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*

-------