Fish
living i" the lakes or streams can take up sludge contaminants into their
tissues; if humans then consume those fish, they can be exposed.
5-15
1592H
-------
B897H
Table 5-7. Assumptions and Parameter Values - Landfills: Surface Water Pathways
Input
parameter
Cover
Vegetal Ion
Distance fro» site to surface water (Meters)
"P" ratio
"C" ratio
Drainage area (hectares)
tn
i
JjJ Population density (people/square Mile)
Percent of population served surface Hater
Fish to sediment ratio - dioxin
Fish to sedinent ratio - Furan
Percent organic carbon in soil
Concentration in filet vs.
Estimate for
typical individual
(tone
90S
152
1:1
1:1
1.295.000
68
49
1
1
0.01
0.5
Estimate for
HEI Individual
None
None
30
1:1
10:1
4.047
NA
HA
10
10
0.001
0.5
Notes/exp lanat i on/ { reference )
USEPA (1988b).
USEPA (1988b).
Assures no support practices.
All scenarios assume that the surrounding drainage area
is pasture land (USDA 1978).
Typical scenario assures the sites drain into a major
Hatemay. The HEI scenario assures the sites drain into
a relatively small strean.
USDOC (1987).
USeS (1985).
USEPA (1988b).
USEPA (1988b).
USEPA (I988b).
See Section 2 of this report.
concentration in Hhote fish
Percent of water and fish ingested that
is contanrinated
100
100
Assures that the affected population that is served by
surface water consunes 100X contaminated water.
Fish consult ion (g/day)
6.5
140
Average U.S. consumption; subsistence fisher.
-------
methodology used here is quite similar to that discussed in (3), in that
both methodologies begin by estimating sediment concentrations of
2,3,7,8-TCDD and 2,3,7,8-TCDF in water bodies as a result of runoff from
landfills. Once sediment concentrations have been estimated, however,
the methodology departs from that described in (3), and uses fish to
sediment bioconcentration factors and estimates of human fish consumption
contaminant doses to humans.
Table 5-6 presents assumptions and input parameters used to calculate
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in surface water and fish
tissue.
(a) Estimating 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations 1n
sediments and fish tissue. This analysis used the Universal Soil Loss
Equation, together with estimates of sediment delivery ratios, to estimate
the fraction of a lake or stream's sediment that originates from the
landfill. By multiplying this fraction by the original concentration of
2,3,7,8-TCDD and 2,3,7,8-TCDF in sludge or soil particles on the surface
of the landfill, estimates of the concentration of contaminants in the
sediment were derived. Empirical sediment to fish bioconcentration
factors were used to estimated concentrations of the contaminant in
fish. The concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in the muscle
tissues of fish (consumed by humans) were assumed to be fifty percent
lower than the whole body concentrations of these contaminants.
(b) Estimates of exposures and risks. Individual exposure and
cancer risk were calculated as follows:
Dosep - *• —'*•
F (BW)(LE)
where:
Dosep
BW
LF
Average lifetime dose from consumption of fish (pg/kg/day)
Concentration of contaminant in fish tissue (pg/g)
Individual's daily fish consumption (g/day)
Bioavailability of 2,3,7,8-TCDD or 2,3,7,8-TCDF from fish
(unitless); assumed to be 1.0.
Human body weight (assumed to be 70 kg)
number of days of exposure in a lifetime (assumed to be
70 years x 365 days/year)
life expectancy in days in a lifetime (assumed to be 70 years)
and:
IRF - (Dosep)(qi*) (5-8)
5-17
1592H
-------
where:
IRC = Lifetime individual cancer risk from ingestion of fish
qi* = Cancer potency factor for 2,3,7,8-TCDD or 2,3,7,8-TCDF
(pg/kg/day)'1
The MEI analysis assumes a subsistence fisher consuming 140 grams/
day. For the typical individual exposure analysis, fish consumption was
assumed to be 6.5 grams per day.
(c) Estimates of populations exposed. As a reasonable worst
case, the size of the total general population potentially exposed to
contaminated fish was assumed to be the same size as estimated in
Section 5.2.1.3 above.
(5) Summary of exposures and risks. Table 5-8 presents estimates
of human risk for the potential pathways of exposure considered for
landfill ing of sludge. As can be seen from the tables, exposure and
risks to the "most exposed individual" from the landfill ing of sludge are
highest from pathways associated with surface runoff. Estimated risks
through these pathways for the MEI are based on an extreme scenario in
which runoff from the site reaches a stream of relatively small drainage
area, and the MEI is assumed to take drinking water or fish from the most
contaminated segment of the stream. Average risks through surface water
pathways are estimated based on the assumption of larger drainage areas,
and are considerably lower.
5.2.2 Exposure and Risks from Disposal of Paper Products in Municipal
Landfills
As shown in Table 5-9, the U.S. generates about 130 million metric
tons of municipal waste per year, of which about 45 million tons, or
36 percent, is pulp and paper. After they are used, these products are
recycled, incinerated, landfilled, or otherwise disposed. About Table 5-8
30 percent of the paper products generated can be expected to contain
2,3,7,8-TCDD and 2,3,7,8-TCDF. To the extent that humans come into
contact with contaminants released from these wastes, risks to human
health may result. This subsection considers potential health risks from
the disposal of paper products in municipal landfills.
Two pathways of potential human exposure to 2,3,7,8-TCDD and
2,3,7,8-TCDF from paper products in municipal landfills are considered:
• Contaminants from the paper are released into leachate within
the landfill and seep into an aquifer beneath the facility.
Nearby residents ingest drinking water from the aquifer and are
potentially exposed.
5-18
1592H
-------
Table 5-8. Estimates of Health Risks to the General Population from Landfill Disposal of
Pulp and Paper Sludges Contaminated with 2,3,7,8-TCOD and 2,3,7,8-TCDF
Exposure pathway
.
latin*1 exposure from volatilization from
dfllls In which sludge Is disposed
/r,*ent TCOD)C
(percent
ME I
risk3
(per
lifetime)
5 x 10"7
(4.0)
Typical Exposed
risk3 population
(per
lifetime)
1 x 10~9 12,800,000
(4.4)
Total
risk
(cases/
year)b
2 x 10"4
(4.4)
ingest
ton exposure from drinking ground water
aW1nated by leachate from landfills in
sludge is disposed
(percent TCOD)C
tlon exposure from drinking surface water
mtnated by surface runoff from landfills
sludge is disposed
•cent TCOO)C
(Per
(Per
cent
ton exposure from fish caught in surface
contaminated by runoff from landfills in
sludge is
TCOO)C
1 x 10'9 1 x 10"10 19,000
(2.0)
(63)
(8)
8 x 10"4 5 x 10"8 6,980.000
(0.6) (0.6)
5 x 10"2 8 x 10"
(65)
14,200,000
3 x 10
-8
(8)
5 x 10
-3
(0.6)
2 x 10
(65)
-2
-4
lated as: (Estimated Dose) x 1.6 x 10 per (pg/kg)/day.
la ted as: (Typical Risk x Exposed Population) / (Life Expectancy)
(Exposure to TCDD)
s: X (Exposure to TCDD) + (1/10) (Exposure to TCDF)
5-19
-------
Table 5-9. Site and Waste Characteristics for Municipal Landfills
Receiving Waste Paper Contaminated With TCDD and TCDF
Site/Waste Characteristic Value
Total municipal solid waste (MSW) (kkg/yr)e i.3x!08
Total pulp and paper waste (kkg/yr)e 4.5x10
Total bleached kraft paper, paper-board, pulp (kkg/yr) 2.3x10
Bleached kraft paper as fraction of MSW (percent) 10.7
Maximum concentration of 2,3,7,8-TCDD in paper (ppt) 36a
Maximum concentration of 2.3,7,8-TCDD In MSW (ppt) 4b
Maximum concentration of 2.3,7,8-TCDF in paper (ppt) 333a
Maximum concentration of 2.3,7,8-TCDF in MSW (ppt) 36b
Area potentially affected for each landfill (ha) 500C
Density of persons using ground water (persons/ha) 4
Number of landfills 9.284
Maximum size of exposed population (persons) 18,500,000
aHighest pulp concentration reported 104 Mill Study (USEPA).
Includes only contribution from paper.
cCortservatfve assumption.
Maximum size of exposed population » (area affected per landfill)
x (number of landfills) x (population density).
eUSEPA (1988a).
fAOL (1987).
5-20
903SH
-------
• Contaminants from the paper are released Into soil moisture and
then volatilize from the landfill site to ambient air. Nearby
residents inhale the contaminated air and are potentially exposed.
Human exposure through both of these pathways depends, in part, on the
fraction of each landfill's contents consisting of contaminated paper
products. If it is assumed that only bleached paper products contain
2,3,7>8-TCDD or 2,3,7,8-TCDF, then one can estimate the fraction of total
disposed paper products likely to contain these contaminants. No
information was found describing the fraction of all paper wastes
originating from the bleached kraft process. Consequently, it was
assumed that the fraction of paper product waste that originated from the
bleached kraft process is the same as the fraction of bleached paper
product production. To the extent that bleached paper products are
exported, recycled, or otherwise disposed at higher or lower rates than
other paper products, this assumption may over- or under-predict the
fraction of municipal waste consisting of bleached paper products.
About 30 percent of all pulp and paper is produced from the bleach
kraft process. If this same fraction also applies to solid wastes, then
one would expect bleached paper wastes to account for about II percent of
total municipal waste. The further assumption that bleached paper
products', contribution to municipal landfills does not differ from its
contribution to the total municipal waste stream would imply that
bleached kraft accounts for about 11 percent of a typical municipal
landfill's contents.
Consequently, since roughly one-tenth of the total paper products
disposed of in a municipal landfill comprises bleached paper products,the
average concentration over all disposed paper products can'be assumed to
be one-tenth the concentration in bleach paper products. The highest
reported concentrations for 2,3,7,8-TCDD and 2,3,7,8-TCDF in bleached
nulp froni tne 104 Mil1 study are 36 PPt and 333 PPt» respectively. If
these same concentrations are assumed to apply to the pulp and paper
oroducts disposed in landfills, then one would expect the average
concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF to be 3.6 ug and 33.3 ug
per metric ton of paper wastes, respectively.
To assess potential human exposures and risks from disposal of waste
naper in municipal landfills, a series of conservative assumptions was
used to estimate contaminant concentrations in air above a landfill and
4n around water beneath it. Details of these calculations are presented
]n Subsections (1) and (2).
(1) Estimates of exposures and risks from Inhalation of vapors.
Humans can be exposed to potential health risks if 2,3,7,8-TCDD and
5 3 7,8-TCDF volatilize from paper in municipal landfills. An upper
bound estimate of the extent of these risks can be derived by combining
5-21
1592H
-------
consistently conservative assumptions into mathematical models for
estimating the rate of emissions from these landfills and then estimating
the extent to which emitted contaminants are diluted before inhalation.
In Table 5-9, conservative estimates of 2,3,7,8-TCDD and 2,3,7,8-TCDF
concentrations in a municipal landfill that might result from the
disposal of paper products are presented. Based on these concentrations,
this analysis uses a set of equations to predict emissions from a
landfill site.
(a) Methods for estimating volatile emissions from surface
impoundments. A numerical solution of a partial differential equation
was used to determine the reduction in volatile emissions that would be
expected following the addition of a soil layer to the top of a landfill
(USEPA 1988b). For a 10 to 25 centimeter soil cover, emission rates were
estimated to be reduced by 75 to 80 percent, given a contaminated layer
thickness of 8 feet. This analysis assumed that municipal landfills
apply such a cover and that emissions are reduced accordingly.
As shown in Table 5-3, organic carbon partition coefficients for
2,3,7,8-TCDD and 2,3,7,8-TCDF are assumed to be 1 x 10' and 3.5 x
104, respectively. The contents of the landfill are (conservatively)
assumed to be only one percent organic carbon, resulting in Kn
estimates of only 1 x 104 and 3.5 x 101 for 2,3,7,8-TCDD and
2,3,7,8-TCDF, respectively. For 2,3,7,8-TCDD, the KQ estimate is
further reduced to correspond to the lowest reported partition
coefficient between 2,3,7,8-TCDD in paper and in liquid or about 2,000
g/g. After adjustment for a 10 to 25 centimeter cover layer, it was
estimated that emissions of 1 x 10"17 and 7 x 10"15 g/niz/second
will occur for 2,3,7,8-TCDD and 2,3,7,8-TCDF, respectively.
(b) Methods for estimating wind transport of volatile
emissions. As a conservative estimate of ambient air concentrations of
these contaminants near the landfill, an atmospheric box model (USEPA
1988b) was used to estimate concentrations of contaminants in air above
the site. The equation used as the basis of the model is as follows:
r Q (5-9)
La = (LS)(V)(HH)
where:
Q - total emissions from site (g/sec)
Ca - ambient air concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF at
the exposure location (g/nr),
LS - equivalent side length of the site perpendicular to the wind
(m),
MH - mixing height before being inhaled by an individual (m), and
V • average wind speed at the inhalation height (m/s)
5-22
1592H
-------
It has been reported that 95 percent of all municipal landfills are
less than 100 acres (41 hectares) in area (USEPA 1988c). This analysis
considers a landfill of 41 hectares, represented as a square 640 meters
wide. If the average wind velocity at the site is 4.5 meters per second,
then the mixing volume of air within a height of 1.5 meters above the
site is about 1,340 cubic meters per second. Given total emissions of
2,3,7,8-TCDD and 2,3,7,8-TCDF of 1 x 10-17 and 7 x 10'15 grams per
second from the landfill area, the box model predicts ambient air
concentrations of about 1 x 10"3 and 9 x 10"z pg/nr, respectively.
(c) Estimates of exposures and risks. Based on the estimated
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in ambient air,
individual exposures and cancer risks were calculated using Equations 5-1
and 5-2.
Based on assumed inhalation rates of 23 m3 per day of outdoor air
directly above the landfill site, these concentrations would result in
upper bound cancer risks of less than 1 x 10"'. Because the risk to a
most exposed individual for this very conservative "high risk" scenario
is so low, estimate of risk to the typical individual was not derived.
(2) Estimates of exposures and risks from ingestion of ground
Mater. Landfill disposal of paper or paperboard containing 2,3,7,8-TCDD
and 2,3,7,8-TCDF can result in human exposure and risk if these
contaminants migrate from the landfill to groundwater, and are then
transported to nearby drinking water wells.
Table 5-10 presents information about models used to estimate
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in groundwater and a few
key assumptions used to estimate those concentrations.
(a) Method for estimating contaminant loadings to groundwater.
The estimated average concentration in municipal solid waste of
2 3,7,8-TCDD and 2,3,7,8-TCDF, adjusted by 10.7 percent to account for
other waste in the landfill, were 4 and 36 ng/kg, respectively (see
Table 5-91. The partition coefficients for 8 percent ethanol were used
(2 000 cm3/g) (see Appendix A.I). The maximum equilibrium
concentration of the two chemicals in leachate, estimated from their
concentration in municipal solid waste divided by their partition
meffjcient, was estimated to be 2xlO"5ug/l for 2,3,7,8-TCDD and
2xlO"Wl or 2,3,7,8-TCDF.
(b) Method for estimating groundwater concentrations.
Ninety-five percent of all municipal landfills are less than 100 acres
?41 hectares) in area (USEPA 1988c). If leachate from the landfill
Contains 2,3,7,8-TCDD and 2,3,7,8-TCDF at the estimated maximum
roncentrations, a 100 acre landfill site with 43 cm/year of recharge
vJould release a maximum loading of 1.6 x 10'11 kg/hour of TCDD and
5-23
159ZH
-------
8897H
Table 5-10. Assunptions and Parameter Values - Paper Wastes in Municipal Landfills: Ground-Water Pathway
tn
Input
parameter
Node) used for transport through
saturated zone
Results used for exposure estimates
Distance to aquifer (•)
Aquifer medium
Depth of aquifer
width of aquifer (•)
Distance to model wells (m)
Decay
Effective porosity of aquifer mediui
Hydraulic conductivity
Hydraulic gradient
Longitudinal dispersivity (m)
Percent organic carbon in soil
Water consumption (liters/day)
Estimate for
typical individual
NA
KA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.01
2
Estimate for
MEI individual Notes/explanation/freference)
AT123D
Steady state Conservative.
0 Conservative.
Silty sand
10
500
200
0 No decay rate available for TCDO or TCOF in ground water.
0.1 Appropriate for silty sand.
1 "High risk" estimate appropriate for silty sand.
IX Typical value.
10 Within range of "typical" values reported for sand or silt.
0.001
2
-------
Table S-1O, (continued)
tfl
1
INS
in
Input Estimate for Estimate for
parameter typical individual MCI individual Motes/explanation/ (reference)
Human absorption of water 100 100
Transverse UK 1 Within range of "typical" values reported for sand or silt.
Vertical dispersivity (ID) HA 1
TCDD KQ in aquifer HA 10,000 Does not affect steady state results.
TCDF Kg in aquifer HA 35 Does not affect steady state results.
HA - Not applicable.
-------
1.5 x 10~10 kg/hour of TCDF (from paper wastes) to an underlying
aquifer. For a hypothetical landfill of 410 meters width and length (100
acres), and for an aquifer with characteristics described by the "high
risk" scenario in Table 5-9, the AT123D model was used to estimate
estimate well concentrations at 200 meters from the edge of the
facility. Maximum predicted concentrations at 200 meters distance were 1
x 10'4 pg/1 for 2,3,7,8-TCDD and 1 x 10'3 pg/1 for 2,3,7,8-TCDF.
(c) Estimates of exposures and risks. Based on assumed rates of
individual water ingestion per day (2 liters/day) and duration of
exposure (70 years) individual exposures and cancer risks were calculated
using Equations 5-4 and 5-5, respectively.
(3) Summary of results. Table 5-11 summarizes estimates of human
risk for the disposal of waste paper in municipal landfills. Based on
these results, 2,3,7,8-TCDD and 2,3,7,8-TCDF in waste paper products do
not appear to result in significant human risks to human health when
received by municipal landfills. Using conservative assumptions, it was
estimated that maximum cancer risks to human health from the
volatilization of 2,3,7,8-TCDD and 2,3,7,8-TCDF from municipal landfills
would be expected to be lower than 1 x 10"6 for a "most exposed
individual" who lives 24 hours per day on the top of the landfill. Risks
from groundwater contamination appear to be lower still, with cancer
risks to the MEI of less than 1 x 10'9.
5.2.3 Exposures and Risks from Disposal of Pulp and Paper Sludge In
Surface Impoundments
Surface impoundments are defined as facilities in which pulp and
paper mill sludge is stored or disposed on land without a cover layer of
soil. For this analysis, it was assumed that sludge contained in such
facilities was of higher moisture content than the sludge deposited in
landfills, at least in the active phase of the surface impoundment.
Twenty facilities in the 104-Mill Study (USEPA 1989a) reported using
surface impoundments for their sludge.
This analysis estimated general population exposures and risks to
2,3,7,8-TCDD and 2,3,7,8-TCDF through four exposure pathways associated
with surface impoundments:
• Volatilized 2,3,7,8-TCDD and 2,3,7,8-TCDF are emitted from the
impoundment surface. These chemicals are transported downwind to
nearby areas. Humans inhale the contaminated ambient air and are
exposed.
• Contaminants from sludge placed in the impoundment are dissolved
in water seeping through the bottom of the impoundment.
Contaminated water enters an aquifer beneath the impoundment, and
5-26
1592H
-------
Table 5-11. Estimates of Health Risks to the General Population from Landfill
Disposal of Paper Contaminated with 2,3.7.8-TCDD and 2.3,7.8-TCDF
.—
Exposure pathway
•
lation exposure from volatilization from
nJcipal landfills in which paper is disposed:
disposed:
percent TCDD)C
stlon exposure from drinking ground water
laminated by leachate from municipal land-
fills m *h1ch PaPer 1s deposed:
(pcrcent TCOD)C
Typical
ME I risk3
risk3 (per Exposed
(per lifetime) lifetime) population
<6xlO~7 <6xlO"7 NAd
(16) (16)
IxlO"9
-------
flows down-gradient to drinking water wells. Humans withdraw
drinking water from the contaminated aquifer and are exposed.
• Surface runoff carries particles of sludge from the surface of
the impoundment to a nearby lake or stream. 2,3,7,8-TCDD and
2,3,7,8-TCDF adsorbed to these particles enter the surface water
body, and some of it dissolves in the surface water. Humans
withdraw surface water for drinking and are exposed.
• Surface runoff carries particles of sludge from the surface of
the impoundment to a nearby lake or stream, where the particles
are suspended or settle to bottom sediment. Fish absorb and
bioconcentrate 2,3,7,8-TCDD and 2,3,7,8-TCDF from the sludge
particles. Humans ingest the fish, and are exposed.
Each of these potential pathways of exposure to the general population is
described below. Table 5-12 summarizes the characteristics of
hypothetical surface impoundments intended to be representative of those
that receive pulp and paper sludge.
(1) Estimates of exposure and risk from Inhalation of vapors.
Modeling potential exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF vapor from
surface impoundments involved two steps. First, the rate at which each
contaminant is emitted from the surface of an impoundment facility was
estimated. Second, wind transport of emitted contaminants affecting
ambient air concentrations in surrounding areas was estimated.
Information about models used to estimate volatile emissions and a few
key assumptions used to estimate release from volatilization are
presented in Table 5-13.
(a) Methods for estimating volatile emissions from surface
Impoundments. Several authors have proposed methods for estimating
emissions from liquids contained in surface impoundments. Common to
most methodologies is the use of a two-layer resistance model to estimate
volatile emissions from a liquid surface. In general, the methodologies
assume that emissions occur by molecular diffusion through non-turbulent,
viscous sublayers on either side of the air/liquid interface, and that
these two films form the dominant resistances to mass transfer across the
interface. Background contaminant concentrations in ambient air are
assumed negligible. This study used a two-layer resistance model with
estimated mass transfer coefficients (USEPA 1989b).
Estimation of emissions required additional assumptions for
estimating the dissolved concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF.
For example, concentrations of dissolved contaminant may vary with depth
in the lagoon or over time. If the contents of the Impoundment were well
mixed, if sludge was deposited only once, and if no contaminant was lost
to other loss processes, then the concentration of TCDD in the
5-28
1592H
-------
Table S-12. Asst^^itions and Parameter Values - Surface In*x>um*nents: All Exposure Pathways
C7I
vo
Input
parameter
Number of impoundments per site
Area of each impoundment (hectares)
Depth of impoundment (n)
Fraction organic carbon in sludge
Estimate for
typical individual
3
25
3
25X
Estimate for
HEI individual Notes/explanation/ (reference)
3 USEPA (I985a).
50 Typical estimate from USEPA (1985). HEI estimate is typical
estimate + 100X.
3
25X Midpoint of range reported bv HCASI (1984).
(X dry wt)
Sludge concentration TCDO (ppt)
Sludge concentration TCDF (ppt)
Quantity of sludge received (DMT/yr)
163
885
28.653
293
91.250
Mean and 90th percent!1e reported concentrations (USEPA 1989a),
Mean and 90th percent!le reported concentrations (USEPA 1989a).
Mean and highest reported quantities. (USEPA 1989a).
-------
8897H
Table 5-13. Assumptions and Parameter Values - Surface Iimwumhents: Volatilization Pathway
Input
parameter
Estimate for
typical individual
Estimate for
MEI individual
Notes/explanat ion/{reference)
Model used to estimate emissions
Model used to estimate wind transport
and human exposure
Temperature CO
Average windspeed (m/sec)
Two-phase
resistance
ISCLT
25
4.5
Two-phase
resistance
ISCLT
25
4.5
USEPA (1989b)
Bowers et al. (1980)
Single average windspeed used to estimate emissions (USEPA
19886). Site-specific windspeeds and direction used for ISCLT
calculations.
in
u>
o
-------
impoundment would diminish as a result of continued emissions. If sludge
were instead deposited regularly in the impoundment, then the dissolved
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF might remain constant or
increase. Unfortunately, sufficient data were not available for
detailed, site-specific modeling of the time path of dissolved
contaminant concentrations in existing pulp and paper sludge
impoundments. This analysis therefore attempted to approximate
concentration and emission estimates based on idealized scenarios.
It was assumed that the impoundment was well mixed, with regular
additions of sludge containing 2,3,7,8-TCDD and 2,3,7,8-TCDF in known
(dry weight) concentrations. It was assumed that 2,3,7,8-TCDD and
2,3,7,8-TCDF were partitioned at equilibrium between adsorbed and
dissolved phases in the impoundment.
(b) Methods for estimating wind transport of volatile
emissions. Once 2,3,7,8-TCDD or 2,3,7,8-TCDF vapor is emitted from a
surface impoundment, it can be transported downwind to nearby residents,
resulting in potential human exposure and health risks. This analysis
used a Gaussian plume dispersion model to estimate the extent to which
air concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF would be reduced in
the process of wind transport. Calculations were performed by the area
source version of ISCLT (Industrial Source Complex, Long Term) model
maintained in the USEPA/OTS Graphical Exposure Modeling System, or GEMS.
Because specific information required to estimate emissions of
2 3,7»8-TCDD and 2,3,7,8-TCDF to air at each site was not available, all
sites in the inventory were assumed to be of identical size, to contain
sludge with the same concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF, to
use the same management practices, and to emit contaminants at identical
rates. These assumptions varied between "typical" and "MEI" estimates,
but were applied consistently across all facilities.
(c) Estimates of human exposures and risks. Based on
estimated concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in ambient air,
individual exposure and cancer risk were calculated using Equations 5-1
and 5-2, respectively.
ISCLT provided estimates of population-weighted average
concentrations of the 2,3,7,8-TCDD or 2,3,7,8-TCDF In ambient air
Surrounding all surface impoundments. Aggregate exposures and cancer
risks-were calculated using Equations 5-3 and 5-4, respectively.
(d) Estimates of populations exposed The area source version
£ the model estimates ambient air concentrations at selected locations
in a polar grid centered on an area source of emissions and extending 560
kilometers in all directions. It then maps those air concentrations onto
ctual human populations for the regions Involved. Population data are
5-31
1592H
-------
drawn from the 1980 Census, mapped to the level of Census block group and
enumeration district.
(2) Estimates of exposures and risks from incestion of ground water.
If 2,3,7,8-TCDD and 2,3,7,8-TCDF are dissolved in water that seeps from
the bottom of a surface impoundment, these contaminants may be
transported to an aquifer beneath the facility. If those chemicals are
then transported down-gradient by ground water, then they may reach
drinking water wells near the site. This section outlines the methods
used by this analysis to assess the extent of this potential route of
exposure and risk. Table 5-14 lists key assumptions and input parameters
used to estimate exposures and risks from ingestion of ground water.
(a) Methods for estimating seepage beneath a surface
impoundment. Potential ground-water contamination from surface
impoundments was estimated similarly to that from landfills, with one
important difference. If a surface impoundment contains a significant
volume of water, drainage from the pond may result in increased recharge
and contaminant loading to the underlying aquifer. The extent to which
the downward flux of water beneath a site is increased by water from the
impoundment will depend on:
• the amount of water in the impoundment,
• whether a natural or synthetic liner is present,
• whether the water content of the impoundment is periodically
restored by additional deposits of sludge,
• the extent to which the solids layer on the bottom of the
impoundment inhibits water flux out of the impoundment, and
• the hydraulic conductivity of the medium between the impoundment
and the water table.
Of course, if aquifer recharge beneath an Impoundment significantly
exceeds the sum of net precipitation and the annual influx of water from
sludge, then the impoundment will soon dry out. A sustained and
substantial increase in ground-water recharge beneath an impoundment
therefore requires repeated additions of water to an active Impoundment.
Movement of TCOD (and to a lesser extent TCDF) through soil is
retarded by its high soil/water partition coefficient. Given a
significant soil layer between the impoundment and the water table,
steady-state loadings of these contaminants to groundwater beneath an
impoundment may not be reached until many years after the sludge 1s last
placed in the facility. If the active lifetime of the impoundment ends
before 2,3,7,8-TCDD and 2,3,7,8-TCDF reach the underlying aquifer, then
5-32
1592H
-------
: ffrauad-tater fytbay
\nput
parameter
Estimates for
Typical individual
Estimate for
ME1 individual
Notes/explanation/(reference)
Model used for transport through
unsaturated zone
Result used for exposure estimates
Vertical distance to aquifer (m)
Aquifer recharge (cm/yr)
w Vadose zone medium
SESOIL
SESOIL
CO
CO
Hydraulic conductivity of unsaturated
zone (•/sec)
Effective porosity of aquifer medium
Hydraulic conductivity (m/hr)
Hydraulic gradient
Longitudinal dispersivity (•)
Steady state
3
43
Sand & gravel.
silt & clay
1 x 10"7
0.25
10
IX
20
Steady state
3
43
Sand & gravel
silt & clay
1 x 10~7
0.25
10
IX
ZO
Low and best est imates assure that long tern average
recharge to aquifer is not affected by water in
impoundment. High estimate assumes recharge limited by
hydraulic conductivity in unsaturated zone.
Assumes 6-meter depth to water table (DRASTIC data base).
Estimate represents typical recharge rate for areas with
surface impoundments. Assunes long-term recharge is not
affected by Mater in impoundment.
Host cannon medium in counties containing surface
impoundments (DRASTIC data base).
Selected value is between ranges typical for sand and
for clay. Would result in complete drainage of inactive
lagoon within three years (Freeze and Cherry 1979).
Appropriate for sand and gravel (Freeze and Cherry 1979.
Teh 1981).
Appropriate for productive sand and gravel aquifer
(Freeze and Cherry 1979).
"Typical" value (USEPA 1988b).
Within range of "typical" values reported for sand.
Equals appro*. 1/10 distance to receptor well (Teh 1981).
-------
8897H
Table 5-14. (Continued)
Input
parameter
Estimates for
Typical individual
Estimate for
ME I individual
Notes/explanat i on/(reference)
Transverse dispersity (•)
Vertical dispersivity (•)
Within range of "typical" values reported for sand.
Equals 1/3 longitudinal dispersity (Teh 1981).
Within range of "typical" values reported for sand (Yen
1981).
cn
CO
TCDD Kg in aquifer
TCDF Kg in aquifer
Distance to model wells (•)
Distances represented (•)
Exposed populations: 0-400 •
400-2.000 •
2.000-4.000 m
10.000 10.000 Does not affect steady-state results.
35 35 Does not affect steady-state results.
200 200 Midpoints of ranges listed below.
1.200
3.000
0-400 200
400-2.000
2.000-4.000
63 — Assumes density of 68 persons/sq. mile (USDOC 1987).
1.500
4.670
MA = Hot applicable.
-------
the water content of a surface impoundment may not result in appreciable
increases in the peak or "steady state" loading of sludge contaminants to
the aquifer.
Ideally, ground-water contamination beneath a surface impoundment
could be modeled separately for the active phase of the impoundment
(during which sludge is regularly added to the impoundment and the liquid
content of the impoundment is maintained or increased), and the inactive
phase of the impoundment (during which no further sludge is added, and
the liquid content is decreasing or is at equilibrium with precipitation
less evaporation). For lack of sufficient data, the present analysis did
not attempt to model these two periods separately, or to quantify the
exact extent to which water in an impoundment would increase recharge
(and contaminant loading) to an underlying aquifer.
The analysis assumed that the impoundment would be inactive during
the period in which maximum loading of 2,3,7,8-TCDD and 2,3,7,8-TCDF to
the aquifer takes place. If so, then the resulting loading to ground
water could be calculated with methods identical to those used to
estimate ground-water contamination beneath sludge landfills. Estimates
of exposure and risk were based on results from SESOIL simulations
similar to those described in Section 5.2.1.
(b) Contaminant concentrations In seepage beneath an
impoundment. Contaminant concentrations in a surface impoundment are
likely to vary with depth and the age of the impoundment; dissolved
concentrations may be highest at the bottom of the impoundment, where
sludge has settled and the solids are concentrated. As an upper bound
estimate of water concentrations at the bottom of the impoundment, this
analysis assumed that concentrations were unlikely to exceed values
suggested by the ratio: dry weight concentration of contaminant in
sludge/soil-water partition coefficient for contaminant.
As an alternative approach, it was reasoned that 2,3,7,8-TCDD and
2 3,7,8-TCDF from a typical surface impoundment may not reach the aquifer
during the active lifetime of the facility. If so, then long-term
loadings of 2,3,7,8-TCDD and 2,3,7,8-TCDF may be better modeled based on
tne assumption that the surface impoundment had lost its excess moisture
and behaved as a soil column similar to those modeled in Section 5.2.1
for landfills. The SESOIL model allows a more comprehensive approach to
contaminant transport through soil layers. The model considers monthly
climate data, and maintains a mass balance for contaminant transport
through multiple soil layers. Estimates of potential human exposure and
risks through the groundwater pathway were derived by using the SESOIL
model to simulate 2,3,7,8-TCDD and 2,3,7,8-TCDF transport to the aquifer
once steady state conditions had been reached in the unsaturated zone.
Loadings to ground water were proportional to assumed recharge beneath
5-35
1592H
-------
the landfill. A value of 43 centimeters recharge per year was selected
based on GEMS data for a county in Wisconsin.
(c) Use of AT123D to predict contaminant transport through the
aquifer. From estimated loadings of 2,3,7,8-TCDD and 2,3,7,8-TCDF to
groundwater, the AT123D model was used to predict contaminant
concentrations at wells down-gradient of each site. As with landfills,
this analysis considered only the "steady state" concentrations predicted
by the model, without regard to the amount of time required to reach
steady state.
(d) Estimates of exposures and risks. Based on assumed rates
of individual water ingestion per day (2 liters per day) and exposure
over an entire lifetime, individual exposure and cancer risk were
calculated using Equations 5-5 and 5-6, respectively.
Exposure and risk were estimated separately for persons taking
drinking water at each of the three model distances from a surface
landfill site. HEI exposure and risk were assumed to occur at the
nearest well location.
(e) Estimates of exposed populations. To calculate the size
of the exposed populations, the size of the affected area downgradient of
each site (within 400, 2000, or 4000 meters) Is multiplied by the average
population density (68 persons per square mile) (USDOC 1987). The
results are multiplied by 20 (i.e., the number of facilities reporting to
use impoundments) (USEPA 1989a).
(3) Estimates of exposures and risks from Inoestlon of surface
water. The extent of exposure and risks associated with surface water
pathways for surface impoundments will depend on the characteristics of
individual surface impoundment sites, and on management practices used to
contain the sludge. If a facility is surrounded by a substantial berm,
for example, runoff from the impoundment will be minimized. This
analysis conservatively assumed that runoff from an inactive surface
impoundment would result in the same amount of soil transport from the
facility per unit area as estimated for landfills without cover. The
methodology used for this analysis is thus nearly identical to the
approach discussed in Section 5.2.1.
Table 5-15 presents the assumptions and input parameters used to
calculate exposures and risks as a result of ingestion of surface water
from disposal of TCDO- and TCDF-contaminated pulp and paper sludge in
surface impoundments.
(a) Estimating 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations In
surface water. In general, these calculations used the Universal Soil
Loss Equation, together with estimates of sediment delivery ratios, to
5-36
1S92H
-------
: Surface tatec fyttey
Input
parameter
Cover
Vegetation
Distance fro* site to surface water
(•eters)
"C" ratio
"p" ratio
Drainage area (hectares)
in
i
u>
•^j
Population density (people/ square Mile)
Percent of population served by surface
Mater
Fish consumption (g/day)
Percent of water and fish consumed
that is contaminated
Fraction organic carbon in soil
Fish to sediment ratio: TCOO
TCOF
Estimate for
Typical individual
Hone
SOX
152
1.1:1
1:1
1.295.000
68
49
6.5
100
0.01
1
1
Estimate for
MEl individual
None
None
30
10:1
1:1
4.047
MA
NA
140
100
0.001
10
10
Motes/explanat i on/ ( reference )
USEPA (19B8b).
USEPA (1988b).
All scenarios assure that the surrounding drainage area
is pasture land (USDA 1978).
Assures no support practices.
The typical scenario assumes the sites drain into a major
waterway. The HE I scenario assumes the sites drain into
a smaller tributary.
USOOC (1987).
USGS (1985).
Average U.S. consumption; subsistence fisher.
USEPA (1988b)
USEPA (1988b)
MA = Hot applicable
-------
estimate the fraction of a lake or stream's sediment that originated from
the landfill. By multiplying this fraction by the original concentration
of 2,3,7,8-TCDD and 2,3,7,8-TCDF in sludge or soil particles on the
landfill surface, the methodology derived estimates of the concentration
of contaminants in the sediment. This contaminant load was then
partitioned between adsorbed and dissolved phases, based on the
assumption of equilibrium partitioning between the two phases.
(b) Estimates of exposures and risks. Individual exposure and
cancer risks were calculated using Equations 5-5 and 5-6, respectively.
(c) Estimates of populations exposed. The population exposed
to contaminated water is estimated by multiplying the area of the
drainage basin by the estimated population density (68 persons/sq.
mile). This estimated population was then multiplied by the fraction of
the population that takes its drinking water from surface supplies (49
percent).
(4) Estimates of exposure and risks from Ingestion of fish from
surface water sources. Where pulp and paper sludge is deposited in
uncovered landfills, particles of sludge or soil from the landfill
surface can be transported by erosion to nearby lakes or streams. If the
sludge contains 2,3,7,8-TCDD or 2,3,7,8-TCDF, then those particles can
carry these contaminants to the surface water bodies. Fish living 1n the
lakes or streams can take up sludge contaminants into their tissues; if
humans then consume those fish, they can be exposed.
This section discusses methods that were used to estimate the extent
of this potential exposure, and its associated risks to human health.
The methodology was quite similar to that discussed in Section 5.2.3(3),
1n that both methodologies began by estimating sediment concentrations of
2,3,7,8-TCDD and 2,3,7,8-TCDF in water bodies as a result of runoff from
landfills. Once sediment concentrations were estimated, however, the
methodology departed from that described in Section 5.2.3.(3), and used
fish-to-sediment bioconcentration factors and estimates of human fish
consumption contaminant doses to humans. The last step in the
methodology involved estimating the sizes of exposed populations, and
combining these results with estimates of individual dose and health risk
to derive total health risks to the entire exposed population.
(a) Estimating the concentration of 2,3,7,8-TCDD and
2,3,7,8-TCDF 1n sediments and fish tissues. This analysis used the
Universal Soil Loss Equation, together with estimates of sediment
delivery ratios, to estimate the fraction of a lake or stream's sediment
that originated from the landfill. By multiplying this fraction by the
original concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF 1n sludge or soil
particles on the surface of the landfill, the methodology derived
estimates of the concentration of contaminants in the sediment. These
5-38
1592H
-------
sediment concentrations were then used to estimate contaminant
concentrations in the tissues of fish.
Based on the assumption that sediment concentrations were the best
predictor of fish concentrations of hydrophobic compounds like
2,3,7,8-TCDD and 2,3,7,8-TCDF, the methodology used empirical
fish-to-sediment bioconcentration factors to estimate concentrations of
contaminant in freshwater fish as a function of concentrations in stream
or lake sediment. The concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
the muscle tissues of fish (consumed by humans) were considered to be
fifty percent lower than the whole body concentrations of these
contaminants.
(b) Estimates of exposures and risks. Individual exposure and
cancer risk were calculated using Equations 5-7 and 5-8, respectively.
(c) Estimates of populations exposed. The population exposed
to fish containing 2,3,7,8-TCDD and 2,3,7,8-TCDF was estimated by
multiplying the area of the drainage basin containing each facility by
the estimated population density (68 persons/sq. mile).
(5) Summary of results. Table 5-16 presents risks to humans from
the four potential pathways of exposure considered for disposal of sludge
in surface impoundments. Exposure through ingestion of fish from surface
rtater sources showed the highest risk, followed by ingestion of drinking
water. Highest total population risk was observed through the
volatilization pathway.
5 2.4 Exposures and Risks from Land Application of Pulp and Paper Hill
Sludge
Land application of sludge, while an alternative disposal method, also
fertilizes and conditions soil and allows the sludge to be used as fill,
According to the 104-M111 Study and follow-up conversations with state
environmental offices, sludge from four mills in three states is used on
forests; sludge from two mills in two states is applied to agricultural
land; and sludge from two mills in two states is used to reclaim abandoned
mine sites. Conversations with state environmental offices indicate that
approximately 325,000 dry metric tons of sludge are being land applied per
Sear with about 10 percent to mine reclamation, 10 percent to agriculture,
and 80 percent to forests. This analysis examines the risks to human
health from the land application of pulp and paper mill sludges.
In this assessment, two application rates are used - one for the
aflH cultural application scenario and one for the mine and forest
scenario. To estimate reasonable values for these generic
JJJpli cation rates, the rates of agricultural application obtained from
tales were averaged and the rates of forest and mine application
5-39
1592*
-------
Table 5-16^ Estimates of Health Risks to the General Population from Surface
Impoundment of Pulp and Paper Sludges Contaminated with 2,3,7.8-TCDD
and 2.3.7,8-TCOF
Exposure pathway
Inhalation exposure from volatilization from
surface impoundments in which sludge is disposed
(Percent TCDO)C
ME I
risk3
(per
lifetime)
IxlO'6
(0.6)
Typical
risk3
(per Exposed
lifetime) population
5xlO"8 7.100,000
(0.7)
Total
risk
(cases/
year)
5xlO"3
(0.7)
Ingest ion exposure from drinking ground water
contaminated by leachate from surface impound-
ments in which sludge is disposed
(Percent TCDO)C
3xlO"8 5xlO"10 6.000
(0.4) (0.7)
-8
4x10
(0.7)
Ingest ion exposure from drinking surface water
contaminated by surface runoff from surface
impoundments in which sludge Is disposed
(Percent TCOO)C
2xlO"3 6xlO"8 2.330.000 2xlO*3
(0.6)
(0.7)
(0.7)
Ingest ion exposure from fish caught In surface 1x10
water contaminated by runoff from surface
impoundments in which sludge is disposed
(Percent TCDD)C (63)
-1
IxlO"7 4.760.000 7xlO"3
(65)
(65)
a Calculated as: (Estimated Exposure) x 1.6x10 per (pg/kg)/day.
13 Calculated as: (Typical Risk x Exposed Population) / (Life Expectancy)
c (Exposure to TCDD)
Calculated as: 100 x (Expogure tfl TC[)n) + (1/1Q) (Exposure to TCOF)-
5-40
903 5H
-------
obtained from states were averaged. Application rates that were
estimated based on practices in other states were not included in the
averages.
It is assumed that sludge that is applied agriculturally is
incorporated to a depth of six inches. Sludge applied to reclaim mines
is assumed to be top-dressed. Silvicultural application is analyzed as a
unit with mine reclamation, and conservatively assumed not be be
incorporated with the soil. Calculations of soil concentrations for mine
reclamation and Silvicultural applications assume a one-time
application. Agricultural applications of sludge are assumed to continue
for 70 years in the MEI analysis and for 20 years in the typical
analysis. These assumptions are summarized in Table 5-17.
Since likely exposure pathways differ depending on the type of land
receiving the sludge, exposure pathways considered in this analysis
differed for agricultural and mine/forest application. In addition,
farmers will be exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF from
agricultural application through pathways that are not relevant to the
population at large. The following pathways of exposure were considered
in this analysis:
Risk .estimates for forest application/mine reclamation considered two
pathways for both a total population and a most exposed individual (MEI)
risk:
• Contaminated soil erodes from the forest/mine site and contami-
nates surface water and stream sediment. The surface water serves
as a drinking supply.
• Contaminated soil erodes from the forest/mine site and contami-
nates surface water and sediment. 2,3,7,8-TCDD and 2,3,7,8-TCDF is
incorporated into fish tissue and fish are consumed by humans.
Risk estimates from agricultural application considered the two pathways
above and added the following pathway for both a total population and a
subsistence farmer's (MEI) risk:
• Small amounts of contaminant are taken up Into the tissues of
crops. These crops are then either consumed or fed to animals which
bioconcentrate the contaminant and produce a meat or dairy product
.that is consumed.
Other exposure pathways particular to farming were:
• Children and adults in the farming household come into direct
dermal contact with the sludge-amended soil in both outdoor and
indoor settings. 2,3,7,8-TCDD and 2,3,7,8-TCDF from the sludge is
5-41
1592H
-------
8897H
Table 5-17. Assumptions and Parameter Values - Land Application: All Exposure Pathways
tf*
i
rs>
Input
parameter
Application rate (DMT/HA)
Forest/nine
Agriculture
Tears land receives sludge
Forest/Mine
Agriculture
Incorporation depth (cm)
Forest/Mine
Agriculture
TCOO concentration in soil (ppt)
Forest/nine
Agriculture
TCOF concentration in soil (ppt)
Forest/Mine
Agriculture
Estimate for
typical individual
776
38
1
20
0
15
163
35
885
188
Estimate for
ME I individual
776
38
1
70
0
15
293
107
1.760
645
Notes /exp lanat 1 on/ reference )
Average of individual states (State Environmental
Officials)
Estimates based on information front individual states
Estimates based on information from individual states
Average and 90th percent i le values for all mills
1989a).
See Section 5.2.4(l)(a).
Average and 90th percent i le values for all mills
1989a).
See Section 5.2.4(l)(a).
(USEPA
(USER A
-------
so
absorbed through the skin. Children ingest small amounts of the
sludge/soil mixture through normal mouthing behavior. Adults al
inadvertently ingest small quantities of sludge/soil.
• 2,3,7,8-TCDD and 2,3,7,8-TCDF applied to the farmland volatilizes
from the sludge into the air. Residents of the farm inhale the
volatilized 2,3,7,8-TCDD and 2,3,7,8-TCDF.
• Particles of the sludge/soil mixture become suspended in the air
during application. Residents of the farm inhale the contaminated
particles.
The following sections describe the methods and data used to estimate
risks from land application of sludge through these pathways. Results are
summarized in the final section.
(1) Estimates of exposures and risks from dermal contact with
skin. Humans coming in direct contact with sludge contaminated soils
jjjay~absorb 2,3,7,8-TCDD and 2,3,7,8-TCDF through their skin. The amount
of 2,3,7,8-TCDD and 2,3,7,8-TCDF absorbed depends on the area of skin
exposed and on the length of time that the contaminated soil is in
contact with the skin. The following discussion summarizes the process
used to estimate exposure through dermal contact. The values used for
each model input are summarized in Table 5-18.
To estimate exposure through direct contact with soil containing
2 3,7»8-TCDD and 2,3,7,8-TCDF, various methodologies were used. The
dermal exposure model used empirically-derived information on the amount
of soil or dust that adheres to a square centimeter of skin, the area of
skin exposed in various settings, and the absorption rate of 2,3,7,8-TCDD
or 2,3»7,8-TCDF through skin to derive the dose of 2,3,7,8-TCDD or
2 3,7»8-TCDF from dermal contact with contaminated soil or dust.
The estimation of dermal exposure proceeded in two steps. First the
average daily exposure from dermal contact was calculated as the product
of area of skin affected, the contact rate, the dermal absorption rate
and the duration of contact. Second, the risk from dermal contact was
Calculated using the estimate of daily exposure and the potency of
o\ 7,8-TCDD and 2,3,7,8-TCDF.
c.»J»''
(a) Methods for estimating soil concentrations. The methods of
cludge application considered were (1) application as a top dressing, and
(2) incorporation into a layer of soil. Concentrations in the soil for
11 top dressing scenarios were assumed equal to concentrations in the
fand-appl^d sludge. For sludge that was incorporated Into the soil, the
nncentrations were dependent on the depth of the sludge incorporated in
the soil* frequency of application, and initial sludge concentration.
5-43
1592H
-------
8897H
Table 5-18. Assumptions and Parameter Values - Land Application: Dermal Pathway
Input
parameter
Estimate for
Typical individual
Estinte for
HE I individual
Notes/explanat ion/(reference)
in
COHCEMTMTIOH
2
Contact rate (mg/cm), outdoor,
child
o
Contact rate (mg/cm ]. outdoor.
older child
Contact rate (mg/cm ). outdoor,
work.
Contact rate (mg/cm ). indoor, child
Contact rate (mg/cm ). indoor, older
child
Contact rate (mg/cm2):
- indoor, adult, living space
- indoor, adult, attic
Ratio of contaminant concentration
indoors to soil concentration out-
doors
EXPOSURE
Time spent outdoors, child (hr)
0.5
0.5
3.5
O.OS6
O.OS6
0.056
1.8 (110 mg/hr)
0.80
1.565
1.5
1.5
3.5
0.06
0.06
0.06
0.85
2.190
Schaum (1984).
Schaum (1984).
Estimate front Haw ley (1985) based on adults doing yard
Estimate from Hawley (1985) assuming dust fall indoors = 20X
outdoors, with cleaning every two weeks.
Estimate Tram Hawley (1985) assuming dust fall indoors = 20X
outdoors, with cleaning every two weeks.
Contact rate for adults in living space same as for
children; for work in attic value of 1.8 ing/on for
direct contact; for indirect contact with dusty air. contact
estimated as 110 mg for 1 hour of attic work (Hawley 1985).
Ha*ley (1985).
Typical estimate: assumes 5 days/wk. 6 months/yr. 12 hrs/day
(12 hrs before soil is washed off). HEI: 7 days/wk. 6
months/yr. 12 hrs/day soil contact (Hawley 1985).
-------
Input
parameter
Estimate for
Typical individual
Estimate for
MEI individual
Notes/explanat ion/(reference)
EXPOSURE
Ti«e spent outdoors, older child (hr)
1.824
Tiae spent outdoors, adult (hr)
Ti«e spent indoors, child (hr)
1.565
4.380
cn
i
Ti*e spent indoors, older child (hr)
Tine spent indoors, adult (hr). living
space, attic
1.460
4.380
48
Area of skin exposed outdoor, child
(a.2)
2.100 cm2
2.190
1.565
6,570
4,380
6.570
72
2.800 of-
Typical: child outdoors from Nay to September every day (150
days). 12 hrs of soil contact before washing. MEI: child
outdoors every day for 6 mntns. 12 hrs soil contact (Hawley
1985).
Assunes 12 hrs soil contact before washing. 5 days/**.. 6
•onths/yr (130 days) (Hawley 1985; Keenan et al. 1989).
Typical: 12 hrs/day of indoor dust contact all year. HEI:
assuoes 24 hrs/day of indoor dust contact for 6 winter months.
plus 12 hrs/day for 6 suner Months (Hawley 1985; Keenan et
al. 1989).
Typical: assumes 4 hrs/day of indoor dust contact all year.
NEI: assumes 12 hrs/day of indoor dust con-
tact all year (Hawley 1985).
Living space: low and best estimate assunes indoor dust
contact 12 hrs/day, all year; high assunes indoor dust contact
24 hrs/day. 6 winter •onths, plus 12 hrs/day. 6 sunoer
•nnths. Attic: low assunes 1 day in attic with soil left on
skin for 12 hrs while in attic plus 4 hrs before washing; best
estimate assumes 12 days with
soil on skin 4 hrs/day; high assunes 12 days with soil on skin
6 hrs/day (Hawley 1985).
Low assunes only hands exposed (child wearing pants and
long-sleeved shirt). Best estimate assumes both hands, legs,
and feet exposed during play. High assunes both hands, arms.
legs, and feet exposed (Hawley 1985).
-------
8897H
Table 5-18. (continued)
Input
parameter
Estimate for
Typical individual
Estimate for
ME I individual
Notes/explanation/(reference)
EXPOSURE
Area of skin exposed outdoor, older
child (c*2)
1.600
3.200
Typical: estimate assures both hands, forearms, legs from
knees down exposed during play. NEI: assumes both hands.
legs, and feet exposed during play (Hawley 1985; Keenan et al.
1989).
tn
i
o»
Area of skin exposed outdoor, adult
(cm2)
Area of skin exposed indoor, child
(cm2)
1.700
500
2.940 cm Typical: assumes both hands, most of forearms exposed.
MEI: assumes adult is wearing short-sleeved shirt, with an
open neck, pants, shoes, no gloves or hat (Hawley 1985; Schaum
1984).
2,800 cm Typical: estimate assumes area exposed equals one half
of surface area of child's feet, hands, and forearms. MEI:
assumes all of area of feet, hands, and forearms (Hawley 1985).
Area of skin exposed indoor, older
child (cm2)
Area of skin exposed indoor, adult
(cm2)
Bioavailability through skin
Availability of dioxin from soil
matrix
400
1.700 cm2
(attic)
900 cm2
(living space)
0.012 h
-1
3.200
2.940 cm£
in both attic
and living
space
0.024 h
1
(children
only)
15X
Typical: estimate assume only hands are exposed.
MEI: assumes area exposed indoors sane as outdoors (Hawley
1985).
Typical: estimate assumes hands exposed in living
space, while adult wears short-sleeved shirt, with an
open neck, pants, shoes, no gloves or hat, to work in
attic. NEI: assumes area exposed indoors sane as outdoors
(Hawley 1985).
Typical: data from Appendix A.I. MEI: assumed that skin
of children has twice the absorption of adults (Hawley 1985).
Recomnended value: IX (Appendix A.I).
-------
The mass of the contaminant added was calculated by multiplying the
application rate by the contaminant concentration in the sludge. The
volume of soil with which the sludge was mixed was determined by
multiplying the incorporation depth by the incorporation area. The
volume was then multiplied by the soil density to obtain the mass of the
soil with which the sludge was incorporated. Average soil concentrations
for a given year were estimated by adding the mass of contaminant applied
to the land for that specific year to the mass present from previous
applications and then dividing by the mass of the receiving soil plus the
mass of the applied sludge.
(b) Estimates of exposures and risks. The indoor dust and
outdoor soil contaminant concentrations were used to estimate exposure
and risk from dermal contact with these media. Daily doses were
estimated for three age groups: young children (ages 1-6), older
children (ages 7-11), and adults (ages 12 and older). The dose for each
age group was calculated as:
DOSEg = [[(C0)(CR0rg)(SAgUt>g)(H0>g)(ABd)(M)] (5-10)
where:
AB(j • systemic absorption rate through the skin
BWn - body weight = 70 kg for adult, 16 kg for young child,
9 35 kg for older child
C0 - concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in soil
outdoors, mg/mg
Cin - concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in dust
indoors, mg/mg
CR.jp q - contact rate of soil with skin for age group g, indoors,
mg/cnr
CR0 q • contact rate of soil with skin for age group g, outdoors,
mg/cnr
DOSEq * dose from outdoor exposure for age group g, mg/kg/day
HJ q - hours spent indoors for age group g
u g - hours spent outdoors for age group g
SAin.g " surface area of skin exposed to soil for age group g,
indoors, cm
H • fraction ot 2,3,7,8-TCDD or 2,3,7,8-TCDF that migrates
from the soil matrix and contacts skin
SAOL|t g - surface area of skin exposed to soil for age group g,
outdoors, cnr
For each age group, the soil contaminant concentration (expressed in
/mg for ease of calculation) was multiplied by the soil contact rate
utdoors (mg/cnr) and by the area of the skin exposed during outdoor
5-47
-------
activity (cm2) to obtain the total quantity of soil-bound 2,3,7,8-TCDD
or 2,3,7,8-TCDF adhering to the skin (mg). The quantity of contaminant
on the skin was then adjusted by two factors: the fraction of the
contaminant that migrates from the soil matrix and comes into contact
with the skin; and the fraction of 2,3,7,8-TCDD or 2,3,7,8-TCDF that
absorbs through the skin. Since the dermal absorption rate was expressed
as the fraction of 2,3,7,8-TCDD or 2,3,7,8-TCDF that is absorbed through
the skin per hour of contact, it must be multiplied by the hours that the
soil was assumed to be in contact with the skin. The same calculations
were also performed for exposures in indoor settings, using the
corresponding indoor values for the model input parameters. The total
daily dermal absorption of 2,3,7,8-TCDD or 2,3,7,8-TCDF was the sum of
indoor absorption and outdoor absorption. Dividing the total dermal
absorption for each age group by the body weight for that age yielded a
daily dose of 2,3,7,8-TCDD or 2,3,7,8-TCDF through dermal absorption in
mg/kg/day.
To obtain the weighted average dose over the lifetime of an individ-
ual, the following calculation was used:
DOSEavg - I (FRACg)(DOSEg) (5-11)
where:
DOSEaVg - weighted average daily dose for an individual, mg/kg/day
DOSEg = daily dose for individual in age group g
FRACg - fraction of lifetime spent in age group g
Once the daily dose estimate was obtained, it was combined with
information about the potency of 2,3,7,8-TCDD and 2,3,7,8-TCDF to obtain
an estimate of lifetime risk from dermal exposure to these contaminants.
Maximum and average individual risks were calculated as:
1C » (DOSEavg)(fll*) (5-12>
where:
DOSEaVg - weighted average daily dose for an individual, pg/kg/day
1C • individual cancer risk over lifetime from DOSEava of
2,3,7,8-TCDD or 2,3,7,8-TCDF y
qi - incremental lifetime risk per pg/kg/day dose of
2,3,7,8-TCDD or 2,3,7,8-TCDF
Individual cancer risk for an average exposed individual was converted to
annual total population risk (in cases per year) by multiplying the number
of persons exposed by the individual risk and dividing by the average
person's lifespan, as described in the following equation:
5-48
1592H
-------
PC = (ICHPOP1 (5-13)
LS
where:
LS = average lifespan of an individual • 70 years
PC = population risk, cancer cases per year
POP = population exposed to DOSEaVg
(c) Estimates of populations exposed. The population exposed
to 2,3,7,8-TCDD and 2,3,7,8-TCDF through dermal contact was limited to
the population residing on the agricultural land application sites. The
number of sites applying kraft mill sludge to land was equal to the total
number of acres applied with sludge in the state divided by the average
number of acres per site. Values for both the total acres and the acres
per site were obtained through conversations with state officials in
Mississippi and Pennsylvania, the states where agricultural land
application is currently practiced. In Mississippi, 1,000 acres are
applied with the sludge from one mill, with an estimated 100 acres per
site, yielding an estimate of 10 sites in Mississippi. Pennsylvania has
50 acres covered with sludge from one mill, with an average of 12 acres
per site, giving a total of 4 sites. The total number of sites in each
state was then multiplied by the number of people living on each site to
obtain the exposed population. According to the 1980 U.S. Census, the
average number of persons per household is 2.7. In Pennsylvania, the
exposed population is approximately 11 persons, while in Mississippi, the
exposed population is approximately 27. The total population exposed was
estimated to be 40 persons.
(2) Estimates of exposures and risks from Incestlon of produce.
m(int, and dairy products grown on sludge-amended land. Sludge is
TppHed to various types of land, including forest, abandoned mines,
nasture, and land used for the production of animal feed or human food
crops. This analysis evaluated the application of sludge to pasture and
and examined the following potential pathways of exposure:
Sludge is incorporated into the soil of farmland used for produc-
ing food crops. Contaminants in the sludge are drawn from the soil
to the tissue of those crops, and are then ingested by humans who
consume the crops directly.
Sludge is incorporated into the soil of farmland used for produc-
ing animal feed or pasture. Contaminants in the sludge are absorbed
into the tissues of these feeds or pasture grasses, which are then
consumed by livestock. The meat and dairy products produced by
these livestock are consumed by humans.
5-49
159ZH
-------
• Sludge is applied to the surface of pasture land, and adheres to
the pasture grasses. Grazing cattle or sheep ingest the sludge
directly as a fraction of their pasture consumption. Contaminated
beef or dairy products are then consumed by humans.
The concentration of contaminants in food products from land applica-
tion of paper and pulp mill sludge was estimated using a model that uses
information regarding sludge application rates and sludge contaminant
concentrations to calculate the uptake of contaminants by crops and by
animals feeding on crops and pasture. Exposure was calculated using data
on dietary consumption of these meats and crops. For each scenario, the
model was given input data for sludge application rates, concentrations
of individual contaminants in the sludge, uptake rates of soil
contaminants to various crop tissues, uptake rates of contaminants in
animal feed to meat or dairy products, the fraction of each type of feed
in animal diets, the production yield of animal product per unit of food,
human dietary data, the acreage of sludge-amended land devoted to each
crop, the productivity of land for each crop, and the population of the
distribution area. Values assumed for these parameters that were used to
estimate MEI and typical risks are presented in Table 5-19. The model
returned exposure estimates which were then used to estimate risk.
Exposure and risk were calculated for both a typical individual and a
MEI. The MEI was assumed to be a subsistence farmer applying sludge to
all crop types grown on his/her land. Each exposure calculation
consisted of three steps. First, the model calculated tissue
concentrations of contaminants in each crop as a result of the land
application of sludge. Second, the model estimated concentrations of the
contaminants in meat or dairy products. Third, the model summed the
amount of each contaminant in all crops and animal products ingested to
estimate typical exposure or MEI exposure. The estimated soil
concentrations for each land application site are presented in Table 5-1'•
(a) Methods for estimating son concentrations. The concentra-
tion of 2,3,7,8-TCDD and 2,3,7,8-TCDF in the soil outdoors was derived as
described in Section 5.2.4(1)(a).
(b) Methods for estimating tissue concentrations in crops.
Each crop's uptake rate was applied to the soil contaminant concentra-
tions to estimate the concentration of each contaminant per unit dry
weight of crop tissue. Dry weight tissue concentrations were converted
to fresh weight concentrations to match units of human consumption data.
(c) Methods for determining tissue concentrations in meat and
dairy products. To calculate meat and dairy product contaminant
concentrations, the model derived average concentrations of each
contaminant in each animal's feed mix. Average concentrations were
calculated by taking a weighted average of the contaminant concentration5
5-50
159ZH
-------
Table 5-19. Assuo^it Ions and Parameter Values - Agricultural Application: Dietary Pathway
Input Estimate for Estimate for
parameter typical individual ME! individual Notes/explanation (reference)
Percent of cow's consumption that is
soil (dry weight)
Animal product yields:
Ib beef/lb feed
Ib milk/lb feed
Ib hog/lb feed
Ib chicken/ Ib feed
Ib catfish/ Ib feed
Percent of product that is fat:
beef
milk
hog
en chicken
ui
Fish filet concentration as a percent
of whole body concentration
Animal consumption (% of total dry weight
consumption)
- corn
beef
dairy
hogs
chicken
- soybeans
catfish
- pasture
beef
dairy
8% 8X USEPA (I988b)
H.A. USDA (1987)
0.07
0.97
0.17
0.40
0.5
10 13 Kimbrough et al. (1984); NCASI (1987); Wisconsin DHSS (1989)
4 4
7 8
10 10
50 50 See Section 2 of this report.
USEPA (1989c); Delta Western (1989)
2 2
8 8
77 77
59 59
50 50
38 38
38 38
Size of population over which produce
is distributed
239,000,000
H.A.
Assumes national distribution
-------
8897H
Table 5-19. (Continued)
Input Estimate for
parameter typical individual
Uptake rates - (dry weight)
wheat 0.02
corn 0.02
grass 0.02
soybeans 0.02
Fish BCF 1
Beef fat BCF (fresh weight) 4
Nilk fat BCF (fresh weight) 4
Hog fat BCF (fresh weight) 4
Chicken fat BCF (fresh weight) 4
i
tn
r\> X dry Matter
grains 0.26
soybeans 0.08
grass 0.05
% hc«e grown ".A.
grains
legures
•ilk
•eat
poultry, fish
Consumption (ag/kg/day) M.A.
wheat
corn
beef fat
dairy fat
hog fat
* Chicken '
wytatri atl
Estimate for
ME I individual Notes/explanation (reference)
See Appendix A. 4
0.02
0.02
0.02
0.02
10 USEPA (19885)
6 Fries (1982)
5 Fries (1982)
6 Assured to be equal to beef fat
6 Assured to be equal to beef fat
FDA "Reana lysis of FDA Revised Total Diet Food List" as
0.26 cited in USEPA (1987)
0.08
0.05
USEPA (1987a)
2
17
40
44
34
Highest values for any age group are used for each food
3.154 (USEPA 1987b)
i nn
1 . UU
669
1.424
4Z5
796
i ,as»
-------
parameter
f or
indwduaA
for
MET individual
Notes/explanation (reference)
in
CO
Total sluidge area (ha)
Amount of beef produced fro* sludge
amended pasture (kgs)
435
5.084
Crop yields (bushels/ha):
wheat
corn
soybean
X of area producing:
wheat
corn
soybeans
pasture
X wheat to animals
X corn to annuls
X soybeans to animals
X of animal corn fed to:
beef cows
hogs
chickens
dairy cows
94
222
59
22X
33X
36X
9X
0
92X
40X
74X
3X
IX
22X
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
State contacts: Bureau of Waste Management, Wilkesbury, PA.
Regional Office; and Bureau of Pollution Control, HS Dept. of
Environmental Quality
State contacts listed above. County Extension Agent for Perry
County. MS. and USDA (1987). Divided Perry and Forest
Counties sludge-amended pasture by total pasture in the
counties. Applied this percentage to marketed cows for the 2
counties. Multiplied by kilograms per cow.
USDA (1987)
State contacts: Bureau of Pollution Control, Mississippi
Dept. of Environmental Quality; County Extension Agent for
Perry County, MS and Bureau of Waste Management. Wilkesbury,
PA Regional Office
County Extension Agency for Perry County. MS and Bureau of
Waste Hangement. Wilkesbury. PA Regional Office
County Extension Agent for Perry County, HS: Harrisburg, PA
Dept. of Agriculture; and USDA (1987). MS: State contact
said there are not conmercial dairy cows though fanners may
have them for own use. From agriculture statistics,
multiplied beefs, hogs, and chickens in the state by the
anount they each consumed. Suraned to get total consumption
and then took X's consumed by beefs, hogs, chickens.
PA: 100X of corn is assumed to be fed to dairy cows.
-------
8897H
Table 5-19. (Continued)
Input
parameter
Estimate for
typical individual
Estimate for
MEI individual
Notes/explanation (reference)
X of animal soybeans fed to catfish
100
N.A.
The MS Perry County extension agent reports that approximately
SOX of soybeans fed to animals serve as catfish feed. The
remainder are fed to hogs, broilers and laying hens. In the
absence of information on quantities fed to each animal and on
the animal yield per unit soybean feed, all soybeans fed to
animals are assumed to be fed to catfish.
N.A. = Not applicable.
tn
in
-------
in each animal's food sources. An additional source of contaminant,
direct soil ingestion, was added to the dose derived from food. The feed
contaminant concentrations were multiplied by animal bioconcentration
factors to determine fresh weight concentrations of contaminants in each
meat or dairy product.
(d) Estimates of exposures and risks (HEI). Methods of
calculating MEI exposure and population exposure differed. The
calculation for MEI exposure proceeded in three steps. First, exposure
through direct consumption of crops was calculated. To obtain the
exposure from direct crop consumption, the daily dietary consumption of
each crop was multiplied by the fraction of that crop produced in sludge
amended soil, by the fresh weight contaminant concentration of the crop,
and by the bioavai lability of the pollutant when consumed with the crop:
DC = II [(CWijMFCiMDCiMBiMlO-)] (5-14)
where:
Bi = Bioavai lability of pollutant when consumed in crop 1
(unitless); assumed to be 1.
CWjj - Tissue concentration of pollutant j in crop i (mg/kg fresh
weight),
Dc * Dose of pollutant j from crops produced with
sludge -amended soil (mg/kg/day),
DCi » Daily dietary consumption of crop i (mg/kg/day fresh
weight),
FCj - Fraction of dietary consumption of crop i grown 1n
sludge- amended soil (unitless)
In the second step, the MEI dose of the contaminant through
consumption of animals raised on contaminated feed was determined. The
equations were similar to those discussed for dose from crop consumption:
Da - Zk [(CWjk)(FCk)(DCk)(Bk)(10-6)] (5-15)
where:
Bk - Bioavai lability of pollutant when consumed in meat or
dairy product k (unitless); assumed to be 1.
CW-jk - Tissue concentration of pollutant j in meat or dairy
product k (mg/kg fresh weight),
Da • Exposure to pollutant j from animals produced with
sludge- amended soil (mg/kg/day),
DCk • Dally dietary consumption of meat or dairy product k
(mg/kg/day fresh weight),
pck • Fraction of dietary consumption of animal product k
produced from sludge -amended soil (unitless).
5-55
159ZH
-------
In the third step, doses from crop and animal consumption were summed:
Dj = Dc + Da (5-16)
where:
Dj = Total exposure to pollutant j from crops, meat and dairy
products produced with sludge-amended soil (mg/kg/day).
Once the daily dose estimate to the MEI was obtained, it was combined
with information about the potency of 2,3,7,8-TCDD and 2,3,7,8-TCDF to
obtain an estimate of lifetime risk from dietary exposure to these
contaminants. The calculation of maximum and average individual risks
was:
1C » (DOSEavg)(qi*) (5-17)
where:
DOSEavg = weighted average daily dose for an individual, pg/kg/day
1C - individual cancer risk over lifetime from DOSEava of
^ 2,3,7,8-TCDD or 2,3,7,8-TCDF y
qi » incremental lifetime risk per pg/kg/day dose of
2,3,7,8-TCDD or 2,3,7,8-TCDF
(e) Estimates of exposures and risks (population). Population
contaminant dose was calculated for each state in which sludge was
applied agriculturally and then summed over these states. The dose from
the dietary pathway was calculated by summing three inputs: dose of
contaminant bioavailable in crops for direct human consumption; dose of
contaminant bioavailable in animal products contaminated by crop consump-
tion; and dose of contaminant bioavailable in animal products contami-
nated by grazing. The average population dose was calculated as:
TDj - (Dj + AJ + Gj)/BW/DP/DY (5-18)
where:
AJ - Dose of pollutant j in animal products contaminated by
crop consumption (mg/year)
DP » Population over which the crops and animal products are
distributed
BW - Body weight (kg)
DY » Days per year
Dj • Dose of pollutant j in crops for direct human consumption
(mg/year)
5-56
1592H
-------
G.J = Dose of pollutant j in animal products contaminated by
grazing (includes soil adherence and grass uptake)
(rag/year)
TD-j - Total exposure to pollutant j from crops, meat and dairy
products produced with sludge-amended soil (mg/kg/day)
The following discussion describes the methods used to obtain each of the
three components of the estimation of total dose.
(i) Method for determining population contaminant dose from
4pgest1on of crops. The dose of contaminant bioavailable in crops for
direct numan consumption was calculated by multiplying the mass of each
crop grown on sludge- amended land that was directly consumed by the
tissue concentration of the crop. The result was multiplied by a
bioavai lability factor:
(5-19)
where:
B.J = Bioavailability of pollutant when consumed in crop i
(unitless); assumed to be 1.
CW-jj « Tissue concentration of pollutant j in crop i (mg/kg
fresh weight),
MH-j = Mass of crop i grown on sludge-amended land that is
consumed directly by humans (kg fresh weight/year).
The mass of each crop grown on sludge- amended land that was consumed
directly was calculated by multiplying total acres of the crop receiving
sludge by the crop yield. This mass was then multiplied by the percent
Of the crop consumed directly:
M1 - (AtMYj) (5-20)
where:
A^ » Sludge-amended land area on which crop i is grown
(hectares/year),
Y^ - Yield per area of crop i (kg/hectare),
(5-21)
where:
Mi - Mass of crop i grown on sludge-amended land (kg/year),
- Percent of crop i consumed directly by humans (unitless).
5-57
1592H
-------
(11) Method for determining population contaminant dose from
ingestion of animals fed on contaminated crops. To determine the dose
of contaminant available through consumption of animal products produced
from sludge-amended land, the mass of each contaminated crop fed to
animals was multiplied by an animal product yield from each kilogram of
contaminated feed adjusted by dividing by the fraction of animal diet
that the crop constitutes. To obtain a dose to humans who ingest the
animal products, this yield was multiplied by contaminant concentration
in the animal tissue, percent of the animal that is fat, and a
bioavailability factor:
(5-22)
Aj • Iik[(MAik)(YAk/FAki)(CWjk)(Bk)(PFk)]
where:
Bk = Bioavailability of pollutant when consumed in meat or
dairy product k (unitless}; assumed to be 1.
cwjk = Tissue concentration of pollutant j in meat or dairy
product k (nig/kg fresh weight),
FAk^ - Fraction of dietary consumption of crop i for animal k
(unitless),
MA-jk = Mass of crop i grown on sludge-amended land that is fed
to animal k (kg/year),
PFk - Percent fat in animal product k (unitless),
YAk - Yield of animal k per unit of corn-equivalent feed
(kg/kg).
To obtain the mass of each crop fed to each animal the mass of each
crop fed to all animals was divided between the animals according to each
animal's percentage of total consumption:
MA, - Mi - MHj (5-23)
where:
MA^ - Mass of crop i grown on sludge-amended land that is fed
to animals (kg fresh weight/year)
MA1k - HA, [(Nk)(Ck) / Xk[(Nk)(Ck)]] (5-24)
where:
• Food consumed per animal k (kg/year)
• Mass of crop i grown on sludge-amended land that is fed
to animals (kg fresh weight/year)
Nk - Number of animal k in the state
5-58
1592H
-------
(111) Method for determining population contaminant dose from
•ingestion of animals grazing on contaminated pasture land. To obtain
the dose of contaminant from animal products raised on sludge-amended
pasture land, the number of contaminated animals marketed was multiplied
by an average production weight to yield a total mass of meat available
for consumption. This mass was then multiplied by the tissue contaminant
concentration, the percent of the animal that is fat, and a bioavaila-
bility factor:
GJ - Ig[(NSg)(WHg)(CWjk)(Bk)(PFk)] (5-25)
where:
NSg - Number of each animal, g, grown on sludge-amended land
and marketed (animals/year),
WHg = Average U.S. production of grazing animal g per head
marketed (kg/animal).
The number of each type of animal raised on sludge-amended land was
the product of the percentage of pasture land that receives contaminated
sludge and the total number of animals grazed on pasture land. The
resulting number of animals was then multiplied by the percent of grazed
animals that are marketed each year:
NSg - (SPC/ Pc)(Ng)(PMg) (5-26)
where:
Nq = Number of grazing animals (g) in the counties with
y sludge-amended pasture,
Pc - Total pasture land in all counties with sludge-amended
pasture land in state (hectares),
PMq - Percent of grazing animals (g) marketed per year on a
national basis,
SPC * Sludge-amended pasture land in all counties in state
applying sludge agriculturally (hectares).
Once the daily dose estimate was obtained, it was combined with
•information about the potency of 2,3,7,8-TCDD and 2,3,7,8-TCDF to obtain
Ip estimate of lifetime risk from dietary exposure to these
contaminants. The maximum and average individual risks were calculated
as:
1C - (DOSEavg)(qi*) (5-27)
5-59
1592H
-------
where:
DOSEavg = weighted average daily dose for an individual, pg/kg/day
1C = individual cancer risk over lifetime from DOSEava of
^ 2,3,7,8-TCDD or 2,3,7,8-TCDF y
qi = incremental lifetime risk per pg/kg/day dose of
2,3,7,8-TCDD or 2,3,7,8-TCDF
Individual cancer risk for an average exposed individual was converted to
annual total population risk (in cases per year) by multiplying the number
of persons exposed to the individual risk and dividing by the average
person's lifespan, as described in the following equation:
PC = [(IC)(POP)] / LS (5-28)
where:
LS = average lifespan of an individual - 70 years
PC = population risk, cancer cases per year
POP - population exposed to DOSEaVg
(f) Estimates of populations exposed. To determine typical
exposures,, total available quantity of contaminated food was assumed to
be distributed nationally. This assumption does not affect population
risk since the quantity of contaminated food is determined independently
of the population exposed and risk is assumed to be a linear function of
exposure.
(3) Estimates of exposures and risks from direct 1noest1on of
sludae. Direct ingestion of soil can occur when sludge is applied to
sites where people may live and work, such as a family farm. This
analysis assumes that both the typical and MEI individual consume an
average of 0.1 grams/day of soil over a lifetime. It is assumed that all
ingested soil originates from outdoor sources. The values used for each
model input are summarized in Table 5-20. Average soil concentrations
are presented in Table 5-17.
(a) Methods for estimating soil and Indoor dust concentrations.
The concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in outdoor soil were
estimated as described in Section 5.2.4(1)(a).
(b) Estimates of human exposures and risks. Exposures and
risks were calculated as follows. The dally dose was calculated as:
(5-29)
DOSEg « [(C0)(DCg)] (ABg1J / BWg
5-60
1592H
-------
/M/e S-Sff. 4fsa^f/a>jy #rf ftrMffcr Kt/ies •
&// /agestfoff Wfmity
Input
parameter
Estimate for
Typical individual
Estinate for
ME1 individual
Notes/explanat i on/(reference)
ui
i
Ot
EXPOSURE
AMNint soil ingested (g/day) 0.1
Percent of daily ingest ion fro* outdoors 100X
Fraction of ingested soil that is fron 0.10
sludge-anended land
0.1
100X
1.0
-------
where:
ABgi = systemic absorption rate from gastrointestinal tract;
assumed to be 1
BWq = body weight of individual
Cg = concentration of contaminant in soil, mg/kg
DCg = daily soil ingestion rate
DOSEg = daily dose to individual (mg/kg/day)
Individual and population risks were calculated using Equations 5-12
and 5-13, respectively.
(c) Estimates of populations exposed. The population exposed
to 2,3,7,8-TCDD and 2,3,7,8-TCDF through ingestion was limited to the
population residing on the agricultural land application sites as
discussed in Section 5.4.2.(1)(d).
(4) Estimates of exposures and risks from Inhalation of sludge-
contaminated participates. 2,3,7,8-TCDD and 2,3,7,8-TCDF adhering to
soil particles can become suspended in the air near a land application
site. Transport downwind will dilute the concentration of particles from
a land application site; these particles will also re-deposit on
surfaces. Residents living on or near the land application sites may be
exposed to 2,3,7,8-TCDD or 2,3,7,8-TCDF by inhaling these particles.
This section describes the methods used to estimate the emissions of
particles from a land application site and the subsequent human exposure
to these emissions. This analysis only considered exposure to inhaled
particulates for residents onsite. The data inputs for the particle
inhalation model are presented in Table 5-21 except for respiration
rate. This parameter is discussed below in Section 5.2.4(4)(c). The
estimated soil concentrations are presented in Table 5-17.
The estimation of risks from inhalation of particles required several
steps. First, the emissions of particles from the treated area was
estimated. Next, the indoor and outdoor concentrations of particles
onsite were calculated. The concentrations were combined with information
about the length of time spent indoors and outdoors, respiratory rate,
and the slope factor of 2,3,7,8-TCDD and 2,3,7,8-TCDF to yield the
estimated cancer risks.
(a) Methods for estimating soil and Indoor dust concentrations.
The concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in outdoor soil were
estimated as described in Section 5.2.4(l)(a).
(b) Methods for estimating emissions and participate concentra-
tions. The method used for estimating emissions due to wind erosion
assumed that the uncrusted contaminated surface is exposed to the wind
and consists of finely divided particles. This created a condition that
5-62
1592H
-------
\nput
parameter
Estimate for
Typical individual
Estimate for
(€1 individual
Notes/explanation/(reference)
cn
i
-------
B897H
Table 5-21. (continued)
Input
parameter
Estimate for
Typical individual
Estimate for
MEI individual
Notes/explanation/(reference)
EXPOSURE
tn
i
o>
Hours per day and number of days spent
outdoors, child
Hours per day and number of days spent
outdoors, older child
Hours per day and number of days spent
indoors, adult, attic, living space
8 hrs/day
130 days
5 hrs/day
152 days
12 hrs/yr in
attic; 12 hrs/
day. 130 days
plus 24 hrs/day.
235 days.
farmer
12 hrs/day Typical: assumes child plays outdoors 8 hrs/day,
182 days May-October, 5 days per wk. ME1: assumes child plays
outdoors 12 hrs/day. Hay-October. 7 days per wk.
12 hrs/day Typical: assumes child plays outdoors an average
152 days of 5 hrs/day. Hay-September. HEI: assumes 12 hrs/day,
Hay-September (Hawley 1985).
12 hrs/yr in Attic: assumed that adult spends 12 hrs/yr (either 1 day
attic; 12 hrs/ for 12 hrs or 1 hr for 12 days) in attic. Living space:
day, 130 days 12 hrs/day in living space for 130 days in sinner months
plus 24 hrs/ plus 24 hrs/day in living space for rest of year (Hawley 1985).
day 235 days
Hours per day and number of days spent
indoors, child
16 hrs/day.
130 days, plus
24 hrs/day,
235 days
12 hrs/day. HEI: assimes child plays indoors 12 hrs per day, J days
182 days, plus per wk. during Hay-October, rest of year indoors 24 hrs
24 hrs/day, per day. Typical: assumes indoors IB hrs/day,
182 days 5 days/wk during Hay-October, indoors 24 hrs/day for re-
mainder of days.
Hours per day and number of days spent 16 hrs/day for
indoors, older child 365 days
12 hrs/day,
152 days plus
24 hrs/day for
213 days
Typical: assumes child is indoors 16 hrs/day. HEI: assumes
child is indoors an average 12 hrs/day from Hay-September, and
is indoors for the rest of the year 24 hrs/day (Hawley 1985).
COHCEHTRA1TOH
Indoor contaminant concentration as
function of outdoor contaminant con-
centration
indoor =
BOX outdoor
indoor =
100X outdoor
Typical: from Hawley (1985). HEI:
value possible.
estimate is max in
-------
results In maximum wind-caused dust emissions. The model for estimating
particulate concentration incorporated information on wind speed and
percent vegetation cover to estimate the flux of small particles (i.e.,
less that 10 /^m) from an area of land. Soil amended with paper mill
sludge may not have the characteristics assumed by the model; to the
extent that the surface of a sludge-amended site consists of coarser
particles, the model was likely to overestimate emissions.
To estimate particulate concentration, the calculated emission rate
Was used as input to a box model of atmospheric mixing. The box model
ignored any atmospheric dispersion downwind, and was only appropriate for
estimating onsite concentrations. The model used wind speed, size of the
site and the mixing height to yield an onsite particulate concentration.
/\s an alternative approach to estimating onsite particulate
concentration, another model, which used measured values of total
suspended particles adjusted by the fraction of particles assumed to be
derived from local soils, was applied.
The indoor suspended particle concentration was derived by applying
the ratio of suspended particulate concentration indoors to the suspended
particulate concentration outdoors. Since only a portion of indoor dust
Was assumed to originate from outdoor sources, the contaminant
concentration in indoor dust was adjusted by a fraction representing the
ratio of indoor dust contaminant concentration to the outdoor soil
contaminant concentration.
(c) Estimates of exposures and risks. Once the concentration
of contaminants in particulates was estimated, exposure to contaminated
particulates was calculated. In the MEI estimate of risk scenario, the
particulate concentration was estimated based on total particulates.
[Jnen calculating exposure from this estimate of particulate
concentration, the first step was to determine the concentration of
particles that were respirable. In the typical estimate scenarios, all
of the emissions were assumed to be respirable.
The next step in the estimation of human exposure to 2,3,7,8-TCDD and
2,3>7,8-TCDF through the inhalation of particulates was the estimation of
the daily dose. The daily dose was calculated for three age groups:
young children (ages 1-6), older children (ages 7-11), and adults (ages
12 and older).
(5-30)
POSE - I[(RC )(D )(AB )(H )] + [(RC )(D )(AB )(H )]] V
o,g o 1 1 o.g q qj qj o.q q
BWg
5-65
J59ZH
-------
where:
ABn = systemic absorption rate through the lung; assumed to be
1.0.
ABqi = systemic absorption rate through the gastrointestinal
tract; assumed to be 1.0.
BUq = body weight of individual in age group g
D-j = fraction of respired particles retained by the lung
Dqj = fraction of respired particles swallowed (fraction of
particles to gastrointestinal tract)
OOSE0 q = dose to individual in age group g, outdoors, mg/kg/day
HQ q = hours spent outdoors for individual in age group g
RC0 = respirable particulate concentration outdoors, mg/nr
Vq = weighted average ventilation rate for individual in age
group g, nr/day
(5-31)
DOSE -[[(Re ){D)(AB)(H )] + [(RC ) (D )(AB )(H )]] V
i,g _ in 1 1 L_q _ in qi ai i ,g _ a
BWg
where:
OOSEj q = dose to individual in age group g, indoors, mg/kg/day
= hours spent indoors for individual in age group g
= respirable particulate concentration indoors, mg/nr
In this equation, the concentration of the contaminant adhering to
particles was multiplied by the volume of air inhaled each day and by the
fraction of the day spent outdoors. Similarly, the quantity of
particulates inhaled indoors each day was the product of the indoor
respirable concentration, the volume of air inhaled each day, and the
fraction of the day spent indoors. The total quantity of particles
inhaled each day was then partitioned between the lung and the
gastrointestinal tract. A gastrointestinal absorption fraction was then
applied to the portion swallowed, while a respiratory absorption fraction
was applied to the portion remaining in the lung.
A weighted average dose for an individual over the entire lifetime
was derived by weighting the daily dose received during each age interval
by the fraction of the individual's lifespan spent in that age group,
This calculation is described in the following equation:
,g + DOSEitg)(Fg) (5-32)
5-66
159ZH
-------
where:
DOSEavg = weighted average daily dose over lifetime, mg/kg/day
DOSE0)g = dose to individual in age group g, outdoors, mg/kg/day
DOSE^'g = dose to individual in age group g, indoors, mg/kg/day
Fg ' = fraction of lifespan spent in age group g
Individual and population risks were calculated using Equations 5-12
and 5-13, reespectively.
(d) Data sources and model inputs for respiration rate.
Respiration rate was used in the model to assess the total daily volume
of particles inhaled. For adults, the average respiration rate was
calculated to be 23 m3 per day. This value was calculated using data
on the ventilation rates during different levels of activity, and the
amount of time spent per day engaging in these levels of activity, to
obtain a daily total. The ventilation rate of young children engaged in
light activity is 7.6 1/min, while the ventilation rate during rest is
2.8 1/min; assuming children spend roughly one-third of their day engaged
in light activity and two-thirds at rest, the total ventilation rate is
6.3 m3 per day. For older children, the ventilation rate is 11.6 1/min
during light activity and 4.3 1/min at rest, with a total ventilation
rate of 8.4 nr per day.
(e) Estimates of populations exposed. The population exposed
to 2,3,7,8-TCDD and 2,3,7,8-TCDF through inhalation of particulates was
limited to the population residing on the agricultural land application
sites as discussed in Section 5.4.2(l)(d).
(5) Estimates of exposures and risks from inhalation of vapors.
Residents of land application sites may incur risk from the inhalation of
volatilized 2,3,7,8-TCDD and 2,3,7,8-TCDF. Because actual locations of
the land application sites were not known, the ISCLT model was not used
to estimate concentrations downwind from such sites. As a result, this
analysis estimated only exposures to onsite residents, using an
atmospheric box model to obtain the onsite concentrations from the
estimated emissions.
The calculation of risks from the inhalation of vaporized
2,3,7,8-TCDD or 2,3,7,8-TCDF required the estimation of emissions, and
the calculation of indoor and outdoor onsite concentrations. The
concentrations were combined with data on time spent indoors and
outdoors, respiration rate and slope factor of 2,3,7,8-TCDD and
2,3,7,8-TCDF to obtain the estimated cancer risk from this pathway of
exposure. Table 5-21 summarizes key assumptions and input parameters for
estimating exposure through the vapor inhalation pathway.
(a) Methods for estimating vapor emissions. Information on
the partitioning of 2,3,7,8-TCDD and 2,3,7,8-TCDF between soil and air
5-67
1592H
-------
and between water and soil was used to estimate emissions of 2,3,7,8-TCDO
and 2,3,7,8-TCDF vapor per nr area. The emissions estimate was then
multiplied by the area of the site, in mz, to obtain the total
emissions of vapor from the land application site. The emission rate was
then coupled with a box model to obtain the onsite concentrations of
vapor.
(b) Methods for estimating vapor concentrations. The indoor vapor
concentration was derived by applying the ratio of vapor concentration
indoors to the vapor concentration outdoors. It was assumed that the
relationship between vapor concentrations indoors and outdoors was
similar to the ratio between indoor and outdoor particulate
concentrations (that is, indoor concentrations are approximately
75 percent of outdoor concentrations with a range from 70 to 85 percent).
(c) Estimates of exposures and risks. Once the concentration
of contaminant in the air was estimated, the calculation of exposure and
risks from the inhalation of vapor then proceeded in the same manner as
the exposure and risk from the inhalation of particles, described in
Section 5.2.4(4). In some cases, the data inputs used for the estimation
of exposure and risk were different than those used in Section 5.2.4(4).
The emissions model required the soil/water partition coefficient as
one input. This partition coefficient was, in turn, based on the
fraction of organic carbon (fc) in the soil. For land application
sites where sludge is soil-incorporated, it was assumed that the fraction
of organic carbon in the sludge-soil mixture was approximately equal to
the fraction of organic carbon in the soil alone. Estimates of fc for
soils of I percent and 4 percent were assumed for typical and MEI
estimates of exposure, respectively. The organic carbon content of
sludge has been reported to range from 14 to 40 percent. Estimates of
fr for sludge of 25 percent and 40 percent were assumed for typical and
Mtl estimates of exposure, respectively.
The data inputs and model sources for the vapor exposure estimate
were the same as those described in Section 5.2.4(4), with two
exceptions. The first exception is that 100 percent of vapor emissions
were assumed to be respired. The second exception is that all of the
vapor was absorbed through the lung; none was absorbed through the GI
tract. This analysis also assumed that all volatilized 2,3,7,8-TCDD and
2,3,7,8-TCDF was completely absorbed into the system when inhaled, due to
its high lipophllicity.
(d) Estimates of populations exposed. The population exposed
to 2,3,7,8-TCDD and 2,3,7,8-TCDF through inhalation of vapors was limited
to the population residing on the agricultural land sites as discussed in
Section 5.4.2(l)(d).
5-68
1592H
-------
(6) Estimates of exposures and risks from inaestion of around
watec. Land application of sludge is not expected to present
significant risk to human health through contamination of ground water.
Conservative, high risk estimates of ground-water contamination from
sludge in industrial landfills yielded risk estimates on the order of
10"7 for a most exposed individual; typical estimate assumptions
yielded lower risk estimates. Consideration of some major differences
between landfill ing and land application of pulp and paper sludge
suggests that risks of ground-water contamination and health risks from
land application would be lower than those estimated for landfills and
are therefore too low to justify more detailed evaluation.
Land application and landfill sites differ in at least four important
respects: (1) land application sites may be larger than landfills; (2)
local geo-hydrological or weather conditions may differ between landfill
and land application sites; (3) sludge may be placed in landfills to a
significant depth below ground level (land-applied sludge is generally
applied to the ground surface, or incorporated into a relatively shallow
surface soil layer); and (4) the quantities of sludge applied to a
hectare of treated land tend to be much lower than the quantities
deposited in a hectare of landfill.
(7) Estimates for exposures and risks from inoestion of drlnklno
gater from surface water sources. Uhere Pulp anH papt>r ci.^jn 1;
applied to land, particles of sludge or soil from the surface can be
transported by erosion to nearby lakes or streams. If humans consume
water or fish from these lakes or streams, they may be exposed to
2,3,7,8-TCDD and 2,3,7,8-TCDF from the land applied sludge.
The methodology to estimate exposure consisted of three general
steps: (1) based on sludge concentrations of 2,3,7,8-TCDD and
2,3,7,8-TCDF, local topography, land use and other factors, it estimated
contaminant concentrations in sediments, surface water, and fish; (2) it
used these estimated concentrations, assumptions about individual
ingestion of drinking water, and assumptions about the bioavailability
and the slope factor of 2,3,7,8-TCDD and 2,3,7,8-TCDF, to estimate
individual health risks for humans potentially exposed; and (3) it
combined these results with estimates of the size of exposed populations
to derive estimates of total human health risks in the U.S. population.
Table 5-22 summarizes the values used for each model input parameter.
The application rates and soil incorporation depths for the sites
receiving the sludge are displayed in Table 5-17, along with soil and
sludge concentrations.
(a) Methods for estimating surface water concentrations. The
Universal Soil Loss Equation was used along with estimates of sediment
delivery ratios, to estimate the fraction of a lake or stream's sediment
that originates from the landfill. By multiplying this fraction by the
5-69
1592H
-------
8897H
cn
««j
o
Table 5-Z2. Assumption and Parameter Values - Land Application: Surface Water Pathways
Input
parameter
COHCEHTttATlOH
"C" ratio
Agriculture
Forest/nine reclamation
-P" ratio
Agriculture
Forest/mine reclamation
Drainage area (ha)
Percent organic carbon in soil
Fish to sediment ratio - dioxin
Fish to sediment ratio - foran
Concentration in filet vs. concentra-
tion in whole fish
Sludge-amended area (hectares per yr)
Agriculture
Forest/mine reclamation
Distance to stream (meters)
Agriculture
Forest/nine reclamation
Estimate for
Typical individual
4
1
0.75
1
1.295.000
0.01
1
1
0.5
435
1.805
30
1.355
Estimate for
ME I individual Notes/explanation/ (reference)
USDA (1978)
6
1
1 Typical: estimate assumes contouring; HEI: assumes no support
1 pract ices .
4.047
0.001 USEPA (1986b)
10 USEPA (I968b)
10 USEPA (1988b)
0.5
405 From each state's permit data or best estimate.
1.012
30
46
EXPOSURE
Water consumption (liters/clay)
-------
\nput
parameter
Estimate for
Typical individual
Estimate for
MEI individual
Notes/explanation/(reference)
EXPOSURE (continued)
Fish consumption (grams/day)
Percent of fish ingested that is
contaminated
Percent of population served by sur-
face water
Percent of water ingested that is
contaminated
6.5
100X
49X
100X
140
100X
N/A
100X
U.S. average; subsistence fisher consumption.
ASSUMES that for the population in the drainage area,
100X of freshwater and estuarine fish consumption is from the
contaminated source.
US6S (1985)
Assumes that the affected population that is served by
surface water consumes 100X contaminated water.
-------
original concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF in sludge or soil
particles on the landfill surface, the methodology derived estimates of
the concentration of contaminants in the sediment. This contaminant load
was then partitioned between adsorbed and dissolved phases, based on the
assumption of equilibrium partitioning between the two phases.
(b) Estimates of exposures and risks. Exposures and risks
were calculated using Equations 5-5 and 5-6, respectively.
(c) Estimates of populations exposed. The population exposed
to contaminated water was estimated by multiplying the area of the
drainage basin above each sludge management area (SMA) drainage point by
the national population density (68 persons/sq. mile) (USDOC 1987). This
estimated population was then multiplied by the fraction of the
population that takes its drinking water from surface supplies:
PEW - (AB)(PD)(PSW) (5-33)
where:
PEW = Population exposed to contaminated water
AD = Area of the drainage basin (ha)
PD = Population density (persons/ha)
PSW - Percent of population served by surface water
Only a portion of this population will rely on surface water for
their drinking water. Therefore, the estimated population exposed was
reduced by multiplying by the average percentage of population served by
surface water. Forty-nine percent of the population was assumed to be
served by surface water.
(8) Estimates of exposures and risks from Inoestion of fish from
surface water sources. Where pulp and paper sludge is applied to land,
particles of sludge or soil from the surface can be transported by
erosion to nearby lakes or streams. If the sludge contains 2,3,7,8-TCDD
or 2,3,7,8-TCDF, then those particles can carry these contaminants to the
surface water bodies. Fish living in the lakes or streams can take up
sludge contaminants into their tissues; if humans then consume those
fish, they can be exposed.
The methodology used to estimate exposure and its associated risks
began by estimating sediment concentrations of 2,3,7,8-TCDD and
2,3,7,8-TCDF in water bodies as a result of runoff. The methodology also
used fish to sediment bioconcentration factors and estimates of fish
consumption contaminant doses. In addition, the methodology Involved
estimating the sizes of exposed populations, combining these results with
estimates of individual dose and health risk to derive total health risks
to the entire exposed population. The values used for each model input
are summarized in Table 5-22.
5-72
1S92H
-------
(a) Methods for estimating sediment concentrations. The
method for estimating concentrations used the Universal Soil Loss
Equation, together with estimates of sediment delivery ratios, to
estimate the fraction of a lake or stream's sediment that originates from
the landfill. Estimates of the concentration of contaminants in the
sediment were determined by multiplying this fraction by the original con-
centration of 2,3,7,8-TCDO and 2,3,7,8-TCDF in sludge or soil particles
on the surface of the landfill. These sediment concentrations were then
used to estimate contaminant concentrations in the tissues of fish.
(b) Methods for estimating fish tissue concentrations. The
methodology used empirical fish to sediment bioconcentration factors to
estimated concentrations of contaminant in freshwater fish as a function
of concentrations in stream or lake sediment. This was based on the
assumption that sediment concentrations were the best predictor of fish
concentrations of hydrophobic compounds like 2,3,7,8-TCDD and
2,3,7,8-TCDF. The concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in the
muscle tissues of fish (consumed by humans) were considered to be fifty
percent lower than the whole body concentrations of these contaminants.
(c) Estimates of exposures and risks. Exposures and risks
were calculated using Equations 5-7 and 5-8, respectively.
(d) Estimates of exposed populations. The population exposed
to fish containing 2,3,7,8-TCDD and 2,3,7,8-TCDF was estimated by
multiplying the area of the drainage basin containing each facility by an
estimated population density of the regions containing the SMA's. This
estimated population was then multiplied by the fraction of the
population that takes its drinking water from surface supplies:
PEW- (AB)(PD)(PSW) (5-34)
where:
PEW - Population exposed to contaminated water
AR - Area of the drainage basin (ha)
PSW - Percent of population served by surface water
PD • Population density
This analysis assumed that all fish were consumed regionally. An
alternative approach would be the assumption that the fish were distrib-
uted nationally. Therefore, the percent of the freshwater fish each
nerson consumed from the contaminated stream would be calculated by using
the ratio of the drainage area of the contaminated stream to the drainage
area of the entire United States. This percentage could be used as the
nercent of contaminated freshwater fish consumed by the entire U.S. popu-
lation. However, the current methodology used drainage area to determine
the proportion of U.S. citizens who were exposed to contaminated water
5-73
J592H
-------
and fish. This population was assumed to consume 100 percent of their
freshwater fish from the contaminated stream.
The receiving stream for each SMA fill was assumed to be a major
stream with a 5,000 square mile watershed area. Another assumption was
that the population exposed was positively correlated with stream size
and that stream size was positively correlated with drainage area i.e.,
the larger the drainage area, the more people were likely to receive
their drinking water supply from the stream. To quantify this
relationship, population exposed was modeled as a function of drainage
area. Each unit area of the watershed was multiplied by the average
population density for the regions through which the waterways flow to
yield population exposed.
The entire exposed population was assumed to ingest fish at
concentrations appropriate for the "point" of entry of the SMA runoff
into the stream. Since the population exposed inhabited an area of
approximately 70 by 70 miles, this assumption was conservative, and tends
to overstate exposure and risk. Dilution and dispersion of the
contaminant would have occurred before much of the population was exposed.
(9) Summary of results. The risks resulting from exposures
associated with the land applications of sludge are summarized in
Table 5-23. Typical risk to individuals are low in all pathways analyzed
for this disposal/re-use practice. Highest typical risks are estimated
for persons living on the land application site, exposed through direct
ingestion vapor inhalation, and dermal contact with contaminated soil.
The highest typical risk (i.e., 1 x 10"5) is associated with the vapor
inhalation pathway. However, because the size of the population exposed
through this pathway is small, the total annual cancer risk resulting
from this exposure is estimated to be only 7 x 10"5 cases per year.
The highest MEI risks are from the ingestion of fish caught in
contaminated surface water bodies, and from the ingestion of produce,
meat, and dairy products grown on sludge-amended land. The MEI is
assumed to take fish from a relatively small stream from the location in
the stream with the maximum dissolved concentrations of 2,3,7,8-TCDD and
2,3,7,8-TCDF. For the produce and meat ingestion pathway, the MEI is
assumed to be a subsistence farmer who raises his/her own meat and dairy
products on sludge-amended agricultural land. The MEI risk associated
with the fish ingestion pathway is 1 x 10"1, while the MEI risk from
the ingestion of produce and meat is 1 x 10~z.
5.2.5 Exposures and Risks from Distribution and Marketing of Pulp and
Paper Sludge
Sludge that is composted and marketed can be used as a soil amendment
1n residential settings as well as for agricultural and commercial
5-74
1S92H
-------
Table 5-23. Estimates of Health Risks to the General Population from Land Application of
Pulp and Paper Sludges Contaminated with 2,3.7,8-TCDO and 2,3,7,8-TCDF
Exposure pathway
rtnal exposure from contact with soil
contaminated by land application of sludge:
(Percent TCOD)C
Exposure from direct ingestion of soil
ontaminateci by land application of sludge:
(Percent TCDO)C
halation exposure to air contaminated by
latilization from soil contaminated by land
application of sludge:
(Percent TCDD)C
latiort exposure to particulates from soil
'"ontaminatert by land application of sludge:
(percent TCDD)C
stion exposure from drinking surface water
t minated by surface runoff from soil contaml-
C°ted by agricultural land application of sludge:
fl£ p
(percent TCOD)C
tion exposure from drinking surface water
In96 inated by surface runoff from soil con-
c°n. ateo- by land application of sludge to
(percent TCOD)C
•On exposure from drinking ground water
jnges te{) (jy leaching from soil contaminated
^"land application of sludge
Uji/ 1 O* ' ^
rent TCDO)C
(Percent
exposure to fish contaminated by surface
In9eSff°from areas contaminated by agricultural
run° Application of sludge
1a"cent TCDD)°
- - oKDOSure to fish contaminated by
ME!
risk3
(per
lifetime)
5xlO"5
(62)
5xlO~5
(62)
2xlO'4
(4)
5xlO~6
(62)
2xlO"3
(0.6)
3xlO"3
(0.6)
<3xlO'7
(0.2)
IxlO"1
(63)
2X10"1
Typical
risk3
(per Exposed
lifetime) population
3xlO"7 40
(65)
IxlO"6 40
(65)
IxlO'5 40
(4)
8xlO"7 40
(65)
3xlO'7 333,000
(0.6)
3xlO'7 833.000
(0.6)
<3xlO'7 NAd
(0.2)
5xlO'7 679.000
(65)
5xlO*7 1,700.000
Total
risk
(cases/
year)b
2xiO"7
(65)
6xlO"7
(65)
6xlO'6
(1)
5xlO"7
(65)
IxlO"3
(0.6)
4xlO"3
(0.6)
NAd
5xlO"3
(65)
IxlO"2
S0
rface
runoff from areas contaminated by
Heat ion of sTudge to mines/forests
(63)
(65)
(65)
5-75
9035H
-------
Table 5-23. (Continued)
Exposure pathway
ME I
riska
(per
lifetime)
Typical
risk6
(per
1 if et ime)
Exposed
populat ion
Total
risk
(cases/
year)15
Dietary exposure from produce grown in soil con- 1x10
laminated by land application of sludge
(Percent TCDO)C (62)
-2
2x10
(65)
-10
240,000,000 7x10
(65)
-4
3 Calculated as: [Estimated Dose] x 1.6 x 10"4 per (pg/kg)/day.
Calculated as: [Typical Risk x Exposed Population] / [Life Expectancy].
[Exposure to TCDD1
Calculated as: 100 x
Not applicable.
[Exposure to TCDD] + (1/10) [Exposure to TCDF]'
5-76
9035H
-------
purposes. According to the 104-Mill Study, seven mills in five states
distribute and market at least a portion of their sludge. Based on data
from the 104-Mill Study, the total volume of sludge distributed and
marketed by these plants is estimated to be 208,000 dry metric tons per
year. In some cases, the plants in this study reported two methods of
sludge disposal, but did not provide a break-down of the quantities of
sludge disposed by each method. In these cases, it was assumed the
entire quantity of sludge by the plant produced is distributed and
marketed. To the extent that this assumption overestimates the quantity
of sludge distributed and marketed every year, the population risk
estimates derived from this analysis overestimate the true population
risk.
This analysis estimates risks to members of households using
composted sludge from the following routes of exposure:
• Home gardeners incorporate distributed and marketed sludge into
home gardens. The home-grown crops incorporate small amounts of
contaminant into their tissues. Household residents then consume
the home-grown crops.
• Home gardeners incorporate distributed and marketed sludge into
their home gardens, or use it for other home gardening purposes,
such as lawns or flower beds. Children and adults in the gardening
household come into direct dermal contact with the sludge.
2,3,7,8-TCDD and 2,3,7,8-TCDF from the sludge is absorbed through
the skin. Children ingest small amounts of the sludge/soil mixture
through normal mouthing behavior. Adults also inadvertently ingest
small quantities of sludge/soil.
. 2,3,7,8-TCDD and 2,3,7,8-TCDF in distributed and marketed sludge
volatilizes from the sludge into the air. Residents of the
household inhale the volatilized 2,3,7,8-TCDD and 2,3,7,8-TCDF.
• Distributed and marketed sludge is applied to home gardens or
other home uses. Particles of the sludge/soil mixture become
suspended in the air. Members of the household inhale the
contaminated particles.
Since the actual users of distributed and marketed sludge are not
known, a generic scenario was used to estimate risks from the
Distribution and marketing of sludge. In this scenario, a household was
assumed to use composted sludge as a soil amendment for ornamental or
vegetable gardening. The following discussion briefly describes the
scenario considered. The estimated soil concentrations used in the
scenarios are shown in Table 5-24. Other parameters that describe the
generic scenario used in this assessment are presented in Table 5-25.
5-77
-------
8897H
Table 5-24. Distribution and Marketing Sludge and Soil Contaminant Concentration*
Estimate for Typical Individual: Estimate for ME I:
Soil concentration Soil concentration using
using Mean cone., and 90th percenti le cone.
Contaminant
TCOO
TCOF
Mean sludge
concentration
163
885
90th percent lie
sludge concentration
293
1760
assuring 6 inches (15 en)
depth of incorporation
10
55
assuming no
soil incorporation
293
1760
•Average concentrations over 70 years of exposure, assuming sludge is applied to land for 20 years.
00
-------
: ff/str/dut/aa jm/fhrketiaq
Input
parameter
tst\«wte for
typical individual
Estimate for
MEI individual
Notes/explanat ion/reference)
en
i
~-i
\o
Percent of sludge to home
garden vs. other uses
Soil incorporation assumptions
Garden size
(hectares)
Application rate
(DMT/year)
SOX
6 inches
0.016
10
100X
0 inches
0.022
20
Typical estimate: based on analysis of National
Garden Survey (1987). MCI: maximum value possible.
USDA (1979)
National Garden Survey (1987)
USOA (1979)
-------
According to recent data, 34 million of the 69 million households
involved in gardening activity in 1986 grew vegetables (National
Gardening Survey 1987). Based on these data, this analysis assumed that
approximately one-half of the distributed and marketed sludge was used at
households with vegetable gardens, while the other half was used at
households that use the sludge for ornamental gardening. Furthermore, the
analysis assumed that households using sludge for ornamental gardening
apply sludge at the same rate as households with vegetable gardens. In
the MEI scenario, it was assumed that all of the distributed and marketed
sludge was used at households with vegetable gardens.
The concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in sludge from
plants that distribute and market sludge were obtained from the 104-Mill
Study. The methods for calculating the soil concentration of 2,3,7,8-TCDD
and 2,3,7,8-TCDF in soils amended with the composted sludge were
described in Section 5.2.4. The soil concentration model requires inputs
for the initial 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations, length of
the application period and the depth of incorporation with background
soils. Decay of 2,3,7,8-TCDD and 2,3,7,8-TCDF during the composting
process was assumed to be negligible. The analysis assumed that
composted sludge is applied to a home garden for 20 years. The home
gardener was assumed to continue to use the garden for an additional
fifty years.
In the typical scenarios, the analysis assumed that sludge is soil
incorporated. For the MEI scenario, it was assumed that sludge is
applied only to the top layer of soil. Recommended depths of
incorporation for the maintenance of crops range from approximately 6 to
10 inches (USDA 1979). In the typical scenario, this analysis used 6
inches of incorporation. The top-dressing scenario was used to model
risks to the MEI.
This analysis assumed that the home gardener applies 10 dry metric
tons of sludge per hectare per year to his or her home garden. The
recommended application rates for home uses ranges from 5 to 20 dry
metric tons per hectare (USDA 1979). The MEI was assumed to apply 20
metric tons per year to his/her home garden.
The size of the garden assumed in the generic scenario affects the
individual risk estimates and influences the estimate of the size of the
exposed population. The average garden size for combined rural and urban
vegetable gardens is 0.016 hectares. For the MEI scenario, the average
rural garden size (0.022 hectares) was assumed.
Using the assumptions described above, this analysis estimated
exposures from dermal contact, vapor and particulate Inhalation, and
direct ingestion for young children (ages 1-6), older children
(ages 7-12) and for adults, while dietary exposures were estimated for
5-80
1593q
-------
young children (1-6) and adults. The average daily dose over a lifetime
from each of these pathways was the weighted average of the daily doses
during these stages of life. These doses were combined with cancer
potency estimates to obtain incremental lifetime risk from 2,3,7,8-TCDD
and 2,3,7,8-TCDF exposure. The following sections describe the methods
and data used to estimate risks from home uses of distributed and
marketed sludge through each these pathways. Results are summarized in
the final section.
(1) Estimates of exposures and risks from dermal contact with
skin. Humans coming in direct contact with sludge contaminated soils
^y~absorb 2,3,7,8-TCDD and 2,3,7,8-TCDF through their skin. The amount
of 2,3,7,8-TCDD and 2,3,7,8-TCDF absorbed will depend on the area of skin
exposed and on the length of time that the contaminated soil is in
contact with the skin. The following discussion summarizes the model
used to estimate dose through dermal contact with composted pulp and
paper mill sludges used in residential settings.
The model used to estimate dose through direct contact with soil
containing 2,3,7,8-TCDD and 2,3,7,8-TCDF used empirically derived
information on the amount of soil or dust that adheres to a square
centimeter of skin, the area of skin exposed in various settings and the
absorption rate of 2,3,7,8-TCDD or 2,3,7,8-TCDF through skin to derive
the dose of 2,3,7,8-TCDD or 2,3,7,8-TCDF from dermal contact with
contaminated soil or dust. The values used for each model input are
presented in Table 5-26.
The calculation of dermal exposure proceeded in two steps. First the
average daily exposure from dermal contact was calculated as the product
of the area of skin affected, the contact rate, the dermal absorption
rate and the duration of contact. Second, the risk from dermal contact
was calculated using the estimate of daily dose and the slope factor of
2,3,7,8-TCDD and 2,3,7,8-TCDF.
(a) Methods for estimating soil concentrations. The methods of
sludge application considered were (1) application as a top dressing, and
/2) incorporation into a layer of soil. Concentrations in the soil for
ail top dressing scenarios were assumed equal to concentrations in the
land-applied sludge. For sludge that was incorporated into the soil, the
concentrations were dependent on the depth of the sludge incorporated in
the soil, frequency of application, and initial sludge concentration.
The mass of the contaminant added was calculated by multiplying the
application rate by the contaminant concentration in the sludge. The
volume of soil with which the sludge was mixed was determined by
multiplying the incorporation depth by the Incorporation area. The
volume was then multiplied by the soil density to obtain the mass of the
soil with which the sludge was incorporated. Average soil concentrations
5-81
1593Q
-------
8897H
Table 5-26. Assumptions and Parameter Values - Distribution and Marketing: Dermal Pathway
en
i
00
Input
parameter
Contact rate
Contact rate
child
Contact rate
Contact rate
0
mg/cm . outdoor, child
mg/cm , outdoor, older
mg/cm . outdoor, adult
mg/cm2. indoor, child
Estimate for
typical individual
0.5
0.5
3.5
0.056
Estimate for
ME I individual
1.5
1.5
3.5
0.06
Notes/explanation/ (reference)
SchauM (1984).
Schaun (1984).
Based on adults doing yard work (Haw ley 1985) .
Typical: assuming dust fall indoors = 20% out-
Contact rate mg/cm . indoor, older
child
Contact rate mg/cm2. indoor, adult
living space
indoor, adult, attic
Ratio of contaminant concentration in-
doors to soil concentration outdoors
Tine spent outdoors, child (hr)
Time spent outdoors, older child (hr)
Time spent outdoors, adult (hr)
0.056
0.056
1.8 (110 mg/hr)
0.80
1.565
0.06
0.06
0.85
2.190
1.824
2.190
348
1.S65
doors, with cleaning every two weeks (Hawley 1985).
Typical: assuming dust fall indoors = 20X out-
doors, with cleaning every two weeks (Hawley 1985).
Contact rate for adults in living space same as for
P
children; for work, in attic, a value of 1.8 mg/cm
used for direct contact; for indirect contact with dusty
air, estimated contact is 110 mg for 1 hour of attic work
(Hawley 1985).
Hawley (1985).
Typical: estimate assumes 5 days/wk, 6 months/yr. 12
hrs/day (12 hrs before soil is washed off). ME I:
7 days/wk, 6 months/yr, 12 hrs/day soil contact (Hawley
1985).
Typical: child outdoors from May to September every day
(150 days). 12 hrs of soil contact before washing. ME]:
child outdoors every day for 6 months. 12 hrs soil
contact (Hawley 1985).
Typical: assimes 8 hrs of soil contact before
washing. 2 days/wk for 5 Months. WEI: assunes 12 hrs
moll contact before washing. 5 d»y»/xk. 6 mnthv/yr (13O
-------
Table S-2G. (continued)
Input
parameter
Estimate for
typical individual
Estimate for
ME I individual
Notes/explanation/(reference)
en
i
oo
Time spent indoors, child (hr)
Time spent indoors, older child (hr)
Time spent indoors, adult (hr)
living space
attic
Area of skin exposed outdoor, child
(cm2)
Area of skin exposed outdoor, older
child (cm2)
Area of skin exposed outdoor, adult
(cm2)
4.380
1.460
4,380
48
2.100 cm*
1.600 on*
1.700 car
6,570
4,380
6.570
72
2,800 cm'
3.200 aif
2.940 caT
Typical: 12 hrs/day of indoor dust contact all year.
MEI: assumes 24 hrs/day of indoor dust contact for
6 winter months, plus 12 hrs/day for 6 sunroer months
(Hawley 1985; Keenan 1989).
Typical: assumes 4 hrs/day of indoor dust contact all
year. MEI: assumes 12 hrs/day of indoor dust con-
tact all year (Hawley 1985).
Living space: Typical: assumes indoor dust
contact 12 hrs/day. all year; high assumes indoor dust
contact 24 hrs/day, 6 winter months, plus 12 hrs/day,
6 sunnier months. Attic: Typical: assumes 12 days with
soil on skin 4 hrs/day. MEI: assumes 12 days with soil
on skin 6 hrs/day (Hawley 1985).
Typical: estimate assumes both hands, legs, and feet
exposed during play. MEI: assumes both hands, arms.
legs, and feet exposed (Hawley 1985).
Typical: estimate assumes both hands, forearms, legs
from knees down exposed during play. MEI: assumes both
hands, legs, and feet exposed during play (Hawley 1985;
Keenan 1989).
Typical: assumes both hands, most of forearms exposed.
MEI: assumes adult is wearing shortsleeved shirt, with
an open neck, pants, shoes, no gloves or hat (Hawley
1985; Schaum 1984).
-------
889 7H
Table 5-26. (continued)
Input
parameter
Estimate for
typical individual
Estimate for
ME! individual
Notes/exp lanat i on/ ( reference )
Area of skin exposed indoor, child
(cm2)
Area of skin exposed indoor, older child
(cm2)
Area of skin exposed indoor, adult
(cm2)
500
2,800
CO Bioavailability through skin
Availability of dioxin from soil
matrix
400
1.700 01
(attic)
900 cm2
(living space)
0.012 h'
IX
3.200
2.940 cm2 in
both attic and
living space
(living space)
0.024 h"1
(children only)
I5X
Typical: estimate assumes area exposed equals one half
of surface area of child's feet, hands and forearms.
HEI: assumes area exposed indoors same as outdoors
(Haw ley 1985).
Typical: estimate assumes only hands are exposed.
MCI: assumes area exposed indoors same as outdoors.
Typical: estimate assumes hands exposed in living
space, while adult wears short-sleeved shirt, with an
open neck, pants, shoes, no gloves or hat, to work in
attic. HEI: assumes area exposed indoors same as
outdoors.
Typical: see Appendix A.I. HEI: assumed that skin of
children has twice the absorption of adults.
Recommended value: IX (see Appendix A.I).
-------
for a given year were estimated by adding the mass of contaminant applied
to the land for that specific year to the mass present from previous
applications and then dividing by the mass of the receiving soil plus the
mass of the applied sludge.
(b) Estimates of exposure and risks. The indoor dust and
outdoor soil contaminant concentrations were used to estimate human
exposure and risk from dermal contact with these media. Daily doses were
estimated for three age groups: young children (ages 1-6), older children
(ages 7-11) and adults (ages 12 and older). The dose for each age group
tfas calculated using Equation 5-10.
To obtain the weighted average dose over the lifetime of an
individual, Equation 5-11 was used. Individual and population risks were
calculated using Equations 5-12 and 5-13, respectively.
(c) Node! inputs for calculating size of the exposed
population. The size of the potentially exposed population was
calculated in the following manner. First, the total tons of sludge to
distribution and marketing from each plant engaged in this practice were
obtained from the 104-Mill Study. This analysis assumed that sludge was
applied at a rate of 10 DMT per hectare. Dividing tons by the
application rate yielded the number of acres covered by the distributed
and marketed sludge. Next, the size of the average garden was used to
determine the number of households using distributed and marketed
sludge. The average garden size for combined rural and urban vegetable
gardens was assumed to be 0.016 hectares. Dividing acres covered by
sludge by the number of acres per household gave the number of households
affected. Finally, the number of households was multiplied by the
average number of persons per household to obtain the total number of
persons affected by distributed and marketed sludge.
(2) Estimates of exposures and risks from Inoestlon of homo-grown
0£gduce. To model the risks associated with ingestion of home-grown
crops, a computer model, executed on an IBM personal computer, was
used. Risks through the dietary pathway were calculated by estimating
the contaminant concentration in homegrown crops, and then multiplying
this concentration by the daily consumption of home-grown vegetables.
This analysis assumed that only the residents of the household using the
composted sludge were exposed to sludge contaminants. Furthermore, the
analysis assumed that home gardeners do not produce meat or dairy
products with the distributed and marketed sludge. The total number of
households using distributed and marketed sludge was determined based on
the quantity of compost going to residential uses, the average
application rate in a residential setting, and size of the average
qarden; the number of persons potentially exposed was then derived by
jnultiplying the number of households by the average number of persons per
household.
5-85
J593Q
-------
For households with vegetable gardens, the calculations proceeded in
three steps. The first calculation estimated the sludge 2,3,7,8-TCDD and
2,3,7,8-TCDF concentrations in the tissues of crops grown in
sludge-amended home gardens. Next, individual risks were estimated based
on dietary ingestion of each crop. Finally, risks from all crops were
summed to estimate the total cancer risk from 2,3,7,8-TCDD and
2,3,7,8-TCDF through dietary exposures. The calculations for each of
these three components are described below.
(a) Determining tissue concentrations of contaminants for produce
grown in sludge-amended gardens.
where:
(Cs)(Ui)
(5-35)
CM
tissue concentration (dry weight) of 2,3,7,8-TCDD or
2,3,7,8-TCDF in crop i, ug/g dry
soil concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF,
adjusted for additional mass from added sludge (mg/kg)
rate of uptake of 2,3,7,8-TCDD or 2,3,7,8-TCDF into tissue
of crop i (ug/g dry weight per mg/kg in soil).
I (5-36)
where:
CW.j = tissue concentration (wet weight) of 2,3,7,8-TCDD or
2,3,7,8-TCDF in crop i, ug/g wet
KDWj - constant for converting dry weight concentration to fresh
weight concentration for crop i
The calculations for determining the soil concentration of 2,3,7,8-TCDD
or 2,3,7,8-TCDF are described in Section 5.2.4. Once the contaminant
soil concentration had been determined, each crop's uptake rate was
applied to contaminant amounts to estimate the concentration of
2,3,7,8-TCDD or 2,3,7,8-TCDF per unit of dry-weight of crop tissue.
Finally, dry weight tissue concentrations were converted to wet weight
concentrations.
(b) Determining exposures and risks from contaminant ingestlon
through foods grown in sludge-amended gardens. Exposures were
calculated using Equation 5-14. Daily doses were estimated for children
ages 1-6 and for individuals over age 7. Doses of 2,3,7,8-TCDD or
2,3,7,8-TCDF from each food were estimated by multiplying fresh weight
contaminant concentrations by the amount of that food crop consumed in
the average diet for each age group and by the fraction of the daily
5-86
1593q
-------
quantity consumed that came from the sludge-amended garden. The dose for
each food was combined with the dose of 2,3,7,8-TCDD or 2,3,7,8-TCDF from
other garden produce to yield a total dietary ingestion of 2,3,7,8-TCDD
and 2,3,7,8-TCDF.
The weighted average daily dose of contaminant over an individual's
lifetime was calculated as the sum of the daily doses for each age group
weighted by the fraction of the individual's lifespan spent as a member
of that age group, as described in the following calculation:
DOSEavg = I (Fg) (DOSEg) (5-37)
where:
DOSEaVg = average daily dose over lifetime, pg/kg/day
DOSEg = daily dose for individual in age group g, pg/kg/day
Fg - fraction of an individual's lifetime spent in age group g.
The calculation of individual and population risks were performed using
Equations 5-12 and 5-13, respectively.
The input parameters used to estimate dose through ingestion of
contaminated home-grown produce are summarized in Table 5-27.
(c) Model inputs for calculating size of the exposed
population. The size of the potentially exposed population was
calculated in the following manner. First, the total tons of sludge to
Distribution and marketing from each plant engaged in this practice were
obtained from the 104-Mill Study. This analysis assumed that sludge was
applied at a rate of 10 DMT per hectare. Dividing tons by the
application rate yielded the number of acres covered by the distributed
and marketed sludge. Next, the size of the average garden was used to
determine the number of households using distributed and marketed
sludge- The average garden size for combined rural and urban vegetable
gardens was assumed to be 0.016 hectares. Dividing acres covered by
sludge by the number of acres per household gave the number of households
affected. Finally, the number of households was multiplied by the
average number of persons per household to obtain the total number of
persons affected by distributed and marketed sludge.
(3) Estimates of exposures and risks from direct Inqestlon of
slujUfe' Direct ingestion of soil can occur when sludge is applied to
*^tes where people may live and work, such as a family farm. To model
the risks from the direct Ingestion of sludge contaminated with
2 3,7,8-TCDD and 2,3,7,8-TCDF, this analysis adapted a model which
accounted for differences in dose from exposure to indoor and outdoor
concentrations of soil contaminants. Children ingest far more soil on
average than adults; however, adults may also Inadvertently ingest soil
that adheres to food or cigarettes.
5-87
1593^
-------
8897H
Table 5-27. Assumptions and Parameter Values - Distribution and Marketing: Dietary Pathway
Input
parameter
Estimate for
typical individual
Estimate for
MEI individual
Notes/explanat ion/(reference)
Plant uptake rates
Fraction of vegetables fro* hoae
garden
Above-ground:
O.OZ;
root: 0.5
0.27 for all
vegetables but
legtMes; 0.15
potatoes; 0.07
dried legwes.
Above-ground:
0.02;
root: 0.5
0.6 for all
vegetables but
legwes; 0.45
potatoes; 0.17
dried leguaes.
See Appendix A.4.
Typical: estimate values represent rural, nonfarm
households; HE1: represent rural, farm households
{USEPA 1988e).
Adult and child consumption rates
in
i
Values based
on TAS Dietary
Data Base
Values based
on TAS Dietary
Data Base
-------
The calculation of risks from direct ingestion of sludge was
straightforward. First, the soil concentrations outdoors and the dust
concentration indoors were estimated. The outdoor contaminant
concentration was multiplied by the quantity of dirt consumed outdoors,
wnile the indoor contaminant concentration was multiplied by the quantity
of indoor dust ingested daily. Risk was estimated based on the daily
quantity of soil and dust ingested, the gastrointestinal absorption of
2 3,7,8-TCDD and 2,3,7,8-TCDF from soil, and the potency of 2,3,7,8-TCDD
and 2,3,7,8-TCDF.
(a) Methods for estimating soil concentrations. The
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in outdoor soil were
estimated as described in Section 5.2.4.
(b) Exposure was estimated using Equation 5-29. Individual
and population risks were calculated using Equations 5-12 and 5-13,
respectively. The values used for each model input are summarized in
Table 5-28.
(c) Model inputs for calculating size of the exposed
population. The size of the potentially exposed population was
estimated in the following manner. First, the total tons of sludge to
distribution and marketing from each plant engaged in this practice were
obtained from the 104-Mill Study. This analysis assumes that sludge was
applied at a rate of 10 DMT per hectare. Dividing tons by the
application rate yielded the number of acres covered by the distributed
and marketed sludge. Next, the size of the average garden was used to
determine the number of households using distributed and marketed
sludge. The average garden size for combined rural and urban vegetable
Gardens was assumed to be 0.016 hectares. Dividing acres covered by
sludge by the number of acres per household gave the number of households
affected. Finally, the number of households was multiplied by the
average number of persons per household to obtain the total number of
persons affected by distributed and marketed sludge.
(4) Estimates of exposures and risks to the general population from
i
of sludge-contaminated participates. 2,3,7,8-TCDD and
23,7»8-TCDF adhering to soil particles can become suspended in the air
near a site treated with sludge. Transport downwind will dilute the
concentration of particles from a treated area; these particles will also
redeposit on surfaces. Residents using composted pulp and paper mill
sludge on their home gardens may be exposed to 2,3,7,8-TCDD or
2 3,7,8-TCDF by inhaling these particles. This section describes the
methods used to estimate the emissions of particles from a treated site
and the subsequent human exposure to these emissions. This analysis only
considered exposure to inhaled particulates for residents onsite.
5-89
-------
01
I
10
o
8897H
Table 5-28. Assumptions and Parameter Values * Distribution and Marketing: Soil Ingest ion Pathway
Input
parameter
Estimate for
typical individual
Estimate for
ME I individual
Notes/explanat ion/ ( reference )
Amount soil
ingested, g/day.
0.1
0.1
Single value used to represent average daily consumption
over lifetime. (OSWFR Directive 9850.4 "Interim Final
Guidance for Soil Ingest ion Rates," J. Winston Porter,
January 27. 1989).
Percent of daily ingest Ion fro*
outdoors, child
100X
100X
Analysis assumes 100X of soil from outdoor sources.
Amount of soil ingested fro* sludge-
amended land
0.10
1.0
Typical value of 10X Mas used in evaluation of reuse of
municipal sewage sludge. ME1 values arbitrary.
-------
To estimate the suspended participate concentration at treated sites,
a model for emissions from wind erosion was chosen since the analysis
focuses on average exposures over the long-term. The assumptions
underlying the model are described as follows: "This method assumes that
the uncrusted contaminated surface is exposed to the wind and consists of
finely divided particles. This creates a condition defined ... as an
"unlimited reservoir" and results in maximum wind-caused dust emissions."
The model incorporated information on wind speed and percent vegetation
cover to estimate the flux of small particles (i.e., less than 10 pm)
from an area of land. Soil amended with paper mill sludge may not have
the characteristics assumed by the model; to the extent that the surface
Of a sludge-amended site consists of coarser particles, the model was
likely to overestimate emissions.
To obtain particulate concentration, the calculated emission rate was
used as input to a box model of atmospheric mixing. The box model
ignored any atmospheric dispersion downwind, and was only appropriate for
estimating onsite concentrations. The model used wind speed, size of the
site and the mixing height to yield an onsite particulate concentration.
As an alternative approach to estimating onsite particulate
concentration, an additional model was applied that uses measured values
Of total suspended particles adjusted by the fraction of particles
assumed to be derived from local soils.
The calculation of risks from inhalation of particles required
several steps. First, the emissions of particles from the treated area
were estimated. Next, the indoor and outdoor concentrations of particles
onsite were calculated. The concentrations were combined with
information about the length of time spent indoors and outdoors,
respiratory rate, and the slope factor of 2,3,7,8-TCDD and 2,3,7,8-TCDF
to yield the estimated cancer risks.
(a) Method used to estimate the concentration of 2,3,7,8-TCDD
and 2,3,7,8-TCDF in participates. The first step in the calculation of
concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF in particulates was to
estimate the emissions of particulates from the treated areas. The
method used estimated the flux of dust particles from the surface as a
function of (1) the vegetative covering of the surface, and (2) the cube
Of the ratio of the windspeed to the threshold wind velocity (the
velocity required to initiate erosion). The dust flux was then converted
to a contaminant emission rate.
The next step was to estimate the concentrations of particulates on
tne land-treatment site. Both outdoor concentrations and indoor
concentrations were calculated. The outdoor concentrations were derived
by dividing the emissions by the product of the length of one side of the
treated area by mixing height and by windspeed.
5-91
15930
-------
An alternative method of calculating outdoor contaminant particulate
concentrations was to adjust the measured total suspended particulate
concentration at the site by the fraction believed to originate from
local (contaminated) soils.
The indoor suspended particle concentration was derived by applying
the ratio of suspended particulate concentration indoors to the suspended
particulate concentration outdoors. Next, since only a portion of indoor
dust was assumed to originate from outdoor sources (the rest is derived
from smoking, cooking, etc.) the contaminant concentration in indoor
dust was adjusted by a fraction representing the ratio of indoor dust
contaminant concentration to the outdoor soil contaminant concentration.
(b) Estimates of exposures and risks. Once the concentration
of contaminants in particulates was estimated, human exposure to
contaminated particulates can be estimated. In the HE I scenario, the
particulate concentration was estimated based on total particulates.
When calculating exposure from this estimate of particulate
concentration, the first step was to determine the concentration of
particles that are respirable. The respirable concentration was
estimated as:
RC0 - (C0)(FR)
and
where:
(5-38)
(5-39)
C0 - concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in suspended
particles outdoors, mg/m3
PR - fraction of suspended particles that are respirable
RC0 - respirable particulate concentration outdoors, mg/m3
Cin - concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in suspended
particles indoors, mg/m3
RCjn - respirable particulate concentration indoors, mg/m3.
In the typical scenarios, the estimation of contaminant emissions
adhering to respirable particles assumed that all of the emissions were
respirable.
The next step in the calculation of human exposure to 2,3,7,8-TCDD
and 2,3,7,8-TCDF through the inhalation of particulates was the
estimation of the daily dose. The daily dose was calculated for three
age groups: young children (ages 1-6), older children (ages 7-11), and
adults (ages 12 and older) using Equation 5-30.
1593q
5-92
-------
A weighted average dose for an individual over the entire lifetime
was derived by weighting the daily dose received during each age interval
by the fraction of the individual's lifespan spent in that age group.
This calculation is described in Equation 5-32. Individual and
population risks were calculated using Equations 5-12 and 5-13,
respectively.
The data inputs for the particle inhalation model are presented in
Table 5-29.
(c) Model inputs for calculating size of the exposed
population. The size of the potentially exposed population was
estimated in the following manner. First, the total tons of sludge to
distribution and marketing from each plant engaged in this practice were
obtained from the 104-Mill Study. This analysis assumes that sludge was
applied at a rate of 10 DMT per hectare. Dividing tons by the
application rate yielded the number of acres covered by the distributed
and marketed sludge. Next, the size of the average garden was used to
determine the number of households using distributed and marketed
sludge. The average garden size for combined rural and urban vegetable
gardens was assumed to be 0.016 hectares. Dividing acres covered by
sludge by the number of acres per household gave the number of households
affected. Finally, the number of households was multiplied by the
average number of persons per household to obtain the total number of
persons affected by distributed and marketed sludge.
(5) Estimates of exposures and risks from Inhalation of vapors.
Residents using distributed and marketed sludge may incur risk from the
inhalation of volatilized 2,3,7,8-TCDD and 2,3,7,8-TCDF. Because actual
locations of the homes using composted sludge are not known, the ISCLT
model could not be used to estimate downwind concentrations. As a
result, this analysis estimated only exposures to onsite residents, using
a box model to obtain the onsite concentrations from the emissions
estimates.
The calculation of risks from the inhalation of vaporized
2,3,7,8-TCDD and 2,3,7,8-TCDF required first the estimation of emissions,
then the calculation of indoor and outdoor onsite concentrations. The
concentrations were combined with data on time spent indoors and
outdoors, respiration rate and potency of 2,3,7,8-TCDD and 2,3,7,8-TCDF
to obtain the estimated cancer risk from this pathway of exposure.
(a) Methods for estimating vapor emissions. This analysis
assumed that emissions from home gardens were a function of the
partitioning of 2,3,7,8-TCDD and 2,3,7,8-TCDF between soil and air and
between water and soil. The emissions estimate was then multiplied by
the area of the home garden, in nr, to obtain the total emissions of
vapor from the residential site. The emission rate was then coupled with
a box model to obtain the onsite concentrations of vapor.
5-93
-------
8897H
Table 5-29. Assertions and Parameter Values - Distribution and Marketing: Vapor and Particulate Inhalation Pathways
Input
parameter
Estimate for Estimate for
Typical individual MEI individual Notes/exp1anation/(reference)
CONCENTRATION
Method for calculating participate
emissions
Indoor TSP concentration as function
of outdoor TSP concentration
Assume
SOX vegetative
cover
0.75
Assume SOX of Typical: calculated assuming SOX vegetation.
total TSP from MEI: assume SOX of total TSP from local soil; local TSP =
local soil 64 ug/m3 (Hawley 1985).
0.85 Hawley (1985).
in
i
EXPOSURE
Percent TSP respirable
Fraction of TSP inhaled to lung
Absorption rate, lung
N/A
1/3 of
respired TSP
Fraction of TSP inhaled to SI tract 2/3 of
respired TSP
1.0
Hours per day and number of days spent 8 hrs/day
outdoors, adult 40 days
0.90
0.90 (1001 of
respired TSP)
OX
1.0
12 hrs/day
130 days
Typical: estimate for emissions is from information
that gives estimates of small (<10 urn) particle emissions.
All of these particles are assumed to be respired. The range
given here applies only to TSP. which consists of large and
small particles (i.e., this range applies only to the MEI
estimate of emissions).
Typical: one-third of total TSP is initially found in lungs;
MEI: 100X (SchauM 1984).
Value from bioava liability is applied to the particles that
reach the alveoli.
Typical: assumes adult works 2 days per wk.
5 days/**. 12 hrs/day (Hawley 1985).
HEI: 6 months,
-------
Table 5-23. (continued)
en
i
to
en
Input
parameter
EXPOSURE
Hours per day and number of days spent
outdoors, child
Hours per day and number of days spent
outdoors, older child
Hours per day and number of days spent
indoors, adult, attic, living space
Hours per day and nunber of days spent
indoors, child
Hours per day and nunber of days spent
indoors, older child
COHCEHTRAITON
Indoor contaminant concentration as
function of outdoor contaminant con-
Estimate for
Typical individual
8 hrs/day
130 days
5 hrs/day
152 days
12 hrs/yr in
attic; 12 hrs/
day, 130 days
plus 24 hrs/day.
235 days.
farmer
16 hrs/day.
130 days, plus
24 hrs/day.
235 days
16 hrs/day for
365 days
indoor =
BOX outdoor
Estimate for
MEI individual
12 hrs/day
182 days
12 hrs/day
152 days
12 hrs/yr in
attic; 12 hrs/
day, 130 days
plus 24 hrs/
day 235 days
12 hrs/day,
182 days, plus
24 hrs/day.
182 days
12 hrs/day.
152 days plus
24 hrs/day for
213 days
indoor =
100X outdoor
Notes/explanation/(reference)
Typical: assumes child plays outdoors 8 hrs/day,
Hay-October, 5 days per wk. HEI: assumes child plays
outdoors 12 hrs/day, Hay-October. 7 days per wk.
Typical: assumes child plays outdoors an average
of 5 hrs/day. Hay-September. HEI: assumes 12 hrs/day.
Hay-September (Haw ley 1985).
Attic: assumed that adult spends 12 hrs/yr (either 1 day
for 12 hrs or 1 hr for 12 days) in attic. Living space:
12 hrs/day in living space for 130 days in summer months
plus 24 hrs/day in living space for rest of year (Haw ley
MEI: assumes child plays indoors 12 hrs per day. 7 days
per wk during May-October, rest of year indoors 24 hrs
per day. Typical: assumes indoors 16 hrs/day.
5 days/wk during May-October, indoors 24 hrs/day for re-
mainder of days.
1985).
Typical: assumes child is indoors 16 hrs/day. MEI: assumes
child is indoors an average 12 hrs/day from Hay-September, and
is indoors for the rest of the year 24 hrs/day (Haw ley 1985).
Typical: from Hawley (1985) . MEI: estimate is maximum
value possible.
centration
-------
The indoor vapor concentration was derived by applying the ratio of
vapor concentration indoors to the vapor concentration outdoors. It was
assumed that the relationship between vapor concentrations indoors and
outdoors was similar to the ratio between indoor and outdoor particulate
concentrations (that is, indoor concentrations are approximately
75 percent of outdoor concentrations with a range from 70 to 85 percent).
(b) Estimates of exposure and risk. Once the concentration of
contaminant in the air was estimated, the calculation of exposure and
risks from the inhalation of vapor then proceeds in the same manner as
the exposure and risk from the inhalation of particles, described in
Section 5.2.5(4). In some cases, however, the data inputs used for the
estimation of exposure and risk were different than those used in Section
5.2.5(4).
(6) Summary of results. Health risks from the distribution and
marketing of paper mill sludge containing 2,3,7,8-TCDD and 2,3,7,8-TCDF
are summarized in Table 5-30. In general, risks for the "maximum exposed
individual" (MEI) were two to three orders of magnitude higher than the
risks for a typical individual. Estimated risks for the MEI were lowest
for the particle inhalation and dietary pathways, and were highest for
the pathways involving direct human contact with contaminated soil (i.e.,
the dermal and direct ingestion pathways).
5.3 Analysis of Uncertainty
Total risk estimates for ingestion of contaminated ground water were
prepared for the disposal of sludge in landfills, disposal of sludge in
surface impoundments, and the land application of sludge. For landfills
and surface impoundments, risk estimates varied by one to two orders of
magnitude. Results varied significantly between calculations involving
the SESOIL model and those based on simpler, but less data intensive,
assumptions about the equilibrium partitioning of contaminants between
dissolved and adsorbed phase. Low estimates of risk were not prepared,
but estimates of zero risk through this pathway could be easily derived
if landfills are assumed to be located in areas without productive
aquifers.
Total risks from inhalation of particulates were estimated for
persons living on a land application site, or applying pulp and paper
sludge to home gardens. The estimated range of uncertainty was
relatively low for these estimates (one to two orders of magnitude), and
reflected differences In results based on different mathematical models
for estimating particulate suspension. Other key variables were the
ratio of indoor to outdoor concentrations, and assumed absorption rates
in the lungs and gastrointestinal tract.
5-96
1593q
-------
Table 5-3C. Estimates of Health Risks to the General Population from Distribution and Marketing
of Pulp and Paper Sludges Contaminated with 2.3.7,8-TCDD and 2.3,7.8-TCDF
— -"^
Exp0sure pathway
i exposure from contact with soil contaminated
0ertt1Histributed and marketed sludge
/percent TCDO)C
jre from direct ingestion of soil contaminated
E*P . istributed and marketed sludge
percent TCDD)C
tlon exposure from air contaminated by
In filiation from soil contaminated by
V° * ibuted and marketed sludge
d'st Trnn>c
rcent TCDD'
tlon exposure from partlculates from soil
I"*13 -itnated by distributed and marketed sludge
c°pt .. rrnnl0
rcent TCDOI
exposure from produce grown in gardens
D^e --mated by distributed and marketed sludge
'C°rc*nt TCDD)C
ME I
riska
(per
lifetime)
IxlO'4
(63)
IxlO"4
(57)
6xlO"7
(5)
2xlO"7
(64)
ZxlO"8
(62)
Typical Total
risk3 risk
(per Exposed (cases/
lifetime) population year)
3xlO"8 3,500,000 2xlO"3
(59) (59)
3xlO"7 3,500.000 2xlO"2
(56) (56)
SxlO"8 3.500,000 3xlO*3
(6) (6)
5xlO"9 3,500,000 3xlO"4
(71) (71)
5xlO"J1 3,500,000 3xlO"6
(H) (71}
.lated as: [Estimated Dose] x 1.6x10-* per (pg/kg)/day.
a £a icu
ulated as: [Typical Risk x Exposed Population] / [Life Expectancy].
b C«1C fTCDD Dosel
l1culated as:
C»]
[1/10)[TCDF Dose]
5-97
-------
Risk estimates for inhalation of volatilized 2,3,7,8-TCDD and
2,3,7,8-TCDF were also relatively stable with respect to the assumptions
tested for land application and distribution and marketing scenarios, and
represented variation in assumptions about time spent indoors and
outdoors, ratios between indoor and outdoor concentrations, and the
fraction of organic carbon in the soil. For landfills, the wider range
of estimates for exposure and risks from volatilized contaminants
reflected the testing of two different models for estimating volatile
emissions from soil. Both typical estimates of risk and MEI estimates of
risk from emissions from surface impoundments were based on a two film
resistance model; differences in risk estimates were therefore
attributable to differences in selected parameter values.
Precise estimation of total risks from ingestion of surface water
contaminated by runoff from sludge re-use or disposal would require
extensive, site-specific data for each sludge management site. Without
detailed data about the location of each site, distances to surface
water, site and surrounding topography, hydrology of nearby surface
water, locations (if any) of withdrawal of surface water for drinking
water sources, and numerous other key data, one must rely on simple
screening models to derive rough estimates of potential risk. The wide
range of risk estimates reported for this exposure pathway highlights the
fact that without such site-specific data, precise quantification was
impossible. The possibility of significant risks through this pathway
cannot be ruled out on the basis of existing data. Risk estimates are
sensitive to the size of the drainage area assumed for the stream
receiving runoff from each site.
Similarly, precise estimation of total risk from fish ingestion was
impossible without detailed, site-specific information. Depending on the
assumptions chosen for modeling exposure and risks through this pathway,
resulting estimates can vary by four orders of magnitude or more. As
with risks from the ingestion of drinking water from surface water
sources contaminated by surface runoff, significant risks from ingestion
of fish cannot be dismissed on the basis of this analysis.
Total risks from dietary consumption of vegetables, meat or dairy
products grown from sludge amended land (or from feeds grown on
sludge-amended land) varied by about two orders of magnitude for land
application sites, depending on the assumptions chosen. Key assumptions
included the percentage of animal diet consisting of sludge, and
bioconcentration factors. For distribution and marketing, dietary risk
estimates varied by four orders of magnitude, depending on which set of
assumptions was used. Results were greatly influenced by whether we
assume soil incorporation, and by the fraction of each household's diet
assumed to be homegrown.
5-98
1593q
-------
Estimated total risks from dermal absorption of 2,3,7,8-TCDD and
2,3,7,8-TCDF in sludge or soil were also subject to the selection of
parameter values. This range was attributable to numerous differences in
the two scenarios, including whether or not the sludge was assumed to be
soil incorporated, the area of skin exposed, contact rates and absorption
rates for skin, and numerous other factors. Estimated total risks from
direct ingestion of soil varied by two to three orders of magnitude,
depending on whether the sludge was assumed to be soil incorporated,
assumed rates of soil ingestion for children and adults, and the fraction
of daily soil ingestion originating from the treated area.
Considerable uncertainty is implicit in the risk estimates derived in
this chapter. In the absence of detailed site-specific data for the
numerous sites at which sludge is re-used or disposed, such broad ranges
of uncertainty in exposure and risk estimates are unavoidable.
5.4 Conclusions
Table 5-31 compares human health risk estimates for each waste
disposal method analyzed. For each waste management practice, it reports
results for the exposure pathway found to result in highest risks to the
"most exposed individual" and to the total population. The possibility
Of simultaneous exposure through multiple pathways was ignored. By
comparing maximum estimated risks for each management practice,
fable 5-31 allows a simple comparison of risks associated with each
practice.
As can be seen from the table, estimated risks to the typical exposed
individuals were typically three to six orders of magnitude lower than
risks to the MEI, except for the landfill ing of paper wastes, for which
separate population risk estimates were not performed. As shown by the
table* none of the waste disposal methods and exposure pathways analyzed
v*as expected to result in a total population cancer risk of more than one
expected incremental cancer case per twenty years of sludge or paper
disposal.
In summary, this analysis found that typical risks to human popula-
tions exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF from pulp and paper mill
sludge are generally low. Risks to most exposed individuals, however,
could be significant, depending on site-specific circumstances and
individual human behavior; it is not certain that the hypothetical
circumstances depicted in estimating these risks actually occur. As
discussed in Section 5.5, the best estimates of human exposure and risk
provided in this report should not be interpreted as precise
quantification of exposure and risks, but rather as general indicators of
the magnitude of those risks.
5-99
-------
Table 5-31. Maximum General Population Health Risks from Exposure to TCDD and TCDF, by Uaste Management Method
Exposure pathway
Potential human exposure from disposal of pulp
and paper sludge 1n landfills
(Percent TCOD)f
Potential human exposure from disposal of paper
wastes 1n landfills
(Percent TCDD)f
Potential human exposure from disposal of pulp
and paper sludge in surface impoundments
(Percent TCDD)
Human exposure from land application of pulp
and paper sludge
(Percent TCDO)f
Human exposure from distribution and marketing
of pulp and paper sludge
(Percent TCDD)f
ME I
risk3'"
(per
lifetime)
5xlO"2
(63)
<3xlO~7
(17)
IxlO"1
(63)
ZxlO"1
(63)
IxlO'4
(63)
Typical
Hska'c
(per Exposed
lifetime) population
8xlO"8 14,200,000
(65)
<3xlO~7 Not
estimated
(17)
IxlO"7 4,760.000
(65)
IxlO*5 1,700,000
(4)
3xlO"7 3,500,000
(56)
Total
risk
(cases/
year)6
ZxlO"2
(65)
Not
estimated
7xlO"3
(65)
IxlO"2
(65)
IxlO"3
(56)
a -4
Calculated as: (Estimated Exposure) x 1.6x10 per (pg/kg)/day.
Represents highest risk to most exposed individual through any single exposure pathway associated with this
waste disposal or re-use method. All risk estimates based on best estimate Input parameters.
c Represents risk to typical exposed individual for single exposure pathway associated with highest total
risk for this waste disposal or re-use method. All risk estimates based on best estimate input parameters.
Total exposed population for single exposure pathway associated with highest typical risks for this sludge
re-use or disposal method.
e Calculated as: (Average Risk x Exposed Population) / (Life Expectancy).
(Exposure to TCDD)
Calculated as: 100 x
(Exposure to TCDD) + (1/10) (Exposure to TCDF)
903 5H
5-100
-------
5.5 References
ADL. Arthur D. Little, Inc. 1987. Exposure and risk assessment of TCDD
in bleached kraft paper products. U.S. EPA, Office of Water Regulations
and Standards. Contract No. 68-01-6951.
Bonazountas M, Wagner JM. 1984. SESOIL. A seasonal soil compartment
model. U.S. EPA, Office of Toxic Substances. Contract No. 68-01-5271.
Bowers JF, et al . 1980. Industrial source complex (ISC) dispersion
model user's guide. U.S. EPA, Research Triangle Park, NC. PB 80-133044.
pelta Western Feed Mills. 1989. Personal communication with Abt
Associates. July 1989.
Eduljee G. 1987. Volatility of TCDD and PCBs from soil. Chemosphere.
16(4): 907-920.
freeze RA, Cherry JA. 1979. Groundwater. Prentice Hall, Inc.,
Englewood Cliffs, NJ.
Fries G. 1982. Potential polychlorinated biphenyl residues in animal
products from application of contaminated sewage sludge to land. J. of
Environ. Qua!. 11(1).
Hawley JK. 1985. Assessment of health risk from exposure to
contaminated soil. Risk Analysis. 5(4): 289-302.
iCeenan RE, Sauer M, Lawrence F, Rand E, Crawford C. 1989. Examination
Of potential risks from exposure to dioxin in sludge used to reclaim
abandoned strip mines. In: The Risk Assessment of Environmental and
Human Health Hazards: A Textbook on Case Studies. D..J. Paustenback,
ed. J. Wiley and Sons, New York. pp. 935-998
Kimbrough R, Falk H, Stehr P. 1984. Health implications of 2,3,7,8-TCDD
contamination of residential oil. J. Tox. Enviro. Health 14:47-93.
Lyman WJ, Reehl WF, Rosenblatt DH. 1982. Handbook of chemical property
estimation methods: environmental behavior of organic compounds.
McGraw-Hill Book Company, New York.
NCASI- 1984. National Council of the Paper Industry for Air and Stream
improvement. The lard application and related utilization of pulp and
paper mill sludges. New York, New York. Technical Bulletin No. 439.
1987. National Council of the Paper Industry for Air and Stream
Improvement, Inc. Assessment of human health risks related to exposure
to dioxin from land application of wastewater sludge 1n Maine. New York,
York. Technical Bulletin No. 525.
5-101
l593q
-------
National Gardening Survey Assoc., Inc. 1987. National Gardening survey.
Schaum J. 1984. Risk analysis of TCDD contaminated soils. U.S. EPA,
Office of Health and Environmental Assessment. EPA 600/84-4-031.
USDA. 1978. U.S. Department of Agriculture. Predicting rainfall
erosion losses.
USDA. 1979. U.S. Department of Agriculture. Use of sewage sludge
compost for soil improvement and plant growth. ARM-NE-G.
USDA. 1987. U.S. Department of Agriculture. Agricultural statistics -
1987. United States Government Printing Office. Washington, D.C.
USDOC. 1987. U.S. Department of Commerce. Bureau of the Census,
Statistical Abstract of the United States. 1987.
USEPA. 1985. Environmental Protection Agency. Summary of data on
industrial nonhazardous waste disposal practices. Office of Solid Waste
and Emergency Response.
USEPA. 1987a. U.S. Environmental Protection Agency. Development of
risk assessment methodology for the land application and distribution and
marketing of municipal sludge. Final draft. Cincinnati, OH: Office of
Health and Environmental Assessment, Evaluation and Criteria Assessment
Office.
USEPA. 1987b. U.S. Environmental Protection Agency. Comparison of food
consumption data—Tolerance Assessment Program. Washington, D.C.:
Office of Pesticide Programs.
USEPA. 1988b. U.S. Environmental Protection Agency. Estimating
exposures to 2,3,7,8-TCDD. External review draft. Office of Health and
Environmental Assessment.
USEPA. 1988c. U.S. Environmental Protection Agency. Report to
Congress: Solid waste disposal in the United States. Volume 1.
USEPA. 1988d. Environmental Protection Agency. Development of risk
assessment methodology for land application and distribution and
marketing of municipal sludge. Office of Research and Development,
Environmental Criteria and Assessment Office. EPA/600/6-89/001.
USEPA. 1988e. U.S. Environmental Protection Agency. Technical support
document for the land application and destruction and marketing of sewage
sludge. EPA Office of Water Regulations and Standards.
5-102
1593q
-------
USEPA. 1989a. U.S. Environmental Protection Agency. U.S. EPA - Paper
Industry Cooperative Dioxin Study. 104-Mill Data Base, July 17 revision,
Office of Water Regulations and Standards.
USEPA. 1989b. U.S. Environmental Protection Agency. Hazardous waste
treatment, storage, and disposal facilities (TSDF) - air emissions
model. Office of Air Quality Planning and Standards. EPA-450/3-87-026.
USEPA. 1989c. U.S. Environmental Protection Agency. Human health risk
assessment for municipal sludge disposal: benefits of alternative
regulatory options. Washington, D.C.: Office of Water Regulations and
Standards.
USGS. 1985. U.S. Geological Survey. National Water Summary - 1985.
Wisconsin DHSS. 1989. Wisconsin Department of Health and Social
Services. Human exposure assessment for dioxin and furan contaminated
papermill sludge applied to soils. Final draft. Madison, WI. January
1989.
Yen GR-. 1981. AT123D: Analytical transport one-, two- and
three-dimensional simulation of waste transport in the aquifier system.
Oak Ridge.National Laboratory. Publication No. 1439.
5-103
-------
6. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM THE DISCHARGE
OF EFFLUENTS FROM THE PULP AND PAPER INDUSTRY
6.1 Introduction
6.1.1 Purpose
The purpose of this assessment was to develop reasonable worst-case
estimates of exposures and risks to humans from 2,3,7,8-TCDD and
2,3,7,8-TCDF discharges from chlorine-bleaching pulp and paper mills.
The assessment focused on the highest estimated in-stream contaminant
concentrations immediately downstream of each mill discharge point
(assuming steady-state, fully mixed conditions) and the potential human
health impacts resulting from the consumption of 2,3,7,8-TCDD and
2,3,7,8-TCDF contaminated fish and drinking water associated with these
exposures. Because no comprehensive studies on 2,3,7,8-TCDD and
2,3»7,8-TCDF build-up in sediments and bioaccumulation up the food chain
exist, only the water column was investigated as a potential route of
exposure and uptake of 2,3,7,8-TCDD and 2,3,7,8-TCDF by exposed fish.
Carcinogenic and non-carcinogenic effects in humans are considered.
This chapter is a condensed version of the following report prepared
by the U.S. EPA Office of Water Regulations and Standards (OWRS) as part
of the Interagency Dioxin-in-Paper Workgroup:
USEPA. 1990. U.S. Environmental Protection Agency. Risk assessment
for 2,3,7,8-TCDD and 2,3,7,8-TCDF contaminated receiving waters from
U.S. chlorine bleaching pulp and paper mills. Washington, D.C.:
Office of Water Regulation and Standard, U.S. Environmental
Protection Agency. August 1990.
6.1.2 Scope
Two approaches were used to estimate and compare exposures to
2,3,7,8-TCDD and 2,3,7,8-TCDF resulting from surface water effluent
discharges from pulp and paper mills. In the first approach, a simple
dilution calculation was conducted to estimate the in-stream contaminant
concentration after the effluent is mixed in the receiving water. This
calculation assumed 100 percent of the in-stream contaminants (both
dissolved and adsorbed to suspended solids) are bioavailable to fish.
The second approach used the Exposure Assessment Modeling System
(EXAMS II) (Burns et al., 1982; Burns and Cline, 1985; Harrigan and
Battin, 1989) to partition in-stream steady-state water column
contaminant concentrations between dissolved and particulate forms.
However, only the dissolved contaminant concentration predicted by EXAMS
H was considered in determining exposure and risk. Both the simple
dilution and EXAMS II in-stream exposure methods were used to estimate
6-1
1601P
-------
the potential human health risks associated with ingestion of
contaminated fish and drinking water. No attempt was made in the EXAMS
II approach to estimate fish exposure to contaminants associated with
suspended particulates, bed sediments, or the food chain due to lack of
sufficient and appropriate data.
The EXAMS II method, on the other hand, provides a more reasonable
estimate of the direct exposure of fish to the contaminants from water
only. One result of this analysis is an upper bound estimate of the
potential risk of cancer over the lifetime of a hypothetically exposed
individual. No attempt has been made to characterize the human
population potentially at risk. For these risk estimations, reasonable
worst-case ambient and effluent characterizations were used, along with
best estimates of physical and chemical properties of 2,3,7,8-TCDD and
2,3,7,8-TCDF. Because not all of the parameter values used in this
assessment are "worst-case," the hypothetically exposed individual is not
considered the "most exposed individual."
The probability of an individual developing cancer in a lifetime due
to the ingestion of contaminated fish or drinking water was calculated
based on exposure estimates and the EPA carcinogenic potency factor.
Also, the data for exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF from
bleached paper mills were screened for exposure scenarios exceeding an
average of 100 pg/kg/day, an estimated one-day Health Advisory (HA) for
2,3,7,8-TCDD for protection against human liver effects (see Section 3.3.2
for details on this HA). Exposure scenarios exceeding this level were
examined in more detail to determine whether the cancer or non-cancer
endpoint is the most sensitive indicator of risk.
6.2 Exposure and Risk Assessment Methodology Requirements
The approach taken was designed to incorporate an appropriate balance
between the difficulty (detail) of the analysis and the accuracy of the
results. The critical factors considered in the development of the
analytical approach were: 1) in-stream chemical transformation processes,
2) applicability of calculation methods, 3) availability of environmental
data, and 4) model sensitivity.
The chemical/physical processes thought to most significantly
influence the fate of 2,3,7,8-TCDD and 2,3,7,8-TCDF in the aquatic
environment are dilution and adsorption of the contaminants to
particulates. Other processes, such as volatilization, hydrolysis,
photolysis, and biotransformation do not appear to significantly affect
the fate of the contaminants.
A simple dilution calculation method for estimating water column
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF provides total in-stream
contaminant concentration without consideration of the effects of
adsorption to particulates and eventual sedimentation or other fate
6-2
1601q
-------
processes. This method of predicting exposure results in worst-case
water column exposure estimates. Because 2,3,7,8-TCDD and 2,3,7,8-TCDF
appear to have a high affinity for adsorption to particulates, other
methods of estimating contaminant fate and transport are necessary to
consider partitioning between the dissolved and solid forms of the
contaminants. The Exposure Assessment Modeling System (EXAMS II), a
state-of-the-art surface water contaminant modeling system, is capable of
estimating the partitioning of a contaminant between its dissolved form
in the water column and that portion that associates with suspended and
benthic solids.
Many parameters describing the physical/chemical properties of
2,3,7,8-TCDD and 2,3,7,8-TCDF and mill-specific effluent and receiving
water characteristics are required to operate EXAMS II. STORET (a water
quality data base maintained by EPA's Office of Water that can access
water quality sampling data from monitoring stations around the country)
was used to access data on flow, total suspended solids (TSS), pH, and
other parameters required to operate EXAMS II for the receiving waters of
most mills.
A model sensitivity analysis was conducted to determine which
environmental data parameter variations had the greatest influence on
EXAMS II contaminant concentration estimation results under steady-state
conditions and given known 2,3,7,8-TCDD and 2,3,7,8-TCDF physical/chemical
properties. The analysis indicated that variations in receiving water
suspended solids levels produced the greatest variations in resulting in-
stream dissolved contaminant concentrations. Therefore, mill-specific
values were obtained and used in all subsequent EXAMS II analyses. All
other environmental parameters, except for mill-specific contaminant
leadings, and receiving water flow rates, were assigned default values.
g.3 Exposure Assessment Methodology
(1) In-stream contaminant concentrations. In this investigation,
two approaches are used to estimate and compare exposures to 2,3,7,8-TCDD
and 2,3,7,8-TCDF resulting from surface water effluent discharges from
pulp and paper mills. The first approach consists of a simple dilution
calculation conducted to estimate the total, steady-state in-stream
concentration of the contaminants after the effluent is mixed in the
receiving water. This calculation assumes 100 percent of the in-stream
contaminants (both dissolved and adsorbed to suspended solids) are
bioavailable to fish. In the second approach, the Exposure Assessment
Modeling System (EXAMS II) is used to partition in-stream steady-state
concentrations of the contaminants between dissolved and particulate
forms.
EXAMS II accounts for the high affinity of 2,3,7,8-TCDD and
2 3,7,8-TCDF for solids and, therefore, the likelihood that a percentage
Of the contaminants will be associated with suspended and benthic
6-3
I601q
-------
solids. Both the simple dilution and EXAMS II approaches are used to
estimate and compare the potential human health risks associated with
ingestion of contaminated fish tissue and drinking water.
The following simple dilution equation was used to provide an
estimate of the concentration of a contaminant downstream from a point
source release into a flowing water body after dilution of the substance
by the receiving water (USEPA 1988b):
'e'e
(6-1)
where:
C = concentration of substance in stream (mass/volume),
Ce - concentration of substance in effluent (mass/volume),
Qe - effluent flow rate (volume/time), and
Qt = combined effluent and stream flow rate (volume/time).
Although this calculation is easily executed and provides a
quantitative estimate of in-stream contaminant concentration which is
limited in precision only by the precision of the input parameters, this
calculation provides only the total in-stream contaminant concentration
attributable to the point source. It does not provide a distribution of
the contaminant between the dissolved and adsorbed states.
EXAMS II, on the other hand, is a sophisticated computer modeling
system capable of computing parameters of exposure, fate, and persistence.
Once input parameters describing the environment (temperature, stream
compartment geometry, receiving water flow, solids, organic carbon
fraction, etc.), the chemical contaminant characteristics (molecular
weight, vapor pressure, Henry's Law constant, Kow, Koc, solubility,
etc.), and the loadings are entered, the model produces a report
detailing the three sets of computations described above (i.e., exposure,
fate, and persistence).
For each mill, the calculated water column concentrations were used
as the basis for further calculations. The estimated concentrations were
considered 100 percent available to the aquatic organisms living in the
receiving waters, 100 percent available to humans using the water as a
drinking water source, and 95 percent available to humans through fish
tissue consumption.
For this assessment, the mills were grouped Into one of three
categories: direct dischargers to free flowing streams, direct
dischargers to open waters (e.g, oceans, lakes, reservoirs), and Indirect
dischargers [dischargers to POTWs) to either free-flowing streams or open
6-4
1601q
-------
waters. Contaminant concentrations resulting from direct discharges to
free-flowing streams were calculated directly using the simple dilution
and EXAMS II water column methods when adequate environmental data were
available for the site. Contaminant concentrations resulting from direct
discharges to open water bodies were calculated using the simple dilution
method, based on zone of initial dilution factors for the mills that were
provided by EPA Regions (Albright 1990; Davis 1989; Derose 1989; Fisher
1989; Greenburg 1989; Greenfield 1990; Hall 1989; Hangarden 1989; Henry
1989; Hyatt 1989; Keefler 1989; Loster 1989; Menzardo 1989; Tingperg
1989; and Weeks 1989). The zone of initial dilution is the region of
initial mixing surrounding or adjacent to the end of the outfall pipe in
which aquatic inhabitants may be chronically exposed to concentrations of
pollutants in excess of water quality standards. Initial dilution is
defined by USEPA (1982) as the flux-averaged dilution (averaged over the
cross-sectional area of the plume) achieved during the period when
dilution is primarily a result of plume entrainment, and is not dominated
by ambient conditions.
EXAMS II requires stream flow data as input to calculate in- stream
contaminant concentrations. Because flows for open water bodies were not
available, it was necessary to back-calculate "surrogate" water body
flows for direct dischargers to open water bodies based on known mill
olant flows and the dilution factors for the mills. The following
calculation was used to determine surrogate water body flows for direct
open water discharges:
F0 • (D * Fp) - Fp (6-2)
where:
f « surrogate open water body flow
f° » mill plant flow
Q" - dilution factor
Contaminant concentrations resulting from indirect discharges to
either free-flowing streams or open water bodies were calculated using
e same methods described above, except that loadings were decreased to
and 25 percent of the total to account for the effects of treatment on
discharge effluent stream.
In-stream contaminant concentrations were calculated using the
harmonic mean flow for the receiving water. The harmonic mean flow 1s
defined as the reciprocal of the mean value of the reciprocal of
individual values.
(2) Whole-body and fish filet contaminant concentrations. Tissue
residue levels for fish exposed to the 1n-stream contaminant
concentrations estimated above were calculated by multiplying the
6-5
1601Q
-------
contaminant concentration by estimated bioconcentration factors (BCFs)
for 2,3,7,8-TCDD and 2,3,7,8-TCDF. Estimated fish tissue residue levels
resulting from exposure to these contaminants in the water column were
based on BCFs of 5,000 (2,3,7,8-TCDD, filet only), 100,000 (2,3,7,8-TCDD,
whole body) and 3,900 (2,3,7,8-TCDF, whole body). For example, assuming
a 2,3,7,8-TCDD water column concentration of 2.2 x 10"8ng/l, and a BCF
of 5,000, whole body fish tissue concentration would be calculated as
follows:
(water column concentration) (BCF) » filet concentration
(2.2 x 10-8ng/l)(5,000) . 1.1 x 10'4ng/kg
The BCF for 2,3,7,8-TCDD of 5,000 (USEPA 1984) is based on fish filet
residue levels, not whole body levels. This BCF, in combination with a
moderate fish tissue consumption rate (6.5 g/day), was the basis for
estimating human health impacts from consumption of contaminated fish in
EPA's ambient water quality criteria for 2,3,7,8-TCDD.
A second BCF of 100,000 was developed primarily from the results of
the EPA Duluth Laboratory's most recent studies on the bioconcentration
of 2,3,7,8-TCDD by fish (Cook et al. Unpublished). During these
investigations, BCFs for carp and fathead minnows were determined through
laboratory studies with exposures of up to 71 days in duration. The BCF
levels presented in the Cook et al. study ranged from 65,900 ± 9,300
to 159,000 ± 40,000. These BCF values were developed from whole body
tissue residue levels. The Cook et al. study is preliminary and has not
been peer-reviewed. Based on this information, the present study
selected a reasonably conservative BCF value of 100,000 to represent a
more extreme worst-case bioaccumulation potential that, in combination
with higher fish tissue consumption rates, result in higher estimated
human health risks.
Results of a recent literature review by Nabholz et al. (Unpublished)
were used as the basis for selecting BCF values to determine whole body
contaminant concentrations in fish exposed to 2,3,7,8-TCDF in the water
column. Only three measured fish BCF values for 2,3,7,8-TCDF were
identified, two from water exposures and one from a dietary source. The
geometric mean of the measured BCF values for water exposure (3,900) was
used in the present study.
Contaminants are generally distributed unequally among the tissues in
the fish after their ingestion of the contaminants. For instance, high
concentrations of many contaminants accumulate in the fish liver,
generally an Inedible portion of the fish. Also, whole body residue
levels include the viscera, which contain significant quantities of
sediments ingested during feeding. Because of the affinity of
2,3,7,8-TCDD and 2,3,7,8-TCDF for sediment, high concentrations of
6-6
ISOlq
-------
contaminants would be found in this inedible portion. Using the
estimated whole-body concentration of a contaminant would therefore not
accurately reflect human exposure as a result of consumption of the
edible portion of the fish (the filet). In general, the concentration of
2,3,7,8-TCDD in fish muscle is about 50 percent of whole fish
concentration (Branson et al. 1985). To compensate for the unequal
partitioning of contaminants between the edible and inedible fish
tissues, the estimated whole-body BCFs of 100,000 (for 2,3,7,8-TCDD) and
3,900 (for 2,3,7,8-TCDF) were multiplied by 0.5 to estimate the
concentration in the edible portion of the fish. Thus, the effective
BCFs are 50,000 for 2,3,7,8-TCDD and 1,950 for 2,3,7,8-TCDF.
The 5,000 BCF for 2,3,7,8-TCDD is based on fish filet residue levels,
and, consequently, no adjustments in the fish tissue 2,3,7,8-TCDD
concentration estimates are necessary. It should be noted that for some
species of shellfish (e.g., mollusks) the whole body (minus the shell) is
consumed by humans, and the whole-body contaminant concentration would
more accurately reflect human exposure.
(3) Drinking water concentration. Drinking water contaminant
concentrations are assumed to be the same as the in-stream receiving
water concentrations which were calculated using the simple dilution and
EXAMS II water column (i.e., dissolved) approaches. It is assumed that
M) water that is ingested is taken from the point of highest in-stream
contaminant concentration after the effluent is fully mixed in the
receiving stream and (2) the untreated water is ingested as raw stream
water with no removal of contaminants.
(4) Human exposures from 1noest1on of contaminated fish tissue and
ing water. Human exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF from
consumption of contaminated fish tissue is estimated based on fish
tissue consumption rates of 6.5, 30, and 140 g/day. The 6.5 g/day
consumption rate is equivalent to less than two 1/4 Ib meals per month
and is cited by USEPA (1980) as the average level of fish and shellfish
consumption in the United States. The 30 g/day consumption rate is
equivalent to approximately eight 1/4 Ib meals per month and is
considered applicable for typical recreational fishers. The 140 g/day
consumption rate is equivalent to approximately thirty-eight 1/4 Ib meals
Der month and is considered a high consumption rate applicable for
subsistence fishers. The 30 g/day and 140 g/day consumption rates are
the values used in this report to represent consumption rates for
recreational fishers in any area with a large water body and where
Widespread contamination is evident (USEPA 1989a).
Estimates of human exposure to 2,3,7,8-TCDF through the consumption
Of contaminated fish tissue are based on the three consumption rates
(6.5, 30, and 140 g/day) in combination with fish tissue 2,3,7,8-TCDF
6-7
16014
-------
concentration estimates based on a single BCF (1,950) for fish exposed to
the contaminant in the water column.
The average daily lifetime exposure (mg of contaminant/kg of body
weight/day over a 70-year lifetime) is calculated by multiplying the
chemical concentration in the edible fish tissue by the ingestion rate
and dividing by an average adult body weight of 70 kg. The equation is
represented as follows:
C x R /c •}>
i/mc — — {*'*'
LAUt ~ Body weight (kg) x Lifetime (70 yrs) x (365 days/yr)
where:
LADE - Lifetime average daily exposure
C = Concentration in tissue (pg/kg)
R = Consumption rate (kg/day)
The average daily lifetime human exposure to 2,3,7,8-TCDD and
2,3,7,8-TCDF from the ingestion of contaminated drinking water is based
on a 2 liter/day average lifetime ingestion rate (NAS 1977). The average
daily lifetime exposure for a 70 kg adult is determined by multiplying
in-stream chemical concentrations by a 2 liter/day average lifetime
ingestion rate and then dividing by 70 kg, as follows:
C x R
LADE " Body weight (kg) x Lifetime (70 yrs) x (365 days/yr)
where:
C » Concentration of chemical in stream (pg/1)
R - Consumption rate (I/day)
6.4 Risk Assessment Methodology
6.4.1 BioavaiTable Dose From Ingestion Of Contaminated Fish Tissue And
Drinking Hater
Not all contaminants that are ingested with fish tissue are available
for uptake by humans. Results of a recent review conducted by Boyer
(1989) suggest that 85 to 95 percent absorption is a reasonable estimate
of 2,3,7,8-TCDD bloavailability in humans from the ingestion of fatty or
oily foods, especially milk, fish, and meats. For the present study, the
conservative upper limit of this range of bioavailabillty (95 percent)
was used. Therefore, the estimated exposure of humans to 2,3,7,8-TCDD
and 2,3,7,8-TCDF from the consumption of contaminated fish is multiplied
6-8
1601q
-------
by 0.95 to arrive at an estimated human dose. Further characterizations
of doses to specific target organs, for example, via a pharmacokinetic
analysis, were not conducted.
Boyer (1989) also investigated bioavailability of 2,3,7,8-TCDD from
water. Although the author could find no data that specifically
addressed the bioavailability of 2,3,7,8-TCDD from drinking water, he
assumed that the contaminant would be present at its maximum solubility
in water and, therefore, would be 100 percent bioavailable for absorption
to the gastrointestinal tract. The present study also assumes that
contaminants in drinking water are 100 percent bioavailable.
(1) Estimated cancer risk from Incestion of contaminated fish tissue
fipH drinking water. The average daily lifetime bioavailable dose (for
'both fish tissue and drinking water contamination) is multiplied by the
£PA carcinogenic potency factor for 2,3,7,8-TCDD to calculate a
conservative (upper bound) estimate of the hypothetically exposed
•individual's cancer incidence rate above background incidence rates due
to 2,3,7,8-TCDD. The probability of developing cancer in a lifetime due
to a given dose of contaminant is represented by the following formula,
which estimates a plausible upper limit to excess lifetime risk of cancer
at low doses:
(b)(d) (6-5)
where:
R - cancer risk
b • the EPA carcinogenic potency factor, (1.6 x 10"4 (pg/kg/day)"1)
d - lifetime average daily bioavailable dose.
To estimate the combined 2,3,7,8-TCDD/2,3,7,8-TCDF cancer risk,
2 3,7,8-TCDF doses are converted to 2,3,7,8-TCDD toxicity equivalences
(TEQs) using a multiplier of 0.1. The TEQ value is then multiplied by
the carcinogenic potency factor for 2,3,7,8-TCDD to obtain the combined
2,3,7,8-TCDD/ 2,3,7,8-TCDF risk. The TEQ 1s generated by using the
toxicity equivalency factor (TEF) recommended in Barnes et al. (1989).
In this study, TEQ represents only the contribution of 2,3,7,8-TCDD
and 2,3,7,8-TCDF to risk. There are likely to be additional risk
contributions from other chlorinated dibenzo-p-dioxins and furans
associated with discharges from chlorine-bleaching pulp and paper mills
that are not addressed here. However, 2,3,7,8-TCDD and 2,3,7,8-TCDF
generally account for more than 90 percent of the TEQ in effluents from
Chlorine-bleaching pulp and paper mills.
(2) Non-cancer health risks from ingestion of contaminated fish
+j5£jj£. For risk associated with 2,3,7,8-TCDD and 2,3,7,8-TCDF
exposures, cancer is generally considered the most sensitive endpoint.
6-9
-------
It is assumed that if individuals are protected from significant concern
for cancer, they will also be protected from other endpoint risks such as
developmental toxicity, reproductive effects, liver toxicity,
immunotoxicity, etc. However, cancer risks are computed assuming an
average daily dose over a lifetime of exposure. If individuals were
exposed infrequently to relatively high doses over a short period of
time, the risks associated with that level of exposure, when averaged
over a lifetime, might not be significant in terms of carcinogenic risk.
However, the individual could be at risk for other health effects from
the short-term exposure.
EPA issues Health Advisories (HAs) for such short-term exposures.
HAs are considered doses likely to be without appreciable risk for
deleterious effects. The HAs are appropriate for comparison with a
single dose or single-day intakes, or short-term exposure. For the
purpose of this assessment, one-day and ten-day HAs have been developed
for exposure to 2,3,7,8-TCDD. They were derived from animal toxicity
data and incorporate uncertainty factors intended to take into account
differences in sensitivity between animals and humans, variability in
susceptibility within human populations, and other factors. The HAs for
exposures to 2,3,7,8-TCDD are as follows: 1 day - 300 pg/kg/day for
protection against developmental toxicity effects; 1 day - 100 pg/kg/day
for protection against liver effects; and 10 day - 10 pg/kg/day for
protection against liver effects (USEPA 1989b; Farmon 1989).
For this assessment, the data for exposures to 2,3,7,8-TCDD and
2,3,7,8-TCDF from the paper mills were screened for exposure scenarios
exceeding an average of 100 pg/kg/day. Exposure scenarios exceeding this
level were examined in more detail to determine whether the cancer or
non-cancer endpoint was the more sensitive indicator of risk.
6.5 Results of the Assessment
The results of this Investigation are presented in two parts. The
first part addresses exposure estimates and compares the results of the
in-stream contaminant concentration calculations that were performed
using the two exposure assessment approaches (simple dilution and EXAMS
II water column). The second part of the results presents estimated
human health risks (Individual cancer risks as well as non-cancer health
effects) associated with the ingestion of 2,3,7,8-TCDD and 2,3,7,8-TCDF
contaminated fish tissue and drinking water.
Sufficient data were not available for all mills investigated to
allow a complete evaluation and comparison of results for each of the 104
facilities. For example, for several mills discharging to open waters
(i.e., lakes, open ocean), information was not available on the zone of
initial dilution in the receiving water. This parameter 1s necessary for
6-10
1601q
-------
calculating effluent dilution. For other mills, the accuracy of the data
was questionable and new samples were being taken. However, the results
of the new sample evaluations were not available for inclusion in this
study. In addition, for some mills there was sufficient information to
predict risks based on the simple dilution method, but insufficient
information to predict risk based on the EXAMS II method. Also, harmonic
mean flow data were not available for several facilities. Actual fish
tissue concentration data from the National Bioaccumulation Study (NBS)
were also evaluated as part of this assessment. The number of facilities
evaluated was dependent on the number and location of samples taken as
part of the study.
6.5.1 Exposure Assessment Results
(1) In-stream contaminant concentrations. The estimated in-stream
2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations for each of the samples from
the 104 mill sites are based on harmonic maan flow (in pg/1). In some
instances, more than one sample result is presented for a given mill.
Results are ranked by mill in decreasing 2,3,7,8-TCDD and 2,3,7,8-TCDF
concentration order (based on the highest sample value per mill).
For each of the samples, estimated in-stream 2,3,7,8-TCDD and
2 3,7,8-TCDF concentrations based on harmonic mean flow were highest when
calculated using the simple dilution exposure assessment method.
jn.stream 2,3,7,8-TCDD concentrations estimated using the simple dilution
method ranged from a high of 3.2 x 10Z pg/1 to a low of 4.1 x 10~5
na/1- In-stream 2,3,7,8-TCDF concentrations ranged from a high of 8.0 x
10* pg/1 to a low of 1.0 x 10'4 pg/1. Using the EXAMS II water
column method, estimated 2,3,7,8-TCDD concentrations ranged from a high
of 8.3 x 101 pg/1 to a low of 3.4 x 10'5 pg/1. Estimated
2,3,7,8-TCDF concentrations ranged from 7.1 x 10Z pg/1 to 1.1 x 10"3
pg/1 •
The estimated distribution of mills for which discharges result in
2 3,7,8-TCDD and 2,3,7,8-TCDF concentrations falling within specific
concentration ranges (based on harmonic mean flow) are presented using
the simple dilution method (Figure 6-1) and EXAMS II water column method
^Figure 6-2).
(2) F1sh tissue contaminant concentrations. The mill-specific
fish tissue concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF were also
letinated using the two exposure assessment methods. The actual fish
tissue concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF were measured
during the National Bioaccumulation Study.
6-11
160W
-------
2378-TCDF
2378-TCDD
CD
e
n
oc
c
o
u
o
O
_c
•*-
i
j2
I
"o
o
.0
1E+02 1E+01 1E*00 1E-01 1E-02 1E-03 1E-04
Concentration Rang* (pg/l)
FIGURE 6-1. Distribution of the number of mills for which discharges
would result In a given range of water column contaminant concen-
trations as estimated by the simple dilution method.
ToUl number of mill* •viMMUdj • §7.
Citlme tee beeed *n HermenM Meen Flew •« receiving wlt«r>
Number •( mill! wllhln cone.nlrillon rin«*« for whtoh 237S-TCDD and/or
237i-TCDF w«r» net detected In the eHluenl and therefore water column
coiwentrellon eetlmates «re based en effluent eoMentrstloni ef 1/2 the
detection Hmlt:
1E»0 11-1 11-t 11-3
TCOO 1 • 7 4
TCDF 3 3 1
6-12
-------
2378-TCDD
2378-TCDF
33 —
30 —
I
s
u
o
O
M
I
"o
w
21
18
ta
12
12
11*02 1E*01 11*00 11-01 11-02 11-03 11-04
Concentration Rang* (pg/l)
Figure 6-2. Distribution of the number of mills for which discharges
would result In a given range of water column contaminant concen-
trations as estimated by the EXAMS II water column method.
*)•!••:
Total number o( mllle evaluated • ST.
E»tlm«(»i bated en Harmonic Mean Flaw of receiving water*.
Num*er of mill* within concentration rengte (ef which 2371-TCOD end/or
2371-TCOP were not detected In the el fluent end there»e»e w»ter column
concenlritlen eetlmetei are bated en effluent eencentraUena el 1/2 the
deteetlon limit.
TCDO
tear
11*0
i
i
11-1
7
:
11-2
•
4
11-3
4
i
11-4
2
6-13
-------
The highest fish tissue concentrations due to in-stream exposure to
the contaminants were estimated by the simple dilution method. The
2,3,7,8-TCDD fish tissue concentrations estimated using the BCF of 5,000
ranged from a high of 1.6 x 103 ng/kg to a low of 2.05 x 10~4 ng/kg.
Using the BCF of 50,000, 2,3,7,8-TCDD fish tissue concentrations ranged
from a high of 1.6 x 104 ng/kg to a low of 2.0 x 10'J ng/kg. Use of
the simple dilution method estimated 2,3,7,8-TCDF concentrations in fish
tissue (using the single BCF of 1,950) ranging from 1.6 x 103 ng/kg to
2.0 x 10'4 ng/kg.
The EXAMS II water column method resulted in fish tissue
concentrations of 2,3,7,8-TCDD ranging from a high of 4.2 x 10' ng/kg
to a low of 1.71 x 10*4 ng/kg using the 5,000 BCF and from 4.2 x 10J
ng/kg to 1.2 x 10'3 ng/kg using the 50,000 BCF. The 2,3,7,8-TCDF fish
tissue concentrations estimated by the EXAMS II water column method
ranged from 1.4 x 103 ng/kg to 1.5 x 10~3 ng/kg.
Measured 2,3,7,8-TCDD concentrations in fish tissue from the National
Bioaccumulation Study ranged from a high of 7.2 x 101 ng/kg to a low of
2.0 x 10"1 ng/kg. 2,3,7,8-TCDF measured values ranged from 2.1 x 10'
ng/kg to 1.3 x 10"1 ng/kg-
(3) Drinking water contamination. This study assumes that the
concentrations of the contaminants expected to be found in drinking water
are the same as those predicted in-stream. The distribution of the
number of mills for which discharges result in in-stream concentrations
of the contaminants within specific concentration ranges are illustrated
in Figure 6-1 (for the simple dilution method) and Figure 6-2 (for the
EXAMS II water column method}.
6.5.2 Risk Assessment Results
(1) Bloavailable dose from ingestion of fish tissue and drinking
water. The bioavailable dose to humans from consumption of
contaminated fish tissue was calculated based on 95 percent
bioavailability and three fish tissue consumption rates: 6.5 g/day in
combination with fish tissue concentrations based on fish
bioconcentration factors of 5,000 for 2,3,7,8-TCDD and 1,950 for
2,3,7,8-TCDF; and 30 and 140 g/day in combination with fish tissue
concentrations based on fish bioconcentration factors of 50,000 for
2,3,7,8-TCDD and 3,900 for 2,3,7,8-TCDF. The bioavailable dose from
drinking water was calculated based on a drinking water ingestion rate of
2 L/day and a 100 percent oral dose bioavailability. The mill-specific
bioavailable doses of 2,3,7,8-TCDD and 2,3,7,8-TCDF from consumption of
contaminated fish tissue are estimated based on both the simple dilution
and EXAMS II methods. Mill-specific bioavailable doses from ingestion of
contaminated drinking water are also estimated. These values were used
to predict the hypothetically exposed individual's upper bound cancer
risks associated with discharges from each mill.
6-14
1601q
-------
(2) Estimated cancer risk from inaestion of contaminated fish tissue
_ drinking water. The mi 11-specific upper bound lifetime risks of
cancer to the hypothetically exposed individual from consumption of
contaminated fish tissue are predicted based on the simple dilution and
EXAMS II methods. The mill-specific upper bound risks of cancer from
ingestion of contaminated drinking water are also predicted . The values
presented are in toxicity equivalents (TEQs), representing the combined
impacts of 2,3,7,8-TCDD and 2,3,7,8-TCDF. The cancer risks associated
with contaminated fish consumption are presented for 6.5 g/day, 30 g/day,
and 140 g/day consumption rates. The percent 2,3,7,8-TCDD contributing
to TEQ is also estimated for contaminated fish tissue consumption and for
contaminated drinking water ingestion.
(a) Contaminated fish tissue. Figures 6-3 through 6-6
present the estimated distribution of the number of mills for which
discharges would result in a given range of estimated lifetime cancer
risks for the hypothetically exposed individual due to the consumption of
contaminated fish tissue based on the simple dilution exposure assessment
method and the EXAMS II water column exposure assessment method.
Estimated values are in TEQs.
The results of calculations using the 6.5 g/day fish tissue
consumption rate In combination with the BCF of 5,000 are considered to
be reasonable worst-case scenarios. The results of these calculations
are therefore presented separately from the results of calculations using
the 30 and 140 g/day consumption rates in combination with the BCF of
100,000, which are considered more extreme worst-case scenarios.
Using the simple dilution exposure assessment estimates, the 6.5
g/day fish tissue consumption rate, and the fish tissue contaminant
concentrations based on a BCF of 5,000, the upper bound mill-specific
cancer rates for the hypothetically exposed individual range from the
j0-/ to 10"° risk levels {Figure 6-3). Risk levels associated with
discharges from 80 of the 97 mills evaluated (82 percept) fell within the
jO'4 to 10'° risk levels, with 35 mills within the 10'5 risk level.
Hill-specific cancer rate estimates using the 30 g/day fish tissue
consumption rate and fish tissue contaminant concentrations based on a
BCF of 50,000 range from the slO'1 to 10'6 risk levels (Figure 6-4).
Seventy of the 97 mills (71 percent) were associated with risk levels
Between 10'J to 10'4, and 40 of these 70 fell within the 10*3
range. Using the 140 g/day fish tissue consumption rate and fish tissue
contaminant concentrations based on the 50,000 BCF, risk levels range
from slO'1 to 10'b (Figure 6-4). Sixty-six out of the 97,mills ,
(68 percent) were associated with risk levels between 10"z to 10"3
39 within the 10"3 range.
6-15
-------
6.5 g/day consumption
40
35
o
O)
c
a
cc 30
J2
E
£ 25
w 20 H
! 15
i 10
5 -
1E-02 1E-03
1E-04 1E-05
Risk Range
1E-06 1E-07 1E-08
FIGURE 6-3. Distribution of the number of mills for which discharges
would result In a given range of lifetime cancer risk due to the
consumption of contaminated fish tissue as estimated by the
simple dilution method (6.5 g/day consumption rate and BCF
of 5,000 for 2378-TCDD*).
Not**:
Total number of mill* •viiuiUd • i7.
Combined 2378 -TCDD/-TCOF rl.k pr.dict.d u.lng TEQ.
Numb*r of mill* within rl*k r«na.. tor which aaTi-TCOD «nd/or 237S-TCOF w.r«
not dotoetod In tho •(flu.nt *nd thor*foro rl*k **tlmat** *r* ba**d on offlu*nt
concentration* of 1/2 th* dotoetlon limit:
TCOO
rcor
TCOO * TCOF
1E-4 16-6 1E-I TE-7
27*3
1 1
2 2 1
• Rooont laboratory ovMone* Indtoat** that a BCF Mgh*r than 8,000
for 2378-TCDO (*.a^ 60,000) mor* aeauratoly rofloot* uptako of 2378-TCDO by
flih. U** of • BCF of 80,000 for 2378-TCOO woiiM bMroaao rl*k by an ord*r of
mafnltudo.
6-16
-------
140 g/day consumption
30 g/day consumption
I
(A
cc
40 U
35 -
30 -
25 -
20 -
- 15 -
10
5
ME-01 1E-02 1E-03 1E-04 1E-OS
Risk Range
1E-06
FIGURE 6-4. Distribution of the number of mills for which discharges
would result In a given range of lifetime cancer risk due to the
consumption of contaminated fish tissue as estimated by the
simple dilution method (30 and 140 g/day consumption rates
and BCF of 50,000 for 2378-TCDD).
T»UI number •( mill* •v«ki*U4 • »7.
C*mMn*4 tiri -TCDO/-TCDF rltk pr .*•<•< u«to« TIG.
Numb.f •« mill* within risk r.nv .. (or wntaft 137 •-TCDO «nd/«r
2 371-TCDF w»f. n.l d« !••!.« In lh« •ffkiMt aitri thwvtata risk
• •llmal*. «r. bn.« M .tlkitnl «en«*n(rill««« •( 1 /2 th.
d»l»«ll«i Mult:
11-1 H-3 11-4 1I-« 11-t
30t/»1«y
TCDO 7432
TCOr> 1 ,
TCOO A TCOF 1 3 1
140j/
-------
6.5 g/day consumption
40
-------
140 g/day consumption
30 g/day consumption
0)
O)
KJ
cc
cc
c
£
0
j>»
I
'o
L»
-Q
Z
ME-01 1E-02 1E-03 1E-04 1E-05 1E-06 1E-07
Risk Range
FIGURE 6-6. Distribution of the number of mills for which discharges
would result In a given range of lifetime cancer risk due to the
consumption of contaminated fish tissue as estimated by the
EXAMS II method (30 and 140 g/day consumption rates and
BCF of 50,000 for 2378-TCDD).
Notoo:
Total number of mill* evaluated • i7.
Combined 237• -TCOD/-TCDF risk predicted uetn( TEQ.
Number of mill, within rnk range* lor which 237t-TCOD and/or
237S-TCOF were not detected In the effluent and therefore rttk
eattmatea are baaed on effluent oonoontratlona ef 1/2 tho doteetlon
limit:
11-2 11-3 11-4 1C-5 1I-« 11-7
30l/d«y
TCDO 4 • 2 2
TCOF 1 1
TCDDITCDF 2 2 1
140l/diy
TCDO
TCDF
TCDD * TCOF
7411
1 1
1 3 1
6-19
-------
Mill-specific upper bound cancer rate estimates for the
hypothetically exposed individual using the EXAMS II water column
exposure assessment method, the 6.5 g/day fish tissue consumption rates,
and the fish tissue contaminant concentrations based on a BCF of 5,000
range from the 10"3 to 10'8 risk levels (Figure 6-5). Seventy of the
87 mills evaluated (80 percent) were associated with risk levels between
1(T5 (33 mills) to 110~1 to 10'7 risk levels (Figure 6-6). Sixty-four of
the 87 mills (74 percent) were associated with risk levels within the
10"3 to 10'* range, and 40 of these fell within the 10"* range.
Cancer risk estimates using the 140 g/day fish tissue consumption rate
and the BCF of 50,000, range from the >10-1 to 10'° risk levels
(Figure 6-6). Sixty-two of the 87 mills (71 percent) were associated
with risk levels between the 10'3 and 10'4 range, and 37 of these
fall within the 10~3 range.
(b) Contaminated drinking water. Figures 6-7 and 6-8 present
the distribution of the number of mills for which discharges are
estimated .to result in a given range of upper bound lifetime cancer risks
to the hypothetically exposed individual due to the ingestion of
contaminated drinking water. Only those facilities discharging to fresh
water lakes, rivers, and streams are included in this analysis. No
discharges to marine or estuarine waters are included, since these water
bodies would not be used as drinking water sources.
Use of the simple dilution method estimated that the cancer risks
associated with the 69 mills evaluated range from the 10~* to 10"a
risk levels (Figure 6-7). The greatest percentage of these mills (44, or
64 percent) were associated with risk levels within the 10'° (23 mills)
to 10'7 (21 mills) range. Use of the EXAMS II water column method
estimated that the risk levels associated with the 64 mills evaluated
range from the 10'5 to 10'9 levels (Figure 6-8). Fifty of these
mills (78 percent! were associated with risk levels between the 10"°
(18 mills) to ID'7 (32 mills) range.
(3) Non-cancer health effects from Ingestion of contaminated fish
tissue. The mill-specific human doses from the consumption of a single
115 gram (1/4 pound) portion of contaminated fish tissue (using a BCF of
50,000 for 2,3,7,8-TCDD and a BCF of 1,950 for 2,3,7,8-TCDF) were
estimated based on the simple dilution and EXAMS II water column exposure
assessment methods. The values are presented In 2,3,7,8-TCDD and TEQ
doses. Results are reported in pg/kg/day for comparison to the estimated
one-day Health Advisory for protection against liver effects (100
pg/kg/day).
6-20
1601q
-------
9
0
C
JC
JA
oc
C
JO
i
»-
o
k.
-------
o
O)
c
(Q
tr
x
«
tr.
_M
i
"o
O
IE-OS 1E-06 1E-07
Risk Range
1E-08 1E-O9
Figure 6-8. Distribution of the number of mills for which discharges
would result In a given range of lifetime cancer risk due to the
Ingestlon of contaminated drinking water as estimated by the
EXAMS II method.
Net**:
Total number of mill* •v«lu*t*d • «4.
Combine* 237 • -TCDD/-TCDF rlak •replete* ustna TIQ.
••••a on « 2 L/day Inflection rat*.
Number of milla within risk rangea fer whtoh 237S-TCDD and/or
2 37S-TCOF were net detected In the effluent and therefore rlik
estimate* are baaed on effluent oonaentratlona of 1/2 the
detection limit:
1E-S 1E-« 1E-7 1C-8 1C-*
TCOO
TCDF
TCOO & TCDF
4 2 1
1 1
3 1 1
6-22
-------
Based on the simple dilution method results (Figure 6-9}, the dose
associated with discharges from 29 of the 97 mills evaluated (27 percent)
would equal or exceed the estimated one-day HA dose for protection from
liver effects (100 pg/kg/day). Use of the EXAMS II method (Figure 6-10)
estimates that the dose associated with discharges from nine mills of the
37 evaluated (10 percent) would equal or exceed the 100 pg/kg/day dose
1 evel.
g.6 Discussion of Results
6.6.1 Assumptions, Limitations, and Uncertainties
This section presents the assumptions that were made during the
planning and conduct of this study and discusses significant results and
the limitations and uncertainties associated with those results. The
following is a list of assumptions used in this investigation:
• Mill-specific, five-day composite effluent contaminant
concentrations were multiplied by mean plant flow rates to
determine contaminant load. This resulting load to the receiving
water is assumed to be continuous. The representativeness of the
effluent sample as reflecting long-term mill operations is
unknown; since then, the mills may have made plant process or
operation changes to reduce dioxin and furan formation.
• The highest estimated in-stream concentrations in the immediate
vicinity of the discharges (assuming steady-state, fully mixed
conditions) were considered for fish exposure.
• Receiving water stream flow rates for estimating human health
risks were calculated using the harmonic mean of historic flow
measurements from nearby stream gaging stations. These flows may
not be the same as those used by specific states to assess risk.
• Three bioconcentration factor (BCF) values were used for
estimating 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations in fish:
two for 2,3,7,8-TCDD and one for 2,3,7,8-TCDF. The resulting fish
tissue concentrations were used to estimate human exposure to the
contaminants through consumption of fish tissue. For 2,3,7,8-TCDD,
a BCF of 5,000 was used in combination with a human consumption
rate of fish tissue of 6.5 g/day, and a BCF of 50,000 was used in
combination with consumption rates of 30 g/day and 140 g/day. The
6.5 g/day fish tissue consumption rate in combination with the BCF
of 5,000 is considered a reasonable estimate for an average
consumer of locally caught fish. The 30 and 140 g/day consumption
rates in combination with the BCF of 100,000 are used as
sensitivity comparisons and represent more extreme exposure
scenarios for recreational and subsistence fishers. A single BCF
for 2,3,7,8-TCDF of 3,900 was used in combination with each of the
three consumption rates.
6-23
-------
34
0)
O)
nj
cc
o
w
o
0
32 L
28 I—
24 I—
12 h—
a —
4 —
1E*04 1E*03 1E«02 1E*01
Dose Range(pg/kg/day)
11*00
1E-01
1E-02
FIGURE 6-9. Distribution of the number of mills for which discharges
would result In a given range of human doses from a one-time
exposure to contaminated fish tissue as estimated by the simple
dilution method.
Not**:
T«l*l numb*r of mlllt •viluaUd • »7.
Combined 2378 -TCDD/-TCDF do** predicted u*tn« TEQ.
Bi*«d on th* «on*umptlon of • ilngl* 11 8 g portion o« cont*mln*t«d ll*h tl**u*
• n* using • fHh fll*t ICF of 60,000 for 2 37S-TCDO.
Numb*r of mill* within do** ring** for which 2378-TCDO *nd/or
2 371-TCDF w*r* not d*t*et*d In th* •ffhiont «nd thorofor* d«*»
••tlmat** *r* b***d on *fflu*nt eono»ntr*tlon* of 1/2 tho dotoatlon
limit:
1E»2 1E*1 1E*0 1E-1
TCDO 1743
TCOF 1 1
TCDO * TCOr- 2 3 1
6-24
-------
ID
0>
c
ra
OC
<3>
W
o
Q
4>
-Q
E
3
z
1E+03 1E+02 1E+01 1E+00 1E-01 1E-02
Dose Range (pg/kg/day)
FIGURE 6-10. Distribution of the number of mills for which discharges
would result In a given range of human doses from a one-time
exposure to contaminated fish tissue as estimated by EXAMS II
method.
NoUs:
Total numbor of mill* ovaluatod • 87.
Combined 2379 -TCDD/-TCOF d«»« pr.otcUd usiftf TEQ.
Bmutd on th« consumption of • «lnal« 118 g portion of eontamlnatod f l*h tltiuo
•nd using • fish f Hot BCF of 60,000 for 2378-TCDO
Numbor of mills within doso r»ng«« for wM«n 2378-TCOD and/or
2378-TCOF woro not dotoetod In th« offki*nt and Mtorofero dos*
• stlmatos aro basod on offhiont oonoontratl«na of 1/2 tho dotootlon
limit:
TCDD
TCDF
TCDD 4
1E*2
1
1 TCDF
1E*1
5
1
1
1E»0
S
3
1E-1
2
1
1
1E-2
1
1
6-25
-------
• A drinking water ingestion rate of 2 I/day was used to estimate
human exposures through ingestion of contaminated drinking water.
It was assumed that the water consumed is taken from the point of
highest in-stream pollutant concentration after the effluent is
fully mixed in the receiving stream, and no treatment of the water
is undertaken to remove contaminants prior to ingestion.
• Fish tissue contaminant bioavailability for humans was assumed
to be 95 percent of oral dose. Contaminants in water were assumed
to be 100 percent bioavailable to both fish and humans.
• Fish were assumed to be exposed to contaminants only in the
water column. No food chain or sediment associated exposures were
considered, other than for the simple dilution method in which the
total in-stream contaminant level (both dissolved and adsorbed to
suspended solids) were bioavailable.
In evaluating the results of this assessment, it should be noted that
BCFs are highly species specific. The BCF for a contaminant in a given
fish species is dependent on fish tissue lipid content, mode of
contaminant uptake, and other factors. Thus, using a single BCF does not
take into account interspecies differences in the rate and degree of
contaminant bioconcentration. For example, the study conducted by Cook
et al. (Unpublished) indicates that BCF values of 200,000, higher than
those used in this study, may be applicable for 2,3,7,8-TCDD for some
species of fish. Also, the 50,000 BCF for 2,3,7,8-TCDD used in
conjunction with fish consumption rates of 30 and 140 g/day for
recreational and susistence fishers is based on the assumption that only
the filet portion of the fish is consumed. However, some subpopulations
of subsistence fishers and certain ethnic groups eat whole fish in which
the concentration of contaminants is likely to be higher than in the
filet alone. Therefore, the use of a BCF of 50,000 may underestimate
risks to these subpopulations.
The predictions also do not take into consideration the mobility of
fish in the receiving waters. Both resident and migrating species will
move in and out of the discharge area. This analysis assumed that the
fish remain exposed to the predicted contaminant concentration up to the
time they are caught, thus resulting in a conservative estimate of
aquatic life impacts and human health risk.
No attempt has been made to estimate fish exposure to contaminants
associated with suspended part-iculates, bed sediments, or the food chain
becuase of a lack of sufficient and appropriate scientific data and
understanding of the bioaccumulation of these contaminants by fish
through these routes of exposure. An exception is in considering the
results of the simple dilution method in which total contaminant
concentrations, both dissolved and adsorbed to suspended particulates,
6-26
1601q
-------
are evaluated. It is evident that food and sediment provide exposure
routes to fish downstream, where the amount of dissolved 2,3,7,8-TCDD and
2,3,7,8-TCDF available for uptake across gills becomes much less; thus,
the assumption that fish remain in the area immediately downstream from
the point of discharge is sufficiently conservative to compensate for any
Tack of food chain or sediment associated exposure components. In
addition, as a check and a sensitivity comparison, the results of the
simple dilution calculation are considered to provide an upper bound on
fish tissue contaminant levels since 100 percent of the in-stream
contaminants are assumed to be bioavailable.
The assumed fish tissue consumption rates also have an impact on
results of this assessment. The fish tissue consumption rate of 6.5
q/day (or less than two 1/4 lb meals per month) is considered an average
level of fish and shellfish consumption in the United States. However,
there appears to have been a significant increase in fish consumption
rates in the United States since 1980 when the national average fish
consumption rate of 6.5 g/day was derived. Therefore, risks estimated
based on this consumption rate may, in some cases, significantly
underestimate risk.
It should also be noted that, if multiple discharges to the same
waterbody are present, the actual risk associated with a waterbody may be
substantially greater than estimated in this study. For example, there
are several chlorine-bleaching pulp and paper mills that discharge to the
Columbia River basin. Calculations in this report assume that each mill
discharges to a receiving stream with no background level of
contamination. Therefore, in the case of multiple discharges to a
receiving stream, estimating risks from one mill alone can result in a
Significant underestimate of risk.
The simple dilution method assumes that all contaminants in the water
column, both dissolved and adsorbed to suspended solids, are bio-
As a result, for each of the mills analyzed, the simple
dilution exposure assessment method resulted in higher contaminant
concentrations and human health risks than did the EXAMS II water column
method. The EXAMS II water column method, on the other hand, considers
only those contaminants in the dissolved phase. In cases where the
receiving water TSS was relatively low, the simple dilution and EXAMS II
Later column results are comparable. However, when suspended solids
concentrations were high, the EXAMS II method estimated risks
significantly lower than those predicted by the simple dilution method.
Therefore, in those water bodies with relatively high suspended solids
content, the EXAMS II method likely underestimated human health risks
from consumption of contaminated fish tissue, since fish exposure to
pediment-absorbed contaminants was not considered.
6-27
16011
-------
Results of this assessment indicate that the fish tissue exposure
route poses a greater human cancer risk to the hypothetically exposed
individual than does the drinking water exposure route. However, the
upper bound cancer risk estimated from consumption of contaminated fish
tissue based on the 6.5 g/day consumption rate and the BCF of 5,000 are
relatively close to the cancer risk estimates based on ingestion of
contaminated drinking water. It should be noted that fish tissue
consumption may not pose a greater risk to the entire population than
ingestion of contaminated drinking water. Determining which exposure
route poses the greatest risk to the entire population requires knowledge
of the number of persons eating contaminated fish tissue versus the
number of persons who use contaminated surface water as a drinking water
source. Such a population assessment was not conducted for this
assessment.
A comparison of the cancer versus non-cancer risks associated with
2,3,7,8-TCDD and 2,3,7,8-TCDF discharges from pulp and paper mill
effluents indicates that more mills would result in potential cancer
risks than would result in non-cancer risks. However, the non-cancer
risk may actually be the more sensitive end point. The cancer risk was
estimated for the lifetime of a continuously exposed individual. The
non-cancer risk, on the other hand, was predicted based on the
consumption of a single portion of contaminated fish tissue. More of the
population would likely be exposed to a single dose of contaminated fish
tissue than to a lifetime of consuming contaminated fish tissue or
drinking water taken from the vicinity of certain mills. In addition,
the single dose used to predict the noncancer effects was a relatively
modest serving of 115 g (about 1/4 lb.) which is less than an
enthusiastic person might eat at one sitting.
It should be noted that the fish tissue contaminant concentrations
and subsequent risk estimates developed from the NBS data may not be
representative of ambient conditions at a given mill. The NBS samples
that were used for this evaluation were taken from sites close to pulp
and paper mills using chlorine for bleaching. However, the sites may
have been several miles from the mill, and not immediately downstream.
In some cases the samples were taken several miles upstream of the
mills. In addition, the NBS sample analyses were performed on composites
composed of several fish of different sizes (within a given range} from
which aliquots were prepared and analyzed and, therefore, would tend to
"average" contaminant concentration values. In addition, finfish migrate
in and out of an area; therefore, the fish sampled from the NBS were not
likely to be exposed to a constant level of contamination throughout
their lifetime, as was assumed for the simple dilution and EXAMS II
assessments.
The prediction of human health risk presented in this assessment
apply to hypothetically exposed individuals in the Immediate vicinity of
discharges only, not to the entire population. Predictions of the
6-28
-------
population exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF in the environment
using site specific effluent and receiving stream characteristics were
beyond the scope and resources of this study. A consequence of not
conducting a population assessment is the uncertainty concerning the
extent of human exposure and total population risks associated with
discharges of 2,3,7,8-TCDD and 2,3,7,8-TCDF.
6.6.2 Conclusions
Taking into account the above assumptions, simplifications, and
limitations, the results of this assessment indicate the potential exists
for high levels of 2,3,7,8-TCDD and 2,3,7,8-TCDF contamination in the
water column resulting from surface water effluent discharges from many
Of the chlorine-bleaching pulp.and paper mills investigated. These
predicted contaminant concentrations could represent significant
^plications for human health. Each of the exposure assessment
approaches used in this analysis predict upper bound risks that should be
carefully considered by risk managers while assessing potential impacts
associated with the discharge of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
chlorinerbleaching pulp and paper mill effluents.
6.7
References
Albright R. 1990. Personal communication. Environmental Scientist,
U.S. Environmental Protection Agency Region X.
Barnes DG, Kutz FW, Bottimore DP. 1989. Interim procedures for
estimating risks associated with exposures to mixtures of chlorinated
Hibenzo-p-dioxins and -dibenzofurans (CDDs and CDFs) and 1989 update.
625/3-89-016.
Boyer U. 1989. Bioavailability of ingested 2,3,7,8-TCDD and related
substances (Draft). Prepared for EPAs Working Group on the
gioavailability of Dioxins in Paper Products. (See Appendix A of this
assessment.)
Branson OR* Takahashi IT, Parker WM, Blau BE. 1985. Bioconcentration
kinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rainbow trout.
Toxicol. Chem. 4(6): 779-788.
L-A, Cline DM, Lassiter RR. 1982. Exposure analysis modeling
(EXAMS): User manual and system documentation. Athens, GA:
nffice of Research and Development. EPA 600/3-82-023.
Burns LA, Cline DM. 1985. Exposure analysis modeling system (EXAMS):
Reference Manual for EXAMS II. Athens, GA: Office of Research and
pevelopment
6-29
-------
Cook PM. Unpublished (1989). Bioaccumulation and toxicity of PCDDs and
PCDFs for freshwater fish. Duluth, MN: U.S. Environmental Protection
Agency
Davis S. 1989. Personal Communication. California Regional Water Board
No. 5.
Derose J. 1989. Personal Communication. Environmental Scientist, US
EPA Region V.
Farmon G. 1989. Personal Communication. [Affiliation, City, St.].
Fisher C. 1989. Personal Communication. Environmental Engineer, US
EPA Region X.
Greenfield J. 1990. Personal communication. Environmental scientist,
U.S. Environmental Protection Agency Region IV.
GSC. 1986. General Sciences Corporation. Graphical Exposure Modeling
System (-GEMS) Users Guides, Volume 2: Modeling (DRAFT). Contract No.
68-02-3770. Washington, DC: U.S Environmental Protection Agency.
GSC. 1988. General Sciences Corporation. GEMS User's Guide, Contract
No. 68-02-4281. Washington, DC: U.S. Environmental Protection Agency.
Greenburg K. 1989. Personal Communication. Environmental Scientist, US
EPA Region IX.
Hall D. 1989. Personal Communication. Environmental Scientist,
Minnesota Pollution Control Board.
Hangarden J. 1989. Personal Communication. Environmental Scientist,
USEPA Region IX.
Harrigan P, Battin A. 1989. Training Materials for GEMS and PCGEMS:
Estimating Chemical Concentrations in Surface Waters. Washington, DC:
U.S. Environmental Protection Agency, Office of Toxic Substances.
Henry T. 1989. Personal Communication. Environmental Scientist, USEPA
Region V.
Hyatt M. 1989. Personal Communication. Environmental Scientist, USEPA
Region IV.
Keefler J. 1989. Personal Communication. Environmental Scientist,
USEPA Region IX.
6-30
1601q
-------
Loster J. 1989. Personal Communication. Chief of Planning, USEPA
Region II I •
Menzardo A. 1989. Personal Communication. Chief of Permits, USEPA
Region V.
Nabholz JV. Unpublished (1989). Bi concentration factors for 2,3,7,8 -
chlorinated dibenzodioxin and 2,3,7,8-chlorinated dibenzofurans.
Washington, DC: U.S. Environmental Protection Agency, Office of Toxic
Substances.
MAS. 1977. National Academy of Sciences. Drinking water and health.
Washington, DC: NRC Press.
Tingperg K. 1989. Personal Communication. Staff Engineer, US EPA
Region II.
USEPA. 1980. Water Quality Criteria Documents. Fed. Reg.
45:79318-79379. November 28.
USEPA. 1982. Revised Section 301(h) Technical Support Document.
Washington, DC: U.S. Environmental Protection Agency, Office of Water
program Operations.
USEPA. 1984. Ambient Water Quality Criteria for 2,3,7,8-tetrachloro-
dibenzo-p-dioxin. Washington, DC: U.S. Environmental Protection Agency,
Office of Water Regulations and Standards.
USEPA. 1988a. Paper industry cooperative dioxin screening study.
Washington, DC: Office of Water Regulations and Standards. EPA
440-1-88-025.
. 1988b. Superfund Exposure Assessment Manual. Washington, DC:
Office of Remedial Response. EPA 540 1-88-001.
USEPA. 1989a. Exposure Factors Handbook. Washington, DC: Office of
Health and Environmental Assessment. EPA 600 8-89-043.
USEPA. 1989b. Aquatic life hazard assessment (including BCF values) for
Hioxins in paper (draft). Washington, DC: Office of Pesticides and
Toxic Substances.
USEPA- 1989c. National Bioaccumulation Study (Draft). Washington, DC:
U.S. Environmental Protection Agency, Office of Water Regulations and
Standards.
Weeks C. 1989. Personal Communication. Environmental Engineer, US EPA
VI.
6-31
-------
7. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FRON EXPOSURE TO
DIOXINS AND FURANS RESULTING FROM PULP/PAPER WASTEWATER SLUDGE
INCINERATION
7.1 Introduction
This chapter provides estimated exposures and risks to humans associ-
ated with emissions of dioxins and furans that may result from incinerat-
ing pulp and paper mill wastewater sludges. The chapter is a condensed
version of the following report prepared by EPA's Office of Air Quality
planning and Standards (OAQPS) as part of the Interagency Dioxin-in-Paper
Workgroup:
Dusetzina M. 1989. Human health exposure and risk assessment for
dioxins-pulp/paper waste water sludge incineration-subtask 5. U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards. October 1989.
Ousetzina (1989) used air dispersion modeling to predict potential
inhalation exposures to dioxins and furans to populations surrounding each
of the twenty-one pulp and paper mills that reported employing incinera-
tion as a wastewater sludge disposal method in the 104 Mill Study (USEPA
1988)> Both maximum individual cancer risks and aggregate (or population)
cancer risks were then estimated. Section 7.2 of this chapter summarizes
the methodology used by Dusetzina (1989). Section 7.3 and Section 7.4
summarize the results and uncertainties, respectively, of the assessment.
7.2 Methodoloav
7.2.1 Unit Risk Estimate
The numerical constant that defines the exposure-risk relationship
used by OAQPS in its analysis of carcinogens is called the unit risk esti-
mate. The unit risk estimate 1s the lifetime cancer risk occurring in a
hypothetical population in which all individuals are exposed throughout
their lifetimes (about 70 years) to an average concentration of 1 M9/m of
tne agent in the air which they breathe. An upper bound incremental unit
cancer risk estimate of 3.3 x 10'' pg/nr was developed by EPA for
3,7,8-TCDD using a multistage extrapolation model that is linear at low
' /1 iff n* i AOJI \ T * ..*.»*» •*.!«.«* •* *. MUMMJ i*»* cnn A. L. ^ A. xi__ __ uj_i. .£_,._
Hoses (USEPA 1984). It was also assumed by EPA that the cancer risk from
inhalation of 2,3,7,8-TCDF 1s one-tenth that of 2,3,7,8-TCDD or 3.3 x
I0-o pg/m3.
7.2*2 EPA Human Exposure Model (HEN) (Background)
The EPA HEM is a general model capable of producing quantitative
-voressions of general population exposure to ambient air concentrations
|,f pollutants emitted from stationary sources (USEPA 1986). The HEM
7-1
J5991
-------
contains (1) a sector-averaged atmospheric dispersion model, with
meteorological data included, and (2} a population distribution estimate
based on Bureau of Census 1980 Master Area Reference File (MARF) data.
The input data needed to operate this model are source data (e.g., plant
location, height of the emission release point, volumetric rate of release
and temperature of the off-gases). Based on the source data, the model
estimates the magnitude and distribution of ambient air concentrations of
the pollutant in the vicinity of the source and enumerates the population
around each source. The model is programmed to estimate these
concentrations for a specific set of points within a radial distance of
50 kilometers from the source.
The HEM uses the estimated ground level concentrations of a pollutant
together with population data to calculate general population exposure.
For each of 160 receptors located around a plant, the concentration of
the pollutant and the number of people estimated by the HEM to be exposed
to that particular concentration'are determined. The HEM multiplies
these two numbers to produce exposure estimates and sums these products
for each plant.
7.2.3 Pulp and Paper Hill Source Data
Twenty-one incinerators were included in the analysis. The locations
and names of the mills that incinerate wastewater sludge, according to
the 104-Mill Study, are listed in Table 7-1. Table 7-2 contains
incinerator stack parameter values for these mills as reported to EPA by
the American Paper Institute (Van Hook 1989).
According to Van Hook (1989) actual data on stack gas concentrations
of dioxins were available for only one incinerator. Therefore, emission
rates were derived for each incinerator by two different methods. The
first method (i.e., the stack gas method) used data on the concentration
of CDDs/CDFs in the stack gas from the incinerator for which API provided
data (Van Hook 1989). Emission rates were calculated by multiplying each
incinerator's operating hours by each incinerator's stack gas volume (see
Table 7-3) and the 2,3,7,8-TCDD equivalent stack gas concentration
obtained for the incinerator for which API provided measured data (see
Table 7-4). Emission rates were developed for both dioxins and furans in
units of TEQ.
The second method (I.e., the sludge concentration method) used
wastewater sludge concentration data from the 104-Mill Study. Since
information on the effectiveness of existing control devices in removing
CDOs/CDFs was unavailable, it was assumed as a worst case that these
devices were unable to remove CDDs/CDFs and that all 2,3,7,8-TCDD/TCDF
present in pulp and paper mill wastewater sludge was emitted to the
atmosphere. Table 7-5 presents the amount of wastewater sludge
7-2
1599q
-------
8925H
Table 7-1. Locations of Pulp and Paper Mill Wastewater Sludge Incinerators
Company3
Potlach Corp.
International Paper Co.
Alabama River Pulp
Scott Paper Co.
Boise Cascade Corp.
Scott Paper Co.
International Paper Co.
Potlach Corp.
International Paper Co.
International Paper Co.
Champion Intn'1
Chesapeake Corp.
Longview Fibre Co.
Simpson Paper Co.
Ketch tkan Pulp Co.
Alaska Pulp Corp.
Scott Paper Co.
ITT-Rayonier. Inc.
ITT-Rayonier. Inc.
Scott Paper Co.
Appleton Papers. Inc.
Location8
Lewiston, ID
Texarkana, TX
C la i borne, AL
Hinckley, HE
Jackson. AL
Mobile. AL
Pine Bluff. AR
Cloquet. MM
Moss Point. MS
Georgetown, SC
Houston, TX
Vest Point, VA
Longview. WA
Tacoma, WA
Ketchikan. AK
Sitka. AK
Everett, WA
Hoqulam. WA
Port Angeles. WA
Westbrook. ME
Roaring Springs. PA
Latitude6
(deg/min/sec)
46/25/28
33/18/59
31/34/54
44/41/56
31/29/37
30/44/09
34/13/11
46/43/29
30/25/3Z
33/21/53
29/52/51
37/32/21
46/06/15
47/16/02
55/23/30
57/02/50
47/59/02
46/58/03
48/07/00
43/41/04
40/20/12
Longitude*1
(deg/min/sec)
116/58/26
94/04/47
87/29/54
69/38/51
87/53/56
88/02/59
91/54/32
92/25/46
88/30/56
79/18/08
95/06/32
76/48/19
122/55/07
122/25/39
131/44/30
135/13/35
122/12/58
123/51/45
123/24/25
70/21/08
78/24/21
* From 104-Mill Study (USEPA 1988) and Van Hook (1989).
From USGS topographical maps. These locations should be fairly
representative of the plant if not the incinerator.
7-3
-------
B925H
Table 7-2. Pulp and Paper Mill Incinerator Stack Parameters9
Mill
Lewiston
Texarkana
C la i borne
HinckW*
Jackson"
Mobile
Pine Bluff
Cloquet
Moss Point
Georgetown
Houstonb
West Point
Longview
TICOM"
K»tchtUnb
Sltka"
Everett
Hoqutasi
Port Angeles
Vntbrook
Roaring Spring
Stack
height
(«)
91.4
64.3
66.8
69.5
84.0
69.5
88.7
76.2
76.2
43.0
85.3
69.5
76.2
31.0
69.5
69.5
69.5
54.0
30.2
51.5
45.7
107.6
109.6
69.5
Stack
diameter
(•)
4.10
3.95
3.63
3.51
3.51
3.51
5.60
3.58
2.44
2.44
5.20
3.51
3.51
2.67
3.51
3.51
3.51
4.00
2.13
2.43
1.83
5.49
3.20
3.51
Stack
exit
velocity
(•/sec)
13.00
10.78
15.00
13.55
28.50
13.55
15.70
8.40
15.20
22.90
14.20
13.55
13.55
13.55
5.00
13.55
13.55
13.55
14.00
12.30
9.00
1.10
23.80
13.55
Stack
gas
temperature
(Kelvin)
462
445
338
425
443
425
454
333
449
452
464
425
450
331
425
425
425
450
502
336
472
373
466
425
a from Van Hook (1989).
Information not available - stack parameters assigned are an average
calculated from data provided by Van Hook (1989).
7-4
-------
8925H
Table 7-3. Dloxln TEQ and Furan TEQ Emission Rates Based on Reported Stack Gas Concentrations
Hill
Lewiston
Texarkana
Claiborne
Hinckley
Jackson
Mobile
Pine Bluff
Cloquet
Moss Point
Georgetown
Houston
West Point
Longview
Tacomab
Ketch ikan
Sitkab
Everett
HoquiW"
port Angeles
Westbrook
Roaring Springs0
Operating
hours per
a
year
8.400
6.915
7.990
7.942
8.500
7,942
8.496
8,430
8.640
8,400
8.616
7.942
8,520
8,040
7,942
7.942
7,942
8.712
3,000
7.345
8,400
8.568
7.942
Stack gas
volume8
(acm/min)
10.328
4.673
9.057
8,202
13.800
8.202
23.442
4.953
3.541
6,860
17,999
8.202
9.204
9.628
8.202
8.202
8,202
5.700
3.000
4.248
1.536
11.470
8.202
Dloxln TEQ
emission rate
(kg x 10"8/yr)
19.
7.0
16.
15.
2.5
1.5
42.
9.1
6.5
12.
34.
15.
17.
17.
15.
15.
15.
11.
4.4
6.7
2.7
21.
15.
Furan TEQ
emission rate
(kg x 10'8/yr)
12.
4.4
9.9
9.5
16.
9.5
27.
5.8
4.2
7.9
21.
9.5
11.
11.
9.5
9.5
9.5
6.8
2.8
4.3
1.7
13.
9.5
a From van Hook (1989).
b stack Parameter data not available - emission rates assigned are an average of rates
calculated.
7-5
-------
8925H
Table 7-1. Dioxin/Furan Emissions and Stack Flow Rate Date Used to Calculate
Emission Rates3 Based on Stack Gas Concentrations
I saner Concentration (na/dsem)
I saner
Z.3.7.8-TCDD
1.2.3.7,8-PeCDD
1.2.3.4,7.8-HxCOO
I.2.3.6,7.8-HxCDD
1,2,3.7,8,9-HxCDD
1.2.3.4.6,7.8-HpCDD
2.3.7.8-TCDF
1.2.3.7.8-PeCDF
2,3,4.7,8-PeCDF
l.2.3.4.7.8-HxCOF
1,2.3,6.7,8-HxCDF
1,2,3.7.8.9-HxCDF
2.3,4.6.7.8-HxCDF
1.2,3.4,6.7,8-HpCDF
1.2,3.4.7,8,9-HpCOF
Stack Flow Rate Data
Dry
standard
condition
68.130 dscfm
1,927.4 dscfm
Toxic Equivalency Factor
Run 1
(max)
<0.021
<0.016
<0.03S
0.16
0.10
0.71
0.043
<0.024
<0.039
0.065
0.080
0.089
<0.024
0.63
0.22
s
=
Dloxins/furans stack gas
dloxins
furans
Average
(nq/acm)
0.0187
0.0147
Run 2 Run 3
<0.0054 <0.010
<0.0067 <0.015
<0.0085 <0.0095
<0.010 <0.0075
<0.016 <0.011
0.488 <0.0024
<0.0050 <0.011
0.0063 <0.0086
0.0067 <0.0066
0.015 <0.013
0.021 <0.012
<0.0096 <0.016
<0.0076 <0.0090
0.078 0.037
0.026 0.0097
Actual
condition
124.940 acfm
3.538.3 acmn
concentrations
Maximum
(nq/acm)
0.0358
0.0228
Average
0.0121
0.0126
0.0177
0.0592
0.0423
0.4001
0.0197
0.0130
0.0174
0.031
0.0377
0.0382
0.0135
0.2483
0.0852
Toxic
equiv.
factor
1.0
0.5
0.1
0.1
0.1
0.01
0.1
0.05
0.05
0.1
0.1
0.1
0.1
0.01
0.01
a Research Triangle Institute (RTI 1989) used the API data (Van Hook 1989) to
create this table.
7-6
-------
89Z5H
Table 7-5. Z.3,7,8-TCDO and 2.3,7,8-TCDF Emission Rates Based on Site-Specific Concentrations in Sludge
Hill
lewiston
Texarkana
Claiborne
Hinckley
Jackson
Mobile
Pine Bluff
Cloquet
Moss Point
Georgetown
Houston
West Point
Longview
Tacoma
Ketchikan'
Sitka
Everett
Hoquiam
Port Angeles
Westbrook
Roaring Springs
Sludge
burned3
(t/yr)
11.680
47.046
9.379
27,025
6.961
19.173
37,237
10.907
2,333
16.736
33.095
17.663
5.475
12.045
14.600
26.280
11.526
2.615
17.800
19.749
11.700
2,3.7.8-TCOD
conc.a
(ppt)
78
86
81
36
18
9.5
185
5
161
62
106
14
35
34
3.5
4.7
46b
4.8
47
13
5
2,3.7.8-TCDF
cone . a
(PPt)
639
1.000
313
128
169
18
2.940
Z5
1.020
161
144
77
375
98
366b
42
72
25
65
55
113
2.3,7.8-TCDO
emission rate
(kgx!0'3/yr)
0.827
3.674
0.690
0.883
0.107
0.165
7.255
0.050
0.341
0.942
3.185
0.225
0.174
0.372
0.046
0.112
0.481
0.011
0.760
0.233
0.053
2,3.7,8-TCDF
emission rate
(kgxlO'3/yr)
6.777
42.718
3.311
3.141
1.068
0.313
99.404
0.247
2.161
2.447
4.327
1.235
1.864
1.072
4.852
1.002
0.754
0.059
1.051
0.986
1.201
a From 104-Hill Study (USEPA 1988) and Van Hook (1989).
b information not available - emission rates assigned are an average of rates calculated.
7-7
-------
incinerated at each pulp and paper mill, concentrations of 2,3,7,8-TCDD
and 2,3,7,8-TCDF in sludge and estimated emission rates for 2,3,7,8-TCDD
and 2,3,7,8-TCDF. The emission rates presented in Table 7-5 were
estimated by multiplying the amount of sludge burned at each mill by the
concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF reported to be present in
each mill's sludge. For those pulp and paper mills for which wastewater
sludge concentratons were not available at the time of this assessment,
the average emission rate for all other pulp and paper mills incinerating
sludge was used. Although in this method it was assumed as a worst case
that all 2,3,7,8-TCDD and 2,3,7,8-TCDF present in pulp and paper mill
wastewater sludge was emitted to the atmosphere, the potential for
secondary formation of CDDs/CDFs as products of incomplete combustion was
not considered.
7.2.4 Risk Calculations
Aggregate risk was calculated by multiplying the total exposure
produced by HEM (for a single source, a category of sources, or all
categories of sources) by the unit risk estimate. The product was cancer
incidences among the population exposed for 70 years. The total
exposure, as calculated by HEM, is illustrated by the following equation:
N
Total Exposure = X (PiC-j)
1-1 (7-1)
where
£ = Summation over all grid points where exposure is calculated
P^ = Population associated with grid point i
Cj = Long-term average 2,3,7,8-TCDD/TCDF concentration at grid
point i
N = number of grid points to 2.8 kilometers and number of block
group/enumeration district (BG/ED) centroids within 2.8 and
50 kilometers of each source.
Individual risk, expressed as "maximum lifetime risk," was calculated
by multiplying the highest concentration to which the general population
is exposed, as reported by HEM, by the unit risk estimate. The product,
a probability of getting cancer, applies to the number of people which
HEM reported as being exposed to the highest listed concentration. The
concept involved a simple proportioning from the 1 jig/nr on which
the unit risk estimate was based to the highest listed concentration. In
other words:
Maximum Lifetime Risk m The Unit Risk Estimate
Highest Concentration to * , M/_3
which People are Exposed l "9/m
7-8
1599q
-------
7.3 Results
Tables 7-6 through 7-9 present the total number of people encompassed
by the exposure analysis, the total exposure, and the highest concentra-
tion to which anyone is predicted to be exposed. This means that exposure
results were not reported for each incinerator, but were reported by
source category (all 21 incinerators in the analysis). The tables separ-
ately show the results for 2,3,7,8-TCDD or dioxin TEQs, and 2,3,7,8-TCDF
or furan TEQs using the two methods of estimating emissions. Table 7-10
presents the combined estimated dioxin/furan risks for each incinerator.
7,4 Analytical Uncertainties
7.4.1 The Unit Risk Estimate
The uncertainties associated with evaluating the human hazard poten-
tial of 2,3,7,8-TCDD and 2,3,7,8-TCDF are discussed in Chapter 3 of this
Integrated Assessment.
7.4.2 Emission Estimates
Literature reviews suggest that chlorinated dibenzo-p-dioxin/
chlorinated dibenzofuran (CDD/CDF) formation and destruction are affected
in the following ways:
• Chlorine must be available to produce CDDs/CDFs.
• Organic compounds capable of being transformed to rings must be
present.
• Temperature must be sufficient to initiate and carry out the
formation reaction to an appreciable degree within the residence
time of the combustion and post-combustion process.
• Temperatures above a certain level will destroy CDDs/CDFs and
their precursor compounds.
• CDD/CDF formation probably takes place on surfaces with active
reaction sites (e.g., fly ash) rather than in the gas phase.
• Competing species, such as sulfur compounds, can occupy reaction
sites and reduce CDD/CDF formation.
• Release of CDD/CDF molecules from reaction sites to the gas phase
is temperature-dependent.
7-9
-------
89Z5H
Table 7-6. Site-Specific Exposure Analysis Based on 2.3,7,8-TCDD
Concentrations in Sludge (Maximum Radius = 50 km)
Annual
average
concentration
(ug/m3xlO~8)
1.00
0.5
0.25
0.1
O.OS
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
emulative
popu 1 at ion
exposed
(persons)*
8
521
5,820
102.000
213.000
556.000
2.620,000
3,790,000
5.670,000
7,220.000
7,830.000
8.310,000
8.640,000
8,680,000
8,690,000
Cumulative
exposure
(persons
ug/m3)b
0.0000000857
0.00000343
0.0000195
0.000153
0.000235
0.000345
0.000654
0.000743
0.000810
0.000836
0.000840
0.000842
0.000843
0.000843
0.000843
a This column displays the computed value rounded to the nearest whole
number of the cumulative number of people exposed to the matching and
higher concentration values found in the first column.
The last column displays the computed value of the cumulative exposure
to the matching and higher concentration values found In the first
column.
7-10
-------
8925H
Table 7-7. Site-Specific Exposure Analysis Based on 2,3.7.8-TCDF
Concentrations in Sludge (Maxlmun Radius - 50 km)
Annual
average
concentration
(ug/m3xlO~7)
1.00
0.5
0.25
0.1
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
Cumulative
population
exposed
(persons)9
219
545
16.000
122,000
199,000
293,000
434.000
723,000
1,470,000
4,520.000
6,610.000
7,800.000
8,480.000
8.60G.OOO
8,690.000
Cumulative
exposure
(persons
ug/m3)b
0.0000262
0.0000537
0.000537
O.OOZ24
0.00280
0.00315
0.00337
0.00358
0.00383
0.00431
0.00446
0.00450
0.00452
0.00452
0.00452
4 This column displays the computed value rounded to the nearest whole
number of the cumulative number of people exposed to the matching and
higher concentration values found in the first column.
b The last column displays the computed value of the cumulative exposure
to the matching and higher concentration values found In the first
column.
7-11
-------
8925H
Table 7-8. Site-Specific Exposure Analysis Based on Dioxin
Concentrations in Stack Gas (Expressed as dicxin
TEOs) - Maximum Values (Maximum Radius = 50 km)
Annua 1
average
concentration
(ug/m3x!0~12)
3.51
2.50
1.00
0.5
0.25
0.1
0.05
0.025
0.01
0.005
O.OOZ5
0.001
0.0005
Cumulative
population
exposed
( persons )a
9
60
481
11.900
43,000
109,000
323,000
896,000
3,250,000
6,690,000
8,420.000
8,680,000
8.690,000
Cumulative
exposure
( persons
ugV)b
IxlO'8)
0.00320
0.0161
0.0783
0.831
1.91
2.90
4.36
6.40
9.90
1Z.4
13.0
13.1
13.1
a This column displays the computed value rounded to the nearest whole
number of the cumulative number of people exposed to the matching and
higher concentration values found In the first column.
The last column displays the computed value of the cumulative exposure
to the matching and higher concentration values found in the first
column.
7-12
-------
89Z5H
Table 7-9. Site-Specific Expsoure Analysis Based on Furan
Concentrations in Stack Gas (expressed as dioxin
TEQs) - Maximum Values (Maximum Radius = 50 km)
Annua 1
average
concentration
(ug/m3xlO~1Z)
3.27
1.00
0.5
0.25
0.1
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
Cumulative
population
exposed
(persons)3
9
119
2.900
22.200
65.000
169.000
525.000
1.690.000
4.350.000
7.630.000
8.630,000
8.690.000
Cumulative
exposure
(persons
ug/m3)b
(xlO'8)
0.00207
0.0174
0.182
0.827
1.54
2.22
3.44
5.18
7.05
8.26
8.46
8.47
a This column displays the computed value rounded to the nearest whole
number of the cumulative number of people exposed to the matching and
higher concentration values found in the first column.
The last column displays the computed value of the cumulative exposure
to the matching and higher concentration values found In the first
column.
7-13
-------
B925H
Table 7-10. Contained Oioxin/Furan Risks and Annual Incidence3
Maximum individual risk
Annual Incidence
Pine fl luff
Port Angeles
Longview
Texarkana
Everett
Lew is ton
Houston
Noss Point
Uestbrook
Gsorgctown
Ketchikan
Roaring Springs
TacoH
West Point
Hinckley
Claibome
Jackson
Cloquet
Mobile
Hoquia*
Sitka
Sludge
cone, method
(xlO'7)
9.3
3.1
2.5
l.B
1.4
1.4
1.2
0.95
0.16
0.15
0.15
0.12
0.11
0.056
0.049
0.049
0.039
0.038
0.025
0.021
0.012
Stack gas
•ethod
IxlO'7)
0.000055
0.00026
0.0012
0.000042
0.00029
0.000053
0.000055
0.000049
0.000021
0.000044
0.000045
0.00011
0.000037
0.000029
0.000011
0.000012
0.000029
0.0000035
0.000057
0.000056
0.0000091
Sludge
cone. method
0.00015
0.000013
0.000027
0.000098
0.000042
0.0000097
0.0002
0.000012
0.000014
0.0000016
<0. 000001
0.0000015
0.000034
<0. 000001
0.0000027
<0. 000001
<0. 000001
<0. 000001
0.0000023
<0. 000001
«0. 000001
Stack gas
•ethod
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
<0. 000001
«0. 000001
<0. 000001
<0. 000001
<0. 000001
* Risks presented Mere calculated using the EPA unit risk estimate for 2.3,7.8-
TCDD and the TEQ sethod. This unit risk estimate was derived using the EPA
carcinogenic potency estimate for 2.3,7.8-TCDO (1.56 x 10~4 (po/kg/d)'1). Had
risks been calculated using FDA's potency estimate (1.75 x 10~5 (pg/kg/d)"1).
then the risks and incidences would be a factor of 8.9 lover than those pre-
sented in the table. Had risks been calculated using CPSC's potency estimate
(6.7 x 10"5 (pg/kg/d)'1). then the risks would be "at least* a factor of
2.3 lower than those presented in the table. The tens "at least" is used be-
cause, as discussed in Section 3.3 of this report, CPSC does not place the same
emphasis on risks calculated by the TEQ Bathed as it does for 2.3.7,8-TCOD
itself when estimating carcinogenic potency.
7-14
-------
• Overall formation of CDDs/CDFs is probably controlled by the
number of active sites available given a conductive time/
temperature profile.
• Gaseous and particulate matter control devices after combustion
sources may or may not be effective in removing or preventing
formation of CDDs/CDFs after the combustion zone but prior to
stack entry. Wet processes (scrubbers and spray drier/baghouse
combinations) are probably more effective.
The above information suggests that, given the chemical constituents,
CDD/CDF formation is maximized in processes that supply particulate
matter for reaction sites and occur at moderate temperatures, in short,
inefficient combustion. The two major sources of inefficient combustion
should be open fires and poorly designed or operated boilers, furnaces,
and incinerators.
7.4.3 Sludge Burning at Municipal Incinerators
It is known that at least one pulp and paper mill sends primary
sludge to a municipal incinerator for incineration. However, OAQPS has
no information on dioxin/furan emissions from incineration of pulp and
paper sludges off-site at municipal incinerators.
7.4.4 Exposure Assumptions
(1) The public. The following are relevant to the public as dealt
witn in this analysis:
• The basic assumptions implicit in the methodology are that all
exposure occurs at people's residences, that people stay at the
same location for 70 years (approximate human lifetime) to
maintain consistency with URE (lifetime annual studies), that the
ambient air concentrations and the emissions which cause these
concentrations persist for 70 years (URE), and that the
concentrations are the same inside and outside the residences.
From this, it can be seen that public exposure is based on a
hypothetical premise.
. The people dealt with in the analysis are not located by actual
residences. As explained previously, people are grouped by census
districts and these groups are located at single points called the
population centroids. The effect is that the actual locations of
residences with respect to the estimated ambient air concentrations
are not known and that the relative locations used In the exposure
model may have changed since the 1980 census.
7-15
J5991
-------
(2) Ambient air modeling. The following are relevant to the esti-
mated ambient air concentrations of dioxins/furans used in this analysis:
• Flat terrain was assumed in the dispersion model.
Concentrations much higher than those estimated would result if
emissions impact on elevated terrain near a plant.
• The estimated concentrations do not account for the additive
impact of emissions from plants located close to one another.
Several incinerators in Washington State are located such that
their concentration distributions overlap, e.g., they are sited
within 50 kilometers of each other. This will not increase
exposure significantly because they are not sited extremely close
together, and the additional contribution to those exposed to
emissions from more than one incinerator would be extremely small.
• The increase in concentrations that could result from re-entrain-
ment of dioxins/furans dust from, e.g., city streets, dirt roads,
and vacant lots, is not considered.
• Meteorological data specific to plant sites are not used in the
dispersion model. As explained, HEM uses the meteorological data
from the STAR station nearest the plant site. Site-specific
meteorological data could result in significantly different
estimates, e.g., the estimated location of the highest
concentrations.
• The dioxin/furan emission rates are estimates that are based on
assumptions rather than on measured data.
7.4.5 Conclusions
Population risks from inhalation of dioxin/furan emissions from
incineration of pulp and paper sludge were estimated by two methods. The
estimated risks appear to be low for both. These two methods were used
to assess risks for several reasons. Relevant, although limited, data
were used in each. Although the stack gas concentration data were more
limited (stack gas concentrations were available only at one stack) than
the concentration of 2,3,7,8-TCDD/TCDF in sludge, it was probably more
relevant since destruction efficiency, partitioning to fly ash and bottom
ash, and distribution of CODs/COFs in gaseous and particulate phases did
not have to be characterized. Uncertainties regarding these parameters
are important for the sludge concentration method. The major problem
associated with the stack gas method was, of course, the limited data.
Questions concerning secondary formation and effectiveness of control
devices, particularly if a significant fraction of the dioxins/furans are
in a gaseous phase, are also important for the sludge concentration
method.
7-16
1599q
-------
OAQPS contends that the first assessment method (i.e., the stack gas
method) provides a better estimate of the performance of the boilers used
at the 21 pulp and paper mill facilities that incinerate sludge than the
worst-case estimates from the second method which assumes no destruction
of CDDs/CDFs in the sludge feed. The sludge charged to the power boiler
for which stack gas concentration data are available uses a high sludge
feed content (10 to 15 percent of feed) relative to the feed content used
by the other facilities (2 to 17 percent) and therefore may represent an
overestimation of typical emissions. The second method (i.e., the sludge
concentration method), although more conservative since no destruction of
CDDs/CDFs is assumed, does not account for secondary formation of
CDDs/CDFs as products of incomplete combustion.
7.5 References
Research Triangle Institute. 1989. Second study of combustion
sources of polychlorinated dibenzo-p-dioxins and polychlorinated
(jibenzofurans to support risk assessment. Draft report. Prepared for
U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC. September 1989.
USEPA. 1984. U.S. Environmental Protection Agency. Health assessment
document for polychlorinated dibenzo-p-dioxins. U.S. Environmental
protection Agency, Office of Research and Development. EPA 600/8-84-014F.
USEPA. 1986. Environmental Protection Agency. User's manual for the
Human Exposure Model (HEM). U.S. Environmental Protection Agency, Office
JJf Air Quality Planning and Standards. EPA 450/5-86-001.
USEPA. 1988. U.S. Environmental Protection Agency. U.S. EPA/Paper
Industry Cooperative Dioxin Study. Data submittals during 1988 and 1989
to EPA, Office of Water Regulations and Standards.
Van Hook MB. 1989. Letter from M.B. Van Hook, American Paper Institute,
to Dwain Winters, USEPA, Office of Toxic Substances. August 11, 1989.
7-17
-------
8 ASSESSMENT OF CANCER RISK FROM EXPOSURE TO PCDDs AND PCDFs IN
CONSUMER PRODUCTS
gej Introduction
Exposure to 2,3,7,8-TCDD and 2,3,7,6-TCDF in consumer products may
potentially occur by means of dermal absorption (ADL 1987, NCASI 1987).
That is, when paper products come into contact with the skin, dioxin in
the paper may migrate to the skin surface where it may subsequently be
absorbed. The risk of dioxin exposure may be greater when the product is
contacted by liquid, as may occur with disposable infant diapers, because
the liquid may enhance the transfer of dioxin to the skin. This chapter
-js a condensed version of the following report prepared by the Consumer
product Safety Commission (CPSC) as part of the Interagency
pioxin-in-Paper Workgroup:
Babich MA. 1989. CPSC staff assessment of the risks to human health
from exposure to chlorinated dioxins and dibenzofurans in paper
products. Memorandum from Dr. Michael A. Babich (CPSC) to Lois Dicker
(EPA/OTS). January 25, 1990.
The paper products considered in Babich (1989) were limited to prod-
ucts under CPSC jurisdiction. They were divided into product categories,
depending on the assumed exposure mechanism. Exposure was assumed to
occur by means of either liquid mediated exposure or dry contact
(explained under Subsection 8.2). Products involving liquid-mediated
absorption included: disposable infants diapers, paper towels, facial
tissue, and toilet tissue. Paper towels were further divided into two
scenarios, drying hands and household cleaning. Exposure by means of
contact with dry paper included dinner napkins and communication paper
(i.e., uncoated sheets), such as bond paper, books, magazines, and news-
print, except for coated sheets. Exposure to communications paper was
assessed for exposures at home and in school.
Qt2 Methodology
g.2.1 Exposure Assessment
(1) Assumptions common to all exposure scenarios. For the purpose
Of exposure assessment, dermal exposure was treated as a two-step process:
/I) migration or extraction of dioxin from paper or pulp into a liquid
contacting the skin or to the surface of the skin itself, followed by
12) percutaneous absorption. The migration step may occur by either of
two general mechanisms, liquid-mediated extraction or skin contact with
paper (unmediated diffusion).
(a) Percutaneous absorption. Estimates of the rate and extent
Of dermal absorption for use 1n exposure assessment were based on prelimi
8-1
16001
-------
nary data with human skin In vitro (Weber et al. 1989). These experiments
were done using both acetone and mineral oil as vehicles and using both
intact and "damaged" skin.
Measures of dermal exposure included the rate of absorption and the
extent of absorption at a given time. The exposure model used and the
exposure data which were available determined which measure would be
required. The rate of absorption was given as the transfer coefficient,
defined by:
J = kA (8-1)
where J is the flux (ug/cm2/M, k is the transfer coefficient (/h), and
A is the specific dose (ug/cmz) (Scheuplein and Ross 1974). The extent
of absorption was expressed as the percent of the applied dose which is
absorbed by 24 h post exposure. Parameters for dermal absorption were
based on solvent deposition experiments, i.e., experiments where volatile
solvents such as methanol or acetone were used as vehicles.
On average, 18.5 percent of 2,3,7,8-TCDD was absorbed by 17 h post
exposure (estimated from Weber et al. 1989). This estimate is based on
the assumption that 2,3,7,8-TCDD in both the dermis and epidermis is
available for systemic absorption. With a transfer coefficient k of
0.012 h"1, the fraction of 2,3,7,8-TCDD estimated to be absorbed at
24 h post exposure is -0.25. For the purpose of risk assessment, it was
assumed that 2,3,7,8-TCDD and 2,3,7,8-TCDF are absorbed at a rate of
0.012 h"1 and that 25 percent is absorbed in 24 hours.
(b) Concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF. The
limited data on 2,3,7,8-TCDD and 2,3,7,8-TCDF levels in consumer products
which are available are summarized in Table 8-1. Some of the data are for
European products and, therefore, cannot be used to assess the risk to
U.S. consumers. However 2,3,7,8-TCDD and 2,3,7,8-TCDF levels were meas-
ured in bleached pulp at all 104 bleached kraft and bleached sulfite proc-
ess paper mills in the U.S. (USEPA 1988a) In the absence of additional
data, dioxin levels in pulp were used as surrogates for product levels.
Since information regarding the mill sources used in manufacturing specif-
ic products is, in general, unavailable, the average 2,3,7,8-TCDD and
2,3,7,8-TCDF levels over all the mills were used (8.6 ppt for 2,3,7,8-TCDD
and 103.6 ppt for 2,3,7,8-TCDF). In calculating these averages, non-
detects were averaged as one-half the detection limit.
(2) Disposable Infant diapers. Disposable Infant diapers consist of
both core components and ancillary components (Mizutani 1987). Core com-
ponents are: a top sheet of non-woven polypropylene which contacts the
skin; fluff pulp fiber, which may be thicker in the middle; super-
absorbent diapers contain absorbent gelling material (AGM), a crosslinked
polyacrylate gel, mixed into the fiber; and a polyethylene back sheet.
8-2
IBOOq
-------
89ZOH
Table 8-1. Concentrations (ppt) of 2,3.7,8-TCDO and
2.3,7,8-TCDF In Consumer Products
Product
2.3.7.8-TCDD 2.3.7,8-TCDF
Reference
Disposable diapers
Paper towels
Bond paper
Facial tissue
Scrap paper
Newsprint
Tissue
NO (2.6)
NO (2.1)
NO
3.7
13
13
1.1
0.6
ND
1.3
8.8
7.2
3.7
32
290
240
13
13
ND
31.1
NCAS1 (1987a); Blosser (1987)
NCASI (1987a); Blosser (1987)
WJLA TVa
NCASI (1987)
NCASI (1987)
NCASI (1987)
Beck et al. (1988)
Beck et al. (1988)
Beck et al. (1988)
LeBel et al. (1989)
a Assays for VJLA TV were performed by Triangle Laboratories, Research Triangle
Park. NC.
ND. nondetectable; the number In parentheses Is the detection limit.
8-3
-------
Ancillary components include the fasteners, elastic, and adhesive. AGM
has a higher affinity for liquids than fluff pulp. The total capacity of
a superabsorbent diaper is -300 g, depending on its ability to distribute
the moisture throughout the core. Absorbed liquid may also be released
from both the pulp and AGM under pressure, such as occurs when the child
sits.
(a) Exposure and risk calculations. Exposure was assumed to be
a 2 step process. First, dioxins in the pulp (solid, or adsorbed phase)
are extracted into urine (aqueous phase). Aqueous phase dioxins are then
available for absorption by the skin. Dermal absorption is assumed to be
the rate limiting step. Extraction of dioxins from the pulp is assumed to
be a rapid process relative to dermal absorption and quickly approaches a
steady state.
Infants were assumed to use an average of six diapers per day (includ-
ing four daytime, one naptime, and one overnight diaper) for three years.
The average daily dose (ADD) and the lifetime average daily dose
(LADD) (pg/kg-d) were calculated as follows:
ann
ADD
kHCHFaUFbHFc) UfUsdUTdl +
(B)(K)
fUsnHTn) + (UsoHTon
,Q ,*
(8-2)
where
k
C
Fa, Fb, and Fc
Usd, Usn, and Uso
Td, Tn, and To
B
K
the rate of dermal absorption;
the initial dioxin concentration in the pulp;
correction factors to account for such conditions
as anatomical site, damaged skin, and age,
respectively;
the amounts of urine in skin contact for daytime,
naptime, and overnight diapers, respectively;
the exposure durations for daytime, naptime, and
overnight diapers, respectively;
the average body weight; and
the urinerpulp equilibrium partition coefficient.
BfkUCUFaHFbUFcl
where
3
70
UfUsdHTdl
70(B)(K)
- the years of exposure; and
- the average life expectancy.
+ (UsnHTnl + fUsoWToll
(8-3)
Because different partition coefficient values are used for 2,3,7,8-
TCDD and 2,3,7,8-TCDF, ADD and LADD must be calculated separately for
each. Then the total ADD and total LADD in terms of TEQs are calculated.
8-4
1600q
-------
The dioxin TEQ concentration in diapers which would result in a
lifetime risk of 1 x 10"5, or the "risk specific dioxin concentration",
Crs, was given by:
_ 70fBHKHRsdl _ ,ft 4,
Crs = 3(k)(Fa)(FbJ(Fc) [4(Usd)(Td) + (Usn)(Tn) + (UsoJ(To)] {*~*>
where:
Rsd - the risk specific dioxin TEQ dose (i.e., the dose at which the
lifetime cancer risk is 1 x 10"").
K = partition coefficient for 2,3,7,8-TCDD, in this case.
(b) Parameters. The values of the parameters from which expo-
sure and risk are estimated are given in Table 8-2. The rationale for the
values assigned to each parameter is presented in detail in Babich (1989).
In general, the parameters are based on infants wearing medium size
diapers, i.e., infants from 12 to 24 pounds. It is assumed that these
represent the average or typical infant in the age range from birth to
3 years, the time when infants are assumed to be in diapers (ADL 1987;
NCASI 1987).
Certain parameters such as urine load were specific for the time of
day. The abbreviations for these parameters were distinguished by the
use of the suffixes d, n, and o for day, nap, and overnight, respectively.
(3) Paper towels, facial tissue, and toilet tissue. Exposure is a
two-step process: (1) liquid-mediated extraction of dioxin from the
paper or pulp, followed by (2) dermal absorption. The extraction
step (Step 1) was assumed to be rapid, i.e., it was assumed to reach
equilibrium, even though exposures for these products were brief. How-
ever, this model differed from the diaper model in that there were two
components to the dermal absorption step (Step 2) (NCASI 1987).
During Step 1, dioxin is extracted from the pulp by a given amount of
liquid phase (L). In general, some portion of the liquid phase is in con-
tact with the skin (Is). During Step 2a» the amount of dioxin dissolved
in Ls is available for dermal absorption, at rate k, for the length of
time that the product and the liquid phase are in contact with the skin
/i.e., the exposure duration). Up to this point, the exposure model is
similar to that for diapers. During Step 2b, it was assumed that a
oortion of Ls remains on the skin for 24 h (Lsr) (NCASI 1987) and that a
fraction, f, of the 2,3,7,8-TCDD and 2,3,7,8-TCDF dissolved in Lsr will
be absorbed.
(a) Exposure and risk calculations. The average daily dose
(ADD)> lifetime average daily dose (LADD), and the risk specific dioxin
concentration (Crs), were calculated as follows:
ADD ' t(k)(Ls)(T) + (f)(Lsr)] (8-5)
B-5
I600q
-------
8920H
Table 8-2. Parameters for Estimating Exposure to 2.3.7,8-TCDO
and 2,3.7.8-TCDF in Disposable Infant Diapers
Parameter Value
c.
k,
K,
B.
Fa.
Fb.
Fc.
Ud.
Un.
Uo.
Ed.
En.
Eo.
Td.
Tn.
To.
Usd.
Usn,
Uso,
2,3,7,8-TCDD concentration (pg/g)
2,3.7.8-TCDF concentration (pg/g)
TEQ concentration (pg/g)
Dermal absorption rate (/h)
Pu1p:urine partition coefficient
2,3.7.8-TCDD
2.3.7.8-TCDF
Average body weight (kg)
Anatomic site correction
Damaged skin correction
Infant skin correction
Urine load, day (n-4) (g)
Urine load, nap (n»l) (g)
Urine load, overnight (n»l) (g)
Wear time, day (h)
Wear time, nap (h)
Wear time, overnight (h)
Wet time, day (h)
Wet time, nap (h)
Wet time, overnight (h)
Urine in skin contact, day (g)
Urine in skin contact, nap (g)
Urine in skin contact, overnight (g)
3a
25a
5.5
0.012
14.300b
6.300b
10
2
1.5
1.5
90C
105C
160C
2.6°
3.6C
10. Oc
0.8C
1.4C
7.0C
d d
Suoerabsorbent Convent iona 1
0.20 0.36
0.28 0.42
0.55 4.8
a Average values. Based on data provided in the 104 Mill Study
(USEPA 1988a) and on data provided by diaper manufacturers directly to
CPSC.
b Coefficients were determined at 32*C using synthetic urine and
Southern softwood fluff pulp (NCASI 1989).
c O'Reilly (1989).
d Based on NCASI (1987) and O'Reilly (1989).
8-6
-------
where
C
N
K
B
k
f
F
Ls
Lsr
T
the concentration of dioxin in the pulp,
the number of units per day,
the pulprliquid partition coefficient,
the average body weight,
the dermal absorption rate,
the fraction of 2,3,7,8-TCDD absorbed by the skin in 24 h,
a correction factor applied to k and f,
the amount of liquid phase in skin contact,
the amount of Ls which remains on the skin for 24 h, and
the exposure duration.
LADD
(cwmmm
(70)(K)(B)
[(k)(Ls)(T) + (f)(Lsr)]
(8-6)
where:
Y
70
the number of years that the product is used, and
the average life expectancy.
Crs
(70URsdUKUBl
[(k)(Ls)(T) + (f)(Lsr)]
(8-7)
where:
Rsd
the risk specific dioxin TEQ dose (i.e., the dose at which the
lifetime cancer risk is IxlO"5).
(b) Parameters. The parameters used in estimating exposure to
in paper towels, facial tissue, and toilet tissue are given in
Tables 8-3 through 8-5, respectively.
(4) Communications paper and paper napkins. The exposure model for
these products differed from the models discussed above, because there is
no lio.1^ Pnase to promote the migration of dioxin from the pulp to the
ckin. Hence, migration (Step 1) occurs by passive diffusion. This was
ssumed to be the rate limiting step. The amount of dioxin which migrates
depends on the exposure duration. A fraction (f) of the dioxin which is
transferred is absorbed by the skin (Step 2).
(a) Exposure and risk calculations. The average dally dose
/ADD), lifetime average daily dose (LADD), and risk specific dioxin TEQ
concentration in the product (Crs), were calculated as follows:
ADD
(rHfHCHPHTHAsHFHm
(Ap)(B)
(8-8)
8-7
-------
89ZOH
Table 8-3. Parameters for Estimating Exposure to 2,3,7.8-TCDD
and 2.3.7,8-TCDF in Paper Towels
Parameter
C.
N.
T,
K,
L,
Is.
Lsr.
k.
f.
F.
B,
f.
Rsd.
As.
2.3,7.8-TCDD concentration (pg/g)
2.3,7,8-TCDF concentration (pg/g)
TEQ concentration (pg/g)
Number of units per day
Contact duration (sec)
Liquid phase surrogate
Partition coefficient:
2,3,7.8-TCDD
2,3,7.8-TCDF
Mass of liquid phase (g)
Liquid in skin contact
Liquid remaining on skin
Dermal absorption rate (/h(
Fraction dermal absorption at 24 h
Correction factor for k and f
Average body weight (kg)
Tears used
Risk specific dose (pg/kg/d)
Skin surface are exposed (cm )
Drying
hands
8.6
103.6
19
5
15a'b
water9
>13.000a
29,000a
15a
1.28
0.6d
0.012
0.25
1
70a'c
70a
0.015
800C
Household
cleaning
8.6
1C3.6
19.0
5
60
8X EtOH*
*2,000a
2.000a
15a
1.2a
0.3d
0.012
0.25
1
70a'c
70a
0.015
400
8 NCASI (1987).
b ADL (1987).
c USEPA (1989).
Assuming 0.1 g per 100 cm2 of exposed skin surface.
8-8
-------
8920H
Table 8-4. Parameters for Estimating Exposure to Z.3.7.8-TCDD
and 2.3,7,8-TCDF in Facial Tissue
Parameter
C.
N.
T.
K.
L.
Ls.
Lsr.
k,
f.
F.
B.
Y.
Rsd.
As.
2,3,7.8-TCDD concentration (pg/g)
2.3.7.8-TCOF concentration (pg/g)
TEQ concentration (pg/g)
Number of units per day
Contact duration (sec)
Liquid phase surrogate
Partition coefficient:
2.3.7.8-TCDO
2.3.7.8-TCDF
Mass of liquid phase (g)
Liquid in skin contact
Liquid remaining on skin
Dermal absorption rate (/h)
Fraction dermal absorption at 24 h
Correction factor for k and f
Average body weight (kg)
Years used
Risk specific dose (pg/kg/d)
Skin surface are exposed (cm )
Normal
use
8.6
103.6
19
6a
20a
saline8
a!4.300e
5.300e
la
0.5a
0.005d
0.012
0.25
2
70a'c
70a
0.015
5
Make-up
removal
8.6
103.6
19.0
la
300a
8X EtOH8
22.0008
2.000*
la
la
0.3d
0.012
0.25
2
65C
50a
0.015
300°
8 HCASI (1987).
b am
f 1QO7>
c USEPA (1989).
Assuming 0.1 g per 100 cm of exposed skin surface.
e HCASI (1989).
8-9
-------
8920H
Table 8-5. Parameters for Estimating Exposure to 2.3.7,8-TCDD
and 2.3,7.8-TCOF in Toilet Tissue
Parameter Male Female
c.
N.
T,
K.
L.
Ls,
Lsr.
k.
f.
F,
"B.
Y,
Rsd,
As.
2,3,7.8-TCOD concentration (pg/g)
2.3,7,8-TCDF concentration (pg/g)
TEQ concentration (pg/g)
Nunber of units per day
Contact duration (sec)
Liquid phase surrogate
Partition coefficient:
2.3.7,8-TCDD
Z.3.7.8-TCDF
Mass of liquid phase (g)
Liquid in skin contact
Liquid remaining on skin
Dermal absorption rate (/h)
Fraction dermal absorption at 24 h
Correction factor for k and f
Average body weight (kg)
Years used
Risk specific dose (pg/kg/d)
Skin surface are exposed (cm )
8.6
103.6
19.0
30a
10a
urine3
2l4.300e
6,300e
la
la
0.025d
0.012
0.25
2
75C
70*
0.015
50
8.6
103.6
19.0
60a
10a
urine9
>14,300e
6.3006
la
la
0.05d
0.012
0.25
2
65C
70*
0.015
100
8 HCASI (1987).
b AOL
(1987).
c USEPA (1989).
Assuming 0.1 g per 100 on of exposed skin surface.
e NCASI (1989).
8-10
-------
where:
C
P
r
f
T
As
Ap
N
B
F
the concentration of dioxin In the pulp,
the mass of pulp in the product,
the rate of transfer of 2,3,7,8-TCDD from the product to skin,
the fraction of 2,3,7,8-TCDD absorbed by the skin in 24 h,
the exposure duration,
the skin surface area contacted by the product,
the surface area of the product (one side;,
the number of units per day,
the average body weight, and
correction factor.
LADD
(rHfHCHPHTUAsUFHNHYl
(70)(Ap)(B)
(8-9)
where:
Y
70
the number of years that the product is used,
the average life expectancy.
and
Crs
(70URsdUApHBl
(r)(f)(P)(T)(As)(F)(N)(Y)
(8-10)
where
Rsd
the risk specific dioxin TEQ dose.
(b) Parameters. The parameters used in estimating exposure to
Hioxin in communications paper and dinner napkins are summarized in
Tables 8-6 and 8-7, respectively. The surface area contacted by dinner
napkins was assumed to be equal to the area of one side of both hands.
a 2.2 Risk Assessment
/I) ^dividual cancer risk. The individual cancer risk, that is
tne average lifetime excess cancer risk (R), was calculated by:
(8-11)
D 11 flnnwnifhurcan absorption fraction 1
R » lLAUUnq)|an.mal absorption fraction]
LADD is the lifetime average daily dose (pg/kg/d), calculated from
ions 8-3, 8-6,,or 8-9. The carcinogenic potency estimate, Q (in
"nUs of (pg/kQ/d)-1), for 2,3,7,8-jTCDD is 6.7 x 10's (CPSC Babich
iQ88), 1.8xlO'5 (FDA), and 1.6xlO'4 (EPA). The "human absorption
fraction" is assumed to be 1.0 for the dermal exposures assessed here.
The "animal absorption fraction" 1s an estimate of the absorption which
curred during the animal experiments from which the carcinogenic potency
8-11
-------
8920H
Table 8-6. Parameters for Estimating Exposure to Z.3.7.8-TCDD
and 2.3,7,8-TCDF in Communications Paper
Parameter Hone School
Product parameters
C, 2.3,7.B-TCDD concentration (pg/g) 8.6 8.6
2.3.7.B-TCDF concentration (pg/g) 103.6 103.6
Toxic equivalents (pg/g} 19.0 19.0
Mass of product (g) 4.5a 4.5a
X Pulp 90b 90b
P. Mass of pulp (g) 4.05 4.05
Ap. Surface area (cm2) 603*>b 603a
-------
89ZOH
Table 8-7. Parameters for Estimating Exposure to 2.3.7.8-TCDD
and 2.3.7.8-TCDF in Paper Napkins
Product parameters
C. Z.3.7.8-TCDD concentration (pg/g) 8.6
2.3,7,8-TCDF concentration (pg/g) 103.6
Toxic equivalents (pg/g) 19.0
Mass of product (g) 9.5a
X Pulp 98a
P. Mass of pulp (g) 9.3
Ap. Surface area (cm2) 1,865*
N, Number of units/day 3a
Subject parameters
T,
As.
r.
f.
F.
(fed.
B.
r.
Contact duration (mln)
Skin surface area contacted (on2)
Rate of TCDD transfer to skin (/h)
Fraction dermal absorption
Correction factor
Risk specific dose (pg/kg-d)
Average body wt. (kg)
Years used
Za
300
0.0005C
0.25
1
0.015
70a.b
70a
a AOL (1987).
b USEPA (1989).
c Data on migration of tr1s-(Z.3-d1bromopropyl)phosphate (a.k.a.. TRIS)
from dry cloth to rabbit skin Mere used as a surrogate for 2.3.7,8-TCDO
(Ulsamer et al., 1978).
8-13
-------
estimates were derived. CPSC assumed an animal absorption fraction of
0.75 whereas EPA and FDA assumed 0.55 (Farland 1987b).
(2) Population cancer risk. The population risk, i.e., the number
of cancers per year in the U.S. (N), was estimated by:
N = R x P / 70 (8-12)
where:
R = the individual risk,
P - the exposed population, and
70 » the average lifespan.
(3) Non-cancer endooints. The risk of non-cancer adverse effects
is represented by the hazard index (HI):
HI = ADD/HA (8-13)
where ADD is the average daily dose, and HA is the estimated health
advisory level developed by EPA, for the purpose of this assessment, for
protection against liver toxicity (10 days at 10 pg/kg/d) (USEPA 1988b;
Lee 1989). When the HI is less than or equal to 1, risk is assumed to be
absent or, at most, trivial.
8.3 Results
8.3.1 Individual Cancer Risk
The exposures and individual cancer risks estimated to result from
2,3,7,8-TCDD and 2,3,7,8-TCDD in consumer paper products under CPSC
jurisdiction are summarized in Table 8-8. Table 8-8 presents risk
estimates for each product type using the carcinogenic potency estimates
of CPSC, FDA, and EPA. As discussed in Section 3.3 of this report, CPSC
does not place the same emphasis on risks calculated by the TEQ method as
it does for 2,3,7,8-TCDD itself when estimating carcinogenic potency.
Therefore, Table 8-8 presents CPSC risk estimates based on 2,3,7,8-TCDD
alone.
Estimates of individual cancer risk (using CPSC's cancer potency
estimate) range from 10-in-a-trillion for facial tissues (normal use
scenario) to 2-in-a-billion for paper towels. For all products combined,
the individual risk is estimated to be 5 per billion. Use of the EPA
cancer potency factor and TEQ method results in slightly greater risks.
Use of the FDA cancer potency factor and TEQ method results in slightly
lower risks.
8-14
1600q
-------
\a\>\e Vft. \n&m&A\ UteXAie and fopAation Cancer Risks frw 7.3.7 .a-TCBu and 2.3,7.8-1CDF in Consumer Paper Products
oo
i
Product
Superabsorbent Diapers
Conventional Diapers
Paper Towels
(Hand Drying)
Paper Towels
(Cleaning)
Facial Tissue
(NorMl Use)
Facial Tissue
(Makeup Removal)
Toilet Tissue
(Hales)
Toilet Tissue
(Females)
CoMwnication Paper
(Homes)
Coraunication Paper
(School)
Paper Dinner Napkins
2.3.7.8-TCOO
LADD
(pg/kg/d)
2.4xlO~7
1.7xlO~6
7.1xlO'6
2.3xlO~5
1.3xlO~7
7.2xlO~6
3.0xlO'G
l.«xlO~5
S.BxlO'6
6.8xlO"6
2.3xlO"6
TEQ
LADO
(pgAg/d)
6.8xlO"7
S.OxlO"6
l.lxlO"5
S.lxUT5
5.6x10 7
l.SxlO'5
l.lxlO"5
5.2xlO"5
l.3xlO"5
l.SxlO"5
S.lxlO"6
Lifetime
CPSC
2-lxlO"11
(100)
l.SxlO'10
(100)
6.3xlO-10
(100)
2.1xlO'9
(100)
1.2x10 U
(100)
6.4xlO-10
(100)
2.7xlO'i0
(100)
1.2xlO"9
(100)
S.lxlO'10
(100)
e.ixio-10
(100)
2.1X10-10
(100)
individual
EPA
Z.OxlO'10
(35)
1.4xlO~9
(35)
S.lxlO"9
(64)
l.SxlO"8
(45)
t.exiQ-10
(23)
4.6xlO"9
(45)
3.lxlO~9
(22)
l.SxlO"8
(22)
3.7xlO"9
(45)
4.3xlO"9
(45)
1.4xlO'9
(45)
c
cancer risk
FDA
2.2xlO"U
(35)
1.6X10'10
(35)
3.6xlO-10
(64)
l.SxlO"9
(45)
l.BxtO"11
(23)
5.2x10-'°
(45)
3.6xlfl-10
(22)
1.7xlO"9
(22)
4.2X10-10
(45)
4.9x10-'°
(45)
1.7*10'°
(45)
Exposed
population
-
1.0xl07d
2.4xl08
2.4xl08
1.2xl08
1.2xl08
1.2xl08
1.2xl08
2.4xl08
2.4xl08
2.4xl08
Excess cancers per year
CPSC EPA FDA
-
0.00002 0.0002 0.00002
0.002 0.011 0.001
0.007 0.051 0.005
0.00002 0.0003 0.00003
0.001 0.008 0.0009
0.0005 0.005 0.0006
0.002 0.026 0.003
0.002 0.013 0.001
0.002 0.015 0.002
0.0007 0.005 0.0006
-------
8920H
Table 8-8. (continued)
2.3.7.8-TCDO
LAOO
Product (pg/kg/d)
All Products (Hale)3 S.OxlO'5
AH Products (Faule)b 6.8xlO~5
TEQ Lifetim
LAOO
(pg/kg/d) CPSC
l.lxlO'4 4.5x10 9
(IOOJ
l.BxlO'4 6-lxlO"9
(100)
c
e individual cancer risk Excess cancers ner vear
EPA
3.2xlO"8
(45)
4.6xlO"8
(45)
Exposed
FDA population CPSC
3. 5x10 9 1.2xl08 0.008
(45)
5.2xlO~9 l.ZxlO8 0.010
(45)
EPA FDA
0.055 0.006
0.079 0.009
00
i
a Includes all products except superabsorbent diapers, facial tissues (makeup removal), and toilet tissue (females).
Includes all products except superabsorbent diapers and toilet tissue (Males).
0 Calculated using Eqn. 9-12 and each Agency's cancer potency estimate. Umbers in parentheses are the percent of estimated risk due to 2.3.7.8 ICDD.
Assures use only of conventional diapers. Resident population age 0 to 3 years.
-------
8.3.2 Population Cancer Risk
Excess cancer risks in the U.S. population are given in Table 8-8 in
terms of excess cancers per year. Less than one cancer per year is
expected to occur from 2,3,7,8-TCDD and 2,3,7,8-TCDF in consumer paper
products. Additivity was assumed in combining risks from different
products or different scenarios.
8.3.3 Non-Cancer Endpoints
The ADD values for all products are well below the estimated health
advisory level for protection against liver toxicity (10 days at
10 pgA9/d). Tnus» tne hazard index, is much less than one, indicating
that the risk for non-cancer effects is absent or, at most, trivial. ADD
values and hazard index values are given in Table 8-9 for all products.
g^4 Uncertainty Analysis
8.4.1 Liquid Mediated Extraction
K values were used to estimate the concentration of 2,3,7,8-TCDD/F in
a liquid medium (Caq) in contact with the skin. This concentration may
he regarded as an upper limit, since equilibrium conditions are not
necessarily expected to occur during the course of exposure. For
exposures less than 2 h, the extent of migration of 2,3,7,8-TCDD from
oaper may be less than that predicted by equilibrium partition
coeffic1ents» leading to an overestimate of exposure. Any error
introduced by this assumption would be greater for products involving
brief exposures (up to 5 minutes), such as paper towels, facial tissue,
and toilet tissue than for diapers, where the exposure duration is
greater (up to 10 h).
Even when exposures are of sufficient duration for equilibrium to
occur, Caq may still be overestimated by assuming equilibrium
conditions. The steady-state value of Caq depends in part on the rate at
wnich aqueous phase TCDD is removed from solution by dermal absorption.
If the rate of dermal absorption were similar to the extraction rate, the
steady state value of Caq would be much smaller than the equilibrium
Ja1ue. However, it was assumed that dermal absorption 1s rate limiting.
Therefore, assuming an equilibrium value for Caq will not lead to a
Substantial overestimate of exposure.
One of the assumptions of the disposable diaper exposure assessment
-is that aqueous phase 2,3,7,8-TCDD is 1n equilibrium with 2,3,7,8-TCDD
bound to the pulp. In other words, as aqueous phase 2,3,7,8-TCDD 1s
removed by dermal absorption, it may be replenished by reextractlon of
the pulp* Although the extraction process is assumed to be continuous,
ic replenishment (as may occur when the child sits, for example)
result in similar exposures.
8-17
-------
8920H
Table 8-9. Risks of Non-Cancer Adverse Effects from 2.3.7.8-TCDD
and 2,3,7,8-TCDF in Conswner Paper Products
Product
Superabsorbent Diapers
Conventional Diapers
Paper Towels (Hand
Drying)
Paper Towels
(Cleaning)
Facial Tissue
(Normal Use)
Facial Tissue
(Makeup Renoval)
Toilet Tissue
(Males)
Toilet Tissue
(Fanles)
Gamin icat ion Paper
(Ho»)
CoMwnication Paper
(School)
Paper Dinner-napkins
2,3.7,8-TCDO
ADOa
(pg/kg/d)
5.3xlO"6
4.0xlO"5
7.1xlO"6
2.3xlO"5
1.3xlO~7
l.OxlO"5
3.0xlO"6
1.4xlO"5
7.7xlO"6
4.0xlO"5
2.3xlO~6
TEQ
ADOa
(pg/kg/d)
l.SxlO"5
1.2xlO"4
l.lxlO'5
S.lxlO"5
5.6xlO"7
2.2xlO"5
l.lxlO"5
5.2xlO~5
1.7xlO"5
8.8xlO"5
S.lxlO"6
2.3.7.8-TCDD
Hazard
Index6
5.3xlO"7
4.0xlO"6
7.1xlO"7
2.3xlO"6
1.3xlO"8
l.OxlO"6
3-OxlO"7
1.4xlO'6
7.7xlO"7
4.0xlO"6
2.3xlO"7
TEQ
Hazard
Indexb
l.SxlO"6
1.2xlO"5
l.lxlO"6
B.lxlO"6
5.6xlO"8
2.2xlO"6
l.lxlO"6
5.2xlO"6
1.7xlO"6
B.SxlO"6
S.lxlO"7
a ADO is the average daily dose.
b The hazard index is the ratio of ADD to the EPA 10-day health advisory of
10 pg/kg/d.
8-18
-------
8.4.2 Unmediated Migration
Like 2,3,7,8-TCDD,,JRIS is a nonvolatile, hydrophobia compound.
Thus, it does not appear unreasonable to employ TRIS as a surrogate for
2,3,7,8-TCDD. Of course, the appropriateness of polyester cloth as a
surrogate for paper is of concern. The basis weight of the cloth (-0.01
q cm'2) is roughly equivalent to that of bond paper. However, the
extent to which the two matrices act as barriers to diffusion may be
different. In addition, the mechanism by which 2,3,7,8-TCDD is adsorbed
to paper, as well as the strength of the interaction, may differ from the
interaction between TRIS and cloth. Thus, it is impossible to be certain
whether the TRIS cloth model adequately represents 2,3,7,8-TCDD exposure.
3.4.3 Use of in vitro Percutaneous Absorption Data
Data obtained from human studies are preferred for use in assessing
human exposure and risk. In addition, the available dermal penetration
data for 2,3,7,8-TCDD with human skin {Weber et al., 1989) have the
advantage that the time course of 2,3,7,8-TCDD absorption was determined
over a.time range (0.5 to 17 h) which is particularly relevant to human
exposure scenarious. The primary disadvantage of the human data is that
they were obtained in vitro. However, In vitro percutaneous absorption
studies have been shown to correlate well with in vivo studies, including
studies with hydrophobic compounds (Bronaugh & Franz, 1986; Bronaugh &
Stewart, 1986). The primary disadvantage of the in vivo studies with
rats is that dermal absorption was generally determined at one time point
only. In addition, the applicability of rat data to human exposure
assessment is uncertain, because human skin is generally considered to be
less permeable than rat skin (based on studies with various penetrants,
but not 2,3,7,8-TCDD; Bartek et al., 1972; Wester & Maibach, 1983a,b).
In consideration of the advantages and disadvantages of both kinds of
studies, the human data were used to assess human exposure.
0.4.4 Factors Affecting Percutaneous Absorption
Numerous factors have been reported to affect either the rate or
extent of percutaneous absorption (Wester and Maibach 1983a,b),
including; anatomic site, occlusion and hydration, the vehicle, diseased
Or damaged skin, drug concentration and skin surface area, multiple or
chronic dosing, and age. Examples of each factor and a discussion of the
extent to which each may influence the rate of 2,3,7,8-TCDD absorption
are discussed below.
(a) Anatomic site. The rate and extent of percutaneous
absorption has been shown to depend on the anatomical site to which the
penetrant is applied (Feldman and Maibach 1967; Maibach et al. 1971).
percutaneous absorption rates are highest on the skin of the genitals
et al. 1980; Hurley 1985). The greatly enhanced permeability of
8-19
1600Q
-------
the genitals should be considered in assessing exposure to 2,3,7,8-TCDD
in disposal diapers. However, the genitals are not the only portion of
the infants' skin which is exposed to urine. Therefore, a correction
factor for anatomic site (Fa) of 2 was used in this assessment for
diapers and toilet tissue.
(b) Diseased and damaged skin (Fb). Diseased or damaged skin
is generally more permeable than normal healthy skin (Wester and Maibach
1983a,b). Thus, it is likely that diaper dermatitis will increase the
permeability of infant skin toward 2,3,7,8-TCDD. If we assume that
diaper dermatitic skin is 5-fold (i.e., 500 percent) more permeable to
2,3,7,8-TCDD than healthy skin, that the incidence of dermatitis is
20 percent (Jordan et al. 1986), and that 50 percent of the urine exposed
skin is dermatitic, then the permeability of diapered skin will average
50 percent or 1.5-fold greater than that of undiapered skin. Thus, a
correction factor for diaper dermatitis (Fb) of 1.5 was assumed in this
assessment.
(c) Correction factor for age (Fc). The skin of preterm
infants is more permeable than that of full term infants during the first
three weeks after birth (Wilson and Maibach 1982; Rutter 1987). It is
unclear, .however, whether the skin of full term infants differs from that
of adults, as the data are contradictory. The degree of barrier function
in infant skin may be specific for certain compounds (McCormack et al.
1982) or for certain classes of compounds (Wilson and Maibach 1982).
However, it is quite possible that the rate of absorption of 2,3,7,8-TCDD
could be greater in infant skin than in adult skin. The difference in
absorption rate between infant and adult skin is likely to be no more
than 10-fold. For the purpose of this assessment, it was assumed that
infant skin is 1.5-fold more permeable to 2,3,7,8-TCDD than adult skin,
i.e., Fc = 1.5.
8.5 References
ADL. Arthur D. Little, Inc. 1987. Exposure and risk assessment of
dioxin in bleached kraft products. Draft report. U.S. Environmental
Protection Agency Contract No. 68-01-6951. June 15, 1987.
Babich, MA. 1988. Unit risk estimate of the carcinogenicity of 2,3,7,8-
tetrachlorodibenzo-p-dioxin [TCDD]. U.S. Consumer Product Safety
Commission. Bethesda, MD.
Bartek, MJ, La Budde OA, Maibach HI. 1972. Skin permeability in vivo:
comparison in rat, rabbit, pig, and man. Journal of Investigative
Dermatology, 58: 114-123.
Beck H, Eckart K, Mathar W, Wittkowski R. 1988. Occurrence of PCDD and
PCDF in different kinds of paper. Chemosphere, 17: 51-57.
8-20
1600q
-------
Blosser RO. 1987. Communication to U.S. Consumer Product Safety
Commission (CPSC). September 25, 1987.
Britz MB, Maibach H, Anjo DM. 1980. Human percutaneous penetration of
hydrocortisone: the vulva. Archives of Dermatological Research, 267:
313-316.
Bronaugh RL, Stewart RF. 1986. Methods for in vitro percutaneous
absorption studies VI: preparation of the barrier layer. Journal of
pharmaceutical Sciences, 75:487-491.
Bronaugh RL, Franz TJ. 1986. Vehicle effects on percutaneous
absorption: in vivo and in vitro comparisons with human skin. British
journal of Dermatology, 115: 1-11.
peldmann RJ, Maibach HI. 1967. Regional variation in percutaneous
penetration of 14C cortisol in man. Journal of Investigative
Dermatology, 48: 181-183.
Hurley HS. 1985. Permeability of the skin. In: Dermatology.
floschella SL, Hurley HJ, eds. Volume 1, 2nd Edition. W.B. Saunders
Company, Philadelphia. Section IV, pp. 97-103.
Jordan WE, Lawson KD, Berg RW, Franxman JJ, Marrer AM. 1986. Diaper
dermatitis: frequency and severity among a general infant population.
pediatric Dermatology, 3: 198-207.
Le Bel GL, DT Williams DT, FM Benoit FM. 1989. Determination of
chlorinated dibenzodioxins and dibenzofurans in selected paper products.
Ninth International symposium on Chlorinated Dioxins and Related
Compounds, Toronto Ontario, September 17-22, 1989. Abstract PLP23.
CC. 1989. Human health hazard assessment of dioxins/furans. U.S.
, Office of Toxic Substances. Memorandum to L. Dicker, EPA, Office of
toxic Substances, October 31, 1989.
Maibach HI, Feldmann RJ, Milby TH, Serat WF. 1971. Regional variation
in percutaneous penetration in man. Archives of Environmental Health,
23; 208-211.
McCormack JJ, Boisits EK, Fisher LB. 1982. An in vitro comparison of
the permeability of adult versus neonatal skin. In: Neonatal skin.
Maibach H, Boisits EK, Eds. New York: Marcel Dekker, Inc., Chapter 11,
pp. 149-164.
Mizutani, H. 1987. Designing Japanese diapers becomes a growing
concern. Nonwovens World. November, pp. 39-41.
8-21
16004
-------
NCASI. 1987. National Council of the Paper Industry for Air and Stream
Improvement. Assessment of potential health risks from dermal exposure
to dioxin in paper products. Technical Bulletin No. 534. November, 1987.
NCASI. 1989. National Council of the Paper Industry for Air and Stream
Improvement. Interim report on measurement of pulp/aqueous solution
partition coefficients. November, 1989.
O'Reilly JT. 1989. Communicated to the U.S. Consumer Product Safety
Commission by J.T. O'Reilly, The Procter and Gamble Company, May 4, 1989.
Rutter N. 1987. Percutaneous drug absorption in the newborn: hazards
and uses. Clinics in Perinatology, 14: 911-930.
Scheuplein RJ, Ross LW. 1974. Mechanism of percutaneous absorption V.
Percutaneous absorption of solvent deposited solids. The Journal of
Investigative Dermatology 62: 353-360.
USEPA. 1988b. U.S. Environmental Protection Agency. U.S. EPA -- Paper
Industry Cooperative Dioxin Study, Agreement. [Commonly referred to as
the Tier I of 104 Mill Study]. April 25, 1988.
USEPA. 1988c. U.S. Environmental Protection Agency. Risk assessment
for dioxin contamination in Midland, Michigan. EPA 905/4-88-005.
April 1988.
USEPA. 1989. U.S. Environmental Protection Agency. Exposure Factors
Handbook. Office of Research and Development, Washington, DC.
EPA 600 8-89-043. July 1989.
Ulsamer AG, Porter WK, Osterburg RE. 1978. The percutaneous absorption
of radiolabeled TRIS from flame-retarded fabric. Journal of
Environmental Pathology and Toxicology 1: 543-549.
Weber LWD, Zesch A, Rozman K. 1989. Penetration of TCDD into human skin
in vitro. Annual Meeting of the Society of Toxicology, Atlanta, GA,
February 27 - March 3, 1989. Abstract No. 472.
Wester RC, Maibach HI. 1983a. In vivo percutaneous absorption. In:
Dermatotoxicology. Marzulli FN, Maibach HI, eds. 2nd Edition,
Hemisphere Publishing Company, Washington. Chapter 5, pp. 131-146.
Wester RC, Maibach HI. 1983b. Cutaneous pharmacokinetics: 10 steps to
percutaneous absorption. Drug Metabolism and Reviews, 14: 169-205.
Wilson DR, Maibach H. 1982. An in vivo comparison of skin barrier
function. In: Neonatal skin. Maibach H, Boisits EK, Eds. New York:
Marcel Dekker, Inc. Chapter 5, pp. 101-113.
8-22
1600q
-------
9.1
ASSESSMENT OF RISKS TO THE GENERAL POPULATION EXPOSURE TO
DIOXINS AND FURANS RESULTING FRON THE USE OF PULP-CONTAINING
MEDICAL DEVICES
Introduction
This chapter provides estimated exposures and risks associated with
use of medical devices containing bleached wood pulp. This chapter
.js, in large part, a condensed version of the following report prepared
bv/EPA as part of the Interagency Dioxin-in-Paper Workgroup:
USEPA. 1989. U.S. Environmental Protection Agency. Assessment of
exposures and risks to the general population from the use of pulp-
containing medical devices. Draft Report. Washington, D.C.: U.S.
Environmental Protection Agency, Office of Toxic Substances.
Contract No. 68-02-4254.
The scope of this assessment is limited to "patients" (nonoccupation-
1); occupational exposures and risks were not estimated. This assessment
fas been developed in cooperation with the Food and Drug Administration
(fDW Center for Devices and Radiological Health (CRDH). The Center
identified those medical devices that contain bleached wood pulp, and
lrovided numerous parameters/assumptions concerning product use and wood
[julp content, and general guidance on how each product is used.
Table 9-1 provides a list of those medical devices believed to contain
Beached wood pulp and the corresponding use for each device. Note, how-
*,ver, that FDA is not certain whether other medical devices may contain
hieached wood pulp because manufacturers are not required to provide FDA
"nformation regarding the ultimate source of the raw materials. There-
lore, the products listed in this table are FDA's best estimate of what
Ir-oducts contain bleached wood pulp. These devices are used at medical
facilities, and several may be purchased over the counter for home use.
This section is organized into three parts. Section 9.1 contains
vnosure and risk estimates for each device listed in Table 9-1., along
h input parameters used to derive these estimates. A discussion on
ertainties is presented in Section 9.2, and Section 9.3 presents con-
2 Estimates of Exposures and Risks from Dermal Contact with Pulo-
9' Containing Medical Devices
2.1 Exposure Parameters
The exposure parameters used to estimate exposure and risks from der
al contact with pulp-containing medical devices are listed in Table 9-2.
9-1
-------
Table 9-1. Medical Devices for Which Exposures and Risks Were Estimated and Their Corresponding Uses
Medical device
Use
vo
i
no
Unscented Menstrual pad
Scented Menstrual pad
Unscented Menstrual tanpon
Scented Menstrual tanpon
Alcohol pads
Skin preparation for dressing wounds
Absorbable henostatic agents
Wound dressings containing carbcncyMethyl
cellulose
Surgical apparel
Adult diapers
Medical disposal bedding
Medical absorbent fiber
Absorbent tipped applicator
Examination gown
DphthalMlc sponges
HydroxpropyMethyl cellulose
Cottoooid paddie
Electroconductive Media
Cutaneous electrode
Anesthetic conduction filter
Breathing circuit bacteria filter
Heat and Moisture condensers
Isolation gowns
To absorb Menstrual discharge
To absorb Menstrual discharge
To absorb Menstrual discharge
To absorb Menstrual discharge
To apply alcohol or other disinfectants to the surface of the skin
To clean cuts or wounds before applying a permanent bandage
A small sponge used during surgery
To cover cuts or wounds
Worn fay surgeons, nurses, and patients during surgery (e.g.. hoods, caps, masks, gowns, foot coverings, drapes)
To absorb urine or feces uncontrollably released by adults
To cover Mattresses
Cotton-like pads used to apply Medication or to absorb snail amounts of fluid from a patient's body surface
To apply Medications or ranove specimens from a patient
Worn by patients during examinations
Snail sponges used to absorb fluids during eye surgery
To replace fluids in the eye lost during surgery
To absorb body fluids (i.e.. a cotton ball)
Conductive creams or gels used to reduce the impedance to the electrode from the surface of the skin
An electrode applied directly to the skin to either record physiological signals or apply stimulation
A Microporous filter used to remove participates froM anesthesia or other gases
To filter Microbiological and particulate matter frow a breathing circuit (which administers medical gases to a
patient)
To preserve the purity and physical state of gases used in a respirator or as an anesthesia
Worn to isolate patients at a hospital
-------
Device name3
Unscented Menstrual Pad
Scented Menstrual Pad
Unscented Menstrual Tampon
Scented Mentrual Tampon
Alcohol Pads
Skin Prep. Wipe for
Dressing Wounds
Absorbable Hemostatic Agents
(e.g.. SurgicelR. Oxycel)
Wound Dressings Containing
Carboxymethyl Cellulose
Surgical Apparel: Hood. Cap.
Masks , Gowns , Foot Cov . . Drapes
Adult Diapers
Medical Disposable Bedding
Medical Absorbent Fiber
Absorbent Tipped Applicator
Examination Gown
Ophthalmic Sponges
Hydroxypropymethyl Cellulose
Cottonoid Paddie
Electro Conductive Media
Cutaneous Electrode
Anesthetic Conduction Filter
Breathing Circuit Bacteria Filter
Heat & Moisture Condensers
Isolation Gowns
Contact type3
Skin
Skin
Intact Nat. Channel
Intact Nat. Channel
Skin
External, Short Term
Internal, Short Term
Compromised Tissue
External
Skin
Skin
Skin
Skin
Skin
Surgical Aids
Intraocular Surg Aid
Compromised Tissue
Skin Surf ace (Intact)
Skin Surf ace (Intact)
Ho Direct Contact
tto Direct Contact
Ho Direct Contact
Externa 1
Device
mass3
(gn)
10
10
3-5
3-5
0.5-1
2
3-5
4
150 (GUNS)
7-10 (MSKS)
113.5
113.5
<0.5
0.25
113.5
0.5
<1 ml
2
1-5
1-5
2-3
2-3
2-3
150
Pulp in
product3
(X)
90
90
90
90
100
100
100
100
90
100
100
50
100
100
100
<1
<1
<1
100
100
100
100
Pulp
mass in
product3
(9")
9
9
3.6
3.6
0.75
4
150
8.5
102.2
113.5
0.5
0.12
113.5
0.5
1
0.002
0.003
0.003
2.5
2.5
2.5
150
Exposure
durat ion3
(days/
lifetime)
2.400
2.400
2.400
2.400
6
NA
NA
NA
0.17
730
1
17.7
17.7
0.6
0.08
0.08
0.5
2
2
0.17
Volume of
liquid on
skin/
total
volume
(X)
25
25
100
too
100
50
100
50
NA
0.017
HA
50
100
NA
100
100
100
100
100
NA
Wetting
factor
(X)
10
10
100
100
100
10
100
50
NA
10
NA
100
100
NA
100
100
100
100
100
NA
Absorbt ion
rate Partition _
through
skinc
(X)
25
25
100
100
25
25
100
100
0.30
25
0.30
25
25
0.30
100
100
100
25
25
0.30
c
coefficient
TCDD
14.300
14.300
14.300
14.300
2.000
14.300
14.300
14.300
NA
14.300
NA
14.300
2,000
HA
14.300
14.300
2.000
2.000
2.000
HA
TCDF
5.300
5.300
5,300
5,300
2.000
5.300
5.300
5.300
NA
6.300
NA
5.300
2.000
NA
5.300
5,300
2,000
2.000
2.000
NA
NA - Not applicable
3 Data obtained from FDA/CDRH (Stratmeyer (1989) or telephone conversations between Versar and FDA).
Assumptions by Versar and FDA based on best available data and expected use patterns.
c Based on data obtained from Babich (1989) and Babich et al. (1989) (see Chapter 9 and Appendix A of this assessment)
-------
In addition to the exposure parameters listed in Table 9-2, the
industry average concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCOF found in
pulp in the 104 Mill Study were used to estimate exposures and risks for
all medical devices, except those made from rayon. This was necessary
because, in most cases, concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
pulp at individual mills could not be traced to specific medical devices.
In calculating the average values, one-half the detection limit was
substituted for nondetected values (see Table 9-3). However, the average
concentrations were similar to average concentrations calculated without
nondetected values.
For those medical devices made from rayon, the identities of those
mills that produce dissolving cellulose pulp used to make rayon were
identified by the American Paper Institute. The locations of the sites
and the concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in pulp from
those sites as found in the 104 Mill Study are presented in Table 9-4.
Of the devices listed in Table 9-2, the following subset belongs in
the category of rayon-containing devices:
Unscented Menstrual Tampon
Scented Menstrual Tampon
Wound Dressings Containing Carboxymethyl Cellulose
Medical Absorbent Fiber
Hydroxpropymethyl Cellulose
The exposure parameters in Table 9-2 that require further explanation
are detailed below.
(1) Exposure duration (davs). Depending on the specific situation,
alcohol pads are used rarely to daily. As a worst case assumption, it is
assumed that each application of alcohol pads lasted 30 seconds and will
be administered 365 days per year for 50 years.
The surgical apparel and isolation gowns are used only during surgery.
It is assumed that surgery lasts 2 hours or 0.083 day and occurs twice
over a 70-year lifetime. In addition, exposure to medical disposable
bedding will occur for hours on a rare basis. It is assumed that exposure
to medical disposable bedding would last 12 hours per visit and would
occur twice over a 70-year lifetime. The examination gowns used by
patients are worn occasionally for hours. It is assumed that the gowns
are worn for 1 hour every 5 years over a 70-year lifetime.
Exposures to medical absorbent fiber and absorbent tipped applicators
occur for seconds on an occasional to daily basis. As a worst-case
assumption, it is assumed that these devices are used for 60 seconds at a
rate of 365 days per year over 70 years.
9-4
1591q
-------
8710H
Table 9-3. Average Concentrations of 2,3.7.8-TCDD and 2.3.7,8-TCDF In
Pulp Calculated Based on Results from the 104-Mill Data Base
Average Average
Highest cone, without cone, with
cone. nondet. nondet.
(pg/g) (pg/g) (pg/g)
2,3.7,8-TCDD 116 8.4 8.5
2,3.7,8-TCDF 2,620 84.4 84.4
9-5
-------
8710H
Table 9-4. Concentrations of 2.3.7,8-TCDD and 2.3.7,8-TCOF in
Pulp at Pulp Hills that Produce Dissolving Cellulose
Company
Alaska Pulp Corp.
International Paper Co.
ITT Rayon ier. Inc.
Ketchlkan Pulp & Paper
Proctor & Gamble Co.
Weyerhaeuser Co.
Hill location
Sitka. AK
Natchez, MS
Fernandina Beach, FL
Jesup, GA
Port Angeles, WA
Ketchlkan, AK
Mehoopany, PA
Cosmopolis. WA
2,3,7.8-
TCDD
cone.
(pg/g)
0.7 (HD)
3.6
2.2
0.2 (ND)
0.6 (ND)
0.3 (ND)
0.7 (ND)
0.7 (ND)
0.6 (ND)
0.3 (ND)
2.0
1.0 (ND)
NQ
0.3 (ND)
0.3 (ND)
2,3.7,8-
TCDF
cone.
(pg/g)
1.4
15.0
3.0
0.5 (ND)
0.8 (ND)
0.8
0.6
0.9
2.1
0.3 (ND)
1.1
6.3
6.4
2.9
3.1
AVERAGE CONCENTRATION3
0.8
3.0
ND -= Non-Detect.
NQ > Not Quantified.
a In calculating the average concentrations, ND values were assumed to
be one-half the detection limit.
9-6
-------
Both ophthalmic sponges and hydroxypropymethyl cellulose are used
during eye surgery. Eye surgery lasts less than 1 hour and occurs once
Or twice per lifetime. Therefore, it is assumed that eye surgery will
last 1 hour and that it occurs twice over a 70-year lifetime.
Cottonoid paddies are used several minutes to hours on an infrequent
basis; therefore, it is assumed that the paddies are used once for a
total of 12 hours over a 70-year lifetime. Electro conductive media and
cutaneous electrodes are also used on an infrequent basis for durations
of minutes to days. It is assumed that the exposure duration for these
devices occur once for 2 days over a 70-year lifetime.
(2) Absorption rate. The rate of 2,3,7,8-TCDD transferred to the
over a 24-hour period from surgical apparel, medical disposable
ng* and examination and isolation gowns is calculated as 0.0005/hr x
hr - 0.012. In addition, it was assumed that 25 percent of this amount
l be absorbed. There is no partition coefficient since this is based
dry skin transfer. Therefore, assuming dry dermal contact, the amount
transferred and absorbed over a 24 hour period is 0.012 x 0.25 « 0.003 (or
0.3*)-
For those products in contact for long periods of time with internal
hody fluids- or in contact with compromised tissue 1n a wetted state,
?00 percent absorption was assumed.
(3) Partition coefficient. The partition coefficients used are those
ported for paper pulp using ethanol , synthetic urine, or saline solution
4n Babich (1989a). The partition coefficient used for alcohol pads is
iksed on the ethanol results. Ethanol closely approximates the rubbing
icohol solution actually used. The transfer medium for the use of
absorbent tipped applicators, cottonoid paddies, electro conductive media,
nd cutaneous electrode is assumed to be analogous to the transfer medium
by Babich (1989b) for make-up removal using facial tissues (etha-
«1). This assumption provides a worst-case scenario for the partition
Jjifficlent.
For all other medical devices, with the exception of diapers, saline
nlution was assumed to be the most representative partitioning/transfer
fldium- For adult diapers, the results from the urine partitioning exper-
were used.
The- final general point about Table 9-2 is that when no actual data
e available, reasonable or reasonable worst-case assumptions were used.
example, for the "volume of liquid on skin/total volume" and the
factor," reasonable worst-case assumptions were used. For parti-
coefficients, the most reasonable case was selected; however, if no
choice could be made, the worst-case option was used. The estima-
'on f ^e exposure duration was also based on the most "reasonable"
V *
9-7
-------
assumptions. However, if accurate data were not available, reasonable
worst-case assumptions were used.
9.2.2 Exposure/Risk Assessment for Medical Devices
Table 9-5 lists the exposure/risks associated with the use of the
medical devices listed in Table 9-1. A few general points should be noted
when reviewing this table. First, lifetime average daily dose (LADO) was
estimated using three slightly different methods, depending on the way the
product is used and the type of data available. The most common method
was as follows:
|ann
LAUU = Body Weight (kg) x Lifetime (70 years) x 365 Days/Year
where:
C = Concentration (pg/g) of 2,3,7,8-TCDD or TCDF
PM » Pulp Mass (g/day)
ED Exposure Duration (days/lifetime)
V
WF
PC
AR
•Volume of liquid on skin/total volume
Wetting factor (unitless)
Partition coefficient (unitless)
Absorption Rate (%)
This method estimates the amount of 2,3,7,8-TCDD/TCDF available on
the skin surface, the transfer rate of dioxin from the medical device to
the surface of the skin (partition coefficient), and the absorption rate
through the skin. For several products (skin preparation for dressing
wounds, absorbable hemostatic agents, and wound dressing containing
carboxymethyl cellulose), FDA provided the total mass of product an
individual may reasonably be exposed to over a lifetime. Therefore, this
altered the way that the amount of 2,3,7,8-TCDD/TCDF available on the skin
surface was estimated. For these products, LADD was estimated as follows:
,.nn (CHTMlfVWWFUl/PCHARl ,q ?»
LMUU * Body Weight (kg) x Lifetime (70 years) x 365 days/year W
where:
C
TM
V
WF
PC
AR
Concentration (pg/g)
Total Mass Exposed
Volume of liquid on skin/total volume
Wetting factor (unitless)
Partition coefficient (unitless)
Absorption Rate (%)
9-8
1591q
-------
Table 3-5- Estimates of Risks to the General Population from the Use of Pulp-Contain
ing Medical Devices
Lifetime average
da ily dose
(LAOO)fc
c.d.e
(DQ/ka/dav)
Device name
Unscented Menstrual Pad
Scented Menstrual Pad
Unscented Menstrual Tampon
Scented Menstrual Tampon
Alcohol Pad
Skin Prep. Wipe for Dressing
Wounds
Absorbable Henostatic Agent
(e.g.. SurgicelR. Oxyccl)
Wound Dressing Containing
Carboxymethyl Cellulose
Surgical Apparel: Hood. Cap.
Mask. Gown. Foot cov.. Drape
Adult Diaper
Medical Disposable Bedding
Medical Absorbent Fiber
Absorbent-Tipped Applicator
Examination Gown
Ophthalmic Sponge
Hydroxypropymethy 1 Ce 1 lu lose
2.3.7.8-
TCDO
4.49E-08
4.49E-08
2.70E-07
2.70E-07
2.67E-09
2.08E-10
1.66E-08
7.82E-12
3.64E-07
1.05E-10
1.62E-06
3.46E-11
1.26E-09
9.71E-07
1.58E-11
2.50E-12
TEQ
1.65E-07
1.65E-07
5.43E-07
5.43E-07
5.32E-09
7.64E-10
6.11E-08
1.57E-11
7.25E-07
3.43E-10
3.23E-06
6.96E-11
2.51E-09
1.94E-07
5.15E-11
5.03E-12
4
4
1
1
1
2
1
4
2
9
9
1
7
5
1
1
2
EPA
-68E-11
.68E-11
.54E-10
.54E-10
.51E-12
-17E-13
.73E-11
.46E-15
.06E-10
.73E-14
.15E-10
.97E-14
.13E-13
.49E-10
.46E-14
.43E-15
.3.7.8-
TCDO
(X)
27
27
50
50
50
27
27
50
50
31
50
50
50
50
31
50
FDA
5.25E-12
5.25E-12
1.73E-11
1.73E-11
1.70E-13
2.43E-14
1.95E-12
5.01E-16
2.31E-11
1.09E-14
1.03E-10
2.22E-15
8.00E-14
6.16E-11
1.B4E-15
1.60E-16
2.3.7,8-
TCDD
(X)
27
27
50
50
50
27
27
50
50
31
50
50
50
50
31
50
CPSC
4.01E-12
4.01E-12
2.41E-11
2.41E-11
2.39E-13
1.86E-14
1.48E-12
6.99E-16
3.25E-11
9.41E-15
1.45E-10
3.09E-15
1.13E-13
8.67E-11
1.42E-15
2.24E-16
2.3.7.8-
TCDD
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Potentially
exposed
popu lat ion
3.96E+07
3.71E+07
2.83E+07
5.20E+06
l.OE+06 -
Mi 11 ions
Millions
Hundreds
Millions
Thousands
.OE+06 -
.OE406 -
.OE+06 -
.OE+06 -
.OE+06 -
l.OE+07
of Thousands
(patients)
(health care)
.OE+07
.OE+07
.OE+07
.OE+07
.OE+07
1.5 Million Cataract
Oper./Year
Cottonoid Paddie
Electro-Conductive Media
Cutaneous Electrode
Anesthetic Conduction Filter9
Breathing Circuit Bacteria Fltr.a
Heat & Moisture Condensers9
Isolation Gown
2.38E-12
3.56E-12
3.56E-12
3.64E-07
4.74E-12
7.10E-12
7.10E-12
7.25E-07
1
2
2
2
.34E-15
.01E-15
.01E-15
06E-10
SO
50
50
50
1.51E-16
2.26E-16
2.26E-16
2.31E-11
50
50
50
50
2.12E-16
3.18E-16
3.18E-16
3.25E-11
100
100
100
100
Millions
Mi 11 ions
Mi 11 ions
U * 1 1 *
Mi 11 ions
Millions
Millions
Millions (patients)
Thousands
(health care)
-------
Table 9-5. (continued)
a There will be no direct contact for these products. The only potential exposure route is through inhalation of dioxin that leaves the filter or
condenser and enters the indoor air. Exposure through this pathway is expected to be negligible because only a very small amount of dioxin will leave
these products and enter the air. and of the amount that does enter indoor air. very little will actually enter the lungs and be absorbed.
b LAOOs Mere calculated as follows:
fpql f q 1
(Concentration r^ x Pulp Mass 7~~i x Exposure Duration (Days) x Volume of Liquid on Skin/Total Volume x Vetting Factor (unitless)
U J I'M
x I/Partition Coefficient (unitless) x Absorption Rate (X))
Body Weight (Kg) x Lifetime (70 years) x 365 days/year
There were two exceptions, however. The first exception was the method to estimate LADD for surgical apparel, medical disposable bedding, examination
gowns, and isolation gowns, and this was explained in Equation 8-3. The other exception was for products where FDA already estimated the total mass of the
product available for exposure (skin prep, wipe for dressing wounds, absorbable hemostatic agents, and wound dressings containing carboxymethy1 cellulose).
In this case, LADD was estimated as follows:
(Concentration x Total Mass Exposed x Volume of Liquid on Skin/Total Volume x wetting Factor x I/Partition Coefficient x Absorption Rate)
Body Weight x 70 years x 365 days/year
C The slope factors are as follows for Z.3.7.8-TCDO: EPA = 1.56xlO~4 (pg/kg day)"1; FDA = 1.75xlO~5 (pg/kg day)"1; CPSC = 6.7xl65 (pg/kg/dayf1.
d The slope factors are as follows for 2.3.7.8-TCDF: EPA = 1.56xlO~5 (pg/kg day)"1; FDA = 1.75xlO~6 (pg/kg day)"1; CPSC = 0.
e For EPA and FDA cancer slope factors, risk was estimated as follows: Risk = potency factor (pg/kg-day)" x LADD (pq/kg-day)/0.55. However, for the
CPSC cancer slope factor, risk was estimated as follows: Risk = potency factor (pg/kg-day)"1 x LADO (pg/kg-day) / 0.75. The divisor is changed to
0.75 (from 0.55) because a different bioassay was used. The total risk is the sum of the risks from TCDO and TCDF.
-------
For four other devices (surgical apparel, medical disposable bedding,
isolation gowns, and examination gowns), the rate of 2,3,7,8-TCDD/TCDF
transferred to the skin and the absorption rate were combined. This
transfer and absorption rate was used by Babich (1989), and it applies to
products that will undergo dry contact with the skin surface. In these
situations, LADD was estimated as follows:
(CUPMUEDUTRUARi
Body Weight (kg) x Lifetime (70 years) x 365 days/year ^y"J)
where:
C
PM
ED
TR
AR
Concentration (pg/g) of 2,3,7,8-TCDD or TCDF
Pulp Mass (g/day)
Exposure Duration (days/lifetime)
Transfer Rate (unitless)
Absorption Rate (%)
As shown in Table 9-5, LADDs for 2,3,7,8-TCDD were found to range
from 2.38 x 10'" pg/kg-day for cottonoid paddies to 1.62 x 10'6
pg/kg-day for medical disposable bedding. LADDs for 2,3,7,8-TCDFs were
found to range from 2.36 x 10'11 pg/kg-day for cottonoid patties to
1.61 x 10'3 pg/kg-day for medical disposable bedding. The other
categories with the highest exposure levels are examination gowns,
isolation gowns, surgical apparel, and tampons. Exposures for medical
absorbent bedding, examination gowns, isolation gowns, and surgical
apparel were estimated using the transfer and absorption rate of 0.3
percent used by Babich (1989) because they involve dry skin c™*3"*
This method may be yielding unrealistically high estimates si
expected that, in reality, dry skin contact would yield a low
since it is
ower dose.
r!6
Estimated risks were found to vary from 2.22 x 10-JO to
Q 15 x 1°"J7 Usin9 EPA slope factors. They were found to vary from
2*49 x 10"1' to 1.03 x 10"1U using FDA slope factors and 5.07 x
10-1' to 1.45 x 10'1U to using the CPSC factor. As discussed in
Section 3.3. of this report, CPSC does not place the same emphasis on
risks tabulated by the TEQ method as it does for 2,3,7,8-TCDD itself when
estimating carcinogenic potency. Therefore, Table 9-5 presents CPSC risk
estimates based on 2,3,7,8-TCDD
alone.
9.3
Uncertainty Analysis
The goal of an analysis of uncertainties is to provide decision makers
with the complete spectrum of information concerning the quality of an
assessment, including the variability in the estimated exposures and
risks, the inherent variability in the input parameters, data gaps, and
the effect these gaps have on the accuracy or reasonableness of the expo-
9-11
-------
sure and risk estimates developed. The general causes of uncertainty in
an exposure/risk assessment are as follows:
• Measurement error
• Use of indirect empirical or generic data
• Variability
• Use of models to estimate exposure/risk
• Use of professional judgment/disagreement
For this assessment, uncertainties will occur from all of the above
areas. All areas are important, with the possible exception of measure-
ment errors. Measurement errors will occur (e.g., in determining the
product mass), but compared to other errors, they will usually be insig-
nificant. The remainder of this section discusses how the specifics of
this assessment apply to the major areas of uncertainty.
Indirect or empirial data create uncertainties when the surrogate data
used do not directly apply. The most important example is the partition
coefficient. Most partition coefficients used were not estimated using
the transfer medium in which the exposure will take place. It is antici-
pated that the partition coefficient can affect the results by over an
order of magnitude, and this may be the single most important area of
uncertainty.
Use of models to approximate the process of transfer and absorption
of dioxin thru human skin introduces uncertainty into the assessment.
Uncertainty may be further compounded by the selection of the input param-
eters because errors associated with these parameters may be propagated
by the use of these models.
Variability and professional judgment are most important in terms of
the input parameters used in the exposure and risk models. All parameters
are affected to some degree by these two areas of uncertainty, with expo-
sure duration likely to have the largest effect on the results. For some
categories (e.g., menstrual products), exposure duration is known within
reasonable limits. In most other categories, however, a wide range of
possible exposure durations is expected, and thus a high level of uncer-
tainty will occur. Professional judgment is also particularly important
for "volume of liquid on skin/total volume" and "wetting factor," since
in most cases, measured data were not available.
9.4 Conclusions
Based on the analysis presented in this chapter, risks from individual
medical devices are very small. The most significant risk, medical
disposable bedding, was found to be 9.15 x 10'10. It 1s possible that
risks to health care workers could be greater than other subpopulations
because this population will have significantly higher exposure durations
9-12
ISSlq
-------
and may be exposed to multiple medical devices. Unfortunately, this sub-
population could not be characterized with the existing data. If addi-
tional work is done on risks from dioxins and furans in medical devices,
additional data should be gathered, and risks to health care workers
should be characterized.
9.5 References
pabich MA, Adams M, Cinalli C, Galloway D, Hoang K, Huang S, Rogers P.
1989- Common assumptions for the assessment of human dermal exposure to
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and 2,3,7,8-tetratetrachloro-
jibenzofuran (TCDF). Interagency Dioxin-in-Paper Workgroup, Dermal
pioavailability Workgroup, December 12, 1989.
gabich MA. 1989. CPSC staff assessment of the risks to human health
from exposure to chlorinated dioxins and dibenzofurans in paper products.
Memorandum from Dr. Michael A. Babich (U.S. Consumer Product Safety
Commission) to Lois Dicker (U.S. Environmental Protection Agency, Office
of Toxic Substances). January 25, 1990.
ctratmeyer ME. 1989. Letter from Dr. Mel E. Stratmeyer (U.S. Food and
Drug Administration, Center for Devices and Radiological Health) to
rreg Schweer (U.S. Environmental Protection Agency, Office of Toxic
Substances). June 5, 1989.
9-13
-------
10. ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM DIOXINS AND
FURANS IN FOODS PACKAGED IN OR CONTACTING BLEACHED PAPER PRODUCTS
10.1 Introduction
This chapter is a condensed version of a report prepared by the
Quantitative Risk Assessment Committee (QRAC) of the U.S. Food and Drug
Administration (FDA) as part of the Interagency Dioxin-in-Paper Workgroup:
USFDA. U.S. Food and Drug Administration. 1990. Carcinogenic risk
assessment for dioxins and furans in foods contacting bleached paper
products. Report of the Quantitative Risk Assessment Committee.
April 20, 1990.
Trace amounts of dioxin, mostly 2,3,7,8-chlorine-substituted
congeners, are formed during the bleaching step in the manufacture of
paper pulp. Studies carried out by the paper industry and the EPA in
1987-1988 confirmed the presence of trace amounts of dioxins and furans
in bleached pulp and in bleached paper used for food packaging, as well
as for personal care and other products. In addition, a 1988 preliminary
Canadian survey of milk packaged in polyethylene-coated bleached paper
cartons (Ryan et al. 1988) indicated that dioxin transfer to milk could
occur.
At the request of EPA and FDA, paper manufacturers and the American
Paper Institute (API) initiated studies to determine the migration of
dioxin from various paper packaging materials into food. The possibility
that dioxin congeners might transfer to food from paper articles was
verified in brewing experiments with coffee filters conducted at Wright
State University (Tiernan 1987) and by the National Council of the Paper
Industry for Air and Stream Improvement (NCASI 1988). Also, in the
spring of 1989, FDA collected and analyzed samples of homogenized whole
milk packaged in polyethylene- coated bleached paper cartons for dioxins
and furans. Based on these data, FDA performed a quant1*"™6."8* .ne
assessment for cancer risk to consumers of milk contaminated with dioxins
(USFDA 1989).
Because of the enormous size of the paper market, FDA requested that
the paper industry provide information which could be used for assessing
(1) which food-contact articles had the greatest potential for
contaminating food with dioxin congeners and (2) which food-contact
situations needed to be the subject of detailed migration studies.
Based on the results of a detailed survey by the American Paper
Institute of the end use market of paper products in contact with food,
FDA evaluated the need for additional migration tests based on the
Potential for human exposure to dioxin congeners from paper food-contact
articles. The following criteria were used to decide which articles were
°f greatest concern:
10-1
1590q
-------
• the concentration of the dioxin congeners in the pulp used to
manufacture the food-contact article;
• the basis weight of the paper in the article;
• the presence or absence of a coating on the paper;
• the nature of the coating;
• the temperature of the food during contact with the paper article;
• the duration of food contact;
• the food weight-to-surface area ratio during contact with the
paper;
• the nature of the food; and
• the amount of food consumed per day that might contact the paper
article.
The paper industry has conducted a series of migration tests on the
high priority paper food-contact articles identified by FDA. These
studies were performed under conditions that were intended to simulate as
closely as possible actual food-contact applications. The results of
these studies have formed the basis of a series of research reports from
NCASI. These reports also served as the primary basis for our estimates
of dioxin migration and exposure from paper food packaging. This
information was analyzed by the Food and Color Additives Review Section
(FCARS) and the Division of Nutrition and communicated to the
Quantitative Risk Assessment Committee (QRAC). Because migration levels
to food are affected by the level of residual dioxin congeners in the
packaging material, changes in bleaching techniques that are currently
underway in the paper industry are expected to reduce the amount of
dioxin migrating to food and therefore estimates of exposure to dioxin
congeners.
10.2 Dioxin Concentrations 1n Bleached Hood Pulp and Paper Food-
Contact Articles
Bleached wood pulp and paper food-contact articles have been analyzed
using similar analytical methods. The high resolution GC/MS methods
described in NCASI Technical Bulletin No. 551 (NCASI 1989a) ("NCASI
Procedures for the Preparation and Isomer-Speciflc Analysis of Pulp and
Paper and Industry Samples for 2,3,7,8-TCDD and 2,3,7,8-TCDF") have been
used for the development of essentially all of the analytical data
submitted to FDA as well as for most of the pulp data analyzed under the
Industry/EPA Cooperative Study Agreement (the 104-Mill Study). The paper
10-2
1590q
-------
Industry has conducted full congener analyses of bleached pulp that
demonstrate that other chlorinated dioxins and furans may also be present
in bleached pulp (NCASI 1989b); however, the two congeners covered by the
above methodology account for 93 to 100 percent of the dioxin toxic
equivalents (TEQ).
The results from the 104-Mill Study provide one view of the range of
dioxin congener levels in bleached pulp and a perspective on the range of
values that might be encountered in food packaging. A limited set of
analyses of dioxin residues in "representative" paper articles, as shown
in Table 10-1, performed as part of the various NCASI studies of dioxin
migration from high priority food-contact articles, indicate that dioxin
residues in paper articles contacting food are comparable to or slightly
less than the average pulp levels found in the 104-Mill Study.
Recent information from the paper industry (API 1990) breaking out
the results from the 104-Mill Study according to those mills producing
pulp f°r specific applications was used by FCARS for the exposure
assessment. This group of food-contact materials included coffee
filters, microwave popcorn bags, ovenable board stock, paperplate stock,
and cup- stock. For the remaining articles (bakery cartons, ice cream
cartons, tea bags, and margarine wrap), the average dioxin residue levels
for the 104-Mill Study were assumed (i.e., 8.5 ppt 2,3,7,8-TCDD and 85
npt 2,3,7,8-TCDF; 17 ppt dioxin TEQ). The assumption for this last group
of typical "food-contact" articles is based on FCARS' observations that
the dioxin levels in those "representative" articles used in the
migration studies are reasonably close to the average levels of the
104-Mill Study.
In addition, the paper industry has recently submitted updated dioxin
residue data for all manufacturers of milk carton stock, which is also
used for the packaging of juice products and half and half. Recent
analyses of bleached pulp and paperboard show consistently low dioxin
residue levels for all milk carton manufacturers (< 1 ppt
2 3,7,8-TCDD and < 0.6-2.0 ppt dioxin TEQ). Because a previous FDA
risk assessment (Carcinogenic Risk Assessment for Dioxins in Milk Based
on the FDA 1989 Survey, memorandum dated August 2, 1989, QRAC to R.J.
Scheuplein, Ph.D.) for milk packaged in paperboard cartons was based on
the results of milk samples obtained from cartons manufactured prior to
these substantial reductions in dioxin residue levels, revised estimates
of dioxin exposure reflective of present carton manufacturing practices
are presented in this assessment.
10>3 Food-Paper Migration Studies
10.3»l Types of Articles Investigated
Essentially all of the information on migration of dioxin congeners
from paper food-contact articles has been derived from studies conducted
10-3
1590"
-------
9071H
~aole LC-1. Representative Dioxir, Congener Levels in Pulp and
Paper Matrices
2.3,7,8- 2,3.7,8-
TCDD TCOF
Product type (ppt) (ppt)
Bleached wood pulp <1 - 116a <1 - 2B2Da
8.5b 85b
Coffee filters
Coffee cupsc
Cups - soupc
Trays dual ovenable PET coated
Plates - uncoated and clay coated
Milk cartons
Half and half cartons
Juice cartons EVA6 coated
1 - 4
4-13
11 - 13
0,5 - 3
5 - 10
1 - 12
4
5
4 -
5 -
86
1 -
11
1 -
25
44
23
115
- 99
4
- 90
74
aRange of values from the 104-Hr 11 Study.
Average value from the 104-Mill Study.
Low density polyethylene - coated cups.
Polyethylene terephthalate (uncoated base stock analyzed).
Ethylene vinyl alcohol.
10-4
-------
by NCASI. These studies were conducted with the intent of providing the
FDA information on the extent of migration from high priority
food-contact articles. Protocols for the migration studies were designed
with FDA input with the intent of simulating as closely as possible
actual use conditions, or in some instances, the worst use conditions
that the article was likely to experience with food. NCASI has submitted
the results of migration studies for the following food-contact
articles: coffee filters, milk cartons, cream cartons, orange juice
cartons, paper cups for hot beverages, paper cups for soups, paper plates
for hot foods, and dual ovenable trays (NCASI 1988, 1989d-1989i, 1990a).
Results of these studies and preliminary findings in a study of microwave
popcorn (NCASI 1990b) have been evaluated by FCARS.
FCARS assumed that the food migration data developed for one type of
paper article could serve as a surrogate for related food-contact
articles. The results of some of the migration studies as well as
certain assumptions have been used to estimate migration levels for those
food-contact articles (cartons for bakery products, cartons for ice
cream, tea bags, and wraps for butter and margarine) that were not
subject to detailed migration studies. In other instances, FCARS assumed
that potential dioxin congener transfer would either be low (e.g., from
dry food packages such as flour bags) or the amount of food exposed would
he low. In either case, FCARS considered that migration studies would be
unnecessary because the expected contribution to dietary exposure to
dioxin congeners would be insignificant relative to the exposures that
are likely for the high priority food-contact articles.
jO.3.2 Conclusions from the Migration Studies
All of the migration studies demonstrated detectable levels of
transfer of the dioxin congeners to the test foods. Migration studies
•Lith milk in paperboard were able to demonstrate that the migration
levels to milk increased during refrigerated storage as a linear function
of the square root of the storage time (thus establishing that the
transfer process to food of dioxin congeners followed familiar migration
kinetics).
The temperature of foods, their composition, the nature of the paper
article, as well as the dioxin congener levels in the paper articles wen
ii found to influence the extent of transfer to test foods. In the case
f coffee filters, the action of the hot liquid flowing through the
^iter during brewing was able to facilitate nearly quantitative
itraction of 2,3,7,8-TCDF ( and presumably 2,3,7,8-TCDD). When hot
nffee was held in polyethylene-coated heavy cup stock, that is, not
Slowing through the paper, very little extraction of either congener was
hserved (s 0.4 percent of the amounts in the cup). When the test
2 od was changed to hot chicken soup held under similar conditions, the
rcent extraction from the heavy cup stock was found to increase to 6
10-5
-------
percent for 2,3,7,8-TCDD and 10 percent for 2,3,7,8-TCDF. The highest
extraction levels were found with clay-coated paper plates exposed to hot
corned beef hash (13 percent 2,3,7,8-TCDD extraction and 18 percent
2,3,7,8-TCDF extraction). Dioxin congener levels in milk were found to
be roughly proportional to the levels in cartons used to hold the milk.
The test articles used in the migration studies were selected by
NCASI and FCARS as "representative" of the articles in use. FCARS
observed that the dioxin congener levels in the test articles were
somewhat different from the average levels reported in pulp from the 104-
Mill Study for all mills production pulp used in each of the
applications. The predicted dioxin TEQ levels in foods used in this
assessment represent migration levels that were either determined during
migration studies performed under "worst-case" scenarios that each paper
article might experience with food or estimated from migration values
developed for one type of paper article that was assumed to be a
surrogate for another.
One of the complicating factors encountered in the migration studies
was the presence of dioxin congeners in food matrices prior to their
contact.with paper articles. Although NCASI made no attempt to identify
the environmental sources of these contaminants, they attempted to
account for the background levels of dioxin congeners in food when
evaluating the contribution of paper articles to human exposure to dioxin
congeners. FCARS believes that such an adjustment is essential. Where
adequate data were available, FCARS used such corrected figures for
estimating exposure.
10.4 Food Intake Information
The estimation of exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF (i.e.,
exposure to dioxin TEQ) from consumption of foods contacting paper
articles requires consideration of appropriate food intake information.
Available nationally representative food consumption data bases, however,
do not provide information on whether foods are sold, held, heated,
cooked, or served in contact with different types of materials. Although
production data on paper articles were provided by industry , use of
these data to adjust food consumption intake to account for the many
food-contact scenarios was not possible. Therefore, for estimating the
total dioxin TEQ intake from all foods that may contact bleached paper,
available food consumption data for average consumers of each of the
foods were used in conjunction with the assumption that all such foods
have indeed been in contact with bleached paper prior to being consumed.
This necessary assumption leads to an overestimate of consumption of food
that has been in contact with paper and, hence, to an overestimate of
total dioxin TEQ exposure.
This upper-bound estimate of total dioxin TEQ exposure was adjusted
through a comparison with a lower-bound estimate of dioxin TEQ exposure,
10-6
1590q
-------
specifically, a per capita exposure, obtained using industry-provided
production figures and estimates of the amount of food that might contact
the paper. Additionally, exposure to dioxin TEQ on a food-by-food basis
for individuals consuming mean and 90th percentile amounts of foods
contacting each paper article was also calculated.
Because 100 percent of the U.S. population are "eaters" of at least
one food contacting a bleached paper article, it was assumed that 100
percent of the U.S. population are "eaters" of dioxin. Therefore,
estimates of food intake used to estimate total dioxin TEQ intake should
he representative of the total population. We selected the Market
Research Corporation of America's (MRCA) data base, five-year Menu Census
1982-87 (MRCA 1987) as our source of food intake data. This data base
contains 14 consecutive days of data and is likely to capture a
representative diet including both frequently and infrequently consumed
foods.
10.5
Estimated Exposures
in 5.1 Estimated Total Dioxin TEQ Intake: All Foods Contacting
Bleached Paper
Total -sample intakes for each food of interest in the MRCA data base
ere multiplied by the dioxin TEQ levels in the corresponding foods (see
Table 10-2). The resulting dioxin TEQ values were summed to provide a
total population dioxin TEQ intake of 12.7 pg per person per day. This
ran be considered to be an upper-bound estimate of mean dioxin TEQ intake
because of the assumption that all food had been in contact with bleached
paper Prior to being consunied'
The availability of industry production figures (NCASI 1989c) allowed
for the estimate of a lower-bound exposure to dioxin TEQ of 5.5
nfl/person/day (see Table 10-3). Recognizing that a reaonable estimate
Phould fall between the upper-and lower-bound estimates, we were able to
a "best estimate" of mean total dioxin TEQ exposure of 9.1
e .
^/person/day. This value is the a mid-point between the upper- and
Tbwer-bound estimates.
The lower-bound estimate was obtained in two steps. First, the same
• ox-jn TEQ food levels as used for the upper-bound estimate were combined
]th per capita food intakes derived from industry production figures and
ctimates of the amount of food that might contact each of the paper
eJticles of interest. In three instances (i.e., bakery cartons, tea
as *nd rnarg3^06 wraP)» where no data were provided, we used the total
Sole population MCRA intake for the appropriate food categories. The
5 r capita dioxin TEQ values obtained for each food were added to obtain
phe total per capita dioxin TEQ disappearance (11 pg/person/day)
"
10-7
-------
9071H
Table 10-2. Carcinogenic Risk for Consumers Resulting from Total Dioxin TEQ Intake
from All Foods Contacting Bleached Paper ("mean consumer - total sample basis")
Paper TEQb
Food3
(paper article)
1.
2.
3.
4.
5.
6.
7.
8.
Milk (cartons)
Coffee (f i Iters)
Cream (cartons)
Juice (cartons)
Coffee (cups)
Soup (cups)
Meals-seasoned meat , vegs.
(dual-oven trays)
Meals-seasoned meat
levels
(ppt)
2
8.8
2
2
10.1
10.1
10.6
7.9
TEQ levels
in food
(ppq)
5
3.2
5
15
0.8
23
35
140
Food intake
per eating
event (g)
191
332
23.8
190
332
292
215
340C
Avg. daily
food intake
per person
(grams/day)
124
136
1.4
36
136
56
61
37
Daily
TEQ intake
per person
(pg/day)
0.62
0.44
0.007
0.54
0.11
1.3
2.1
5.2
(paper plates)
9. Popcorn (microwave bags) 5.9
10. Donuts, sweet rolls
(bakery cartons)
11. Frozen dairy desserts
13.8
45
50
67.8
1.6
7.7
0.072
0.39
(ice cream cartons) 13.8
12. Tea (bags) 17
13. Margarine (wrap) 17
50 110 23
8 301 22
82 10.8 7.3
Upper-bound
Lower-bound
Beat estimate
1.2
0.12
0.60
12.7
5.5
9.1**
"The corresponding upper bound lifetime risks were estimated to be 2.4x10" based on FDA's
slope factor, Z.lxlO"5 based on EPA's slope factor, and 6.7xlO"6 based on CPSC's slope
factor.
aFood intake reported by Market Research Corporation of America (MRCA) obtained by multiplying the HRCA
mean frequency of eating occasions (1982-87 5-year Menu Census) by the mean grams/eating occasions fr*
USDA/NFCS. 1977-78 (Pao et al. 1982). Data are for the 2+ years age group, males and females, total saW
population.
b104-mill average for producers (except updated for milk, cream and juice carton producers) (API 19901-
The value for items 10-13 is the total 104-Mtll Study average for bleached wood pulp.
cFood intakes based on NCASI production of 20 billion plates/yr and assumptions of 80 plates/person/year.
2 plates/per eating occasion, and 340 g food per eating occasion.
dNCASl per capita consumption estimate based on 1.4 billion bags sold 1n the U.S. in 1989. No survey
data available for microwaveable popcorn.
eEsttmate based on industry production figures for paper products.
Mldrange of lower and upper bound estimates.
10-8
-------
9071H
Table 10-3. Industry-Based Per Capita Dioxin TEQ Exposures
Paper TEQa
Food
(paper article)
I.
2.
3.
4.
5.
6.
7.
8.
9.
Milk (cartons)
Coffee (filters)
Cream (cartons)
Juice (cartons)
Coffee (cups)
Soup (cups)
Meals-seasoned meat, vegs.
(dual-oven trays)
Paper plates
Microwave popcorn
levels
(ppt)
2
8.8
2
2
10.1
10.1
10.6
7.9
5.9
TEQ levels
in food
(ppq)
5
3.2
5
15
0.8
23
35
140
45
Food intake
(g/p/day)
ioob
176°
1
59d
45e
45f
5.99
h
37
1.61
TEQ
intake
(pg/p/day)
0.5
1.0
0.9
0.04
1.0
0.2
5.2
0.07
10. Donuts, sweet rolls
(bakery cartons) 13.8
11. Frozen dairy desserts
(ice cream cartons) 13.8
12. Tea bags 17
13. Margarine (wrap) 17
50
50
8
82
7.7
k
j
23
22"
7.3j
0.4
1.1
0.18
0.60
a104-mill average for producers (except updated for milk, cream and juice carton
producers) (API 1990). The value for Items 12 and 13 Is the total 104-mlll average for bleached
wood pulp.
bNCASI estimate based on total production of milk in paperboard cartons.
CNCASI estimate based on coffee consumption data from the International Coffee
Organization, fraction of the population that consumes coffee, and fraction of
coffee that is brewed using filters.
CNCASI estimate based on total production of paperboard cartons used to package
juice.
eNCASI based on estimate on all cup stock used only for coffee consumption.
CNCASI based its estimate on use of all cup stock for soup consumption only.
9NCASI used production figures for dual-ovenable trays to estimate per capita annual
use of 6.4 trays. This was combined with the assumption of 340 g of food contacting
each tray.
hNCASI estimated annual production of 20 billion plates or 80 plates/person/year.
They assumed two plates are used to hold 340 g of food.
'NCASI estimate based on 1.4 billion bags (100 g/bag) sold 1n U.S. fn 1989.
JNO NCASI estimate available. Figure Is from MRCA (see Table 10-2, footnote3).
kNCASI estimate based on production figures from the International Ice Cream
Association.
CNCASI did not estimate. FCARS concluded that the per capita Intake would be quite
small, and would be subsumed within the figure provided for milk In paperboard
cartons.
10-9
-------
unadjusted for waste, inventory stock, and any other diversions from the
consumer. Second, in order to adjust for such wastage and losses, we
have divided the 11 pg/person/day by 2 to obtain the lower-bound of 5,5
pg/person/day.
This adjustment by a factor of 2 was used because comparisons of per
capita estimates of intake (or "disappearance") for food components and
nutrients with available survey data on food intake by individuals show
that per capita estimates generally exceed survey intake estimates by 1.5
to 3-fold (Park and Yetley 1990, USDA 1984, USDHHS/USDA 1986, Glinsman et
al. 1986). Most of the comparisons examined show approximately a 2-fold
difference. Finally, as noted above, we selected a best estimate of
dioxin TEQ exposure to fall midway between the upper- and lower-bound
values,
10.5.2 Estimated Dioxin TEQ Intake: Food-by-Food Basis for Individual
Paper Products
Although the above analysis is intended to provide a picture of
overall dioxin TEQ exposure for the total U.S. population from
consumption of foods that may have contacted bleached paper, the question
arises about potential dioxin exposure to individuals who specifically
have been identified as "eaters" of the various foods. The MCRA data
base contains food intake information for the "eaters-only" population as
well as for the total sample population. "Eaters-only" food intake data
were combined with the dioxin TEQ levels in the corresponding foods to
obtain potential dioxin TEQ intakes that would result from individuals
consuming the foods at the mean and the 90th percentile levels of intake
(see Table 10-4). Again, it was assumed that all foods consumed have
indeed been in contact with bleached paper prior to being consumed. The
other assumptions used for the total population exposure analysis also
apply to the "eaters-only" analysis. Therefore, the dioxin TEQ intakes
in Table 10-4 may also be considered as upper-bound estimates of mean and
90th percentile for "eaters" of each of the foods. The dioxin TEQ
exposures in Table 10-4 must not be summed because the population of
eaters is not the same for each food cateoorv.
10.6 Risk Assessment
10.6.1 Cancer Risks
Table 10-2 presents the results of the cancer risk assessment for the
"mean consumer - total sample basis." The estimates of lifetime upper
bound individual lifetime cancer risk range from 2.4 x 10~b (using
FDA's cancer potency slope factor) to 2.1 x lO"1* (using EPA's cancer
potency slope factor).
10-10
1590q
-------
Contwrtnatwl m\tt\ ft\ox\n Vwters only, - food-by-food tes\$"}
TEQ levels
Food3'1* in food
(paper article! (ppq)
Milk (cartons) 5
Coffee (filters) 3.2
Cream (cartons) 5
Juice (cartons) 15
Coffee (cups) 0.8
Soup (cups) 23
Meals-seasoned meat. vegs.
(dual-oven trays) 35
Metals-seasoned meat
(paper plates) 140
Popcorn (microwave bags) 45
Donuts, sweet rolls
(bakery cartons) 50
Frozen dairy desserts
tire cream cartons) 50
Food intake0
(9/P/d)
raean/90th Xtile
170/408
278/641
7.3/18.7
72/179
278/641
74/148
64/108
37/74d
16/326
15/29
32/63
f.g
Uooerbound lifetime risk
FDA
2.2 x IO"7
2.4 x IO"7
1.0 x 10~8
2.9 x IO"7
5.9 x 10"8
4.5 x 10"7
5.9 x 10'7
1.4 x 10"6
1.9 x 10'7
2.1 x 10"7
4.3 x 10"7
Mean intake
EPft
2.0 x 10~6
2.1 x 10"6
8.9 x 10"8
2.6 x 10"6
5.2 x 10"7
4.0 x 10'6
5.2 x 10"6
1.2 x 10"5
1.7 x 10"6
1.9 x 10"6
3.8 x 10"6
CPSC
6.2 x
6.7 x
2.8 x
8.1 x
1.6 x
1.3 x
1.6 x
3.9 x
5.3 x
5.9 x
1.2 x
ID'7
ID"7
to'8
io-7
ID'7
lO'6
ID"6
ID'6
lO'7
ID'7
ID'6
90th percent ile intake
FDA EPA
5.5 x
5.3 x
Z.6 x
6.0 x
1.4 x
9.1 x
9.9 x
2.7 x
3.7 x
4.0 x
8.5 x
IO'7 4.9 x IO"6
1C"7 4.7 x IO"6
ID"8 2.3 x IO"7
IO"7 5.3 x IO"6
IO"7 1.2 x IO"6
IO"7 8.1 x IO"6
IO"7 8.8 x IO"6
IO"6 2.4 x IO"5
10"7 3.3 x IO"6
IO"7 3.6 x IO"6
IO"7 7.6 x IO"5
CPSC
1.5 x IO"6
1.5 x IO"6
7.3 x IO"8
1.7 x IO"5
3.9 x IO"7
2.5 x IO"6
2.8 x IO"6
7.5 x IO"6
1.0 x IO"6
1.1 x IO"6
2.4 x IO"6
-------
907IH
Table 10-4. (continued)
o
i
ro
Food4'*1
(paper article)
Tea (bags)
Margarine (wrap)
TEQ levels
in food
(ppq)
B
82
Food intake0
(q/p/d) Mean intake
mean/90th Xtile FDA EPA
120/284 2.6 x 10~7 2.3 x ID"6
9/19 2.0 x 10~7 1.8 x l(f6
Upper bound
CPSC
7.3 x 10~7
5.6 x 10"7
f.g
lifetime risk
90th percent i le intake
FDA EPA
6.1 x 10~7 5.4 x 1C"6 1
4.3 x 10~7 3.8 x 10~6 1
CPSC
.7 x 1(T6
.2 x 1CT6
aFood intake obtained by multiplying the MRCA mean and 90th percent!le frequencies of eating occasions (14-day average. 1982-87 5-year Menu Census) (MRCA
1987) by the mean grams/eating occasion from USOA/NFCS. 1977-78 (Pao et al. 1982). Data are for the 2+ years age group, males and females, eaters-only
population.
bSee Table 15, footnote b for TEQ levels in the paper article.
cHean/90th percentile values.
See Table 15, footnote c. Because of the conservatism of the per capita estimate, we have selected 37 g/p/d to represent the mean food intake
eaters-only value as well. The 90th percentile value was assumed to be 2 times the mean. This is reasonably consistent with the relationship between the
mean and 90th percentile figures for the other entries in the table.
eNCASI estimated per capita consumption of 1.6 g/person/day based on 1.4 billion bags sold in the U.S. in 1989. To obtain the eaters-only mean intake, we
have assumed that all microwaveable popcorn is consumed by only 10X of the U.S. population. The 9Qth percent!le value was assumed to be 2 times the mean.
Assumes a typical body weight of 60 kg for an adult.
9EPA classifies 2.3.7,8-TCOO as a "B2" carcinogen.
-------
Table 10-4 presents the results of the cancer risk assessment for the
"eaters only - food-by-food basis." The maximum estimated risk for mean
consumers (eaters only) of any one of the food products is less than
1 4 x 10";? using FDA's cancer potency slope factor and less than
1.2 x 10"5 using EPA's cancer potency slope factor. The risks for 90th
percentile (eaters-only) consumers of individual foods are approximately
two times greater than risks for the mean (eaters-only) consumer.
10.6.2 Non-Cancer Risks
FDA used an ADI of 1-10 pg/kg/day to assess non-cancer risks of
potential exposures. This is the most sensitive non-cancer toxicological
endpoint associated with dioxin exposure in animal studies. Although the
estimated daily exposures in units of pg/kg/body weight/day are not shown
•in Tables 10-2 and 10-3, the calculated exposures were all less than the
ADI.
10.6.3 Uncertainties
The major uncertainties inherent in this assessment concern
assumptions regarding food intake rates and dioxin migration rates. The
estimation of exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF from consumption
Of all foxjds contacting paper articles requires consideration of
appropriate food intake information. Available nationally representative
food consumption data bases, however, do not provide information on
•Lhether foods are sold, held, heated, cooked, or served in contact with
Different types of materials. Therefore, for estimating the total dioxin
2rn intake from all foods that may contact bleached paper, available food
ronsumption data for average consumers of each of the foods were used in
conjunction with the assumption that all such foods have indeed been in
rontact with bleached paper prior to being consumed. This necessary
ssumption leads to an overestimate of consumption of food that has been
fn contact with paper and, hence, to an overestimate of total dioxin TEQ
exposure.
This upper-bound estimate of total dioxin TEQ exposure (12.7
/p/day) was adjusted through a comparison with a lower-bound estimate
Pfdioxin TEQ exposure, specifically, a per capita exposure, obtained
-------
population exposure analysis also apply to the "eaters-only" analysis.
Therefore, the dioxin TEQ intakes in Table 10-4 may also be considered as
upper-bound estimates of mean and 90th percent!le for "eaters" of each of
the foods. The dioxin TEQ exposures in Table 10-4 must not be summed
because the population of eaters is not the same for each food category.
All of the migration studies demonstrated detectable levels of
transfer of the dioxin congeners to the test foods. The temperature of
foods, their composition, the nature of the paper article, as well as the
dioxin congener levels in the paper articles were all found to influence
the extent of transfer to test foods. Although recent production lots of
paper and paperboard used in food application have been shown to have
significantly lower levels of dioxin congeners than the levels used for
this assessment and risks from these articles should be expected to be
lower, FDA has not used these recently submitted figures to try to
estimate lower risks because the quality and the completeness of these
reports have not been assessed and migration studies which might show the
extent of reduced exposure (and risk) have not been conducted.
Even considering these uncertainties, the results of the assessment
of individual lifetime cancer risks indicate a potentially significant
risk is posed by "current" (i.e., as used in this assessment) levels of
dioxin in paper food contact product. Although individual risks are not
high, the potentially exposed population is much larger than any other
exposed population group addressed in the Integrated Assessment.
10.7 References
API. iggo. American Paper Institute, Inc. Letter of 3-15-90. Cavaney,
R. to Shank, F.
Glinsman WH et al. 1986. J. Nutri., November 1986.
MRCA. 1988. Market Research Corporation of America. Report to the Food
and Drug Administration. Frequency distributions of the total number of
eating occasions in an average 14-day period, by eaters only produced
from the five menu census studies of July 1982 through June 1987 in
partial fulfillment of Contract #223-87-2088, Task II, Section C.1.D(2),
and Sections F.I.2(5-6). December. 1988.
NCASI. 1988. National Council of the Paper Industry for Air and Stream
Improvement. Assessment of the risks associated with potential exposure
to dioxin through the consumption of coffee filters brewed using bleached
Paper coffee filters. Technical Bulletin in No. 546.
NCASI. 1989a. National Council of the Paper Industry for Air and Stream
Improvement. NCASI procedures for the preparation and isomer-specific
analysis of pulp and paper industry samples for 2,3,7,8-TCDD and
2,3,7,8-TCDF. Technical Bulletin No. 551.
10-14
1590q
-------
NCASI. 1989b. National Council of the Paper Industry for Air and Stream
Improvement. Summary of results of the analysis of tetra through octa
dioxins and furans. Letter of 8-2-89. Gillespie, W. to Cramer, G.
NCASI. 1989c. National Council of the Paper Industry for Air and Stream
Improvement. Assumptions for analysis of exposures to dioxin from foods
contacting paper products. August, 1989.
NCASI. 1989d. National Council of the Paper Industry for Air and Stream
Improvement. First progress report NCASI milk carton migration study.
August, 1989.
NCASI. 1989e. National Council of the Paper Industry for Air and Stream
Improvement. Interim report NCASI hot beverage paper cup coffee exposure
scenario migration study. October, 1989.
NCASI. 1989f. National Council of the Paper Industry for Air and Stream
Improvement. Interim report NCASI hot beverage paper cup chicken broth
exposure scenario migration study. November, 1989.
NCASI. 1989g. National Council of the Paper Industry for Air and Stream
Tmorovement . Interim report NCASI dual ovenable tray migration study.
November, 1989.
NCASI. 1989h. National Council of the Paper Industry for Air and Stream
improvement. Interim report NCASI plates/trays/dishes migration study.
November, 1989.
NCASI. 19891. National Council of the Paper Industry for Air and Stream
Tmorovement . Interim report NCASI orange juice carton migration study.
December, 1989.
MCASI. 1990a. National Council of the Paper Industry for Air and Stream
Trnorovement . Summary of microwave popcorn migration study method
Development study data. Letter of 3-6-1990. La Fleur, L. to Cramer, G.
•urASl. 1990b. National Council of the Paper Industry for Air and Stream
Vmnrovement . Interim report NCASI milk carton migration study of half
half. January, 1990.
EM, Flemin KH, Guenther PM, Mickle SJ. 1982. Foods commonly eaten
\* individuals: amount eaten per day and per eating occasion. Home
?rononrics Research Report Number 44. Washington, D.C.: U.S. Department
~f Agriculture
rk YK and Yetley EA. 1990. Amer. J. of Clin. Nutri., May, 1990.
10-15
-------
Ryan JJ, Panopio LG, Lewis D. 1988. Bleached pulp and paper as a source
of PCDD's and PCDF's in foods. Paper presented at DIOXIN '88, The Eighth
International Symposium on Chlorinated Dioxins and Related Compounds.
Umea, Sweden, August 21-26, 1988.
Tiernan TO. Wright State University, Letter of 12-3-1987 to Amendola,
G., USEPA, Region 5.
USDA. 1984. U.S. Department of Agriculture. NFCS 1977-78 Report No.
1-2. 1984.
USDHHS/USDA. 1986. U.S. Department of Health and Human Services/U.S.
Department of Agriculture. Nutrition monitoring in the United States.
USFDA. 1989. U.S. Food and Drug Administration. Quantitative risk
assessment for dioxins in milk based on the FDA 1989 survey. Memorandum
dated August 2, 1989, Quantitative Risk Assessment Committee to R.J.
Scheuplein.
10-16
1590q
-------
ii ASSESSMENT OF RISKS TO THE GENERAL POPULATION FROM DIOXINS AND
FURANS IN CELLULOSE DERIVATIVES USED IN FOOD, DRUG, AND COSMETIC
FORMULATIONS
jl.l Introduction
This chapter is a condensed version of two reports prepared by the
Quantitative Risk Assessment Committee (QRAC) of the U.S. Food Drug
Administration (FDA) as part of the Interagency Dioxin-in-Paper Workgroup:
USFDA. U.S. Food and Drug Administration. 1990a. Carcinogenic risk
assessment for dioxins and furans in cosmetic products containing
cellulose derivatives produced from bleached wood pulp. Report of
the Quantitative Risk Assessment Committee. March 20, 1990.
USFDA. U.S. Food and Drug Administration. 1990b. Carcinogenic risk
assessment for dioxins and furans in cellulose derivatives used in
foods and ingested drug products. Report of the Quantitative Risk
Assessment Committee. April 24, 1990.
The Center for Food Safety and Applied Nutrition's (CFSAN)
Quantitative Risk Assessment Committee (QRAC) assessed the potential
risks to persons using various food, drug, and cosmetic products
rontaining cellulose derivatives prepared from bleached wood pulp that
may be contaminated with dioxin congeners. Information on exposure to
Hioxin congeners from the use of these products was provided to the QRAC
the Food and Color Additives Review Section of CFSAN.
.. * 9 Exposure to Dioxins and Furans from Use of Cosmetic Products
jj. C. -
11.2.1 Identity and Use of Cellulose Derivatives
Much of the information on cellulose derivatives usage in cosmetic
roducts was provided to CFSAN by the paper industry. Powdered cellulose
nd various cellulose ethers are used in a wide range of leave-on type
nsmetic products such as lotions, creams, and powders and in wash-off
C^oducts such as shampoos, conditioners, and non-fluoride toothpastes.
?he leave-on products are reported to contain less than 2 percent
2llulose derivatives, while wash-off products contain less than
f t>ercent. The cellulose derivatives that are used in cosmetic products
\ Jji into two different classes: water-insoluble cellulose derivatives
•eluding powdered cellulose and microcrystalline cellulose, and
ir>er- soluble, or dissolving cellulose derivatives, that include a
v*a iety of cellulose ethers and esters.
9 2 Dioxin Concentration in Cellulose Derivatives
i't'
Although analytical information is not available on the concentra-
ns of dioxin congeners in cellulose derivatives, information is
11-1
-------
available on the concentration of dioxin congeners in pulp that is
subsequently converted into the various cellulose derivatives. The
available data from'the 104-Mill Study show that none of the pulps used
to prepare cellulose-derived products that are added to cosmetic
formulations contain detectable levels of either 2,3,7,8-TCDD or
2,3,7,8-TCDF. Rather than assuming that these congeners are not present
at any level in the cellulose derivatives, FDA assumed their presence at
no greater than one-half of the average detection limit (range 0.2-1 ppt,
avg. 0.6 ppt) of the analytical method used, or about 0.3 ppt of dioxin
TEQ.
11.2.3 Cosmetic Product Use
The paper industry indicated that a wide range of cosmetic products
contains cellulose derivatives in their formulation; however, the
information was not product specific. Therefore, it was assumed that
(1) all cosmetic products applied to the skin are likely to contain such
derivatives and (2) each cosmetic product contains these derivatives at
the maximum level reported by the industry.
Information on the frequency of use and amount of cosmetic product
applied per use was derived from USFDA (1983). To account for the high
amount of cosmetic products used by some individuals, a 90th percentile
daily use rate for each of the cosmetic products was used. This rate was
derived by combining information on the average amount of cosmetic
product used per application with 90th percentile frequency of use. The
amount of cosmetic product used each day was multiplied by the use level
of the cellulose derivative and the assumed maximum concentration of
dioxin in the cellulose derivative to obtain an upper bound estimate of
the potential dioxin exposure from each cosmetic product. However, to
estimate probable dioxin exposure, it is necessary to account for the
extent to which wiping, wearing off, or washing remove the cosmetic
products from the skin (thus, decreasing the potential for exposure to
dioxin contaminants). Correction factors proposed by the Color Additive
Scientific Review Panel in September 1985 to describe the percent of the
cosmetic product that is available for absorption are used in the risk
calculations.
When information on the amount of cosmetic product used each day was
combined with information on the extent of cosmetic product that remains
available for absorption, a few cosmetic products were found to dominate
the potential for dioxin exposure. Rather than examining the potential
dioxin exposure that may occur from each cosmetic product, only those
cosmetic products thought to present the greatest potential for consumer
exposure to dioxin have been addressed. Examples are dentifrices, body
lotions, and hair shampoos. Other cosmetic products have either lower
daily use rates, lower levels of the cellulose derivative, or most of the
product is washed off following use. It was concluded that dioxin and
11-2
1597q
-------
furan exposures that may occur with other cosmetic products will be
subsumed within the exposures estimated to occur from dentifrices, body
lotions, and hair shampoos.
11.2.4 Dermal Absorption of Dioxin from Cosmetic Products Applied to
the Skin
Limited information is available on the rate of percutaneous
absorption of dioxin. Many factors affect this rate, such as the nature
of the cosmetic product, the duration of dermal contact, the amount of
dioxin deposited on the skin, the anatomic site, and the age and health
of the skin. The Consumer Product Safety Commission (CPSC) has assembled
a review of available information on studies of dioxin dermal transport.
CPSC reported that in in vitro experiments with human skin (Weber et al .
1989) and in absorption studies with rats (Poiger and Schlatter 1980), 19
to 29 percent solvent-deposited dioxin was absorbed. CPSC used
25 percent as a measure of the fraction of dioxin likely to be absorbed
by human skin 24 hours after dioxin contact. This value is used in this
assessment as well. Because of the limited amount of data on dioxin skin
absorption and the low levels of exposure estimated for individuals using
high amounts of cosmetic products, no attempt was made to develop
exposure estimates for individuals with damaged or aged skin.
11.2.5 Dioxin Exposure from Cosmetic Products Applied to the Skin
Table 11-1 contains a summary of the average daily TEQ exposures that
were estimated for users of dentrifrices, lotions, and shampoos as well
as the parameters used to calculate the exposures. Daily exposures were
pstimated to range from 0.0001 pg/person/day from use of shampoos to
0.005 pg/person/day from use of lotions.
jl.3 Exposure to Dioxins and Furans from Use of Cosmetic Met Wines
11.3.1 Identity and Use of Cosmetic Met Wipes
Cosmetic wet wipes used to wipe hands and in diaper changes are
manufactured using either synthetic fibers or bleached pulp. Information
„„ the composition and use of these wipes was provided by the National
Council of the Paper Industry for Air and Stream Improvement (NCASI
V<)89)- wiPes dl*Pe"sed from pop-up containers do not contain any pulp;
iowever, those which are dispensed from tubs in which the wipes lie flat
V« a stack are composed of bleached pulp (75 to 85 percent of dry weight)
ith the remaining material being binders and synthetic fibers.
potential exposure to dioxin congeners in the wet wipes made from
Beached pulp can occur as a result of migration from the web into the
?Ition during ext<;!]d^ storage, transfer of the lotion to the skin during
a"e"PtdUnn9 thi "m between Wet Wle usae
1J-3
[5971
-------
8913H
Table 11-1. Dioxin Exposure From Cosmetic Products Applied to the Skin
Product
type
Dentifrices
Lotions
Shampoo
Grams used
per day3
0.15
7.0
16.4
Cone, of
deriv.
in prod.
.Li*
_<_ZX
.Ll*
Percent
available
for
absorption
100
50
1
Percent
absorbed
100
25
25
Dioxin
TEQ
absorbed0
(pg/p/day)
0.0005
0.005
0.0001
a90th percentile (USFDA 1983).
bHart et al. (1986).
clt is assumed that the dioxin TEQ level in cellulose derivatives is
0.3 ppt.
11-4
-------
11.3.2 Dioxin Concentration In and Extraction From Wet Wipes
Analyses of pulp and web from two manufacturers have shown
2 3 7,8-TCDD levels in the 0.5 ppt to 4 ppt range and 2,3,7,8-TCDF levels
ranging from 2 ppt to 39 ppt (NCASI 1989). Due to the lack of analytical
data for all manufacturers of wet wipes, it was assumed that all of the
wet wipes are manufactured from pulp containing dioxin congeners at a
level equivalent to the average pulp value found in the 104-Mill Study
(17 ppt dioxin TEQ) rather than the highest level in webs analyzed by the
paper industry (8 ppt dioxin TEQ) (NCASI 1989).
Information on the extent of extraction of dioxin congeners from the
web during storage and on the skin penetration rate by dioxin is
currently being developed by the paper industry. However, the paper
industry has provided preliminary findings from experiments conducted
with two representative types of wipe lotions and two levels of
surfactants (1 and 3 percent) for periods up to 90 days. The partition
coefficients (the ratio of the concentration of the dioxin congener in
the pulp to the concentration in the lotion) for the dioxin and the furan
were approximately 900 for the 1 percent solution and approximately 300
for the 3 percent solution.
11.3.3 Wet Wipe Use Information
Total wet wipe use for six daily diaper changes was estimated to be
eight uses per day (NCASI 1989). Wet wipe use information for periods
other than during diaper changes has not been developed; it was assumed
that because of the high frequency and duration of wet wipe use during
vears of diaper changes that wet wipe use for other periods will be much
less frequent and will contribute little to total exposure. Therefore,
it was further assumed that infants will be exposed to wet wipes for a
neriod of three years and that primarily one adult will use wet wipes for
six years caring for an average of two children.
The amounts of lotion deposited on the skin of an infant during
diaper changes and on the hands of adults during these periods of wet
2ipe use are dependent on the area of exposed skin area and the duration
"f skin contact during wiping. NCASI (1989) has estimated the amount of
?0tion transferred to skin for each use scenario, accounting for such
factors as weight of lotion lost by evaporation during wiping, whether
Ihe wipe came from the top, middle or bottom of the tub, and differences
Between three different wet wipe manufacturers. The average amount of
Siuid transferred from wipes to the skin was 0.83 ml/166 in2.
NCASI (1989) also used the surface area of a medium-sized diaper
, 1 95 inz) *s a measure of the area of skin that is likely to be
iintacted by the lotion of a wet wipe. It was estimated that 0.98 ml of
n will be transferred to infants for each diaper change. Using the
11-5
-------
surface area of the palm and outstretched fingers of adult males and
females (23 in2 and 19.4 in2, respectively), it was estimated that
fluid transfer for males and females will be 0.117 ml and 0.097 ml per
wipe, respectively (NCASI 1989). For simplicity, this assessment is
based on conversion of these figures to percent of lotion in the wet wipe
that is left on the skin - 20 percent for infant use and 2 percent for
adult use.
11.3.4 Dermal Absorption of Dioxin Congeners
Studies of the effect of skin contact time on dermal absorption of
dioxin for children vs. adults have not been performed. Therefore, it
was assumed that all of the 2,3,7,8-TCDD and 2,3,7,8-TCDF in the applied
lotion is available for absorption. The absorption coefficient of 25
percent has been used as an estimate of the fraction of dioxin and furan
absorbed during the shorter exposure intervals associated with infant and
adult use of wet wipes. Using the 24-hour absorption coefficient for
modeling dermal uptake during shorter periods is assumed to be
conservative. Therefore, a correction factor to account for the
relatively fast dermal absorption rates of damaged skin that might be
experienced by infants with diaper rash was not used.
11.3.5 Dioxin Exposure from Met Wipes
Table 11-2 contains a summary of the average daily dioxin TEQ
exposures that were estimated for infants and adults for periods of wet
wipe use during diaper changes as well as the parameters used to
calculate the exposures. Daily infant exposures were estimated to be
0.13 picograms (pg) dioxin TEQ which is 10 times higher than the figure
(0.013 pg) calculated for adults.
11.4 Exposure to Oioxins and Furans in Cellulose Derivatives Used In
Foods and Drug Products
11.4.1 Identity and Use of Cellulose and Cellulose Derivatives
Cellulose (powdered and microcrystalline) and a variety of
cellulose-derived esters and ethers are used in food and drug
formulations to accomplish a variety of technical effects. The paper
industry has indicated that food uses of powdered cellulose include baked
goods (e.g., bread, cookies, rolls, crackers, pie fillings, icings),
dairy products (e.g., ice cream, whipped toppings, milk shakes), pasta,
sausage casings, diet beverages, tablets, and "miscellaneous items" such
as candy, dried fruits, and flavor carriers. Microcrystalline cellulose
and cellulose-derived ethers and esters are added to baked goods (similar
uses as powdered cellulose), dairy products and related substitutes,
pasta, tablets, fried foods, syrups, soups, sauces, dressings, and
11-6
1597q
-------
8913H
Table 11-2. Dioxin Exposure From Wet Wipes
User WP CP WL
PD K PA FU DA
Infants
Adults
2-1 17 6.0 20 300 25 8
2-1 17 6.0 2 300 25 8
0.13
0.013
OA = Dioxin TEQ absorbed/person/day in picograms (pg)
= WL x PD x CL x PA x FU
WL x PD x CP x WP x PA x FU
(K x WP) + Wl
WL = Weight of lotion per wet wipe in grams
PD = Percent of lotion deposited on the skin as decimal
CL = Concentration of dioxin TEQ in lotion (ppt)
CL = CP x WP
(K x WP) + WL
CP = Concentration of dioxin TEQ in web used for wet wipes (ppt)
WP = Weight of paper used for each wet wipe in grams
K = Partition coefficient of dioxin between paper and lotion
PA = Percent of dioxin on skin absorbed as decimal
FU = Number of wet wipes used per day
Sample Calculation for Wet Wine Exposure (Infants)-
DA VI- x PO x CP x WP x PA )r Fli
(K x WP) + WL
= (6.0 g lotion/wet wipe) x (0.20) x (17 x 10'12 g dioxin TEQ) x
(2.1 g pulp/wipe) x (0.25) x (8 wipes/day) divided by [(300 x 2.1) + 6.0]
= 0.13 x 10"12 g dioxin
TEQ
11-7
-------
"miscellaneous items" such as candy, dried fruits, and flavor carriers.
The paper industry also provided general information on the range of use
levels for these derivatives in most of the food categories. For these
applications, the technical effects accomplished by these cellulose
derivatives include those of an anticaking agent, formulation aid,
processing aid, stabilizer and thickener.
Cellulose derivatives are also used widely in the formulation of drug
tablets, suspensions, and creams. Although the cellulose derivatives
serve as inactive ingredients in nearly all of these applications,
methyl cellulose and sodium carboxymethylcellulose serve as active
ingredients when they are used as bulking agents in laxatives. Because
of the large amounts required to accomplish the bulking effects,
laxatives represent the greatest potential for exposure to cellulose
derivatives from drug products.
11.4.2 D1ox1n Concentration In Cellulose Derivatives
The available data from the 104-Mill Study show that none of the
pulps used to prepare cellulose-derived products that are added to foods
and ingested drug formulations contain detectable levels of 2,3,7,8-
TCDD. But, one of the mills reported low levels of 2,3,7,8-TCDF
(equivalent to less than 0.5 ppt dioxin TEQ). Rather than assuming that
these congeners are not present at any level in the pulps used to prepare
the cellulose derivatives, FDA assumed that they are present at one-half
of the average detection limit (range 0.2 - 1 ppt, ave. 0.6 ppt) of the
analytical method used (i.e., about 0.3 ppt of dioxin TEQ).
Further, FDA assumed that the chemical and mechanical processing
steps used to prepare the cellulose derivatives do not increase the
residual levels of dioxins and furans above the levels assumed to be
present in the bleached wood pulp used to manufacture these derivatives.
Using dioxin equivalency factors, total dioxin TEQ in cellulose
derivatives used in food and ingested drug products is estimated by FDA
to be no greater than 0.3 ppt.
11.4.3 Oioxin Exposure Estimates
(1) General approach used. One approach to the estimation of
possible exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF from use of cellulose
derivatives in food and drug applications involves the use of information
on the annual poundages of the cellulosics directed for use in foods and
drugs. Such information allows a per capita exposure estimate assuming
that total production of cellulose derivatives is evenly distributed
across the entire U.S. population. Per capita exposures, however, tend
to underestimate exposure to individuals consuming those items in which a
particular substance is most heavily used, or to individuals who consume
many products at high levels in which the substance is used. Per capita
11-8
1597q
-------
exposures (or per capita disappearances) can be useful, however, as a
check or validation of other types of exposure estimates that more
accurately reflect the characteristics (e.g., age, sex, ethnicity) of the
exposed population.
For the current problem, the paper industry has provided data only
for total bleached wood pulp-derived cellulose (59,000 tons) of which
approximately 25 percent was reported to go towards foods, drugs, and
cosmetics. Furthermore, the MAS 1987 food additive/ingredient poundage
survey of industry contains information on annual poundages of a number
of cellulosics directed into the food supply. These data are known to be
underreported. Therefore, FDA has foregone the presentation of a per
capita exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF and has based the
assessment on food consumption survey data in the case of exposure to
in TEQ from foods containing cellulosics and on information from the
Center for Drug Evaluation on drug use.
(2) Exposure estimates for foods. The estimation of exposure to
2,3,7,8-TCDD and 2,3,7,8-TCDF from consumption of foods containing
cellulose derivatives requires consideration of the appropriate food
intake information and information on the specific foods to which
cellulosics may be added. The paper industry has provided some
information on the food uses of cellulosics. Additional information on
use is contained in the National Academy of Sciences 1977 Survey of
Industry and in agency memoranda in CRP 4T0133. In general, cellulosics
are used in a wide variety of food products and can be considered to be
ingested by a broad segment of the U.S. population- -with perhaps one
exception- -aJ/jfca-cellulose used to make the "high fiber" breads.
These breads are selected by consumers for their high fiber, i.e., their
cellulose content.
The use of available nationally representative food consumption data
for those foods to which cellulosics would be added, when combined with
appropriate use level information, will lead to an overestimate of
exposure to cellulose derivatives and, hence, to an overestimate of total
dioxin TEQ exposure. This is because of the necessary assumption that
al1 foods which may contain a cellulosic do contain a cellulosic.
„ Because we believe that 100 percent of the U.S. population are
eaters" of at least one food containing a cellulose derivative on a
Sequent or chronic basis, it was assumed that 100 percent of the U.S.
Population are "eaters" of 2,3,7,8-TCDD and 2,3,7,8-TCDF. I^fore,
estimates of food intake used to estimate total dioxin TEQ intake should
°e representative of the total population. We selected the Market
^search Corporation of America (MRCA) database, the five-year Menu
Lensus, 1982-87 as our source of food intake data. This data base
c°ntains 14 consecutive days of data and is likely to capture a
Representative dfet including both frequently and infrequently consumed
*
11-9
-------
Total-sample intakes for each food of interest in the MCRA data base
were multiplied by the use levels of the appropriate cellulose derivative
to obtain the intake estimate for the cellulose derivative. The intake
estimate for each cellulosic was multiplied by its dioxin TEQ
concentration to obtain the set of dioxin TEQ intakes that could be
summed to provide an upper-bound total population mean dioxin TEQ intake,
i.e., 1.4 pg/person/day (see Table 11-3).
As seen from Table 11-3, the prime contribution to total mean dioxin
TEQ exposure derives from baked goods, and bread, in particular. Because
of the high probability that cellulose-containing bread would be
selectively consumed by individuals interested in high-fiber foods, it is
evident that for those individuals who do not consume high-fiber bread,
the total mean exposure to dioxin TEQ of 1.4 pg/person/day represents a
substantial overestimate. For those consumers who do select for
high-fiber bread, however, we have looked at potential dioxin TEQ intake
using those individuals identified as "eaters" of bread during the 14-day
MRCA survey. We have assumed that an eater of high-fiber bread consumes
this product at the same level as an eater of "regular" bread, because
the survey data available to us do not contain intake data specific to
high-fiber bread. Thus, the eaters-only bread intakes (white bread)
are: 41 g/person/day (mean) and 73 g/person/day (90th percentile).
Assuming the bread contains 7.5 percent a7p/?a-cellulose with a
dioxin TEQ level of 0.3 ppt, the eaters-only dioxin TEQ exposures are:
0.92 pg/person/day (mean) and 1.6 pg/person/day (90th percentile). The
eaters-only mean exposure of 0.92 pg/person/day is essentially the same
as that for the total-sample value given in Table 11-3 of 0.90
pg/person/day. This is because the percentage of caters of bread in the
MRCA survey is essentially 100 percent.
Although it is evident that the total population dioxin TEQ exposure
is highly conservative, it may not be as evident for the exposure
estimate for those individuals selectively consuming high-fiber bread.
Several factors should be noted that reflect the conservatism of this
exposure as well. First, the assumed use level of a/p/?a-cellulose
(7.5 percent) is believed to be a maximum level rather than a typical
level. Second, it is assumed that all bread consumed by an individual is
high-fiber bread. Third, it is assumed that a consumer of high-fiber
bread consumes this bread at the same rate as one who eats regular
bread. Fourth, the concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
all the cellulose derivatives are expected in nearly all instances to be
substantially lower than the 0.3 ppt value we have used. The current
industry trend appears to be toward ever lower levels as industry more
carefully controls dioxin levels in its various manufacturing processes.
(3) Exposure estimates for drugs. For drug uses of cellulosics,
available information indicates that the major uses are either as an
inactive ingredient as a binder in drug tablets or as an active
11-10
1597q
-------
8913H
Table 11-3. Total Sample Basis Dioxln TEQ Exposure from Foods
Food use
(cell, deriv.
Cell, deriv.
level in food Food Intake0' Cell, deriv. intake Dioxin TEQ intake
(g/P/d) (g/p/d) (mg/p/d) (fg/p/d)
! Bread, wht. or dk.
(alpha)
2 Eng. muf.. bagels
(alpha)
3 Crackers
(alpha!
4 Stuffings
(alpha]
Diet baked goods
(alpha)
Diet baked goods
(HPMC)
Other baked goods mixes
(alpha)
a Cakes
8 (SCMC)f
Oonuts
(SCMC)
10. «'"•
(EC)9
Non-stnd. dressing
(SCMC)
Oiet fats, oils
1 (SCMC)
juice drinks, ades
1 (SCMC)
Other frt. juices
(MO
7.5
0.4
0.1
0.1
0.6
0.06
0.06
0.03
0.5
39
3.4
2.1
6.5
0.7
07
12
11
7.7
12
4.5
1.7
8.6
4.3
2.900
170
110
300
40
600
11
7.7
69
2.6
1.0
2.5
900
51
33
90
12
0.9
180
3.3
2.3
21
0.8
0.3
0.8
6.6
11-11
-------
8913H
Table 11-3. (Continued)
Food use
(cell. derw.)a
Cell, deriv.
level in food Food intake1*'
(g/P/d)
(9/P/d)
Cell, deriv. intake Oioxin TEQ intake
(mg/p/d) (fg/P/d)
15. Dried wht. potatoes 0.5
(HPMC)
16. Sugar panned candy 0.2
(HPC)
17. Diet swt. sauces 1.2
(SCMC)
18, Dry beverage mixes 0.01
(SCMC)
19. Diet beverages non-cola 0.02
(SCMC)
20. Toppings, dairy 0.4
(HPMC)
21. Coffee whiteners 1.3
(alpha)
22. Heat products 0.1
(alpha)
23. Nut products 0.6
(alpha)
24. Other foodsh 0.1
(alpha)
0.6
7.0
0.13
0.27
9.5
0.4
0.64
1.7
6.9
2.9
200
1.6
2.7
1.9
1.5
83
1.7
40
ZOO
TOTAL = 1.407 fg/p/d or 1.4 pg/p/d
0.9
4.2
0.5
0.8
0.6
0.5
25
0.6
12
60
a Cellulose derivative abbreviations:
alpha = alpha-cellulose (mtcrocrystal1ine and powdered cellulose share many of the same food uses; we have
assumed substitutabi1ity)
HPMC = hydroxyprcpylmethyIcellulose
SCMC = sodium carboxymethyIcellulose
EC = ethylcellulose
MC = methylcellulose
HPC = hydroxypropylcellulose
11-12
-------
Table 11-3. (Continued)
Footnotes (continued):
b e levels for SCMS, MC, HPMC, EC, and HPC are from the NAS 1977 Survey of Industry. Use levels for alpha-
llulose are from memoranda and information in GRP 4T0133.
c n\ tained by multiplying the MRCA mean frequency of eating occasions (14-day average, 1982-87, 5-year Menu
sus) by the mean grams/eating occasion from USOA/NFCS, 1977-78. Data are for the 2+ years age group,
]es and females, total sample populations.
J ms/eating occasion for numbered items: 1-44.2; 2-57.2; 3-20.0; 4-65.7; 5-37.7; 6-37.7; 7-105; 8-79.6;
Gr"67 8; 10-146; 11-31 5; 12-18.3; 13-276; 14-142; 15-164; 16-51.3; 17-25.8; 18-277; 19-303; 20-13.5; 21-8.2;
22-59.2; 23-40.0.
e The
nioxin TEQ level in all cellulose derivatives was taken as 0.3 ppt. This is one-half of the nondetect
1 (0.6 ppt) for analytical methods used to measure dioxin congener levels in the pulps used to manufacture
ce11ulose derivatives.
MC arid carboxymethylethylhydroxycellulose are assumed interchangeable.
d methylethyIcellulose are assumed interchangeable.
9 €.£•
rding to a memorandum of 8/24/81 (J. Modderman, Ph.D. to L. Gosule, GRP 4T0133), alpha-cellulose
onsiderecl technologically useful in a number of food categories at use levels no greater than
ercent. It was assumed that the Individual dally Intake for these "other foods" would total to
. this represented an "eaters only" value. We have Included this value (200 g) in our tabulation. The
-n TEQ intake from this "category" is insignificant so no effort was made to correct the total sample
11-13
-------
ingredient as a bulking agent in laxatives. A review of FDA files of
drug formulations showed that when various cellulose derivatives (e.g.,
powdered cellulose, microcrystalline cellulose, hydroxypropylmethyl cellu-
lose, and ethyl cellulose) are used as binders of tablets that may be
consumed on a chronic basis, daily ingestion of the cellulose derivatives
will total less than 1 g/person/day. In contrast, the use of either
methyl cellulose or sodium carboxymethylcellulose in laxatives can result
in daily doses of either of these derivatives as high as 6 g/person/day
(Handbook of Non-Prescription Druos. 7th ed., 1982). Although the
recommended use of laxatives containing cellulosics is for a maximum
period of just 2 weeks, the FDA Center for Drug Evaluation has indicated
that elderly individuals (60-70 years) with reduced gastric motility may
use such laxatives on a chronic basis. However, this would involve less
than lifetime exposure.
For estimating the upper-bound intake of dioxin TEQ from drugs, we
have provided separate estimates for the maximum exposures that may occur
from tablet binders and laxatives (Table 11-4). While both estimates are
expected to overestimate actual exposure because they are based on
assumed dioxin TEQ levels in the derivatives of half of the detection
limit, the exposure estimated for laxative use has an additional
exaggeration because lifetime use of such products is not expected.
11.5 Cancer Risk Estimates
Table 11-5 summarizes the estimated exposures (normalized for body
weight) and lifetime cancer risks for cosmetic products, wet wipes, and
food and drug products containing cellulose or cellulose derivatives.
Risk estimates were calculated using the cancer slope factors derived by
FDA, EPA, and CPSC for 2,3,7,8-TCDD. The slope factors, which were
derived from the results of animal experiments where rats were exposed
to 2,3,7,8-TCDD in the diet, are expressed in terms of administered
dose. EPA and FDA have estimated that 55 percent of 2,3,7,8-TCDD
administered to rats in the diet is absorbed into the body. CPSC has
estimated that 75 percent is absorbed. Therefore, in Table 11-5, the
upper bound lifetime risks were corrected for oral bioavailability where
appropriate, i.e., in the case of lotion, shampoo, and wet wipe (adult
and infant) use.
Also, although non-cancer risks were not assessed in the source
document, comparison of the daily exposure estimates in Table 11-5 with
the estimated RFD of 1 pg/kg/day and the estimated health advisories of
10 to 300 pg/kg/day that were developed by EPA for the purposes of this
assessment indicates that the exposures pose minimal risk of non-cancer
effects.
11.6 Analysis of Uncertainties
It should be emphasized that the risks presented in Table 11-5 are
not additive. Several conservative assumptions about exposure have been
11-14
1597q
-------
8913H
Table 11-4. Upper Bound Intake of Dioxin TEQ from Drugs
Upper-bound intake
dioxin TEQ
application (g/person/day) (pg/person/day)
Cellulose product9 Cellulose product intake dioxin TEQ
Tablet binders _<_ 1
Laxatives 6
0.3
1.8b
a It is assumed that the cellulose products contain dioxin TEQ at
0.3 ppt, I.e., one-half of the non-detect level (0.6 ppt) of analytical
methods used to measure dioxin congener levels in pulps used to manu-
facture the cellulose derivatives.
k Estimated intakes of cellulose derivatives and dioxin TEQ from
laxatives exceed lifetime average exposure because chronic use of the
laxatives only occurs for a fraction of a lifetime.
11-15
-------
I.iblf II *• i*pprr Bourn) ( jrt inoqrn u
Cellulose Derivative
i-.k (or 'l-,pr-. ^f I xxi. i)ru)11
S x 10 3 H x 10 " / 1 x 10 '
3 x 10 Z 0 x 10 7 4 4 x 10 f>
1 l> < 10 ''
4 ', * 10 "
1 1 « 10 (>
') b < 10 '
! ! x 10 '
1 4 * 10 B
in parenthesis denotes the daily tlose dur\nq tfie period of exposure (i e . 3 yearr, for infant and h /ear-, lor adult)
Salue in parenthesis denotes exposure and risk, lor 00th percent i le consumption rale
°tPfc has classified ? .3.7 ,8-KDO as a "B?" carcinoqen
-------
made earlier in this report which affect the risk estimates. First,
dioxin congeners have been assumed to be present at one-half the
detection limit when such congeners may be present at a lower level or
not present at all. Second, products which may contain a cellulosic have
been assumed to do so. Third, while 100 percent of the U.S. population
is likely to consume regularly at least one food containing a cellulose
derivative, it is unlikely that 100 percent of the population would
consume all foods containing a cellulose derivative. Fourth, for the
high-fiber bread consumer, such breads have been assumed to contain a
maximum, rather than a typical, level of cellulose; it has also been
assumed that the high-fiber bread consumer eats only high-fiber bread and
at the same rate as one who eats regular bread. Fifth, exposure to
laxatives and tablets containing cellulose tablet binders would be less
than lifetime. The risk figures presented in Table 11-5 are for lifetime
exposure since FDA does not believe that methodology currently exists to
accurately adjust these figures for less-than-lifetime exposure.
11.7 References
Fries 6, Marrow G. 1975. Retention and excretion of 2,3,7,8-TCDD by
rats. J. Agric. Food Chem., 23:265-269.
Hart R, et al. 1986. Final report of the Color Additive Scientific
Review Panel. Risk Anal. 6(2):117-154.
Kociba RJ, et al. 1978. Results of a two-year chronic toxicity and
oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats.
Toxicol. Appl. Pharmacol. 46:279-303.
NATO. 1988. North Atlantic Treaty Organization. Pilot study on the
international information exchange on dioxins and related compound.
Report No. 178.
NCASI. 1989. National Council of the Paper Industry for Air and Stream
Improvement. Interim report for exposure parameters for wet wipe usage.
July 1987.
Poiger H, Schlatter C. 1980. Influence of solvents and absorbents on
dermal and intestinal absorption of TCDD. Food Cosm. Toxicol.
18:477-481.
USFDA. 1983. U.S. Food and Drug Administration. Summary of the results
of the amount and frequency of use of cosmetic products by women. Food
and Drug Administration, Center for Food Safety and Applied Nutrition,
Color Additive Master File No. 9.
USFDA. 1989. U.S. Food and Drug Administration. Food Additive Petition
(FAP) OT4192.
11-17
1597q
-------
Weber LWD, Zesch A, Rozman K. 1989. Penetration of TCDD into human
skin in vitro. Ann. Mtg. Soc. Toxicol., Atlanta, Georgia, February 27
March 3, 1989. Abst. No. 472.
WHO. 1983. World Health Organization. Guidelines for the dietary
intakes of chemical contaminants. Geneva, Switzerland: Gobal
Environmental Monitoring System, pp. 19-50.
11-18
1597q
-------
12 FDA ASSESSMENT OF RISKS FROM EXPOSURE TO DIOX1NS AND FURANS IN
FISH CONTAMINATED BY BLEACHED KRAFT PULP AND PAPER HILLS
12.1 Introduction
This chapter is a condensed version of a report prepared by the
Quantitative Risk Assessment Committee (QRAC) of the U.S. Food and Drug
Administration (FDA) as part of the Interagency Dioxin-in-Paper Workgroup:
USFDA. U.S. Food and Drug Administration. 1990. Carcinogenic risk
assessment for dioxins and furans in fish contaminated by
bleached-paper mills. Draft Report of the Quantitative Risk
Assessment Committee. January 19, 1990.
Historically, FDA has dealt with hazards and risks of contaminants in
fish as part of the agency's mission to protect the public from food-borne
hazards. Fish collected in the vicinity of pulp mills have been found to
contain levels of dioxins and related furans that are substantially higher
than those levels in fish caught in areas remote from the mills. The
copulations judged by QRAC to be most at risk are subsistence fishers and
sports fishers since they are likely to consume above-average amounts of
fish on a regular basis.
FDA had previously addressed the issue of dioxin exposure resulting
from the ingestion of fish contaminated with dioxin. As part of congres-
sional testimony on the issue of a Great Lakes fish advisory, on June 30,
1983. Dr. Sanford Miller, Director of the Bureau of Foods, presented
before the Subcommittee on Natural Resources, Agriculture, Research and
Environment, Committee on Science and Technology, a safety assessment in
cUPP°rt °f an advisoiry tnat expressed concern about persons consuming fish
n a chronic basis containing dioxin at greater than 25 parts per trillion
(ppt).
It is important to note that QRAC does not have accurate information
the numbers of subsistence and sports fishers in the vicinity of pulp
ls. QRAC also lacks accurate fish consumption data for these persons,
s well as a thorough profile of dioxin congener levels in fish living in
Caters near pulp mills. In an attempt to bridge these gaps in knowledge,
nRAC has used estimates discussed in detail in the Food and Color
Additives Review Section (Center for Food Safety and Applied
idutrition-CFSAN) memorandum of November 13, 1989. (Memorandum, G.
rramer, Ph.D. to S. Henry, Ph.D.). These estimates will be summarized in
following discussion.
Levels of Dioxin Congeners In Fish
*__
EPA's Environmental Research Laboratory in Duluth, Minnesota, recently
nducted the National Bioaccumulation Study (USEPA 1989a). This study
d**
12-1
-------
involved the analysis of fish tissue for dioxin and furan (and about 65
other pollutants) in whole body and fillet portions of bottom-feeding and
predator fish collected at 400 locations across the U.S. Levels of
2,3,7,8-TCDO and furan and other isomers and congeners were reported as
dioxin toxic equivalents (TEQ). This study confirmed earlier findings
that fish obtained from waters in the vicinity of pulp mills contain high-
er levels of dioxin and furan than fish caught in areas remote from the
mills (USEPA 1988).
EPA's fish analyses included a variety of predator and bottom-feeding
fish, such as bass, walleye, perch, trout, catfish, carp, and buffalo
fish. The highest levels of dioxin TEQs in fillets were found in perch
(24.3 ppt), trout (22.5 ppt), crappie (22.1 ppt), and large mouth bass
(20.4 ppt); the lowest values were reported to be less than 0.1 ppt.
The levels of dioxin and furan congeners in fish samples as part of
the National Bioaccumulation Study, although limited, offer the best
available measure of likely levels of dioxin congeners in fish near pulp
mills. Only a few samples of fish were collected at most of the mills and
most analyses were done on whole fish. Since dioxin and furan congeners
concentrate in the organs of the fish and people typically eat cleaned
fish without the organs, the results of fillet composites from the few
samples collected in the region of pulp mills have been used. Dioxin
TEQs were calculated using EPA's 1987 interim procedures for estimating
risks from exposure to mixtures of dioxins and furans (USEPA 1989b).
Calculating dioxin TEQs using the EPA's March 1989 revised procedures
would not substantially change the TEQ values.
Although some fishers probably consume on a chronic basis fish caught
downstream from only one of these mills, QRAC does not believe that
there are adequate fish data for the individual mills to warrant a mi 11-
by-mill analysis of possible exposure to dioxin and furan congeners.
Therefore, all the fish fillet data have been combined in order to develop
an average concentration of TEQs in fish located in the region of pulp
mills. By multiplying the levels of TEQs in 47 fish fillet composites by
the number of fish in the composite, summing these values, and dividing
by the total number of fish analyzed, an estimated weighted mean level of
6.5 ppt TEQs in fish downstream from pulp mills is obtained, as shown in
Table 12-1. The highest levels in fish fillets reported above are about
3 times higher, i.e., .about 20 ppt.
12.3 Sources of Information on Fish Intake
In general, quantitative information on fish consumption by individ-
uals is limited. Additionally, data on consumption by specific subgroups,
such as sports fishers and subsistence fishers and their families, are
virtually nonexistent. Reliable data on fish consumption have been
difficult to obtain for a variety of reasons. Within the general popula-
12-2
1595q
-------
9029H
Table 1E-1. Dioxin TEQ Intake by Subsistence and Sports Fishers
Fisher
Subsistence
Average
90th percent ile
Sports
Average
90th percent 11e
Fish intake
(g/person/day)
69
116
13
39
Dioxin
TEQ levels1
(ppt)
6.5, 20
6.5, 20
6.5. 20
6.5. 20
Dioxin
TEQ intakes2
(pg/person/day)
450, 1380
750. 2320
85, 260
254, 780
In fish fillets: first value is weighted mean level and second is
highest level.
First value assumes weighted mean level in fillets and second assumes
highest level In fillets.
12-3
-------
tion, finfish and shell fish are eaten sporadically and selectively.
Therefore, fish intake is often not adequately reflected in established
and widely used national consumption surveys which provide data for
relatively short time periods, usually 3-14 days. More importantly, the
commonly available nationally representative surveys contain very few
persons who are representative of the at-risk subgroups, i.e., sports
fishers or subsistence fishers. Furthermore, descriptions of the fish
products consumed often are not sufficiently detailed to allow for
estimation of fish intake on the basis of species or source. For some
foods, data on national availability or production of the food can be used
to derive per capita availability estimates which can be helpful in esti-
mating intakes by individuals when specific information on consumption is
unavailable. However, there are currently no reliable sources of general
poundage data relative to non-commercially caught freshwater fish. There-
fore, the available data have been considered, and consumption levels have
been estimated using "common sense" scenarios.
Three established data bases which contain information on fish con-
sumption were initially reviewed. These data bases were: The Fish Con-
sumption Study conducted during 1973-74 for the Tuna Research Institute
by NPO Research, Inc.; the Nationwide Food Consumption Survey (MFCS) con-
ducted by the U.S. Department of Agriculture; and the 1977-78 Menu Census
conducted by the Market Research Corporation of America (MRCA). These
surveys were carried out using different methodologies and for different
periods of time, but they all were based on nationally representative
samples and were not targeted to certain at-risk groups, e.g., subsistence
and sports fishers.
The NPD data indicate that intake of all freshwater fish species aver-
aged 20.2 g/person/day and the 90th percentile averaged 41.7 g/person/day.
The NFCS reports a mean and 90th percentile intake for finfish and shell-
fish combined (all age categories) of 48 and 94 g/person/day, respective-
ly. The corresponding intakes from this survey for finfish, other than
canned, dried, and raw, are 54 and 96 g/person/day. Data from the MRCA
survey indicate that consumption of fish (fresh or frozen) averaged
14.3 g/person/day with a 90th percentile of 25.3 g/person/day.
The data from these surveys are reflective of the national consumption
levels for primarily marine species which were commercially obtained.
These data were not considered directly useful for risk estimation
purposes because intakes relative to freshwater sports and subsistence
fishing were lacking.
Three other sources of data on fishing practices and consumption were
available: a 1982 study by Puffer et al.; a Canadian study of fishermen
conducted in 1984 by the Ontario Ministry of the Environment; and a U.S.
Department of Interior data base on fishing and hunting practices in the
U. S. (USDOI 1988).
12-4
1595q
-------
The first two of these sources can be characterized as relatively
informal surveys, the results of which have been reported by others. The
third source considered was not oriented toward food consumption. There-
fore, the reliability and appropriateness of these data for QRAC purposes
are not clear and the data were reviewed from this perspective.
The Puffer et al. (1982) study conducted for EPA was reportedly based
on interviews with 1,059 fishers in southern California. According to
the report, 50 percent of the sample fished 1-2 times per week and
14 percent reported fishing 3-7 times per week, thus suggesting that the
sample may have contained a mixture of sports fishers and subsistence
fishers. The self-reported intake of fish by these persons was 37 g/
person/day for the median intake and 225 g/person/day for the 90th percen-
tile of intake.
The information available to FDA from the Ontario Ministry of the
Environment survey of fishers did not include information on survey
methodology. The fishers reported eating locally caught fish once every
21 days, on average. The mean amount of fish consumed per eating occasion
was reported to be 10 ounces, which reflects an intake of 13.8 grams per
day. About one quarter of the respondents reported consuming an average
of one pound of fish or more per eating occasion which results in an in-
take of about 21.6 g/person/day.
The U.S. Department of the Interior did not obtain consumption data
in their data collection; their report indicates that U.S. sports fishers
average 21 days of fishing per year. In order to use this information to
estimate consumption, assumptions about the frequency of fishing and the
amount of fish consumed per day of fishing are needed.
12.4 Estimation of Fish Intake bv Subsistence and Snorts Fishers and
Their Families
12.4.1 Subsistence Fishers
Those persons turning to local waters on a regular and frequent basis
as their predominant source of flesh food are expected to consume consid-
erably more fish than Individuals with a broader source of food. In the
typical U.S. diet, the predominant form of flesh food 1s red meat as
documented by several national consumption surveys. Thus, in order to
eStimate intake of fish among subsistence fishers and their families, a
••common sense" scenario has been derived by assuming that fish intake
among this group is equal to the consumption of red meat among the general
nopulation. In order to estimate intake, data on red meat consumption
JJere obtained from the 1977-78 MRCA survey. The MRCA intake figures are
based on food consumption during a 14-day survey period, and therefore
^re likely to provide a reasonable estimate of chronic meat Intake. Mean
f.ed meat consumption for all age groups in the MRCA survey was reported
12-5
-------
to be 69 g/person/day and the 90th percent!le intake was reported to be
116 g/person/day. Chronic fish consumption by subsistence fishers and
their families is estimated to be:
mean consumption 69 g/person/day
90th percentile consumption 116 g/person/day
As expected, for both mean and 90th percentile values, these figures
range from 1.2 to 4.8 times greater than the amount of fish estimated to
be consumed by the general population based on national surveys.
12.4.2 Sports Fishers
Fish intake by sports fishers is likely to be lower than fish intake
by subsistence fishers, but the intake by sports fishers may be quite
variable. Given the lack of data concerning fish consumption by sports
fishers, a "common sense" scenario based on fishing practices as re-
ported by the U.S. Department of the Interior has been developed. Their
1985 survey indicated an average of 21 days of fishing among sports
fishers. It was assumed that each day of fishing would result in one
eating occasion. It is likely that some fishing trips would result in no
catches, while others could result in a large catch of fish which could
be taken home, frozen and eaten later, if at all. But, in any case, one
eating occasion per trip was considered reasonable. Therefore, frequency
of eating recreationally caught fish was estimated to be 21 times per
year. In order to estimate serving size, data from the U.S. Department
of Agriculture's 1977-78 NFCS were used. The 90th percentile of serving
size for filleted finfish was reported to be 8 ounces (227 g) and was the
serving size value used in this scenario because the perishable nature of
fish is likely to encourage large serving size. This serving size value
is also in line with the 10-ounce mean serving size reported in the
Canadian study as documented by EPA. Eight ounces of recreationally
caught fish consumed 21 times per year results in an intake of 13.1 g/
person/day. This value is very close to the Ontario study which reported
13.8 g/person/day. For QRAC's purposes, the value of 13 g/person/day
derived from this scenario will be used. The available data did not allow
a direct estimate of a 90th percentile of intake for fishers and their
families who are more frequent consumers of recreationally caught fish.
However, a World Health Organization document (MHO 1983) suggests that
for most foods the 90th percentile level of consumption is approximately
2.5 to 3 times the value for mean consumption, and use of this factor in
the absence of data has been CFSAN policy for some time. The value of
3 times the mean estimated consumption relative to intakes by sports
fishers was used. About one quarter of the respondents in the Canadian
study consumed at least 21.6 g/person/day which is slightly less than and
consistent with the 90th percentile estimated via the CFSAN scenario. We
estimate that chronic fish consumption among sports fishers is:
12-6
1595q
-------
mean consumption 13 g/person/day
90th percent!le consumption 39 g/person/day
Given the lack of available data, estimating consumption by family
members of sports fishers was felt to be unjustifiable. It was assumed
that lifetime consumption for family members would be no higher and most
probably lower than the estimate for the sports fishers because of the
possibility that at least some or all of the catch would be consumed by
the fisher during the fishing expedition. Also, it should be noted that
the CFSAN scenario assumes that sports fishing practices and there-
fore consumption are constant throughout the life span. Fishing practices
may not necessarily remain constant. But, because there are no data which
would allow for a reasonable adjustment of the estimate, the assumption is
used.
12.5 Risk Assessment for Cancer
Levels of TEQs in fish fillets have been given earlier, i.e., an esti-
mated weighted mean level of 6.5 ppt TEQ in fish fillets and the highest
levels in fish fillets reported of about 20 ppt. The fish consumption
figures for subsistence and sports fishers have been discussed above.
The estimates of exposure to dioxin TEQ of subsistence and sports fishers
are presented in Table 12-1.
FDA has used a level of 0.064 pg/kg/day of 2,3,Z,8-TCDD as the level
tfhich will result in at most a lifetime risk of 10"6 of cancer. This
equates to a carcinogenic unit risk of 16xlO"6 risk for an intake of
I pg/kg/day of 2,3,7,8-TCDD. This level is based on a linear-at-low-dose
extrapolation from animal bioassays. EPA and CPSC have estimated this
level to be 0.006 pg/kg/day and 0.015 pg/kg/day, respectively, based on
similar modeling of animal bioassay data (see Section 3 of this report for
more details). Using the dioxin TEQ intakes and fish consumption figures
presented in Table 12-1, the upperbound lifetime risks to subsistence and
sports fishers are presented in Table 12-2.
Interpreting these risk figures is very difficult. Risks appear to be
^igh for subsistence fishers consuming average amounts of fish contami-
nated with dioxin TEQ at the weighted mean level; risks for sports fishers
are somewhat lower. The risks calculated in Table 12-2 must be seen in
the context of the following factors.
First, epidemiology studies have not as yet indicated increased risk
of cancer for humans exposed to dioxins; such studies are equivocal at
nest. Second, the risks presented in Table 12-2 are upperbound risks
hased on exposure data which would not normally be considered by the
agency as adequate for use in a quantitative risk assessment. The popula-
tions of sports fishers and subsistence fishers are poorly defined; the
numbers of fish samples collected for analysis for dioxin were too small
12-7
-------
9029H
Table 12-2. Cancer Risk for Subsistence and Sports Fishers
Dioxin
Fish TEO intake2
intake (pg/kg/day)
Fisher
Subsistence
Average
90th percent ile
Sports
Average
90th percenti le
(q/ka/dav) mean
1.2 7.5
1.9 12.5
0.22 1.4
0.65 4.2
hiqh
23
38.7
4.3
13
Upoer bound 1
FDA
mean
1.2xlO~4
2.0xlO"4
2.2xlO'5
6.5xlO'5
hiqh
3.6xlO"4
6.0x10"*
6.7xlO"5
2.0xlO'4
ifetime risk
EPA
mean
l.ZxlO"3
2.1xlO"3
2.3X10'4
7.0xlO'4
hiqh
3.8xlO~3
6.4xlO"3
7.2xlO'4
2.2xlO'3
3
CPSC
mean hiqt
S.OxlO"4 l.SxlO"3
8.3xlO~4 2.6xlO"3
9.3xlO"5 2.9xlO~4
2.8xlO'4 8.7x10'
2
From Table 12-1, Subsistence (average): 69 g/person/day / 60 kg body wt. = 1.2 g/kg/day. Subsistence
(90th percent! le): 116 g/person/day / 60 kg body wt. - 1.9 g/kg/day. Sports (average): 13 g/person/day / 60 kg
body wt. = 0.22 g/kg/day. Sports (90th percentile): 39 g/person/day / 60 kg body wt. = 0.65 g/kg/day.
First value assumes a weighted mean level (mean) of 6.5 ppt in fillets and the second assumes the highest level
(high) in fillets of 20 ppt.
Sample calculation: [(Daily intake of 23 pg/kg/day for average subsistence fisher) + (FDA daily TEQ dose
equivalent to 10'6 cancer risk: 0.064 pg/kg/day)] x (10~6) = (Z3/0.064) x (10'6) = 3.6 x 10'4
12-8
-------
to represent an adequate profile of dioxin congeners in fish near pulp
mills. Third, the risk calculations are based on linear-at-low-dose,
no-threshold extrapolations from animal bioassay data; the risk calcula-
tions may be several orders of magnitude greater than the actual risk.
FDA has no data which indicate that humans are more sensitive than animals
to the effects of dioxin.
However, QRAC has frequently been required to perform quantitative
risk assessments with imperfect and incomplete data. It is necessary to
emphasize that the risk numbers presented in Table 12-2 are upper bound
risks, and that the true carcinogenic risk to humans from dioxins and
furans lies between 0 and the risk numbers presented.
12.6 Non-Cancer Toxlcologlcal Effects of Dioxins
2,3,7,8-TCDD has been shown to produce toxicological effects other
than cancer, primarily in animal models. These include teratogenic,
fetotoxic and reproductive effects in rodents and subhuman primates. An
acceptable daily intake (ADI) of 1-10 pg/kg bw/day can be derived for the
most sensitive non-cancer endpoint of 2,3,7,8-TCDD, namely reproductive
toxicity. In deriving this ADI, the no-observed-effeet-level reported in
a three-generation rat feeding study can be used, along with an appropri-
ate uncertainty factor, either 100 or 1,000 (Cordle 1981). Reproductive
effects seen included significant decreases in fertility and neonatal sur-
vival in the fQ-generation rats receiving 0.1 ug dioxins/kg/day, and, at
0.01 ug dioxins/kg/day, a significant decrease in fertility in the f\
and f£ generations, decreases in litter size at birth, gestational sur-
vival, and neonatal survival and growth. Immunotoxic effects of 2,3,7,8-
tCDD have also been reported at low doses (low ng amounts/kg bw/day) in
animals. Other non-cancer effects that have been noted in animal models
at low dose levels (ng/kg bw/day) include dermal lesions, toxic hepatitis
and amyloidosis of the kidney, liver and spleen. The dose ranges of the
non-cancer effects noted above appear to be higher than those for repro-
ductive effects, and therefore, the ADI of 1-10 pg/kg bw/day developed for
reproductive effects should subsume these other non-cancer endpoints. It
•js also important to note that this ADI range is applicable to durations
Of exposure of months or longer and that for shorter durations of exposure
a higher ADI would be more appropriate.
There is little evidence that 2,3,7,8-TCDD causes reproductive effects
in humans, other than the finding of increases in minor birth defects
among'Vietnam veterans exposed to Agent Orange. The epidemiologic data
-n the association between adverse reproductive outcomes and 2,3,7,8-TCDD
cxp°sure suffer from limitations in assessing and quantifying exposure,
~rid in assessing the effect of multiple chemical exposures. In addition,
reproductive endpoints such as early fetal loss, infertility, and minor
birth defects are extremely difficult to ascertain. Given the high back-
ground rate for these events, most studies are not sensitive enough to
y
12-9
-------
detect small increases. Furthermore, the vast majority of the data on
reproductive events and dioxin exposure is concerned with paternal expo-
sure only. In contrast, toxicological data provide evidence for adverse
reproductive outcomes associated with maternal exposure to 2,3,7,8-TCDD.
Changes in immunological parameters in humans have been shown to occur
after exposure to 2,3,7,8-TCDD, but there has been no indication that
clinical disease has occurred as a result. The observed immune effects
associated with short-term 2,3,7,8-TCDD exposure seem to disappear after
the cessation of exposure. It is unknown how repeated doses or low doses
given chronically may affect the immune system. Also, perinatal exposure
or exposure during childhood may be more pronounced to the immune system,
but there are problems with measuring such an effect. Immunological base-
line data as well as normal response data are generally not available for
children. As a result, currently no unequivocal data exist indicating
that 2,3,7,8-TCDD is a human immunologic hazard.
In summary, based on laboratory studies in animals, an ADI of
1-10 pk/kg bw/day can be developed for the reproductive toxicity induced
by 2,3,7,8-TCDD, which is the most sensitive non-cancer toxicological
endpoint. Epidemiological studies of exposed populations suffer from a
number of limitations, including a lack of sensitivity, and therefore, do
not provide data that are definitive enough to assess the reproductive
and immunological hazard of 2,3,7,8-TCDD.
12.7 Risk of Non-Cancer Toxicoloaical Effects to Subsistence and
Sports Fishers
QRAC has used an ADI of 1-10 pg/kg bw/day dioxin for reproductive
effects of dioxin; this is the most sensitive non-cancer toxicological
endpoint associated with dioxin exposure in animal studies. The dioxin
TEQ exposure levels calculated in Table 12-2 for subsistence fishers,
both at the average and 90th percentile levels, clearly exceed 10 pg/kw
bw/day, except for the dioxin TEQ intake based on an average fish consump-
tion level {not 90th percentile) and weighted mean dioxin fillet level.
These estimates would seem to indicate that subsistence fishers who
either consume fish which are caught in waters in the vicinity of pulp
mills and which are contaminated with 20 ppt dioxin TEQ, or who consume
greater than average amounts of these fish, are at risk for reproductive
effects if humans are more sensitive than rats to reproductive effects.
Various assumptions as described in this report have been made to bridge
information gaps, and these assumptions generally would tend to make the
risk estimates more conservative. There is little evidence, as pointed
out earlier, that dioxins cause reproductive effects in humans.
ni
Risks of reproductive effects to sports fishers are much less sig-
ficant. Fish intake for sports fishers in Tables 12-1 and 12-2 was
12-10
1595q
-------
calculated on the basis of lifetime fishing, which is fairly unlikely;
therefore the risk may be overestimated.
12.8 References
Cordle F. 1981. The use of epidemiology in the regulation of dioxins in
the food supply. Reg. Tox. & Pharm. 1: 379-387.
Puffer HW, Duda MJ, Azen SP. 1982. Consumption rates of potentially
hazardous marine fish caught in polluted coastal waters of Los Angeles
County. N. Am. J. Fish Manag. 2: 74-79.
USDOI. 1988. 1985 National survey of fishing, hunting, and wildlife
associated recreation. Fish and Wildlife Service. Issued November 1988
(p.18).
USEPA. 1988. U.S. Environmental Protection Agency. EPA/Paper industry
cooperative dioxin screening study. Washington, DC: Office of Water
Regulations and Standards. EPA 440/1-88-025.
USEPA. 1989a. U.S. Environmental Protection Agency. National
hioaccumulation study. Data provided to FDA, Center for Food Safety and
Applied Nutrition (March 1989) by EPA Office of Water Regulations and
Standards.
USEPA. 1989b. U.S. Environmental Protection Agency. Interim procedures
for estimating risks associated with exposures to mixtures of chlorinated
Hibenzo-p-dioxins and dibenzofurans (CDDs and CDFs) and 1989 update.
EPA 625/3-89-016.
WHO. 1983. World Health Organization. Guidelines for the study of
dietary intakes of chemical contaminants. Geneva, Switzerland: Global
Environmental Monitoring System, pp. 19-50.
12-11
-------
13. ESTIMATES OF RISKS TO TERRESTRIAL AND AVIAN WILDLIFE FROM LAND
APPLICATION OF PULP AND PAPER MILL SLUDGE AND TO AQUATIC LIFE
FROM DISCHARGE OF EFFLUENTS
13.1 Introduction
This chapter provides estimated exposures and risks to wildlife from
land application of pulp and paper mill sludge and to aquatic life from
discharge of pulp and paper mill effluents containing 2,3,7,8-TCDD and
2,3,7,8-TCDF. The information presented was compiled from:
NYOEC. 1987. New York Department of Environmental Conservation.
Niagara River biota contamination project: fish flesh criteria for
piscivorous wildlife. Technical report 87-3. Division of Fish and
Wildlife, Bureau of Environmental Protection.
Rabert WS. 1990. An update on the environmental effects of TCDD and
TCDF releases from pulp and paper mills on aquatic and terrestrial
animals. U.S. EPA, Office of Toxic Substances, Health and
Environmental Review Division. Memorandum to P. Jennings, EPA,
Exposure Assessment Branch. June 26, 1990.
USEPA. 1990a. U.S. Environmental Protection Agency. Risk
assessment for 2,3,7,8-TCDD and 2,3,7,8-TCDF contaminated receiving
waters from U.S. chlorine-bleaching pulp and paper mills.
Washington, DC: Office of Water Regulation and Standard, U.S.
Environmental Protection Agency. August 1990.
USEPA. 1990b. U.S. Environmental Protection Agency. Assessment of
risks from exposure of humans, terrestrial and avian wildlife, and
aquatic life of dioxins and furans from disposal and use of sludge
from bleached kraft and sulfite pulp and paper mills. Washington,
DC: Office of Toxic Substances and Office of Solid Waste.
EPA 560/5-90-13.
The land application of contaminated sludges to agricultural sites,
mine reclamation sites, and silvicultural sites can lead to wildlife
exposures to 2,3,7,8-TCDD and 2,3,7,8-TCDF. Such exposure may have
adverse effects on individual organisms and may also affect the overall
structure and health of the ecosystem. This analysis examines exposures
and risks resulting from direct ingestion of soils and ingestion of prey
items that have bioconcentrated 2,3,7,8-TCDD or 2,3,7,8-TCDF. Other
routes of exposure exist (e.g., direct contact with soil). However, these
two exposure pathways were chosen for assessment because of the potential
for significant terrestrial wildlife exposure via these pathways, and
because of the availability of data needed to perform the exposure and
risk assessment. Furthermore, this assessment estimates only the effects
Of 2,3,7,8-TCDD and 2,3,7,8-TCDF on individual organisms and their
13-1
-------
ability to produce viable offspring. The analysis does not attempt to
predict the effects of the pollutants on whole populations of wild
species of birds and mammals, or on ecosystems.
Similarly, the discharge of wastewater effluents containing
2,3,7,8-TCDD and 2,3,7,8-TCDF from pulp and paper mills can exert adverse
effects on both the individual organisms downstream from the discharge
point and on the overall structure and health of the ecosystem. The
resident aquatic life (i.e., fish, plants, and benthic organisms) can
also be exposed, in addition to those organisms feeding on aquatic life
(i.e., fish-, or plant-, eating birds and mammals).
13.2 Terrestrial and Avian Wildlife Risk Assessment
13.2.1 Development of Benchmark Doses to Which Terrestrial Wildlife
Exposures (Adjusted for Absorption} Are Compared
Estimated wildlife exposures (adjusted for absorption) are compared
to benchmark doses that have been identified as causing adverse effects
in laboratory species. When these exposure levels approach or exceed the
selected benchmark, the exposed animal is at risk for experiencing
adverse effects. Where possible, doses observed to cause adverse
reproductive effects were selected as benchmarks; the exposure of a
number of individual members of a species to a dose exceeding such a
reproductive effect benchmark may lead to adverse overall population
effects.
For adult birds the estimated daily dose is compared to the
concentration level that had no observable adverse effects in laboratory
experiments (the NOAEL). The concentration in bird eggs is compared to
the lowest concentration level observed in a laboratory that caused
observed adverse effects (the LOAEL). For mammals, the dose is compared
to the lowest dose observed to cause reproductive effects in laboratory
animals. These methods assume that the wild species are as sensitive or
more sensitive to 2,3,7,8-TCDD and 2,3,7,8-TCDF than the laboratory
species.
(1) NOAEL for birds. In order to compare the estimated exposure to
wildlife species from the food chain to the NOAEL benchmark, the exposure
is adjusted by the percent of 2,3,7,8-TCDD or 2,3,7,8-TCDF assumed to be
absorbed from the diet. The NOAEL benchmark was determined from a study
where Schwertz et al. (1973) administered 100 ng/kg body weight/day of
2,3,7,8-TCDD 1n a corn oil/acetone vehicle to 3-day old white leghorn
chickens. This dose was administered for 21 days and produced no adverse
effects. It is assumed that 2,3,7,8-TCDD is 100 percent absorbed from
the corn oil/acetone vehicle (USFDA 1989). However, it is not expected
the absorption of 2,3,7,8-TCDD from food would be the same as the
absorption from a corn oil vehicle. Accordingly, the estimated dose to
wildlife species from the ingestion of prey items 1s adjusted by the
13-2
1594q
-------
percent of 2,3,7,8-TCDD assumed to be absorbed from the diet. Values for
this percentage are found in a recent review of the literature performed
by USFOA (1989). In addition, the laboratory dose must be converted to
an equivalent dose over the length of time that wild species of birds are
exposed to 2,3,7,8-TCDD from the sludge applied to the land-treated
area. All of the migratory birds in this analysis are assumed to reside
In the land-treated area for 6 months (180 days); for these birds, the
NOEL is adjusted by a factor of 180/21, or about 9 (Keenan 1986). In
this case, the adjusted NOAEL is 100/9 ng/kg/day, or 11 ng/kg/day. The
loggerhead shrike is assumed to remain onsite for the entire year. For
this non-migratory species, the NOAEL is adjusted by a factor of 365/21,
or about 17. In this case, the adjusted NOAEL is 100/17 ng/kg/day, or 6
ng/kg/day. The toxicity of 2,3,7,8-TCDF is assumed to be 1/10 that for
2,3,7,8-TCDD as is assumed for toxicity to humans (see Chapter 3).
Therefore, the NOAEL value used for comparison to doses of 2,3,7,8-TCDF
estimated from this analysis are 10 times the NOAEL for 2,3,7,8-TCDD.
(2) LOAEL for bird eoos. Bird eggs can contain 2,3,7,8-TCDD
transferred from the mother's body burden of 2,3,7,8-TCDD. Eggs are an
important endpoint to consider because of their sensitivity to
2,3,7,8-TCDD. The LOAEL used for bird eggs is 65 ppt, based on a study
by Sullivan et al. (1987) that found a 2-fold increase in cardiovascular
malformations in chicken embryos at an estimated egg concentration of
65 P9/9 (65 PPt)- Although effects were found at lower concentrations of
2,3,7,8-TCDD, Sullivan et al. (1987) concluded that the evidence for
effects at these lower levels was inconclusive. The 65 ppt value is used
•jn this analysis for comparison with predicted egg concentrations for
Wj1d species. The toxicity of 2,3,7,8-TCDF is assumed to be 1/10 that
for 2,3,7,8-TCDD as is assumed for toxicity to humans (see Chapter 3).
Therefore, the LOAEL value used for comparison to doses of 2,3,7,8-TCDF
estimated from this analysis are 10 times the LOAEL for 2,3,7,8-TCDD, or
650 ppt.
(3) LOAEL for mammals. This analysis compares exposures (adjusted
for absorption) for small mammals (i.e., mammals less than 1 kg) to the
lowest observed adverse reproductive effect level in laboratory rats.
The LOAEL for small mammals is 10 ng/kg/day, based on a study by Hurray
et al. (1979) (as cited by Kociba and Schwetz (1982)) in which rats were
administered 2,3,7,8-TCDD at 100, 10, or 1 ng/kg/day through the diet to
study the effects on subsequent generations. This analysis compares
exposure estimates to the 10 ng/kg/day level, at which Murray et al.
(1979) found decreased fertility in the fj and fg generations.
For larger mammals, the expected dose for wild species is compared to
ttie lowest dose observed to produce adverse reproductive effects in
^hesus monkeys. The LOAEL for large mammals is 1.7 ng/kg, based on a
study by Schnantz et al. (1982) (as cited by Kociba and Schwetz (1982))
? which rhesus monkeys given 1.7 ng/kg body weight of 2,3,7,8-TCDD in
13-3
-------
the diet showed that four of seven pregnancies had been terminated in
abortion. This value is used in this analysis for comparisons with doses
received in larger wild mammals.
In both the Murray et al . (1979) and the Schnantz et al . (1982)
laboratory studies, doses were administered in the diet. It is assumed
that absorption from a laboratory diet is similar to the absorption from
a wild diet and that these doses are directly comparable to the daily
exposures in wild species from the ingestion of prey items. The toxicity
of 2,3,7,8-TCDF is assumed to be 1/10 that of 2,3,7,8-TCDD as is assumed
for toxicity to humans (see Chapter 3). Therefore, the LOAEL values used
for comparison to doses of 2,3,7,8-TCDF ingested by large and small
mammals are 10 times the LOAEL values for 2,3,7,8-TCDD.
13.2.2 Estimating Exposures to Terrestrial Wildlife
To assess the potential for wildlife exposure to sludge contaminated
with 2,3,7,8-TCDD and 2,3,7,8-TCDF, this analysis adopts elements of
models used by Sullivan et al . (1987) to estimate the potential
2,3,7,8-TCDD and 2,3,7,8-TCDF exposure to wild birds, and methods
employed by Keenan et al. (1989) to estimate the 2,3,7,8-TCDD uptake by
wild turkeys and deer. In addition, this analysis also incorporates work
by Thiel et al. (1988) and the Ontario Ministry of the Environment (OME
1985) concerning the estimation of the steady-state and nonsteady-state
body burden of 2,3,7,8-TCDD and 2,3,7,8-TCDF.
(1) Methodology. The calculations for estimating wildlife
exposures proceed in four steps. First, the dose to an individual animal
is calculated based on the contaminant concentrations in soil, the
2,3,7,8-TCDD and 2,3,7,8-TCDF uptake rates by prey items, and the amount
of each prey item ingested daily. In the second step, the estimated dose
is compared to the LOAEL or NOAEL. Next, the steady state body burden
for the animal is calculated. For migratory species, a body burden is
calculated based on the length of time spent in the treated area.
Finally, egg concentrations are estimated based on the body burden of the
female.
(a) Calculation for estimating dally dose. The dose of
2,3,7,8-TCDD or 2,3,7,8-TCDF to an animal may be calculated as follows:
DOSE - (Pa)(TC) [II to inCjHBCF^tFCtj)] + (Cs)(FSj)] Abg/BWj
where:
Abg - gastrointestinal absorption rate of 2,3,7,8-TCDD or
2,3,7,8-TCDF, percent
- body weight of animal j, kilograms
13-4
1594q
-------
FCi
FS.
Pa*
T?
concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in soil, ng/kg
bioconcentration factor of food source i
fraction of animal j's diet that consists of food source i
fraction of animal j's diet that consists of soil
percent of food originating from the land treated area
total daily quantity of food consumed by the animal, kg
In this calculation, the soil concentration of 2,3,7,8-TCDD or
2,3,7,8-TCDF (CJ is combined with bioconcentration factors (BCFs) of
food items to yield the concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in
these organisms. Birds and small mammals ingest 2,3,7,8-TCDD or
2,3,7,8-TCDF when they prey on these items. Animals may also directly
ingest some sludge if they graze on the application site (e.g., deer), or
if they dig for prey or burrow directly on the application site. To
derive the total amount of 2,3,7,8-TCDD or 2,3,7,8-TCDF ingested daily by
an individual of the species, the percent of the diet represented by each
contaminated food source is multiplied by the concentration of
2,3,7,8-TCDD or 2,3,7,8-TCDF in that food source and by the total daily
quantity of food consumed from the contaminated site. If necessary for
consistency in comparison with LOAEL and NOAEL values derived in the
literature, this estimated daily intake is adjusted by the
bioavailability of 2,3,7,8-TCDD and 2,3,7,8-TCDF from food items.
(b) Comparison of estimated dose to the LOAEL or NOAEL. The
estimated daily dose is then compared to the selected benchmark LOAEL or
NOAEL, using the following equation:
or
DOSE%LOAEL " (DOSE/LOAEL) x 100
DOSE%NOAEL " (DOSE/NOAEL) x 100
where:
D°SE%LOAEL
DOSE%NOAEL
DOSE
NOAEL
LOAEL
dose to wild animal expressed as a fraction of the
LOAEL
dose to wild animal expressed as a fraction of the
NOAEL
dose to the animal, ng/kg/day
dose at which no adverse effects were observed in
laboratory species, ng/kg/day
lowest dose at adverse effects were observed in
laboratory species, ng/kg/day
The daily dose of 2,3,7,8-TCDD or 2,3,7,8-TCDF 1s compared to
literature values for doses that cause reproductive effects. The dose
oredicted for wild species is expressed as a fraction of the lowest dos
observed to cause adverse effects (LOAEL), or as a fraction of the dose
0t?served to cause no adverse effects (NOAEL).
13-5
-------
(c) Calculation for steady state body burden. In order to
determine the concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in eggs laid
by birds exposed to these contaminants, the body burden resulting from
the ingestion of the daily dose must be calculated. If the bird ingests
food contaminated with 2,3,7,8-TCDD and 2,3,7,8-TCDF for a sufficient
length of time to achieve steady-state, then the body burden is estimated
as follows:
Bss - 1.443 (DOSE) (T1/2) / t
where:
B.c = steady- state body burden, ng
DOSE - dose to animal, ng/day
T1/2 - half-life of 2,3,7,8-TCDD or 2,3,7,8-TCDF, days
t » time between doses, days
The steady-state body burden is calculated as the dose multiplied by
the half-life (in days) divided by the length of time between doses (in
days). In this analysis, the dose is ingested daily, so t is equal to
one day.
For migratory birds who arrive at the land application site only a
few weeks before egg-laying, the body burden may not reach steady-state.
For these birds, the body burden is calculated by the equation of Thiel
et al. (1988) as follows:
Bns - [DOSE /(0.693/T1/2)] [l-(l/2)n]
where:
BQS - nonsteady- state body burden, ng
DOSE - dose to animal, ng/day
T1/2 - half-life of 2,3,7,8-TCDD or 2,3,7,8-TCDF
n - (time spent on treated site)/Tj/2
The larger the ratio n, the closer the body burden will be to the
steady-state body burden.
(d) Calculation for estimating egg concentrations. The
concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in bird eggs is derived
using the following equation:
cegg " (Bns) TC / wegg
or, "
cegg " (Bss) Tc / wegg
where:
Bss * steady-state body burden, ng
Bns - nonsteady-state body burden, ng
13-6
1594q
-------
concentration in egg, ng/kg (ppt)
transfer coefficient from mother to egg, expressed as a
fraction of body burden
wegg = wei9nt °f tne e99» kilograms
To predict the concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in bird
eggs, a transfer rate from the female to eggs is used to estimate the
total quantity of 2,3,7,8-TCDD or 2,3,7,8-TCDF in the egg, in ng. The
total quantity in the egg is then divided by the weight of the egg to
obtain an egg concentration of 2,3,7,8-TCDD or 2,3,7,8-TCDF in ng/kg.
The estimated egg concentrations can then be compared to lowest
concentrations observed to cause adverse effects in laboratory studies:
cegg%LOAEL - (Cegg/LOAELegg) x 100
where:
Ceaa%LOAEL * concentration in egg expressed as a fraction of the
33 LOAEL
Cpgy - concentration in egg, ng/kg (ppt)
LOAELggg - lowest concentration at which adverse effects were
observed in eggs in laboratory, ng/kg (ppt)
The 2,3,7,8-TCDD or 2,3,7,8-TCDF concentration in the egg is compared
to a selected literature value for the lowest concentration observed to
produce adverse effects in embryos in laboratory studies. The
concentration predicted for wild bird species is expressed as a percent
Of LOAEL for eggs.
(2) Data Used To Estimate Exposures to Terrestrial Wildlife. The
data used for estimating wildlife exposures are summarized in
fable 13-1. The following subsections describe various factors used in
estimating wildlife exposure.
(a) Selection of species examined. To select the species of
interest, this analysis relied on the expertise of biologists with the
Natural Heritage Programs in the seven states of interest, namely,
Georgia, Maine, Maryland, Mississippi, Ohio, Pennsylvania, and Wisconsin.
jn each state, the Natural Heritage experts provided a list of common
avian and mammalian species as well as endangered species believed to
•inhabit regions of the state where land application of paper and pulp
^-(11 sludge is practiced. Data on the occurrence of mammals in each
state were also obtained from Caras (1967). From these species
identified, nine avian species and seven mammalian species were selected
study. They are the following: loggerhead shrike fLanlus
. American woodcock (Scoolax rusticola). pine warbler
pinusl. eastern meadowlark (Sturnella maana). great crested
ycatcher (Mviarchus crinitus). tree swallow (Iridoprocne bicolorl.
13-7
-------
B899H
Table 13-1. Assumptions and Par-Meter Values - Land Application: Wildlife Exposures
Input
parameter
LOM
Best
estimate
High
Hotes/explanat ion/references
Uptake rates for vegetation eaten by 0.01X
wildlife
2X
15X
BCF for earthworms
0.2
3.5
10
Assumes wildlife eat above-ground plants only.
Low: Uipf et al. (1982)
Best: Sargeant (1989) (See Appendix A.4)
High: Young (1983)
The best estimate Has reported at a site where paper
•ill sludges Mere applied. High and low values are
range reported in the literature. Hart in et al.
(1987)
BCF for insects
0.4
1.5
i
00
BCF for small
BCF for fish fro* sediment (fish is
food source for river otter)
Duration of Food Sources from Sludge
Applied land:
1.3
0.0967
1.4
0.0967
1.4
This is the approximate range of BCFs for insects at
Elgin AFB. Fl. Low is for a composite of soil and
plantborne insects, best estimate is for burrow
spiders, high is for insect grubs. Young and
Cockerham (1985)
These values are for the whole body rather than just
the liver. Martin et al. (1987)
Low and best: Nabholz (1989)
High: USEPA (1988)
Large mammals
River otter
0.5
0.06
1.0
1.0
1.0
1.0
Whole-body elimination rate half life.
TCDO:
Birds
15 days
21 days
31 days
Large Manuals feed from treated area exclusively in
best and high cases. Low is same as for birds. Low
river otter X calculated as size of one side of
50 acre treatment site/size of otter's home range.
Chapman and Feldhamer (1982)
Best estimate: calculated based on information
provided on bluebirds in the literature. Low and
tliglh eattiute* ar« for ••all vertebrates. Lo« *n
OME. ll«a&) B*r»t: Ttttvl et al- 113B81
-------
Table 13-1. f
Input
parameter
Low
Small mammals 15 days
Large manna Is 30 days
Transfer of dioxin from hen to eggs: 3.3X
(X of body burden)
Best
estimate
High
31 days
60 days
4.8X
37 days
365 days
6.2X
Notes/explanat ion/references
Best: data from rat oral; low from hamster oral; high
from mouse intraperitoneal. ONE (1985)
Best = 2 x data for rat oral; low = 2 x hamster oral;
high = monkey. ONE (1985)
Thiel et al. (1988)
OJ
to
Time spent on-site before egg-laying
weeks
Egg mass
Percent of dioxin absorbed from food
6-8
10
0.6
Total food consumption for birds
Data from Nice
(1939) as cited
in Kenaga (1973)
Total food constant ion for manna Is
Nix of food sources
Species-specific
0.7
Data from Nice
(1939) as cited
in Kenaga (1973)
Used data from
Davis a Golly
(1963) and from
Mild Kama Is of
North Alter ira
Species specific
data
data
.95
Data from
Kendeigh (1960)
as cited by
Kenaga (1973)
Best: 6-8 weeks depending on species, except logger-
headshrike. which is not migratory in its range. Low
and high are arbitrary. Bent (1955. 1962. 1963a.
1963b. 1964)
Schonwetter (1960-1983). Thiel et al. (1987). Bent
(1964)
Low: low end of range for food given by workgroup
for non-oily foods. Best: high end of range for non-
oily foods. High: high end of range for absorption
of oily foods. Boyer (1989) (See Appendix A.2)
Kenaga (1973)
Where specific species were not included on Davis and
Golly table, the ratio of consumption to body weight
for a similar animal, and then multiplied by the
weight of the animal of interest. Davis and Golly
(1963)
Toweill and Tabor (1982) as cited in Chapman and
Feldhamer (1982). Hamilton (1979). Bent
(1955. 1962. 1963a. 19B3b. 1964)
-------
8899H
Table 13-1. (continued)
Input
par-Meter
Low
Best
estimate High
Notes/explanat ion/references
Percent of diet that is soil:
Eastern Meadowlark
0.1X
IX
10X
Nine-banded armadillo
IX
10X
2QX
Eastern mole
J^ Virginia opossum
OJ
o
0.1X
IX
Fraction of food sources from sludge-
apolied land;
Birds
Small
0.5
0.5
IX
n
1.0
1.0
10X
10X
1.0
1.0
Southern meadowlarks are described as "fastidious
preeners" that ingest soil during preening. No range
for soil ingest ion given, so range here is
arbitrary. Young and Cockerham (1985)
Described as often ingesting "large amounts of
soil." No range of percentages were given, so range
here is arbitrary. Galbreath (1982) as cited in
Chapman and Feldhamer (1982)
Beachmouse exposed to soil through burrowing,
preening. Hole may have similar exposure. Range
given here is arbitrary. Young and Cocker-ham (1985)
Best estimate: analysis of stomach contents for
opossum in Pennsylvania revealed 7X sand and stones.
Low and high arbitrary. Gardener et al. (1982) as
cited in Chapman and Feldhamer (1982)
Best and high assme birds feed from treated area
only. Low estimate from the literature. Sullivan
et al. (1987)
Small mamals assumed to feed completely from treated
area. Low estimate same as birds. Sullivan et al.
(1987)
-------
American robin (Turdus miqratpriusK wood thrush (Hvlocichla musteline).
and eastern bluebird fSiala sialis sialis). The mammalian species chosen
were: nine-banded armadillo (Dasvous novemcinctusK least shrew
(Cryptotis pan/a), eastern mole (SealOPUS aauaticus). striped skunk
(Mephitis mephitis). Virginia opossum (Didelohis virqiniana). river otter
(Lutra canadensisK and the grey bat (Hvotis arisescens). Only species
judged to be at risk for exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF were
selected. In general, species were selected based on their dietary
habits. Those species that ingest significant quantities of prey items
that bioconcentrate 2,3,7,8-TCDD and 2,3,7,8-TCDF, such as soil
invertebrates, earthworms, insects and small mammals, were thought to be
at most risk.
For avian species, the analysis focused on female birds for two
reasons: female birds have the potential to transfer a portion of their
body burden to eggs, and female birds might receive a higher dose per
kilogram body weight than males during the breeding season. To the
extent that males may be more sensitive than females to the adverse
effects of 2,3,7,8-TCDD or 2,3,7,8-TCDF, this analysis may underestimate
risk to more sensitive individuals (adult males or rapidly-growing
nestlings) of these species.
Soil-dwelling organisms such as earthworms are directly exposed to
contaminated soil. However, Reinecke and Nash (1984) (as cited in Eisler
(1986))reported that two species of earthworms fAllolobophora calioinosa
and Lumbricus rubellusl were held in soils containing 5 parts per million
(ppm) of 2,3,7,8-TCDD and showed no adverse effects. Since a
concentration of 5 ppm is 7300 times higher than the highest sludge
concentration reported for any mill that currently land-applies sludge,
it is assumed that current land application practices would have no
adverse effects on the two species of earthworms tested.
Many predatory/scavenger species could also be exposed to 2,3,7,8-TCDD
or 2,3,7,8-TCDF through the food chain. Because the land application of
sludge is a localized practice, typically covering an areas smaller than
a few hundred acres, species with large hunting territories, such as
osprey, bald eagles, and herons, are unlikely to obtain a large fraction
of their diets from a single sludge land application site. Thus, these
types of organisms are not quantitatively evaluated here. However, it is
possible that these species may accumulate significant levels of
2,3,7,8-TCDD or 2,3,7,8-TCDF because of their position at the top of the
food web; for this reason, these species may warrant further analysis,
especially if future land application of sludge 1s practiced on larger or
on multiple contiguous tracts of land. The analysis does estimate
exposure to one mammal species with a territory relatively large compared
to the land application site: the river otter. This species, which
ingests fish that bioconcentrate 2,3,7,8-TCDD and 2,3,7,8-TCDF from river
sediments, is considered a threatened species in an area of one state
where land application is practiced.
13-11
lS94q
-------
(b) Soil concentrations. This analysis uses the soil
concentrations at the land application sites during a single year as the
basis for the wildlife assessment. The average 2,3,7,8-TCDD and
2,3,7,8-TCDF concentrations over one year for each land application site
are summarized in Table 13-2.
(c) Body weights of animals. In order to compare the amount
ingested by a wild animal to the doses which induce effects in laboratory
animals, dose must be expressed in terms of milligrams per kilogram of
body weight per day. The body weights of all species are summarized in
Table 13-3, along with their daily food consumption levels. The body
weights of female birds were obtained from a monograph produced by the
Western Bird Banding Association, which lists the average body weights of
686 species of North American birds. The body weights of all mammals
except bats and shrews were obtained from Chapman and Feldhamer (1982).
Body weights for bats and shrews were obtained from Hamilton (1979).
(d) Estimating the fraction of food from treated areas. In
this analysis, the "high estimate of risk" and "best estimate of risk"
scenarios assume that the species considered obtain all of their food
from the treated area. This assumption is derived from the fact that
home ranges of most of these species could be encompassed by the sludge
treated area. Even those animals with home ranges larger than the
treated area are likely to be attracted to the treated area for foraging,
since the presence of sludge nutrients may increase the availability of
food in the treated area compared to surrounding areas. For the "low
estimate of risk," it is assumed for all species except otters that 50
percent of an animal's diet originates in the treated area, which is an
assumption consistent with that used by Sullivan et al. (1987). River
otters have a much larger home range than any other of the species
considered in this analysis. The home range of otters varies from 7 to
j5 kilometers (Towel11 and Tabor 1982 as cited in Chapman and Feldhamer
1982). Assuming an average site of 20.2 acres would affect 450 meters of
an adjacent river, the percent of an otter's range that would be affected
tfould vary from 3 to 6.4 percent. Therefore, for the low estimate of
risk, it is assumed that 6 percent of an otter's diet is affected by the
treated area.
(e) Estimating mixes of food sources. Food sources are
assigned to the following categories: earthworms, Insects, plant matter,
soil, small mammals and fish (river otters only). The food mixes for
each species are summarized 1n Table 13-4. Data on the percent of each
type of food consumed by each bird species was obtained from the series
£ books on the life histories of birds by Bent (1955, 1962, 1963a,
l963b, 1964). The one exception was the food consumption data for the
ioodcock, because woodcocks are avid consumers of earthworms. Since the
hi concentration of 2,3,7,8-TCDD and 2,3,7,8-TCDF in earthworms is higher
than other food sources, this analysis estimates woodcock 2,3,7,8-TCDD
13-12
-------
Tattle 13-2.
Characteristics of Land Application Sites and Soil Concent rat: Ions Used In Ul Id life Analysis
State
Georgia
Maine
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
Type of
application
Forest
Forest
Nine
Agriculture
Nine
Agriculture
Forest
Application
rate
(DMT/Ha)
2.065
40
224
58
224
18
40
Years
land
receives
sludge
1
1
1
70
1
20
1
Incorporation
depth (era)
2.5
2.5
0
15
0
15
2.5
Area
receiving
sludge
(Ha/Year)
81
607
27
405
81
30
1.012
Sludge
concent rat ion
2.3.7.8-TCDD
220
13
80
681
145
34
109
Sludge
concentration
2.3.7.8-TCOF
610
55
471
0
795
10
1,300
Soil
2.3.7.8-TCDD
cone.
(ppt)
181
1
80
14
145
0.2
9
Soil
2.3.7.8-TCOF
cone."
(ppt)
501
5
471
0
795
0.07
106
•— aSoi1 concentrations at the land application site are based on a single year.
u>
-------
8BS9H
Table 13-3. Body Weights and Daily Food Conswption of Animals Affected by Land Application of Sludge
Birds
Bluebird
American Robin
6reat Crested Flycatcher
Loggerhead Shrike
Eastern Neadowlark
Tree Swallow
Pine Warbler
UOOQCOCK
Wood Thrush
Wine-Banded Armadillo
6rey Bat
Eastern Mole
River Otter
Virginia Opossui
Least Shrew
Striped Skunk
Range of daily food
Body weight. consiMption grams. Daily food consumption
arans dry Meioht gran, wet weioht Reference
Dice Kendeigh
(1939) (1969)
31.6 4.522 8.915
77.3 8.84 14.73
33.5 5.09 9.21
47.4 6.4 11.19
76.0 8.74 14.59
20.1 3.63 6.92
11.9 2.58 5.18
219.0 --* — *
47.4 6.4 11.19
4.000 120 Davis and Golly (1963)
8 8 Hamilton (1979)
50 25 Chapman and Feldhaaer (1982)
9.000 1.000 Chapun and Feldhaner (1982)
3.500 105 Davis and Golly (1963)
5 S.5 Davis and Golly (1963)
2,300 46 Davis and telly (1963)
* Assumed only constant ion of earthworms and ignored other food sources.
-------
8900H-
Table 13-4. Mix of Food Sources for Birds and Manuals
irda
Bluebird
American Robin
great Crested Flycatcher
Loggerhead Shrike
Eastern Meadow lark
Tree Swallow
pine Warbler
American Woodcock*
^ood Thrush
-jSSUa
nine-Banded Armadillo
firey Bat
gastern Hole
Virginia Opossum
Le»st Shrew
striP«l Skunk
giver Otter
Earthworm
0
2
0
0
0
0
0
100
0
0
0
34
8
50
0
0
Insects
68
98
93.7
72
74
BO
95
0
62
45
100
50
9
50
90
0
Food tvoe.
Plants
32
0
6.3
0
25
20
5
0
38
0
0
15
22
0
10
0
oercent
Small wun
0
0
0
28
0
0
0
0
0
45
0
0
30
0
0
0
nals Soil
0
0
0
0
1
0
0
0
0
10
0
1
7
0
0
0
Fish
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
100
only at consumption of earthworm; Ignored other food sources.
13-15
-------
and 2,3,7,8-TCDF exposure from its consumption of earthworms alone. It
is assumed other food sources contribute relatively little to total
woodcock 2,3,7,8-TCDD and 2,3,7,8-TCDF exposure. Sheldon (1967) reported
that the total daily consumption of earthworms by woodcock is 150 grams.
Data on the mix of food sources for all mammals except shrews were
obtained from Chapman and Feldhamer (1982). The data on mix of food
sources for shrews were obtained from Hamilton (1979). When data on mix
of food sources were reported for more than one area of the country, the
data from a state where land application is practiced, or a nearby state,
were used.
(f) Estimating the fraction of the diet consisting of soil.
Some mammals and birds will ingest soil inadvertently while consuming
ground-dwelling prey or while preening or burrowing. Young and Cockerham
(1985) reported relatively higher liver concentrations of 2,3,7,8-TCDD
for Southern meadowlarks (a subspecies of the eastern meadowlark, one of
the bird species selected for detailed study in this analysis) residing
around a 2,3,7,B-TCDD-contaminated area at Elgin AFB in Florida. Young
and Cockerham (1985) hypothesized that the birds ingest some soil while
preening. Based on the Young and Cockerham report, this analysis assumes
that the eastern meadowlark, a relative of the Southern meadowlark,
ingests a small amount of soil during preening. For this analysis, it is
assumed that between 0.1 and 10 percent of the diet consists of soil,
with a best estimate of 1 percent.
Young and Cockerham (1985) also postulated that beachmice living in
the same area may have elevated liver concentrations of 2,3,7,8-TCDD due
to their burrowing ard preening behavior. Eastern moles, who are also
burrowers, may have similar opportunities for inadvertent soil
ingestion. For this analysis, it is assumed that from 0.1 to 10 percent
of a mole's diet consists of soil ingested while foraging or burrowing,
with a best estimate of 1 percent.
Gal breath (1982) (as cited in Chapman and Feldhamer 1982) reported
that armadillos often ingest large amounts of soil, although the percent
of the diet consisting of soil was not reported. This analysis assumes
that an armadillo's diet could consist of 1 to 20 percent soil, with a
best estimate of 10 percent.
Gardener (1982) (as cited in Chapman and Feldhamer 1982) reported
that analysis of the stomach contents of Virginia opossums found in
Pennsylvania contained approximately 7 percent sand and stones. This
value is used as an estimate of the percent of the opossum's diet
consisting of soil, and 1 and 10 percent are used for the low and high
estimates, respectively.
(g) Biconcentration factors.
13-16
1594q
-------
• Earthworms. The tendency of earthworms to bioconcentrate
2,3,7,8-TCDD has been shown in several studies. Many of the
studies that yielded high bioconcentration factors were conducted
at sites where soil 2,3,7,8-TCDD concentrations were quite high
such as Seveso, Italy; these values were considered inappropriate
for use in this analysis, where much lower soil concentrations of
2,3,7,8-TCDD are expected. Reinecke and Nash (1984) reported
earthworm concentrations of 2,3,7,8-TCDD of 0.2 to 10 times higher
than soil concentrations. The best estimate of the bioconcentration
of 2,3,7,8-TCDD and 2,3,7,8-TCDF in earthworms, reported by Martin
(1987) at an actual site where paper mill sludges containing
2,3,7,8-TCDD and 2,3,7,8-TCDF had been applied, is 3.5 times higher
than the soil concentration. The low and high estimates are 0.2
and 10, respectively.
• Insects. Young and Cockerham (1985) reported average
concentrations of 2,3,7,8-TCDD in a number of species and families
of insects at a 2,3,7,8-TCDD-contaminated site at Elgin Air Force
Base in Florida. Comparing the insect concentrations to the
average soil concentrations reported for the same area, the
bioconcentration factors for insects varies from zero for
grasshoppers, to 0.4 for a composite of soil and plant-borne
insects, to a high of 1.5 for insect grubs. For insects, a value
of 1 is used as the best estimate for the bioconcentration factor,
0.4 and 1.5 are used for the low and high estimates, respectively.
• Small Mammals. Martin et al. (1987) reported that the whole
body bioconcentration factor for deer mice was about 1.4, and
compared this value to the whole body bioconcentration factor of
1.3 reported for field mice taken from the Seveso area. In
contrast, Thalken and Young (1983) reported values for beachmice
liver tissue that ranged from 6.7 for females to 18 for males. For
this analysis, the whole body bioconcentration factors were judged
to be more appropriate than liver-only values. Low, best, and high
bioconcentration factors are 1.3, 1.4, and 1.4, respectively. It
is important to note that these bioconcentration factors were
derived from small mammal species that may get little 2,3,7,8-TCDD
or 2,3,7,8-TCDF exposure from their diet (i.e., mice eat large
quantities of seeds which would not be expected to contain
significant amounts of 2,3,7,8-TCDD or 2,3,7,8-TCDF). Small mammals
that consume prey that bioconcentrate 2,3,7,8-TCDD and 2,3,7,8-TCDF
may have higher levels of 2,3,7,8-TCDD and 2,3,7,8-TCDF 1n their
bodies.
• Fish. The fish to sediment ratio used in both the low estimate
of risk and best estimate of risk scenarios to estimate the
concentration of 2,3,7,8-TCDD in fish consumed by river otters is
assumed to be 0.0967 (whole body, wet weight), as derived from a
13-17
-------
recent EPA literature review (USEPA 1989b). In the high estimate of
risk scenario, the fish to sediment ratio, 5, is obtained from
another EPA review of literature (USEPA 1988). Otters are assumed
to eat the entire body of the fish. For simplicity, it is assumed
that the sediment 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations near
all land application sites are, on average, 1/1000 the 2,3,7,8-TCDD
and 2,3,7,8-TCDF concentrations in the soil at the land application
site. This value is based on the average sediment to soil ratios
calculated in this report for the seven land application sites.
(h) Plant uptake rates. The low, best, and high estimates for
the uptake of 2,3,7,8-TCDD or 2,3,7,8-TCDF from soil into plant tissues
is 0.01, 2, and 15 percent. The low estimate for the uptake of
2,3,7,8-TCDD or 2,3,7,8-TCDF from soil into plant tissues (0.01 percent)
is derived from Wipf (1982). The high estimate of the plant uptake rate
is taken from a study by Young (1983). For the best estimate of plant
uptake, this analysis uses the value suggested by USEPA (1989a) (see
Appendix A) for above-ground plants. Since all of these values were
derived from studies on cultivated plants, the use of these values in
this analysis is based on the assumption that wild plants take up
2,3,7,8-TCDD and 2,3,7,8-TCDF at the same rate as cultivated crops.
Furthermore, the use of this range of estimates assumes that wild animals
consume only above-ground crops.
(1) Total food consumption per day. A summary of the data used
for total daily food consumption estimates is presented in Table 13-3.
The total consumption values for the armadillo, opossum and striped skunk
were estimated by applying the food intake/body weight ratio for raccoons
(Chapman and Feldhamer 1982) to the body weights of these animals. The
total food consumption data for other species were taken directly from
the literature, with otters and moles from Chapman and Feldhamer (1982),
bats from Hamilton (1979), and least shrew from Davis and Golly (1963).
Data from Nice (1939) and Kendeigh (1969) were used by Kenaga (1973)
to predict a regression equation relating the log of body weight to the
log of the ratio of food consumption to body weight. The body weights
for birds from Dunning (1984) were used in these equations to predict the
total daily food consumption, in dry weight. The Nice (1939) data are
used for the best estimate, and the Kendeigh (1969) data are used as the
high estimate. To use the values derived, dry weight values must be
converted to wet weight values. To convert the dry weight values to wet
weight, earthworms are assumed to be 83 percent water by weight (French
et al. 1957 as cited by Kenaga 1973). No data on the wet weight of other
food sources were found; therefore, the percent water of other food
sources is arbitrarily assumed to be 50 percent.
Sheldon (1967) reported that the American woodcock consumes 150 grams
per day of earthworms. Since earthworms bioconcentrate 2,3,7,8-TCDD and
2,3,7,8-TCDF at a higher rate than any other food source, the exposure
13-18
1594q
-------
estimate of woodcocks to 2,3,7,8-TCDD and 2,3,7,8-TCDF is based on its
consumption of earthworms alone; other food sources are assumed to
contribute relatively little to the total dose.
(j) Absorption of 2,3,7,8-TCDD or 2,3,7,8-TCDF from
gastrointestinal tract. The absorption of 2,3,7,8-TCDD and 2,3,7,8-TCDF
from food sources is needed in order to compare the dose ingested by wild
birds to results from laboratory studies where 2,3,7,8-TCDD was delivered
to chickens in a corn oil matrix. USFDA (1989) recently reviewed studies
on the bioavailability of TCDD ingested in a variety of matrices, and
concluded that 60 to 70 percent of 2,3,7,8-TCDD is absorbed by the
gastrointestinal tract from non-oily foods, while 85 to 95 percent is
absorbed by the gastrointestinal tract from oily foods. For the best
estimate, it is assumed that absorption of 2,3,7,8-TCDD and 2,3,7,8-TCDF
from all food sources in a wild animal's diet is 70 percent, while values
of 60 percent and 95 percent are used to represent the low and high
estimates, respectively.
For mammals, no adjustment is necessary, because, in the laboratory
studies to which wildlife exposure estimates are compared, researchers
administered 2,3,7,8-TCDD to the animals through the diet. This analysis
assumes that the 2,3,7,8-TCDD and 2,3,7,8-TCDF absorption rate from the
wild species diets is the same as the 2,3,7,8-TCDD and 2,3,7,8-TCDF
absorption rate from the laboratory diet.
(k) Estimating body burdens of 2,3,7,8-TCDD and 2,3,7,8-TCDF.
Egg concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF are a function of the
body burden of 2,3,7,8-TCDD and 2,3,7,8-TCDF in the female laying the
eggs. In order to calculate the transfer of 2,3,7,8-TCDD and
2,3,7,8-TCDF from the female bird to her offspring, the body burden of
the female bird must be calculated. The equation for calculating steady-
state body burden is derived in the Ontario Ministry of the Environment
Scientific Criteria Document for PCDDs and PCDFs (OME 1985). However,
not all organisms will be exposed to the contaminated area long enough to
reach steady state. In fact, of all bird species analyzed here, only the
loggerhead shrike is considered nonmigratory. For migratory organisms, a
pharmacokinetic model described in Thiel et al. (1988) is used to predict
the body burden at the time of reproduction. One input to this model is
the length of time exposed to 2,3,7,8-TCDD and 2,3,7,8-TCDF before
reproduction. For a migratory bird, it is assumed that the exposure
Begins upon arrival in the northern portion of the range and continues
until the time when the eggs are laid. For the bird species analyzed,
^his is approximately 6-8 weeks, depending on the species. This value was
obtained from calculating the number of weeks between the first spring
sightings of the birds in the states where land application is practiced
flnd reported egg-laying dates in these states (Bent 1955, 1962, 1963a,
*g63b, 1964).
13-19
-------
Another input required for the pharmacokinetic model is the half-life
of 2,3,7,8-TCDD and 2,3,7,8-TCDF in wild birds. OME (1985) reported the
half-life for whole body elimination of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
several species, including rats, mice, hamsters, and monkeys. No data
were found for wild birds. However, from an analysis by Thiel et al.
(1988), an estimate of the half-life of 2,3,7,8-TCDD in birds was
indirectly estimated. Because the 2,3,7,8-TCDD soil concentrations were
not reported by Thiel et al. (1988), the average soil concentrations over
one year were derived using the sludge concentrations of 2,3,7,8-TCDD
applied to the treated area, assuming a 2,3,7,8-TCDD half-life in soil of
10 years, and assuming that the sludge was incorporated with 1 inch of
forest floor litter. Information on the dietary habits of bluebirds from
Bent (1964) and an estimate of the bluebird's total consumption of food
per day from Kenaga (1973) were used to estimate the daily dose of
2,3,7,8-TCDD to the bluebird at the site. In addition, information found
in Bent (1964) on the time of arrival and egg dates in Wisconsin, the
state where the Thiel et al. (1988) study was conducted, was used to
estimate the length of time bluebirds were residing in the treated area
before reproducing. Finally, the average weights of bluebirds eggs and
the percent of 2,3,7,8-TCDD transferred from the bluebird hen to her eggs
was obtained from Thiel et al. (1988). All of the information was
entered into the pharmacokinetic model, and the value for the half-life
of 2,3,7,8-TCDD was adjusted until the model yielded values that
corresponded to the actual 2,3,7,8-TCDD concentration in bluebird eggs
reported in Thiel et al. (1988). The value estimated for half-life is 21
days. This value is in good agreement with values reported in OME (1985)
for other small vertebrates, such as rats and mice, which range from 17
to 31 days, and is used as the best estimate. For a low and high
estimate of the half-life of 2,3,7,8-TCDD in wild birds, data from OME
(1985) are used, which reports a range of 2,3,7,8-TCDD half-lives for
small vertebrates from 17 to 31 days. Martin et al. (1987) stated that
the half-life for the whole body elimination of 2,3,7,8-TCDF is
one-eighth the half-life of 2,3,7,8-TCDD. Under this assumption, the
estimated half-life for 2,3,7,8-TCDF in wild birds is 2.6 days, and
ranges from 2.1 to 3.9 days.
For the estimate of half-life of 2,3,7,8-TCDD in mammals, data
reported in OME (1985) are used. For small mammals, a value of 31 days
represents the best estimate. This value was observed in rats
administered 2,3,7,8-TCDD orally. The low estimate is 15 days, the value
observed for hamsters administered 2,3,7,8-TCDD orally, while the high
estimate 37 days, the value observed in mice administered with
2,3,7,8-TCDD i.p. For larger mammals (i.e., over 1 kg), the best
estimate for half-life of 2,3,7,8-TCDD is 60 days. This value is about
twice the half-life observed in rats given 2,3,7,8-TCDD orally. The low
estimate is 30 days, twice the hamster value, while the high estimate of
365 days is the half-life of 2,3,7,8-TCDD observed in monkeys. Again, it
is assumed that the half-life for the whole body elimination of
13-20
1594q
-------
2,3,7,8-TCDF is one-eighth the half-life of 2,3,7,8-TCDD (Martin et al .
1987). Under this assumption, the half-life of 2,3,7,8-TCDF in small
mammals is 3.9 days (with low and high estimates of 1.9 days and 4.6
days, respectively). The half-life for 2,3,7,8-TCDF in large mammals is
7.5 days (ranging from 3.8 days to 45.6 days).
(1) Transfer coefficient from the hen to eggs. The transfer
coefficient of 2,3,7,8-TCDD from the hen to the egg was estimated by
Thiel et al . (1988) by measuring the 2,3,7,8-TCDD concentration in the
body of the bluebird and the concentration in the eggs. Thiel et al .
(1988) reported that a mean of 4.8 percent of the female's body burden of
2,3,7,8-TCDD was transferred to eggs, with a range from 3.3. to 6.2
percent. A value of 4.8 percent was derived as the best estimate of
transfer rate for 2,3,7,8-TCDD and 2,3,7,8-TCDF, while values of
3.3 percent and 6.2 percent are used for the low and high estimates,
respectively.
(m) Egg weights of birds. The quantity of 2,3,7,8-TCDD and
2,3,7,8-TCDF transferred from the female bird to the eggs must be divided
t>y the weight of the egg to obtain an estimated egg concentration. Egg
weights were reported for bluebirds and tree swallows in Thiel et al,
(1938) and for eastern me ad owl arks in Bent (1964]. All other egg weights
were obtained from Schonwetter (1960-1984).
13.2.3 Summary of Results: Terrestrial Wildlife.
The risks to terrestrial wildlife from land application of pulp and
paper sludge are presented as best estimates of risk, as well as low and
high estimates. For adult birds and bird eggs, these three sets of risk
estimates are presented in Table 13-5 and for mammals the estimates are
presented in Table 13-6. Table 13-7 presents a summary of the results of
the best estimate of risk analysis for birds foraging from land
application sites. This table shows the highest estimates of the daily
dose (expressed as a percent of the NOAEL) among the seven land
application sites assessed in this analysis. The table also indicates
the states where the highest values occur. Similarly, Table 13-8
summarizes the risks to bird eggs, while Table 13-9 presents the risks to
species.
The results confirm that those species whose diets consist largely of
prey species that bioconcentrate 2,3,7,8-TCDD and 2,3,7,6-TCDF are at the
greatest risk from the land application of sludges containing
2,3,7,8-TCDD and 2,3,7,8-TCDF. For example, at all seven sites assessed,
the avian species at greatest risk is the American woodcock, a species
that consumes relatively large quantities of earthworms. At the land
application site with the highest estimated soil concentrations of
2 3,7,8-TCDD, the best estimate of the daily dose of 2,3,7,8-TCDD
^gested by this species is 27.6 times the estimated NOAEL; the daily
of ingested 2,3,7,8-TCDF is 7.6 times the estimated NOAEL. The eggs
13-21
-------
8900H-22
Table 13-5. Estimates of Risks to Adult and Hatching Birds from Exposure to 2.3,7,8-TCDD as a Result of
Land Application of Pulp and Paper Mill Sludge
Species/state
Anerican Robin
Georgia
Maine
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
llnnrinnnlr
VOOOCOCK
Georgia
Maine
Maryland
Wisconsin
Eastern Bluebird
Maine
Maryland
Ohio
Pennsylvania
Wisconsin
Great Crested Flycatcher
Georgia
Maryland
Mississippi
Ohio
Pennsylvania
Loggerhead Shrike6
Maryland
Mississippi
Ohio
Pennsylvania
Eastern Neadowlark
Maryland
Mississippi
Ohio
Pennsylvania
Sludge
cone.
(ppt)
220
13
80
681
145
34
109
220
13
80
109
13
80
145
34
109
220
80
681
145
34
80
681
145
34
80
681
145
34
Exposure
cone.
(ppt)
181
1
80
14
145
0.2
9
181
1
80
9
1
80
145
0.2
9
181
80
14
145
0.2
80
14
145
0.2
80
14
145
0.2
Adult risks3
(tines NOAEL)
Low Best High
0.67
0
0.30
0.05
0.53
0
0.03
0.68
0
0.30
0.03
0
0.29
0.53
0
0.03
0.56
0.25
0.04
0.45
0
0.70
0.13
1.28
0
0.15
0.03
0.2B
0
4.13
0.02
1.83
0.33
3.31
0.01
0.21
27.61
0.15
12.20
1.37
0.02
1.70
3.09
0
0.19
3.28
1.45
0.26
2.63
0
2.87
0.51
5.20
0.01
0.88
0.16
1.60
0
6.57
0.04
2.90
0.52
5.26
0.01
0.33
78.89
0.44
34.87
3.92
0.03
2.61
4.72
0.01
0.29
4.93
2.18
0.39
3.95
0.01
3.78
0.68
6.85
0.01
1.33
0.24
2.42
0
Enbryo risks6
(tines LOAEL)
Low Best High
1.36
0.01
0.60
0.11
1.09
0
0.07
1.36
0.01
0.60
0.07
0
0.27
0.49
0
0.03
1.36
0.38
0.07
0.68
0
0.90
0.16
1.63
0
0.29
0.05
0.53
0
10.59
0.06
4.68
0.84
8.49
0.01
0.53
62.37
0.34
27.57
3.10
0.05
3.72
6.75
0.01
0.42
6.24
2.76
0.49
5.00
0.01
3.97
0.71
7.20
0.01
2.11
0.38
3.82
0.01
30.16
0.17
13.33
2.38
24.16
0.04
1.50
358.32
1.98
158.37
17.82
0.13
10.19
18.47
0.03
1.15
16.80
7.43
1.33
13.46
0.02
7.88
1.41
14.29
0.02
5.71
1.02
10.34
0.02
13-22
-------
ggOOH
-23
Table 13-5. (Continued)
Species/state
-r-oe Swal1ow
fre*5
Georgia
M-ryland
nhio
Wisconsin
P10C ^leP
jne
ryland
nhio
wi9Con»in
^_d Thrush
*eeorgia
^ine
Dryland
Wisc°"sin
Sludge
cone.
(ppt)
220
13
80
IK
109
220
13
80
145
109
220
13
80
145
109
Exposure
cone.
(ppt)
161
1
80
14S
9
181
1
80
14S
9
181
1
80
14S
9
Low
0.57
0
0.25
0.46
0.03
0.82
0
0.36
0.65
0.04
0.33
0
0.15
0.27
0.02
Adult risks"
(times NOAEL)
Best
3.34
0.02
1.47
2.67
0.17
4.76
0.03
2.10
3.81
0.24
1.94
0.01
0.86
1.55
0.10
High
5.05
0.03
2.23
4.05
0.25
7.15
0.04
3.16
5.73
0.36
2.98
0.02
1.32
2.39
0.15
Low
1.12
0.01
0.49
0.89
0.06
0.92
0.01
0.41
0.74
0.05
0.55
0
0.24
0.44
0.03
Embryo risks
(tines LOAEL)
Best
8.18
0.05
3.62
6.55
0.41
6.79
0.04
3.00
5.44
0.34
4.05
0.02
1.79
3.25
0.20
High
22.19
0.12
9. SI
17.77
1.10
18.26
0.10
8.07
14.63
0.91
11.15
0.06
4.93
8.93
0.55
factors represent the comparison factor between the estimated exposure to birds and the NOAEL for
Z(3.7.8-TCDD of 11 ppt for migratory birds and 6 ppt for nonmigratory birds.
These factors represent the comparison facator between the estimated exposure to bird eggs and the LOAEL for bird
for 2-3-7-8- of 65 ppt-
Loggerhead Shrike is considered to be a threatened species in the State of Maryland.
13-23
-------
8900H-20
Table 13-6. Estimates of Risks to Manuals from Exposure to
2,3,7,8-TCOO as a Result of Land Application of Pulp
and Paper Mill Sludge
Species/state
Least Shrew
Georgia
Maine
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
6rey Batb
Georgia
Maine
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
Eastern Mole
Georgia
Maine
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
Virginia ODOSSUM
Georgia
Malm
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
Striped Skunk
Georgia
Maine
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
Sludge
cone.
(ppt)
220
13
80
681
145
34
109
ZZO
13
80
681
145
34
109
220
13
80
681
145
34
109
220
13
80
681
145
34
109
220
13
80
681
145
34
109
Soil
cone.
(PPt)
181
1
80
14
145
0.2
9
181
1
80
14
145
0.2
9
181
1
80
14
145
0.2
9
181
1
80
14
145
0.2
9
181
1
80
14
145
0.2
9
Adult risks"
(tiws LOAEL)
Low Best High
2.99
0.02
1.32
0.24
2.39
0
0.15
3.62
0.02
1.60
0.29
2.90
0
0.18
1.22
0.01
0.54
0.10
0.98
0
0.06
0.72
0
0.32
0.06
0.56
0
0.04
0.38
0
0.17
0.03
0.31
0
0.02
44.80
0.25
19.80
3.54
35.89
0.06
2.23
18.10
0.10
8.00
1.43
14.50
0.02
0.90
15.40
0.09
6.81
1.22
12.34
0.02
0.77
2.88
0.02
1.27
0.23
2.31
0
0.14
1.92
0.01
0.85
0.15
1.54
0
0.10
114.48
0.63
50.60
9.04
91.71
0.14
5.69
27.15
0.15
12.00
2.15
21.75
0.03
1.35
38.50
0.21
17.02
3.04
30.85
0.05
1.91
4.69
0.03
2.07
0.37
3.76
0.01
0.23
2.89
0.02
1.28
0.23
2.32
0
0.14
13-24
-------
8900H-Z1
Table 13-6. (Continued)
Species/state
Sludge
cone.
(ppt)
Soil
cone.
(PPt)
Adult risks9
(tines LOAEL)
Low Best High
Nine-Banded Armadillo
Mississippi
681
14
0.11
0.31
0.37
River Otter
Georgia
Maine
Maryland
Mississippi
Ohio
Pennsylvania
Wisconsin
220
13
80
681
145
34
109
181
1
80
14
145
0.2
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.06
0
0.03
0
0.05
0
0
These factors represent the comparison factor between the estimated
exposure to mamuls and the LOAEL for 2.3,7,8-TCDD of 10 ppt for anil
mammals and the LOAEL of 1.7 ppt for large mamals. For enable, the best
estimate of exposure of the Least Shrew to 2.3.7,8-TCDD fro* sludge
applied to land in Georgia is 44.8 times greater than the LOAEL for small
mannals (i.e.. 10 ppt).
The Grey Bat Is considered to be an endangered species in the State of
Georgia.
13-25
-------
8900H-9
Table 13-7. Sumnary of Risks to Birds ("Best Estimate")
Comparison of Estimated Dose to NOAEL
Species
Eastern Bluebird
American Robin
Great Crested Flycatcher
Loggerhead Shrike
Eastern Meadow lark
Tree Swallow
Pine Warbler
Woodcock
Wood Thrush
2.3.7.8-TCDD
high dose
factor
3.1
4.1
3.3
5.2
1.6
3.3
4.8
27.6
1.9
2,3.7.8-TCDF
high dose
factor
1.7
1.1
0.9
2.9
0.9
0.9
1.3
7.6
0.5
State
where
highest
value
occurs
OH
GA
GA
OH
OH
GA
GA
GA
SA
a These factors represent the comparison factor between the estimated dose to birds and the NOAEL value
for 2.3,7,8-TCDD of 11 ppt for migratory birds and 6 ppt for non-migratory birds and the NOAEL value for
2,3.7,8-TCDF of 110 ppt for migratory birds and 60 ppt for non-migratory birds. For example, the
estimated exposure level of 2.3,7.8-TCOO in the state of Georgia for the woodcock is 27.6 times greater
than the NOAEL for birds.
13-26
-------
8900H-10
Table 13-8. Sunmary of Risks to Bird Eggs ("Best Estimate"
a
Comparison of Estimated Dose to LOAEL
Species
faat.ern Bluebird
African Robin
Great Crested Flycatcher
L oggerhead Shrike
f astern Meadowlark
TrCe Swallow
p1ne Warbler
woodcock
wood Thrush
2,3,7.8-TCDD
high dose
factor
6.8
10.6
6.2
7.2
3.8
8.2
6.8
62.4
4.1
2.3.7,8-TCDF
high dose
factor
0.6
0.4
0.3
0.5
0.3
0.3
0.3
2.9
0.2
State
where
highest
value
occurs
OH
GA
GA
OH
OH
GA
6A
GA
GA
/•these factors represent the comparison factor between the estimated dose to bird eggs and the LOAEL
value for bird eggs of 65 ppt for 2.3,7.8-TCDD and 650 ppt for 2.3.7.B-TCOF.
13-27
-------
8900H-11
Table 13-9. Surma ry of Risks to Manna Is ("Best Estimate")
a
Comparison of Estimated Dose to LOAEL
Spec ies
Nine-Banded Armadillo1*
Grey Gat
Eastern Hole
River Otter
Virginia Opossum
Least Shrew
Striped Skunk
2.3,7,8-TCDD
high dose
factor
0.3
18.1
15.4
c
2.9
44.8
1.9
2,3.7,8-TCDF
high dose
factor
0
5.0
4.3
c
0.8
12.4
0.5
State
where
highest
value
occurs
NS
GA
GA
c
GA
GA
GA
"These factors represent the comparison factor between the estimated dose to mammals and the
LOAEL value for 2,3,7,8-TCDD of 10 ppt for small manuals and 1.7 ppt for large manna Is and the
LOAEL value for 2,3,7,8-TCDF of 100 ppt for small manna Is and 17 ppt for large manuals.
Armadillo only occurs In Mississippi.
GRiver otter dose less than 0.005 in all states.
13-28
-------
of woodcocks residing on this site are estimated to have 2,3,7,8-TCDD
concentrations 62.4 times higher than the 2,3,7,8-TCDD LOAEL of 65 ppt
and a 2,3,7,8-TCDF concentration that is 2.9 times higher than the
2,3,7,8-TCDF LOAEL of 650 ppt.
Similarly, the mammalian species at greatest risk from land
application of 2,3,7,8-TCDD- and 2,3,7,8-TCDF-contamiriated sludge is the
least shrew. Fifty percent of this species' diet consists of earthworms.
This species also consumes large quantities of food relative to its body
weight, leading to a greater dose per body weight than other species. At
the site with the highest 2,3,7,8-TCDD concentrations, the estimated
daily dose of 2,3,7,8-TCDD ingested by the shrew is 44.8 times higher
than the 2,3,7,8-TCDD LOAEL for small mammals.
The wildlife risk assessment results also show that species whose
diets consist of only moderate amounts of prey species that bioconcentrate
2,3,7,8-TCDD and 2,3,7,8-TCDF may exceed toxicity thresholds if the
concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF are sufficiently high.
for example, at the site with the highest 2,3,7,8-TCDD and 2,3,7,8-TCDF
concentrations, all avian species assumed to reside there exceed the
2,3,7,8-TCDD NOAEL for adult birds. The eggs of these species are
estimated to have concentrations of 2,3,7,8-TCDD that exceed the
2,3,7,8-TCDD LOAEL for eggs. Furthermore, all of the mammals at this
site, except the otter and armadillo, also exceed the 2,3,7,8-TCDD LOAEL.
The wildlife risk assessment results imply that individual members of
certain wildlife species are at risk for reproductive and other effects
from the land application of pulp and paper mill sludges containing
2,3,7,8-TCDD and 2,3,7,8-TCDF. This result assumes that wild species are
at least as sensitive to the effects of 2,3,7,8-TCDD and 2,3,7,8-TCDF as
laboratory species. Adverse effects on individuals may be important if
the individuals affected are members of species that are endangered or
threatened. Table 13-10 presents results of the preliminary search for
endangered and threatened species found in seven counties where the eight
pulp and paper mills are located that apply pulp and paper mill sludge to
land.
This assessment does not attempt to quantify the effects of
2,3,7,8-TCDD and 2,3,7,8-TCDF on populations or ecosystems. However, the
results of assessment show that at certain land application sites, the
Reproductive capability of individuals of certain terrestrial species may
be affected, assuming that wild species are at least as sensitive to the
effects of 2,3,7,8-TCDD and 2,3,7,8-TCDF as laboratory species. Effects
ofl the reproductive capability of a sufficient number of individual
^embers of a species may lead to overall population effects for that
5pecies in that area.
j3.3 Aouatic Life Risk Assessment
i3.3.1 Risks to Fish
Currently, sufficient data are not available concerning the chronic
Affect of 2,3,7,8-TCDD and 2,3,7,8-TCDF on aquatic life to derive
e 13-29
-------
8900H-1Z
Table 13-10. Results of Preliminary Search3 for Endangered (E) and
Threatened (T) Species Found in the Counties Where Pulp
and Paper Mills Are Located that Apply Dioxin- and
Furan-Contamlnated Pulp and Paper Mill Sludge to Land
Endangered and
threatened species
States with soil application
GA
ME
MD
MS
OH
PA
WI
Mammals
Indiana bat (E)
West Indian Manatee (E) K
P P P
Birds
Bald Eagle (E)
Piping Plover (ED)
Wood Stork (E)
Red-Cockaded Woodpecker (E)
Reptiles
Eastern Indigo Snake (T)
Gopher Tortoise (T)
Kemp's Ridley Sea Turtle (E)
Leatherback Sea Turtle (E)
Loggerhead Sea Turtle (T)
Fish
Shortnose Sturgeon
Invertebrates
Iowa Pleistocene Snail
(terrestlal) (E)
Plants
Harperalla (E)
Small Whorled Pogonla (E)
3 Based on information dated October 26, 1989.
bPulp mill sites:
Camden County. Georgia
Cumberland County, Maine
Allegheny County, Maryland
Perry County, Mississippi
Ross County. Ohio
Wyoming County, Pennsylvania
Wood County, Wisconsin
P - Possibly present In county
K - Known to be present in county
13-30
-------
national water quality or sediment criteria for these contaminants.
However, several studies have been conducted that provide some
information concerning the long-term effects of 2,3,7,8-TCDD and
2,3,7,8-TCDF on fish, that is, from subacute exposures longer than short-
term acute tests, but not full-life cycle chronic tests. EPA/OTS has
developed estimated chronic toxicity values for 2,3,7,8-TCDD and
2,3,7,8-TCDF based on these existing exposure studies. Potential impacts
on fish were determined by comparing estimated in-stream concentrations
of 2,3,7,8-TCDD and 2,3,7,8-TCDF to these estimated chronic toxicity
values: 0.038 pg/1 for 2,3,7,8-TCDD and 0.41 pg/1 for 2,3,7,8-TCDF.
Site-specific water column contaminant concentrations were calculated by
USEPA (1990a) using the simple dilution exposure assessment approach and
low (7Q10) receiving stream flow conditions.
Water column concentrations of 2,3,7,8-TCDD immediately downstream
from 80 out of 90 mills evaluated (89 percent) were estimated to exceed
the estimated chronic toxicity value of 0.038 pg/1. Seventy-four mills
(82 percent) exceeded the estimated chronic toxicity value of 0.41 pg/1
for 2,3,7,8-TCDF.
The 7Q10 is used as a design flow for stressed aquatic systems;
however, use of 7Q10 receiving water flow rates does not necessarily
result in the extreme worst-case scenario for aquatic life impacts. 7Q10
is defined as the lowest consecutive seven-day average flow over a
jO-year period. Streamflows less than or equal to the 7Q10 flow
(expressed as a daily flow) can occur multiple times within a given year
for periods of 1 to several days. It is possible that even brief
exposures (i.e., less than seven days) to high concentrations of
2,3,7,8-TCDD and 2,3,7,8-TCDF can result in toxic effects to aquatic
organisms such as migrating fish, and such effects may occur after an
appreciable delay following only brief exposures.
Taking into account the above assumptions, simplifications, and
limitations concerning the risks to fish from effluent discharges of
2,3,7,8-TCDD and 2,3,7,8-TCDF, the results of this assessment indicate
that the levels of 2,3,7,8-TCDD and 2,3,7,8-TCDF contamination in the
^ater column resulting from surface water effluent discharges from many
Ch1orine-blcaching pulp and paper mills could be exerting significant
adverse effects on fish.
J3.3.2 Risks to Aquatic Plants and Herbivores
Physical properties of TCDD and TCDF, such as low water solubility
and high log P, are associated with chemicals which are expected to sorb
*o organic material. The sorption of 2,3,7,8-TCDD and 2,3,7,8-TCDF to
suspended matter, plants, and sedments in an aquatic environment may pose
A potential risk to plants, herbivores and benthic organisms. No toxic
effects on plants were identified for 2,3,7,8-TCDD. While active uptake
systemic root absorption 1s not expected to yield much 2,3,7,8-TCDD
13-31
-------
in plants, 2,3,7,8-TCDD is likely to sorb to organic surfaces of plants
immediately downstream of pulp mills. The amounts of 2,3,7,8-TCDD and
2,3,7,8-TCDF that would adsorb to plants has not been estimated, nor have
samples been taken for measurement.
The question of risks to herbivores, however, is not unimportant.
Dermal exposure to dioxins and consumption of 2,3,7,8-TCDD-contaminated
plants both represent likely routes of exposure which might produce
lethal or sublethal effects. If exposure levels in plants and dermal
uptake can be measured or estimated, potential adverse effects on aquatic
herbivores should be considered further.
13.3.3 Risks to Benthic Organisms
Sediments downstream of effluent discharges from pulp mills are
predicted to be contaminated by dioxins and furans. If the half-life in
sediments are similar to the 10 to 15 years in soils for 2,3,7,8-TCDD,
the dioxin- and furan-contaminated sediments may pose a problem for many
years. The extent of the contaminated sediments produced from pulp mill
discharges were not determined by monitoring or modelling. At the
predicted soil degradation rates, in ten years 2,3,7,8-TCDD levels in
sediments would be more than 7 times higher than the annual deposition
levels. Risks to benthic species cannot be determined at this time,
because sediment toxicity data on 2,3,7,8-TCDD are not available for
benthic species. Also, concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF
in sediments are not available for the 104 pulp mills.
However, 2,3,7,8-TCDD concentrations have been measured in the
sediments in the Fox River, Wisconsin; the mean 2,3,7,8-TCDD level for 13
sites was 3.1 pg/g (ppt) and ranged from undetected at 1.4 pg/g to 7.4
P9/9 (Ankley et il., in press). The 2,3,7,8-TCDF levels at the same
sites average 7.6 times higher than 2,3,7,8-TCDD and ranged from 0.1 to
61.1 pg/g. The highest 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations were
found at the site of a pulp mill. The numbers of benthic fauna were
significantly reduced at the pulp mill site compared to the control
site. The average number of chironomids and oligochaetes at the pulp
mill site were significantly lower than the controls. Numbers of
organisms in both taxa were less than 1 percent of the control levels.
The numbers of benthic fauna were also significantly reduced at all 9
sites downstream of the pulp mill. Other pollutants measured at the
sites included penta-, hexa-, hepta-, and octa-chlorinated forms of
dioxin and furans.
The contaminated sediments can also be expected to provide a
reservoir from which dioxins and furans will enter the food web and be
bioaccumulated by benthic organisms and bottom feeders, such as carp,
catfish, and other edible fish species. Dioxins and furans are expected
to be transferred from one trophic level to another. Toxic effects might
be anticipated in those 2,3,7,8-TCDD-sensitive species throughout the
13-32
1594q
-------
food web, especially in the top carnivores such as salmon, bass, pike,
greyling, and fish-eating birds and mammals, including humans.
13.3.4 Risks to Fish-Eating Birds and Mammals
The New York Department of Environmental Conservation (NYDEC 1987)
reviewed non-cancer toxicity data for dioxins for piscivorous mammals and
birds and arrived at a dietary criterion for consumption of fish. They
determined that wildlife feeding primarily on fish with 2,3,7,8-TCDD body
burden concentrations of greater than 3 ppt were at risk. Comparison of
the 3 ppt toxicity value with measured 2,3,7,8-TCDD concentrations in
whole fish (Table 13-11) sampled near pulp mills in the National
Bioaccumulation Study indicates that 66 percent (57/86) of the fish
samples exceeded that threshold toxicity value. Over 38 percent of the
fish sampled (33/86) contained twice the 2,3,7,8-TCDD level compared with
the criterion. Distribution of whole fish contaminated with more than 3
ppt 2,3,7,8-TCDD include 21 states from Florida to Minnesota and from
Maine to California. The fish with low 2,3,7,8-TCDD levels (i.e., less
than 3 ppt) were usually collected from large bodies of water such as
lakes and the Puget Sound area. The mean measured 2,3,7,8-TCDD
concentration in whole fish is about 7 ppt. Since typically only a
single sample of whole fish was analyzed from each site, the range and
distribution of 2,3,7,8-TCDD concentrations in the fish at these sites
cannot be determined.
Measured 2,3,7,8-TCDD levels in whole fish from several areas appear
high enough to pose an unacceptable risk to resident wildlife species
even with large feeding ranges. For example, in the area around the
first 12 pulp mills in Table 13-11, wildlife (e.g., a bald eagle) would
only have to obtain about one-tenth of its dietary food intake from the
receiving waters to be at risk.
j3.4 References
/Uikley GT, Balcer MD, Brooke LT, Call DJ, Carlson AR, Cook PM, Johnson RD,
Kreis Jr. RG, Lodge K, Neimi GJ. In press. Integrated assessment of
contaminated sediments in the lower Fox River and Green Bay, Wisconsin.
Submitted to J. Great Lakes Research.
pent AC. 1955. Life Histories of North American Wagtails, Shrikes,
Vireos, and Their Allies. New York: Dover Publications.
pent AC. 1962. Life Histories of North American Shore Birds, part 1.
York: Dover Publications.
pent AC. 1963a. Life Histories of North American Flycatchers, Larks,
Shallows, and Their Allies. New York: Dover Publications.
13-33
-------
8900H-9
Table 13-11. Distribution of Dioxin Concentrations in Whole Fish
Sampled in the National Bioaccumulation Study
2.3.7,8-TCDO
concentration
(ppt)
117.89
107.02
75.70
67.18
58.21
40.96
34.40
33.86
32.69
30.04
28.66
24.04
24.01
22.07
21.01
16.60
16.08
15.31
14.75
13.69
13.19
9.10
8.58
8.54
7.97
7.87
7.82
6.76
6.40
6.35
6.00
5.79
5.23
5.20
5.12
5.02
.88
.75
.73
.50
.42
Fish species
Carp
Blue catfish
Sucker
Carp
White sucker
Sucker
Catfish
Blue catfish
White sucker
Carp
Carp
Carp
Spot
Carp
Carp
Sm. buffalo
Sucker
Carp
Channel catfish
Channel catfish
Bowfln
Carp
Carp
Carp
White sucker
Sucker
Sucker
Shorthead redhorse
White sucker
Sucker
Carp
White sucker
Sucker
White sucker
Sucker
White carp
Spotted sucker
Carp
Carp
Carp
Carpsucker
State
Louisiana
South Carolina
North Carolina
Wisconsin
Maryland
Maine
Mississippi
Arkansas
Minnesota
Alabama
Alabama
Florida
Georgia
North Carolina
Michigan
Alabama
Maine
South Carolina
Ohio
Louisiana
Florida
South Carolina
Ohio
Wisconsin
Maine
Massachusetts
New Hampshire
Virginia
Maine
California
Mississippi
Pennsylvania
Washington
Maine
Washington
Wisconsin
Georgia
Kentucky
Arkansas
Georgia
Kentucky
Water body
Wham Brake
Samp It River
Pigeon River
Wisconsin River
N. Br. Potomac River
Androscoggin River
Escatawpa River
Arkansas River
Rainy River
Coosa River
Alabama River
Elevenmlle Creek
Turtle River
Pigeon River
Menominee River
Chicasaw River
Androscoggin River
Catawbee River
Scioto River
Bayou Anacoco
Fenholloway River
Wateree River
Scioto River
Peshtigo River
Penobscot River
Millers River
Androscoggin River
James River
Kennebec River
Sacramento River
Mississippi River
Clarion River
Columbia River
Presumoscot River
Columbia River
Wisconsin River
Altamana River
Mississippi River
Mississippi River
Savannah River
Ohio River
13-34
-------
8900H-10
Table 13-11. (continued)
2.3.7.8-TCOD
concentrat ion
(PPt)
4.30
4.17
3.97
3.92
3.85
3.80
3.80
3.62
3.50
3.47
3.46
3.13
3.10
2.78
2.40 nd
2.01
1.79
1.71
1.69
1.58
.57
.51
.40 nd
.20
.20 nd
.20 nd
.10 nd
.11 nd
1.10 nd
1.00 nd
0.90 nd
0.76
0.70
0.67 nd
0.59
0.55
0.50
0.46
0.45
0.41
Fish species
Carp
Sucker
Carp
Carp
Carp
Carp
Carp
Carp
Hardhead catfish
Sucker
Carp
Hardhead catfish
Carp
Sucker
Carp
Redhorse sucker
Flathead catfish
White sucker
Carp
Catfish
St. flounder
St. flounder
Largescale sucker
Goldfish
Carp
Red drui
Catfish
Atlanta saloon
Flathead sole
St. flounder
Flathead sole
Sucker
Carp
White sucker
Redhorse sucker
Quillback carp
White sucker
Sucker
St. flounder
Sm. buffalo
State
Alabama
Arkansas
Tennessee
Wisconsin
Michigan
Alabana
Alabama
Arkansas
Florida
California
Michigan
Florida
Alabama
Oregon
Wisconsin
North Carolina
Louisiana
Pennsylvania
Georgia
Texas
Washington
Washington
Montana
Nev York
Tennessee
Georgia
Virginia
Washington
Alaska
Washington
Alaska
Oregon
Texas
Maine
Pennsylvania
Alabama
Minnesota
Idaho
Washington
Texa*
Water body
Tonblgbee River
Red River
Hivassee River
Lake Superior
Ashland Harbor
Escanaba River
Tombigbee River
Alabama River
Ouachita River
St. Josephs Bay
San Joaquin River
Muskegon River
St. Andrew Bay
Tombigbee River
Columbia River
Wisconsin River
Neuse River
Mississippi River
Hotter Creek
Flint River/
Lake Blackshear
Lake Sam Rayburn
Steamboat Slough
Commencement Bay
Clark Fork River
Hudson River
Holston River
North River
Pamunkey River
Port Angeles Bay
Ward Cove
Grays Harbor
Silver Bay
Willamette River
Sulfur River
St. Crolx River
Susquehamu River
Conecun River
St. Louis River
Snake River
Grays Harbor
Neche* River
nd - Not detected.
13-35
-------
Bent AC. 1963b. Life Histories of North American Wood Warblers, parts 1
and 2. New York: Dover Publications.
Bent AC. 1964. Life Histories of North American Thrushes, Kinglets, and
Their Allies, parts 1 and 2. New York: Dover Publications.
Brewster DW, Matsumura F, Akera T. 1987. Effects of 2,3,7,8-
tetrachloro-dibenzo-p-dioxin on guinea pig heart muscle. Toxicol. Appl.
Pharmacol. 89:408-417.
Boyer (1989) see FDA.
Caras RA. 1967. North American mammals: fur-bearing animals of the
United States and Canada. New York: Galahad Books.
Chapman JA, Feldhamer GA, editors. 1982. Wild mammals of North
America. Baltimore: John Hopkins University Press.
Davis DE, Golly FB. 1963. Principles of mammology. New York: Reinhold
Publishing Company.
Dunning JB, Jr. 1984. Body weights of 686 species of North American
birds. Western Bird Banding Association Monograph Number 1, May.
Eisler R. 1986. Dioxin hazards to fish, wildlife, and invertebrates: a
synoptic review. U.S. Department of the Interior, Fish and Wildlife
Service, Biological Report 85 (1.8), May.
Hamilton WJ. 1979. Mammals of the Eastern United States, Ithaca, NY:
Cornell University Press.
Hebert DC, Birnbaum LS. 1987. The influence of aging on intestinal
absorption of TCDD in rats. Toxicol. Lett. 37:47-55.
Keenan RE. 1986. Testimony before the Maine Bureau of Environmental
Protection, March 16, 1986. In: Paper Industry Information Office.
Potential Impacts on Wildlife from Dioxin Containing Sludges: A
Compilation of Testimony and Exhibits from the BEP Hearing Record.
(3/16/86 to date).
Keenan RE, Sauer M, Lawrence F, Rand E, Crawford D. 1989. Examination
of potential risks from exposure to dioxin in sludge used to reclaim
abandoned strip mines. In: The Risk Assessment of Environmental and
Human Health Hazards: A Testbook of Case Studies. D.J. Paustenback, ed.
New York: J. Wiley and Sons, pp. 935-998.
Kenaga EE. 1973. Factors to be considered in the evaluation of the
toxicity of pesticides to birds in their environment. In: Coulston F,
F. Kprte, eds. Environmental Quality and Safety; Global Aspects of
Chemistry, Toxicology and Technology as Applied to the Environment. New
York: Academic Press.
13-36
1594q
-------
Kenaga EE, Norris LA. 1983. Environmental toxicity of TCDD. In:
Tucker RE, Young AL, AG Gray, eds. Human and Environmental Risks of
Chlorinated Dioxins and Related Compounds. New York: Plenum Press.
Kociba RJ, Schwetz BA. 1982. Toxicity of 2,3,7,8-tetrachlorodibenzo-p-
dioxin (TCDD). Drug Metabolism Rev. 13: 387-406.
Martin SG, Thiel DA, Duncan JW, Lance WR. 1987. Effects of a paper
industry sludge containing dioxin on wildlife in red pine plantations.
In: Proceedings of 1987 TAPPI Environmental Conference.
Miller R, Norria L, Hawkes C. 1973. Toxicity of 2,3,7,8-tetrachloro-
dibenzo-p-dioxin (TCDD) in aquatic organisms. Environ. Health Perspec.
September, pp. 177-186.
Nabholz V. 1989. Bioconcentration factors for 2,3,7,8-chlorinated
dibenzodioxin and 2,3,7,8-chlorinated dibenzofuran. Unpublished.
Washington, DC: U.S. Enviornmental Protection Agency, Office of Toxic
Substances.
Newton M, Snyder SP. 1978. Exposure of forest herbivores to 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in areas sprayed with 2,4,5-T. Bull.
Environ. Contam. Toxicol. 20: 743-750.
Nikolaidis E, Brunstrom B, Dencker L. 1988. Effects of TCDD and its
congeners 3,3,4,4-tetrachlorazoxybenzene and 3,3,4,4-tetrachlorobiphenyl
on lymphoid development in the thymus of avian embryos. Pharmacol.
Toxicol. 63: 333-336.
OME. 1985. Ontario Ministry of the Environment, Intergovermental
Relations and Hazardous Contaminants Coordination Branch. Scientific
Criteria Document for Standard Development No. 4-84; Polychlorinated
Dibenzo-p- dioxins (PCDDs) and Dibenzofurnas (PCDFs).
Schonwetter M. 1960-1983. Handbuck der Oologie. Berlin.
Schwetz BA, Norris JM, Sparschu GL, Rowe VK, Gehring PJ, Emerson JL,
Gerbig CG. 1973. Toxicology of chlorinated dibenzo-p-dioxins. Environ.
Health Prespect. September, pp. 87-99.
Sheldon WG. 1967. The Book of the American Woodcock. Amherst, MA:
University of Massachusetts, University Press.
Sullivan JR, Kubiak TJ, Amundson TE, Martini RE, Olson LJ, Hill GA.
1987. A wildlife exposure assessment for landspread sludges which
contain dioxins and furans. In: Preceedings of the Tenth Annual Madison
International Waste Conference: Municipal and Industrial Waste.
September 29-30, 1987. University of Madison, Madison, Wisconsin.
13-37
I594q
-------
Thalken CE, Young AL. 1983. Long-term field studies of a rodent
population continuously exposed to TCDD. In: Tucker RE, Young AL, Gray
AP, eds. Human and Environmental Risks of Chlorinated Dioxins and
Related Compounds. New York: Plenum Publishing Corp.
Thiel DA, Martin SG, Duncan JW, Lemke MJ, Lance WR, Peterson R. 1988.
Evaluation of the effects of dioxin-contaminated sludges on wild birds.
In: Proceedings of the 1988 TAPPI Environmental Conference.
USEPA. 1988. Estimating exposure to 2,3,7,8-TCDD. Draft report.
Washington, DC: Office of Health and Environmental Assessment, Exposure
Assessment Group.
USEPA. 1989a. Memorandum to Dioxin-in-Paper Workgroup, on the
bioavailability of dioxins in paper products, dated June 23 from C.
Cinalli and C. Flessner.
USEPA. 1989b. Memorandum: OTS/EEB Aquatic Life Hazard Assessment
(Including BCF Values) for "Dioxin in Paper". Office of Pesticides and
Toxic Substances. Washington, D.C.
USEPA. 1989c. 104-Mill Data Base. Office of Water Regulations and
Standards, July 17 version.
USEPA. 1989d. National bloaccumulation study (draft). Washington, DC:
U.S. Environmental Protection Agency, Office of Water Regulations and
Standards.
USFDA. 1989. U.S. Food and Drug Administration. Bioavailability of
ingested 2,3,7,8-TCDD and related substances. Draft memo dated June 22
from Ivan Boyer.
Wipf HK, Homberger E, Neuner N, Ranalder UB, Vetter W, Vuilleumier OP.
1982. TCDD levels in soil and plant samples from the Seveso area. In:
Huntiziger 0, Frei RW, Merian E, Pocchiari F, eds. Chlorinated Dioxins
and Related Compounds: Impact on the Environment. New York: Pergamon
Press.
Young AL. 1983. Long-term studies on the persistence and movements of
TCDO in a natural ecosystem. In: Tucker RE, Young AL, Gray AG, eds.
Human and Environmental Risks of Chlorinated Dioxins and Related
Compounds. New York: Plenum Press.
Young AL. 1984. A case study 1n ecotoxicology: long-term field exposure
of Peromvsus polionotus in dioxin. In: Hommage au Professor Rene
Truhaut. Paris, France: Academie des Sciences, pp. 1229-1233.
13-38
1594q
-------
Young AL, Cockerham LG. 1985. Fate of TCDD in field ecosystems -
assessment and significance for human exposures. In: Kamrin MA ad
Rodgers PW, eds. Dioxins in the Environment. New York: Hemosphere
Publishing Corp., pp. 153-171.
13-39
15941
-------
APPENDIX A
Unavailability
-------
TABLE OF CONTENTS
Page No.
A.I ASSESSMENT OF BIOAVAILABILITY VIA DERMAL EXPOSURES A-l
A. 1.1 Introduction A-l
A.1.2 Migration from Sol id Matrices A-2
A.1.2.1 Liquid Mediated Extraction A-2
A.1.2.2 Dry Transfer from Paper A-4
A.1.2.3 Exposure to PCDD/F Contaminated Soil and
SI udge A-6
A.1.3 Percutaneous Absorption A-8
A.1.3.1 Dermal Absorption Studies A-8
A.1.3.2 Parameters for Exposure Assessment A-9
A.1.3.3 Factors Affecting Percutaneous Absorption .. A-11
A.1.3.4 Future Experiments A-ll
A.1.4 Discussion A-ll
A. 1.5 References A-12
A.2 BIOAVAILABILITY OF INGESTED 2,3,7,8-TCDD A-14
A.2.1 Gastrointestinal Absorption of 2,3,7,8-TCDD in a
Vol unteer A-14
A.2.2 Gastrointestinal Absorption of 2,3,7,8-TCDD in
Experimental Animals A-15
A.2.2.1 2,3,7,8-TCDD in Vegetable Oil, in Other
Solvents, or in the Diet A-15
A.2.2.2 2,3,7,8-TCDD in Soil A-16
A.2.2.3 2,3,7,8-TCDD and Related Substances in
Soot A-20
A.2.2.4 2,3,7,8-TCDD and Related Substances in Fly-
Ash A-21
A.2.3 Bioavailability of Ingested 2,3,4,7,8-PeCDF
(pentachlorodibenzofuran) A-22
A.2.4 Bioavailability of Ingested OCDD (octachlorodibenzo-
p-dioxin) A-22
A.2.5 Evaluation of Bioavailability in Humans Based on
Animal Data A-24
A.2.5.1 Estimated Bioavailability of 2,3,7,8-TCDD
from Foods A-24
A.2.5.2 Estimated Bioavailability of 2,3,7,8-TCDD
from Soil A-24
-------
TABLE OF CONTENTS (continued)
Pace No.
A.2.5.3 Estimated Bioavailability of 2,3,7,8-TCDD
from Paper A-27
A.2.5.4 Estimated Bioavailability of 2,3,7,8-TCDD
from SIudge A-28
A.2.5.5 Estimated Bioavailability of 2,3,7,8-TCDD
from Water A-28
A.2.6 Conclusions A-28
A.2.7 References A-28
A.3 BIOAVAILABILITY OF INHALED VAPORS AND PARTICLES CONTAINING
2,3,7,8-TCDD AND 2,3,7,8-TCDF A-33
A.3.1 Introduction A-33
A.3.2 Inhalation of Vapors and Bioavailability A-34
A.3.3 Particulate Inhalation and Deposition A-34
A.3.4 Bioavailability of Particulate-Bound 2,3,7,8-TCDD and
2,3,7,8-TCDF A-37
A.4 UPTAKE IN TERRESTRIAL PLANTS A-37
A.4.1 Root Crops A-38
A.4.2 Aboveground Crops A-42
A.4.3 References A-45
-------
LIST OF TABLES
Page No.
Table A-l. Paper:Solvent Equilibrium Partition Coefficients (Kp) A-3
Table A-2. Some Properties of TCDD and TRIS A-5
Table A-3. In Vivo Dermal Absorption of PCDD and PCDF in Rats .. A-7
Table A-4. Bioavailability of Ingested Dioxins and Related
Compounds in Vegetable Oils or in the Diet A-25
Table A-5. Bioavailability of Ingested Dioxins and Related
Compounds in Soils, Soot and Fly-Ash, and Other
Substances A-26
Table A-6. Bioavailability of Ingested 2,3,7,8-TCDD from Various
Media A-30
Table A-7. Percent of Total Dust in Different Aerodynamic
Diameter (da) Size Fractions A-36
Table A-8. Root-to-Soil Ratios for 2,3,7,8-TCDD and PCBs A-39
-------
Appendix A
Bioavailability
A.I ASSESSMENT OF BIOAVAILABILITY VIA DERMAL EXPOSURES
A.1.1 Introduction
The purpose of this section is to provide a set of common assumptions
for use in assessing the human dermal exposure to polychlorinated dibenzo-
p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF). This section
was compiled from:
Babich M, Adams M, Cinalli C, Galloway 0, Hoang K, Huang S, Rogers P.
1989. Common Assumptions for the Assessment of Human Dermal Exposure
to Polychlorinated Dibenzo-p-Dioxins and Dibenzofurans. Interagency
Dioxin-in-Paper Workgroup, Dermal Unavailability Workgroup.
December 12, 1989.
This integrated risk assessment assesses exposure from a variety of
sources such as communications paper, personal care products, contaminated
soil, and paper mill wastes. Exposure may occur by a diversity of scenar-
ios, ranging from occupational exposure at paper mills to the use of dis-
posable diapers made of bleached fluff pulp. However, all of these sce-
narios may be divided into three general categories: (1) liquid mediated
extraction from pulp or paper products; (2) dry transfer from paper
products; and (3) exposure to contaminated soil or paper mill sludge.
Scenario 3 could have been treated as a special case of scenario 1,
because exposure is frequently to wet material; however, soil and sludge
are treated separately from paper products, in part because of the types
and format of the data available.
For the purpose of this multiple source (i.e., multimedia) risk
assessment, dermal exposure will be treated as a two-step process:
(1) migration or extraction of PCDD/F from the appropriate matrix (e.g.,
soil, sludge, or paper) onto the skin, followed by (2) percutaneous
absorption. Moreover, dermal absorption studies have been done with only
a few vehicles (mainly volatile solvents) and only one solid matrix
(soil). It would be unrealistic either to conduct dermal absorption
studies or to expect to find published studies using matrices and/or
vehicles specific for every exposure scenario. On the other hand, there
have been studies of the extraction of dioxin from pulp and paper using
several different solvents. Therefore, Step 1, migration/extraction, will
be treated as scenario specific, whereas Step 2, percutaneous absorption,
will be treated as common to all exposure scenarios.
A-l
159Sq
-------
A.1.2 Migration from Solid Matrices
A.1.2.1 Liquid Mediated Extraction
Liquid mediated extraction of 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD) and 2,3,7,8-tetrachlorodibenzofuran (TCDF) has been characterized
by paperrsolvent equilibrium partition coefficients (Kp) (NCASI 1987a),
defined by:
Kp - [TCDD]paper/[TCDD]solvent (A-l)
where [TCDD]QaDer and [TCDD]soiyent represent the equilibrium concen-
trations of TCDD (or TCDF) in ihe paper and aqueous phase, respectively.
Some of the Kp values in Table A-l were determined with ground bond paper,
at room temperature, and with 24 h for equilibration (NCASI 1987a). Kp
values for synthetic urine and isotonic saline at 32" were determined
with soft-wood fluff pulp (NCASI 1989). The ethanol Kp was determined
with bleached paperboard strips, after 96 h at 49CC (NCASI 1987a).
In some cases, 2,3,7,8-TCDD was not detected in the solvent phase. Thus,
Kp values for 2,3,7,8-TCDD are lower limits, calculated from the limits
of detection in the solvent.
The time course of extraction was reported for synthetic urine
(NCASI 1989) and 8 percent ethanol (NCASI 1987b). Equilibrium apparently
was reached within 2 h with urine and at 4-to-12 h with 8 percent ethanol.
That ethanol required more time to reach equilibrium may be due to the
fact that paperboard strips were used as the solid phase, whereas fluff
pulp was used with urine. It appears reasonable to assume that equilib-
rium is reached within 2 h with aqueous solvents under most conditions.
Kp values may be used to estimate the concentration of TCDD/F in a
liquid medium in contact with the skin. This concentration may be
regarded as an upper limit, since equilibrium conditions are not necessar-
ily expected to occur during the course of exposure. This concentration
may be combined with a measure of percutaneous dermal absorption to
estimate exposure.
(1) Factors affecting liquid mediated extraction. The factors which
are expected to affect the rate or extent of extraction include tempera-
ture, solvent, and time. Kp is expected to decrease at higher tempera-
tures, at least with aqueous solvents. (A smaller Kp means more PCDD/F is
extracted into the liquid phase). The rate of extraction is expected to
increase at higher temperatures. For example, at room temperature and
allowing 24 h for equilibration, the paperrwater Kp for TCDF was estimated
to be 29,000 (NCASI 1987a). Thus, -1/29,000 or only 0.003 percent of TCDF
was extracted from the paper into the liquid phase. In contrast, up to
100 percent of TCDF was extracted from coffee filters in a few minutes,
with water at 90°C (NCASI 1988).
A-2
1598q
-------
890ZH
Table A-l. Paper: Solvent Equilibrium Partition Coefficients (Kp)
Solvent
Solid
phase
Temp. Time
CC) (h)
TCDO
TCOF
Water"
I sot on ic
saline8
Isotonic
sa1ineb
Synthetic
urine"
Synthetic
urineb
8X Ethanol8
Ground bond
paper
Ground bond
paper
Softwood
fluff pulp
Ground bond
paper
Softwood
fluff pulp
Paperboard
strips
21
21
24 >-13.000 29,000
21 24 x*19.000 63,000
32 24 14.300 5,300
24 >-14,000 48,000
32 2-to-24 14.300 6.300
49 96 >- 2.000 2.000
1987a
bMCASI 1989
A-3
-------
Extraction is also dependent on the solvent. Extraction from pulp or
paper is favored with solvents which can swell the paper fibers, such as
alcohols (NCASI 1987b). Thus, 95 percent alcohol was a better extractant
than benzene for quantitative 2,3,7,8-TCDD extraction (NCASI 1987b).
Extraction by aqueous media is decreased at higher ionic strength. For
example, the Kp for TCDF at 21°C is 29,000 with water and 63,000 with
isotonic saline (Table A-l).
A.1.2.2 Dry Transfer from Paper
There are no empirical data relating to the "dry transfer" of PCDD/F
from paper to skin, such as that which might occur during contact with
communications paper. In the absence of appropriate data, previous risk
assessments have relied upon various assumptions regarding the possible
mechanisms and rates of dry transfer.
(1) Previous risk assessments. In a risk assessment prepared for
the U.S. Environmental Protection Agency (EPA), Arthur D. Little, Inc.
assumed that dry transfer is actually mediated by skin oil (ADL 1987). In
other words, "dry transfer" is fundamentally similar to liquid mediated
extraction, but with skin oil as the liquid phase. Because exposure to
communications paper is brief (15 sec), the liquid phase penetrates only
1 percent of the depth of the product. It was further assumed that ex-
traction of TCDD by skin oil is 100 percent efficient and that 0.1 percent
of the liquid phase remains on the skin. Thus, the fraction of TCOO
transferred to the skin in a 15 sec exposure would be:
0.01 x 1 x 0.001 - 1 x 10'5 (A-2)
In the U.S. Paper Industry risk assessment, it was assumed that paper
is roughly equivalent to soil, except that sheets of paper allow less
opportunity for skin contact than granular soil (NCASI 1987a). Thus, the
bioavailability of TCDD from sheet paper was assumed to be 50 percent of
the bioavailability from soil.
(2) TRIS flame retardant as a model for TCDD. In the absence of
data on the unmediated migration of TCDD from paper, the migration of the
flame retardant tr1s-(2,3,dibromopropyl)phosphate (TRIS) from dry cloth
will be used as a surrogate (Ulsamer et al. 1978). Like 2,3,7,8-TCDD,
TRIS is a nonvolatile, hydrophobic compound (Table A-2). Thus, it is
reasonable to employ TRIS as a surrogate for TCDD, but the appropriate-
ness of polyester cloth as a surrogate for paper is of concern. The
basis weight of the cloth (-0.01 g cm"2) is roughly equivalent to that
of bond paper. However, the abilities of these two matrices as barriers
to diffusion or to adsorb TCDD or TRIS are likely to differ. Thus, it is
not possible to be certain whether the TRIS cloth model would adequately
estimate exposure to TCDD from paper.
A-4
1598q
-------
890ZH
Table A-2. Seme Properties of TCOD and TRIS
TCDDa
THIS"
Molecular Height
Vapor pressure.
MI Hg 0 25'C
Physical state at
room temperature
Melting/free/ing point
Soluble/wiscible in:
321.97
,-9
Water solubility
1.4x10
solid
305'C
o-dichlorobenzene
chlorobenzene
benzene
CHC13
acetone
19.3 ng/1
697.7
,-4
1.9x10
viscous liquid
5.5'C
CC14
CHC13
CH2C12
insoluble
"Deal and Basu 1989
bIARC 1979
A-5
-------
From the data of Ulsamer et al. (1978), it may be calculated that
4.63 percent of the total amount of radiolabeled TRIS in a dry cloth
(applied at a concentration of 160 ug/cmz) was transferred to rabbit
skin after 96 h (Ulsamer et al. 1978). Thus, the rate of transfer was
-0.0005 h"1, assuming that the amount transferred is approximately
linear with time during this time period, which is a reasonable
assumption, because <10 percent of the initial TRIS dose was transferred
after 96 h.
For purposes of this exposure assessment, it will be assumed that
TCDD (or TCDF) is transferred from dry paper to skin at the same rate of
0.0005 h"1. Based on this methodology, the fraction of TCDD transferred
to the skin in 15 sec would be:
0.0005 h'1 x 15/3600 h = 2 x 10'6 (A-3)
which is one-fifth the amount estimated in the ADL (1987) risk
assessment. The amount of TCDD transferred to the skin may be multiplied
by the fraction of TCDD absorbed to estimate dormal exposure.
A.1.2.3 Exposure to PCDD/F Contaminated Soil and Sludge
Dermal absorption of 2,3,7,8-TCDD in soil has been measured iji vivo
by direct exposure of rats to soil-bound TCDD, using either dry soil (Shu
et al. 1988) or soil-water paste (Poiger and Schlatter 1980). The
percentage of soil bound TCDD found in the liver ranged from 0.05 percent
to 2.2 percent with wet soil and from 0.65 percent to 1 percent with dry
soil (Table A-3). The extent of release of 2,3,7,8-TCDD from soil was
estimated by comparing the dermal absorption of soil-bound 2,3,7,8-TCDD
with 2,3,7,8-TCDD dissolved in methanol (Poiger and Schlatter 1980). By
this methodology, the release of soil-bound TCDD can be estimated to range
from 0.3 percent to 15 percent, based on a 24 h exposure.
For the purpose of this exposure assessment, it will be assumed that
1 percent of soil bound TCDD (or PCDD/F) is available for dermal absorp-
tion. For sludge, since the physical structure of sludge presumably would
allow easier access to TCDD, the high end value of 15 percent release of
TCDD is recommended. These factors can be combined with either the rate
or fraction of percutaneous absorption to estimate dermal exposure. For
comparison, the U.S. Environmental Protection Agency (EPA), Office of
Research and Development (ORD) has proposed using 0.5 percent as the over-
all dermal exposure (i.e., including migration and dermal absorption) of
TCDD from contaminated soil (USEPA 1988).
Since oral bioavailability of soil-bound 2,3,7,8-TCDD varies greatly
among the different types of soil, this phenomenon may apply to dermal
bioavailiability as well. Oral bioavailability of TCDD is reduced in
soil with high organic content and by increasing the soil residence time,
A-6
1598q
-------
8902H
Table A-3. In Vivo Dermal Absorption of PCDD and PCDF in Rats
Chemical
TCOO
TCOO
TCDD
TCOF
Applied
dose
ug
0.026
0.026
0.35
0.35
0.026
0.35
1.3
0.026
1.3
0.028
0.011
0.072
0.72
7.2
36
72
7
34
69
%
Applied
dose
absorbed
14. 8a
1.4
9.3
14.1
-0.05
1.7
2.2
<0.05
<0.05
0.62*
1.00
38.3b
40.3
27.4
17.8
19.1
17.3
48. 8b
17.9
11.3
Exposure
duration
h Vehicle Reference
24 nethanol Poiger & Schlatter 1980
petroleum jelly
PEG 1500
PEG-water
so 11 -water
soil -Mater
soil -water
activated carbon
activated carbon
4 soil Shu et al. 1988
24
72 acetone Brewster et al. 1989
72 acetone Brewster et al. 1989
Based on liver level.
Based on body burden.
A-7
-------
whereas bioavailibility is enhanced by the presence of oil (Poiger and
Schlatter 1980; Umbreit et al. 1986, 1988). Furthermore, dermal absorp-
tion was undetected when TCDD vas applied as an activated carbon-water
paste (Poiger and Schlatter 1980) (see Table A-3). However, the presence
of used motor oil (up to 2 percent) did not significantly affect dermal
absorption of soil-bound TCDD (Shu et al. 1988). While the presence of
oil appears to enhance gastrointestinal absorption, use of mineral oil or
petroleum jelly as vehicles reduced the dermal absorption of TCDD relative
to methanol or acetone, respectively (Poiger and Schlatter 1980; Weber et
al, 1989).
The Exposure Assessment Group of EPA/ORD through a contractor is con-
ducting dermal absorption studies with 2,3,7,8-TCDD-contaminated soil.
In addition to studying the effects of different soil types, different
concentration levels, and the duration of contact, the study also involves
in vivo (rats) and in vitro (human cadaver skin) studies of interspecies
variation. In this study, total tissue distribution, not just liver
levels, will be analyzed. The laboratory experiments should have been
completed during the summer of 1989.
A.1.3 Percutaneous Absorption
A.1.3.1 Dermal Absorption Studies
(1) Animal studies. Several laboratories have studied the in vivo
percutaneous absorption of 2,3,7,8-TCDD in rats (Banks et al. 1989;
Brewster et al. 1989; Poiger and Schlatter 1980; Shu et al. 1988), while
one laboratory has also studied 2,3,7,8-TCDF, 1,2,3,7,8-penta-chloro-
dibenzofuran (1-PeCOF), and 2,3,4,7,8-pentachlorodibenzofuran (4-PeCDF) as
well (Banks et al. 1989; Brewster et al. 1989). Host studies have been
done by the "finite dose" technique, in which the penetrant (i.e.,
PCDD/F) was applied to the skin in a volatile solvent, either acetone or
methanol, and the solvent was allowed to evaporate. Nonvolatile vehicles
(polyethylene glycol, polyethylene glycol-water, and petroleum jelly) have
also been studied (Poiger and Schlatter 1980), while two laboratories
studied TCDO-contaminated soil (Poiger and Schlatter 1980; Shu et al.
1988). In these studies, absorption was reported as either the fraction
of the applied dose found in the liver (Poiger and Schlatter 1980) or as
a more complete tissue distribution (Banks et al. 1989; Brewster et al.
1989) at 24 to 72 h post exposure. For the most part, the time course of
absorption was not studied, although Shu et al. (1988) reported liver
levels were reported at both 4 and 24 h following exposure to TCDO
contaminated soil. The effect of aging of organisms on degree of
absorption was also studied by Banks et al. (1989).
The results from these studies (Table A-3) show that up to 40 percent
of TCDD and 48 percent of TCDF were absorbed by 72 h after initiation of
A-8
159Bq
-------
exposure (Brewster et al. 1989). In addition, the percent dermal absorp-
tion (i.e., the percentage of the applied dose which is absorbed)
increased as the applied dose decreased, as is generally the case
(Scheuplein and Ross 1974). This is important, because the TCDD levels
to which humans are expected to be exposed are much lower than the levels
used in these laboratory studies.
The only other species studied in vivo is the rhesus monkey. Dermal
absorption of 1-PeCDF was studied concomitantly with oral absorption of
4-PeCDF in a rhesus monkey which had survived a previous intravenous
4-PeCDF exposure at a dose which killed two of three animals (Brewster et
al. 1988). This study is not appropriate for purposes of a risk
assessment, because it involved only one animal and the animal had been
previously exposed.
(2) In vitro studies with human skin. One laboratory has reported
preliminary data on the penetration of human skin by TCDD in vitro (Weber
et al. 1989). Experiments were done using both acetone (i.e., finite
dose technique) and mineral oil as vehicles and using both intact and
"damaged" skin. The distribution of TCDD in stratum corneum, epidermis,
and dermis were reported at times from 30 min to 17 h post exposure.
Interpretation of these in vitro studies is complicated by the
distribution of 2,3,7,8-TCDD among the stratum corneum (SC), epidermis,
and dermis. Only the portion of TCDD which migrates to the dermis is
generally expected to be available for systemic absorption (Weber et al.
1989). However, one might also expect TCDD in the epidermis to be
absorbed, because the SC is the primary barrier to percutaneous
absorption. Some of the TCDD in the SC may also be absorbed since the SC
has been reported to act as a reservoir for certain penetrants (Stoughton
1989). Therefore, in interpreting these in vitro data, we will assume
that TCDD in both the dermis and epidermis will be absorbed systemically,
but that TCDD in the SC will not be absorbed.
A.1.3.2 Parameters for Exposure Assessment
Measures of dermal exposure include the rate of absorption and the
extent of absorption at a given time. Either measure may be required,
depending on the exposure model used and the exposure data which are
available. Thus, estimates of both will be presented. The rate of
absorption will be given as the transfer coefficient, defined by:
J - kA (A-4)
where,
j is the flux (ug cm'2 h"1), k is the transfer coefficient (h'1),
and A is the specific dose (ug cm"2) (Scheuplein and Ross 1974). The
A-9
159*1
-------
extent of absorption will be expressed as the percent of the applied dose
which is absorbed by 24 h post exposure. Parameters for dermal absorption
will be based on "finite dose" experiments (i.e., experiments where
volatile solvents such as methanol or acetone were used as vehicles).
Data obtained from human studies are preferred for use in assessing
human exposure and risk. In addition, the dermal penetration data with
human skin have the advantage that the time course of 2,3,7,8-TCDD
absorption was determined over a range of times (0.5 to 17 h) which is
particularly relevant to human exposure scenarios. The primary disadvan-
tage of the human data is that they were obtained in vitro. However, IQ.
vitro percutaneous absorption studies have been shown to correlate well
with in vivo studies (Feldman and Maibach 1970; Franz 1978), including
studies with hydrophobic compounds (Bronaugh and Franz 1986; Bronaugh and
Stewart 1984, 1986; Cruzan et ,al. 1986). The primary disadvantage of the
in vivo studies with rats is that dermal absorption was generally deter-
mined at one time point only. In addition, the applicability of rat data
to human exposure assessment is uncertain, because human skin is generally
considered to be less permeable than rat skin, based on studies with
various penetrants, but not TCDD (Bartek et al. 1972; Wester and Maibach
1983). -In consideration of the advantages and disadvantages of both kinds
of studies, the human data will be used in assessing human exposure.
Based on the in vitro data with human skin (Weber et al. 1989), it is
estimated that 2,3,7,8-TCDD was absorbed at an average rate of approxi-
mately 0.012 h"1 over the time period from 0.5 h to 17 h. This estimate
is based on the assumption that TCDD in both the dermis and epidermis is
available for systemic absorption. However, it has been assumed by
others that only TCDD in the dermis is available for absorption, and that
TCDD was absorbed at a 10-fold lower rate (Weber et al. 1989).
On average, 18.5 percent of TCDD was absorbed by 17 h post exposure
(Weber et al. 1989). Assuming first order kinetics, the fraction (F) of
the applied dose which would be absorbed at a given time is:
F - 1 - e-kt. (A-5)
Thus, with a transfer coefficient k of 0.012 h"1, the fraction of
2,3,7,8-TCDD absorbed at 24 h post exposure is -0.25. Thus, it may be
assumed that TCDD is absorbed at a rate of 0.012 h'1 and that 25 percent
of the applied dose is absorbed by 24 h post exposure.
For comparison, Table A-3 shows that 14.8 percent of TCDD was found
in the livers of rats at 24 h post exposure (Poiger and Schlatter 1980).
Assuming that liver accounts for roughly half of the body burden (Brewster
et al. 1989), -29 percent of TCDD was absorbed at 24 h, in good agreement
with the value of 25 percent estimated from the human in vitro studies.
A-10
1598q
-------
Dermal absorptions of several other PCDD and PCDF isomers in rats have
also been studied by Brewster et al. (1989). At a dose of 5 ug cm"2,
the extent of absorption of 72 h post exposure decreased in the order:
TCDF (49 percent) > 4-PeCDF (34 percent) > 1-PeCDF (25 percent) > TCDD
(18 percent).
A.1.3.3 Factors Affecting Percutaneous Absorption
Numerous factors are known to affect the rate or extent of percutan-
eous absorption (Wester and Maibach 1983). Percutaneous absorption is
highly dependent on the vehicle (Bronaugh and Franz 1986). Dermal absorp-
tion is expected to decrease when 2,3,7,8-TCDD is applied with hydrophobic
vehicles, due to its hydrophobicity. The extent of absorption of 2,3,7,8-
TCDD decreased 10-fold with a petroleum jelly vehicle, relative to solvent
deposition from methanol (Poiger and Schlatter 1980). The rate of
2,3,7,8-TCDD absorption in vitro also decreased 10-fold with a mineral
oil vehicle, as compared to solvent deposition from acetone (Weber et al.
1989). However, dermal absorption is generally increased by skin hydra-
tion, occlusion of the application site, and the presence of diseased or
damaged skin (Wester and Maibach 1983; Bronaugh et al. 1986). Percutan-
eous absorption also depends on the anatomical site of application. For
example, the scrotum may be 40-fold more permeable than the forearm
(Wester and Maibach 1983). Dermal absorption may occur more rapidly in
preterm infants, because the SC is not fully developed (Wester and Maibach
1983), while absorption of 2,3,7,8-TCDD in rats was reported to decrease
with age (Banks et al. 1989). Finally, the extent of absorption is
expected to increase as the surface of exposed skin increases (Noonan and
Wester 1980).
A.1.3.4 Future Experiments
NCASI has agreed to fund additional research on percutaneous absorp-
tion of TCDD. This work will be done in vitro with human skin and will
involve the use of aqueous vehicles, including synthetic urine (Bond
1989).
A.1.4 Discussion
Dermal exposure to PCDD and PCDF may be affected by many factors, for
example, the matrix, vehicle, exposure duration, and temperature. Thus,
dermal exposure 1s highly dependent on the exposure scenario. The effect
of these variables and others on the common assumptions discussed must be
considered in any exposure assessment. In addition, the validity of
these assumptions should be reevaluated in light of any new data which
become available.
A-ll
-------
A.1.5 References
ADL. 1987. Arthur D. Little, Inc. Exposure and risk assessment of
dioxin in bleached kraft products. U.S. Environmental Protection Agency
Contract No. 68-01-6951. June 15, 1987.
Banks YB, Brewster DW, Birnbaum LS. 1989. Age-related changes in dermal
absorption of TCDD and 2,3,4,7,8-pentachlorodibenzofuran (Pe4CDF).
Annual Meeting of the Society of Toxicology, Atlanta, GA, February 27 -
March 3, 1989. Abstract No. 469.
Bartek, MJ, La Budde JA, HI Maibach. 1972. Skin permeability in vivo:
comparison in rat, rabbit, pig, and man. Journal of Investigative
Dermatology 58:114-123.
Bond GP. 1989. Communication to the U.S. Consumer Product Safety
Commission (CPSC). September 12, 1989.
Brewster DW, Banks YB, Clark AM, Birnbaum LS. 1989. Comparative dermal
absorption of 2,3,7,8-tetrachlorodibenzo-p-dioxin and three
polychlorinated dibenzofurans. Toxicology and Applied Pharmacology
97:156-166.
Brewster DW, Elwell MR, Birnbaum LS. 1988. Toxicity and deposition of
2,3,4,7,8-pentachlorodibenzofuran (4PeCDF) in the rhesus monkey (Macaca
mulatta). Toxicology and Applied Pharmacology 93:231-146.
Bronaugh RL, Franz TJ. 1986. Vehicle effects on percutaneous
absorption: in vivo and in vitro comparisons with human skin. British
Journal of Dermatology 115:1-11.
Bronaugh RL, Stewart RF. 1984. Methods for in vitro percutaneous
absorption studies III: Hydrophobic compounds. Journal of
Pharmaceutical Sciences 73:1255-1258.
Bronaugh RL, Stewart RF. 1986. Methods for in vitro percutaneous
absorption studies VI: preparation of the barrier layer. Journal of
Pharmaceutical Sciences 75:487-491.
Bronaugh RL, Weingarten DP, Lowe NJ. 1986. Differential rates of
percutaneous absorption through the eczematous and normal skin of a
monkey. Journal of Investigative Dermatology 87:451-453.
Cruzan G, Low LK, Cox GE, Meeks JR, MacKerer CR, Craig PK, Singer EJ,
Mehlman MA. 1986. Systemic toxicity from subchronic dermal exposure,
chemical characterization, and dermal penetration of catalytically
cracked clarified slurry oil. Toxicology and Industrial Health 2:429-444.
A-12
1598q
-------
Feldmann RJ, Maibach HI. 1970. Absorption of some organic compounds
through the skin of man. Journal of Investigative Dermatology 54:399-404.
Franz TJ. 1978. The finite dose technique as a valid in vitro model for
the study of percutaneous absorption in man. Current Problems in
Dermatology 7:58-68.
IARC. 1979. International Association for Research on Cancer.
Tris(2,3-dibromopropyl)phosphate. Geneva, Switzerland: World Health
Organization. IARC Monographs 20:575-588.
NCASI. 1987a. National Council of the Paper Industry for Air and Stream
Improvement. Assessment of potential health risks from dermal exposure
to dioxin in paper products. Technical Bulletin No. 534. November 1987.
NCASI. 1987b. National Council of the Paper Industry for Air and Stream
Improvement. First progress report on the assessment of potential health
risks from use of bleached board and paper food packaging and food
contact products. Special Report 87-11. November 1987.
NCASI. . 1988. National Council of the Paper Industry for Air and Stream
Improvement. Assessment of risks associated with potential exposure to
dioxin through the consumption of coffee brewed using bleached paper
coffee filters. Technical Bulletin No. 546. May 1988.
NCASI. 1989. National Council of the Paper Industry for Air and Stream
Improvement. Interim report on measurement of pulp/aqueous solution
partition coefficients. November 1989.
Neal MU, Basu DK. 1987. Toxicological Profile for 2,3,7,8-
Tetrachloridibenzo-p-dioxin. Agency for Toxic Substances and Disease
Registry (ASTDR), U.S. Public Health Service. Contract No. 68-03-3228.
November 1987.
Noonan PK, Wester RC. 1980. Percutaneous absorption of nltroglycerin.
journal of Pharmaceutical Sciences 69:365-366.
poiger H, Schlatter Ch. 1980. Influence of solvents and adsorbent on
dermal and intestinal absorption of TCDD. Food and Cosmetic Toxicology
18:477-481.
Scheuplein RJ, Ross LW. 1974. Mechanism of percutaneous absorption V.
percutaneous absorption of solvent deposited solids. Journal of
Investigative Dermatology 62:353-360.
Shu H, Teitelbaum P, Webb AS, Marple L, Brunck B, Del Rossi D, Murray FJ,
paustenbach DJ. 1988. Fundamental and Applied Toxicology 10:355-343.
A-13
-------
Stoughton RB. 1989. Percutaneous absorption of drugs. Annual Reviews
of Pharmacology and Toxicology 29:55-69.
USEPA. 1988. U.S. Environmental Protection Agency. Estimating
exposures to 2,3,7,8-TCDD (Draft). Office of Research and Development,
Office of Health and Environmental Assessment. EPA/600/6-88/005A,
External Review Draft. March 1988.
Ulsamer AG, Porter WK, Osterburg RE. 1978. The percutaneous absorption
of radiolabeled TRIS from flame-retarded fabric. Journal of
Environmental Pathology and Toxicology 1:543-549.
Umbreit TH, Hesse EJ, Gallo MA. 1986. Bioavailability of dioxin in soil
from a 2,4,5-T manufacturing site. Science 232:497-499.
Umbreit TH, Hesse EJ, Gallo MA. 1988. Bioavailability and cytpchrome
P-450 induction from 2,3,7,8-tetrachlorodibenzo-p-dioxin contaminated
soils from Times Beach, Missouri, and Newark, New Jersey. Drug and
Chemical Toxicology 11:405-418.
Weber LWD, Zesch A, Rozman K. 1989. Penetration of TCDD into human skin
in vitro. Atlanta, GA: Annual Meeting of the Society of Toxicology,
February 27 - March 3, 1989. Abstract No. 472.
Wester RC, Maibach HI. 1983. In vivo percutaneous absorption. In: FN
Marzulli and HI Maibach, eds. Dermatotoxicology, 2nd Edition.
Washington, DC: Hemisphere Publishing Company. Chapter 5, pp. 131-146.
A.2 BIOAVAILABILITY OF INGESTED 2,3,7,8-TCDD
This section was compiled, with the author's approval, from:
Boyer I. 1989. Bioavailability of Ingested 2,3,7,8-TCDD and Related
Substances. U.S. Food and Drug Administration. Draft report
submitted to U.S. EPA, Office of Toxic Substances, June 22, 1989.
A.2.1 Gastrointestinal Absorption of 2.3.7.8-TCDD in a Volunteer
After an overnight fast, a healthly male volunteer ingested 105 ng of
[1,6-3H] 2,3,7,8-TCDD (1.14 ng TCDD/kg b.w.) dissolved in 6 ml of corn
oil (Poiger and Schlatter 1986). About 11.5 percent of the administered
3H-activity was excreted in the feces during the first three days after
exposure, followed by a considerable decrease in the rate of 3H
elimination; 3.5 percent of the dose was excreted from day 7 to day 125
post-exposure. Of the 3H-activity excreted during the first three days,
most was probably from non-absorbed TCDD. Poiger and Schlatter (1986)
suggested that the almost complete absorption (88.5 percent) of 2,3,7,8-
TCDD observed in this study was probably due to the use of corn oil as
A-14
1598q
-------
the vehicle, as well as to the fasted state of the subject, and lower
absorption may be observed when the 2,3,7,8-TCDD is ingested in food or
other carriers. The results also showed that very low doses of 2,3,7,8-
TCDD can be highly absorbed in the human gastrointestinal tract.
A.2.2 Gastrointestinal Absorption of 2.3.7.8-TCOD in Experimental
Animals
A.2.2.1 2,3,7,8-TCDD in Vegetable Oil, In Other Solvents, or In the Diet
Male and female Sprague-Dawley rats exposed to a single oral dose of
1.0 ug HC-labelled 2,3,7,8-TCDD/kg body weight (b.w.) in a mixture of
acetone and corn oil (1:25, v:v) absorbed about 66 percent to 93 percent
(mean ± SD = 84 ± 11 percent) of the total dose (Rose et al.
1976). The
mean absorption was 78 ± 13 percent for males and 90 ± 3 percent
for females. After repeated oral dosing at 0.1 or 1.0 ug/kg b.w./day,
Monday through Friday for 7 weeks, the calculated absorption was 86 ±
12 percent; the mean absorption was 87 ± 6 percent for males and 91
± 1 percent for females at 0.1 mg/kg b.w./day, and 73 ± 10
percent for males and 93 ± 2 percent for females at 1 ug/kg b.w./day.
Male Sprague-Dawley rats were exposed to a single dose of 50 ug
[14C]TCDD/kg b.w. in corn oil by stomach tube (Allen et al. 1975).
Approximately 25 percent of the dose was eliminated with the feces within
the first 3 days. About 1 to 2 percent/day was excreted during the subse-
quent 18 days. During the first three days, the fecal content probably
represented irainly un-absorbed 2,3,7,8-TCDD. Thus, more than 75 percent
of the administered dose was absorbed.
Twenty-four hours after male Syrian Golden hamsters were orally
exposed to a single 650 ug/kg dose of [l,6-3H]-labelled 2,3,7,8-TCDD in
olive oil, 74 ± 23 percent (mean ± SD) of the 3H-activity
remained unexcreted (Olson et al. 1980). Thus, approximately 74 percent
of the dose was absorbed.
Male Sprague-Dawley rats were administered 50 ug [14C]TCDD/kg b.w.
in acetone:corn oil (1:9, v:v) by gavage (Piper et al. 1973). Almost
30 percent of the administered dose was eliminated in the feces during the
first 48 hours. Most of this probably represented un-absorbed 2,3,7,8-
TCDD. About 1 to 2 percent of the l4C-activ1ty was excreted each day
in the feces over the remaining 19 days. The results suggest that about
70 percent of the dose was absorbed.
An unspecified amount of 14COabelled 2,3,7,8-TCDD was administered
to guinea pigs by gavage, and "only half of the dose was absorbed" (Nolan
et al. 1979). However, the vehicle and the method used for calculating
the percentage absorbed were not given in the report.
A-15
l598q
-------
From 54 to 67 percent of the dose was absorbed when ^C-labelled
2,3,7,8-TCDD was administered to male and female Sprague-Dawley rats in
the diet to provide 0.5 or 1.4 ug TCDD/kg b.w./day for 42 days (Fries and
Marrow 1975). The estimates were obtained by comparing total retention
at steady state IRSX%), calculated from actual total retention data and
assuming a single compartment and first-order rates, with the steady state
total retention predicted assuming 100 percent G.I. absorption (RS100%^'
Rsx% ran9ed from 9,4 to 11.6, expressed as multiples of the daily
intake, while Rsioo% ran9eo< ^rom I?-2 f°r females to 21.7 for males.
Thus, absorption was 54 to 55 percent for males and 55 to 67 percent for
females.
Male ICR/Ha Swiss mice were exposed to a single oral dose of 135 ug
[14C]TCDD/kg b.w. in 1 percent ethanol (95 percent)/10 percent Tween
80/89 percent saline {Koshakji et al. 1984). About 71 percent of the
administered 14C-activity was excreted in the feces during the first
24 hours, and an additional 1 percent during the next 10 days. Koshakji
et al. (1984) suggested that the [14C]-activity excreted in the first
24 hours after exposure represents un-absorbed 2,3,7,8-TCDD. Thus,
approximately 29 percent of the ingested dose had been absorbed.
A.2.2.2 2,3,7,8-TCDD in Soil
Soil containing 81 ± 8 ppb 2,3,7,8-TCDD from a "highly contami-
nated" area in Seveso, Italy, was suspended in 10 ml of water and
administered by gavage to albino male rabbits to provide 80 ng TCDD/day
(30,8 ng/kg b.w./day) for 7 days {Bonaccorsl et al. 19B4), For compari-
son, other rabbits were treated with 80 ng TCDD/day in alcohol/water
(1:1). The level of 2,3,7,8-TCDD in the rabbit liver after the 7-day
treatment with Seveso soil was 0.88 ± 0.28 ppb, compared to 2.7 ±
0.5 ppb for treatment with the 50 percent alcohol mixture. Thus, the bio-
availability of 2,3,7,8-TCDD from the Seveso soil appeared to be 32 per-
cent (99 percent confidence interval 5 to 60 percent) of the bioavailabil-
ity from the 50 percent alcohol mixture.
In comparison, 2,3,7,8-TCDD-free Seveso soil, with added 2,3,7,8-TCDD
("lab-contaminated soil") to provide 80 ng TCDD/day, yielded liver
retention of 56 percent compared with that observed from exposure to the
50 percent alcohol mixture (Bonnaccorsi et al. 1984). Also, rats exposed
to 40 ng TCDD/day in lab-contaminated soil exhibited only 74 percent of
the liver retention observed in rats exposed to 40 ng TCDD/day either in
the 50 percent alcohol mixture or in the acetone/vegetable oil (1:6)
mixture. Rats exposed to 20 ng TCDD/day from the lab-contaminated soil
exhibited liver retention equal to that observed in rats exposed to 20
ng/day in the acetone/vegetable oil mixture.
Guinea pigs were exposed orally to various samples of soil from either
the Times Beach or the Minker Stout site, containing 770 and 880 ppb
A-16
159Bq
-------
2,3,7,8-TCDD, respectively, to provide exposures of 1, 3, or 10 ug TCDD/kg
b.w. (McConnell et al. 1984), These soils also contained 40 to 80 ppb
dibenzofurans and 2 to 4 ppm PCBs. Control guinea pigs received soil
samples in which 2,3,7,8-TCDD, PCBs, or PCDFs were not detected. Another
group of guinea pigs received 2,3,7,8-TCDD in corn oil at 0, 1, or 3 ug
TCDD/kg b.w. IMcConnell et al. 1984). The guinea pigs were observed for
30 days after treatment with the single doses. The estimated LD5QS
were 1.75 ug/kg b.w. for 2,3,7,8-TCDD in corr oil, 7,15 ug/kg b.w. for
Times Beach soil, and 5.50 ug/kg b.w. for the Hinker Stout site soil.
Bioavailability was not calculated from these results. However, an exami-
nation of the hepatic 2,3,7,8-TCDD content in the 2,3,7,8-TCDD-exposed
guinea pigs revealed highly efficient absorption of 2,3,7,8-TCDD from the
soil samples.
Female Sprague-Dawley rats were exposed to a single oral dose of
either 2,3,7,8-TCDD in corn oil or TCDD-contaminated soil from the Minker
Stout site (McConnell et al. 1984). Induction of aryl hydrocarbon
hydroxylase (AHH) was measured in the rat livers 6 days after treatment.
The doses ranged from 0.04 to 5.0 ug TCDD/kg b.w. AHH induction in rats
receiving 2,3,7,8-TCDD in corn oil was similar to induction in rats
exposed to contaminated soil, when comparing equivalent doses of 2,3,7,8-
TCDD. In rats treated with the soil, the AHH activity was about 51 to
114 percent (overall mean * 84 percent) of the activity observed in rats
treated with 2,3,7,8-TCDD in corn oil (97 percent at 5 ug/kg b.w.;
51 percent at 1 ug/kg b.w.; 74 percent at 0.2 ug/kg b.w.; 114 percent at
0.04 ug/kg b.w.). Less than 2 percent of the inductive effect of the con-
taminated soil could be accounted for by the dibenzofurans present, and
not more than 0.2 percent by the PCBs. Thus, the TCDD in the Minker
Stout site soil was nearly as potent an inducer of AHH activity as TCDD
in corn oil, suggesting high bioavailability from the soil.
Guinea pigs were exposed to single oral doses of 2,3,7,8-TCDD in soils
(10 percent suspensions in 5 percent gum acacia) from heavily contaminated
sites in Newark, New Jersey, including a 2,4,5-T manufacturing site and a
metal salvage site (Umbreit et al. 19B6a). More than 50 different
dibenzodioxins and dibenzofurans were detected in the soil samples. The
negative control was a decontaminated soil sample from the manufacturing
site. The positive controls were 2,3,7,8-TCDD in a 1:9 mixture of acetone
and corn oil, and 2,3,7,8-TCDD placed on decontaminated soil 1 hour before
dosing ("recontaminated soil").
The soil from the 2,4,5-T manufacturing site was substantially less
toxic than equivalent doses of 2,3,7,8-TCDD in corn oil (Umbreit et al.
I986a). The 2,3,7,8-TCDD in either corn oil or recontaminated soil, both
providing 6 ug TCDO/kg b.w., proved highly toxic. These positive control
treatments produced death with typical signs of 2,3,7,8-TCDD toxlcity in
Over half the treated animals (6 ug/kg b.w.). In contrast, guinea pigs
treated with soil from the manufacturing site, at up to 12 ug/kg b.w.,
A-17
1596Q
-------
showed some signs of toxicity, but none that could be attributed to
2,3,7,8-TCDD ("TCDD syndrome"); all survived 60 days, except for those
that died as a result of gavage error. Thus, the lethal dose for 2,3,7,8-
TCDD in the manufacturing site soil was clearly greater than 12 ug/kg b.w.
Accordingly, at 60 days, liver TCDD content was 18,000 ppt in the rats
treated with recontaminated soil containing 6 ug TCDD/kg b.w., compared to
90 ppt in rats treated with manufacturing site soil containing 12 ug/kg
b.w. (Umbreit et al. 1986a). Other compounds (including other dibenzo-
dioxins and dibenzofurans) in the contaminated soil from the manufacturing
site did not appear to alter the toxicity of the 2,3,7,8-TCDD in the soil,
by inhibiting 2,3,7,8-TCDD absorption in the G.I. tract or by competing
with 2,3,7,8-TCDD at the receptor level. Thus, the relatively low toxici-
ty and low liver content of 2,3,7,8-TCDD in the rats treated with the
manufacturing site soil were attributed to low bioavailability.
In a separate experiment, soil from the metal salvage site was admin-
istered to rats to provide 0.32 ug TCDD/kg b.w. (Umbreit et al. 1986a).
Neither the contaminated soil nor the negative control (decontaminated
soil) caused any signs of toxicity. However, the liver TCDD-content was
230 ppt after treatment with metal salvage site soil (0.32 ug TCDD/kg
b.w.), compared to a TCDD liver content of 90 ppt after treatment with
manufacturing site soil (12 ug TCDD/kg b.w.). Thus, although the TCDD
content of the metal salvage site soil was lower, the bioavailability of
2,3,7,8-TCDD in this soil was substantially greater than in the manufac-
turing site soil.
In another study, Umbreit et al. (1986b, 1988b) reported that 8 of 14
guinea pigs treated with Times Beach soil at 10 ug TCDD/kg b.w. died of
TCDD intoxication within the first 20 days (average time-to-death = 14.3
± 3.2 days). In comparison, 19 of 20 positive control (recontami-
nated soil) guinea pigs died within 20 days (average time-to-death « 9.8
± 3.2 days). In contrast, the manufacturing site soil (at 10 ug
TCDD/kg b.w.) produced only a single death from apparent TCDD intoxication
out of the 18 guinea pigs treated, and that one guinea pig died 57 days
after the initiation of treatment.
Liver-TCDD contents were measured in male and female Hartley guinea
pigs for up to 60 days after exposure to a single dose of contaminated
soil from Times Beach, and the measurements were compared to the liver
contents after exposure to the manufacturing site soil (Umbreit et al.
1988a). Soils from both sources are contaminated with several chlorinated
dioxins and numerous other compounds. On the basis of liver analyses,
the bioavailability of 2,3,7,8-TCDD in the Times Beach soil (10 ug TCDD/kg
b.w.) was approximately 29.5 percent, and the bioavailability of 2,3,7,8-
TCDD in the manufacturing site soil (10 ug/kg b.w.) was about 1.6 percent,
of the bioavailability from recontaminated soil (6 ug TCDD/kg b.w.). The
liver TCDD-contents were as follows: recontaminated soil, 56,000 ppt;
Times Beach soil, 27,500 ppt; 2,4,5-T manufacturing site soil, 1,500 ppt.
A-18
1598q
-------
When an animal fed PCBs undergoes weight loss as a result of illness
or restricted food intake, the fat is metabolized and PCBs in the fat are
transferred to the liver. This process typically produces a decrease in
the PCB concentration in adipose tissue, and a corresponding increase in
the liver PCB concentration. By analogy, a toxic quantity of Times Beach
soil (at 10 ug TCDD/kg b,w.) would be expected to produce a higher liver
TCDD-content than a non-toxic quantity of the manufacturing site soil
(also at 10 ug TCDD/kg b.w.), simply because of the transfer of TCDD from
adipose tissue to liver in those animals succumbing to TCDD toxicity.
When male Sprague-Dawley rats were exposed orally to 50 ug of radiolabeled
TCDD/kg b.w. in corn oil (approximately twice the LDcn) and observed
for up to 21 days post-treatment, the rats lost weight and exhibited a
decided depletion of depot fat (Allen et al. 1975). However, the concen-
tration of radioactivity in the adipose tissue remained constant, and the
concentration in the liver showed a 62 percent decrease. Thus, the redis-
tribution of TCDD at toxic doses may not be analogous to the redistribu-
tion of PCBs.
Umbreit et al. (1986a) suggested there are several possible explana-
tions for the observed differences in soil-TCDD bioavailability. For
example, the bioavailability of dioxin may depend on the nature of the
soil and matrix material. The soil at the manufacturing site contained
asphalt and concrete as well as coarse sand-soil fill. Umbreit et al.
(1986a) suggested that the "carbonaceous" materials in the asphalt and
tar could enhance binding to the matrix and thus reduce bioavailability.
Accordingly, much less than 0.07 percent of a 14.7 ng dose administered
to rats by gavage was retained in hepatic tissues when the 2,3,7,8-TCDD
was mixed in an aqueous suspension of activated carbon (25 percent w/w)
and stored for 15 to 20 hours at room temperature before dosing. In con-
trast, 36.7 ± 1.2 percent was retained in hepatic tissues when the dose
was administered in 50 percent ethanol. These results suggest that
absorption was almost completely eliminated by the presence of activated
carbon.
Likewise, Silkworth et al. (1982) and Kaminsky et al. (1985) reported
that the 42-day LDsg values for female Hartley guinea pigs were
2.5 ug/kg b.w. for 2,3,7,8-TCDD administered by gavage in corn oil, and
19 ug/kg b.w. for 2,3,7,8-TCDD in an aqueous 0.75 percent methyl cellulose
mixture, a 7.6-fold decrease 1n toxicity associated with the presence of
the organic methyl cellulose carbon.
Also, bioavailability may be affected by differences in the applica-
tion of the 2,3,7,8-TCDD to the soil and by the residence time of 2,3,7,8-
TCDD in the soil (McConnell et al. 1984). The soil at the 2,4,5-T site
v*as contaminated over decades from a generally aqueous medium by leaks in
the manufacturing stream and spills. Furthermore, this site was open to
the environment, and solvent materials may have enhanced the percolation
into the soil and thus facilitate the binding to soil components. In
A-19
1598Q
-------
contrast, the Times Beach site was contaminated by spraying waste oil,
contaminated with 2,3,7,8-TCDD and other chlorinated dibenzo-p-dioxins,
directly onto a sandy loam-soil to control road dust (McConnell et al.
1984). The presence of the waste oil and the absence of certain other
types of material may largely account for the relatively high bioavail-
ability of 2,3,7,8-TCDD from the Times Beach soil.
The comparative extractability data also support the hypothesis that
the differences in bioavailability between Times Beach and Newark soils
are a function of soil binding. 2,3,7,8-TCDD is weakly sorbed onto Times
Beach soil as evidenced by the ease with which it is extracted. In con-
trast, 2,3,7,8-TCDD is very difficult to extract from the Newark 2,4,5-T
manufacturing site soil, indicating that it is tightly sorbed onto this
soil. Umbreit et al. (1986a) reported that Soxhlet extraction of the
Times Beach soil yielded a similar quantity of 2,3,7,8-TCDD (950 ppb) as
compared to the solvent extraction of this soil (770 ppb), as reported by
McConnell et al. (1984). In contrast, the Soxhlet-extractable 2,3,7,8-
TCDD in the 2,4,5-T manufacturing site soil was 2200 to 2280 ppb, and only
a small fraction, ">2.5 ppb," was extractable by the solvent extraction
methodology used by McConnell et al. (1984).
In addition, oral exposures of female Sprague-Dawley rats to uncontam-
inated Seyeso soil spiked with 2,3,7,8-TCDD produced lower hepatic levels
of 2,3,7,8-TCDD than exposures to 2,3,7,8-TCDD in 50 percent ethanol
(Poiger and Scnlatter 1980). Comparison of the percentages of the doses
found in the liver indicates that 2,3,7,8-TCDD absorption from the spiked
soil was 68 to 73 percent (doses » 14.7 to 22.9 ng) of the absorption from
the solvent vehicle. Aging the spiked soil for 8 days at 30 to 40°C
following the addition of 2,3,7,8-TCDD decreased the uptake into the liver
to 44 percent of the uptake of 2,3,7,8-TCDD from the 50 percent ethanol
solution. This observation is consistent with reports showing that
2,3,7,8-TCOD from environmental soil (naturally aged) was generally less
bioavailable than 2,3,7,8-TCDD freshly added to clean samples of these
soils. Likewise, Philippi et al. (1981) and Huetter and Philippi (1982)
have shown that radiolabeled 2,3,7,8-TCDD added to soil becomes progress-
ively more resistant to extraction with time.
A.2.2.3 2,3,7,8-TCDD and Related Substances 1n Soot
Kaminski et al. (1985) and Silkworth et al. (1982) have examined the
toxicity of soot containing dibenzodioxins and dibenzofurans from a poly-
chlorinated biphenyl (transformer) fire. Hartley guinea pigs were exposed
to one of the following In a single oral dose: 1) soot in an aqueous
0.75 percent methyl cellulose vehicle, or 2) Soxhlet benzene extract of
the soot in the aqueous 0.75 percent methylcellulose vehicle. The soot
contained approximately 1.2 ppm 2,3,7,8-TCDD, 48 ppm 2,3,7,8-TCDF, and
5,000 ppm PCBs, in addition to many other chlorinated dibenzodioxin and
dibenzofuran congeners. 2,3,7,8-TCDF was estimated to be the most
hazardous toxic component of the soot because of its high toxicity to
A-20
1598q
-------
guinea pigs and the relatively large quantities present. The 42-day
LD5Q of the soot in the aqueous methyl cellulose vehicle was 410 mg
soot/kg b.w. (95 percent confidence limits, 281 and 604), a dose which
would provide approximately 0.5 ug TCDD/kg b.w. (confidence limits, 0.3
and 0.7), 20 ug TCDF/kg b.w. (confidence limits, 13 and 29), and 2,050 ug
PCBs/kg b.w. (confidence limits, 1,404 and 3,020). In comparison, the ex-
tract of the soot in the aqueous methylcellulose vehicle yielded a 42-day
LDcQ of 327 mg soot equivalent/kg b.w. (95 percent confidence limits,
183 and 583), a dose which would provide approximately 0.4 ug TCDD/kg b.w.
(confidence limits, 0.2 and 0.7), 16 ug TCDF/kg b.w. (confidence limits,
9 and 28), and 1,635 ug PCBs/kg b.w. (confidence limits, 915 and 2,915).
Thus, in the 0.75 percent aqueous methylcellulose suspension, the carbon-
like matrix of the soot had little, if any, effect on the oral toxicity of
the substances in the soot.
A.2.2.4 2,3,7,8-TCDD and Related Substances in Fly-Ash
Van den Berg et al. (1985) demonstrated that 2,3,7,8-TCDD and related
substances were bioavailable from fly-ash mixed 25 percent w/w with
standard laboratory diet consumed by guinea pigs over 31 to 95 days.
Approximately 0.9 to 3.7 percent of the total 2,3,7,8-TCDD and approxi-
mately 2.2 to 4.7 percent of the total 2,3,7,8-TCDF were taken up into
the livers of the guinea pigs. Liver contents in the guinea pigs ranged
from 0.5 percent of the total 1,2,3,4,7,8-HxCDD + 1,2,3,6,7,8-HxCDD
(hexachlorinated dibenzo-dioxins) to 11.3 percent of the 1,2,3,7,8-PnCDN
(pentachlorinated dibenzofuran).
Van den Berg et al. (1983) fed Wistar rats for 19 days with a diet
mixed with one of the following: 1) fly-ash containing dioxins and
furans, 2) a Soxhlet/ toluene extract of the fly-ash, or 3) an extract of
the fly-ash purified by means of column chromatography. The rats receiv-
ing food mixed with the purified fly-ash extract retained at least 3 to 5
times as much polychlorinated dibenzodioxins (PCDDs) and polychlorinated
dibenzofurans (PCDFs) in their livers as those rats fed the fly-ash
itself. For PnCDD, HxCDF, and HxCDD isomers, the liver concentrations
were 10 to 20 times higher for food mixed with purified extract vs. the
fly-ash itself. Liver retention in rats fed food mixed with the crude
fly-ash extract was somewhat reduced compared to that in rats fed food
containing the purified extract. However, the doses of PCDDs and PCDFs
in the fly-ash were from 76 to 151 percent of the doses administered by
mixing the crude extract with the food, and from 51 to 112 percent of the
doses administered by mixing the purified extract with the food. Correct-
ing for these differences suggests that the bioavailability of TCDD from
the fly-ash itself was 70 to 75 percent as high as the bioavailability
from the crude extract mixed with the food, and 58 to 75 percent as high
as the purified extract mixed with the food.
A-21
I598q
-------
A. 2. 3 Bioavailabilltv of Ingested 2.3.4,7.8-PeCDF (pentachlorodibenzo-
furan)
Apparently, no data are available for feeding studies with 2,3,7,8-
tetrachlorodibenzofuran.
Male Fischer rats were orally exposed to 34, 170, or 340 ug 2,3,4,7,8-
PeCDF/kg b.w. by gavage (Brewster et al . 1987). The 2,3,4,7,8-PeCDF was
dissolved in acetone/corn oil, and the acetone was removed by evaporation
prior to administration. Approximately 30 percent of the dose was
excreted within 3 days, independent of the magnitude of the dose. For
rats exposed to 34 ug PeCDF/kg b.w. by intravenous (i.v.) injection, about
3 percent of the 2,3,4,7,8-PeCDF was eliminated in the feces within
3 days. The results showed that more than 70 percent of the oral dose was
absorbed.
A. 2. 4 Bioavailabilitv of Ingested OCDD foctachlorodibenzo-D-diox1n)
Toxicity studies of OCDD have not demonstrated lethality at g/kg b.w.
dosages in rats or mice, and OCDD has been classified as essentially non-
toxic by some investigators (Norback et al. 1975, Williams et al. 1972,
Schwartz et al . 1973). However, several investigators have observed
tissue accumulation following repeated exposure to OCDD (Birnbaum and
Couture 1988, Norback et al . 1975, Williams et al . 1972, Couture et al .
1988), and Schwertz et al . (1973) reported that OCDD was embryotoxlc at
500 mg/kg b.w. /day.
OCDD was administered at 100 ug in 1 ml corn oil to male Sprague-
Dawley rats by gastric intubation (Norback et al . 1975)^ The OCDD had
been bomarded with neutrons to produce 12.6 pg SCDD ([35S] thiohepta-
chlorodibenzo-p-dioxin) to serve as a marker. About 0.56 percent of the
total dose was present in the tissues after 21 days of continuous
treatment, suggesting a low rate of gastrointestinal absorption.
Was dissolved in o-dichlorobenzene (DCB), and male
Fischer 344 rats were exposed by gavage to 50, 500, or 5000 ug OCDD/kg
b.w. (Birnbaum and Couture 1988). The low dose (50 ug/kg b.w.) was
administered in the same vehicle (DCB/Emulphor 620, 1:1) and volume
(0.2 ml/kg b.w.) as for rats exposed to this dose by i.v. injection. The
middle dose (500 ug/kg b.w.) was administered in 1 ml DCB: corn oil
(l:l)/kg b.w., or in 1 ml corn oil/kg b.w. (Birnbaum and Couture 1988).
The highest dose (5000 ug/kg b.w.) was given in 1 ml corn oil/kg b.w.
For repeated exposure, the rats were treated orally once a day, 1, 2, 4,
7, or 10 times, and sacrificed 3 days after the last exposure.
Birnbaum and Couture (1988) reported that more than 85 percent of the
radioactivity was eliminated with the feces within 3 days after oral
treatment with 50 or 500 ug OCDD/kg b.w. (in DCB:corn oil). More than
A-22
1596q
-------
95 percent of the dose appeared in the feces within 3 days after exposure
to 500 and 5000 ug/kg b.w. (in corn oil). In contrast, only 13 percent
of the 50 ug OCDD/kg b.w. administered i.v. was excreted in the feces,
and less than 0.2 percent in the urine, within the first 3 days. By day
56 post-exposure, the rate of excretion in the feces was reduced to
0.04 percent/day. Thus, OCDD is poorly absorbed after oral exposure.
Birnbaum and Couture (1988) also reported that repeated doses of OCDD
resulted in increased body burden proportional to the number of doses
administered. However, as the oral dose was increased, proportionally
less OCDD was found in the tissues. At 500 ug/kg b.w. (in DCB:corn oil),
approximately one-quarter to one-half as much of the dose was found in
the liver as after 50 ug/kg b.w. (in DCB:corn oil). When the dose was
increased to 5000 ug/kg b.w., the proportion of OCDD retained in the
liver decreased further, such that a 10-fold increase in the dose (from
500 ug/kg b.w. to 5000 ug/kg b.w., in corn oil) resulted in less than a
3-fold increase in hepatic OCDD concentrations and no increase in blood
or skin concentrations. The results suggest that absorption is nonlinear
at exposure levels between 0.5 and 5 nig/kg b.w.
[14.C] OCDD was dissolved in acetone/corn oil, and the acetone was
evaporated (Couture et al. 1988). Rats were exposed to 50 ug OCDD/kg
b.w. by gastric intubation. The rats were dosed once a day, 5 days per
week, for a total of 10, 20, 40, or 65 times. The liver was the major
depot, accounting for 97 percent of the total body burden after 65 doses,
with no evidence of saturation of any of the depots. Some alterations in
the liver were observed, such as fatty vacuolization, which may be an
early indicator of potentially severe liver damage. Also, some hematolog-
ic changes were noted. However, no severe toxicity was reported. Couture
et al. (1988) estimated that approximately 10 percent of the total admin-
istered dose was absorbed, on the basis of OCDD liver retention. This
estimate is supported by the observation that absorbed OCDD 1s persistent
in the body; Birnbaum and Couture (1988) calculated that the whole-body
half-life of OCDD was at least 70 days, after exposing rats to 50 ug
OCDD/kg b.w. by i.v. injection.
At a dose of 11.5 ug [14C]OCDD/kg b.w. in corn oil, the concentra-
tion of OCDD in the liver at the higher dose volume (5 ml/kg b.w.) was
more than 6-fold higher than at the lower dose volume (1 ml/kg b.w.)
3 days after exposure (Couture et al. 1988). Thus, the low gastrointesti-
nal absorption of OCDD is probably a function of its extreme insolubility.
As the concentration of OCDD is increased in corn oil, the OCDD is
probably present more in suspension than in solution, a factor which may
account for reduced G.I. absorption at higher concentrations.
A-23
I598q
-------
A.2.5 Evaluation of Bioavailability In Humans Based on Animal Data
A.2.5.1 Estimated Bioavailability of 2,3,7,8-TCDD from Foods
One would expect slow but nearly complete absorption of pure 2,3,7,8-
TCDD from the gastrointestinal tract based on its high lipid solubility
and low solubility in water. Accordingly, the estimates of bioavailabili-
ty presented in Table A-4 indicate that 2,3,7,8-TCDD is readily absorbed
from vegetable oils or laboratory diet. Almost complete absorption
(88.5 percent) of 2,3,7,8-TCDD was noted in a human volunteer who consumed
radiolabelled 2,3,7,8-TCDD in corn oil after an overnight fast (Poiger
and Schlatter 1986). Similarly, absorption of 2,3,7,8-TCDD in various
vegetable oil mixtures ranged from 70 to 93 + 2 percent in rats and
hamsters (Rose et al. 1976, Allen et al. 1975, Piper et al. 1973). These
results suggest that 85 to 95 percent absorption is a reasonable estimate
of 2,3,7,8-TCDD bioavailability from fatty or oily foods, especially milk,
fish, and meats.
Fries and Marrow (1975) reported that rats absorbed 54 to 67 percent
of the 2,3,7,8-TCDD mixed into their unpelleted standard laboratory diet
(Table A-4). The decreased bioavailability of 2,3,7,8-TCDD in the diet is
consistent with other studies showing lower 2,3,7,8-TCDD bioavailability
in various experimental vehicles compared to the bioavailability in corn
oil (Koshakji et al. 1984, Silkworth et al. 1982, Nolan et al. 1979)
(Table A-4). Thus, 60 to 70 percent absorption appears to be a reasonable
estimate of 2,3,7,8-TCDD bioavailability from foods other than those with
high fat or oil content.
A.2.5.2 Estimated Bioavailability of 2.3.7.8-TCDD from Soil
The estimates presented in Table A-5 indicate that 2,3,7,8-TCDD added
to non-contaminated Seveso soil shortly before ingestion ("spiked soil")
was only 56 to 74 percent as bioavailable as 2,3,7,8-TCDD in solvent
vehicles (vegetable oil or 50 percent ethanol), with 95 percent confidence
limits ranging from 32 to 95 percent (Poiger and Schlatter 1980,
Bonaccorsi et al. 1984) (Table A-5). Furthermore, the 2,3,7,8-TCDD in
the spiked Seveso soil at the lowest dose tested appeared to be as bio-
available as an equivalent dose of 2,3,7,8-TCDD in acetone/vegetable oil
vehicle (Bonaccorsi et al. 1984) (Table A-5). These results attest to
the potential for high TCDD-bioavailability from soils.
On the other hand, the bioavailability of 2,3,7,8-TCDD from an
environmentally-contaminated 2,4,5-T manufacturing site soil was reduced
to 2 percent of the bioavailability from the spiked sample of non-
contaminated manufacturing site soil (Umbreit et al. 1988a) (Table A-5).
These results demonstrate the potential for marked decreases in bioavail-
ability as a result of soil-TCDD interactions. Reduced bioavailability
of 2,3,7,8-TCDD in environmentally contaminated soil compared to 2,3,7,8-
A-24
1598q
-------
in Vegetable Oi Is or in tfie Diet
Species Substance
Man 2.3.7.8-TCDD
Rat 2,3.7,8-TCDD
Rat 2.3.7.8-TCDD
Hamster 2,3.7.8-TCDO
Rat 2.3.7.8-TCDD
Guinea pig 2.3,7.8-TCDD
Rat 2,3.7.8-TCDD
House 2.3.7.8-TCDD
Rat 2.3.4.7.8-PeCDF
Rat OCDD
Rat OCDD
Rat OCDD
Dose
(ug/kg)
0.00114
1.0
O.I/day or 1.0/daya'b
50
650
50
*'
0.5 or 1.4/day for 42 days
135
34. 170. or 340
50 or 500
500 or 5.000
50/day. 10-65 times3
Vehicle/ Bioava liability
carrier (%)
corn oil (6 ml) 88.5
acetone: corn oil (1:25) 84±llc(66-93)d
86±12c(73±10-93±2)e
corn oil >75
olive oil 74±23C
acetone: corn oil (1:9) 70
H«f 50
diet 54-67
IX ethanol/lOX Tween 80/89X saline 29
corn oil > 70 for each dose
DCB:corn oi 1 <15
corn oil
-------
Table A-5. Bioavatlability of Ingested Dtoxim and Related Compounds in Soils. Soot and Fly-Ash, and Other Substances
Dose
Species
Substance
Vehicle/
carrier
Bioava liability
(X)
Relative to
Reference
BASIS FOR COMPARISON - Toxtcity (doses = estimated
Guinea pig TCOO 5.50
7.15
M inker Stout soil
Tiaes Beach soil
Guinea pig TCOO 19
32
24
Guinea pig Soot 327.000* 0.7SX aqueous aethyIcellulose 80 (30-207)1
0.75X aqueous aethy Ice llu lose 13
1.75 ug/kg in corn oil
Soxhlet extract of the soot in
0.7SX aqueous nethyIcellulose
ID,
50
2.5 ug/kg in corn oil
NcConnell et al. (1984)
Kaminski et al. (198S)
Silkimrth et al. (1982)
BASIS FOR COMPARISOH - Liver TCPO-concentratlon:
i
ro
o»
Guinea pig TCOO
Rabbit
Rat
Rat
Rat
10
Times Beach soil
Manufacturing site soil
TCOO 0.031/dayc ContaMinated Seveso soil
0.031/dayc Spiked Seveso soil
0.015/dayc Spiked Seveso soil
0.008/dayc Spiked Seveso soil
TCOO 0.04-0.06 Spiked Seveso soil
TCOO 3.2 Purified extract of fly-ash
•Ixed with food
TCOO 0.04
Aqueous activated-carbon
30 Man-contaminated manufacturing site
soil with added TCOD
2 (6 ug/Kg)
32(S-60)d Same dose in: SOX ethanol
56(32-81)d SOX ethanol
74(47-95)d SOX ethanol or veg. oil6
100 veg. oil6
68-73 Same dose in SOX ethanol
58-75 Ingesting the fly-ash itself
(1.6 ug/kg)
<2 Same dose in SOX ethanol
Umbreit et al. (1988a)
Bonaccorsi et al. (1984)
Poiger ft Schlatter (1980)
van den Berg et al.
(1983)
Poiger ft Schlatter (1980)
• Provides 0.5 ug TCOO/kg. 20 ug TCOF/kg. and 2.050 ug PCBs/kg.
b Mean (9SJC confidence interval).
c For 7 days.
d Mean (99X confidence interval).
e Veg. oil = acetone/vegetable oil (1:6).
-------
TCDD in freshly spiked soil is consistent with reports that aging the
spiked soil reduces the TCDD bioavailability (Poiger and Schlatter 1980).
Presumably, 2,3,7,8-TCDD becomes more tightly associated with the soil
over time {Philippi et al. 1981, Huetter and Philippi 1982} by adsorbing
onto soil components which act in a manner analogous to activated carbon
or aqueous methyl cellulose (Kaminski ot al. 1985, Poiger and Schlatter
1980) (Table A-5), thus reducing the bioavailability.
The 2,3,7,8-TCDD in environmentally contaminated soil samples from
the Seveso, Times Beach, and Minker Stout sites was from 24 to 32 percent
as bioavailable as TCDD in corn oil, or in spiked manufacturing site soil,
with 99 percent confidence limits of 5 and 60 percent reported for the
Seveso soil (McConnell et al. 1984, Umbreit et al. 1988a, Bonaccorsi et
al. 1984) (Table A-5). Thus, in the absence of site-specific information
concerning bioavailability, or solvent extractability, to which bioavail-
ability may be correlated (Umbreit et al. 1986a), 45 to 55 percent absorp-
tion appears to represent a reasonable estimate of 2,3,7,8-TCDD bioavail-
ability from environmentally contaminated soil.
A.2.5.3 Estimated Bioavailabiliv of 2.3.7.8-TCDD from Paper
Workers employed in the manufacturing of paper products may inhale
dioxin-contaminated paper dust that would be deposited in the airways and,
through mucocilliary action, can gain entry to the gastrointestinal tract.
Apparently, no information is available which specifically addresses the
bioavailability of 2,3,7,8-TCDD from ingested paper dust.
However, it has been argued that the bioavailability from paper dust
can be estimated from the observed bioavailability from other matrices,
such as soil, fly ash, or soot (NCASI 1988a), and that soil serves as the
best surrogate for paper dust because the primary source of the organic
components of soil, namely decaying plant (cellulosic) materials, would
be expected to be the most similar to the organic components of paper.
On this basis, NCASI (1988a) estimated that the bioavailability of
2,3,7,8-TCDD from soil was 57 percent, based on a review of the available
scientific literature, and proposed 50 percent bioavailability for
2,3,7,8-TCDD from paper. Fifty percent bioavailability from paper dust
is suggested to be an over-estimate of the actual bioavailability because
1) paper dust has a higher "organic" content than soil, and 2) the much
lower doses expected from paper dust may be associated with lower bio-
availability than the higher doses tested in the soil studies.
However, Poiger and Schlatter (1986) showed that even extremely low
doses of 2,3,7,8-TCDD can be highly absorbed in the human G.I. tract.
Moreover, the animal data suggest that decreases in the doses of 2,3,7,8-
TCDD added to non-contaminated soil can exhibit increases In the bioavail-
ability. In addition, although there may be an inverse relationship
between 2,3,7,8-TCDD bioavailability and the presence of certain sub-
A-27
I598q
-------
stances in the soil, such as charcoal, tar, and other "carbonaceous"
materials, it seems unlikely that these "carbonaceous" materials can be
equated with "organic" material, such as the cellulose in paper. Like-
wise, Kaminski et al. (1985) reported that a 0.75 percent aqueous methyl -
cellulose suspension reduced the bioavailability of 2,3,7,8-TCDD compared
to 2,3,7,8-TCDD in corn oil, but the aqueous methyl cellulose suspension
is probably not comparable to the cellulose matrix of paper.
On the other hand, Umbreit et al. (1986a) proposed that a direct rela-
tionship exists between the solvent extractability of 2,3,7,8-TCDD from
soil and the bioavailablity of 2,3,7,8-TCDD from the soil. By analogy,
the degree to which 2,3,7,8-TCDD can be extracted from paper dust may be
related to its bioabailability from the paper matrix.
NCASI (1987) reported that greater amounts of 2,3,7,8-TCDD and
2,3,7,8-TCDF were extracted from a non-barrier food packaging paper in a
50 percent ethanol solution compared with the amounts originally thought
to be present in the packaging material, suggesting highly efficient
extraction. Alcohol solutions apparently swell paper pulp fibers, as does
water itself (Browning 1963, Kress and Bialkowsky 1931, McKenzie 1957).
As the paper fiber swells, the intermolecular bonds break, the degree of
order within the fiber diminshes, the surface area elevates, and the
number of available inter- and intra-molecular bonding sites increases.
The ethanol forms hydrogen-bonds with hydroxy groups in the swollen fiber,
creating a more lipophilic environment at the fiber. The NCASI (1987)
report suggests that the process probably facilitates desorption and
subsequent diffusion of PCDD/PCDF occluded within the fiber, providing a
reasonable explanation for the relatively efficient extraction of TCDD
and TCDF from paper into 50 percent ethanol. This phenomenon may also
account, to some extent, for observations of relatively high migration of
2,3,7,8-TCDD and 2,3,7,8-TCDF from paper coffee filters to the coffee
brewed using these filters, with up to 62 percent migration for 2,3,7,8-
TCDD and up to 79 percent or more for 2,3,7,8-TCDF (NCASI 1988b).
In comparison, bile salts in the intestines combine with lipids to
form micelles (water soluble complexes) from which the lipids can be more
easily absorbed (the hydrotropic effect). The bile salts reduce surface
tension and, in conjunction with fatty acids and glycerides, emulsify
fats and oils prior to their digestion and absorption. Water in the G.I.
tract can swell ingested paper, providing the digestive juices with
increased accessibility to the paper matrix. In turn, the emulsion in
the intestines may present a more lipophilic environment to the swelled
paper matrix. Thus, by mimicking the action of 50 percent ethanol, the
fluids in the G.I. tract may support a relatively high rate of migration
of TCDD and TCDF from the paper to a highly absorbable fraction of the
intestinal fluids.
A-28
1598q
-------
The foregoing argument suggests that the bioavailability of 2,3,7,8-
TCDD from ingested paper may be similar to the bioavailability from foods
with low fat or oil content. Thus, in the absence of adequate data, 60 to
70 percent absorption appears to be a reasonable estimate of 2,3,7,8-TCDD
bioavailability from ingested paper.
A.2.5.4 Estimated Bioavailabilitv of 2.3.7.8-TCDD from Sludge
No information is available which specifically addresses the bioavail-
ability of 2,3,7,8-TCDD from ingested sludge. However, one may suppose
that in the absence of oily substances in the sludge, the bioavailability
of 2,3,7,8-TCDD from the sludge might be less than that from solvent
vehicles, such as corn oil, or from fatty or oily foods. Instead, the
bioavailability of 2,3,7,8-TCDD in sludge might be best compared with the
bioavailability in foods with relatively low fat or oil content. Thus,
in the absence of adequate data, 60 to 70 percent absorption appears to
be a reasonable estimate of 2,3,7,8-TCDD bioavailability from sludge.
A.2.5.5 Estimated Bioavailabilitv of 2.3.7.8-TCDD from Water
No information is available which specifically addresses the bioavail-
ability of 2,3,7,8-TCDD in drinking water. However, one may suppose that
the 2,3,7,8-TCDD in drinking water will be present at the maximum solubil-
ity of 2,3,7,8-TCDD in water, and that the solubilized 2,3,7,8-TCDD is
readily absorbed in the G.I. tract. Thus, in the absence of adequate
data, 100 percent absorption appears to be a reasonable assumption for
2,3,7,8-TCDD bioavailability from drinking water.
A.2.6. Conclusions
Table A-6 presents the bioavailability estimates for ingested 2,3,7,8-
TCDD from various media, based on the information presented or assumptions
made in the foregoing discussion.
A.2.7. References
Allen JR, van Miller JP, Norback DH. 1975. Tissue distribution,
excretion, and biological effects of [I4C] tetrachlorodibenzo-p-dioxin
in rats. Food Cosmet. Toxicol. 13: 501-505.
Birnbaum LS, Couture LA. 1988. Disposition of octachlorodibenzo-p-
dioxin (OCDD) in male rats. Toxicol. Appl. Pharmacol. 93: 22-30.
Bonaccorsi A, di Domenico A, Panelli R, Merli F, Motta R, Vanzati R,
Zapponi GA. 1984. The influence of soil particle absorption on 2,3,7,8-
tetrachlorodibenzo-p-dioxin biological uptake in the rabbit. Arch.
Toxicol., Suppl. 7: 431-434.
A-29
I598q
-------
B902H
Table A-6. Bioavailability of Ingested 2.3.7.8-TCDD from Various Media
Medina
Bioavailability
(X of oral dose)
Drinking Hater
Fatty or oily foods
(e.g., Milk, fish,
Other foods
Paper dust
Sludge
Soil
its)
100
85-95
60-70
60-70
60-70
45-55
A-30
-------
Brewster DW, Birnbaum LS. 1987. Disposition and excretion of 2,3,7,8-
pentachlorodibenzofuran in the rat. Toxicol. Appl. Pharmacol.
90: 243-252.
Browning BL. 1963. The wood-water relationship: IV. Swelling and
shrinkage. In: The Chemistry of Wood, Browning BL. ed., Chapter 9, New
York: Interscience Publishers, Division of John Wiley and Sons,
pp. 429-431.
Couture LA, Elwell MR, Birnbaum LS. 1988. Dioxin-like effects observed
in male rats following exposure to octachlorodibenzo-p-dioxin (OCDD)
during a 13-week study. Toxicol. Appl. Pharmacol. 93: 31-46.
Fries GF, Marrow GS. 1975. Retention and excretion of 2,3,7,8-
tetrachlorodibenzo-p-dioxin by rats. J. Agric. Food Chem. 23: 265-269.
Huetter R, Philippi M. 1982. Studies on microbial metabolism of TCDD
under laboratory conditions. Pergamon Ser. Environ. Sci. 5: 87-93.
Kaminski LS, DeCapiro AP, Gierthy JF, Silworth JF, Tumasonis C. 1985.
The role of environmental matrices and experimental vehicles in
chlorinated dibenzodioxin and dibenzofuran toxicity. Chemosphere
14: 685-695.
Koshakji RP, Harbison RD, Bush MT. 1984. Studies on the metabolic fate
of [14C]2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the mouse.
Toxicol. Appl. Pharmacol. 73: 69-77.
Kress 0, Bialkowski H. 1931. Some chemical and physical observations on
hydration. Paper Trade Jour. 93: 35-44.
McConnell EE, Lucier GW, Rumbaugh RC, Albro PW, Harvan DJ, Mass JR,
Harris MM. 1984. Dioxin in soil: Bioavailability after ingestion by
rats and guinea pigs. Science 223: 1077-1079.
McKenzie AW. 1957. The structure and properties of paper: VI. The
effect of swelling pretreatments on interfibre bonding capacity.
Australian J. Appl. Sci. 8: 35-41.
NCASI. 1987. First progress report on the assessment of potential
health risks for use of bleached board and paper food packaging and
contact products. Special report 87-11. New York: National Council of
the Paper Industry for Air and Stream Improvement, Inc. November 1987.
27 pp.
NCASI. 1988a. Risks associated with dioxin exposure through inhalation
of paper dust in the workplace. Technical Bulletin No. 537. New York:
National Council of the Paper Industry for Air and Stream Improvement,
Inc. January 1988, 28 pp.
A-31
1598q
-------
NCASI. 1988b. Assessment of the risks associated with potential exposure
to dioxin through the consumption of coffee brewed using bleached coffee
filters. Technical Bulletin No. 546. New York: National Council of the
Paper Industry for Air and Stream Improvement, Inc. May 1988, 34 pp.
Nolan RJ, Smith FA, Hefner JG. 1979. Elimination and tissue
distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in female
guinea pigs following a single oral dose. Toxicol. Appl. Pharmacol.
48: A162.
Norback DH, Engblom JF, Allen JR. 1975. Tissue distribution and
excretion of octachlordibenzo-p-dioxin in the rat. Toxicol. Appl.
Pharmacol. 32: 330-338.
Olson JR, Gasiewicz TA, Neal RA. 1980. Tissue distribution, excretion,
and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the
Golden Syrian hamster. Toxicol. Appl. Pharmacol. 56: 78-85.
Philippi M, Krasnobagew V, Zeyer J, Huetter R. 1981. Fate of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) in microbial cultures and soil under
laboratory conditions. FEMS Symp. 12: 2210-2330.
Piper WN, Rose JQ, Gehring PJ. 1973. Excretion and tissue distribution
of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the rat. Environ.
Health Perspect. 5: 241-244.
Poiger H, Schlatter C. 1980. Influence of solvents and absorbents on
dermal and intestinal absorption of TCDD. Fd Cosmet. Toxicol.
18: 477-481.
Poiger H, Schlatter C. 1986.
Chemosphere 15: 1489-1494.
Pharmacokinetics of 2,3,7,8-TCDD in man.
Rose JQ, Ramsey JC, Wentzler TH, Hummel RA, Gehring PJ. 1976. The fate
of 2,3,7,8-tetrachlorodibenzo-p-dioxin following single and repeated oral
doses to the rat. Toxicol. Appl. Pharmacol. 36: 209-226.
Schwetz BA, Norris JM, Spaeschu GL, Rowe VK, Gehring PJ, Emerson JL,
Gerbig CG. 1973. Toxicology of chlorinated dibenzo-p-dioxins. Environ.
Health Perspect. 5:87-99.
Silkworth J, McMartin D, DeCaprio A, Rej R, O'Keefe P, Kaminsky L.
1982. Acute toxicity in guinea pigs and rabbits of soot from a
polychlorinated biphenyl-containing transformer fire. Toxicol. Appl.
Pharmacol. 65: 425-439.
A-32
1598q
-------
Umbreit TH, Hesse EJ, Gallo MA. 1986a. Comparative toxicity of TCDD
contaminated soil from Times Beach, Missouri, and Newark, New Jersey.
Chemosphere 15: 2121-2124.
Umbreit TH, Hesse EJ, Gallo MA. 1986b. Bioavailability of dioxin in
soil from 2,4,5-T manufacturing site. Science 121: 497-499.
Umbreit TH, Hesse EJ, Gallo MA. 1988a. Reproductive studies of C57B/6
male mice treated with TCDD-contaminated soils from a 2,4,5-trichloro-
phenoxyacetic acid manufacturing site. Arch. Environ. Contam. Toxicol.
127: 145-150.
Umbreit TH, Hesse EJ, Gallo MA. 1988b. Bioavailability and cytochrome
P-450 induction from 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)
contaminated soils from Times Beach, Missouri, and Newark, New Jersey.
Drug and Chemical Toxicol. 11: 405-418.
van den Berg M, 01ie K, Hutzinger 0. 1983. Uptake and selective
retention in rats of orally administered chlorinated dioxins and
dibenzofurans from fly-ash and fly-ash extract. Chemosphere 12: 537-544.
van den Berg MN, de Vroom A, van Greevenbroek M, 01ie K, Hutzinger 0.
1985. Bioavailability of PCDDs and PCDFs adsorbed on fly ash in the rat,
guinea pig, and Syrian golden hamster. Chemosphere 14: 865-869.
Walsh J. 1977. Seveso: The questions persist where dioxin created a
wasteland. Science 197: 1064-1067.
Williams DT, Cunningham HM, Blanchfield BJ. 1972. Distribution and
excretion studies of octachlorodibenzo-p-dioxin in the rat. Bull.
Environ. Contam. Toxicol. 7: 57-62.
A.3 BIOAVAILABILITY OF INHALED VAPORS AND PARTICLES CONTAINING
2,3,7,8-TCDD and 2,3,7,8-TCDF
A.3.1 Introduction
The purpose of this section is to provide a set of common assumptions
that can be used for assessing bioavailability of inhaled 2,3,7,8-TCDD and
2,3,7,8-TCDF. This section was compiled from:
NCASI. National Council of the Paper Industry for Air and Stream
Improvement. 1988. Risks Associated with Dioxin Exposure Through
Inhalation of Paper Dust in the Workplace. Technical Bulletin No.
537. January, 1988.
and
A-33
1598q
-------
DiCarlo F. 1989. Dioxins in Paper Products: Bioavailability by
Inhalation. Memorandum from F. J. DiCarlo (EPA, Office of Toxic
Substances) to G. Schweer (EPA, Office of Toxic Substances). June
16, 1989.
and
Griffin S. 1989. Inhalation Exposure to Dioxin Particles and
Vapors. Memorandum from S. Griffin (EPA, Office of Solid Waste) to
G. Schweer (EPA, Office of Toxic Substances). July 27, 1989.
and
Abt Associates. 1989. Multimedia Exposure Assessment for Re-use and
Disposal of Sludge from the Pulp and Paper Industry and Disposal of
Paper Products. Final Draft Report. Prepared for EPA, Office of
Toxic Substances. September 29, 1989. Sect. 2.4.4, pp. 154-171.
The bioavailability of inhaled chemicals is dependent on the physical
and chemical characteristics of the inhaled chemical and on the anatomy
and physiology of the respiratory tract. The deposition and retention of
inhaled chemicals are a function of the anatomic features of the respira-
tory tract including the alveolar surface area, the structure and spatial
relationships of conducting airways leading into the alveoli, and the
overall lung volume. Particle deposition and absorption are also a func-
tion of the respiratory rate and tidal volume. Inhaled gases, vapors,
and aerosols are different, however, and can be absorbed throughout the
respiratory tract, primarily by diffusion. Gases and vapors are generally
absorbed rapidly from the respiratory tract, although the solubility of
the gas or vapor in water is an important characteristic in determining
the relative rate of transport from the gaseous phase into the liquid
phase lining the respiratory tract.
A.3.2 Inhalation of Vapors and Bioavailability
No toxicokinetic data are available for assessing the bioavailability
of inhaled dioxin vapors. However, it may be assumed that, due to their
high lipophilicity, there would be 100 percent absorption of the 2,3,7,8-
TCDD or 2,3,7,8-TCDF vapors entering the respiratory tract (DiCarlo 1989,
Griffin, 1989).
A.3.3 Participate Inhalation and Deposition
The probability of inspiration of a particle 1s a function of its
size. This relationship Indicates that the larger the aerodynamic diam-
eter, the lower the "efficiency" of inhalation (probability of inhalation
into the alveolar spaces). Once a particle is inhaled, its ultimate fate
is again a function of its size (NCASI 1988). Four different modes of
A-34
1598q
-------
deposition may occur. First, a particle may deposit in the head and neck
regions of the airways. This deposition is termed extra-thoracic deposi-
tion (ETD). Second, particles which pass through the head and neck air-
ways may deposit in the thoracic regions of the lung, the trachea to the
terminal bronchi. This deposition is tracheobronchial deposition (TD).
Third, particles which pass through the tracheobronchial airways may
deposit in the respiration regions of the lung, the alveolar or gas-
exchange regions. This is respiratory deposition (RD). Fourth, some
particles may not deposit in any region of the lung and are exhaled
(NCASI 1988).
The particulate deposition pattern and rates for nose breathing are
different than for mouth breathing. Particles entering through the nose
have ETD which is divided between the anterior (front) portion of the nose
and the main nasal passages. .With nose breathing, complete deposition can
be expected for particles larger than 4 /*, and of these particles
deposited, 80 percent impact in the anterior nasal passages and 20 percent
impact in the main nasal passages (NCASI 1988).
The deposition pattern for particles up to 100 /* can be calculated
for mouth breathing. Particle-size normalized deposition for particles of
(a) da < 2 M, (b) 2 n < da < 10 /* and (c) 10 A* > da
having combined ETD + TD or RD have been calculated and are presented in
Table A-7.
In a similar manner, the deposition pattern for nose breathing has
also been calculated (Table A-7). For nose breathing, ETD, TD, or RD were
calculated to determine the particle-size normalized deposition for parti-
cles (a) da < 2 M, (b) 2 n < da < 10 n and (c) 10 n > da.
Just as the site of particle deposition varies with particle size and
the type of breathing (mouth vs. nose), the fate of the particle varies
with the site of deposition. For mouth breathing, particles having ETD
impact in the mouth or throat and are swallowed, resulting in an ingestion
route of exposure. Particles having TD are cleared from the lung via the
muco-ciliary escalator (the mechanism of clearing mucus from the lung)
into the throat, and are ultimately swallowed. Particles having RD could
have either of three fates. They either (I) remain deposited in the
alveoli (the gas-exchange region of the lung) or are phagocytized
(engulfed) by macrophages (cells of the immune system) and (2) enter the
lymphatic system or (3) are cleared via the mucociliary escalator and are
swallowed.
For nose breathing, particles having TD are ingested and those with RD
have pulmonary deposition, the same fate as for mouth breathing. However,
the fate of particles having ETD is different. ETD in the main nasal
passages results in ingestion of these particles. ETD in anterior regions
of the nose results in particles being trapped in nasal mucus and being
A-35
1598q
-------
Table A-7. Size-Weighted Average Participate Deposition
Size
range
(u)
Average
size
(u)
ETO
+
TO
(X)
ETO
(X)
TO
(X)
RD
(X)
Nouth Breathing
10-100 55 52.4 0
2-10 6 69.6 12.2
0-2 1 18.7 53.1
Nose Breathing
10-100
2-10
0-?
55
6
I
52.4
78.7
43.9
0
2.17
6.93
0
2.36
47.9
Source: HCASI (1988).
A-36
-------
cleared by mechanical clearing mucus from the nose (nose wiping, blowing
the nose, or sneezing). Little systemic absorption of toxicants is
expected from deposition in the anterior portions of the nasal passages
because of the thickness of the mucus, the limited blood perfusion, and
short residence time, about 12 hours. Thus, any nose breathing during the
period of exposure to particulates decreases the ingestion of particulates
and the subsequent dioxin exposure.
A.3.4 Bioavailabilitv of Particulate-Bound 2.3.7.8-TCDD and
2.3.7.8-TCDF
The bioavailability of 2,3,7,8-TCDD and 2,3,7,8-TCDF via particulate
inhalation is a function of the pathway of uptake, nasal deposition, in-
gestion or pulmonary deposition, and also the relative rate of adsorption
into the body. However, there is no experimental data on bioavailability
of dioxin from paper dust (or any cellulosic matrix) by any route of expo-
sure nor is there any data for bioavailability from sludge or soil
inhaled from the air.
Based on the limited information available, DiCarlo (1989) utilized
two simplifying assumptions for estimating the bioavailability of dioxins
from inhaled particles. The first is that dioxins on the inhaled smaller
particles (up to 10 n in diameter) are almost completely absorbed
(i.e., nearly 100 percent uptake). This assumption can be justified be-
cause the smaller particles in this range will reach the alveoli where
they may either remain for long periods of time (allowing complete absorp-
tion of dioxins) or undergo engulfment by macrophages followed by trans-
port to the pharynx where they can be swallowed with sputum. The larger
particles in the 2 /j < da < 10 » range will be deposited in the
tracheobronchiolar region of the lung, then carried to the pharynx by the
mucociliary escalator and swallowed.
The second simplifying assumption utilized by DiCarlo (1989) was that
about 25 percent of the dioxins in the larger particles (>10 n) are
absorbed. This assumption can be justified because the particles will be
deposited in the nasopharyngeal region from which about half will be
exhaled or mechanically removed from the body by sneezing or nose blowing
and about half of the particles may be swallowed; about half of the PCODs
and PCDFs on these ingested particles will then be absorbed.
A.4 UPTAKE IN TERRESTRIAL PLANTS
The purpose of this section is to provide plant uptake ratios for
2,3,7,8-TCDD for the edible portions of various types of plants. This
section was compiled from:
Sargeant A. 1989. Dioxin Uptake in Terrestrial Plants. U.S. EPA,
A-37
1598q
-------
Office of Research and Development. Memorandum to C. Cinalli, EPA,
Office of Toxic Substances, June 21, 1939.
Aboveground crops and root crops are treated separately, since the
literature suggests that uptake factors differ. Although it also appears
that uptake into reproductive structures (fruits, seeds) is lower than in
other aboveground parts, the available data are not sufficient to quantify
this difference. Therefore, all aboveground portions are treated as one
group.
A.4.1 Root Crops
A range of root-to-soil ratios can be estimated for fleshy root crops
based on the literature for 2,3,7,8-TCDD, PCBs, and pesticides. The root-
to-soil ratios described below are expressed as the wet-weight concentra-
tion of 2,3,7,8-TCDD in the crop divided by the dry-weight concentration
of 2,3,7,8-TCDD in the soil.
It is assumed that the root vegetables will be scrubbed but not peeled
before consumption. This is an important assumption, since many research-
ers have found higher concentrations of hydrophobic organic chemicals in
the peels of root crops (Iwata et al. 1974, Sacchi et al. 1986, Wipf et
al. 1982). A summary of the estimated root-to-soil ratios is presented
in Table A-8.
As can be seen in Table A-8, the root-to-soil ratios ranged approxi-
mately three orders of magnitude, from 0.003 to 3. The range of ratios
can be reduced somewhat by considering data on outer tissues separately
from data on the whole root. For outer tissues or peels, the range of
values is from 0.2 to 2, with root-to-soil point estimate of 2. It should
be noted, however, that peel-to-soil ratios reported by Wipf et al. (1982)
are well below this range.
For whole roots, the range of values was from 0.01 to 1, with a root-
to-soil point estimate of 0.5. This range includes the higher end of some
reported ratios (Wipf et al. 1982), but the value of 2.83, reported by
Moza et al. (1979) for uptake of 2,4',5-trichlorobiphenyl, is not
included in the range. 2,4',5-trichlorobiphenyl has a lower Kow than
2,3,7,8-TCDD, while the other PCB congener included in the same study,
2,2',4,4',6-pentachlorobiphenyl, has a Kow more comparable to 2,3,7,8-
TCDD, was found to be taken up to a lesser extent than the 2,4',5-
trichlorobiphenyl congener, and was included in the range of values used.
Root-to-soil ratios will depend on characteristics of both the soil
and root. Because these characteristics were not quantitatively consid-
ered, the root-to-soil ratios presented may over- or underestimate uptake
occurring in the field for the following reasons:
A-38
159Sq
-------
090ZH
Table A-8. Root-to-Soil Ratios for 2.3.7.B-TCDD and PCBs
Root:so11 ratio
System
pots
pots
Field
Field
pots
Hydroponic
Chemical
TCOD
TCOD
TCOD
PCBs
PCBs
PCBs
TCDD*1
Plant
Carrots
Maize
Beans
Carrots
Potatoes
Onions
Narcissus
Carrots
Carrots
Sugar beets
Barley
Inner
root
0.0015 to 0.0075
NAC
NA
0.69
0.35
0.56
0.42
NA
NA
NA
NA
Whole
root
0.003 to 0.015
NA
NA
0.14
0.019
0.56
0.43
0.03 to 0.5
0.92 to 2.83
0.061 to 0.16
0.015 to 0.24
Outer
n»tb
0.01 to 0.05
0.98 to 1.8
1.5 to 2.4
1.43
0.19
0.55
0.49
NA
NA
NA
0.11 to 1.78
Reference
Wipf et al. 1982
Facchetti et al. 1986
Cocucci et al. 1979
Iwata fc fiunther 1976
Noza et al. 1979
Briggs et al. 1982
Ratios are expressed as concentration per wet weight of root divided by concentration per dry weight
of soil.
Ratios calculated for peels, outer portions of fleshy roots, and fibrous roots are included under
this heading.
NA « not available.
Ratio for TCDD was calculated based on a correlation derived using pesticides.
A-39
-------
• Attributes cf the roots that may influence uptake will include
their oil content and surface-to-volume ratio. Ratios calculated
from studies using fibrous roots, because of their high surface-to-
volume ratio, were assumed to be more reflective of the peels of
fleshy roots.
• Roots grown in hydroponic systems were assumed to adequately
represent roots grown in soil.
• Soil attributes that will influence uptake include organic carbon
content, water content, and particle size distribution. However,
many investigators did not report information concerning these soil
parameters.
Limited data were available describing uptake of 2,3,7,8-TCDD by root
crops. To expand the number of studies on which to evaluate root-to-soil
ratios, experiments using PCBs and a correlation based on pesticides were
included in the analysis. Not all data in the literature were reported as
wet weight concentration in tissue divided by dry weight concentration in
soil. When necessary, the ratios were converted to comparable units using
standard assumptions for water content of soil. In addition, an attempt
was made to use the data on peels and on inner tissues to calculate ratios
for the whole root. This was done by assuming the peel accounts for
14 percent of the weight of whole carrots (Iwata et al. 1974) and 10 per-
cent of the weight of other vegetables such as potatoes and sugar beets
(USDA 1975). Finally, because of their higher surface-to-volume ratios,
the root-to-soil ratios calculated for fibrous roots were assumed to be
most comparable with the outer tissues of fleshy roots.
Wipf et al. (1982) reported 2,3,7,8-TCDD concentrations in carrots
grown in contaminated soil. Uptake factors ranged from 0.0015 to 0.0075
for the peeled carrots, and 0.01 to 0.05 in peels. The organic carbon
content of the soil was not reported. Using the assumption of Iwata et
al. (1974) that the peel accounts for 14 percent of the weight of the
whole carrot, whole root-to-soil ratios would range from 0.003 to 0.015.
Somewhat higher concentrations of 2,3,7,8-TCDD have been found by
Fachetti et al. (1986) in the roots of maize and soybeans than in the
surrounding soil. It should be noted that root-to-soil ratios were much
higher in "blank" pots. Because these higher root-to-soil ratios may be
a result of analytical limitations, only data from the experimental pots
were.used to this analysis. Root-to-soil ratios from these pots ranged
from 0.98 to 2.4.
Concentrations of 2,3,7,8-TCDD in vegetables grown in the Seveso area
of Italy have been reported by Cocucci et al. (1978). Outer tissues of
carrots, potatoes, and onions did not consistently contain higher residues
than inner tissues. This study should be treated with some caution,
A-40
1598q
-------
because the results as described in the text do not agree with the data
provided in their table. The ratios were calculated based on the data
presented in the table. The ratios were reported as fresh weight of
soil, so the ratios were converted to dry weight by assuming that the
soil was at field capacity, and hence 17 percent water. Root-to-soil
ratios for 2,3,7,8-TCDD for inner tissues were 0.69, 0.35, 0.56, and 0.42
for carrots, potatoes, and onions, and narcissus, respectively, and the
corresponding ratios for outer tissues were 1.43, 0.19, 0.55, and 0.49,
respectively. These data were used to calculate a ratio for whole roots
by assuming that the outer tissues comprised 10 percent of the weight of
the whole root. The resulting whole-root-to-soil ratios were 0.76, 0.33,
0.56, and 0.43 for carrots, potatoes, onions, and narcissus, respectively.
Iwata and Gunther (1976) grew carrots in PCB-contaminated field plots
having very low organic carbon content (0.35 percent). The whole-root-to-
soil ratios ranged from 0.03 to 0.50. Although specific congeners were
not identified, the uptake factors decreased with increasing chlorination.
Moza et al. (1979) grew carrots and sugar beets in outdoor boxes con-
taining either 2,4'5-trichlorobiphenyl (TCB) or 2,2',4,4',6-pentachlorobi-
phenyl (PeCB). Whole root-to-soil ratios for the TCD ranged from 0.16 for
sugar beets to 2.83 for carrots. Uptake of the PeCB was lower; root-to-
soil ratios ranged from 0.061 for sugar beets to 0.92 for carrots. Moza
et al. (1979) suggested that uptake into carrots would be expected to be
greater than sugar beets because of the higher oil content of carrots.
Briggs et al. (1982) studied a variety of pesticides in a hydroponic
system using barley, and the uptake of organic chemicals into barley roots
from solution was very dependent on the octanol:water partition coeffi-
cient (Kow) of the chemical. The wet-weight root concentration factor
(RCF) was defined as the concentration of the chemical in the root divided
by the concentration of chemical in solution surrounding the root, and was
related to Kow by the following equation:
log (RCF - 0.82) - 0.77 log Kow - 1.52
where
RCF - root-.soil solution partition coefficient (mg/kg root divided
by mg/kg solution)
Kow - chemical-specific octanol:water partition coefficient.
According to this equation, only those chemicals that are dissolved
in solution in the Interstitial spaces between soil particles are avail-
able for uptake by plants. The concentration of chemicals in the soil
solution can be calculated (for equilibrium conditions) using a simple
partition model:
A-41
1598q
-------
Cs
Csol =
(Koc) (foc)
where
Csol = concentration of chemical in soil solution (mg/1)
Cs = concentration of chemical in soil (mg/kg)
Koc - organic carbon partition coefficient
foc = fraction organic carbon in soil.
Combining these two equations, a root:soil partition coefficient
(uptake factor or UF) can be calculated using the following formula:
RCF
UF = (A-8)
In order to solve this equation, values are needed for Kow, Koc,
and foc. It was assumed that log Kow of 2,3,7,8-TCDD ranged from 6.15
(Schroy et al. 1985) to 7.02 (Burkhard and Kuehl 1986), log KQC was 5.68
(Schroy et al. 1985), and soil organic carbon content ranged from 0.9 to
3 percent (Brady 1974).
Using these variables, the estimated uptake factor for 2,3,7,8-TCDD
ranges from 0.11 to 1.78. There are some important limitations in using
this information for estimating uptake of 2,3,7,8-TCDD by root crops.
First, barley roots are fibrous and may not accurately represent uptake
into fleshy roots, such as carrots or potatoes. Second, the chemicals
studied were mostly organic pesticides, none of which were as hydrophobic
as 2,3,7,8-TCDD. Finally, it must also be assumed that roots grown
hydroponically adequately represent roots grown in soil.
If it is assumed that the uptake factor applies only to the peel of
root vegetables, and that the peel is 14 percent of the whole carrot
weight, estimated uptake factors for the whole carrot would range from
0.015 to 0.25. The lower end of this range corresponds to soils having
high organic content, and the upper corresponds to soils of low organic
content.
A.4.2 Aboveqround Crops
USEPA (1988) uses a preliminary estimate of 2 percent for the plant-
to-soil ratio for 2,3,7,8-TCDD for aboveground crops, based primarily on
the data presented in Sacchi et al. (1986) and Uipf et al. (1982). There
A-42
1598q
-------
does not appear to be any additional recent work to support modification
of this factor. However, this factor could under- or overestimate uptake
in aboveground portions of plants for the following reasons:
• Some studies have found plant-to-soil ratios much higher than
2 percent (although most of these were connected with the Seveso
accident), so the 2 percent factor may underestimate concentrations.
• 2,3,7,8-TCDD was not detected (detection limit 1 ppt; soil level
-10,000 ppt) in edible aboveground portions (fruits and grain) of
plants (Wipf et al. 1982), which suggests that a 2 percent ratio
may overestimate actual concentrations.
• In contrast, Sacchi et al. (1986) found significant 2,3,7,8-TCDD
uptake in aboveground portions of maize and soybeans, which suggests
that the compound may indeed be translocated, but will partition
unevenly between leaves, fruits, and grain.
• Many workers are convinced that volatilization is a significant
fate of 2,3,7,8-TCDD in the soil environment. Because the compound
is. very lipophilic, it is possible that 2,3,7,8-TCDD volatilizing
from the soil could adsorb onto plant cuticles and perhaps enter
leaves via stomates. Most experiments were not designed to distin-
guish between translocation and volatilization as separate above-
ground contamination pathways.
• Another source of uncertainty comes from the use of radio labeled
2,3,7,8-TCDD. Stock solutions are typically contaminated with
unknown compounds at a rate of about 1 percent. If the unknown
compounds happen to be substances that are readily taken up and
translocated, assays that rely on scintillation counting alone will
not provide an accurate measure of the specific distribution of
2,3,7,8-TCDD in the plant. It is possible to validate scintillation
with additional GC-MS analysis, but this procedure was not always
included in experiments.
• Hydroponic studies, while having advantages, may complicate
analysis because of the difficulty of keeping 2,3,7,8-TCDD in solu-
tion. The chemical is so hydrophobic that it sorbs onto the sur-
faces of plants and containers in a matter of minutes after being
added to the growth solution.
• Finally, it appears that 2,3,7,8-TCDD becomes more strongly ad-
sorbed to soil particles with time. Therefore, studies using soils
spiked in the laboratory with 2,3,7,8-TCDD probably overestimate
the amount of the soil-to-plant partitioning that would occur under
field conditions.
A-43
1598q
-------
In experiments with maize and soybeans, Sacchi et al. (1986) reported
that plants grown in soil accumulated increasing amounts of 2,3,7,8-TCDD
with increases in both time and soil concentration, with uptake ratios of
about 2 percent. In addition, plants grown hydroponically seemed to take
up TCDD via the transpirational stream, based on the observation that
plants kept in the dark and treated with transpiration inhibitors did not
takeup 2,3,7,8-TCDD. Unfortunately, workers at EPA's Corvallis laboratory
have been unable as yet to reproduce the hydroponic portion of this study,
possibly casting some doubt on the soil portion of the study.
Vegetation growing near the ICMESA chemical plant in Seveso, Italy,
where a large quantity of dioxin was accidentally released, has been
analyzed by Cocucci et al. (1978). 2,3,7,8-TCDD concentrations were high-
est in twigs, then cork, then leaves, and lowest in fruit. These data
suggest transpirational flow to, and subsequent elimination from, newly-
formed organs. Other possible explanations for the observed distribution
(e.g., dust deposition on plant surfaces) were not discussed.
Maize and soybeans grown in pots have been analyzed by Facchetti et
al. (1986). Based on the fact that TCDD was volatile at ambient tempera-
tures, Facchetti et al. (1986) theorized that TCDD contamination present
in aboveground portions of plants was the result of volatilization from
contaminated soil.
One early study by Isensee and Jones (1971) includes oats and soybeans
grown in soil and a variety of spiked solutions, including TCDD at
0.06 ppm in a benzene carrier. TCDD uptake from solution reached its
maximum in 24 hr and gradually decreased with time. It was theorized that
movement, volatilization from tissue, and translocation back to roots
could all account for this removal. With regard to uptake from soil,
tissue content decreased as age of the plant increased; this was attrib-
uted to tissue dilution. On potential limitation of this work is that
"Tween," a surfactant, was used to keep the 2,3,7,8-TCDD in solution.
This compound decreases the ability of 2,3,7,8-TCDD to bind to plant
roots and promotes a great deal of bacterial growth, which may also affect
2,3,7,8-TCDD behavior in solution.
Another early study by Helling et al. (1973) concluded that plant
uptake and translocation were "highly unlikely." However, the analytical
detection limits (about 1 ppb) may not have been low enough to detect
uptake.
Another early study by Kearney et al. (1973) concluded that TCDD is
immobile in soils, not readily taken up by plants, persistent in soils,
subject to photodegradation, and slowly degraded in soils to polar
metabplities. Foliar applications were conducted, and there was no trans-
location from leaves. Very little was lost from soybean leaves, but loss
was gradual from oat leaves.
A-44
1598q
-------
Jensen et al. (1983) conducted a study to determine the potential
impacts of 2,4,5-T used on rice crops. No 2,3,7,8-TCDD was found in rice
grain (detection limits were 2-7 ppt), either in treated rice or in rice
purchased in locations throughout the United States.
A.4.3 References
Brady NC. 1974. The nature and properties of soils. New York:
McMillan Publishing Co.
Briggs GG, Bromilow RH, Evans AA. 1982. Relationship between
lipophilicity and root uptake and translocation of non-ionised chemicals
by barley. Pest. Sci. 13:495-504.
Burkhard LP, Kuehl DM. 1986. n-Octanol/water partition coefficient by
reverse-phase liquid chromatography/mass spectrometry for eight tetra-
chlorinated planar molecules. Chemosphere 15:163-167.
Cocucci S, Di Gerolamo F, Verderio A, Cavallero A, Colli G, Gorni A,
Invernizzi G, Luciano L. 1978. Absorption and translocation of
tetrachlorodibenzo-p-dioxin by plants from polluted soil. Experientia
35:482-484.
Facchetti S, Balasso A, Fichtner C, Frare G, Leoni A, Mauri C, Vasconi M.
1986. Studies on the absorption of TCDD by plant species. .In. Rappe C,
Choudhary G, Keith LH (eds). 1986. Chlorinated dioxins and
dibenzofurans in perspective. Chelsea, MI: Lewis Publishers.
Facchetti S, Balasso A. 1986. Studies on the absorption of TCDD by some
plant species. Chemosphere 15:1387-1388.
Freeman RA, Schroy JM. 1985. Environmental mobility of TCDD.
Chemosphere 14:873-876.
Freeman RA, Schroy JM. 1986. Modeling the transport of 2,3,7,8-TCDD and
other low volatility chemicals in soils. Environ. Prog. 5:28-33.
Helling CS, Isensee AR, Wool son EA, Ensor PDJ, Jones GE, Plimmer JR,
Kearney PC. 1973. Chlorodioxins in pesticides, soils, and plants.
Jour. Environ. Qual. 2:171-178.
Isensee AR, Jones GE. 1971. Absorption and translocation of root and
foliage applied 2,4-dlchlorophenol, 2,7-dichlorodibenzo-p-diox1n, and
2,3,7,8-tetrachlorodibenzo-p-dioxin. Jour. Agric. Food Chem.
19:1210-1214.
Iwata Y, Gunther GA, Westlake WE. 1974. Uptake of a PCB (Aroclor 1254)
from soil by carrots under field conditions. Bull. Environ. Contam.
Toxic. 11:523.
A-45
1598q
-------
Iwata I, Gunther FA. 1976. Translocation of the polychlorinated
biphenyl Aroclor 1254 from soil into carrots under field conditions.
Arch. Environ. Contain. Toxicol. 4:44-59.
Jensen DJ, Getzendaner ME, Hummel RA, Turley J. 1983. Residue studies
for (2,4,5-trich1orophenoxy)acetic acid and 2,3,7,8-tetrachlorodibenzo-
p-dioxin in grass and rice. Jour. Agric. Food Chem. 31:118-122.
Kearney PC, Isensee AR, Helling CS, Woodson EA, Plimmer JR. 1973.
Environmental significance of chlorodioxins. Advances in Chemistry
Series #120:105-111.
Kew GA, Schaum JL, White P, Evans TT. 1989. Review of plant uptake of
2,3,7,8-TCDD from soil and pontential influences of bioavailability.
Chemosphere 18:1313-1318.
Hoza P, Schenuert K Klein W, Korte F. 1979. Studies with 2,4'5-
trichlorobiphenyl-14C. and 2,2',4,4',6-pentachlorobiphenyl-iqC in
carrots, sugar beets, and soil. Jour. Agric. Food Chem., 27:1120-1124.
Nash RG, Beall ML, Jr. 1980. Distribution of silvex, 2,4-D, and TCDD
applied to turf in chambers and field plots. Jour. Agric. Food Chem.,
28:614-623.
Sacchi GA, Vigano P, Fortunati G, Cocucci SM. 1986. Accumulation of
2,3,7,8-tetrachlorodibenzo-p-dioxin from soil and nutrient solution by
bean and maize plants. Experientia 42:586-588.
Schroy JM, Hileman FD, Cheng SC. 1985. Physical/chemical properties of
2,3,7,8-TCDD. Chemosphere 14:877-880.
USDA. 1975. U.S. Department of Agriculture. Composition of foods.
Agriculture Handbook No. 8, Washington, DC: U.S. Government Printing
Office.
USEPA. 1988. U.S. Environmental Protection Agency. Estimating
exposures to 2,3,7,8-TCDD (external review draft). Washington, DC:
Office of Health and Environmental Assessment. EPA/600/6-88/005A.
Wipf HK, Homberger E, Neuner N, Ranalder UB, Vetter W, Vuilleumier JP.
1982. TCDD levels in soil and plant samples from the Seveso area. In
Huntiziger 0, Frei RW, Merian E, Pocchiari F (eds). 1982. Chlorinated
dioxins and related compounds: Impact on the environment. New York:
Pergamon Press.
Wipf HF, Schmid J. 1983. Seveso--an environmental assessment, in
Tucker RE, Young Al, Gray AP (eds). 1983. Human and environmental risks
of chlorinated dioxins and related compounds. New York: Plenum Press.
A-46
1598q
-------
APPENDIX B
Summary Information on Chlorinated Chemicals
Other than PCDDs and PCDFs (OCOs)
Identified in Pulp Mill Effluents, Sludges, and Pulps
-------
TABLE OF CONTENTS
Paoe No.
B.I Introduction B-l
B.2 Master List of Other Chlorinated Organic Chemicals (OCOs)
Identified in Pulp Mill Effluent B-l
B.3 Screening-Level Human Health Data B-2
B.4 Screening-Level Environmental Hazards from Chlorinated
Organic Chemicals Other than PCDDs and PCDFs Released by
Pulp/Paper Mills B-3
B.4.1 Hazards to Aquatic and Terrestrial Organisms Posed by
Other Chlorinated Organic Chemicals (OCOs) B-4
B.4.2 Risks Posed by the Release of Effluents Containing
Other Chlorinted Organic Chemicals (OCOs) B-4
B.4.3 Conclusions B-4
6.5 Screening Level Human and Aquatic Exposure to Other
Chlorinated Organic Chemicals (OCOs) in Treated Pulp/Paper
Mill Effluent B-5
B.6 Temporal Data - Ten-Month, Bi-Daily QCO Concentrations in
Pulp Mill Effluent B-6
B.7 Activities and Regulatory Approaches of Various
Governmental Agencies Regarding Other Chlorinated Organic
Chemicals Released into the Environment from Pulp/Paper
Mills B-6
B.7.1 Summary of the Office of Water Regulations and
Standards' (OWRS) Regulatory Approach B-7
B.7.2 Summary of the Office of Air Quality Planning and
Standards (OAQPS) Activities Concerning Emissions
of Chloroform from Pulp/Paper Mills B-8
B.7.3 Summary of the Office of Toxic Substances' Toxic
Release Inventory (TRI) Data Regarding OCO Emissions
from Pulp/Paper Mills B-8
B.7.4 Summary of European and Canadian Regulatory
Approaches B-9
B.8 Conclusions and Recommendations B-10
B.9 References B-ll
-------
LIST OF TABLES
Table B-l.
Table B-2.
Table B-3.
Table B-4.
Table B-5.
Table B-6.
Table B-7.
Table B-8.
Table B-9.
Table B-10.
Table B-ll.
Page No.
Master List of the Chlorinated Compounds Other Than
2,3,7,8-TCDD/F Found in Pulp Mill Effluents B-14
Other Chlorinated Organic Chemicals Cancer Risk — B-20
Other Chlorinated Organic Chemicals Drinking Water
Health Advisories B-21
Other Chlorinated Organic Chemicals Reference
Doses B-22
Species List B-23
Toxicity of Other Chlorinated Organics from Bleached
Wood Products to Aquatic Species B-24
Ranking by Acute Toxicity to Aquatic Organisms B-31
Acute Toxicity of Selected Chlorinated Organic
Chemicals (OCO) to Terrestrial Organisms B-32
Summary of Treated Effluent Chemical
Concentrations B-33
Statistical Analysis of Temporal Changes in Chemical
Concentrations in Pulp Mill Effluent B-35
Regulatory Schedule for OCOs in Some Canadian
Provinces and European Countries B-36
-------
LIST OF FIGURES
Page No.
Figure B-l. 10th% Low Flow Stream Cone vs. Eff Cone B-37
Figure B-2. 50th% Low Flow Stream Cone vs Eff. Cone B-37
Figure B-3. Average LADDs at 10th % Mean Flow B-38
Figure B-4. Average LADDs at 50th % Mean Flow B-38
Figure B-5. Temporal Changes in 3,4,5-Trichlorocatechol Cone. . B-39
Figure B-6. Temporal Changes in Pentachlorophenol Cone B-40
-------
APPENDIX B
B.I Introduction
Risks from exposure to 2,3,7,8-TCDD and 2,3,7,8-TCDF have been the
primary focus of this Integrated Exposure/Risk Assessment. Prior to the
discovery of PCDDs and PCDFs in the effluents and products of the pulp
and paper industry, other chlorinated organic compounds (OCOs), such as
chloroform and chlorinated phenols, had already been identified (Suntio
et al. 1988; Sodergren et al. 1988; USEPA 1988; USEPA 1987). Even though
these chemicals may be present individually at relatively low
concentrations in pulp mill effluents, sludges, and pulps, the cumulative
effect of PCDDs, PCDFs, and the OCOs on exposed populations (human,
aquatic life, and wildlife) could be significant. In order to enable
informed risk management decisions with regard to the pulp and paper
industry, it is important to understand at least on a qualitative basis,
the identity, quantity, impacts, and risks of other chlorinated organic
chemicals present in pulp/paper mill wastes and products.
The purpose of this appendix is to present screening level analysis of
OCOs using monitoring data obtained from industry and the literature and,
also, to describe various governmental action regarding OCOs released by
the pulp and paper industry. The screening level analysis includes the
compilation of readily available information on the properties and toxici-
ty of the individual chemicals and generic exposure estimates for those
chemicals where treated effluent data are available.
This appendix is structured to first provide in section B-2, a master
list of the other chlorinated organic chemicals (Table B.I) that have been
found in pulp mill wastes/products. Section B-3 presents screening level
human health data and section B-4 contains a discussion and presentation
of the readily available data on the aquatic and terrestrial toxicity of
the OCOs. Screening level estimates of stream concentrations and drinking
water exposure are given in section B-5. The graphs presented in
Section B.6 show how certain chemical concentrations varied during a ten
month study of treated pulp mill effluent. Section B.7 presents a summary
of governmental activities on OCOs released by the pulp and paper indus-
try. Lastly, conclusions and recommendations concerning OCOs and the pulp
and paper mill industry are made.
B.2 Master List of Other Chlorinated Organic (OCO) Chemicals
Identified In Pulp M111 Effluent
Table B-l is a master list of chlorinated compounds other than PCDDs/
PCDFs which have been found in pulp mill wastewater. No information was
readily available on OCOs in pulp or wastewater sludge.
B-l
1596q
-------
Of the 263 chemicals/chemical families on the master list, treated
effluent concentration data are available for 31, human health effect in-
formation is readily available for 23, aquatic toxicity data are readily
available for 45, and terrestrial toxicity data are readily available for
10. The last column, "Data Code," lists the readily available information
for each chemical. The master list is partitioned into acidic, phenolic,
and neutral categories for which there were 64, 63, and 130 listings,
respectively. There are six entries of general chemical families (i.e.,
chlorinated phenols, etc.) at the end of the neutrals category.
B.3 Screening Level Human Health Data
A CAS number search of the IRIS data base in May 1989 showed that
there are human health data for 23 of the chemicals on the master list.
The IRIS data base lists all chemicals that EPA has reviewed and given an
official agency evaluated reference dose, cancer slope factor, or other
toxicity evaluation. The 23 chemicals with health data listed in IRIS
are identified in Table B-l by an "H" in the "Data Code" column.
Table B-2 shows the cancer risk associated with these chemicals. The
definitions for the various cancer classification codes can be found in
the Guidelines for Carcinogen Risk Assessment published by EPA. The slope
factor, QI*, is taken as the 95% upper-bound of the potency of the chem-
ical in inducing cancer at low doses, where the upper confidence limit for
the extra risk calculated at low doses is always linear. The unit risk
for an air or water pollutant is defined as the lifetime cancer risk
occurring in a hypothetical population in which all individuals are con-
tinuously exposed from birth throughout their lifetimes (70 years) to a
concentration of 1 ng/m* of the agent in the air they breath, or to
1 /*g/l in the water they drink. The unit risk calculation assumes the
individual in the hypothetical population weighs 70 kg, drinks 2 liters
of water per day, and breaths 20 cubic meters of air per day.
Table B-3 lists values for drinking water standards and health adviso-
ries. Maximum contamination limits (MCLs) are enforceable standards that
must be met in finished drinking water supplies. Several of the values
in this table are only proposed values and are not considered as final
rulings. Water quality criteria for the protection of human health can
be used by the states to develop waste load allocations for surface water
dischargers. There are also ambient water quality criteria for the
protection of aquatic life (not listed here). The water quality criteria
values given here are revised values that will not be published until May
1990. These values are still undergoing review and they may change when
finally published.
The drinking water health advisories represent guidance levels for
drinking water and were developed from data describing noncarcinogenic
toxicity endpoints. The values for the one-day, ten-day, and longer-term
B-2
1596q
-------
exposure periods do not consider other exposures such as food and air,
and therefore assume that 100 percent of the individual exposure comes
from drinking water. The lifetime value is calculated only for adults
but does consider other sources of exposures. The Drinking Water
Equivalent Level (DWEL) adjusts the oral RfD for a 70 kg adult who drinks
2 liters of water a day. The DWEL assumes that 100 percent of an
individual's exposure to that contaminant comes from their drinking water.
Table B-4 lists the available Reference Doses (RfD). Reference doses
are based on noncarcinogenic toxicity endpoints. The RfD is a peer-
reviewed estimate of the time-weighted average daily exposure to the human
population that is likely to be without appreciable risk of deleterious
results during a lifetime. The Uncertainty Factor (UF) consists of divis-
ors of multiples of 10 and is assigned according to guidelines. Each
uncertainty factor represents a specific area of uncertainty inherent in
the available data; for example, these factors take into account the
differences in responsiveness between animals and humans, and between
individuals in the exposed population.
B.4 Screening Level Environmental Hazards from Chlorinated Organic
Chemicals Other than PCDDs and PCDFs Released by Pulp/Paper
Mills
(NOTE: This subsection was extracted from a memorandum from Robert E.
Morcock, Health and Environmental Review Division, U.S. Environmental
Protection, to Thomas M. Murray, Exposure Evaluation Division, U.S. Envi-
ronmental Protection Agency, dated September 29, 1989.)
The ACQUIRE data base (USEPA 1984) (an aquatic information retrieval
system) and the RTECS (Registry of Toxic Effects of Chemical Substances)
were used in developing information on hazards posed by OCO chemicals.
In general, data were based on studies conducted in the United States.
However, two pentachlorophenol studies incorporated results from foreign
studies on amphibians. Of the U.S. studies, only records with
reliability codes of -"1" ("meets all criteria") or "2" ("meets some
criteria") were examined. Studies with rating "1" have followed all
established laboratory procedures. A rating of "2" represents studies
that are generally satisfactory, but one variable may not have been
reported or measured (e.g., water chemistry variables are not reported or
toxicant concentration was unmeasured). The codes were established by
the EPA Environmental Research Laboratory in Duluth. In addition, the
potential impact of nine arbitrarily selected chemicals on terrestrial
organisms was examined.
All species were considered to be relevant indicators and the lowest
concentration at which effects were reported was used. Ranking by toxici-
ty was based on acute and/or chronic studies on thirty-two species.
Table B-5 lists species reported in all of the available studies.
B-3
1596q
-------
B.4.1 Hazards to Aquatic and Terrestrial Organisms Posed by Other
Chlorinated Organic Chemicals (OCOs)
Table B-6 gives a summary of available data on the toxicity of other
chlorinated organic chemicals to aquatic organisms. The chemicals listed
in Table B-6 are limited to those with LC™ data in ACQUIRE (USEPA
1984). Of the chemicals examined, two exhibit little toxicity:
trichloroacetic acid and tetrachloro-acetic acid (2,000 mg/1) (Tadken and
Lewis 1983). Phenols were the most toxic (LCcg); pentachlorophenol,
0.023 mg/L and tetrachlorophenol , 0.14 to 0.17 mg/L. These data are
summarized and ranked by toxicity in Table B-7.
Because of time constraints, the potential impacts of the chemicals
on terrestrial organisms were only briefly evaluated. Like heavy metals,
chlorinated organic chemicals are persistent in the soil. Uptake of
chlorinated organic contaminants by plant roots may be an important bio-
logical exposure pathway and must be considered when evaluating risks from
these compounds.
Table B-8 summarizes terrestrial toxicity data of ten selected chemi-
cals.
B.4.2 Risks Posed by the Release of Effluents Containing Other
Chlorinated Organic Chemicals (OCOs)
At present, available data regarding composition of complex effluents
are insufficient in quantity and quality to enable their use in the devel-
opment of quantitative risk assessments. In addition, synergistic effects
have not been assessed. Finally, it is not practical (within the con-
straints of this project) to develop risk assessments for all of the chem-
icals covered in this appendix. Nevertheless, it is expected that there
will be unacceptable risks to aquatic and terrestrial organisms where
environmental concentrations approach toxic levels reported in this
subsection. Table B-9 compares the available data on treated-ef fluent
chemical concentration to the LCcn. The upper range of pentachloro-
phenol and chloroform treated -effluent chemical concentrations exceed the
LC5Q when fate is not considered.
As indicated in Table B-7, twenty-five chemicals exhibited
values below 10 mg/L, eight between 10 and 100 mg/L, and five between
220 mg/L and 2,000 mg/L.
B.4.3 Conclusions
Information on environmental concentration and potential exposure is
currently too sparse to make quantitative risk assessments, although
potential impacts on both aquatic and terrestrial species conceivably may
occur. More than 10 percent of the chemicals on the original list of 263
B-4
1596q
-------
and 65 percent of the chemicals for which data were available, would have
adverse effects if environmental concentrations were as low as 10 mg/L.
B.5 Screening Level Human and Aquatic Exposure to Other Chlorinated
Organic Compounds (OCOs) in Treated Pulp and Paper Hill
Effluents
In order to provide a qualitative sense of exposures to chemicals for
which treated effluent data are available, a generic exposure scenario
was developed for the pulp and paper mill industry. This scenario
consists of treated effluent being discharged into a "generic" river thus
exposing aquatic life directly and exposing humans through ingestion of
contaminated drinking water. The generic approach has given an
"order-of-magnitude" impression of potential exposures, which are
depicted graphically in Figures B-l through B-4. These figures can be
used along with the data on concentration ranges of the treated effluent
presented in Table B-9 to estimate stream concentrations of chemicals of
interest under low and mean flow conditions. It is important to note
that fate and transport processes (i.e., volatilization, biodegradation,
adsorption, etc.) are not considered in this analysis. Therefore, this
exercise provides only a screening level estimate. Also, it should be
noted that the data provided in Table B-9 is dated and may or may not
reflect actual effluent concentrations for each of the chemicals listed.
Generic exposure assessments are developed from stream flow and
effluent flow data from numerous facilities within the same Standard
Industrial Classification (SIC) code. For the purpose of this
assessment, stream and effluent flows for 63 pulp and paper mills within
SIC code 2611 were obtained from the GAGE and IFD files in EPA's STORET
system. These data were used to generate a statistical distribution of
the stream dilution factors under mean and low (7-Q-10) flow conditions.
The 50th percentile and 10th percentile mean flow dilution factors were
358 and 3, respectively. Stream concentrations were then estimated by
dividing the treated effluent concentration by the dilution factor for
the flow condition of interest. The 10th and 50th percentile dilution
factors were used to estimate exposure. The 10th percentile stream flow
dilution factor was used to represent a reasonable worst case and the
50th percentile stream flow dilution factor represents the median
dilution factor.
Low stream flow (i.e., 7-Q-10) dilution factors were used to estimate
potential aquatic toxicity impacts (see Figures B-l and B-2). Low flow
stream data are typically used to calculate acute toxicity impacts to
aquatic life since the exposure to the fish will be the highest. Mean
stream flow dilution factors were used to estimate lifetime average daily
exposure (LADE) to humans from drinking contaminated water. Again, 10th
and 50th percentile stream flow dilution factors represent worst case and
median exposures (see Figures B-3 and B-4). Estimates of drinking water
B-5
1596q
-------
exposure are based on the assumption that a person consumes two liters of
contaminated drinking water 365 days per year for 70 years.
Figures B-l through B-4 were based on the highest treated effluent
data for all chemicals (excluding chloroform, which was found at much
higher levels than all other compounds and is expected to volatilize
rapidly from water to air). These figures for aquatic and human drinking
water exposure values are generic interpretations (e.g., non-chemical
specific) of the data, since any chemical put into a stream will be
equally diluted because the stream flow will be the same. Due to time
limitations, fate and environmental transport properties for each of the
chemicals listed in Table B-9 (other than dilution) could not be
considered. Processes such as volatilization, sorption to sediment,
solubility of the chemical in water, as well as others could considerably
influence the concentrations of OCOs in surface water, sediment, and air.
B.6 Temporal Data - Ten-Month, B1-Dally OCO Concentrations in Pulp
Mill Effluent
In one study supplied to EPA by NCASI (NCASI 1985a), concentration
data for 18 different chemicals were collected every other day for a
10-month period. Data for two of these chemicals are plotted in Figures
B-5 and B-6 for illustration. Table B-10 shows the statistical analysis
for the data. These data show the minimum and maximum concentrations
vary by a factor of 2 to a factor of 60.
B.7 Activities and Regulatory Approaches of Various Governmental
Agencies Regarding Other Chlorinated Organic Chemicals Released
Into the Environment from Pulp/Paper Hills
The pulp and paper industry has been under investigation by many
government agencies both domestic and foreign in regards to their
production and release of OCOs. Within the U.S. EPA, the Office of Water
Regulations and Standards (OURS) is conducting a thorough investigation
of the pulp and paper industry to support the revision of the effluent
limitations guidelines and standards for the pulp, paper, and paperboard
industrial point source category, the Office of Air Quality Planning and
Standards (OAQPS) has Identified the pulp and paper Industry as the
largest emitter of chloroform, and the Office of Solid Waste is
considering listing paper mill sludge as a hazardous waste based on the
presence of dioxins.
Canadian officials have been studying the release of OCOs for some
time and much of the monitoring data presented in this appendix was
obtained from scientific papers published by Canadian researchers.
Overseas, West Germany, Finland, and Sweden have regulated or announced
Intentions to regulate the pulp and paper industry on the basis of TOX
(Total Organic Halogens) or AOX (Adsorbable Organic Halogens) per ton of
air dried bleached pulp produced.
B-6
1596q
-------
B.7.1 Summary of the Office of Water Regulations and Standards'
(OURS) Regulatory Activities
On January 2, 1990, the Office of Water Regulations and Standards
issued a 304(m) notice announcing that the existing effluent limitations
guidelines and standards for the pulp, paper, and paperboard industrial
point source category will be revised. OWRS has established a Pulp and
Paper Guidelines Review Workgroup and has devised a Development Plan and
Schedule. The Development Plan includes the generation of data to support
the revision of the effluent limitations and to investigate pollution
prevention technology options to reduce the water discharges of conven-
tional, toxic, and nonconventional pollutants. These pollutant categories
include the OCOs as defined in this appendix. Therefore, the water re-
lease of OCOs by pulp and paper mill effluent will be regulated according
to the new standards.
OWRS believes reduction in toxic and nonconventional pollutant loads
from pulp and paper facilities may be best accomplished through alterna-
tive process technology rather than wastewater treatment. Furthermore,
OWRS believes that process changes (e.g., use of chlorine dioxide and
oxygenation instead of chlorine and hypochlorite to bleach the pulp and
paper) should reduce air emissions, lower levels of toxics in solid waste,
and reduce consumer exposure to pollutants in pulp and paper products.
It is anticipated that through short and long term field sampling, comple-
tion of a Pulp, Paper, and Paperboard Industry Survey questionnaire, and
an analysis of in-plant process changes, options will be Identified to
reduce the formation of dioxin and other pollutants of concern in pulp and
paper manufacturing operations.
During rulemaking development, OWRS anticipates that data for more
than 200 chemicals will be collected. Thus far, OCOs are the main pollut-
ants identified for regulation. In addition to or instead of developing
chemical-specific limitations and standards, OWRS is exploring the use of
an alternate control parameter, such as TOX (Total Organic Halogens) or
AOX (Adsorbable Organic Halogens). Some of the regulatory options
approaches OWRS may choose to examine include: (1) limitations based on
modifications to the process, (2) numerical limitations for a limited list
of toxic pollutants (i.e., TCDD/TCDF, volatiles, OCOs, etc.), (3) regula-
tion of toxics through the use of numerical limitations for surrogate
parameters (i.e., TSS (Total Suspended Solids), TOX, and (4) whole efflu-
ent toxicity limits. OWRS anticipates the options selection for BPT (Best
Practical Control Technology Currently Available) will be proposed by
July 1993 (Red Border review completed by March 1993). The BAT (Best
Available Technology Economically Achievable)/NSPS (New Source Performance
Standards)/PSES (Pretreatment Standards for Existing Sources)/PSNS (Pre-
treatment Standards for New Sources) will be developed for the bleached
kraft and sulfite mills only. Red border review will be complete by May
1995 and BAT/NSPS/PSES/PSNS will be promulgated by June 30, 1995.
B-7
1596q
-------
B.7.2 Summary of the Office of Air Quality Planning and Standards
(OAQPS) Activities Concerning Emissions of Chloroform from Pulp/
Paper Hills
In 1985, a chloroform exposure and risk assessment was performed by
OAQPS as part of a process initiated in 1984 to determine if chloroform
should be listed as a Hazardous Air Pollutant under Section 112 of the
Clean Air Act (USEPA 1985). Of the 11 major source categories for
chloroform releases assessed in the report, the pulp and paper industry
was identified as the largest category source of chloroform emissions.
It was estimated to account for about 25 percent of the approximate
14,700,000 metric tons of chloroform emitted annually nationwide from all
sources. The second highest emitter was estimated to be drinking water
treatment facilities.
The maximum individual cancer risk estimated for the pulp and paper
industry by this assessment was 1.1 x 10"z and the annual aggregate
cancer incidence was reported to be 0.76 cases/year. As the result of
this assessment, a Notice of Intent to List chloroform under section 112
of the Clean Air Act was published in 1986. The pulp and paper industry
raised concerns in 1986 regarding the modeling of chloroform emissions
from wastewater ponds (lagoons) used in the assessment.
Since this initial assessment was performed, additional data have been
obtained and a new model has been developed for estimating chloroform
emissions from lagoons. OAQPS requested that pulp and paper plants submit
new emissions data including data on process vent emissions of chloroform
which had not been modelled previously. These data are currently being
analyzed.
The new Clean Air Act is anticipated to be signed into law by Congress
during 1990. Under the new law, OAQPS must promulgate emission standards
for ten source categories within two years of enactment. Of all source
categories for which emissions standards must be promulgated, the worst
25 percent must be done within four years (Cinalli 1989). The pulp and
paper industry is highly likely to be included in the top 25 percent.
Therefore, chloroform emissions from this industry are likely to be
regulated within two to four years. During the process of regulating
source categories, co-pollutants on the "Hazardous Air Pollutant List"
must be considered as well; therefore, the emissions of other chlorinated
compounds on the List may be regulated along with chloroform.
B.7.3. Summary of the Office of Toxic Substances Toxic Release
Inventory (TRI) Data Regarding OCO Emissions from Pulp/Paper
Mills
Under section 313 of the Superfund Amendments and Reauthorization Act
(SARA) of 1986, all facilities must report releases of certain toxic chem-
icals to the EPA, Office of Toxic Substances. A 11st of approximately
B-8
lS96q
-------
300 toxic chemicals is defined under SARA Title III. Of the chemicals
identified in Section B.2 of this appendix, 20 chemicals are on the SARA
Title III list. A search of the Toxic Release Inventory (TRI) database
was performed in November, 1989 for these 20 chemicals being released or
transferred from facilities in SIC codes 2611 (pulp mills), 2621 (paper
mills), and 2631 (paperboard mills). Only two chemicals, chloroform and
1,1,1-trichloroethane, were reported to be released from these facilities.
Chloroform releases to air, water, and land from pulp mills were
prevalent and, in some cases, considerable (especially releases to air).
Sixty-eight of the 76 facilities reporting chloroform releases are in the
104 Mill Study. The largest reported releases in 1987 to air, land and
water according to the TRI database are as follows:
Largest single air release: 1,700,000 pounds per year
Releasing facility: Westvaco-Bleached Board (Covington, VA)
Largest single water release: 140,000 pounds per year
Releasing facility: Simpson Paper (Eureka, CA)
Largest single land release: 31,337 pounds per year
Releasing facility: Stone Container Corp. (Snowflake, Az)
Many other facilities reported large releases to all media; however,
the largest estimated releases across the industry were to air. Consider-
ing the volatility of chloroform, the reported water releases will also
ultimately result in air releases.
Additionally, two facilities from the 104 mill study reported releases
to air of 1,1,1-trichloroethane on their 1987 Toxic Release Report. In-
ternational Paper of Mobile, Alabama, reported releasing 16,000 pounds and
the Mead Corporation in Chillicothe, Ohio reported releasing 499 pounds.
No other OCOs were reported released by the 104 mills. This is not
surprising since the reportable quantity was 50,000 pounds per year for
1987. Also the SARA section 313 list contains less than 8 percent of the
OCO chemicals identified in Table B.I.
8.7.4 Summary of European and Canadian Regulatory Approaches
Other countries have regulated or announced intentions to regulate the
pulp and paper industry on the basis of TOCL (Total Organic Chlorine) or
AOX (Adsorbable Organic Halogens) per ton of air dried bleached pulp pro-
duced. Table B-ll lists regulatory schedules for some Canadian provinces
and European countries (Clodey 1989).
Due to the complexities of the composition of bleached mill wastes,
many other countries have adopted generic measures to determine chlori-
B-9
1596q
-------
nated organic compounds in various media. The following parameters have
been suggested for water and effluents: ATOCL (Total Organically Bound
Chlorine), AOX, and TOX (Total Organic Halogens). For sediments and
tissues, the following parameters have been suggested: EOX (Extractable
Organic Halogen), EOCL (Extractable Organic Chlorine), and EPOCL (Extract-
able Persistent Organic Chlorine). Use of the AOX method of analysis is
becoming more widespread due to its ease of use, high precision, and low
interference by inorganic chloride and ambient humic substances.
Pre-regulatory monitoring conducted by the Ontario Ministry of the
Environment detected approximately 200 organic compounds in various waste-
streams at a variety of pulp and paper mill (Ontario Ministry of the
Environment 1989). Three Canadian provinces are establishing regulations
for allowable AOX levels between 1.5 to 2.5 kg/ton air-dried pulp (see
Table B-ll).
The Swedish Pulp and Paper Research Institute (STFI) has positively
identified 315 individual compounds in wastestreams from pulp bleaching
operations and in whole mill effluent (Sodergren et al. 1988). Adverse
biological effects of pulp and paper mill wastes on several Baltic Sea
species has been documented by the Swedish Environmental Protection
Board's Environment Cellulose Project. Based on these and other studies,
the Swedish government has taken a protective approach and has limited
the amount of OCOs released by regulating AOX to the lowest levels
technologically possible.
B.8 Conclusions and Recommendations
Although there is a copious amount of information on chlorinated
organic chemicals present in pulp mill effluent, the available information
is not adequate to permit accurate assessments of human or environmental
risk to the other chlorinated organics. In addition, no data are avail-
able on OCO concentrations in either sludge or pulp; thus, no attempt is
made in this analysis to assess potential risks to humans or terrestrial
organisms from exposure to sludge or pulp/paper. The estimates presented
in this analysis are screening-level only and should only be used as a
qualitative screening tool.
Some general observations can be made, however, on the OCO data.
First, there are species of chemicals that show significant toxicity at
low levels to aquatic life. The most toxic species are the chlorinated
phenols. These compounds are typically found in concentrations from 4 to
15 ^g/1 in the treated effluents. At low stream flow conditions,
this may result in concentrations that are within an order of magnitude
of LCcn levels. Second, in nearly all cases, insufficient site-
specific data exist to identify plants that may be discharging OCO
chemicals at high levels. In addition, the available data are somewhat
dated and therefore may not necessarily reflect changes in effluent
B-10
1596q
-------
concentrations due to recent changes 1n bleaching processes. Third,
concentrations of OCO chemicals in treated effluents are generally much
higher than concentrations of PCDD and PCDF. OCO concentrations in
treated effluents are typically two to three orders of magnitude higher
than PCDD and PCDF concentrations. Fourth, it should be noted that paper
mill effluents are complex mixtures. They contain not only PCDD and PCDF
but also a wide range of organic compounds. This wide array of compounds
may interact, thus further complicating the analysis of paper mill
effluents. Furthermore, the effluent concentration of chloroform
reported on Table B-9 exceeds the water criteria for human water and fish
consumption as listed on Table B-3. Finally, as illustrated on
Table B-9, pentachlorophenol and chloroform exceed the LC$Q', therefore,
additional sampling and analysis may be warranted on OCO compounds in
paper mill effluents. This is warranted since, at this screening level,
qualitative analysis indicates that potential risks from OCO chemicals
may be significant.
B.9 References
Cinalli.C. 1989. Personal communication between C. Cinalli, EPA/OTS,
and J. Vandenberg, EPA/OAQPS. 12/27/89.
Clodey, AG. 1989. Environmental Impact of Bleached Pulp and Paper Mill
Effluents in Sweden, Finland, and Norway: Implications to the Canadian
Environment. Discussion Paper. Unpublished report of the Industrial
Programs Branch, Environmental Programs Directorate/Conservation and
Protection, Environment Canada.
NCASI. 1976. National Council of the Paper Industry for Air and Stream
Improvement. Proceedings of the 1976 NCASI West Coast Regional Meeting.
Technical Bulletin No. 77-10.
NCASI. 1977. National Council of the Paper Industry for Air and Stream
Improvement. Analyses of volatile halogenated organic compounds in
bleached pulp mill effluent. Technical Bulletin No. 198; August 1977.
NCASI. 1980. National Council of the Paper Industry for Air and Stream
Improvement. Chlorinated organic in bleach plant effluents of pulp and
paper mills. Technical Bulletin No. 332; May 1980.
NCASI.. 1981. National Council of the Paper Industry for Air and Stream
Improvement. Experience with the analysis of pulp mill effluents for
chlorinated phenols using an acetic anhydride derivation procedure.
Technical Bulletin No. 347; June 1981.
NCASI. 1982a. National Council of the Paper Industry for Air and Stream
Improvement. Effects of biologically stabilized bleached kraft mill
effluent on cold water stream productivity as determined in experimental
streams. Technical Bulletin No. 368; April 1982.
B-ll
1596q
-------
NCASI. 1982b. National Council of the Paper Industry for Air and Stream
Improvement. Supplemental data reflective of available technological
capability for separation of chlorinated organics from pulp and paper
industry wastewaters. Technical Bulletin No. 82-01.
NCASI. 1983a. National Council of the Paper Industry for Air and Stream
Improvement. A comparison of results from the analysis of pulp mill
effluents for compounds of pulp mill and bleach plant origin. Technical
Bulletin No. 397; April 1983.
NCASI. 1983b. National Council of the Paper Industry for Air and Stream
Improvement. A study of methods for reducing chloroform concentrations
in bleached pulp mill effluents. Technical Bulletin No. 399; May 1983.
NCASI. 1983c. National Council of the Paper Industry for Air and Stream
Improvement. Effects of biologically stabilized bleached kraft effluent
on warm water stream productivity in experimental streams - Third
progress report. Technical Bulletin No. 414; December 1983.
NCASI. 1984a. National Council of the Paper Industry for Air and Stream
Improvement. Observations of the condition of organs and tissues of fish
exposed to biologically treated bleached kraft mill effluent. Technical
Bulletin No. 419; Janury 1984.
NCASI. 1984b. National Council of the Paper Industry for Air and Stream
Improvement. Effects of biologically treated bleached kraft mill
effluent on cold water stream productivity in experimental stream
channels - Third progress report. Technical Bulletin No. 445; October
1984.
NCASI. 1985a. National Council of the Paper Industry for Air and Stream
Improvement. Effects of biologically treated bleached kraft mill
effluent on cold water stream productivity in experimental stream
channels - Fourth progress report. Technical Bulletin No. 474; November
1985.
NCASI. 1985b. National Council of the Paper Industry for Air and Stream
Improvement. Effects of biologically treated bleached kraft mill
effluent during early life stage and full life cycle studies with fish.
Technical Bulletin No. 475; December 1985.
NCASI. 1986. National Council of the Paper Industry for Air and Stream
Improvement. Procedures for the analysis of resin and fatty adds in
pulp mill effluents. Technical Bulletin No. 501; August 1986.
Ontario Ministry of the Environment. 1989. Development Document for the
Draft Effluent Monitoring Regulation for the Pulp and Paper Sector.
B-12
1596q
-------
Sodergren A, Bengtsson, BE et. al. 1988. Summary of Results from the
Swedish Project Environment/Cellulose. Water Science Tech. Vol. 20,
No. 1.
Suntio, LR, Shui WY, and Mackay D. 1988. A Review of the Nature and
Properties of Chemicals Present in Pulp Mill Effluents. Chemosphere
17(7):1249-1290.
Tatken RL, Lewis RJ, ed. 1983. Registry of toxic effects of chemical
substances. 1981-82 edition. U.S. Department of Health and Human
Services.
USEPA. 1984. ACQUIRE. An aquatic information retrieval system.
USEPA. 1985. Memorandum from Timothy 0. Mohin, OAQPS/PAB, to the files
dated 8/16/85. Subject: Chloroform exposure and risk assessment. RTP,
NC.
USEPA. 1987. The National Dioxin Study, Tiers 3, 5, 6 and 7. Office of
Water Regulations and Standards, U.S. Environmental Protection Agency,
Washington, D.C. EPA 440/4-87-003.
USEPA. 1988. U.S. EPA/Paper Industry Cooperative Dioxin Screening Study.
Office of Water Regulations and Standards, U.S. Environmental Protection
Agency, Washington, DC. EPA 440/1-88-025.
B-13
1596q
-------
Uaatar Lilt of tha Chlorinated
Compound* Othar Than 2,3,7,8-TCOOff'
Found in Pulp Uill Bfluantx
CHEMICAL NAME
_.«—
Chloraaoatio Bold
Diohbreaoatio moid
TriohkxtMMtio moid
Diehloropropanoie acid
2,3-Diohloropropanoig void
CAS NO.
79-11-8
L_ 79-43-6
7S-03-8
13167-36-7
3,3-Diohloroprop»noipaeld ! 1961-20-2
•* 4 U 1 t . 1
Triohtorobutanoio »oid
2,4,4-Triehlorebuteneie aoid
3,4,4-Triohlorobutanore aotd
4,4,4-Trlahlorobutafloio aoid
3,4,4-TrioMoro-3-butanoio moid
4,4,4-Triohlere-2-lMJtanaia aoid
Tctraohlorobutanoio aoid
3,4,4,4-T*trmohloro-2-but»noio laid
2,3,4 ,4-T«tr»ohloTo-2-but»fioio aoid
2,3,4 ,4-T«trmohtofo-3-but»nolo moid
Diohbreataarie «oid
9,10-diohtorattowio aoid
ChloroiMlonio «oid
Chlorahi mario add
ChloroimMo «eid
Diehlorematoie moid
3-Chloramuoonie »oid
Chbro-3,4-dJhydnwytMn»ie moid
2-Chlofo-3,4-dihydnntyt»oa>io «oid
S-Chloro-3,4-dlhydraKybwiaoiQ «oid
S-Chloro-3,4-dihydroxyt.«nJo.o *otd
Diohhxo-3,4-dihydroxyt»nKM«i aoid (3 imrwri)
2,5-Diehk>ra-3,4-dihydreMytoncoie *«id
Triehloro-3,4-dihydrexyfaM»o(e aoid
Diohlwe-4-hydraKybwiwle Bold
2,8-Diohloro-4-hydroxyb«nioio Bold
2,3-Diohlore-4-hydraBcyfa*n»io Bold
3.6-DiehkMO-4-hydraKyfawiiato Boid
ChloravBnillio B«d
aohloroy«iailo«iMd
2.3-OietiloroMnUno Mid
Tr!ohtefO*mnlll!fl aeid
Diohtorofyringic moid
ChtoroiMttwHypratoMtMhuio aeid
Dilora-2-thiophwiie acid
3-Chbra-2-thioph«f)ie teid
4-Chkwo-2-thioph«iio maid
5-Chloro-2-thioph*nie teid
Di«hloro-2-thioph«nio moid
2257-35-*
10313-08-5
21422-57-1
22230-93-0
21422-57-1
2230MB3*>0
96091-03-7
88032-93-*
76144-02-C
2892-54-6
31135-63-4
5629-46-1
600-33-9
617-42-6
617-43-6
606-42-4
20668-88-6
87832-50-1
87932-40-8
87032-81-2
6964B-41-4
108096-45-9
9916B-46-5
35456-33-4
66884-08-*
3336-41-2
62936-23-6
70186-92-4
106544-87-4
9326-00-0
20624-96-4
59337-W-2
596i4-ph«nio moid
3,5-Diohloro-2-lhk>phwitB moid
61209-02-7 .
69166-94-9
CtwmtaBJ
Format*
C2H302CL
C2H2O2CL2
C2H02CL3
C3H2O2CL2
C3H02CL3
C4H3CL302
C4H2CL402
C16H3402CL2
C3H3O4a
C4H3O4CL
C4H3O4CL
C4H204CL2
C6H8O4CL
C7H5O4CL
C7H404CL2
C7H304CL3
C7H4O3CL2
08H7O4a
C8H604CL2
C9H604CU3
C8H6O8CL2
C8H706O.
C5H302O.S
CSH202CL2S
(C*CLf
na.N
3
4
5
(C)D
no. C
2
2
2
5 3
6
7
6
20
4
5
5
6
7
6
9
10
9
e
10
11
11
9
6
7
3
4
4
16
3
4
4
4
6
7
7
7
7
8
8
8
9
8
5
5
I
M.W>
94.5
128.95
163.4
140.95
175.4
189.42
223.67
35339
136.51
150.62
150.52
184.96
176.56
188.57
223.01
223.02
207.02
202.50
237.04
271.48
267.06
Oato
Coda
P
E,P _
E,P __
Pm
Pm
Pm
Pm
Pm __
Pm
Pm
Pm
Pm __
Pm __
Pm _
Pm __
Pm
Pm __
Pm
E,0,Pm
Pm __
P.Pm
P,Pm __
P,Pm
Pm
Pm _
Pm^
Pm
Pm __
Pm
Pm
Pm .
Pm
Pm
Pm .
Pm
Pffl _J
^^^
^ ,
^r—
— ^
^
i
— 1
I
I
B-14
-------
T«bt»B.1(«ont.)
CHEMICAL NAME
~~ Chlorod»hydroid
12-Chlorod«hydro*bi«tio teid
~~ 14-Chlorodihydro«bwtia Mid
' Diohlorod*hydro«bi«tio moid
' 12,14-DioMorDcbhydnMbwtioAOid
• N«o*bi«tio«cid
Triohloro-2-o>eo-3-p»nt«noio aoia icom«r 1
* Triehloro-2-oxo-3-pMitenoioBoidiionMr2
5,5,5-Tricriloro-2-axo-3-p»nt«noio«oid
" 3,4,5,5-T»trmohloro-2-oxo-3-p»nt»noio aoid
•~~~~" Tftmohloro-2-e«o-3-p*ntenoio inmw 1
T«tr»ohloro-2-axo-3-p«rrt«noio itomw 2
3,4,5,5.S-P»nt«ohto»io-2-o>«o-3-p«nt»noie«eid
— ~~ ChlorothiophcndiMrboxylii) toid
" 5-ChlofethieprwndiMrboxylioaoid
-" Chloro-1-<2-h/Orox¥>-iiopropyt-4-m«thy( DOTMIM
"^Ojohlero-1-^2-hydro9(y}-fioprepyl-4-i'Mthyl btnimnt (5 wam*rt)
— 1 ,3-Diohloro-2-prop»nol
— 1,1,3-Triohk>ro-2-prop«nol
— ~~~ Cti*orc»c»t«ld»hyd»
— Diohlora>«tBld*hyd>
TriohloKMwtaldrtyda
— " 2-ChloreprapwMl
-— Chk.robut.rml
•—• DIohkNobutornri
•- 4-Chtoro-2-but«n«l
--• — 2-Chh>ro-2-but«nal
Chlocobsnnld^iyd*
-- — DiehloratMnnld»hyd»
— ChlorMMton*
— DiohlorMMlotM
1,1-DiohloraBoatDn*
— -- 1.3-Diahkmna^en*
Tri«hlora«o*tOM
^- 1,l,1-TrieMarMO«toM
1 , 1 ,3-Triohtoro«o»ton<
*>*• TdnohkMOBMtaxw
^-- — 1,1,1,3->T«traehlara«Mlon*
^-- 1.1,3>3-T«tntohleraua(on«
. PMttaohlonaMten*
H«cMhlora«o«ton»
^-- Triorilorooyolopropanon*
^-- — TriohloreeyelebutMioiM
„ — -"^ Diohkxw9yetopwitoM-t,2-dien«
^- 5,B-Diohloro-3-cyolop«nt»n«-t,2-dion»
^^ TriohtorooyolopOTton«-15-dton«(2iioiii»f»}
^--^ 3 .4.8-TriohtofO-3-qyotop«nt»f>»-1 3-dton«
^- — ^ 3,0,8-Trlohkxo-3-«votop«nt«n«-l^-dk>o«
^-* 4.S,»-Tfiohtero-3-eyelcpwt«n^15-dion«
^ — T«tmohtofocyolop»nt>iM-1^-dlon»
^^ — ~~ D»ohlon>-1^-b«»«oquinon»
CAS No.
57065-3B-6
57065-38-7
66281-77 -8
471-77-2
99165-S7-6
90165-A4-3
09165-85-4
M165-Q3-2
36157-45-6
X-23-1
39978-44-7
107-20-0
79-02-7
75-97-6
683-61-2
24443-10-6
53175-28-3
35013-09-8
311S8-00-*
78-96-6
513-86-2
534-O7-6
918-00-0
918-00-3
921-03-9
31422-61-4
16095-46-0
632-21-3
rrea-31-6
116-16-6
85451-72-2
78099-«-e
04650-07-2
110930-98-6
103354-08-1
89283-14-7
67051-44-4
110030-00-4
4054-42-6
18268-81-0
Ch»mlo^
Formuto
C20H2802CL
C20H2802CL
C20H28O2CL
C20M27O2CL2
C20H27O2CL2
C5H303CL3
CSH303CL3
C5H303CL3
C5H203CL4
C5H203CL4
C5H2O3CL4
CSHO3CL5
C8H3O4SCL
C3H60CL2
C3H40CU
C2H30CL
C2H2OCL2
LC2HOCL3
C3H30CL
C4H5OCL
C4H40CL2
CTHSOa
C7H40CL2
C3H5OCL
C3H4OCL2
C3H4OCL2
C3H4OCL2
C3H30CL3
C3H3OCL3
C3H3OCL3
C3H2OCU4
C3H2OCL4
C3H2Oa4
C3H20CLB
C30CL8
C3HOCL3
C4HOa3
C5H402CL2
CSH302CL3
C5H2O2CL4
C8H202CL2
(C*0]P
no. N
21
21
21
22
22
8
8
8
9
9
9
10
7
5
5
3
4
5
4
5
B
8
9
4
5
5
5
6
6
6
7
7
7
6
0
6
7
7
8
9
8
(Q"»
no. C
20
20
20
20
20
5
5
5
S
5
5
5
6
3
3
?
2
2
3
4
4
7
7
3
3
3
3
3
3
3
3
3
3
3
3
3
4
5
5
5
6
M.W.C
336.88
336.88
336.88
370.32
370.32
217.44
217.44
217.44
251.88
251.88
251.88
286.33
206.6
128.99
163.43
785
112.04
147.30
90.52
104.54
138.09
140,57
175,02
92.53
126.07
126.07
126.07
161.42
161.42
161.42
10546
105.86
105.86
230.3
264.75
159.4
171.41
166.08
201.44
235.88
176.09
D*,d
Cod*
E.Pm
D.Pm (
O.Pm
E.D.Pm
Pm
D.Pm
Pm '
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
E.Pm
Pm
Pm
Pm
H,P,Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
P.Pm I
Pm
Pm
P.Pm
Pm
Pm
P.Pm
Pm
Pm
Pm
Pm
P.Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
P.Pm
B-15
-------
Table B.I (oont.)
CHEUICALNAUE
Telraohloro-1 ,2-bennqulnone
3-Chloro-4-diohlorom«ttiyl-5-hydro. /-2(5H)-furmnon»
Diohloroaoetio aeid methyl alter
Chloraaoetie aoid ethyl eiter
Oiohloroieetio moid ethyl etter
Triohtoroaoetio aoid ethyl etter
Diohloromethane
Chloroform
Carbon tetraohtoride
Bromodiahloro methane
Dibromooh tore methane
1,2-Diohloreethane
1,1,1-Triohloroethane
1 . 1 .2 j-Tetraohloroethane
Triohloroethene
Tetraehloroethene
Tetraohloropropene (3 ieameri)
1 ,2 ,3,3- Tetraohloropropene
1 , 1 ,3,3-Tetraohloropropene
2 ,3 ,3,3-Tetraohloropropene
1 ,3 ,3,3-Tetraohloropropene
1 , 1 ,2,3-Tetraohloropropene
Pentaohloropropene
Diohloroprepmdiene
1 ,3-Diohk»ro-1 ,2-propadiene
1 ,1-Diehlore-U-proDadiene
Triehloropropadiene
Tetraohloropropadiene
Trlohlorabutatriene (2 iaomer*}
Tetraehlorobutatriene
PentaonlorobutBdiene (2 itomeri)
Methyl ohlorobutene (S ieameri}
Tetraohforooyolopentadiene
1 , 1 ,2.5,6,6-H«xaohloro-1 ,3,5-hexatrlene
1,1^.3,5,6-H«xaohloro-1,3,6-hexBtriene
1 ,2,3,4,5,6-Hmaohlofo-l t3,5-tiitttttrt.9rw
H«pteohloroh.ix&tri«n*) (2 iMfrwrt}
Chlorobmww
DiohbretenwM
1 vebHDlOnlOPODeMHWW
1 ,4-Oiohloroben«ene
15.4-Triohlorooenxene
Tetraohlorobeniene
1 ,2,3,4-Tetfaohtorobentene
1 ,2,3,5-Tetraohlorobenaene
1 ,2,4,5-TetraohlorofaenMne
Dtenta i%ti Ifwnl^n m^i^
nenwDfiivroDBfmno
Hexaofilorobenwie
CAS No.
2430-O3-Z
77430-76-0
116-64-1
106-38-5
535-15-0
515-44-4
75-08-2
67-66-3
56-23-5
75-27-4
124-48-1
107-06-2
71-65-6
79-34-5
78-01-6
127-18-4
60320-18-6
20568-65-8
18611-43-3
16500-81-7
19022-22-7
10436-30-2
60102-77-6
76720-38-2
83682-32-0
108562-60-3
20170-70-1
18608-30-6
06418-84-1
10782-18-8
55880-77-8
605-77-2
85418-84-3
2058861 6
101654-40-4
101654-38-0
88465-86-3
05418-06-4
10B-90-7
05-60-1
541-73-1
106-46-7
87-61-«
120-82-1
634-66-2
634-00-2
85-84-3
608-83-6
118-74-1
54411-10-7
4306-80-6
CnemioaJ
Formula
C8O2CL4
CSH303CL3
C3H4O2CL2
C4H702CL
C4H602CL2
C4H5O2CL3
CH2CL2
CHCL3
CCL4
CHBrCL2
CHBr2CU
C2H4CL2
C2H3CL3
C2H2CL4
C2HCL3
C2CL4
C3H2CL4
C3HCL4
C3H2CL2
C3HCL3
C3CL4
C4HCL3
C4CL4
C4HCL5
CSHOCL
C5H2CL4
C6H2CL6
C6HO.7
C8H5ai
C8H4CL2
C6H4CL2
C5H4CL2
C8H3CL3
C8H6CL3
C8H2CL4
06H2CL4
CSH2CL4
C6HCL5
C6CL6
C10H13CL
C10H13CL
{C+d)a
no.N
10
8
7
5
6
7
3
4
S
3
2
4
5
6
5
6
7
7
5
6
7
7
8
0
6
0
12
13
7
10
10
10
11
12
11
11
| CO"
no.C
5
5
3
4
4
4
1
1
1
1
1
2
2
2
2
2
3
3
3
3
3
4
4
4
5
S
6
6
6
6
6
6
6
6
6
6
6
6
6
10
10
u.w.c
740.88
217.44
142.87
122.55
157
101.44
84.83
119.36
163.82
163.83
208.3
08.87
133.41
167.86
131.4
165.83
170.86
Data a
Coda
Pm
Pm
Pm
Pm
Pm
Pm
H.E.P
H.E.P.D
H.P
H,P
H,P
H.E.P
H.E.P
H.E.P
H.P
H.P
Pm
Pm
! Pm
214.31
108.06
143.4
177.84
155.41
180.86
226.32
104.56
203.68
286.6
321.24
113.58
147
147
147
147.01
181.40
181.5
215.0
215.0
215.0
215.8
250.3
284.8
168.67
Pm
Pm
Dm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
H.E,P,Pm
P
H.E.P
C.P
E.P
P
H,P
P
P
P
H,P
H.E.P
HP
P
P
B-16
-------
T«bt»B.1(oont.)
CHEMICAL NAME
Chloro-p-eynwn« (2)
Dichlofo-p-oyrtHin*
Diohloro-p-oym«n» (1)
Diohtoro-p-oymw» (2)
3 ,5-Di ohtoro-p-oy nwrw
— 2,6-Diehloro-p-oym»n»
. TrlohlofO-p-«yin«iw
Chlorodimethyfpropyldihydronaphthftl'irw
Oioh1orodim«thylpropyldihydrofwphth>l«n«
DiohlofOdim«thylpropylnaphthalMW
T*rp*rtM
— - , ChlorotrinwthoxytMnwn*
2-Chtofo-1 ,3,8-tfim«thoxyfa»ni»n>
1-Chtoro-2.3,4-trim«thoxyben««n«
. 1-Chloro-2,4,5-trimrthoxyb»nz«n«
5-Chtoro-1 ,2.3-trim«thaKyb»rn»n«
Triehlorotrlm«thoacyb»ni«n«
1.2.4-Triohloro-3.5,6-trim»thoxyb«n»n«
1 ,2,3-Triohk>fo-4,S,6-trlm»Hx>xyb«ni»n»
1 .3.5-Tfiohlero-2.4.6-trim«thoxyb»n*«n«
1 . 1 -Dichtorodim+thytiulfon*
|— . 1.1.3-Tfiohtefodimsthyltulfon*
Triohlorothioptwn*
- 2.3,4-Trioh)crothioph»n»
2,3,5-Trtohlorothiophxw
r T«tr«ohlorcthioph6:
i- Diohloro-2-formytthioph«n«
-__^ 3.5-Diohloro-2-tormyUhioph«n«
-^__ 3.4-Diohloro-2-fof rnylth toph«n»
• 4.5-Diohtofo-2-forn»vlthloph«n«
Triohloro-2-fefinytthioph«n«
- 2-Ao»tytd»ohloreth>oph«n« (2 i«om«ft)
_ a-Aa*tyrtriehtetothioph«n«
~___ 4.B-PichtofX>-2-«o«tymiiOph«o»
— 3.4-Oichlofo-2-«o»tyHhloph«n«
^_^ ptohtofo-2-prop>ooyithioph«fM (2 itonMf i)
--_____ 2,5-Diehk>fo-3-pfopionyHhioph^>ii
-__ Triohtefov^atrelg
_ T^mehlaravaratrel*
^__ ChtoriFMt^l Ph^ol*
-. CtiU)fin«t»d R^ln Ao>di
Chtorirntod Guaiaoolt
^__ Chteriintod Cateohgti ..._
Chloriratad Vwmtrohw
», Chlorinated Vinlllini
}tMMA»OPMpOINHH
Chteroph^»e>
-^_ 2-Chk>foph«nol
. 3-Chtoroptwxd
~- 4-ChtofOpt^no*
L- 2.3-Dlohk>reph«nol J
CAS No.
4386-79-3
65724-12-1
81686-41-1
81688-44-4
81686w»-S
81686-46-6
66329-06-2
66365-28-4
77241-44-2
67827 -56-fl
54625-83-«
20128-10-6
2676-80-1
77223-86-4
00666-78-1
2930-13-1
1015-86-8
37567-97-4
64568-19-0
26761-63-5
17249-78-4
17249-77-3
6012-07-1
66717-67-1
67482-51-3
67482-60-2
67482-48-0
66729-79-1
66717-48-2
6S434-11-8
57681-60-1
57681-66-0
66717-89-3
32427-79-S
73513-25-4
[944-61 -6 ~_J
[25167-60-0
ao-o>^
100-43-0
576-24-9
Ohmauatl
Forimita
C10H13CL
C10H12CL2
C10H12CL2
C10H12CL2
C10H11CL3
C15H20CL
C15H19CL2
C15H16CL2
CSH1103CL
C9H9CL3
C2SH4O2CL2
C2SH3O2CL3
C4SHCL3
C4SCL4
C5SH2OCL2
CSSCL3
C6SH4OCL2
C6SH3OCL3
C7SH6OCL2
C6H5OCL
C6H5OO.
C6HSOCL
C6H40CL2
(C*CI)a
no.N
11
12
12
12
13
14
16
17
17
10
12
4
5
7
8
7
8
8
' 9~
9
7
7
_
7
0
(C>B
no. C
10
10
10
10
10
14
15
15
15
8
8
2
2
4
4
5
5
6
6
7
6
6
r
6
6
M.W.C
203.12
237.56
224.77
235.78
270.22
267.19
202.64
271.54
163.02
197.46
187.48
221.92
181.04
214.47
185.07
229.52
206.00
128.56
128.56
128.56
163
MB"
Code
P
P
P
P
Pm
Pm '
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm 1
Pm
Pm
Pm
Pm
Pm
Pm i
Pm
Pm
Pm
Pm
Pm
P,Pm
Pm '
Pm
PHI
Pm
Pm
Pm
Pm
Pm
D,Pm
Pm
D.Pm
' D,Pm
0,Pm
i^ip
B-17
-------
B.I (aont.)
CHEMICAL NAME
2,4-Dlohtoroph»nol
2,5-Oiehlorophcnol
2,6-Oieh!oroph*nol
2,3,6-Triohtoroph»nol
2,4,5-Triohlofoph»nol
2,4,6-Trfohlorpphmol
2,3,4.5-T«trmchloroph«nol
2,3.4,6-T«tr»ohforoptwrral
2,3,5,6-Trtrmohlorophwiol
>vntBOfl loropnttftol
DiohlorocBfatohol {2 iiom*r*)
3,4-OioMoraoctBohol
3.6-DfohloraoctMtMl
4,5-Oiohh>roo>t«ohol
Triohloro«>teohol (3 i«om«r«}
3,4.5-Trlohtoroc«t«ohol
' 3,4,6-Trlehlorratmhol
TvtraohlorpMttohal
3,4,.5.6-T»»rmohlofoo«t»ehol
Chlomgrataool
6-OiloregiMMeol
DJahtofoguMaot
4,5-DioMon>guM«i>l
3,4 ,6-TrloMoroouMMol
3,4 ,5-TrioMoraguMoal
4IB,S-Triohtoro9u«i«ed
T*traehloreguBiaeel
3,4,9,6-TvtraohlorogiMtaool
6-Chtorownillln
5,6-Diohlorownillin
DieMorawniHin
Triohlorovwiillm
3.S-Oleh(ore-4-hydreMytoiinldihvdt
e-Chkmpiatowt*ahu«ld^ivcto
OiohJoropreloa^MhiMld^iyd*
ChbroiynngMld*hyd*
• Dtohlonwyr{ngMl(W«yd»
Chtoro- 1 ^Xrihydmyten wrw
Diahlon>-1 2,3-trihydreKytenMiM
Trlohtero-1 ^>trltiydrexytwnMM
Chlor»- 1 3. ,4-triKydrae(ybwiM(w
Diohtero-1 2,4-trihydraxybwiMn*
Triohloro-12.4-trihydroj(yb»n»ft«
3.4,5-Tri«hlera«yrtnga4
OiaMoroKWtetyrinooiM
Diehk»ro-1 ^-dlhydroxy-S-iMthcwybwiMfw
Chloro-3,4-dlhydraKyprepwplMnefM
DiohtefO-3,4-dihydroxypropioph«x»n« (2 Iwmws)
ChloroprapwgiMia MM (ohtoropraptovwillton*}
CAS No.
120-83-Z
563-TB-a
67-66-0
033-75-6
06-05-4
88-06-2
25167-83-3
4901-E1-3
58-80-2
936-05-6
87-86-6
25167-86-5
3878-67-4
3838-16-7
13673-82-2
3428-24-8
25167-84-4
56861-20-7 j
32136-72-3
1188-65-6
1188-65-6
3743-23-5
2460-48-3
2460-48-3
60712-44-8
570B7-83-7
2688-24-8
2530-17-6
2838-17-6
16268-76-3
18268-68-4
18268-60-4
118464-61-6
2314-36-6
32864-11-2
34088- 16-6
37686-56-9
76341-60-4
76330-08-8
75B62-88-8
75662-90-2
58061-21-6
75662-81-3
79662-8S-4
56861-22-fl
2530-26-6
84648-71-6
7666243-6
26068-71-0
66021-60-6
Chmnied
Formula
CBH4OCL2
C6H4OCL2 J
C8H4OCL2
C8H3OCL3
06H30CL3
CSH30CL3
C6H2OCL4
C6H20CL4
C6H2OCL4
C6H20CL4
C8HOCL5
CSH402CL2
C6H4O2C12
C8H302CL3
C8H3O3CL3
C6H303CL3
C6H2O2CL2
C6H202CL2
C7H702CL
C7H6O2CL2
C7H602O2
C7H5O2CL3
C7H502CL3
C7H5O2CL3
C7H4«CL4
C7H4O2CL4
C8H703CL
C8H603O2
C8H6C3CL2
C8H503CL3
C7H4O2CL2
C7H8O3O.
C7H4O3CL2
COHOO4CL
COH8O4CL2
C6HB03CL
C8M4O3CU2
C6H3O3CL3
C8H903CL3
C6H40302
C6H3O3a3
C8H7O3O.3
C10H1004CL2
C7H603CL2
C8H8O3CL
C9H803CL2
(C4d>a
no. N
10
10
10
10
11
8
8
10
10
10
11
11
9
10
10
11
9
6
8
10
11
7
6
8
9
8
0
11
12
0
10
11
(Qb
no. C
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
7
7
7
7
7
7
7
7
a
8
8
8
7
7
7
6
6
a
10
7
0
8
M.W. C
163
163
163
197.45
197 .45
197.45
231.9
231.9
231.8
231. B
266.34
178
178
213.46
213.46
213.46
247.88
24T.88
1S8.6
189.04
193.04
227.48
227.48
227.48
261.83
281.83
166.50
221.04
221.04
256.48
181.01
218.63
251.07
160.56
196.01
22848
160.56
196.01
228.46
267 .5
286.00
208.03
200.62
236.06
— 0*0
Co*
H,P,D
P,Pm
E,P,D
E.P
H,E.P,D_
H.E.P.D
Pw
F>
M,E,P,D
E.P.Pm
H,E,P.D _
P.Pm
D,Pm
D,Pm
E,Pm
D,E.Pm
Pm
D,Pm J
P,D,Pm
E,P,W*i"^
E,P,O.Pm _
Pra _2
Pm __
D,Pm
P,D,P«_
Pm
E.P.O.Pm
E,P,D,PmJ
P.D.Pm
E,0,P»_
D,Pm
D,Pm __
Pm
Pm
Pm
Pm
PlB
Pm
Pm
Pm
Pm
Pm
Pm
Pm
Pm
— — • — —
Pm
Pm
Pm
D,Pm
Pm
Pm
Pm
Pm
DM
B-ia
-------
Tabte B.1 (oont.)
CHEMICAL NAME
Triohlorodahydrooonitoryt aloohol
3,4,5-Triohloro-2,6-«lim«thcKyph«nol
" 3,9-Diohloro-2,6-dim«thoxyph«nol
CAS No.
2930-26-6
78782-46-4
OwmiosJ
Formula
C10H10O3CL2
C10H7O3CL3
C8H7O3CL3
C8H803CL2
(0*0)"
no N
12
13
11
10
(C)b
no. C
10
10
8
8
M.W.C
248.00
281.92
257.5
223.06
^rwm
Codi
Pm I
Pm
Pm
Pm
afjumber of carbon and chlorine atoms in molecule.
Dumber of carbon atoms in molecule.
cHolecular weight.
code:
H - Human health data readily available;
E - Ecotox data readily available;
P - Physical/chemical and fate properties readily available;
P - Minimal p/chem properties readily available; and
D - Treated effluent data available.
ogferences: Suntio, et al . (1988); USEPA (1988); NCASI (1977, 1980, 1981, 1982a, 1982b, 1983a, 1983b,
1983c, 1984a, 1984b, 1985a).
B-19
-------
TABLE B-2
OTHER CHLORINATED ORGANIC CHEMICALS CANCER RISK
,~ Chemical r,- -
Carbon Tetrachtoride
Chloroform
1,1,1 -Trichtoroethane
Bromodichloromethane
Chloral
1 ,2-Dichloroethane
Hexachlorobenzene
1 ,2,4-Trichtorobenzene
2.4-Dichlorophenol
Dibromochtoromethane
Tetrachloroethytene
Pentachlorobenzene
2,3.4.6-Tetrachlorophenol
Dichloromethane
Trichloroethylene
1 ,1 ,2,2-Tetrachloroelhane
Pentachtorophenol
2,4,6-Trichlorophenol
1 ,2-Dichlorobenzene
1 ,3-Dichlorobenzene
1 ,4-Dichlorobenzene
2-Chtorophenol
1 ,2,4,5-Tetrachtorobenzene
2.4,5-Trichlorophenol
Chlorobenzene
Chororacetic acid
CAS
No.
56-23-5
67-66-3
71-55-6
75-27-4
75-87-6
107-06-2
118-74-1
120-82-1
120-83-2
124-48-1
127-18-4
60893-5
58-90-2
75-09-2
79-01-6
79-34-5
87-86-5
88-06-2
95-50-1
541-73-1
106-46-7
95-57-8
95-94-3
95-95-4
108-90-7
79-11-8
Cancer
Classification
B2
B2
0
NA
NA
B2
NA
D
NA
NA
NA
NA
NA
B2
NA
C
NA
B2
NA
NA
NA
NA
NA
NA
NA
NA
Oral
Slope Factor
mg/Kg-day
1.30E-01
6.10E-03
NA
NA
NA
9.10E-02
NA
NA
NA
NA
NA
NA
NA
7.50E-03
NA
2.00E-01
NA
2.00E-02
NA
NA
L NA
NA
NA
NA
NA
NA
Drinking Wafer
Unit Risk
ua/|
3.70E-06
1.70E-07
NA
NA
NA
2.60E-06
NA
NA
NA
NA
NA
NA
NA
2.10E-07
NA
5.80E-06
NA
5.70E-07
NA
NA
NA
NA
NA
NA
NA
NA
Inhalation
Slope Factor
mg/Kg-day
1.30E-01
8.10E-02
NA
NA
NA
9.10E-02
NA
NA
NA
NA
NA
NA
NA
1.40E-02
NA
2.00E-01
NA
2.00E-02
NA
NA
NA
NA
NA
NA
NA
NA
Inhalation
Unit Risk
ug/cu. m.
1.50E-05
2.30E-05
NA
NA
NA
2.60E-05
NA
NA
NA
NA
NA
NA
NA
4.10E-06
NA
5.80E-05
NA
5.70E-06
NA
NA
NA
NA
NA
NA
NA
NA
00
I
rv>
o
NA = Not Available
-------
TABLE B-3
OTHER CHLORINATED ORGANIC CHEMICALS DRINKING WATER HEALTH ADVISORIES
- v- ..v-> „--.,» - ' ; '=
-fc^ Cnerofcaf_/ ^
' , \ , V ^ •• " "
Carbon Tetrachloride
Chloroform
1,1,1 -Trichtoroethane
Bromodichloromethane
Chloral
1 ,2-Dtchloraethane
Hexachlorobenzene
1 ,2,4-Trichtorobenzene
2.4-Dichlorophenol
Dibromochloromethane
Tetrachloroelhylene
Pentacntorobenzene
2,3,4,6-Telrachtorophenol
Dichtoromelhane
Trichloroethylene
1 ,1 ,2,2-Tetrachloroethane
Pentachlorophenol
2,4.6-Trichbrophenol
1 ,2-Dichlorobenzene
1 ,3-Dfchlorobenzene
1 .4-Didilorobenzene
2-Chlorophenol
1 ,2,4,5-Tetrachlorobenzene
2,4,5-Trichtorophenol
Chbrobenzene
Chtoroacetic acid
CAS
-. $* •::
^ - „,;,
56-23-5
67-66-3
71-55-6
75-27-4
75-87-6
107-06-2
118-74-1
120-82-1
120-83-2
124-48-1
127-18-4
608-93-5
58-90-2
75-09-2
79-01-6
79-34-5
87-86-5
88-06-2
95-50-1
541-73-1
106-46-7
95-57-8
95-94-3
95-95-4
108-90-1
79-11-8
Maximum
, Contamination
--, Limit mg/l
5.00E-03
2.00E-01
5.00E-03
O.OOE+00 D
9.00E-03 b
5.00E-03 D
5.00E-03
2.00E-01 ft
6.00E-01 «>
7.50E-02
1.00E-01 b
; Water8 ,
_,/
-------
TABLE B-4
OTHER CHLORINATED ORGANIC CHEMICALS REFERENCE DOSES
Chemical '' .:.-..,,-,
ff -. ., -. % ^ %-w
\ v^J.^v. ^ * ^
1 ,1 ,1 -Trichtoroethane
Carbon Tetrachtoride
Chloroform
Bromodichloromethane
Chloral
1 ,2-Dichloroethane
Hexachlorobenzene
1 ,2,4-Trichlorobenzene
2,4-Dichlorophenol
Dibromochloromethane
Tetrachloroethylene
Pentachlorobenzene
2,3,4,6-Tetrachtorophenol
Dichtoromethane
Trichloroethylene
1 ,1 ,2,2-Tetrachloroethane
Pentachlorophenol
2,4,6-Trichtorophenol
1 ,2-Dichlorobenzene
1 ,3-Dichlorobenzene
1 ,4-Dichlorobenzene
2-Chbroprienol
1 ,2,4,5-Tetrachtorobenzene
2,4,5-Trichlorophenol
Chorobenzene
Chbroacetic acid
,CAS
'" No*
71-55-6
56-23-5
67-66-3
75-27-4
75-87-6
107-06-2
118-74-1
120-82-1
120-83-2
124-48-1
127-18-4
608-93-5
58-90-2
75-09-2
79-01-6
79-34-5
87-86-5
88-06-2
95-50-1
541-73-1
106-46-7
95-57-8
95-94-3
95-95-4
108-90-1
79-11-8
OralflfD
mg/Kfl-day
9.00E-02
7.00E-04
1.00E-02
2.00E-02
2.00E-03
ND
8.00E-04
2.00E-02
3.00E-02
2.00E-03
1.00E-02
8.00E-04
3.00E-02
6.00E-02
NO
ND
3.00E-02
ND
9.00E-02
ND
ND
^ 5.00E-03
3.00E-04
1.00E-01
2.00E-02
ND
Oral
: ur
1,000
1,000
1,000
1,000
10,000
ND
100
1,000
100
1.000
1.000
10.000
1,000
100
ND
ND
100
ND
1.000
ND
ND
1,000
1,000
1.000
1,000
ND
Dermal RfD
mg/Kg-day
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Dermal
UP
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
inhalation RfD
fng/Kg-day
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Inhalation
UF
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
RfD
Date
6/30/88
6/30/88
6/30/88
12/1/88
8/22/88
ND
2/1/89
6/30/88
6/30/88
6/30/88
3/1/88
3/1/88
3/1/88
1/1/89
ND
ND
6/30/88
ND
8/1/89
ND
ND
8/1/89
2/1/89
3/1/88
8/1/89
ND
00
ro
no
ND = No Data
-------
uaowi
Table B-5. Species List*
Code
Latin
Canon name
AF
AM
AS
BBJ
BG
BR
CA
CC
CF
CRV
CS
CV
DM
GA
HF
IP
LM
LP
HA
ME
HIS
NN
NS
00
PAP
PP
PR
RM
SC
ST
TN
XL
Astacua fluviatilis
Apbvstam aexicanui
Art gala salina
Bufo bufo Janonicua
laria q labrata
Brachvdanid rerio
Caraaaiua auratua
Cvorinus caroio
Colisa fasciata
Crassostrea viroinica
Cranqon seotenaoinoaa
Cvorinodon varieaatua
Daohnia •aona
Gasterosteua aculeatus
Heterooneustea foaailia
Ictalurus punctatua
Leooens sacrochirua
Leeaa PBrouailla
Mva arenaria
Mesidotea entanxi
Micropterus aaleoidea
Notopterus notooterua
Nitora aoinlnes
Qphrvotrocha diadana
Palieaonatea euaida
Piegohales pmeelas
Poecilla leticulata
Rivulua sanmratus
Salsn oairdnAri
Salao trutta
Tilaoia •oasaabica
XenoPtii 1aev1«
Crayfiah
Salannder
Brine Shrii?
Frog
Snail
Zebra Fish
Goldfish
COMBO (Colored) Carp
Giant fiouraai
Awriean (Virginia) Oyster
Sand Shri*>
Sheepahead Minnow
Waterflaa
Three Spine Stickleback
Indian Catfish
Channel Catfish
Bluegill
OuckMed
Soft Shell Cla«
Aquatic SoHbug
Largeaouth Baas
Featherback
Copepod
Polychaete
Grass Shriep, Fresh Water Prawn
Fathead Minnow
Guppy
Rtvulus (fish, eaybe Marine Minnow)
Rainbow Trout
Brown Trout
Mozambique Tilapla
Clawed Toad
" Corresponds to species codes used in Table B-6.
B-23
-------
Table B-6. Toxlclty of Other Chlorinated Organics from Bleached Wood Products to Aquatic Species
Other chlorinated organ ics
2,3,4.6-Tetrachloroohenol
2,3,5.6-Tetrachlorophenol
2,3,5-THchlorocatechol
2.3.6-Trichloropnenol
2.4,5-THchlorophenol
2.4.6-Trichlorophenol
2,6-Dlchlorophenol
Species9
DM
ST
CA
CS
HA
LN
LM
LN
RN
OH
DM
LH
LN
DM
AF
ON
ON
LH
ST
CA
LN
DM
ON
MA
ST
CA
CS
LN
LN
PP
PP
PP
PP
DM
ST
CS
Exposure
duration
(hours)
48
24
24
96
96
24
96
24
96
24
48
24
96
24
8 Days
24
48
96
24
24
24
24
48
96
24
24
96
24
96
96
96
6 Days
8 Days
24
24
52
LC50
(•B/1)
0.29
0.5
0.75
11.8
11.8
0.19
0.14
0.14
1.1
2.50
0.57
0.4
0.17
3.39
5.4
3.8
2.7
0.45
0.9
1.7
0.61
5.0
6.0
3.9
1.1
10.0
2.7
0.72
0.32
9.7
8.6
5.8
6.4
13.7
4.0
19.1
Noerinal (n)
vs
•ensured (•)
N
N
N
M
H
N
N
N
N
N
N
N
N
N
M
N
N
N
N
N
N
N
N
N
M
N
N
N
N
N
M
N
N
N
N
M
B-24
-------
Table B-6. (continued)
Other chlorinated organ ics
Chlorophenol
2-Chlorophenol
4-Chlorophenol
2UH + I«J*WU A C £ fc»I»kT-—»»^. 1
-HBtnoJty-4 . 3,D-trlcn loropneno 1
3,4.5,6-Tetrachlorocatochol
Tatracnlorocatechol
Species"
LN
ON
DM
CA
LN
PP
PP
PP
PP
PP
LN
LN
LN
CA
CA
CA
PR
PR
PR
CS
LN
LN
NS
DM
DM
CA
LN
LN
ME
ME
CS
DM
ST
ST
NS
DM
Exposure
duration
(hours)
96
24
48
24
24
96
8 Days
24
96
48
24
48
96
24
48
96
24
48
96
96
24
48
96
24
48
24
24
96
96
7 Days
96
24
24
24
96
24
LC50
(•9/1)
6.6
0.06
2.6
16.0
7.2
11.0
6.3
21.96
11.63
18.0
11.31
10.59
10.0
14.48
12.37
12.37
ZZ.17
20.78
20.17
5.3
8.2
8.2
21.0
8.8
4.1
9.0
4.0
3.8
40.3
37. S
4.6
2.B
1.1
1.1
3.3
2.23
Nominal (n)
vs
•easured (•)
N
N
N
N
N
N
N
N
N
N
N
N
N
N
M
N
N
N
N
N
N
H
N
H
N
N
N
N
N
N
N
H
N
H
N
N
3.4.5,6-Tetrachlorogualocal
S6
96
0.32
B-25
-------
8906K
Table 6-6. (continued)
Other chlorinated organ ics
Tetrach lorogua lacol
3,4,5-Tricnloroguaiacol
3.5-Dichlorocatechol
4,5,6-Trichlorogualacol
4.5-Dichlorocatechol
9.10-Dichlorostearic acid
Carbon tetrachlorlde
Chlorobenzene
Chloroform
Species'
NS
DM
56
ST
MS
DM
DM
ST
DM
SG
LN
MB
DM
DM
DM
DM
PP
SG
56
SB
SG
SG
DM
MIS
MIS
MIS
MIS
IP
IP
IP
IP
LP
LP
56
SG
56
Exposure
duration
(hours)
96
24
96
24
96
24
24
24
24
96
96
96
24
48
24
48
24
28 Days
28 Day»
28 Days
32 Days
32 Days
48
12
24
12
24
24
12
48
96
12
24
48
96
12
LC50
(•g/D
3.9
4.96
0.75
2.9
5.Z
2.9
2.8
2.3
6.64
2.5
125.0
150.0
35. 0
30.0
140.0
86.0
25.4
2.03
Z.O
2.16
1.24
2.03
29.0
S6.2
56.2
50.4
50.4
235.0
126.0
191.0
75.0
17.1
17.1
21.4
18.2
37.1
Noainal (it)
vs
•usured (•)
N
N
N
N
N
N
N
N
N
M
N
N
N
N
N
N
N
N
N
M
M
M
N
M
M
M
M
M
M
M
N
M
M
N
M
M
B-26
-------
won
Table B-6. (continued)
Other chlorinated organ ics
Chloroform
(Continued)
Ethylene dlchlorlde
Oichlorodlhydroabietlc acid
Nethylene chloride
Nonochlorodlhydroabietic acid
Pentach lorobenzene
Pentachlorophenol
Species*
SG
PR
S6
LM
.BR
AS
SG
S6
OH
DM
LN
LM
DM
PP
PP
PP
PP
PP
PP
PP
PP
S6
SG
DM
DM
00
00
US
DM
DM
BBJ
BBJ
IP
00
LM
CA
CS
Exposure
duration
(hours)
24
96
28 Days
7 Days
48
24
96
96
24
48
96
24
24
96
72
48
24
96
24
48
72
96
96
24
48
24
24
96
24
48
24
96
24
24
96
24
66
LC50
(-9/D
26.1
300.0
1.240
2.03
100.0
320.0
0.6
0.8
310.0
220.0
220.0
230.0
2.270.0
193.0
232.0
265.0
268.0
310.0
112.8
99.0
99.0
0.6
0.8
17.0
5.3
1.5
2.4
0.27
1.5
0.68
0.17
0.12
0.12
1.5
0.26
0.27
3.3
Noainal (n)
VS
Measured (•)
M
N
M
M
N
N
N
N
N
N
N
N
N
M
M
N
M
N
H
M
M
M
N
N
N
M
N
N
N
N
N
N
N
H
N
N
N
B-27
-------
oauon
Table B-6. (continued)
Other chlorinated organic: Species1
Pentachlorophenol Nil
(Continued} 48
KN
HF
HF
HF
HF
CF
CF
CF
CO
00
CA
CA
CA
CA
CA
LM
LM
LM
LM
AF
AF
CA
DM
DM
ON
ON
DM
6A
TM
CA
86
BG
CV
CV
AN
PP
BR
CSV
NN
CF
Exposure
duration
(hours)
24
0.109
72
24
48
72
96
24
48
96
48
72
72
96
96
46
72
6
24
48
72
8 Days
8 Days
24
21 Days
24
48
24
48
24
24
24
24
24
96
96
48
8 Days
48
14 Days
96
72
No
LCSO
(•9/D M
0.113
N
0.09
0.41
0.38
0.33
0.29
0.57
0.52
0.45
1.1
0.9
0.057
0.023
0.056
O.OB
0.084
0.057
0.038
0.031
0.025
26.0
9.0
0.052
0.48
1.7
1.0
2.8
1.5
0.37 (LCgg)
0.8
1.6 (LCgj,)
1.0 (LCg,,)
0.25 (LCg^j)
0.329
0.223
0.3
0.2
0.4
0.07
0.083
0.48
•inal (n)
vs
asured (•)
N
N
N
N
N
N
N
N
N
M
N
N
N
N
N
N
N
N
N
N
N
M
M
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
B-28
-------
Table B-6. (continued)
Other chlorinated organ ics
Tetrachloroacetic acid
Trichloroacetic acid
Trlchloroethylene
1.1,1-Trichloroethane
1,1.2.2-Tetrachloroethane
IT n i hi ii ii ••! • • ii i
, z-D ten lorownzene
1,3-Dlchlorobenzene
1 . 4-D Ich lorobenzene
l.Z-Dlchloroethane
Species'
PP
CC
PP
PP
NS
PP
DM
DM
LM
ON
ON
LM
LM
PP
PP
ON
DM
LM
LH
LM
MB
PP
PAP
PAP
PAP
ON
OH
LM
LM
DM
DM
DM
DM
DM
DM
Exposure
duration
(hours)
96
96
96
96
96
48
48
24
96
48
24
24
96
72
96
24
48
24
48
24
96
24
96
48
24
24
48
96
24
24
48
24
48
24
48
LC50
(•5/1)
2.000.0
2,500.0
2.000.0
2,000.0
4,800.0
2,000.0
18.0
22.0
45.0
530
1.300
40
40
55. 4
52.8
18
9.3
21
21
21
7.3
25.4
9.4
10.3
14.3
2.4
2.4
5.6
6.3
48
28
42
11
250
250
Noiinal (n)
V3
•easured (•)
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
M
M
M
N
N
N
N
N
N
N
N
N
M
N
B-29
-------
Table B-6. (continued)
Other chlorinated organics
1,2,3, 5-Tetrachlorobenzene
1 ,3-D1ch1oro-2-proponal
Species*
CA
CV
ON
ON
CA
Exposure
duration
(hours)
3
3
24
48
24
LC50
(•3/1)
2.5
1.58
18
9.7
680
Ho>inal (n)
vs
•euured (•)
N
N
N
N
H
a Species codes are listed on Table B-5
B-3Q
-------
Table B-7. Ranking by Acute Toxicity to Aquatic Organ Is
2
2
3
3
1
1
Chemical
Pentach loropheno 1
,3,4. 6-Tetrach loropheno 1
. 3 . 5 . 6-Tetrach loropheno 1
2 , 4 , 6-Tr ich loropheno 1
. 4 , 5 . 6-Tetrach lorogua iacol
2 . 4 , 5-Trichlorophenol
Dichlorodihydroabietic acid
Honochlorodihydroabietic acid
3,4, 5-Tr ich lorogua iaco 1
, 4 , 5 , 6-Tetrach lorocatecho 1
Chlorofone
4.5-Oichlorocatechol
1 , 2-D i ch lorobenzene
9,10-Oichlorostearic acid
Z-Chlorophenol
4.5.6-Trichloroguaiacol
3,5-Olchlorocatechol
2.3,5-THcnlorocatechol
4-Ch loropheno 1
2. 6-0 ich loropheno 1
Pentach lorobenzene
2.3.6-Trichlorophenol
Ch loropheno 1
,1,2. 2-Tetrach loroethane
,2.3. 5-Tet rach lorobenzene
1 , 4-0 ich lorobenzene
Trichloroethylene
Dichloroacetic acid
Ch lorobenzene
1 ,3-Oich lorobenzene
Carbon tetrachloride
1.1.1-Trichloroethane
Metnylene chloride
1,2-Oichloroethane
Ethylene Oichloride
l,3-Oichloro-2-proponal
Tetrachloroacetic acid
Trichloroacetic acid
Lowest
LC50
(•9/D
0.023
0.14
0.17
0.32
0.32
0.45
0.6
0.6
0.75
1.1
1.24
2.3
2.4
2.5
2.6
2.8
2.9
3.39
3.8
4.0
5.3
5.4
6.6
9.3
9.7
11.0
18.0
23.0
25.4
28.0
30.0
40.0
99.0
220.0
320
680
2,000
2.000
Nunber of
Studies
57
9
4
13
3
6
1
I
1
2
24
2
9
1
21
4
1
1
9
2
2
1
1
6
6
6
3
1
2
2
5
7
13
2
1
1
7
5
B-31
-------
8906H
Table B-8. Acute Toxicity of Selected Chlorinated Organic
Chemicals (OCO) to Terrestrial Organisms
Chemical
2 , 3 , 4 , 5-Tetrach loropheno 1
2.3,4. 6-Tetrach loropheno 1
2.4.5-Trichlorophenol
Chloroform
9 f»U |n«wmli»«n 1
c^Jt loropnBno i
1 ,2-Dlchlorobenzene
4 -Ch loropheno 1
1 .4-Dichlorobenzene
Pentach loropheno 1
Pentach lorobenzane
Lowest
LC50
•gAfl
400
140
250
250
1.620
1,194
80
670
670
500
2.000
261
500
2.950
2.800
56
105
168
1.080
1.175
Exposure
route
Oral
Oral
Skin
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Skin
Oral
Oral
Oral
Organ is*
Mouse
Rat
Rabbit
Guinea pig
Rat
Rat
Mouse
Rat
Mouse
Rabbit
Guinea pig
Rat
Rat
Mouse
Guinea pig
Rat
Rat
Hamster
Rat
Mouse
B-32
-------
TABLE B-9.
SUMMARY OF TREATED-EFFLUENT CHEMICAL CONCENTRATIONS
CO
I
CJ
OJ
^ c**tirt&^' *'«
i~k;£ ^%M^M:- *
2.4 Dfchlorophenol
2,6 DteWorophend
2.4.5-Trichloropheno(
2.4.6-TricNofophenoJ
2.3.4.6-Tetrachlorophenol
PentacNorophenol
3,4-DlcNorocatechol
3,4.5 Trichlorocatechol
3.4.6-TricNorocatechol
TetracNorocatechd
3.4.5.6-TetracNofOcatechol
4.5-DicNoroguaiacol
3,4.5-TricNorogualacol
4,5.6 Trichlorogualacol
TetracNorogualacol
3.4.5.6-Tetrachlofogualacol
6 Chorovanllin
5.6 DIcNorovanllln
9.10 DfcWorostearic acid
12-CNorodehydroabletlc acid
14 Chlorodehydroabtetlc acid
DIcNorodehydroabtetic acid
Neoabietlc acid
Chloroform
TricNoroveratrole
Tetrachloroveratrole
.: •• fiange a VAMKUHIWIUII
" : ^ufl/i*-^'-
f>'- iOU %*?
0.4
1.2
0
1
0.5
0
0.8
1.7
1.7
0.1
2
0
1
0.6
0.4
1
3.8
0.8
10
2
2
1
2
3
33.5
10
IbHgh,:,,
51
12.3
61
5
27
1.4
114
19
88
92
24.3
78
42
50.1
36
21
12.4
115
162
27.3
126
91
1688
59.2
47.8
nunwer
- ^5 of
Samples
39
1
7
59
14
27e
2
29
4
23
6
14
35
33
30
9
5
5
4
20
13
27
16
32
4
4
vofrespunumg »
- MlPwoar- •"'--
If* ..:. :
CEHDED
CEHDED
BK
CEHDED/BK
CEHDED
BK
CEHDED
BS
CEHDED
CEHDED
CEHDED
BK
CEHD/CEHDED
CEHDED
C/DED
CEHDED
BK
BK
CEHDED
BK
CEHDED
BK
BK
NG
NG
NG
HflU
CEHH/CEHD
BS
CEHH/CEHD
BK
BS
NG
BK
CEHD
NG
BK
NG
CEHDED
CEHDED
C(D)E(0)H/D
CEHDED
CEH
C(D)E(O)H/D
NG
BK
NG
CEHDED
BK
K
NG
NG
LCSOc
ug/l
4000 (2)d
450(6)
320 (13)
140 (9)
23 (51)
1100(2)
5200 (3)
320 (3)
2500(1)
600(1)
1240 (24)
-------
TABLE B-9. (continued)
SUMMARY OF TREATEDfFFLUENT CHEMICAL CONCENTRATIONS
Number
1 *-
Sample*
Corresponding b
tow
HgjN
UQ/1
CNorlnated Phenols
2.2
143
13
BK
BK
Chlorinated Resin Adds
0.49
49
BK
NG
CNorlnated Gualacds
5.4
110
10
BK
C(D)E(O)H/D
CNorlnated Catechols
11
236
BK
NG
CNorinatedVanlllns
4.6
36.6
BK
CEH
CD
CO
'Treated effluent concentration does not account for dilution In stn
Hill process abbreviations used:
BK - Bleached Kraft
C - Chlorine
0 - Chlorine Otoxide
E - Extraction with alkali
K - Kraft
0 - Extraction with oxygen used as
•inor with E.
clowest value given here, all values listed in Table B-7.
d( ) - mwber of LCSO values given in Table B-7.
eC«?are» to DWBL of 1050 ug/L for this compound.
BS - Bleached Sulfite
D - Chlorine Dioxide
(0)- Chlroine dioxide used as minor method with C
H - Hypochlorite
NG - Not given
-------
ft V*. ^Xb\.\*t\c*A ta»Ay*\s ot I«VOT«\ Changes \n Chea\ca\ Concentrations In Pulp MiU Effluent"
Che* teal tame
3.4. 5-Trichlorocatechol
Tetrachlorocatechol
Dichlorodihydroafaietic acid
3.4.5-TrichlorogiMtacol
4.5-Dichlorocatechol
Tetrachlorogua lacol
2.4.6-Trtchlorophenol
2.4-01chlorophenol
4.5.6-THchloroguaiacol
3.4.6-Trichtorocatechol
14-Chlorodeliydraabietic acid
4 . 5-0 ich lorogua iaco 1
2.3.4.6-Tetrachlorophenol
Pentad) loropheno 1
3 . 4-0 ichlorocatecho 1
3.6-Dichlorocatechol
2.4.5-THchlorophenol
2 . 6-0 icti loropheno 1
n"
79
79
78
79
79
79
79
79
81
79
75
79
79
79
53
79
79
10
Average
concentration
(«/L)c
107.210
75.717
42.299
29.559
28.406
23.824
22.453
10.924
10.139
8.564
8.365
5.686
5.501
1.749
1.700
1.517
1.504
0.789
Variance
727.769
321.161
269.196
102.192
131.607
33.614
10.891
4.7S4
10.918
6.408
20.313
29.661
2.650
0.138
0.982
0.205
0.160
0.161
Standard
deviation
28.977
17.921
16.407
10.109
11.472
5.798
3.330
2.180
3.304
2.531
4.507
5.446
1.628
0.371
0.991
0.453
0.400
0.401
MiniMM
concent rat ion
(K9/O
53.9
38.3
IS
12.6
4.4
13
14
7
4.1
3.7
1
0.3
2.3
0.8
0.4
0.5
0.5
O.S
MaxiM
concentration
ta/L)
180
119
as
53.4
52.3
44.2
31.5
17.7
18
13.8
25
18
9
2.7
4.1
2.5
2.8
1.9
* Effluent sables were collected every other day during a NCASI study (NCASI technical bulletin 1474. Noveafcer 1985) on cold
water streaa productivity.
" Nuriier of aaspics collected during the ten-month study.
c Average concentration («/L) listed highest to lowest.
-------
Table B-ll. Regulatory Schedule for OCOs in Soae
Canadian Provinces and European Countries'
Location
West Germany
Finland
Sweden
Swden
Sweden
Ontario0
British Colusbiac
British Columbia*
Quebec0
* Source: Clodey (1989).
b KilograjH of TOC1 or AOX
TOC1 or AOXb
Discharge
(kg/t)
1.0
2.5
1.3-5.5
1.5
0.1
2.5 (AOX)
2.5 (AOX)
1.5 (AOX)
1.5 (AOX)
per ton of air dried pulp produced..
Target
date
1989
1994
1988
1992
ca. 2010
1991
1991
1994
1993
0 Regulations pending.
B-36
-------
Figure B-l.
10th% Low Flow Stream Cone vs. Eff Cone
Othvr Chlorinated Chomleal 3tocharg««
tOO 160 200
Treat* EMiMnt Concentration (ug/1)
Figure B-2.
50th% Low Flow Stream Cone vs Eff. Cone
Othw Chlorinate* Chomloal Ohohorg»«
250
280
-------
Figure B-3.
Average LADDs at 10th % Mean Flow
Chlorinated Chamiaal Dlwhargw
Itaatod Effluant ConawitrvtlMi
Figure B-4.
Average LADDs at 50th % Mean Flow
OBwr Chlorinated Chamfea)
-------
Figure B-5"
Temporal Changes in 3,4,5-Trichlorocatechol Cone
DO
I
CO
VO
in
o
b>
o
o
E
180
11111111111111111ii11111ii11n1111 111111111111 ii 111111111111 11 11111111111111
September 1983 - June 1984
-------
Figure B-
Temporal Changes in Pentachlorophenol Cone
CO
O
o>
o
o>
o
b
'§
11 n n in n i u m iu i n n i! 11 n n n n i n i n n n n \*\ \\\ \\ \\ \\\\ \\ \\\\\\\ \\M\\\
S 0
N
D J F M A
September 1983 - June 1984
-------
REPORT DOCUMENTATION '• «PO"T NO z-
PAGE £PA 560/5-90-014
4. Titl« and Subntlt
Background Document to the Integrated Risk Assessment for
Dioxins and Furans from Chlorine Bleaching in Pulp and Paper
MilU
7. AU«,or<,) Greg schweer, Ben Gregg, Lee Schultz, Patricia Wood,
Timothy Leiohton, Carl D'Ruiz. Robert Fares. Geoffrey Huse,
9. Ptrtgrmirtf Organisation Ntm« and AddrMt Cl dy CdrpentSr, J 3(1)65 KOflZ ,
Yersar Inc. Daniel Arrenholz
6850 Yersar Center
Springfield, Virginia 22151
12. Sctcntanng Organisation Nam* and Addrtii
United States Environmental Protection Agency
Office of Toxic Substances
Exposure Evaluation Division
_ Washington, D.C. 20460 .
3. R«cipi*nf« Acc«l»ion No
S. Rtoort oil*
7/90
6.
1. Pcrformini Or(»nintion R»o1. No
la Proitct/Taik/Work Unit No.
Task 34
II. ContrictfC) or Grknt(G) No.
cc>68-D9-0166
(G)
Final Report
14.
15. Suc.fj|»m«nl»rv
EPA Project Officer was Thomas Murray
EPA Task Manager was Pat Jennings
1ft. Abtlract (Limit: 100 word*)
This report presents the detailed summaries of findings, assumptions, and uncertain-
ties of an assessment of risks from exposure of humans, terrestrial and av an wild-
life, and aquatic life to dioxins and furans formed during chlorine bleaching at
kraft and sulfite pulp and paper mills. This report contains condensed versions of
eight major exposure/risk assessments and other support documents P^ePar^Q^ PJ°9rar
offices within the U.S. Environmental Protection Agency (EPAJ, the U.S. Food and
Druo Administration (FDA), and the U.S. Consumer Product Safety Commission (CPSC).
was coordinated by the Federal Interagency Working Group in Dioxin-in-Paper.
Oecurncni Analyni a. 0»«nptoft
Di'oxins/Furans
c. COSATI F!«ld/CroitB
li. Awai|«btlitr SUMmvnt
20. Saeurity Clm tThit
U
<*«« ANSI-U«.H)
In Inttruttion* an tttvcra*
OPTIONAL rOBM 272 (4-77)
(Form«rly NTIS-15)
Department of Cwnmtrc*
------- |