EPA/600/P-03/002F
November 2006
An Inventory of Sources and Environmental Releases of
Dioxin-Like Compounds in the United States for
the Years 1987,1995, and 2000
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
ABSTRACT
The purpose of this document is to present a comprehensive inventory and overview of
sources and environmental releases of dioxin-like compounds in the United States. The major
identified sources of environmental releases of dioxin-like compounds are grouped into six broad
categories: combustion sources, metals smelting, refining and process sources, chemical
manufacturing sources, natural sources, and environmental reservoirs. Estimates of annual
releases to land, air, and water are presented for each source category and summarized for
reference years 1987, 1995, and 2000. The quantitative results are expressed in terms of the
toxicity equivalence (TEQ) of the mixture of polychlorinated dibenzo-p-dioxin (CDD) and
polychlorinated dibenzofuran (CDF) compounds present in environmental releases using a
procedure sanctioned by the World Health Organization (WHO) in 1998. This TEQ procedure
translates the complex mixture of CDDs and CDFs characteristic of environmental releases into
an equivalent toxicity concentration of 2,3,7,8-tetrachorodibenzo-p-dioxin (2,3,7,8-TCDD), the
most toxic member of this class of compounds. Using this WHO procedure, the annual releases
of TEQDF-WHO98 to the U.S. environment over the three reference years are 13,965 g in 1987,
3,444 g in 1995, and 1,422 g in 2000. This analysis indicates that between reference years 1987
and 2000, there was approximately a 90% reduction in the releases of dioxin-like compounds to
the circulating environment of the United States from all known sources combined. In 1987 and
1995, the leading source of dioxin emissions to the U.S. environment was municipal waste
combustion; however, because of reductions in dioxin emissions from municipal waste
combustors, it dropped to the fourth ranked source in 2000. Burning of domestic refuse in
backyard burn barrels remained fairly constant over the years, but in 2000, it emerged as the
largest source of dioxin emissions to the U.S. environment.
Preferred Citation:
U. S. EPA (Environmental Protection Agency). (2006) An inventory of sources and environmental releases of dioxin-
like compounds in the United States for the years 1987, 1995, and 2000. National Center for Environmental
Assessment, Washington, DC; EPA/600/P-03/002F. Available from: National Technical Information Service,
Springfield, VA, and online at http://epa.gov/ncea.
11
-------
CONTENTS
LIST OF TABLES xii
LIST OF FIGURES xxv
LIST OF ABBREVIATIONS AND ACRONYMS xxx
FOREWORD xxxiv
PREFACE xxxvi
AUTHORS, CONTRIBUTORS, AND REVIEWERS xxxvii
ACKNOWLEDGMENT xli
EXECUTIVE SUMMARY xlii
1. BACKGROUND, APPROACH, AND CONCLUSIONS 1-1
1.1. BACKGROUND 1-1
1.1.1. Reference Years 1-3
1.1.2. Regulatory Summary 1-3
1.1.3. Definition of Dioxin-Like Compound 1-4
1.1.4. Toxicity Equivalence Factors 1-8
1.1.5. Information Sources 1-11
1.2. APPROACH 1-12
1.2.1. Source Classes 1-12
1.2.2. Quantitative Method for Inventory of Sources 1-23
1.2.3. Confidence Ratings 1-27
1.2.3.1. Rating Scheme 1-31
1.3. CONCLUSIONS 1-37
1.3.1. Total Environmental Releases 1-37
1.3.2. Time Trends 1-42
1.3.3. Sources Not Included in the Inventory 1-42
1.3.4. Formation Theory 1-43
1.3.5. Congener Profiles of CDD/CDF Sources 1-44
2. MECHANISMS OF FORMATION OF DIOXIN-LIKE COMPOUNDS DURING
COMBUSTION OF ORGANIC MATERIALS 2-1
2.1. MECHANISM 1 (PASS THROUGH): CDD/CDF CONTAMINATION IN
FUEL AS A SOURCE OF COMBUSTION STACK EMISSIONS 2-3
2.2. MECHANISM 2 (PRECURSOR): FORMATION OF CDDs/CDFs FROM
PRECURSOR COMPOUNDS 2-4
2.3. MECHANISM 3 (DENOVO SYNTHESIS): SYNTHESIS OF CDDs/CDFs
DURING COMBUSTION OF ORGANIC MATERIALS 2-12
2.4. THE ROLE OF CHLORINE IN THE FORMATION OF CDDs/CDFs IN
COMBUSTION SYSTEMS 2-20
2.4.1. Review of Laboratory-Scale Studies 2-21
2.4.2. Review of Full-Scale Combustion Systems 2-26
2.5. POTENTIAL PREVENTION OF CDD/CDF FORMATION IN
COMBUSTION SYSTEMS 2-28
THEORY ON THE EMISSION OF PCBs 2-29
in
-------
CONTENTS (continued)
2.7. SUMMARY AND CONCLUSIONS 2-31
2.7.1. Mechanisms of Formation of Dioxin-Like Compounds 2-31
2.7.2. Role of Chlorine 2-33
2.7.3. General Conclusion 2-34
3. COMBUSTION SOURCES OF CDDs/CDFs: WASTE INCINERATION 3-1
3.1. MUNICIPAL WASTE COMBUSTION 3-1
3.1.1. Description of Municipal Waste Combustion Technologies 3-2
3.1.1.1. Furnace Types 3-2
3.1.1.2. Air Pollution Control Devices 3-9
3.1.1.3. Classification Scheme 3-10
3.1.2. Characterization of MWCs in Reference Years 2000, 1995, and 1987 ... 3-11
3.1.3. Estimation of CDD/CDF Emissions from MWCs 3-19
3.1.3.1. Estimating CDD/CDF Emissions from MWCs in Reference
Year 2000 3-19
3.1.3.2. Estimating CDD/CDF Emissions from MWCs in Reference
Years 1995 and 1987 3-21
3.1.4. Summary of CDD/CDF (TEQ) Emissions from MWCs for 2000, 1995,
and 1987 3-23
3.1.5. Congener Profiles of Municipal Waste Combustion Facilities 3-31
3.1.6. Estimated CDDs/CDFs in MWC Ash 3-31
3.1.7. Recent EPA Regulatory Activities 3-40
3.2. HAZARDOUS WASTE INCINERATION 3-41
3.2.1. Furnace Designs forHWIs 3-42
3.2.2. APCDs forHWIs 3-44
3.2.3. Estimation of CDD/CDF Emission Factors forHWIs 3-45
3.2.4. Emission Estimates for HWIs 3-47
3.2.5. Recent EPA Regulatory Activities 3-51
3.2.6. Industrial Boilers and Furnaces Burning Hazardous Waste 3-51
3.2.7. Halogen Acid Furnaces Burning Hazardous Waste 3-53
3.2.8. Solid Waste from Hazardous Waste Combustion 3-55
3.3. MEDICAL WASTE INCINERATION 3-55
3.3.1. Design Types of MWIs Operating in the United States 3-55
3.3.2. Characterization of MWIs for Reference Years 1987, 1995, and 2000 . . . 3-57
3.3.3. Estimation of CDD/CDF Emissions from MWIs 3-59
3.3.4. Summary of CDD/CDF Emissions from MWIs 3-64
3.3.5. Recent EPA Regulatory Activities 3-65
3.4. CREMATORIA 3-65
3.4.1. Human Crematoria 3-65
3.4.1.1. Emissions Data 3-65
3.4.1.2. Activity Level Information 3-73
3.4.1.3. Emission Estimates 3-76
3.4.2. Animal Crematoria 3-76
IV
-------
CONTENTS (continued)
3.4.2.1. Emissions Data 3-76
3.4.2.2. Activity Level Information 3-76
3.4.2.3. Emission Estimates 3-76
3.5. SEWAGE SLUDGE INCINERATION 3-77
3.5.1. Emissions Estimates from Sewage Sludge Incinerators 3-79
3.5.2. Solid Waste from Sewage Sludge Incinerators 3-83
3.6. TIRE COMBUSTION 3-83
3.7. COMBUSTION OF WASTEWATER SLUDGE AT BLEACHED
CHEMICAL PULP MILLS 3-87
3.8. BIOGAS COMBUSTION 3-87
4. COMBUSTION SOURCES OF CDDs/CDFs: POWER/ENERGY GENERATION ... 4-1
4.1. MOTOR VEHICLE FUEL COMBUSTION 4-1
4.1.1. Tailpipe Emission Studies 4-1
4.1.2. Tunnel Emission Studies 4-20
4.1.3. National Emission Estimates 4-28
4.1.3.1. Activity Information for On-Road Vehicles 4-28
4.1.3.2. Activity Information for Off-Road Uses 4-31
4.1.3.3 Emission Estimates 4-32
4.2. WOOD COMBUSTION 4-38
4.2.1. Flue Emissions from Wood Combustion (Residential) 4-38
4.2.1.1. Emissions Data 4-38
4.2.1.2. Activity Level Information 4-41
4.2.1.3. Emission Estimates 4-44
4.2.2. Stack Emissions from Wood Combustion (Industrial) 4-44
4.2.2.1. Emissions Data 4-44
4.2.2.2. Activity Level Information 4-50
4.2.2.3. Emission Estimates 4-51
4.2.3. Solid Waste from Wood Combustion (Residential and Industrial) 4-52
4.3. OIL COMBUSTION 4-60
4.3.1. Institutional/Commercial and Residential Oil Combustion 4-61
4.3.2. Utility Sector and Industrial Oil Combustion 4-64
4.3.3. Used Oil Combustion 4-67
4.4. COAL COMBUSTION 4-67
4.4.1. Utilities and Industrial Boilers 4-68
4.4.2. Residential Coal Combustion 4-74
4.4.3. Solid Wastes from Coal Combustion 4-77
5. COMBUSTION SOURCES OF CDDs/CDFs: OTHER HIGH-TEMPERATURE
SOURCES 5-1
5.1. CEMENT KILNS 5-1
5.1.1. Process Description of Portland Cement Kilns 5-1
5.1.2. Cement Kilns That Burn Hazardous Waste 5-2
-------
CONTENTS (continued)
5.1.3. Air Pollution Control Devices 5-3
5.1.4. CDD/CDF Emissions Data 5-4
5.1.4.1. Emissions Data for 1989 through 1996 5-4
5.1.4.2. Emissions Data for 1999 and 2000 5-7
5.1.4.3. Emission Factor Estimates for Cement Kilns Burning Hazardous
Waste 5-7
5.1.4.4. Emission Factor Estimates for Cement Kilns Burning
Nonhazardous Waste 5-9
5.1.4.5. Confidence Ratings of Emission Factor Estimates 5-10
5.1.5. Activity Level Information 5-10
5.1.6. National CDD/CDF Emission Estimates 5-14
5.1.6.1. Estimates for Reference Years 1987 and 1995 5-14
5.1.6.2. Estimates for Reference Year 2000 5-15
5.1.7. EPA Regulatory Activities 5-16
5.1.8. Solid Waste from Cement Manufacturing: Cement Kiln Dust 5-16
5.2. LIGHTWEIGHT AGGREGATE KILNS 5-21
5.3. ASPHALT MIXING PLANTS 5-22
5.4. PETROLEUM REFINING CATALYST REGENERATION 5-24
5.5. CIGARETTE SMOKING 5-28
5.6. PYROLYSIS OF BROMINATED FLAME RETARDANTS 5-39
5.7. CARBON REACTIVATION FURNACES 5-39
5.8. KRAFT BLACK LIQUOR RECOVERY BOILERS 5-42
5.9. OTHER IDENTIFIED SOURCES 5-46
6. COMBUSTION SOURCES OF CDDs/CDFs: MINIMALLY CONTROLLED
AND UNCONTROLLED COMBUSTION SOURCES 6-1
6.1. COMBUSTION OF LANDFILL GAS 6-1
6.1.1. Emissions Data 6-1
6.1.2. Activity Level Information 6-1
6.1.2.1. Activity Levels for 1987 and 1995 6-3
6.1.2.2. Activity Level for 2000 6-4
6.1.3. Emission Estimates 6-4
6.2. ACCIDENTAL FIRES 6-5
6.2.1. Soot and Ash Studies 6-5
6.2.2. Fume and Smoke Studies 6-7
6.2.3. Data Evaluation 6-8
6.2.3.1. Structural Fires 6-8
6.2.3.1.1. Emissions data 6-8
6.2.3.1.2. Activity level information 6-9
6.2.3.1.3. Emission estimates 6-9
6.2.3.2. Vehicle Fires 6-10
6.3. LANDFILL FIRES 6-11
6.3.1. Emissions Data 6-11
VI
-------
CONTENTS (continued)
6.3.2. Activity Level Information and Emission Estimates 6-12
6.4. FOREST AND BRUSH FIRES 6-12
6.4.1. Emissions Data 6-12
6.4.2. Activity Level Information 6-18
6.4.2.1. Approach for Reference Year 2000 (Office of Air Quality
Planning and Standards [OAQPS]) 6-18
6.4.2.2. Approach for Reference Years 1987 and 1995 6-18
6.4.3. Emission Estimates 6-19
6.5. BACKYARD BARREL BURNING 6-20
6.5.1. Emissions Data 6-20
6.5.2. Activity Level Information 6-21
6.5.2.1. Summary of Barrel Burn Surveys 6-23
6.5.2.2. Estimates of Activity Level 6-24
6.5.2.3. Alternative Approach to Estimating Activity Level 6-25
6.5.3. Emission Estimates 6-26
6.5.4. Composition of Ash from Barrel Burning 6-26
6.6. RESIDENTIAL YARD WASTE BURNING 6-26
6.6.1. Emissions Data 6-26
6.6.2. Activity Level Information 6-28
6.6.3. Emission Estimates 6-29
6.7. LAND-CLEARING DEBRIS BURNING 6-30
6.7.1. Emissions Data 6-30
6.7.2. Activity Level Information 6-30
6.7.2.1. Residential Construction 6-30
6.7.2.2. Nonresidential Construction 6-30
6.7.2.3. Roadway Construction 6-31
6.7.2.4. Fuel Loading Factors 6-31
6.7.3. Emission Estimates 6-32
6.8. UNCONTROLLED COMBUSTION OF POLYCHLORINATED
BIPHENYLS 6-32
6.9. VOLCANOES 6-33
6.10. FIREWORKS 6-34
6.11. OPEN BURNING AND OPEN DETONATION OF ENERGETIC
MATERIALS 6-36
7. METAL SMELTING AND REFINING SOURCES OF CDDs/CDFs 7-1
7.1. PRIMARY NONFERROUS METAL SMELTING/REFINING 7-1
7.1.1. Primary Copper Smelting and Refining 7-1
7.1.2. Primary Magnesium Smelting and Refining 7-4
7.1.3. Primary Nickel Smelting and Refining 7-6
7.1.4. Primary Aluminum Smelting and Refining 7-6
7.1.5. Primary Titanium Smelting and Refining 7-7
7.2. SECONDARY NONFERROUS METAL SMELTING 7-8
Vll
-------
CONTENTS (continued)
7.2.1. Secondary Aluminum Smelters 7-8
7.2.2. Secondary Copper Smelters 7-14
7.2.2.1. Emissions Data 7-15
7.2.2.2. Activity Level Information 7-20
7.2.2.3. Emission Estimates 7-20
7.2.3. Secondary Lead Smelters 7-22
7.3. PRIMARY FERROUS METAL SMELTING/REFINING 7-27
7.3.1. Sinter Production 7-27
7.3.2. Coke Production 7-32
7.4. SECONDARY FERROUS METAL SMELTING/REFINING 7-33
7.5. FERROUS FOUNDRIES 7-36
7.6. SCRAP ELECTRIC WIRE RECOVERY 7-40
7.7. DRUM AND BARREL RECLAMATION FURNACES 7-45
7.8. SOLID WASTE FROM PRIMARY/SECONDARY IRON/STEEL
MILLS/FOUNDRIES 7-48
CHEMICAL MANUFACTURING AND PROCESSING SOURCES 8-1
8.1. BLEACHED CHEMICAL WOOD PULP AND PAPER MILLS 8-1
8.1.1. Estimates of National Emissions in 1987 and 1995 8-8
8.1.2. Estimates of National Emissions in 2000 8-8
8.2. MANUFACTURE OF CHLORINE, CHLORINE DERIVATIVES, AND METAL
CHLORIDES 8-9
8.2.1. Manufacture of Chlorine 8-9
8.2.2. Manufacture of Chlorine Derivatives and Metal Chlorides 8-12
8.3. MANUFACTURE OF HALOGENATED ORGANIC CHEMICALS 8-14
8.3.1. Chlorophenols 8-15
8.3.1.1. Regulatory Actions for Chlorophenols 8-16
8.3.2. Chlorobenzenes 8-24
8.3.2.1. Regulatory Actions for Chlorobenzenes 8-26
8.3.3. Chlorobiphenyls 8-27
8.3.4. Ethylene Dichloride/Vinyl Chloride Monomer/Polyvinyl Chloride
Manufacturing 8-30
8.3.4.1. Water Releases 8-33
8.3.4.2. Land Releases 8-36
8.3.4.3. Air Releases 8-36
8.3.4.4. Transfers to Secure Landfills 8-38
8.3.4.5. Products 8-41
8.3.5. Other Aliphatic Chlorine Compounds 8-44
8.3.6. Dyes, Pigments, and Printing Inks 8-45
8.3.6.1. Dioxazine Dyes and Pigments 8-45
8.3.6.2. Phthalocyanine Dyes and Printing Inks 8-48
8.3.7. TSCA Dioxin/Furan Test Rule 8-50
8.3.8. Halogenated Pesticides and FIFRA Pesticides Data Call-In 8-52
Vlll
-------
CONTENTS (continued)
8.4. OTHER CHEMICAL MANUFACTURING AND PROCESSING
SOURCES 8-73
8.4.1. Municipal Wastewater Treatment Plants 8-73
8.4.1.1. Sources 8-73
8.4.1.2. Releases to Water 8-77
8.4.1.2.1 Emissions data 8-77
8.4.1.2.2 Activity level information 8-78
8.4.1.2.3 Emission Estimates 8-81
8.4.1.3. Sewage Sludge Land Disposal 8-81
8.4.1.3.1 Emissions data 8-81
8.4.1.3.2 Activity level information 8-88
8.4.1.3.3 Emission estimates 8-89
8.4.2. Drinking Water Treatment Plants 8-91
8.4.3. Soaps and Detergents 8-93
8.4.4. Textile Manufacturing and Dry Cleaning 8-94
9. INDICATIONS OF POSSIBLE NATURAL SOURCES OF CDDs/CDFs 9-1
9.1. BIOLOGICAL SOURCES OF CDDs/CDFs 9-1
9.1.1. Biotransformation of Chlorophenols 9-1
9.1.2. Biotransformation of Higher CDDs/CDFs 9-4
9.1.3. Dioxin-Like Compounds in Animal Manure 9-6
9.2. PHOTOTRANSFORMATION OF CHLOROPHENOLS 9-8
9.2.1. Photolysis of Higher CDDs/CDFs 9-10
9.2.2. Photolysis in Water 9-11
9.2.3. Photolysis on Soil 9-11
9.2.4. Photolysis on Vegetation 9-12
9.2.5. Photolysis in Air 9-13
9.3. CDDs/CDFs IN BALL CLAY 9-14
9.3.1. Initial Discovery of CDD/CDF Contamination of Ball Clay 9-14
9.3.2. Characteristics of Mississippi Embayment Ball Clays 9-15
9.3.3. Levels of Dioxin-Like Compounds in Ball Clay 9-15
9.3.4. Evidence for Ball Clay as a Natural Source 9-16
9.3.5. Environmental Releases of Dioxin-Like Compounds from the Mining
and Processing of Ball Clay 9-19
10. SOURCES OF DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (PCBs) 10-1
10.1. GENERAL FINDINGS OF THE EMISSIONS INVENTORY 10-1
10.2. RELEASES OF COMMERCIAL PCBs 10-4
10.2.1. Approved PCB Disposal/Destruction Methods 10-11
10.2.1.1. Approved Incinerators/High-Efficiency Boilers 10-14
10.2.1.2. Approved Chemical Waste Landfills 10-14
10.2.1.3. Other Approved Disposal Methods 10-15
10.2.2. Emission Estimates 10-15
IX
-------
CONTENTS (continued)
10.2.3. Accidental Releases of Still-in-Service PCBs 10-15
10.2.3.1. Leaks and Spills 10-16
10.2.3.2. Accidental Fires 10-18
10.2.4. Municipal Wastewater Treatment 10-19
10.3. CHEMICAL MANUFACTURING AND PROCESSING SOURCES 10-26
10.4. COMBUSTION SOURCES 10-26
10.4.1. Municipal Waste Combustors 10-26
10.4.2. Industrial Wood Combustion 10-27
10.4.3. Medical Waste Incineration 10-29
10.4.4. Tire Combustion 10-30
10.4.5. Cigarette Smoking 10-31
10.4.6. Sewage Sludge Incineration 10-31
10.4.7. Backyard Barrel Burning 10-34
10.4.8. Petroleum Refining Catalyst Regeneration 10-34
10.5. NATURAL SOURCES 10-36
10.5.1. Biotransformation of Other PCBs 10-36
10.5.2. Photochemical Transformation of Other PCBs 10-39
10.6. PAST USE OF COMMERCIAL PCBs 10-41
11. RESERVOIR SOURCES OF CDDs/CDFs AND DIOXIN-LIKE PCBs 11-1
11.1. POTENTIAL RESERVOIRS 11-1
11.2. CHARACTERIZATION OF RESERVOIR SOURCES 11-3
11.2.1. Soil 11-3
11.2.1.1. Potential Mass of Dioxin-Like Compounds Present 11-3
11.2.1.2. Mechanisms Responsible for Releases from Surface Soils .... 11-9
11.2.1.3. Estimated Annual Releases from Soil to Water 11-12
11.2.1.4. Estimated Annual Releases from Soil to Air 11-14
11.2.2. Water 11-17
11.2.2.1. Potential Mass of Dioxin-Like Compounds Present 11-17
11.2.2.2. Mechanisms Responsible for Supply to and Releases
from Water 11-17
11.2.3. Sediment 11-20
11.2.3.1. Potential Mass of Dioxin-Like Compounds Present 11-20
11.2.3.2. Mechanisms Responsible for Supply to and Releases
from Sediment 11-20
11.2.3.3. Releases from Sediment to Water 11-21
11.2.4. Biota 11-22
11.2.4.1. Potential Mass of Dioxin-Like Compounds Present 11-22
11.2.4.2. Mechanisms Responsible for Supply to and Releases
from Biota 11-22
11.2.4.3. Approaches for Measuring and Estimating Releases
from Biota 11-24
11.3. SUMMARY AND CONCLUSIONS 11-26
-------
CONTENTS (continued)
11.3.1. Reservoir Sources 11-26
11.3.2. Implications for Human Exposure 11-27
REFERENCES R-l
XI
-------
LIST OF TABLES
Table 1-1. Nomenclature for dioxin-like compounds 1-2
Table 1-2. Municipal waste combustors3 1-4
Table 1-3. Hazardous waste incinerators and cement kilns and lightweight aggregate kilns
burning hazardous waste 1-5
Table 1-4. Cement kilns not burning hazardous waste 1-5
Table 1-5. Secondary aluminum smelters 1-6
Table 1-6. Medical waste incinerators 1-6
Table 1-7. Pulp and paper mills 1-6
Table 1-8. The TEF scheme for I-TEQDF 1-9
Table 1-9. The TEF scheme for dioxin-like PCBs, as determined by the World Health
Organization in 1994 1-9
Table 1-10. The TEF scheme for TEQDFP-WHO98 1-10
Table 1-11. Known and suspected sources of CDDs/CDFs 1-14
Table 1-12. Inventory of contemporary releases (g/yr) of dioxin-like compounds from known
sources in the United States for reference years 2000, 1995, and 1987 and
preliminary release estimates for 2000 1-17
Table 1-13. Products containing CDDs/CDFs (g TEQDF-WHO98/yr) 1-23
Table 1-14. I-TEQDF emission factors used to develop national emission inventory estimates of
releases to air 1-32
Table 1-15. TEQDF-WHO98 emission factors used to develop national emission inventory
estimates of releases to air 1-34
Table 1-16. Confidence rating scheme for U.S. emission estimates 1-36
Table 1-17. Ranking of sources of dioxin-like compounds based on environmental releases
(from high to low) for reference years 2000, 1995, and 1987 1-38
Table 2-1. Concentration of CDDs/CDFs on municipal incinerator fly ash at varying
temperatures 2-7
xn
-------
LIST OF TABLES (continued)
Table 2-2. CDDs/CDFs formed from the thermolytic reaction of 690 mg benzene + FeCl3
silica complex 2-12
Table 2-3. De novo formation of CDDs/CDFs after heating Mg-Al silicate, 4% charcoal, 7%
Cl, 1% CuCl2 in H2O at 300°C 2-14
Table 3-1. Inventory of municipal waste combustors (MWCs) in 2000 by technology, air
pollution control device (APCD), sizea, and annual activity level (kg/yr) 3-14
Table 3-2. Inventory of municipal waste combustors (MWCs) in 1995 by technology, air
pollution control device (APCD), and annual activity level (kg/yr) 3-16
Table 3-3. Inventory of municipal waste combustors (MWCs) in 1987 by technology, air
pollution control device (APCD), and annual activity level (kg/yr) 3-18
Table 3-4. National average CDD/CDF congener concentrations for
large municipal waste combustors (ng/dscm @ 7% O2) 3-20
Table 3-6. CDD/CDF TEQ emission factors (ng TEQ/kg waste) for municipal solid waste
incineration 3-24
Table 3-7a. Annual I-TEQDF emissions from municipal waste combustors (MWCs) operating
in 1995 3-25
Table 3-7b. Annual TEQDF-WHO98 emissions from municipal waste combustors (MWCs)
operating in 1995 3-27
Table 3-8a. Annual I-TEQDF emissions to the air from municipal waste combustors (MWCs)
operating in 1987 3-29
Table 3-8b. Annual TEQDF-WHO98 emissions to the air from municipal waste combustors
(MWCs) operating in 1987 3-30
Table 3-9. Average and range of CDD/CDF congener groups in fly ash from a municipal
incinerator (|ig/kg) 3-36
Table 3-10. Comparison of the amount of TEQs generated annually in municipal waste
combustor ash 3-38
Table 3-11. Concentration of CDD/CDF congener groups (ng/kg) in fly ash samples from
combustion of municipal solid waste in eight Japanese incinerators 3-39
Xlll
-------
LIST OF TABLES (continued)
Table 3-12. Concentration of CDD/CDF congener groups in fly ash samples from municipal
solid waste 3-39
Table 3-13. CDD/CDF concentrations in municipal solid waste ash from a newly constructed
municipal waste combustor in Japan 3-40
Table 3-14a. CDD/CDF emission factors for hazardous waste incinerators and boilers tested
from 1993 to 1996 3-48
Table 3-14b. CDD/CDF emission factors for hazardous waste incinerators and boilers tested in
2000 3-49
Table 3-15. CDD/CDF emission factors for halogen acid furnaces tested in 2000 3-54
Table 3-16. Estimated breakdown of facilities by air pollution control device (APCD) . . . 3-60
Table 3-17. Summary of annual operating hours for each medical waste incinerator (MWI)
type 3-61
Table 3-18. TEQ emissions from medical waste incinerators (MWIs) for reference
year 1987 3-62
Table 3-19. TEQ emissions from medical waste incinerators (MWIs) for reference
year 1995 3-62
Table 3-20. TEQ emissions from medical waste incinerators (MWIs) for reference
year 2000 3-63
Table 3-21. Congener-specific profile for Camellia Memorial Lawn crematorium 3-70
Table 3-22. Congener-specific profile for the Woodlawn Cemetery crematorium 3-71
Table 3-23. Operational data for the Woodlawn Cemetery crematorium, scrubber inlet . . . 3-73
Table 3-24. Congener-specific profile for the Camellia Memorial Lawn crematorium and
the Woodlawn Cemetery crematorium 3-74
Table 3-25. Congener-specific profile for the University of Georgia Veterinary School . . . 3-77
Table 3-26. CDD/CDF emission factors for sewage sludge incinerators 3-82
Table 3-27. CDD/CDF air emission factors for a tire combustion facility 3-84
XIV
-------
LIST OF TABLES (continued)
Table 3-28. CDD/CDF emission factors for combustion of bleached-kraft mill sludge in
wood residue boilers 3-88
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste
combustors (MWCs) operating in 2000 3-90
Table 4-1. Description and results of vehicle emission testing studies for CDDs
and CDFs 4-3
Table 4-2. CDD/CDF congener emission factors (pg/L) for diesel-fueled automobiles .... 4-5
Table 4-3. CDD/CDF congener emission factors (pg/L) for diesel-fueled trucks 4-7
Table 4-4. CDD/CDF congener emission factors (pg/L) for leaded gasoline-fueled
automobiles 4-9
Table 4-5. CDD/CDF congener emission factors (pg/L) for unleaded gasoline-fueled
automobiles without catalytic converters 4-11
Table 4-6. CDD/CDF congener emission factors (pg/L) for unleaded gasoline-fueled
automobiles with catalytic converters 4-13
Table 4-7. Total dioxin emission concentrations from heavy-duty diesel engines
in Germany 4-21
Table 4-8. Levels of 2,3,7,8-chlorine-substituted congeners and total CDDs/CDFs in
vehicle exhaust particles (pg/g) for gasoline engines and suspended
particulate matter (SPM) 4-22
Table 4-9. Levels of 2,3,7,8-chlorine-substituted congeners and total CDDs/CDFs in
vehicle exhaust particles (pg/g) for diesel engines 4-23
Table 4-10. CDD/CDF concentrations (pg/m3) from European tunnel air studies 4-24
Table 4-11. Baltimore Harbor tunnel study: estimated emission factors (pg/km) for heavy-
duty diesel trucks 4-26
Table 4-12. Average CDD/CDF concentration in flue gas while burning wood and
crops 4-40
Table 4-13. CDD/CDF concentrations (pg TEQ/kg wood) in emissions from residential
wood stoves in Canada 4-42
XV
-------
LIST OF TABLES (continued)
Table 4-14. CDD/CDF mean emission factors (ng/kg wood) for industrial wood
combustors
Table 4-15.
Table 4-16.
Table 4-17.
Table 4-18.
Table 4-19.
Table 4-20.
Table 4-21.
Table 4-22.
Table 4-23.
Table 4-24.
Table 4-25.
Table 4-26.
Table 4-27.
Table 5-1.
Table 5-2.
NCASI CDD/CDF TEQ concentrations and emissions for wood
residue-fired boilers
4-45
4-50
CDD/CDF concentrations (ng/kg) in residential chimney soot from
wood stoves and fireplaces ........................................ 4-53
CDD/CDF concentrations (ng/kg) in bottom ash from residential wood stoves
and fireplaces [[[ 4-54
CDD/CDF concentrations in chimney soot (Bavaria, Germany) ............ 4-56
CDD/CDF concentrations (ng/kg) in fly ash from wood-working industry . . . 4-57
CDD/CDF concentrations (ng/kg) in electrostatic precipitator waste ash from
wood-fired industrial boiler ........................................ 4-59
Estimated CDD/CDF emission factors for oil-fired residential furnaces ..... 4-62
CDD/CDF emission factors (pg/L oil) for oil-fired utility/industrial boilers . . 4-65
CDD/CDF concentrations (pg/Nm3) in stack emissions from U.S. coal-fired
power plants [[[ 4-70
Characteristics of U.S. coal-fired power plants tested by the U.S. Department
of Energy [[[ 4-71
CDD/CDF emission factors (ng/kg coal) for coal-fired utility/industrial
power plants [[[ 4-72
CDD/CDF emission factors (ng/kg coal) for residential coal combustors .... 4-75
CDD/CDF concentrations and amounts disposed of in solid wastes from coal-
fired utilities [[[ 4-79
CDD/CDF emission factors (ng/kg clinker produced) for cement kilns
burning hazardous waste for reference years 1987 and 1995
-------
LIST OF TABLES (continued)
Table 5-3. CDD/CDF emission factors (ng/kg clinker produced) for cement kilns not
burning hazardous waste for reference years 1987, 1995, and 2000 5-12
Table 5-4. National emission estimates for cement kilns for reference years
1987 and 1995 5-15
Table 5-5. National emission estimates for cement kilns for reference year 2000 5-17
Table 5-6. CDD/CDF concentrations (ng/kg) in ash samples from cement kiln
electrostatic precipitator and lightweight aggregate (LWA) kiln fabric filter . . 5-20
Table 5-7. CDD/CDF estimates in cement kiln dust (CKD) for reference years 1987,
1995, and 2000 5-21
Table 5-8. Congener-specific emission factors (ng/kg asphalt produced) derived from the
testing of air emissions at two hot-mix asphalt plants in the United States .... 5-24
Table 5-9. CDD/CDF emission factors (ng/barrel)a for petroleum catalytic reforming
units 5-29
Table 5-10. CDD concentrations in Japanese cigarettes, smoke, and ash 5-32
Table 5-11. CDD/CDF concentrations (pg/cig) in cigarette smoke, normalized to a per-
cigarette basis 5-34
Table 5-12. CDD/CDF concentrations (pg/pack) in tobacco cigarette brands from various
countries 5-37
Table 5-13. CDD/CDF mean emission factors (ng/kg feed) for black liquor
recovery boilers 5-44
Table 5-14. CDD/CDF TEQ emission factors and emission estimates from Kraft recovery
furnaces and Kraft lime kilns 5-47
Table 5-15. CDD/CDF concentrations in candle materials and emissions 5-47
Table 6-1. CDD/CDF emission factors for a landfill flare 6-2
Table 6-2 CDD/CDF mean emission factors (ng/kg) for forest fires 6-16
Table 6-3. Forest fire fuel loading factors (tons/acre) 6-19
XVll
-------
LIST OF TABLES (continued)
Table 6-4. CDD/CDF average air emission factors (ng/kg waste burned) from barrel
burning of household waste 6-22
Table 6-5. CDD/CDF analysis for composite ash samples from barrel
burning (ng/kg ash) 6-27
Table 6-6. PCB analysis for composite ash samples from barrel burning (ng/kg ash) .... 6-28
Table 6-7. CDDs/CDFs in dust fall and ashes from volcanoes 6-34
Table 6-8. Residue of HpCDD/HpCDF and OCDD/OCDF (ng/kg) in paper cartridges
and charges of select pyrotechnic products 6-35
Table 7-1. CDD/CDF emission concentrations (pg TEQ/m3 @ 11% oxygen) for primary
copper smelters 7-2
Table 7-2. CDD/CDF emissions data from primary and secondary copper and secondary
lead smelters 7-3
Table 7-3. CDD/CDF emission factors (ng/kg scrap feed) for secondary aluminum
smelters for 2000
Table 7-7.
Table 7-8.
Table 7-9.
Table 7-10.
Table 7-11.
Table 7-4. CDD/CDF emission factors (ng/kg scrap feed) for secondary aluminum
smelters for 1995 and 1987
7-11
7-12
Table 7-5. CDD/CDF mean emission factors (ng/kg scrap feed) for secondary copper
smelters 7-17
Table 7-6. CDD/CDF emission factors (ng/kg lead produced) for secondary
lead smelters
7-25
Estimated annual TEQ emissions (g TEQ) 7-27
CDD/CDF emission factors (ng/kg sinter) for sintering plants 7-30
Operating parameters for U.S. iron ore sintering plants 7-32
CDD/CDF emission estimates for Canadian coke oven facilities, blast furnace
facilities, and electric arc furnaces 7-34
CDD/CDF emission concentrations and rates for Canadian electric arc
furnaces 7-37
XVlll
-------
LIST OF TABLES (continued)
Table 7-12. CDD/CDF emission factors for a U.S. ferrous foundry 7-39
Table 7-13. Congener-specific profile for ferrous foundries 7-40
Table 7-14. CDD/CDF emission factors for a scrap wire incinerator 7-42
Table 7-15. CDD/CDF concentrations in fly ash and ash/soil at metal recovery sites 7-44
Table 7-16. CDD/CDF emission factors for a drum and barrel reclamation facility 7-46
Table 8-1. CDD/CDF concentrations in pulp and paper mill bleached pulp, wastewater
sludge, and wastewater effluent (circa 1988) 8-2
Table 8-2. CDD/CDF concentrations in pulp and paper mill bleached pulp, wastewater
sludge, and wastewater effluent (mid-1990s) 8-5
Table 8-3. Summary of bleached chemical pulp and paper mill discharges (g/yr) of
2,3,7,8-TCDD and 2,3,7,8-TCDF 8-7
Table 8-4. CDD/CDF TEQ concentrations and emissions for the paper and pulp industry by
source 8-10
Table 8-5. CDD/CDF concentrations (ng/kg) in graphite electrode sludge from chlorine
production 8-11
Table 8-6. Releases of dioxin-like compounds in wastewater discharges from chlor-alkali and
mixed chemical manufacturing facilities to surface water in reference years 2000
and 1995 8-13
Table 8-7. Congener-specific and TEQ annual releases to air (g/yr) from chlor-alkali and
mixed chemical production facilities in 2000 8-14
Table 8-8. CDD/CDF concentrations (|ig/kg) in metal chlorides 8-15
Table 8-9. CDD/CDF concentrations (mg/kg) in mono- through tetrachlorophenols .... 8-17
Table 8-10. CDD/CDF concentrations (historical and current) (|ig/kg) in technical-grade
pentachlorophenol (PCP) products 8-18
Table 8-11. Historical CDD/CDF concentrations (|ig/kg) in pentachlorophenol-Na
(PCP-Na) 8-20
XIX
-------
LIST OF TABLES (continued)
Table 8-12. Summary of specific dioxin-containing wastes that must comply with land
disposal restrictions 8-21
Table 8-13. CDD/CDF concentrations (|ig/kg) in chlorobenzenes 8-25
Table 8-14. Concentrations of CDD/CDF congener groups in unused commercial
polychlorinated biphenyl (PCB) mixtures (mg/kg) 8-28
Table 8-15. 2,3,7,8-Substituted congener concentrations in unused polychlorinated biphenyl
(PCB) mixtures (jig/kg) 8-29
Table 8-16. Reported CDD/CDF concentrations (ng/kg) in wastes from polyvinyl chloride
(PVC) manufacture 8-32
Table 8-17. Releases of dioxin-like compounds (g/yr) in wastewater discharges from
EDC/VCM/PVC and integrated chlorine chemical manufacturing facilities to
surface water in reference years 2000 and 1995 8-34
Table 8-18. Congener-specific releases to land from an EDC/VCM/PVC integrated chemical
manufacturing facilitya in reference years 2000 and 1995 8-37
Table 8-19. Congener-specific and TEQ releases to air (g/yr) from EDC/VCM/PVC integrated
chemical manufacturing facilities in reference years 2000 and 1995 8-39
Table 8-20. Congener-specific and TEQ transfers to secure landfills (g/yr)a from
EDC/VCM/PVC integrated chemical production facilities in 2000 8-40
Table 8-21. CDD/CDF concentrations in products from U.S. EDC/VCM/PVC
manufacturers 8-42
Table 8-22. CDD/CDF concentrations (|ig/kg) in samples of dioxazine dyes and pigments
(Canada) 8-46
Table 8-23. CDD/CDF concentrations (ng/kg) in printing inks (Germany) 8-49
Table 8-24. Chemicals requiring Toxic Substances Control Act Section 4 testing under the
dioxin/furan rule 8-51
Table 8-25. Congeners and limits of quantitation (LOQs) for which quantitation is required
under the dioxin/furan test rule and pesticide Data Call-in 8-52
Table 8-26. Precursor chemicals subject to reporting requirements under Toxic Substances
Control Act Section 8(a) 8-53
XX
-------
LIST OF TABLES (continued)
Table 8-27. Results of analytical testing for dioxins and furans in the chemicals tested to date
under Section 4 of the dioxin/furan test rule 8-54
Table 8-28. CDD/CDF concentrations (jig/kg) in chloranil and carbazole violet samples
analyzed pursuant to the EPA dioxin/furan test rule 8-55
Table 8-29. Status of first pesticide Data Call-in: pesticides suspected of having the potential
to become contaminated with dioxins if synthesized under conditions favoring
dioxin formation 8-62
Table 8-30. Status of second pesticide Data Call-in: pesticides suspected of being
contaminated with dioxins 8-67
Table 8-31. Summary of analytical data submitted to EPA in response to pesticide Data Call-
In^) 8-71
Table 8-32. Summary of results for CDDs/CDFs in technical 2,4-D and 2,4-D ester
herbicides 8-72
Table 8-33. CDD/CDF concentrations (jig/kg) in samples of 2,4-D and pesticide formulations
containing 2,4-D 8-74
Table 8-34. Mean CDD/CDF concentrations and range of detection limits (DLs) (pg/L) in
effluents from nine U.S. publicly owned treatment works (POTWs) 8-79
Table 8-35. Effluent concentrations (pg/L) of CDDs/CDFs from publicly owned treatment
works in Mississippi 8-80
Table 8-36. CDD/CDF concentrations (ng/kg) measured in the 1998/1999 National Sewage
Sludge Survey 8-82
Table 8-37. CDD/CDF concentrations (ng/kg) measured in 99 sludges collected from U.S.
publicly owned treatment works (POTWs) during 1994 8-84
Table 8-38. Concentrations of CDDs/CDFs (ng/kg dry matter) in sewage sludge from publicly
owned treatment works in Mississippi 8-85
Table 8-39. CDD/CDF median concentrations (ng/kg) measured in 1999 from a publicly
owned treatment works facility in Ohio 8-86
Table 8-40. CDD/CDF mean concentrations (ng/kg) measured in the 2001 National Sewage
Sludge Survey 8-87
XXI
-------
LIST OF TABLES (continued)
Table 8-41. Quantity of sewage sludge disposed of and potential dioxin TEQ releases from
primary, secondary, and advanced treatment publicly owned treatment works for
reference year 1987 8-90
Table 8-42. Quantity of sewage sludge disposed of and potential dioxin TEQ releases from
primary, secondary, and advanced treatment publicly owned treatment works for
reference year 1995 8-91
Table 8-43. Quantity of sewage sludge disposed of and potential dioxin TEQ releases from
primary, secondary, and advanced treatment publicly owned treatment works for
reference year 2000 8-92
Table 8-44. Biosolids disposal practices for reference year 2000 8-92
Table 8-45. CDD/CDF concentrations in Swedish liquid soap, tall oil, and tall resin 8-95
Table 9-1. Estimated quantity of animal manure produced in the United States in 2000 . . . 9-7
Table 9-2. CDD and CDF concentrations (ng/kg dry weight) in samples of animal manure in
the United Kingdom 9-8
Table 9-3. Concentrations of CDDs (pg/g, dry weight) in eight ball clay samples in the United
States 9-16
Table 9-4. Comparison of the mean CDD/CDF congener group concentrations in ball clay
with those in urban and rural soils in North America (pg/g, dry weight) 9-17
Table 10-1. Confidence rating classes for 2000 for releases from all known and suspected
source categories of dioxin-like PCBs 10-2
Table 10-2. Inventory of contemporary releases of dioxin-like PCBs in the United States for
1987, 1995, and 2000 and preliminary release estimates of dioxin-like PCBs for
2000 (g TEQp-WHO98/yr) 10-3
Table 10-3. Weight percent concentrations of dioxin-like PCBs in Aroclors, Clophens, and
Kanechlors 10-7
Table 10-4. Disposal requirements for PCBs and PCB items 10-12
Table 10-5. Off-site transfers of PCBs reported in the Toxics Release Inventory (TRI)
(1988-2000) 10-16
XXll
-------
LIST OF TABLES (continued)
Table 10-6. Releases of PCBs reported in the Toxics Release Inventory (TRI)
(1988-2000) 10-17
Table 10-7. Aroclor concentrations (ng/kg) measured in EPA's National Sewage
Sludge Survey 10-20
Table 10-8. Dioxin-like PCB concentrations measured in sludges collected from 74 U.S.
publicly owned treatment works (POTWs) during 1994 10-21
Table 10-9. Dioxin-like PCB concentrations in sewage sludge collected from U.S. publicly
owned treatment works during 1999 10-22
Table 10-10. Quantity of sewage sludge disposed of annually in 1989 by primary, secondary, or
advanced treatment publicly owned treatment works (POTWs) and potential
dioxin-like PCB TEQ releases 10-23
Table 10-11. Quantity of sewage sludge disposed of annually in 1995 by primary, secondary, or
advanced treatment publicly owned treatment works (POTWs) and potential
dioxin-like PCB TEQ releases 10-24
Table 10-12. Quantity of sewage sludge disposed of annually in 2000 by primary, secondary, or
advanced treatment publicly owned treatment works (POTWs) and potential
dioxin-like PCB TEQ releases 10-25
Table 10-13. PCB congener group emission factors for industrial wood combustorsa 10-28
Table 10-14. PCB congener group emission factors for medical waste incinerators
(MWIs) 10-29
Table 10-15. PCB congener group emission factors for a tire combustora 10-30
Table 10-16. Dioxin-like PCB concentrations in cigarette tobacco in brands from various
countries (pg/pack) 10-32
Table 10-17. Dioxin-like PCB concentrations in stack gas collected from a U.S. sewage sludge
incinerator 10-33
Table 10-18. Dioxin-like PCB emission factors from backyard barrel burning 10-35
Table 10-19. PCB congener group emission factors for a petroleum catalytic reforming
unit 10-35
XXlll
-------
LIST OF TABLES (continued)
Table 10-20. Estimated tropospheric half-lives of dioxin-like PCBs with respect to gas-phase
reaction with the OH radical 10-40
Table 10-21. Estimated PCB loads in the global environment as of 1985 10-42
Table 10-22. Estimated domestic sales of aroclors and releases of PCBs, 1957-1974 (metric
tons) 10-43
Table 10-23. Estimated U.S. usage of PCBs by use category, 1930-1975 10-44
Table 10-24. Estimated direct releases of Aroclors to the U.S. environment, 1930-1974a
(metric tons) 10-45
Table 10-25. Estimated releases of dioxin-like PCB TEQs to the U.S. environment,
1930-1977 10-46
Table 11-1. Historical production, sales, and usage of 2,4-dichlorophenoxy-acetic acid
(2,4-D) (metric tons) 11-4
Table 11-2. Historical production, sales, and usage of 2,4,5-trichlorophen-oxyacetic acid
(2,4,5-T) (metric tons) 11-7
Table 11-3. CDD/CDF concentrations (i-ig/kg) in recent sample of 2,4,5-trichloro-
phenoxyacetic acid (2,4,5-T) 11-10
Table 11-4. PCB 138 fluxes predicted by Harner et al. (1995) 11-16
Table 11-5. Summary of flux calculations for total PCBs in Green Bay 11-19
Table 11-6. Comparison of model-estimated PCB concentrations with observed
values 11-22
XXIV
-------
LIST OF FIGURES
Figure 1-1. Chemical structure of 2,3,7,8-TCDD and related compounds 1-7
Figure 1-2. Estimated CDD/CDF I-TEQ emissions to air from combustion sources in the
United States for reference year 1987 1-24
Figure 1-3. Estimated CDD/CDF I-TEQ emissions to air from combustion sources in the
United States for reference year 1995 1-25
Figure 1-4. Estimated CDD/CDF I-TEQ emissions to air from combustion sources in the
United States for reference year 2000 1-26
Figure 1-5. Comparison of estimates of annual I-TEQ emissions to air (g I-TEQ/yr) for
reference years 1987, 1995, and 2000 1-27
Figure 1-6. Estimated CDD/CDF WHO-TEQ emissions to air from combustion sources in the
United States for reference year 1987 1-28
Figure 1-7. Estimated CDD/CDF WHO-TEQ emissions to air from combustion sources
in the United States for reference year 1995 1-29
Figure 1-8. Estimated CDD/CDF WHO-TEQ emissions to air from combustion sources
in the United States for reference year 2000 1-30
Figure 1-9. Comparison of estimates of annual WHO-TEQ emissions to air (g WHO-
TEQ/yr) for reference years 1987, 1995, and 2000 1-31
Figure 1-10. Congener profiles (as percent distributions to the sum of CDDs and CDFs) of
anthropogenic sources of chlorinated dibenzo-p-dioxins and chlorinated
dibenzofurans in the United States 1-45
Figure 2-1. Typical mean distribution of CDD and CDF congeners in contemporary municipal
solid waste 2-3
Figure 2-2. The de novo synthesis of CDDs/CDFs from heating carbon particulate at 300°C at
varying retention times 2-15
Figure 2-3. Temperature effects on CDD/CDF formation 2-16
Figure 3-1. Typical mass burn waterwall municipal solid waste combustor 3-3
Figure 3-2. Typical mass burn rotary kiln combustor 3-4
Figure 3-3. Typical modular starved-air combustor with transfer rams 3-5
XXV
-------
LIST OF FIGURES (continued)
Figure 3-4. Typical modular excess-air combustor 3-6
Figure 3-5. Typical dedicated refuse-derived fuel-fired spreader stoker boiler 3-7
Figure 3-6. Fluidized-bed refuse-derived fuel furnace 3-8
Figure 3-7. Municipal waste combustor design classes for 1987 3-11
Figure 3-8. Municipal waste combustor design classes for 1995 3-12
Figure 3-9. Municipal waste combustor design classes for 2000 3-13
Figure 3-10. Congener and congener group profiles for air emissions from a mass burn
waterwall municipal waste combustor equipped with a dry scrubber and fabric
filter 3-32
Figure 3-11. 2,3,7,8-TCDD frequency distribution (negative natural log concentration) . . . 3-33
Figure 3-12. 1,2,3,7,8-PeCDD frequency distribution (negative natural log concentration) . 3-34
Figure 3-13. Congener profile for air emissions from 17 hazardous waste incinerators tested
from 1993 through 1996 3-50
Figure 3-14. Congener and congener group profiles for air emissions from
boilers and industrial furnaces burning hazardous waste 3-52
Figure 3-15. Congener and congener group profiles for air emissions from medical waste
incinerators without air pollution control devices (nondetects set equal
to zero) 3-66
Figure 3-16. Congener and congener group profiles for air emissions from medical waste
incinerators equipped with a wet scrubber and fabric filter 3-67
Figure 3-17. Congener and congener group profiles for air emissions from the Camellia
Memorial Lawn crematorium and Woodlawn Cemetery crematorium 3-75
Figure 3-18. Congener profile for air emissions from the University of Georgia animal
crematorium 3-78
Figure 3-19. Congener and congener group profiles for air emissions from
sewage sludge incinerators 3-81
xxvi
-------
LIST OF FIGURES (continued)
Figure 3-20. Congener and congener group profiles for air emissions from a tire
combustor 3-85
Figure 4-1. Congener and congener group profiles for air emissions from diesel-fueled
vehicles 4-15
Figure 4-2. Congener and congener group profiles for air emissions from leaded
gas-fueled vehicles 4-16
Figure 4-3. Congener and congener group profiles for air emissions from unleaded
gas-fueled vehicles 4-17
Figure 4-4. Tunnel air concentrations 4-29
Figure 4-5a. Congener and congener group profiles for air emissions from industrial wood
combustors 4-46
Figure 4-5b. Congener and congener group profiles for air emissions from bleached Kraft mill
bark combustors 4-47
Figure 4-6. Congener group profile for air emissions from residential oil-fueled
furnaces 4-63
Figure 4-7. Congener and congener group profiles for air emissions from industrial
oil-fueled boilers 4-66
Figure 4-8. Congener and congener group profiles for air emissions from
industrial/utility coal-fueled combustors (nondetects set equal to zero) 4-73
Figure 4-9. Congener group profile for air emissions from residential coal-fueled
combustors 4-77
Figure 5-1. Congener profile for air emissions from cement kilns burning hazardous waste
for reference years 1987 and 1995 (nondetect set equal to zero) 5-9
Figure 5-2. Congener profile for air emissions from cement kilns burning hazardous
waste for reference year 2000 5-11
Figure 5-3. Congener profile for air emissions from cement kilns burning nonhazardous
waste for reference years 1987, 1995, and 2000 5-13
Figure 5-4. Congener and congener group profiles for air emissions from petroleum
catalytic reforming units 5-30
XXVll
-------
LIST OF FIGURES (continued)
Figure 5-5. CDD profiles for Japanese cigarettes, smoke, and ash 5-33
Figure 5-6. Congener group profiles for mainstream and sidestream cigarette smoke .... 5-35
Figure 5-8 Congener and congener group profiles for air emissions from
Kraft black liquor recovery boilers (nondetect set equal to zero) 5-45
Figure 5-7. Congener group profiles for cigarette tobacco from various countries 5-38
Figure 6-1. Congener profile for landfill flare air emissions 6-3
Figure 6-2. Congener profile for forest fire simulation approach emissions 6-17
Figure 7-1. Congener and congener group profiles for air emissions from secondary
aluminum smelters 7-13
Figure 7-2a. Congener group profile for air emissions from a secondary copper smelter . . . 7-18
Figure 7-2b. Congener and congener group profiles for a closed secondary copper
smelter 7-19
Figure 7-3. Congener and congener group profiles for air emissions from
secondary lead smelters 7-26
Figure 7-4. Congener profiles for air emissions from U.S. iron ore sintering plants 7-31
Figure 7-5. Congener group profile for air emissions from a scrap wire incinerator 7-43
Figure 7-6. Congener group profile for air emissions from a drum incinerator 7-47
Figure 8-1. 104 Mill Study full congener analysis results for pulp (nondetects equal
to zero) 8-3
Figure 8-2. 104 Mill Study full congener analysis results for sludge 8-3
Figure 8-3. 104 Mill Study full congener analysis results for effluent 8-4
Figure 8-4. Congener profile for technical-grade PCP (developed from data in last column in
Table 8-10) 8-57
Figure 8-5. Congener profile for 2,4-D (salts and esters) (based on mean concentrations
reported in Table 8-34) 8-73
XXVlll
-------
LIST OF FIGURES (continued)
Figure 8-6. Congener profiles for sewage sludge (created from data in Table 8-38) 8-88
Figure 11-1. Fluxes among environmental reservoirs 11-2
XXIX
-------
LIST OF ABBREVIATIONS AND ACRONYMS
AHA American Hospital Association
AMSA Association of Metropolitan Sewerage Agencies
APCD Air pollution control device
BDDs polybrominated dibenzo-p-dioxins
BDFs polybrominated dibenzofurans
Btu British thermal unit
CaCl2 calcium chloride
CARB California Air Resources Board
CBI Confidential Business Information
CDD polychlorinated dibenzo-p-dioxin
CDF polychlorinated dibenzofuran
CFR Code of Federal Regulations
CSF Confidential Statement of Formula
CKD cement kiln dust
CO carbon monoxide
CO2 carbon dioxide
CuCl copper (I) chloride
CuCl2 copper (II) chloride
DBF dibenzofuran
DCBz dichlorobenzene
DCI data call-in
DCP dichlorophenol
DL detection limit
dscm dry standard cubic meter
DSI dry sorbent injection
EDC ethylene dichloride
EIA Energy Information Administration
EPA U.S. Environmental Protection Agency
EPRI Electric Power Research Institute
ESP electrostatic precipitator
FF fabric filter
FCEM field chemical emissions measurement
FeCl3 ferric (iron) chloride
XXX
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
FIFRA Federal Insecticide, Fungicide, and Rodenticide Act
GAC granular activated carbon
GC/ECD gas chromatography/electron capture detector
GC/MS gas chromatography/mass spectrometry
HC1 hydrogen chloride
HCBz hexachlorobenzene
HDD halogenated dibenzo-^-dioxin
HDF halogenated dibenzofuran
HWI hazardous waste incinerator
HxCB hexachlorobiphenyl
IUPAC International Union of Pure and Applied Chemistry
KC1 potassium chloride
LOQ limit of quantitation
MB-WW mass burn waterwall
MCBz monochlorobenzene
MgCl2 magnesium chloride
MgO magnesium oxide
MSW municipal solid waste
MWI medical waste incinerator
NaCl sodium chloride
NaOCl soldium hypochlorite
NCASI National Council of the Paper Industry for Air and Stream Improvement
MC12 nickel chloride
NiO nickel oxide
Nm3 standard cubic meter
NMOC nonmethane organic compound
OAQPS Office of Air Quality Planning and Standards
O2 molecular oxygen
OH hydroxide ion
OPP Office of Pesticide Programs
ORD Office of Research and Development
OSW Office of Solid Waste
Pb lead
XXXI
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
PCA Portland Cement Association
PCB polychlorinated biphenyl
PCP pentachlorophenol
PCP-Na pentachlorophenate
PeCB pentachlorobiphenyl
PeCBz pentachlorobenzene
PM paniculate matter
POTW publicly owned treatment works
ppb parts per billion
ppm parts per million
ppmv parts per million (volume basis)
ppt parts per trillion
PVC polyvinyl chloride
QA/QC quality assurance/quality control
RCRA Resource Conservation and Recovery Act
RDF refuse-derived fuel
SIC Standard Industrial Classification
SIP State Implementation Plan
SNUR Significant New Use Rule
SO2 sulfur dioxide
TCBz trichlorobenzene
TCDD 2,3,7,8-tetrachlorobidenzo-p-dioxin
TCDF 2,3,.7,8-tetrachlorobidenzofuran
TeCB tetrachlorobiphenyl
TeCP tetrachlorophenol
TEF toxicity equivalency factor
TEQ toxicity equivalence
TEQ/yr toxicity equivalents per year
TiCl4 titanium tetrachloride
TrCB trichlorobiphenyl
TrCP trichlorophenol
TRI Toxics Release Inventory
TSCA Toxic Substances Control Act
xxxn
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
2,4-D 2,4-dichlorophenoxyacetic acid
2,4-DB 4-(2,4-dichlorophenoxy) butyric acid
2,4-DCP 2,4-dichlorophenol
2,4-DP 2-(2,4-dichlorophenoxy) propionic acid
2,4,5-T 2,4,5-trichlorophenoxy (phenoxy herbicides)
U.K. United Kingdom
USD A U.S. Department of Agriculture
VCM vinyl chloride monomer
WHO World Health Organization
WS wet scrubber
XXXlll
-------
FOREWORD
The purpose of this document is to present an inventory of sources and environmental
releases of dioxin-like compounds in the United States. This inventory is associated with three
distinct reference years: 1987, 1995, and 2000. The presentation of information in this manner
permits the ranking of sources by magnitude of annual release and allows for the evaluation of
environmental trends over time.
The term "dioxin-like" includes congeners of polychlorinated dibenzo-p-dioxins (CDDs),
polychlorinated dibenzofurans (CDFs) having chlorine atoms in the 2,3,7,8 positions on the
molecule, and certain coplanar-substituted polychlorinated biphenyls (PCBs). Dioxin-like refers
to the fact that these compounds have similar chemical structure and physical-chemical properties
and invoke a common toxic response. Because of their hydrophobic nature and resistance
towards metabolism, these chemicals persist and bioaccumulate in fatty tissues of animals and
humans. Consequently, the principal route of chronic population exposure is through the dietary
consumption of animal fats, fish, shellfish, and dairy products. Dioxin-like compounds are
persistent in soils and sediments, with environmental half-lives ranging from years to several
decades. Understanding the sources and environmental releases of dioxin-like compounds is
fundamental to ultimately linking sources with population exposures. It is through such
understanding that actions can be taken to reduce human exposures.
The quantitative results of the inventory are expressed in terms of the toxicity equivalence
(TEQ) of the mixture of polychlorinated dibenzo-^-dioxin (CDD) and polychlorinated
dibenzofuran (CDF) compounds present in environmental releases using a procedure sanctioned
by the World Health Organization (WHO) in 1998. This TEQ procedure translates the complex
mixture of CDDs and CDFs characteristic of environmental releases into an equivalent toxicity
concentration of 2,3,7,8-tetrachlorodibenzo:p-dioxin (2,3,7,8-TCDD), the most toxic member of
this class of compounds. With this procedure, the quantity of the mixture of CDDs and CDFs
present as a release is given the notation grams (g) TEQDF-WHO98.
This inventory of sources and environmental releases of dioxin-like compounds concludes
that, between 1987 and 2000, there was approximately 90% reduction in the release of dioxin-like
compounds to the circulating environment of the United States from all known sources combined.
Annual emission estimates (TEQDF-WHO98) of releases of CDDs/CDFs to air, water, and land
from reasonably quantifiable sources are approximately 1,422 g in reference year 2000; 3,444 g in
reference year 1995; and 13,965 g in reference year 1987. In 1987 and 1995, the leading sources
of dioxin emissions to the U.S. environment were municipal waste combustors. The inventory
also identifies bleached chlorine pulp and paper mills as a significant source of dioxin to the
aquatic environment in 1987 but a minor source in 1995 and 2000. The inventory concludes that
xxxiv
-------
the major source of dioxin in 2000 was the uncontrolled burning of refuse in backyard burn
barrels in rural areas of the United States.
The reduction in environmental releases of dioxin-like compounds from 1987 to 2000 is
attributable to source-specific regulations, improvements in source technology, advancements in
the pollution control technologies specific to controlling dioxin discharges and releases, and the
voluntary actions of U.S. industries to reduce or prevent dioxin releases.
Peter W. Preuss, Ph.D.
Director
National Center for Environmental Assessment
Office of Research and Development
xxxv
-------
PREFACE
This document, An Inventory of Sources and Environmental Releases of Dioxin-Like
Compounds in the United States for the Years 1987, 1995, and 2000, was prepared by the
National Center for Environmental Assessment, which is the health risk assessment program in
the Office of Research and Development. The document presents estimates of annual releases of
dioxin-like compounds specific for each year. It is a detailed compilation and description of all
known U.S. sources and their associated activities that cause these compounds to be released into
the open and circulating environment, i.e., to air, water, and land. The overall purpose of this
report is to document and describe sources in the United States that release dioxin-like
compounds into the open environment, quantify annual releases to the environment from known
sources in a scientific and transparent manner, and provide a reliable basis for observing trends in
environmental releases. To the extent practical, the inventory is a comprehensive analysis of
dioxin sources.
This final document reflects a consideration of all comments received on an External
Review Draft dated March 2005 (EPA600/P-03/002A) provided by an expert panel at a peer-
review workshop held September 13-15, 2005, and comments received during a 60-day public
review and comment period (May 6-July 5, 2005).
Over 800 references were reviewed and cited in the preparation of this document. The
citations generally reflect publications up to and including the year 2003.
xxxvi
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
The National Center for Environmental Assessment (NCEA) was responsible for
preparing this document. Major portions of this report were prepared by Versar, Inc. under EPA
Contract No. 68-W-99-041. David Cleverly of NCEA served as the EPA Work Assignment
Manager, providing overall direction and coordination of the project, as well as author of several
chapters.
Authors
David Cleverly
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
Greg Schweer
Office of Pollution Prevention and Toxics
U.S. Environmental Protection Agency
Washington, DC
Ronald Lee
Versar, Inc.
Springfield, VA
Kelly McAloon
Versar, Inc.
Springfield, VA
Charles Peck
Versar, Inc.
Springfield, VA
Marit Espevik Randall
Versar, Inc.
Springfield, VA
Karie Traub Riley
Versar, Inc.
Springfield, VA
xxxvn
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Contributors
Frank Behan, Office of Solid Waste, Office of Solid Waste and Emergency Response, U.S.
Environmental Protection Agency, Washington, DC
Laurel Driver, Office of Air Quality Planning and Standards, Office of Air and Radiation, U.S.
Environmental Protection Agency, Research Triangle Park, NC
Marietta Echeverria, National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Washington, DC
Joseph Ferrario, Environmental Chemistry Laboratory, Office of Pesticide Programs, Office of
Prevention, Pesticides and Toxic Substances, U.S. Environmental Protection Agency, Stennis
Space Center, MS
Roy Huntley, Office of Air Quality Planning and Standards, Office of Air and Radiation, U.S.
Environmental Protection Agency, Research Triangle Park, NC
John Schaum, National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Washington, DC
Walt Stevenson, Office of Air Quality Planning and Standards, Office of Air and Radiation, U.S.
Environmental Protection Agency, Research Triangle Park, NC
Dwain Winters, Office of Pollution Prevention and Toxics, Office of Prevention, Pesticides and
Toxic Substances, U.S. Environmental Protection Agency, Washington, DC
Reviewers
Internal EPA Review: Internal EPA review occurred from May 2 through June 5, 2003. The
key reviewers included the following individuals and offices:
Elmer Akin, EPA Region 4, Atlanta, GA
Don Anderson, Office of Science and Technology, Office of Water, Washington, DC
John D Bachmann, Office of Air Quality Planning and Standards, Research Triangle Park, NC
Angela L. Bandemehr, Office of International Affairs, Washington, DC
Frank Behan, Office of Solid Waste, Washington, DC
XXXVlll
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
Tom Braverman, Office of Air Quality Planning and Standards, Research Triangle Park, NC
Laurel Driver, Office of Air Quality Planning and Standards, Research Triangle Park, NC
Dale Evarts, Office of Air Quality Planning and Standards, Research Triangle Park, NC
Brian Gullett, National Risk Management Laboratory, Research Triangle Park, NC
Bob Holloway, Office of Solid Waste, Washington, DC
Roy Huntley, Office of Air Quality Planning and Standards, Research Triangle Park, NC
Phil Lorang, Office of Air Quality Planning and Standards, Research Triangle Park, NC
Ed Ohanian, Office of Water, Washington, DC
Debbie Sisco, Office of Pesticide Programs, Washington, DC
Alan Rubin, Office of Water, Washington, DC
John Schaum, National Center for Environmental Assessment, Washington, DC
Laurie Schuda, National Center for Environmental Assessment, Washington, DC
Joe Somers, Office of Transportation and Air Quality, Ann Arbor, MI
Walt Stevenson, Office of Air Quality Planning and Standards, Research Triangle Park, NC
Dwain Winters, Office of Pollution Prevention and Toxics, Washington, DC
Donn Viviani, Office of Policy, Economics, and Innovation, Washington, DC
External Peer Review: This document has undergone two external peer reviews: the first
External Review Draft (EPA/600/P-98/002Aa) published in April 1998 was reviewed by an expert
panel at a public meeting held in Alexandria, Virginia, on June 3-4, 1998. The most recent
External Review Draft (EPA600/P-03/002A) published in March 2005 was released for public
review and comment on May 6, 2005, and was subsequently reviewed at a public meeting held in
Arlington, Virginia, on September 13-15, 2005.
XXXIX
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS (continued)
1. Peer review meeting September 13-15, 2005
Panel Chair
Heidelore Fiedler, Ph.D., Scientific Affairs Officer, United Nations Environment Programme
Chemicals, Chatelaine, Switzerland
Panel Members
Judith A. Abbott, Ph.D., Research Scientist, Bureau of Toxic Substance Assessment, New York
State Department of Health, Albany, NY
Gary A. Amendola, Ph.D., President, Amendola Engineering, Westlake, OH
Patrick Dyke, President, PD Consulting, Oxford, United Kingdom
Christopher C. Lutes, Ph.D., Principal Scientist/Associate Vice President, ARCADIS G&M,
Inc., Durham, NC
Jozef M. Pacyna, Ph.D., Center for Ecological Economics, Norwegian Institute for Air Research,
Kjeller, Norway
2. Peer review meeting June 3-4,1998
Panel Chair
Valerie Thomas, Ph.D., Research Scientist, Center for Energy and Environmental Studies,
Princeton University
Panel Members
Patrick Dyke, President, PD Consulting, Oxford, United Kingdom
Raouf Morcos, Manager, National Office of Pollution Prevention, Environment Canada
Quebec, Canada
William Randall Seeker, Ph.D., Senior Vice President, Energy and Environmental Research,
Irvine, CA
Chun Yi Wu, Senior Engineer, Air Quality Division, Minnesota Pollution Control Agency,
St. Paul, MN
xl
-------
ACKNOWLEDGMENT
The authors would like to acknowledge the contributions of the following individuals:
Terri Konoza of NCEA who managed the document production activities and provided editing,
proofreading, and word processing support; Patricia von Brook of KBM Group who provided
editing support; and Christine Chang of KBM Group who provided word processing support.
xli
-------
EXECUTIVE SUMMARY
This report, An Inventory of Sources and Environmental Releases of Dioxin-Like
Compounds in the United States for the Years 1987, 1995, and 2000, presents estimates of annual
releases of dioxin-like compounds specific to each year. It is a detailed compilation and
description of all known U.S. sources and their associated activities that cause these compounds
to be released into the open and circulating environment, i.e., to air, water, and land. The overall
purpose of this report is to document and describe sources in the United States that release dioxin-
like compounds into the open environment, quantify annual releases to the environment from
known sources in a scientific and transparent manner, and provide a reliable basis for observing
trends in environmental releases. To the extent practical, the inventory is a comprehensive
analysis of dioxin sources.
The term "dioxin-like" refers to chemical compounds that mimic the chemical and
physical properties of dioxin and have similar toxic effects. These include compounds of
chlorinated dibenzo-/?-dioxins (CDDs), chlorinated dibenzofurans (CDFs), and certain coplanar
polychlorinated biphenyls (PCBs). The dioxin-like CDDs and CDFs have chlorine atoms in the
2,3,7,8 positions on the molecule. Dioxin-like PCBs contain zero or one chlorine atom in the
2,2',6 or 6' positions. All together there are 7 CDDs, 10 CDFs, and 12 PCBs that are considered
to be dioxin like. It should be emphasized that releases of dioxin-like compounds presented in
this inventory are, for the most part, for dioxin-like CDDs and CDFs. Sources of dioxin-like
PCBs are generally poorly characterized.
Approach
Only sources judged to have a reasonable likelihood for releases of dioxin-like compounds
to the air, water, and land of the United States are addressed in this report. The release estimates
were derived in one of two ways: (1) dioxin was measured as an actual release from the source
(i.e., points of release from the source were sampled and evaluated), or (2) dioxin releases were
calculated on the basis of an emission factor and activity level. The emission factor is the amount
of dioxin anticipated to be emitted per unit of activity and is derived from measurements made at
sources having similar characteristics. The activity level is the amount of material processed,
produced, or consumed by the source in the course of a year or, in the case of mobile sources, the
number of kilometers driven. It can take several forms, such as kilograms of material processed
per year by an industrial facility, vehicle kilometers traveled per year by trucks and automobiles,
and liters of wastewater discharged into surface water from industrial sources. The activity level
is multiplied by the emission factor to arrive at an estimate of annual dioxin releases from those
sources lacking direct measurement of dioxin emissions.
xlii
-------
Confidence in the accuracy of both the emission factor and the activity level are rated as
low, medium, or high, based on the quality of the data. All the release estimates from sources in
the inventory are assigned an overall confidence rating based on the lowest rating assigned to
either the emission factor or the activity level. In some cases, the data were not adequate to
support even a low confidence rating. These cases were treated in one of two ways. If the data
were sufficient to make an approximate, but clearly nonrepresentative, estimate of releases, the
estimates were labeled as preliminary and were not included in the national inventory. If limited
data suggested that dioxin releases were possible from a source but were not adequate to support
even rudimentary calculations of emissions, the source was labeled as unquantifiable. This
approach resulted in the classification scheme shown below.
Category A High Confidence Included in the national
Category B Medium Confidence quantitative inventory
Category C Low Confidence
Category D Preliminary Not included in the national
Category E Unquantifiable quantitative inventory
Throughout this document, environmental release estimates are presented in terms of
toxicity equivalence (TEQs). TEQs are derived from a toxicity weighting system that converts all
mixture components to a single value normalized to the toxicity of 2,3,7,8-TCDD. This is done
for convenience in presenting summary information and to facilitate comparisons across sources.
For many situations, however, it is important to use the individual CDD/CDF and PCB congener
values rather than TEQs. CDD and CDF congener-specific releases for most sources are given in
tables in each chapter. The summary amounts of dioxin-like compounds released to the
environment are reported in units of grams (g) TEQ, developed by the World Health Organization
(WHO), and are given the abbreviated notation of TEQDF-WHO98 throughout the document.
The major findings of the inventory of sources and environmental releases of dioxin-like
compounds in the United States are:
1. In 1987. 1995. and 2000. approximately 13.965: 3.444: and 1.422 g TEO. respectively.
were released into the U.S. environment from all sources. Figure ES-1 graphically
displays these releases.
xliii
-------
13,965.3
14,000
12,000
10,000-
g 8,000
2 6,000
ro
4,000
2,000
,422.0
Year 1987
Year 1995
Year 2000
Figure ES-1. Total environmental releases of dioxin-like compounds (g TEQ)
from all quantifiable sources during 1987,1995, and 2000.
2. Environmental releases of dioxin-like compounds decreased by approximately 90%
between 1987 and 2000. As shown in Figure ES-1, most of the reductions (75%)
occurred between 1987 and 1995. The overall reduction in releases of dioxin-like
compounds is attributed to the control of air emissions of these compounds from
municipal waste combustors (MWCs), medical waste incinerators, and cement kilns
burning hazardous waste and of wastewater discharges of the compounds into surface
waters from pulp and paper mills using chlorine. These reductions were achieved
through a combination of regulatory activities, improved emission controls, voluntary
actions on the part of industry, and the closing of a number of facilities. Table ES-1
shows the reductions made by the largest sources of dioxin-like compound releases.
Emission estimates for individual sources that could be quantified, i.e., Categories A,
B, and C, are presented in the main text of this report.
The leading source of dioxin-like compounds in 2000 was the backyard burning of
refuse in barrels (498.5 g TEQ. or 35% of total releases), as shown in Table ES-2,
which presents the top 10 sources of releases for 2000, 1995, and 1987. Backyard
barrel burning of refuse is an activity that occurs in rural areas of the United States. It
is unregulated on a national level, but many states have banned or limited the practice
(New Jersey, New York, California, Minnesota, Wisconsin, Massachusetts, and
xliv
-------
Table ES-1. Reductions of releases of dioxin-like compounds to the
environment in reference years 2000 and 1987 from major sources in the
United States
Source category
Municipal waste
combustion
Medical waste
incineration
Cement kilns
burning hazardous
waste
Bleached chemical
wood pulp and
paper mills
Releases to:
Air
Air
Air
Surface
water
2000
(g TEQ)
83.8
378.0
18.8
1.0
1987
(g TEQ)
8,905.1
2,570.0
117.8
356.0
Percent
reduction
>99
85
84
>99
4.
Florida to name a few). In 1995 and 1987 MWCs were the leading source of releases
(1,393.5 g, or 40% of total releases in 1995; 8,905.1 g, or 64% of releases in 1987).
However, due to strict regulatory requirements limiting dioxin emissions, MWCs were
ranked fourth among the top 10 sources in 2000, with emissions of only 83.8 g, or 6%
of total releases. Automobiles burning leaded gasoline were ranked as the eighth
leading source of dioxin in 1987. The phase out of lead in gasoline eliminated this
source by 2000. Cement kilns burning hazardous waste dropped out of the top 10
sources in 2000, due primarily to voluntary actions of industry combined with national
regulatory requirements to reduce dioxin emissions.
Environmental releases of dioxin-like compounds in the United States occur from a
wide variety of sources but are dominated by releases to the air from combustion
sources. Figure ES-2 presents the breakdown of releases to air, water, and land for
each reference year.
5. There are potential sources of dioxin-like compounds that were not included in the
inventory. Significant amounts of the dioxin-like compounds produced annually in the
United States are not considered releases to the open and circulating environment and,
therefore, are not included in the national inventory. Examples include dioxin-like
compounds generated internal to a process but destroyed before release and waste
streams that are disposed of in approved and secure landfills. There are also potential
sources for which no information exists to permit any reliable estimates of
xlv
-------
Table ES-2. Top 10 sources of dioxin-like compound releases and amounts released
(g TEQ) for reference years 2000,1995, and 1987
Rank
1
2
3
4
5
6
7
8
9
10
2000
(1,422 g total)
Backyard barrel burning
of refuse (air)
Medical waste/
pathological incineration
(air)
Municipal wastewater
treatment sludge (applied
to land and incinerated)
(land and air)
Municipal waste
combustion (incineration
of refuse) (air)
Coal-fired utility boilers
(electric generating
plants) (air)
Diesel heavy-duty trucks
(air)
Industrial wood
combustion (air)
Diesel off-road
equipment (includes
ships, farm equipment,
trains) (air)
Ethylene dichloride/vinyl
chloride production
(land, air, water)
Sintering plants (air)
498.5
378.0
89.7
83.8
69.5
65.4
41.5
33.1
30.0
27.6
1995
(3,444 g total)
Municipal waste
combustion
(incineration of refuse)
(air)
Backyard barrel
burning of refuse (air)
Medical
waste/pathological
incineration (air)
Secondary copper
smelting (air)
Cement kilns
(hazardous waste
burning) (air)
Municipal wastewater
treatment sludge
(applied to land and
incinerated) (land and
air)
Coal-fired utility boilers
(electric generating
plants) (air)
Ethylene
dichloride/vinyl
chloride production
(land, air, water)
Diesel heavy-duty
trucks (air)
Bleached chemical
wood pulp and paper
mills (land and water)
1,393.5
628.0
487.0
271.0
156.1
133.3
60.1
35.7
33.3
30.0
1987
(13,965 g total)
Municipal waste
combustion
(incineration of
refuse) (air)
Medical
waste/pathological
incineration (air)
Secondary copper
smelting (air)
Backyard barrel
burning of refuse
(air)
Bleached chemical
wood pulp and
paper mills (land,
water)
Cement kilns
(hazardous waste
burning) (air)
Municipal
wastewater
treatment sludge
(applied to land
and incinerated)
(land and air)
Coal-fired utility
boilers (electric
generating plants)
(air)
Automobiles using
leaded gasoline
(air)
2,4-
Dichlorophenoxy
acetic acid (2,4-D)
(land)
8,905.1
2,570.0
983.0
604.0
370.1
117.8
85.0
50.9
37.5
33.4
xlvi
-------
Year 2000
1.8%
5.7%
2.6%
Year 1995
7.4%
92.4%
90.0%
Year 1987
2.5% 0.9%
Releases to Air
96.5%
Releases to Water
Releases to Land
Figure ES-2. Releases of dioxin-like compounds to air, water, and land in
2000,1995, and 1987.
environmental releases; therefore, these potential sources could not be included in the
inventory. EPA has classified these potential sources as Category D sources.
Examples include forest and grassland fires and accidental fires at municipal solid
waste landfills. Taken together, these sources have the potential to significantly
increase the emission estimates in the present inventory.
6. The amount of dioxin-like PCBs released from man-made sources remains poorly
characterized. Only a total of 19.5, 78.5, and 51.5 g of PCB TEQ could be quantified
for 2000, 1995, and 1987, respectively. To date, only sewage sludge has been
adequately characterized as a source of dioxin-like PCB releases.
xlvii
-------
1. BACKGROUND, APPROACH, AND CONCLUSIONS
1.1. BACKGROUND
This report presents a comprehensive inventory of sources of releases of dioxin-like
compounds in the United States for the years 1987, 1995, and 2000. It is a detailed review and
description of all known sources and their associated activities that cause these compounds to be
released into the "open and circulating environment," i.e., air, water, and land.
The aim of this report is to:
• Document and describe sources that release dioxin-like compounds into the
circulating environment of the United States.
Quantify annual releases to the environment of the United States from known sources
in a scientific and transparent manner.
• Provide a reliable basis for time-trends analyses such as observing changes in total
releases to the circulating environment from 1987 to 2000. Time-trend analyses
provide a quantitative indication of the achievements made (or lack thereof) in
reducing environmental releases of dioxin-like compounds from specific sources in
the United States.
This is the second dioxin source inventory issued by the U.S. Environmental Protection
Agency (EPA, or the Agency). The first one was issued in draft form and covered the years 1987
and 1995 (U.S. EPA, 1998a). The current effort updates this earlier document and adds annual
release estimates for 2000.
The intended audience and users of the dioxin inventory include:
• Members of the general public who are interested in learning more about sources of
emissions of dioxin-like compounds to the U.S. environment and in obtaining peer-
reviewed estimates of releases.
State and local regulatory agencies that are interested in obtaining reliable and peer-
reviewed information on sources and environmental releases of dioxin-like
compounds.
• EPA Regional and Program Offices that are responsible for evaluating the need for
regulating and/or preventing dioxin releases to the environment.
• Risk assessors in the private and public sectors who need reliable information on
sources and releases of dioxin-like compounds to improve quantitative risk
assessments of dioxin sources.
1-1
-------
• Researchers who are interested in documented and time-specific dioxin source and
emissions data to be used in sequential time-trends analyses.
• Private and public stakeholder groups that are interested in obtaining reliable and
peer-reviewed information on dioxin sources and releases and in observing time
trends in environmental releases of dioxin-like compounds from specific source
categories.
A complete listing of the nomenclature used in this report is depicted in Table 1-1.
Table 1-1. Nomenclature for dioxin-like compounds
Term/symbol
CDD
CDF
PCB
M
D
Tr
T
Pe
Hx
Hp
O
2,3,7,8
Congener
Congener group
Isomer
Specific isomer
Definition
Chlorinated dibenzo-/?-dioxin, halogens substituted in any position
Chlorinated dibenzofuran, halogens substituted in any position
Polychlorinated biphenyl
Symbol for mono (i.e., one halogen substitution)
Symbol for di (i.e., two halogen substitution)
Symbol for tri (i.e., three halogen substitution)
Symbol for tetra (i.e., four halogen substitution)
Symbol for penta (i.e., five halogen substitution)
Symbol for hexa (i.e., six halogen substitution)
Symbol for hepta (i.e., seven halogen substitution)
Symbol for octa (i.e., eight halogen substitution)
Halogen substitutions in the 2,3,7,8 positions
Any one particular member of the same chemical family (e.g., there are 75
congeners of CDDs).
Group of structurally related chemicals that have the same degree of chlorination
(e.g., there are eight congener groups of CDDs, monochlorinated [MCDD]
through octachlorinated [OCDD]).
Substances that belong to the same congener group (e.g., 22 isomers constitute
the congener group of TCDDs).
Denoted by unique chemical notation (e.g., 2,4,8,9-tetrachlorodibenzofuran is
referred to as 2,4,8,9-TCDF).
Source: Adapted from U.S. EPA (1989a).
1-2
-------
1.1.1. Reference Years
A central part of EPA's dioxin inventory is the organization of estimates of annual
releases of dioxin-like compounds into reference years 1987, 1995, and 2000. The selection and
use of three reference years provides a basis for comparing environmental releases over time.
The year 1987 was selected as the initial reference year because it was the earliest time
when it was feasible to assemble a reasonably comprehensive inventory. Prior to that time, very
little data existed on dioxin emissions from stacks or other release points. The first study
providing the type of data needed for a national inventory was EPA's National Dioxin Study
(U.S. EPA, 1987a). The year 1987 also corresponds roughly with the time when significant
advances occurred in emissions measurement techniques and in the development of high-
resolution mass spectrometry and gas chromatography, which allowed analytical laboratories to
detect low levels of polychlorinated dibenzo-p-dioxin (CDD) and polychlorinated dibenzofuran
(CDF) congeners in environmental samples. Soon after this time, a number of facilities began
upgrades specifically intended to reduce CDD/CDF emissions. Consequently, 1987 emissions
are representative of levels of emissions that occurred before the widespread installation of
pollution control systems and pollution prevention techniques specifically designed to reduce
dioxin releases from man-made sources into the air, land, and water.
EPA selected 1995 as the second reference year because it reflects the completion time of
the first set of regulatory activities specifically tailored to reduce dioxin releases from major
sources. By 1995, EPA had proposed or promulgated regulations limiting CDD/CDF emissions
from municipal waste combustors (MWCs), medical waste incinerators (MWIs), hazardous
waste incinerators, cement kilns burning hazardous waste, and pulp and paper mill facilities
using bleached chlorine processes.
The year 2000 was chosen as the most current date that could be addressed when this
effort began in 2002. Also, it corresponds to a reasonable time interval since 1995 when one
could expect to see further changes occurring in releases as a result of continuing regulatory
activities, voluntary actions on the part of industry, and facility closures.
1.1.2. Regulatory Summary
Tables 1-2 through 1-7 present a synopsis of EPA emission standards for the control of
dioxin releases. As discussed in Section 1.3.2, these regulations (along with other factors)
contributed to the reductions in dioxin emissions observed over time.
1-3
-------
Table 1-2. Municipal waste combustorsa
Category15
New large
Existing large
With electrostatic precipitators as the APCD
With dry scrubber/fabric filters as the APCD
New small
Existing small
With electrostatic precipitators as the APCD
With dry scrubber/fabric filters as the APCD
Stack emission limit0
(ng total
CDD/CDF/dscm)
13
60
30
13
60
30
Effective date
September 20, 1994d
June 19, 1996e
When SIPs are
approvedf
June 6, 2001s
When SIPs are
approved11
aAir emission standards promulgated December 19, 1995.
bLarge = aggregate capacity >225 tons/day; small = aggregate capacity <225 tons/day.
°ng total CDD/CDF/dscm = nanogram total C14 - C18 CDDs plus CDFs per dry standard cubic meter of stack gas
volume, corrected to 7% O2.
dBegan construction on this date.
TVIodified or upgraded on this date.
fWhen SIPs have been approved by EPA (approx. 3 yr from the final rule or 1998).
Tor facilities constructed on or before this date.
hWhen SIPs have been approved by EPA (approx. 3 yr from the final rule or 2003).
APCD = Air pollution control device
SIP = State Implementation Plan
1.1.3. Definition of Dioxin-Like Compound
This inventory of sources and environmental releases addresses specific compounds in
the following chemical classes: CDDs, CDFs, and polychlorinated biphenyls (PCBs). These
subsets of chemicals are defined as "dioxin like." Dioxin-like refers to the fact that these
compounds have similar chemical structures and physical-chemical properties, and they invoke a
common battery of toxic responses. Because of their hydrophobic nature and resistance towards
metabolism, these chemicals persist and bioaccumulate in the fatty tissues of animals and
humans. The CDDs include 75 individual compounds; CDFs include 135 compounds. These
individual compounds are technically referred to as congeners. Only 7 of the 75 congeners of
CDDs, or of brominated dibenzo-p-dioxins (BDDs), are thought to have dioxin-like toxicity; they
are the ones with chlorine substitutions in—at a minimum—the 2, 3, 7, and 8 positions. Only 10
of the 135 possible congeners of CDFs are thought to have dioxin-like toxicity; they also are the
ones with substitutions in the 2, 3, 7, and 8 positions. This suggests that 17 individual
CDDs/CDFs exhibit dioxin-like toxicity.
1-4
-------
Table 1-3. Hazardous waste incinerators and cement kilns and lightweight
aggregate kilns burning hazardous waste"
Source
Standards for new facilities'5
Standards for existing facilities'5
Hazardous waste
incinerators
0.11 ng I-TEQ/dscm for dry
APCD and/or waste heat boiler
sources
0.20 ng I-TEQ/dscm for all other
incinerators
0.20 ng I-TEQ/dscm and
temperature control <400°F at the
APCD inlet
0.40 ng I-TEQ/dscm when PM
control device operated >400°F
Cement kilns burning
hazardous waste
0.20 ng I-TEQ/dscm and
temperature control <400°F at
the APCD inlet
0.40 ng I-TEQ/dscm when PM
control device operated >400°F
0.20 ng I-TEQ/dscm and
temperature control <400°F at the
APCD inlet
0.40 ng I-TEQ/dscm when PM
control device operated >400°F
Lightweight aggregate kilns
burning hazardous waste
0.20 ng I-TEQ/dscm or rapid
quench below 400°F at kiln exit
0.20 ng I-TEQ/dscm or rapid
quench below 400°F at kiln exit
Air emission standards promulgated September 30, 1999, and December, 2005.
bng I-TEQ/dscm = nanogram I-TEQ per dry standard cubic meter of stack gas volume, corrected to 7% O2.
APCD = Air pollution control device (dry = dry scrubber or fabric filter)
PM = Paniculate matter
Table 1-4. Cement kilns not burning hazardous waste"
Existing cement kilnsb
0.20 ng I-TEQ/dscm and temperature control
<400°F at the APCD inlet
0.40 ng I-TEQ/dscm when PM control device
operated >400°F
New cement kilnsb
0.20 ng I-TEQ/dscm and temperature control
<400°F at the APCD inlet
0.40 ng I-TEQ/dscm when PM control device
operated >400°F
aAir emission standards promulgated June 14, 1999.
bng I-TEQ/dscm = nanogram I-TEQ per dry standard cubic meter of stack gas volume, corrected to 7% O2.
APCD = Air pollution control device
PM = Paniculate matter
There are 209 PCB congeners, of which only 12 are thought to have dioxin-like toxicity;
those with four or more lateral chlorine atoms with one or no substitution in the ortho position.
These compounds are sometimes referred to as coplanar, meaning that they can assume a flat
configuration with rings aligned along the same plane. The physical/chemical properties of each
congener vary according to the degree and position of chlorine substitution.
1-5
-------
Table 1-5. Secondary aluminum smelters3
Process
Sweat furnace
Thermal chip dryer
Scrap dryer/delacquering kiln/decoating kiln
Scrap dryer/delacquering kiln/decoating kiln
equipped with an afterburner
Emission standard
0.8 ng I-TEQ/dscm stack gas corrected to 7% O2
2.50 (ig I-TEQ per metric ton of scrap charged to the
dryer
0.25 g I-TEQ per metric ton of scrap charged to the
kiln
5.0 g I-TEQ per metric ton of scrap charged to the kiln
aAir emission standards promulgated March 23, 2000.
Table 1-6. Medical waste incinerators3
Category15
New
Small
Medium and large
Existing (all sizes)
Standard0
125 ng total CDD/CDF/dscm or
2.3 ng I-TEQ/dscm
25 ng total CDD/CDF/dscm or
0.6 ng I-TEQ/dscm
125 ng total CDD/CDF/dscm or
2.3 ng I-TEQ/dscm
When built
Constructed after June 20, 1996,
or existing units that
commenced modification after
March 16, 1998.
Constructed on or before
June 20, 1996; requires
approval of SIPsd
aAir emission standards promulgated September 15, 1997.
bSmall = capacity <100 kg/hr; medium = capacity > 100 to 227 kg/hr; large = capacity >227 kg/hr.
°ng/dscm = nanogram per dry standard cubic meter of stack gas volume, corrected to 7% O2.
dWhen SIPs have been approved by EPA (approx. 5 yr from the final rule or 2002).
Table 1-7. Pulp and paper mills3
Pollutant
Tetrachlorodibenzo-^-dioxin
Tetrachlorodibenzofuran
Maximum 1-day wastewater discharge
<5 parts per quadrillion
31.9 picograms per liter
aEffluent standards promulgated November 14, 1997.
1-6
-------
Generally speaking, this document focuses on the 17 CDDs/CDFs and a few of the
coplanar PCBs that are frequently encountered in source characterization or environmental
samples.
CDDs and CDFs are tricyclic aromatic compounds that have similar physical and
chemical properties. Certain PCBs (the so-called coplanar or mono-ortho coplanar congeners)
are also structurally and conformationally similar. The most widely studied of this general class
of compounds is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). TCDD, often called simply
"dioxin," represents the reference compound for this class of compounds. The structures of
TCDD and several related compounds are shown in Figure 1-1. Although sometimes confusing,
the term "dioxin" is often also used to refer to the complex mixtures of TCDD and related
compounds emitted from sources or found in the environment or in biological samples. It can
also be used to refer to the total TCDD "equivalents" found in a sample. This concept of toxicity
equivalence is discussed below.
ci
2,3,7,8-Tetrachlorodibenzo-p-dioxin
ci
2,3,7,8-Tetrachlorodibenzofuran
ci
1,2,3,7,8-Pentachlorodibenzo-p-dioxin
2,3,4,7,8-Pentachlorodibenzofu ran
CI
CI
CI
CI
CI/ \CI
S.S'.M'.S.S'-Hexachlorobiphenyl
ci
3,3',4,4',5-Pentachlorobiphenyl
Figure 1-1. Chemical structure of 2,3,7,8-TCDD and related compounds.
1-7
-------
1.1.4. Toxicity Equivalence Factors
CDDs, CDFs, and PCBs are commonly found as complex mixtures when detected in
environmental media and biological tissues or when measured as environmental releases from
specific sources. Humans are likely to be exposed to mixtures of CDDs, CDFs, and dioxin-like
PCB congeners that vary by source and pathway, complicating the assessment of human health
risk assessment. In order to address this problem, the concept of a "toxicity equivalence" (TEQ)
has been considered and discussed by the scientific community, and toxicity equivalence factors
(TEFs) have been developed and introduced to facilitate risk assessment of exposure to these
chemical mixtures.
On the most basic level, TEFs compare the potential toxicity of each dioxin-like
compound in the mixture to the well-studied and well-understood toxicity of TCDD, the most
toxic member of the group. The comparison procedure involves assigning individual TEFs to the
2,3,7,8-substituted CDD/CDF congeners and dioxin-like PCBs. To accomplish this, scientists
have reviewed the toxicological databases and, with considerations of chemical structure,
persistence, and resistance to metabolism, have agreed to ascribe specific "order of magnitude"
TEFs for each dioxin-like congener relative to TCDD, which is assigned a TEF of 1. The other
congeners have TEF values ranging from 1 to 0.00001.
Thus, these TEFs are the result of scientific judgment of a panel of experts using all of the
available data and are selected to account for uncertainties in the available data and to avoid
underestimating risk. In this sense, they can be described as "public health-conservative" values.
To apply this TEF concept, the TEF of each congener present in a mixture is multiplied by the
respective mass concentration, and the products are summed to represent the 2,3,7,8-TCDD TEQ
of the mixture (eq 1-1).
TEQ = ^i-n(Congener, x TEF) + (Congenerv x TEF) + (Congener„ x TEFJ (1-1)
The TEF values for CDDs and CDFs were originally adopted by international convention
(U.S. EPA, 1989a). These values were further reviewed and/or revised, and TEFs were also
developed for PCBs (Ahlborg et al., 1994; Van den Berg et al., 1998). A problem arises in that
past and present quantitative exposure and risk assessments may not have clearly identified
which of three TEF schemes was used to estimate the TEQ. This document uses a new uniform
TEQ nomenclature that clearly distinguishes between the different TEF schemes and identifies
the congener groups included in specific TEQ calculations. The nomenclature uses the following
abbreviations to designate which TEF scheme was used in the TEQ calculation:
-------
I-TEQ refers to the international TEF scheme adopted by EPA in 1989 (U.S. EPA,
1989a). See Table 1-8.
TEQ-WHO94 refers to the 1994 World Health Organization (WHO) extension of the
I-TEF scheme to include 13 dioxin-like PCBs (Ahlborg et al., 1994). See Table 1-9.
TEQ-WHO98 refers to the 1998 WHO update to the previously established TEFs for
dioxins, furans, and dioxin-like PCBs (Van den Berg et al., 1998). See Table 1-10.
Table 1-8. The TEF scheme for I-TEQDF
Dioxin congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
TEF
1
0.5
0.1
0.1
0.1
0.01
0.001
Furan congener
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
TEF
0.1
0.05
0.5
0.1
0.1
0.1
0.1
0.01
0.01
0.001
Table 1-9. The TEF scheme for dioxin-like PCBs, as determined by the
World Health Organization in 1994
Chemical structure
3,3',4,4'-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
IUPAC number
PCB-77
PCB-105
PCB-114
PCB-118
PCB-123
PCB-126
PCB-156
PCB-157
PCB-167
PCB-169
PCB-170
PCB-180
PCB-189
TEF
0.0005
0.0001
0.0005
0.0001
0.0001
0.1
0.0005
0.0005
0.00001
0.01
0.0001
0.00001
0.0001
1-9
-------
Table 1-10. The TEF scheme for TEQDFP-WHO
'98
Dioxin congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
Chemical structure
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,3,3',4,4',5,5'-HpCB
TEF
1
1
0.1
0.1
0.1
0.01
0.0001
]
Furan congener
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
UPAC number
PCB-77
PCB-81
PCB-105
PCB-114
PCB-118
PCB-123
PCB-126
PCB-156
PCB-157
PCB-167
PCB-169
PCB-189
TEF
0.1
0.05
0.5
0.1
0.1
0.1
0.1
0.01
0.01
0.0001
TEF
0.0001
0.0001
0.0001
0.0005
0.0001
0.0001
0.1
0.0005
0.0005
0.00001
0.01
0.0001
The nomenclature also uses subscripts to indicate which family of compounds is included
in any specific TEQ calculation. Under this convention, a subscript D is used to designate
dioxins, a subscript F to designate furans, and a subscript P to designate PCBs. As an example,
TEQDF-WHO98 would be used to describe a mixture for which only dioxin and furan congeners
were determined and where the TEQ was calculated using the WHO98 scheme. If PCBs had also
been determined, the nomenclature would be TEQDFP-WHO98. Note that the designations TEQDF-
WHO98 and I-TEQDF are interchangeable, as the TEFs for dioxins and furans are the same in each
scheme. Note also that in this document I-TEQ sometimes appears without the Dor F subscripts.
This indicates that the TEQ calculation includes both dioxins and furans. This document
emphasizes the WHO98 TEF scheme as the preferred scheme to be used to assign TEQs to
complex environmental mixtures.
1-10
-------
Throughout this document, environmental release estimates are presented in terms of
TEQs. This is done for convenience in presenting summary information and to facilitate
comparisons across sources. For purposes of environmental fate modeling, however, it is
important to use the individual CDD/CDF and PCB congener values rather than TEQs. This is
because the physical/chemical properties of individual CDD/CDF congeners vary and,
consequently, the congeners will behave differently in the environment. For example, the
relative mix of congeners released from a stack cannot be assumed to remain constant during
transport through the atmosphere and deposition to various media. The full congener-specific
release rates for most sources are given in an electronic database that will become available as a
companion to this document.
1.1.5. Information Sources
In general, the literature used to prepare this report includes documents published in 2003
or earlier. Some 2004 documents are cited, primarily in Chapter 2, which covers formation
theory, but a thorough literature review was not extended past 2003.
EPA's Toxics Release Inventory (TRI) began collecting data on PCBs in 1988 and on
CDDs/CDFs in 2000 (U.S. EPA, 2003c). These data were considered in this report for purposes
of identifying possible sources, but they were not used for making quantitative release estimates
because of the following considerations:
• With respect to PCBs, the TRI data are reported as total PCBs rather than on a
congener-specific basis. Thus, it is unknown what portion of these releases are
dioxin-like PCBs, and TEQs cannot be calculated. In their present format, the PCB
TRI data are not readily usable within the structure of this dioxin inventory.
• With respect to CDDs/CDFs, the reporting format under TRI is the sum quantity of
the 17 toxic CDDs/CDFs that are emitted in a given year (i.e., the sum of the 2,3,7,8-
chlorine-substituted compounds). Neither the releases of the individual CDD/CDF
congeners nor the TEQs must be reported; therefore, the dioxin TRI data are not
readily usable within the structure of this dioxin inventory.
• The accuracy of the TRI data is unknown because they are self-reported and are not
required to be based on measurements.
The TRI reports lack specific details and descriptions of the reporting industries. This
information is needed for the dioxin inventory because the calculation of source-
specific emission factors (representative of industrial source categories) strongly
depends on closely matching facilities in terms of similarity of process, production,
and pollution control.
1-11
-------
• The TRI reporting format does not include information on the strengths/weaknesses
of the data, and therefore, it would be difficult to evaluate these data in terms of the
confidence rating scheme developed for this inventory (presented in Section 1.2.3).
1.2. APPROACH
Only sources judged to have a reasonable likelihood for releases to the circulating
environment were addressed in this report. For example:
• CDDs/CDFs and dioxin-like PCBs in air emissions and wastewater discharges are
included, whereas those in intermediate products or internal wastestreams are not.
For example, the CDDs/CDFs in a wastestream going to an on-site incinerator are not
addressed in this report, but any CDDs/CDFs in the stack emissions from the
incinerator are included.
• CDDs/CDFs and dioxin-like PCBs in wastestreams applied to land in the form of
"land farming" are included, whereas those disposed of in permitted landfills were
excluded. Properly designed and operated landfills are considered to achieve long-
term isolation from the circulating environment. Land farming, however, involves the
application of wastes directly to land, clearly allowing for releases to therirculating
environment.
1.2.1. Source Classes
The major identified sources of environmental releases of dioxin-like compounds in the
United States are grouped into five broad categories.
Combustion. CDDs/CDFs are formed in most combustion systems (which can include
those that incinerate wastes such as municipal solid waste, sewage sludge, medical waste, and
hazardous wastes); in other high-temperature sources (such as cement kilns); in poorly or
uncontrolled combustion sources (such as forest fires, brush fires, landfill fires, accidental fires,
building fires, and open burning of wastes); and during the burning of various fuels (such as coal,
wood, and petroleum products).
Metals smelting, refining, and processing. CDDs/CDFs can be formed during various
types of primary and secondary metals operations, including iron ore sintering, lead smelting,
copper smelting, magnesium and titanium dioxide production, steel production, and scrap metal
recovery.
Chemical manufacturing. CDDs/CDFs can be formed as by-products of the
manufacture of chlorine-bleached wood pulp, chlorinated phenols (e.g., pentachlorophenol
[PCP]), PCBs, chlorobenzenes, phenoxy herbicides (e.g., 2,4-D and 2,4,5-T), and chlorinated
aliphatic compounds (e.g., ethylene dichloride, vinyl chloride, polyvinyl chloride).
1-12
-------
Natural sources and processes. The evidence for the widespread existence of natural
sources of dioxin is quite weak. Recent studies suggest that CDDs/CDFs can form under certain
environmental conditions (e.g., composting) from the action of microorganisms on chlorinated
phenolic compounds. Similarly, CDDs/CDFs have been reported to form during photolysis of
highly chlorinated phenols. Certain clays used in ceramics (e.g., ball clay) are believed to have
become contaminated with dioxin as a result of natural processes, but the source of
contamination remains unknown. Although it has been suggested that volcanos may be a natural
source, there is no reliable evidence that volcanos produce and emit significant amounts of
dioxin during eruptions.
Reservoirs. Reservoirs are environmental compartments and materials that have the
capacity to store previously formed CDDs/CDFs or dioxin-like PCBs. These compounds are
thus sequestered from the open and circulating environment. Potential reservoirs include soils,
sediments, and biota as well as some anthropogenic materials, such as PCP treated telephone
poles. Dioxin-like compounds in these reservoirs have the potential for redistribution and
circulation in the environment through the physical processes of leaching, volatilization, erosion,
sedimentation, and deposition. Whenever dioxins are released from their place of storage back
into the circulating environment, the reservoir is considered a source of dioxin.
Sources can also be categorized in terms of when releases occur: (1) contemporary
formation sources (sources that have essentially simultaneous formation and release) and (2)
reservoir sources (materials or places that contain previously formed CDDs/CDFs or dioxin-like
PCBs that are re-released to the environment). The contemporary formation sources are
discussed in Chapters 2 through 10 and the reservoir sources are discussed in Chapter 11.
Table 1-11 provides a comprehensive list of all known or suspected sources of
CDDs/CDFs in the United States. The checkmarks indicate how each source was classified in
terms of the following six categories:
1. Contemporary formation sources with reasonably well-quantified releases (see
Section 1.4.2). These sources are listed in Table 1-11 and release estimates are shown
in Table 1-12.
2. Contemporary formation sources without quantified release estimates. These sources
are listed in Table 1-12.
3. Reservoir sources with reasonably well-quantified releases. These sources would
have been listed in Table 1-11, but none have yet been identified.
1-13
-------
Table 1-11. Known and suspected sources of CDDs/CDFs
Source category
Contemporary formation sources
Quantifiable
(Categories
A, B and C)
Preliminary
estimate
(Category D)
Not
quantifiable
(Category E)
Reservoir sources
Quantifiable
(Categories
A, B and C)
Preliminary
estimate
(Category D)
Not
quantifiable
(Category E)
COMBUSTION SOURCES
Waste incineration
Municipal waste combustion
Hazardous waste incineration
Boilers/industrial furnaces
Medical waste/pathological incineration
Crematoria
Sewage sludge incineration
Tire combustion
Pulp and paper mill sludge incinerators
Biogas combustion
Power/energy generation
Vehicle fuel combustion - leaded8
- unleaded
- diesel
Wood combustion - residential
- industrial
Coal combustion - residential
- industrial/utility
Oil combustion - residential
- industrial/utility
Other high-temperature sources
Cement kilns burning hazardous waste
Cement kilns burning nonhazardous waste
Asphalt mixing plants
Petroleum refining catalyst regeneration
Cigarette combustion
Carbon reactivation furnaces
Kraft recovery boilers
Manufacture of ball clay products
Glass manufacturing
Lime kilns
Rubber manufacturing
£
/
£
£
'
,
/
/
/
/
/
/
'
/
/
/
'
/
/
-------
Table 1-11. Known and suspected CDD/CDF sources (continued)
Source category
Minimally controlled or uncontrolled combustion
Combustion of landfill gas in flares
Landfill fires
Accidental fires, structural
Accidental fires, vehicles
Forest, brush, and straw fires
Backyard barrel burning
Uncontrolled combustion of PCBs
Burning of candles
Contemporary formation sources
Quantifiable
(Categories
A, B and C)
/
Preliminary
estimate
(Category D)
\
Not
quantifiable
(Category E)
Reservoir sources
Quantifiable
(Categories
A, B and C)
Preliminary
estimate
(Category D)
Not
quantifiable
(Category E)
METAL SMELTING/REFINING
Ferrous metal smelting/refining
Sintering plants
Coke production
Electric arc furnaces
Ferrous foundries
Nonferrous metal smelting/refining
Primary aluminum
Primary copper
Primary magnesium
Primary nickel
Secondary aluminum
Secondary copper
Secondary lead
Scrap electric wire recovery
Drum and barrel reclamation
'
/
/
/
'
'
/
-------
Table 1-11. Known and suspected CDD/CDF sources (continued)
Source category
Contemporary formation sources
Quantifiable
(Categories
A, B and C)
Preliminary
estimate
(Category D)
Not
quantifiable
(Category E)
Reservoir sources
Quantifiable
(Categories
A, B and C)
Preliminary
estimate
(Category D)
Not
quantifiable
(Category E)
CHEMICAL MANUFACTURING (releases to the environment)
Bleached chemical wood pulp and paper mills
Mono- through tetrachlorophenols
Pentachlorophenol
Chlorobenzenes
Chlorobiphenyls (leaks/spills)
Ethylene dichloride/vinyl chloride
Dioxazine dyes and pigments
2,4-Dichlorophenoxy acetic acid
Municipal wastewater treatment
Tall oil-based liquid soaps
BIOLOGICAL AND PHOTOCHEMICAL
PROCESSES
/
/
/
/
/
/
/
/
/
/
/
/
RESERVOIR SOURCES
Land
Air
Water
Sediments
Anthropogenic structures
PCP-treated wood
/
/
/
/
/
aLeaded fuel production and the manufacture of motor vehicle engines requiring leaded fuel for highway use are prohibited in the United States. (See Chapter 4,
Section 4.1, for details.)
-------
Table 1-12. Inventory of contemporary releases (g/yr) of dioxin-like compounds from known sources in the
United States for reference years 2000,1995, and 1987 and preliminary release estimates for 2000
Source Category
2000 Inventory
TEQDF-
WH098
I-TEQ
Category
rating3
1995 Inventory
TEQDF-
WH098
I-TEQ
Category
rating3
1987 Inventory
TEQDF-
WHO98
I-TEQ
Category
rating3
2000
Preliminary
indication
Category D
rating
TEQDF-WH098
RELEASES TO AIR
WASTE INCINERATION
Municipal waste combustion
Hazardous waste incineration
Boilers/industrial furnaces
Halogen acid furnaces
Medical waste/pathological
incineration
Crematoria
-human
-animal
Sewage sludge incineration
Tire combustion
Pulp and paper mill sludge
incinerators15
Biogas combustion
83.8
3.2
1.8
0.3
378
0.3
9.6
0.5
76.3
3.2
1.8
0.3
357
0.3
9.4
0.5
A
B
C
C
C
C
B
C
1,393.5
5.8
0.4
NA
487
0.2
14.2
0.1
1,101.3
5.7
0.4
NA
459
0.2
14
0.1
B
B
C
C
C
B
C
POWER/ENERGY GENERATION
Vehicle fuel combustion
- leaded gasoline0
- unleaded gasoline on-road
- unleaded gasoline off-road
- diesel on-road (Trucks)
- diesel off-road
- equipment
- railroad
- commercial marine
vessel
7
0.4
65.4
22
6.8
4.3
6.7
0.4
61.7
21
6.4
4
C
C
C
C
C
C
1.6
4.7
NA
33.3
12
7
4.8
1.3
4.4
NA
31.5
11
6.6
4.5
C
C
C
C
C
C
8,905.1
5.0
0.8
NA
2,570
0.2
5.8
0.1
7,858.8
5.0
0.8
NA
2,440
0.1
5.8
0.1
B
B
C
C
C
B
C
37.5
3.6
NA
27.8
9.4
5.8
3.8
31.9
3.3
NA
26.3
8.8
5.5
3.6
C
C
C
C
C
C
<1
<1
-------
Table 1-12. Inventory of contemporary releases (g/yr) of dioxin-like compounds from known sources in the
United States for reference years 2000,1995, and 1987 and preliminary release estimates for 2000 (continued)
Source Category
Wood combustion
- residential
- industrial
Coal combustion
- utility boilers
- residential6
- commercial/industrial
Oil combustion
- industrial/utility, residual oil
- industrial/utility, distillate oil
- institutional/commercial
heating, residual oil
- institutional/commercial
heating, distillate oil
- residential, distillate oil
2000 Inventory
TEQDF-
WH098
11.3d
41.5
69.5
1.7
7.3
0.7
2.9
4.5
I-TEQ
11.3d
39.4
70.4
1.5
6.3
0.6
2.5
3.6
Category
rating3
C
C
B
C
C
C
C
C
1995 Inventory
TEQnF-
^<-L)r
WH098
15.7d
26.2
60.1
10.7
7.3
0.8
3.1
5.0
I-TEQ
15.7d
24.9
60.9
9.3
6.4
0.7
2.7
3.9
Category
rating3
C
C
B
C
C
C
C
C
1987 Inventory
TEQDF-
WHO98
22d
26.5
50.9
17.8
8.3
1.5
3.7
5.4
I-TEQ
22d
25.2
51.4
15.5
7.2
1.3
3.2
4.2
Category
rating3
C
C
B
C
C
C
C
C
2000
Preliminary
indication
Category D
rating
TFODF-WH09S
<10d
>10d
OTHER HIGH-TEMPERATURE SOURCES
Cement kilns burning hazardous
waste
Lightweight aggregate kilns
burning hazardous waste
Cement kilns burning
nonhazardous waste
Asphalt mixing plants
Petroleum refining catalyst
regeneration
Cigarette combustion
Carbon reactivation furnaces
Kraft recovery boilers
18.8
1.9d
17.2
2.2
0.4
O.ld
0.8
16.6
1.8d
16.6
2.1
0.4
O.ld
0.8
B
C
C
C
C
C
B
156.1
2.4d
16.6
2.2
0.8
O.ld
2.3
145.3
2.4d
15.9
2.1
0.8
O.ld
2.3
C
C
C
C
C
C
B
117.8
3.3d
12.7
2.2
1
O.ld
2
109.6
3.3d
12.3
2.1
1
O.ld
2
C
C
C
C
C
C
B
-------
Table 1-12. Inventory of contemporary releases (g/yr) of dioxin-like compounds from known sources in the
United States for reference years 2000,1995, and 1987 and preliminary release estimates for 2000 (continued)
Source Category
2000 Inventory
TEQDF-
WH098
I-TEQ
Category
rating3
1995 Inventory
TEQDF-
WHO98
I-TEQ
Category
rating3
1987 Inventory
TEQDF-
WHO98
I-TEQ
Category
rating3
2000
Preliminary
indication
Category D
rating
TFODF-WH098
MINIMALLY CONTROLLED OR UNCONTROLLED COMBUSTION1
Combustion of landfill gas
Landfill fires
Accidental fires
- structural
- vehicles
Forest and brush fires'1
Backyard barrel burning1
Residential yard waste burningj
Land clearing debris burning
498.5
472.6
C
628
595
C
METALLURGICAL PROCESSES
Ferrous metal smelting/
refining
- sintering plants
- coke production
- electric arc furnaces
- foundries
Nonferrous metal smelting/
refining
- primary copper
- secondary aluminum
- secondary copper
- secondary lead
- primary magnesium
Drum and barrel reclamation
27.6
0.3d
8.3
0.9
2.5
4.3d
0.6
24.4
0.3d
7.8
0.9
2.4
4.3d
0.6
A
B
C
C
B
A
C
28
<0.5d
19.5
271
1.7
4.1d
0.1
25.1
<0.5d
18.3
266
1.6
4.1d
0.1
B
B
C
C
B
C
C
604
573
C
32.7
<0.5d
10.9
983
1.3
NA
0.1
29.3
<0.5d
10.2
966
1.2
NA
0.1
C
B
C
C
B
C
d
>lpDOOg
>10d
d
^,©00
^19oo
<10d
<100
>10d
-------
Table 1-12. Inventory of contemporary releases (g/yr) of dioxin-like compounds from known sources in the
United States for reference years 2000,1995, and 1987 and preliminary release estimates for 2000 (continued)
Source Category
2000 Inventory
TEQDF-
WH098
I-TEQ
Category
rating3
1995 Inventory
TEQnF-
WH098
I-TEQ
Category
rating3
1987 Inventory
TEQDF-
WHO98
I-TEQ
Category
rating3
2000
Preliminary
indication
Category D
rating
TEQDF-WH098
CHEMICAL MANUFACTURE/PROCESSING SOURCES
Ethylene dichloride/vinyl
chloride/PVC
Chor-alkali facilities
TOTAL RELEASES TO AIR"
5.5d
1.8d
1,314.5
5.5
1.8d
1,243.6
A
A
A,B, C
11.2d
1.8 d
3,239.9
11.2
1.8 d
2,857.1
A
C
A,B,C
NA
NA
13,482.6
NA
NA
12,230.7
A,B, C
RELEASES TO WATER
CHEMICAL MANUFACTURE/PROCESSING SOURCES
Bleached chemical wood pulp
and paper mills
POTW (municipal) wastewater
Ethylene dichloride/vinyl
chloride/PVC
Chor-alkali facilities
TOTAL RELEASES TO
WATER"
1.0
23. ld
1.8d
25.9
1.0
23. 9d
1.8d
26.7
A
A
A
A,B, C
28
23. ld
1.8d
52.9
28
23.9d
d
1 °
53.7
A
C
C
A,B, C
356
NA
NA
356
356
NA
NA
356
A
A,B, C
>10
RELEASES TO LAND
CHEMICAL MANUFACTURING/PROCESSING SOURCES
Bleached chemical wood pulp
and paper mill sludge
Ethylene dichloride/vinyl
chloride/PVC
Municipal wastewater treatment
sludge
Commercially marketed sewage
sludge
0.1
1.4
78.2
1.9
0.1
1.5
78.2
1.9
A
A
A
A
2
1.4
116.1
3
2
1.5
156.5
4
A
B
A
A
14.1
NA
76.6
2.6
14.1
NA
103
3.5
A
A
A
to
o
-------
Table 1-12. Inventory of contemporary releases (g/yr) of dioxin-like compounds from known sources in the
United States for reference years 2000,1995, and 1987 and preliminary release estimates for 2000 (continued)
Source Category
2,4-Dichlorophenoxy acetic acid
TOTAL RELEASES TO LAND"
OVERALL RELEASES TO
OPEN AND CIRCULATING
ENVIRONMENT
2000 Inventory
TEQDF-
WH098
NA
81.6
1,422.0
I-TEQ
NA
81.7
1,352.0
Category
rating3
A,B, C
A,B,C
1995 Inventory
TEQDF-
WH098
28.9
151.4
3,444.2
I-TEQ
18.4
182.4
3,093.2
Category
rating3
A
A,B, C
A,B,C
1987 Inventory
TEQDF-
WHO98
33.4
126.7
13,965.3
I-TEQ
21.3
141.9
12728.6
Category
rating3
A
A,B, C
A,B,C
2000
Preliminary
indication
Category D
rating
TFODF-WH09S
The most reliable estimates of environmental releases are those sources in categories A, B, and C.
Included in estimate for wood combustion, industrial.
°Leaded fuel production and the manufacture of motor vehicle engines requiring leaded fuel for highway use are prohibited in the United States. (See Chapter 4,
Section 4.1 for details.)
dEstimate based on a TEQDF-WHO98 emissions estimate.
Includes combustion of bituminous/subbituminous coal and anthracite coal.
'Refers to conventional pollutant control, not dioxin emissions control. Very few sources listed in this inventory control specifically for CDD/CDF emissions.
BCongener-specific emissions data were not available; the Nordic TEQ estimate was used as a surrogate for the I-TEQDF emissions estimate.
Includes forest wildfires and prescribed burning for forest management.
Term refers to the burning of residential waste in barrels.
Includes burning of brush and leaves.
Total reflects only the total of the estimates made in this report.
Category ratings:
A = Characterization of the source category judged to be adequate for quantitative estimation with high confidence in the emission factor and high
confidence in activity level.
B = Characterization of the source category judged to be adequate for quantitative estimation with medium confidence in the emission factor and at least
medium confidence in activity level.
C = Characterization of the source category judged to be adequate for quantitative estimation with low confidence in either the emission factor and/or the
activity level.
D = Preliminary indication of the potential magnitude of emissions from "Unqualified" (Category D) sources; based on extremely limited data, judged to be
clearly nonrepresentative. D estimates are not included in the inventory of source emissions, but serve the purpose of highlighting sources in need of more
adequate emissions information.
NA = Not available (information is lacking).
POTW = Publicly owned treatment works or sewage treatment plant.
-------
4. Reservoir sources with preliminary release estimates. These sources are discussed in
Chapter 11.
5. Reservoir sources without quantified releases. These sources are discussed in
Chapter 11.
Only contemporary formation sources (numbers 1 and 2 above) are considered for
inclusion in the national inventory. Reservoir sources are not considered because they are not
original releases, but rather the recirculation of past releases. To date, no reliable estimates of
releases from the reservoir sources have been made because information is either lacking or is
inadequate to allow for estimates to be made.
This document includes discussions on products that contain dioxin-like compounds.
Some of these products, such as 2,4-D, are considered to be sources because they are clearly used
in ways that result in environmental releases (e.g., they are sprayed onto agricultural lands for
weed control). If a release from the product occurs, it is added to the national dioxin inventory.
Other products containing dioxin-like compounds, such as vinyl chloride products, do not appear
to have environmental releases and are not considered sources. For all CDD-/CDF-containing
products, this document summarizes the available information about contamination levels and,
where possible, makes estimates of the total amount of CDDs/CDFs produced annually in these
products. Estimates of the CDD/CDF TEQ amounts in products are summarized in Table 1-13.
Table 1-13. Products containing CDDs/CDFs (g TEQDF-WHO98/yr)
Product
Bleached chemical wood pulp
Ethylene dichloride/vinyl chloride
Chloranil
Pentachlorophenol
2,4 -Dichlorophenoxy acetic acid (2,4-D)a
TOTAL
2000
0.58
0.02
1.16
4,395
NA
4,397
1995
40
0.02
64
4,800
28.9
4,933
1987
505
NA
NA
20,000
33.4
20,538
a Only 2,4-D is considered to be an environmental release.
NA = Information not available
1.2.2. Quantitative Method for Inventory of Sources
Some source types have a high percentage of facilities with measured CDD/CDF releases,
such as municipal waste combustion, hazardous waste incineration, and cement kilns that burn
hazardous waste (air emissions), and wastewater releases from chlorine-bleached pulp and paper
mills. In addition, some source activities have been adequately sampled with respect to levels of
1-22
-------
lands and 2,4-D applied to agricultural lands. Other source categories have relatively few tested
facilities and/or the activity has not been comprehensively evaluated for dioxin releases. In these
cases, EPA relies on the use of emission factors to estimate CDD/CDF releases from the untested
sources. This provides a method of extrapolation from tested sources to national estimates of
environmental releases. Many of the national emission estimates, therefore, have been developed
using this "top-down" approach.
The first step in this approach is to derive from the available emissions monitoring data
an emission factor (or series of emission factors) deemed to be representative of the source
category (or segments of a source category that differ in, e.g., configuration, fuel type, air
pollution control equipment). The emission factor relates mass of CDDs/CDFs or dioxin-like
PCBs released into the environment with some measure of activity (e.g., kilograms of material
processed per year, vehicle miles traveled per year, liters of wastewater discharged per year). It is
developed by averaging the emission factors for the tested facilities or activities within the
particular classification of sources. For example, mass burn MWCs equipped with dry scrubbers
(DSs) combined with fabric filters (FFs) will have an average emission factor derived from the
tested facilities within this source classification. This average emission factor is then multiplied
by the measure of activity for the nontested facilities in the class (e.g., total kilograms of material
processed by these facilities annually). Finally, emissions are summed for the tested facilities
and nontested facilities. In general, this procedure can be represented by the followinge quations:
• 'total ~ LJ- 'tested, I Lj 'untested, I
^ total ~ Lj^testedJ L^ i *-i untested
where:
Etotai = annual emissions from all facilities (g TEQ/yr)
Etested j = annual emissions from all tested facilities in class I (g TEQ/yr)
Euntested j = annual emissions from all untested facilities in class I (g TEQ/yr)
EF; = mean emission factor for tested facilities in class I (g TEQ/kg)
Aj = activity measure for untested facilities in class I (kg/yr)
Figures 1-2 through 1-4 and 1-6 through 1-8 depict the various source categories and their
emission factors, activity levels, and annual emissions for reference years 1987, 1995, and 2000,
respectively, in I-TEQ and WHO-TEQ units. Figures 1-5 and 1-9 depict comparisons of the
estimated I-TEQ and WHO-TEQ air emissions for these years.
1-23
-------
Emission Source
(tested / total units)
Municipal Waste Combustion (19/113)
Medical Waste Incineration (8 / 5000)
Secondary Copper Smelting (2 / 4)
Backyard Barrel Burning (NA)
Cement Kilns Burning Haz Waste (10/?)
Utility / Industrial Coal Combustion (11 / ?)
On-road Leaded Gas Fuel Combustion (? / ?)
Iron Ore Sinter Plants (2 / ?)
On-road Diesel Fuel Combustion (NA)
Industrial Wood Burning (9 / ?)
Residential Wood Burning (7 / 25000000)
Off-road Diesel Fuel Combustion (NA)
Utility / Industrial Residual Oil Combustion (? / ?)
Secondary Aluminum Smelting (6 / 67)
Cement Kilns Not Burning Haz Waste (15 / ?)
Utility / Industrial Distillate Oil Combustion (>2 / ?)
Sevvage Sludge Incineration (13/199)
Hazardous Waste Incineration (17/171-227)
On-road Unleaded Gas Fuel Combustion (? / ?)
Manufacture of EDC/VC (? / ?)
Best Estimate of
I-TEQ Emission Factor
(ng/kg or ng/L)
Total Annual "Activity"
(thousand metric tons/yr or
million L/yr)
s s s
^ ° 8
The figures include sources with annual I-TEQ emission estimates greater than 5 g l-TEQ/yr in
one or both of the Reference Years 1987 or 1995. Derivation of the emission factors
and annual "Activity" estimates (e.g., kg of waste incinerated) are presented in the following
chapters of this report. The difference in bar shading indicates the degree of confidence in the
estimate. The set of numbers following the source categories indicates the number of
facilities/sites for which emission test data are available versus the number of facilities/sites
in the category. A question mark (?) indicates that the precise number of facilities/sites
could not be estimated.
Annual I-TEQ Emission
(g l-TEQ/yr)
•
<
»
1 S S S S S i
- o o o o o c
S § § § i
T-" o" O" C
S i
Low Confidence
Medium Confidence
High Confidence
Figure 1-2. Estimated CDD/CDF I-TEQ emissions to air from combustion
sources in the United States for reference year 1987 (municipal solid waste
incineration is currently referred to as municipal waste combustion).
Some source categories are made up of facilities that vary widely in terms of design and
operating conditions. For these sources, as explained above, an attempt was made to create
subcategories that grouped facilities with common features and then to develop separate emission
factors for each subcategory. Implicit in this procedure is the assumption that facilities with
similar design and operating conditions should have similar CDD/CDF release potential.For
most source categories, however, the specific combination of features that contributes mostto
1-24
-------
Best Estimate of
I-TEQ Emission Factor
(ng/kg or ng/L)
Emission Source
(tested / total units)
Municipal Waste Combustion (39 /130)
Medical Waste Incineration (20 / 2400)
Secondary Copper Smelting (2 / 3)
Backyard Barrel Burning (NA)
Cement Kilns Burning Haz Waste (10/34)
Utility / Industrial Coal Combustion (11 / ?)
On-road Leaded Gas Fuel Combustion (? / ?)
Iron Ore Sinter Plants (2/11)
On-road Diesel Fuel Combustion (NA)
Industrial Wood Burning (9 / ?)
Residential Wood Burning (7 / 25000000)
Off-road Diesel Fuel Combustion (NA)
Utility / Industrial Residual Oil Combustion (? / ?)
Secondary Aluminum Smelting (6 / 76)
Cement Kilns Not Burning Haz Waste (15 /178)
Utility / Industrial Distillate Oil Combustion (>2 / ?)
Sewage Sludge Incineration (13 / 257)
Hazardous Waste Incineration (17/162)
On-road Unleaded Gas Fuel Combustion (? / ?)
Manufacture of EDC/VC (? / ?)
The figures include sources with annual I-TEQ emission estimates greater than 5 g l-TEQ/yr in
one or both of the Reference Years 1987 or 1995. Derivation of the emission factors
and annual "Activity" estimates (e.g., kg of waste incinerated) are presented in the following
chapters of this report. The difference in bar shading indicates the degree of confidence in the
estimate. The set of numbers following the source categories indicates the number of
facilities/sites for which emission test data are available versus the number of facilities/sites
in the category. A question mark (?) indicates that the precise number of facilities/sites
could not be estimated.
Total Annual "Activity"
(thousand metric tons/yr or
million L/yr)
- 2 8 8
Annual I-TEQ Emission
(g l-TEQ/yr)
•>
«
>
- o o o o o o
S § § § §
S § §
T- O
Legend
1 | Low Confidence
| | Medium Confidence
^^^^^ High Confidence
Figure 1-3. Estimated CDD/CDF I-TEQ emissions to air from combustion
sources in the United States for reference year 1995 (municipal solid waste
incineration is currently referred to as municipal waste combustion).
CDD/CDF or dioxin-like PCB releases is not well understood. Therefore, how to best
subcategorize a source category was often problematic. For each subcategorized source category
in this document, a discussion is presented about the variability in design and operating
conditions, what was known about how these features contributed to CDD/CDF or dioxin-like
PCB releases, and the rationale for creating subcategories.
1-25
-------
Emission Source
(tested/ total units)
Municipal Waste Combustion (195/251)
Medical Waste Incineration (22 / ?)
Secondary Copper Smelting (1 / 2)
Backyard Barrel Burning (NA)
Cement Kilns Burning Haz Waste (10 / ?)
Utility/ Industrial Coal Combustion (11 /?)
On-road Leaded Gas Fuel Combustion (0 / 0)
Iron Ore Sinter Plants (2/11)
On-road Diesel Fuel Combustion (NA)
Industrial Wood Burning (9 / ?)
Residential Wood Burning (19 / 25000000)
Off-road Diesel Fuel Combustion (NA)
Utility/ Industrial Residual Oil Combustion (?/?)
Secondary Aluminum Smelting (6 /?)
Cement Kilns Not Burning Haz Waste (15/?)
Utility / Industrial Distillate Oil Combustion (>2 / ?)
Sewage Sludge Incineration (14 / ?)
Hazardous Waste Incineration (22 /132)
On-road Unleaded Gas Fuel Combustion (? / ?)
Manufacture of EDC/VC (8/12)
Best Estimate of
I-TEQ Emission Factor
(ng/kg or ng/L)
Total Annual "Activity"
(thousand metric tons/yr or
million L/yr)
Annual I-TEQ Emission
(g l-TEQ/yr)
s s s s s s
^ ? s s s 8
^ ? 8 s
2 8
s s
? 8
The figures include sources with annual I-TEQ emission estimates greater than 5 g l-TEQ/yr in
one or both of the Reference Years 1987 or 1995. Derivation of the emission factors
and annual "Activity" estimates (e.g., kg of waste incinerated) are presented in the following
chapters of this report. The difference in bar shading indicates the degree of confidence in the
estimate. The set of numbers following the source categories indicates the number of
facilities/sites for which emission test data are available versus the number of facilities/sites
in the category. A question mark (?) indicates that the precise number of facilities/sites
could not be estimated.
Legend
Low Confidence
Medium Confidence
High Confidence
Figure 1-4. Estimated CDD/CDF I-TEQ emissions to air from combustion
sources in the United States for reference year 2000.
The emission factors developed for the inventory are intended to be used for estimating
total emissions for a source category rather than emissions from individual facilities. EPA has
made uncertainty determinations for each of these emission factors, based, in part, on the
assumption that by applying them to a group of facilities, the potential for overestimating or
underestimating individual facilities will, to some extent, be self-compensating. This means that
in using these emission factors one can place significantly greater confidence in an emission
1-26
-------
Municipal Waste Combustion
Medical Waste Incineration
Secondary Copper Smelting
Backyard Barrel Burning
Cement Kilns Burning Haz Waste
Utility / Industrial Coal Combustion
On-road Leaded Gas Fuel Combustion
Iron Ore Sinter Plants
On-road Diesel Fuel Combustion
Industrial Wood Burning
Residential Wood Burning
Off-road Diesel Fuel Combustion
Utility / Industrial Residual Oil Combustion
Secondary Aluminum Smelting
Cement Kilns Not Burning Haz Waste
Utility / Industrial Distillate Oil Combustion
Sewage Sludge Incineration
Hazardous Waste Incineration
On-road Unleaded Gas Fuel Combustion
Manufacture of EDC/VC
0.1
10
100
1000
10000
11987 D1995 D2000
Figure 1-5. Comparison of estimates of annual I-TEQ emissions to air
(g I-TEQ/yr) for reference years 1987,1995, and 2000.
estimate for a class than in an estimate for any individual facility. Given the limited amount of
data available for deriving emission factors and the limitations of our understanding about
facility-specific conditions that determine formation and control of dioxin-like compounds, the
current state of knowledge cannot support the development of emission factors that can be used
to accurately estimate emissions on an individual facility-specific basis. The emission factors
developed for each of the categories discussed in this national emissions inventory are listed in I-
TEQ and TEQDF-WHO98 in Tables 1-14 and 1-15, respectively.
1.2.3. Confidence Ratings
Each source emission calculation required estimates of an emission factor and an activity
level. For each emission source, the quantity and quality of the available information for both
vary considerably. Consequently, it is important that emission estimates be accompanied by
some indicator of the uncertainties associated with their development. For this reason, a
1-27
-------
Emission Source
(tested I total units)
Municipal Waste Combustion (19/105)
Medical Waste Incineration (8 / 5000)
Secondary Copper Smelting (2 / 4)
Backyard Barrel Burning (NA)
Cement Kilns Burning Haz Waste (10 / ?)
Utility / Industrial Coal Combustion (11 / ?)
On-road Leaded Gas Fuel Combustion (? / ?)
Iron Ore Sinter Plants (2 / ?)
On-road Diesel Fuel Combustion (NA)
Industrial Wood Burning (9 / ?)
Residential Wood Burning (7 / 25000000)
Off-road Diesel Fuel Combustion (NA)
Utility / Industrial Residual Oil Combustion (? / ?)
Secondary Aluminum Smelting (6 / 67)
Cement Kilns Not Burning Haz Waste (15/?)
Utility / Industrial Distillate Oil Combustion (>2 / ?)
Sewage Sludge Incineration (13/199)
Hazardous Waste Incineration (17/171-227)
On-road Unleaded Gas Fuel Combustion (? / ?)
Manufacture of EDC/VC (? / ?)
Best Estimate of
WHO-TEQ Emission Factor
(ng/kg or ng/L)
Total Annual "Activity"
(thousand metric tons/yr or
million L/yr)
Annual WHO-TEQ Emission
(g WHO-TEQ/yr)
- S § §
§ §
§ §
The figures include sources with annual WHO-TEQ emission estimates greater than 5 g WHO-TEQ/yr in
one or both of the Reference Years 1987 or 1995. Derivation of the emission factors
and annual "Activity" estimates (e.g., kg of waste incinerated) are presented in the following
chapters of this report. The difference in bar shading indicates the degree of confidence in the
estimate. The set of numbers following the source categories indicates the number of
facilities/sites for which emission test data are available versus the number of facilities/sites
in the category. A question mark (?) indicates that the precise number of facilities/sites
could not be estimated.
Legend
Low Confidence
Medium Confidence
High Confidence
Figure 1-6. Estimated CDD/CDF WHO-TEQ emissions to air from
combustion sources in the United States for reference year 1987 (municipal
solid waste incineration is currently referred to as municipal waste combustion).
qualitative confidence rating scheme was developed as an integral part of the emissions estimate
with the following considerations.
Emission factor. The uncertainty in the emission factor estimate depends primarily on
how well the tested facilities represent the untested facilities. In general, confidence in the
emission factor increases with increases in the number of tested facilities relative to the total
number of facilities. Variability in terms of physical design and operating conditions within a
1-28
-------
Emission Source
(tested / total units)
a\ Waste Combustion (39 /130)
Waste Incineration (20 / 2400)
ary Copper Smelting (2 / 3)
•d Barrel Burning (NA)
Kilns Burning Haz Waste (10 / 34)
ndustrial Coal Combustion (11 / ?)
I Leaded Gas Fuel Combustion (? / ?)
i Sinter Plants (2/11)
I Diesel Fuel Combustion (NA)
al Wood Burning (9/?)
itial Wood Burning (7 / 25000000)
I Diesel Fuel Combustion (NA)
ndustrial Residual Oil Combustion (? / ?)
ary Aluminum Smelting (6/76)
Kilns Not Burning Haz Waste (15/178)
ndustrial Distillate Oil Combustion (>2 / ?)
• Sludge Incineration (13 / 257)
)us Waste Incineration (17 /162)
I Unleaded Gas Fuel Combustion (? / ?)
:tureofEDC/VC(?/?)
Best Estimate of
WHO-TEQ Emission Factor
(ng/kg or ng/L)
Total Annual "Activity"
(thousand metric tons/yr or
million L/yr)
Annual WHO-TEQ Emission
(g WHO-TEQ/yr)
588
8 8
= 5 - ° §
§ §
§ § §
- S §
jures include sources with annual WHO-TEQ emission estimates greater than 5 g WHO-TEQ/yr in
both of the Reference Years 1987 or 1995. Delation of the emission factors
inual "Activity" estimates (e.g., kg of waste incinerated) are presented in the following
;rs of this report. The difference in bar shading indicates the degree of confidence in the
ite. The set of numbers following the source categories indicates the number of
;s/sites for which emission test data are available versus the number of facilities/sites
category. A question mark (?) indicates that the precise number of facilities/sites
not be estimated.
Low Confidence
Medium Confidence
High Confidence
Figure 1-7. Estimated CDD/CDF WHO-TEQ emissions to air from
combustion sources in the United States for reference year 1995.
class or subclass must also be considered. The more variability among facilities, the less
confidence that a test of any single facility is representative of that class or subclass. The quality
of the supporting documentation also affects uncertainty. Whenever possible, original
engineering test reports were used. Peer-reviewed reports from the open literature were also used
for developing some emission factors. In some cases, however, draft reports that had undergone
more limited review were also used. In a few cases, unpublished references (such as personal
communication with experts) were used and are clearly noted in the text.
1-29
-------
Emission Source
(tested / total units)
Municipal Waste Combustion (195 / 251)
Medical Waste Incineration (22 / ?)
Secondary Copper Smelting (1 / 2)
Backyard Barrel Burning (NA)
Cement Kilns Burning Haz Waste (20 / 22)
Utility / Industrial Coal Combustion (11 / ?)
On-road Leaded Gas Fuel Combustion (0 / 0)
Iron Ore Sinter Plants (4/11)
On-road Diesel Fuel Combustion (NA)
Industrial Wood Burning (9 / ?)
Residential Wood Burning (19 / ?)
Off-road Diesel Fuel Combustion (NA)
Utility/Industrial Residual Oil Combustion (?/?)
Secondary Aluminum Smelting (6 / ?)
Cement Kilns Not Burning Haz Waste (15/?)
Utility / Industrial Distillate Oil Combustion (>2 / ?)
Sewage Sludge Incineration (14 / ?)
Hazardous Waste Incineration (22 /132)
On-road Unleaded Gas Fuel Combustion (? / ?)
Manufacture of EDC/VC/PVC (8/12)
Best Estimate of
WHO-TEQ Emission Factor
(ng/kg or ng/L)
Total Annual "Activity"
(thousand metric tons/yr or
million L/yr)
Annual WHO-TEQ Emission
(g WHO-TEQ/yr)
o o o o
1111
^ 2 s s
888
° 8
The figures include sources with annual WHO-TEQ emission estimates greater than 5 g WHO-TEQ/yr in
one or both of the Reference Years 1987 or 1995. Derivation of the emission factors
and annual "Activity" estimates (e.g., kg of waste incinerated) are presented in the following
chapters of this report. The difference in bar shading indicates the degree of confidence in the
estimate. The set of numbers following the source categories indicates the number of
facilities/sites for which emission test data are available versus the number of facilities/sites
in the category. A question mark (?) indicates that the precise number of facilities/sites
could not be estimated.
Legend
Low Confidence
Medium Confidence
High Confidence
Figure 1-8. Estimated CDD/CDF WHO-TEQ emissions to air from
combustion sources in the United States for reference year 2000.
Activity level. The uncertainty in the activity level estimate was judged primarily on the
basis of the extent of the underlying data. Estimates derived from comprehensive surveys
(including most facilities in a source category) were assigned high confidence. As the number of
facilities in the survey relative to the total decreased, confidence also decreased. The quality of
the supporting documentation also affects uncertainty. Peer-reviewed reports from the open
literature (including government and trade association survey data) were considered to be the
1-30
-------
Municipal Waste Combustion
Medical Waste Incineration
Secondary Copper Smelting
Backyard Barrel Burning
Cement Kilns Burning Haz Waste
Utility/ Industrial Coal Combustion
On-road Leaded Gas Fuel Combustion
Iron Ore Sinter Plants
On-road Diesel Fuel Combustion
Industrial Wood Burning
Residential Wood Burning
Off-road Diesel Fuel Combustion
Utility/ Industrial Residual Oil Combustion
Secondary Aluminum Smelting
Cement Kilns Not Burning Haz Waste
Utility/ Industrial Distillate Oil Combustion
Sewage Sludge Incineration
Hazardous Waste Incineration
On-road Unleaded Gas Fuel Combustion
Manufacture of EDC/VC
0.1
10
100
1000
10000
11987 D1995 D2000
Figure 1-9. Comparison of estimates of annual WHO-TEQ emissions to air
(g WHO-TEQ/yr) for reference years 1987,1995, and 2000.
most reliable. However, as with the emission factor estimates, draft reports that had undergone
more limited review were used in some cases, and in a few cases unpublished references such as
personal communication with experts were used. These are clearly noted in the text.
1.2.3.1. Rating Scheme
The confidence rating scheme shown in Table 1-16 represents the qualitative criteria used
to assign a high, medium, or low confidence rating to emission factors and activity levels for
those source categories for which emission estimates could be reliably quantified. The overall
confidence rating assigned to an emissions estimate was determined by the confidence ratings
assigned to the corresponding activity level and emission factor. If the lowest rating assigned to
either the activity level or the emission factor is "high," then the category rating assigned to the
emission estimate is high (Category A). If the lowest rating assigned to either the activity level
1-31
-------
Table 1-14. I-TEQDF emission factors used to develop national emission inventory estimates of releases to air
Source category
I-TEQDF emission factor
2000
1995
1987
Emission factor unit
WASTE INCINERATION
Municipal waste combustion
Hazardous waste incineration
Boilers/industrial furnaces
Halogen acid furnaces
Medical waste/pathological incineration
Crematoria - human
-animal
Sewage sludge incineration
Tire combustion
Pulp and paper mill sludge incinerators1"
2.82
2.12
1.21
0.803
a
0.11
6.65
0.282
38.2a
3.83
0.64
a
17,000
598
6.94
0.282
a
3.83
573 0.64
a
1 706
17,000
6.94
0.282
ng TEQ/kg waste combusted
ng TEQ/kg waste combusted
ng TEQ/kg waste combusted
ng TEQ/kg waste feed
ng TEQ/kg waste combusted
ng TEQ/body
ng TEQ/kg animal
ng TEQ/kg dry sludge combusted
ng TEQ/kg tires combusted
POWER/ENERGY GENERATION
Vehicle fuel combustion - leaded0
- unleaded
- diesel
Wood combustion - residential
- industrial
Coal combustion - utility
Oil combustion - industrial/utility
NA
1.5
172
0.5
0.56-13.2d
0.079
0.2
45
1.5
172
2
0.56-13.2d
0.079
0.2
45
1.5
172
2
0.56-13.2d
0.079
0.2
pg TEQ/km driven
pg TEQ/km driven
pg TEQ/km driven
ng TEQ/kg wood combusted
ng TEQ/kg wood combusted
ng TEQ/kg coal combusted
ng TEQ/L oil combusted
OTHER HIGH-TEMPERATURE SOURCES
Cement kilns burning hazardous waste
Lightweight aggregate kilns
Cement kilns not burning hazardous waste
Petroleum refining catalyst regeneration
Cigarette combustion
Carbon reactivation furnaces
Kraft recovery boilers
1.444
2.06
0.27
1.52
0.00043-0.0029
1.2
0.029
1.04-28.58e
0.27
1.52
0.00043-0.0029
1.2
0.029
1.04-28.58e
0.27
1.52
0.00043-0.0029
1.2
0.029
ng TEQ/kg clinker produced
ng TEQ/ kg waste feed
ng TEQ/kg clinker produced
ng TEQ/barrel reformer feed
ng TEQ/cigarette
ng TEQ/kg of reactivated carbon
ng TEQ/kg solids combusted
to
-------
Table 1-14. I-TEQDF emission factors used to develop national emission inventory estimates of releases to air
(continued)
Source category
I-TEQDF emission factor
2000
1995
1987
Emission factor unit
MINIMALLY CONTROLLED OR UNCONTROLLED COMBUSTION
Backyard barrel burningf
72.8
72.8
72.8
ng TEQ/kg waste combusted
METALLURGICAL PROCESSES
Ferrous metal smelting/refining - sintering plants
Nonferrous metal smelting/refining
- primary copper
- secondary aluminum smelting
- secondary copper smelting8
- secondary lead smelters
Drum and barrel reclamation
0.55-4.14
<0.31
4.9
0.05-8.31
16.5
0.55-4.14
<0.31
4.9
0.05-8.31
16.5
0.55-4.14
<0.31
4.9
0.05-8.31
16.5
ng TEQ/kg sinter
ng TEQ/kg copper produced
ng TEQ/kg scrap feed
ng TEQ/kg scrap consumed
ng TEQ/kg lead produced
ng TEQ/drum
CHEMICAL MANUFACTURING/PROCESSING SOURCES
Ethylene dichloride/vinyl chloride
0.95a
ng TEQ/kg EDC produced
different emission factors were derived for various subcategories within this industry; the value listed is a weighted average.
Included in total for wood combustion, industrial.
°Leaded fuel production and the manufacture of motor vehicle engines requiring leaded fuel for highway use are prohibited in the United States.
(See Chapter 4, Section 4.1, for details.)
dEmission factor of 0.56 ng I-TEQDF/kg used for nonsalt-laden wood; emission factor of 13.2 ng I-TEQDF/kg used for salt-laden wood.
eEmission factor of 1.04 ng I-TEQDF/kg used for kilns with air pollution control device (APCD) inlet temperatures less than 232°C; emission factor of 28.58 ng
I-TEQDF/kg used for kilns with APCD inlet temperatures greater than 232°C.
Includes the burning of brush and leaf residential yard waste.
gFacility-specific emission factors were used ranging from 3.6 to 16,600 ng I-TEQDF/kg scrap consumed.
-------
Table 1-15. TEQDF-WHO98 emission factors used to develop national emission inventory estimates of releases
to air
Emission source category
TEQDF-WHO98 emission factor
2000
1995
1987
Emission factor unit
WASTE INCINERATION
Municipal waste combustion
Hazardous waste incineration
Boilers/industrial furnaces
Halogen acid furnaces
Medical waste/pathological incineration
Crematoria - humanb
-animal
Sewage sludge incineration
Tire combustion
Pulp and paper mill sludge incinerators0
3.10
2.13
1.21
0.836
a
630
0.12
6.74
0.281
a
3.88
43.4 0.65
a
17,000
633
7.04
0.281
a
3.88
644 0.65
a
l,^7!'000
7.04
0.281
ng TEQ/kg waste combusted
ng TEQ/kg waste combusted
ng TEQ/kg waste combusted
ng TEQ/kg waste feed
ng TEQ/kg waste combusted
ng TEQ/body
ng TEQ/kg animal
ng TEQ/kg dry sludge combusted
ng TEQ/kg tires combusted
POWER/ENERGY GENERATION
Vehicle fuel combustion - leadedd
- unleaded
- diesel
Wood combustion - residential
- industrial6
Coal combustion - utility
Oil combustion - industrial/utility
NA
1.6
182
b
0.6-13.2
0.5 °-78
0.23
53
1.6
182
b
0.6-13.2
2 0.078
0.23
53
1.6
182
b
0.6-13.2
0.078
0.23
pg TEQ/km driven
pg TEQ/km driven
pg TEQ/km driven
ng TEQ/kg wood combusted
ng TEQ/kg wood combusted
ng TEQ/kg coal combusted
ng TEQ/L oil combusted
OTHER HIGH-TEMPERATURE SOURCES
Cement kilns burning hazardous waste
Lightweight aggregate kilns
Cement kilns not burning hazardous waste
Petroleum refining catalyst regeneration
Cigarette combustion
Carbon reactivation furnaces
Kraft recovery boilers
'1.99
0.26
1.59
0.00044-0.003
1.2b
0.028
l.ll-30.7f
0.26
1.59
0.00044-0.003
1.2b
0.028
l.ll-30.7f
0.26
1.59
0.00044-0.003
1.2b
0.028
ng TEQ/kg clinker produced
ng TEQ/ kg waste feed
ng TEQ/kg clinker produced
ng TEQ/barrel reformer feed
ng TEQ/cigarette
ng TEQ/kg of reactivated carbon
ng TEQ/kg solids combusted
-------
Table 1-15. TEQDF-WHO98 emission factors used to develop national emission inventory estimates of releases
to air (continued)
Emission source category
TEQDF-WHO98 emission factor
2000
1995
1987
Emission factor unit
MINIMALLY CONTROLLED OR UNCONTROLLED COMBUSTION
Backyard barrel burning8
76.8b
76.8b
76.8b
ng TEQ/kg waste combusted
METALLURGICAL PROCESSES
Ferrous metal smelting/refining - sintering plants
Nonferrous metal smelting/refining
- primary copperb
- secondary aluminum smelting
- secondary copper smelting11
- secondary lead smelters
Drum and barrel reclamation
0.62-4.61
0.31
5.2
0.05-8.81
17.5
0.62-4.61
0.31
5.2
0.05-8.81
17.5
0.62-4.61
0.31
5.2
0.05-8.81
17.5
ng TEQ/kg sinter
ng TEQ/kg copper produced
ng TEQ/kg scrap feed
ng TEQ/kg scrap consumed
ng TEQ/kg lead produced
ng TEQ/drum
CHEMICAL MANUFACTURING/PROCESSING SOURCES
Ethylene dichloride/vinyl chloride
0.95a'b
ng TEQ/kg EDC produced
"Different emission factors were derived for various subcategories within this industry; the value listed is a weighted average.
bCongener-specific data were not available; the TEQDF emission factor was used as a surrogate for the TEQDF-WHO98 emission factor.
Included in total for wood combustion, industrial.
dLeaded fuel production and the manufacture of motor vehicle engines requiring leaded fuel for highway use are prohibited in the United States.
(See Chapter 4, Section 4.1, for details).
eEmission factor of 0.6 ng I-TEQDF/kg used for non-salt-laden wood; emission factor of 13.2 ng I-TEQDF/kg used for salt-laden wood.
Emission factor of 1.11 ng I-TEQDF/kg used for kilns with air pollution control device (APCD) inlet temperatures less than 232°C; emission factor of 28.58 ng
I-TEQDF/kg used for kilns with APCD inlet temperatures greater than 232°C.
BThis term refers to the burning of residential waste in barrels.
facility-specific emission factors were used ranging from 3.6 to 16,600 ng TEQDF-WHO98/kg scrap consumed.
-------
Table 1-16. Confidence rating scheme for U.S. emission estimates
Confidence rating
Activity level estimate
Emission factor estimate
Categories/media for which releases can be reasonably quantified
High
Medium
Low
Derived from comprehensive
survey
Based on estimates of average plant
activity level and number of plants
or limited survey
Based on data judged possibly
nonrepre sentative
Derived from comprehensive survey
Derived from testing at a limited but
reasonable number of facilities
believed to be representative of
source category
Derived from testing at only a few,
possibly nonrepre sentative facilities
or from similar source categories
Categories/media for which releases cannot be reasonably quantified
Preliminary estimate
Not quantified
Based on extremely limited data,
judged to be clearly
nonrepre sentative
No data available
Based on extremely limited data,
judged to be clearly
nonrepresentative
( 1 ) Argument based on theory but
no data, or
(2) Data available indicating
formation but not in a form that
allows developing an emission
factor
or the emission factor term is "medium," then the category rating assigned to the emission
estimate is medium (Category B). If the lowest rating assigned to either the activity level or the
emission factor is "low," then the category rating assigned to the emission estimate is low
(Category C). It is emphasized that this confidence rating scheme should not be interpreted as a
statistical measure, but rather as subjective judgment of the relative uncertainty among sources.
For many source categories, either emission factor information or activity level
information was inadequate to support development of reliable quantitative release estimates for
one or more media. For some of these source categories, sufficient information was available to
make preliminary estimates of emissions of CDDs/CDFs or dioxin-like PCBs; however, the
confidence in the activity level estimates or emission factor estimates was so low that they could
not be included in the sum of quantified emissions from sources with confidence ratings of A, B,
and C. These preliminary estimates were given an overall confidence rating of D. The
preliminary release estimates for sources with a confidence rating of D are given in the right-
most column of Table 1-12. Because these are order-of-magnitude estimates, they are made for
2000 only. As preliminary estimates of source magnitude, they can be used to help prioritize
1-36
-------
future research and data collection. The actual magnitude of emissions from these sources could
be significantly lower or higher than these preliminary estimates. Although EPA has chosen not
to include them in the more thoroughly characterized emissions of the national inventory, some
of these poorly characterized sources have the potential of being major contributors of releases to
the environment. It is important to present these estimates because they may help determine
priorities for future data collection efforts. As the uncertainty around these sources is reduced,
they will be included in future inventory calculations.
For other sources, some information exists that suggests that they may release dioxin-like
compounds; however, the available data were judged to be insufficient for developing any
quantitative emissions estimate. These source categories were assigned a confidence rating of E
and also were not included in the national inventory (see the "Not quantifiable" column in Table
1-11).
1.3. CONCLUSIONS
1.3.1. Total Environmental Releases
Nationwide emission estimates of grams I-TEQDF and TEQDF-WHO98 released to the open
and circulating environment of the United States are presented in Table 1-12. For the year 2000,
EPA draws the following conclusions:
• The total releases in the inventory (Categories A, B, and C) were 1,422 g TEQDF-
WHO98/yr. These were dominated by releases to the air (92%). Most of the air
releases were from combustion sources. Table 1-17 presents a ranking of sources for
2000, 1995, and 1987 based on the magnitude of environmental release. The top
three sources were backyard barrel burning of refuse (498.5 g, 32% of total), MWIs
(378 g, 27%), and the incineration and land application of municipal wastewater
treatment sludge (89.7 g, 5%).
• There is a significant potential for release of dioxin-like compounds from Category D
sources. However, these sources a currently poorly characterized. The most important
Category D sources are forest fires and accidental fires at MSW landfills. Research is
recommended to confirm emissions from these sources and to provide a more
accurate assessment of releases.
• A total of 18 contemporary formation sources were classified as Category E.
Information suggests these may be sources of dioxin-like compounds, but it is
insufficient to make a national estimate of releases. Additional research on these
sources is recommended in order to adequately identify them as actual sources and to
provide data for estimating releases.
1-37
-------
Table 1-17. Ranking of sources of dioxin-like compounds based on environmental releases (from high to low)
for reference years 2000,1995, and 1987
2000
Source (released to)
Backyard barrel
burning of refuse (air)
Medical
waste/pathological
incineration (air)
Municipal wastewater
treatment sludge, land
application and
incineration (land, air)
Municipal waste
combustion (air)
Coal fired-utility
boilers (air)
Diesel heavy-duty
trucks (air)
Industrial wood
combustion (air)
Diesel off-road
equipment, ships,
trains, tractors (air)
Releases
(grams)
498.5
378.0
89.7
83.8
69.5
65.4
41.5
33.1
Percent
of total
35.1
26.6
6.3
5.9
4.9
4.6
2.9
2.3
1995
Source (released to)
Municipal waste
combustion (air)
Backyard barrel
burning of refuse (air)
Medical
waste/pathological
incineration (air)
Secondary copper
smelters (air)
Cement kilns burning
hazardous waste (air)
Municipal wastewater
treatment sludge, land
application and
incineration (land and
air)
Coal fired-utility
boilers (air)
Ethylene
dichloride/vinyl
chloride production
(land, air, water)
Releases
(grams)
1,393.5
628.0
487.0
271.0
156.1
133.3
60.1
35.7
Percent
of total
40.5
18.2
14.1
7.9
4.5
3.9
1.7
1.0
1987
Source (released to)
Municipal waste
combustion (air)
Medical waste/
pathological incineration
(air)
Secondary copper
smelters (air)
Backyard barrel burning
of refuse (air)
Bleached chemical
wood pulp and paper
mills (land, water)
Cement kilns burning
hazardous waste (air)
Municipal wastewater
treatment sludge, land
application and
incineration (air, land)
Coal fired-utility boilers
(air)
Releases
(grams)
8,905.1
2,570.0
983.0
604.0
370.1
117.8
85.0
50.9
Percent
of total
63.8
18.4
7.0
4.3
2.7
0.8
0.6
0.4
oo
-------
Table 1-17. Ranking of sources of dioxin-like compounds based on environmental releases (from high to low)
for reference years 2000,1995, and 1987 (continued)
2000
Source (released to)
Ethylene dichloride/
vinyl chloride
production (water,
land, air)
Sintering plants (air)
Cement kilns burning
hazardous waste (air)
Cement kilns burning
nonhazardous waste
(air)
Residential wood
combustion (air)
Secondary aluminum
smelting (air)
Industrial/utility oil
combustion, distillate
oil (air)
Automobiles using
unleaded gasoline
(air)
Residential heating,
distillate oil (air)
Primary magnesium
production (air)
Releases
(grams)
30.0
27.6
18.8
17.2
11.3
8.3
7.3
7.0
4.5
4.3
Percent
of total
2.1
1.9
1.3
1.2
0.8
0.6
0.5
0.5
0.3
0.3
1995
Source (released to)
Diesel heavy-duty
trucks (air)
Bleached chemical
wood pulp and paper
mills (land, water)
2,4-Dichlorophenoxy
acetic acid (land)
Sintering plants (air)
Industrial wood
combustion (air)
Diesel off-road
equipment: ships,
trains, tractors (air)
Secondary aluminum
smelters (air)
Cement kilns burning
nonhazardous waste
(air)
Residential wood
combustion (air)
Industrial/utility oil
combustion, residual
oil (air)
Releases
(grams)
33.3
30.0
28.9
28.0
26.2
23.8
19.5
16.6
15.7
10.7
Percent
of total
1.0
0.9
0.8
0.8
0.8
0.7
0.6
0.5
0.5
0.3
1987
Source (released to)
Automobiles using
leaded gasoline (air)
2,4-Dichlorophenoxy
acetic acid (land)
Sintering plants (air)
Diesel heavy-duty trucks
(air)
Industrial wood
combustion (air)
Residential wood
combustion (air)
Diesel off-road
equipment: ships, trains,
tractors (air)
Industrial/utility oil
combustion, residual oil
(air)
Cement kilns burning
nonhazardous waste
(air)
Secondary aluminum
smelting (air)
Releases
(grams)
37.5
33.4
32.7
27.8
26.5
22.0
19.0
17.8
12.7
10.9
Percent
of total
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.1
VO
-------
Table 1-17. Ranking of sources of dioxin-like compounds based on environmental releases (from high to low)
for reference years 2000,1995, and 1987 (continued)
2000
Source (released to)
Chlor alkali facilities
(air, water)
Hazardous waste
incineration (air)
Institutional/cornmerci
al heating, distillate
oil (air)
Secondary lead
smelting (air)
Petroleum refining
catalyst regeneration
(air)
Lightweight aggregate
kilns burning
hazardous waste (air)
Boilers/industrial
furnaces (air)
Industrial/utility oil
combustion, residual
oil (air)
Bleached chemical
wood pulp and paper
mills (land, water)
Secondary copper
smelting (air)
Releases
(grams)
3.6
3.2
2.9
2.5
2.2
1.9
1.8
1.7
1.1
0.9
Percent
of total
0.3
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
1995
Source (released to)
Industrial/utility oil
combustion, distillate
oil (air)
Hazardous waste
incineration (air)
Residential heating,
distillate oil (air)
Automobiles using
unleaded gasoline (air)
Primary magnesium
production (air)
Chlor-alkali facilities
(air, water)
Institutional/commercia
1 heating, distillate oil
(air)
Lightweight aggregate
kilns burning
hazardous waste (air)
Kraft recovery boilers
(air)
Petroleum refining
catalyst regeneration
(air)
Releases
(grams)
7.3
5.8
5.0
4.7
4.1
3.6
3.1
2.4
2.3
2.2
Percent
of total
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
1987
Source (released to)
Industrial/utility oil
combustion, distillate oil
(air)
Residential heating,
distillate oil (air)
Hazardous waste
incineration (air)
Institutional/commercial
heating, distillate oil
(air)
Automobiles using
unleaded gasoline (air)
Lightweight aggregate
kilns burning hazardous
waste (air)
Petroleum refining
catalyst regeneration
(air)
Kraft recovery boilers
(air)
Industrial/utility oil
combustion, residual oil
(air)
Secondary lead smelting
(air)
Releases
(grams)
8.3
5.4
5.0
3.7
3.6
3.3
2.2
2.0
1.5
1.3
Percent
of total
0.1
0.1
0.04
0.03
0.03
0.02
0.02
0.01
0.01
0.01
-------
Table 1-17. Ranking of sources of dioxin-like compounds based on environmental releases (from high to low)
for reference years 2000,1995, and 1987 (continued)
2000
Source (released to)
Kraft recovery boilers
(air)
Institutional/commerci
al heating, residual oil
(air)
Drum and barrel
reclamation (air)
Tire incineration (air)
Cigarette combustion
(air)
Unleaded gasoline
off-road equipment
(air)
Halogen acid furnaces
(air)
Primary copper
smelting (air)
Crematoria, human
(air)
Carbon reactivation
furnaces (air)
Automobiles using
leaded gasoline (air)
TOTAL
Releases
(grams)
0.8
0.7
0.6
0.5
0.4
0.4
0.3
0.3
0.3
0.1
0
1,422.0
Percent
of total
0.1
0.05
0.04
0.04
0.03
0.03
0.02
0.02
0.02
0.01
0.00
100
1995
Source (released to)
Secondary lead
smelting (air)
Automobiles using
leaded gasoline (air)
Industrial/utility oil
combustion, residual
oil (air)
Cigarette combustion
(air)
Primary copper
smelting (air)
Boilers/industrial
furnaces (air)
Crematoria, human
(air)
Tire incineration (air)
Carbon reactivation
furnaces (air)
Drum and barrel
reclamation (air)
Releases
(grams)
1.7
1.6
0.8
0.8
0.5
0.4
0.2
0.1
0.1
0.1
3,444.2
Percent
of total
0.05
0.05
0.02
0.02
0.01
0.01
0.01
O.01
O.01
O.01
100
1987
Source (released to)
Cigarette combustion
(air)
Boilers/industrial
furnaces (air)
Primary copper smelting
(air)
Crematoria, human (air)
Carbon reactivation
furnaces (air)
Tire incineration (air)
Drum and barrel
reclamation (air)
Releases
(grams)
1.0
0.8
0.5
0.2
0.1
0.1
0.1
13,965.3
Percent
of total
0.01
0.01
<0.01
O.01
O.01
O.01
<0.01
100
-------
Releases from reservoir sources could significantly add to the inventory. However,
environmental reservoirs as sources of the redistribution of previously formed dioxins
into the open and circulating environment are currently poorly understood and poorly
characterized. This report suggests that urban runoff to surface water and rural soil
erosion to surface water can be significant reservoir sources. Releases from reservoirs
(air, sediment, water, and biota) could not be reliably quantified, given the lack of
information in this area.
The amount of dioxin-like PCBs released from man-made sources remains poorly
characterized. Only a total of 19.5, 78.5, and 51.5 g of PCB TEQ could be quantified
for the years 2000, 1995, and 1987, respectively. To date, only sewage sludge has
been adequately characterized in terms of the amount of dioxin-like PCBs that may be
released from a source.
1.3.2. Time Trends
A significant reduction in total CDD/CDF environmental releases has occurred since
1987. EPA's best estimates of releases of CDDs/CDFs to air, water, and land from reasonably
quantifiable sources (Categories A, B, and C) are approximately 1,422 g TEQDF-WHO98 in
reference year 2000; 3,444.2 g in reference year 1995; and 13,965.3 g in reference year 1987.
From 1987 to 2000 there was an approximately 90% reduction in releases to all media. Most of
the reduction in dioxin releases (75%) occurred between 1987 and 1995.
In 1987 and 1995, municipal waste combustion was the leading source of dioxin
emissions to the U.S. environment; however, because of reductions in dioxin emissions from
MWCs, it dropped to the fourth ranked source in 2000. Burning of domestic refuse in backyard
burn barrels remained fairly constant over the years, but in 2000 it emerged as the largest source
of dioxin emissions.
Reductions in environmental releases of dioxin-like compounds are attributed primarily
to reductions in air emissions from MWCs, MWIs, and cement kilns burning hazardous waste,
and in wastewater discharged into surface waters from pulp and paper mills using chlorine.
These reductions have occurred from a combination of regulatory activities (see Section 1.1.2),
improved emission controls, improved industrial technologies, voluntary actions on the part of
industry, and the closing of a number of antiquated facilities.
1.3.3. Sources Not Included in the Inventory
Significant amounts of the dioxin-like compounds produced annually in the United States
are not considered releases to the open and circulating environment and are not included in the
national inventory. Examples include dioxin-like compounds generated internal to a process but
destroyed before release and waste streams that are disposed of in approved landfills.
1-42
-------
The only product judged to have the potential for environmental release—and therefore
considered for the inventory—was the herbicide 2,4-D. Release estimates are provided for 1987
and 1995. Since 1995, the chemical manufacturers of 2,4-D have been undertaking voluntary
actions to significantly reduce the dioxin content of the product. No information is available on
the extent of these reductions and, therefore, no release estimate could be made for 2000.
Regarding other products, data are presented on the amounts of CDDs/CDFs contained in
bleached pulp, ethylene dichloride/vinyl chloride, PCP-treated wood, and dioxazine dyes and
pigments. None of these products, however, was considered to have release potential; they were
not included in the inventory.
A number of contemporary formation sources were classified as D or E and therefore
were not included in the inventory. The largest contemporary formation Category D sources are
forest fires and accidental fires at MSW landfills. Taken together, these sources have the
potential to significantly increase the present inventory if preliminary release estimates are
confirmed.
The possibility remains that truly undiscovered sources exist. Many of the sources that
are well-accepted today were discovered only in the past 20 years. For example, CDDs/CDFs in
stack emissions from MWCs were not detected until the late 1970s; CDDs/CDFs in the
wastewater effluent from bleached pulp and paper mills were found unexpectedly in the mid-
1980s; iron ore was not recognized as a source until the early 1990s.
1.3.4. Formation Theory
Current theory proposes that CDDs/CDFs are formed within the cool-down region of
combustion processes, either de novo or from dioxin precursors. De novo synthesis involves
solid-phase reactions with carbon, chlorine, and oxygen on combustion-generated particles
promoted by copper chloride as a catalyst. A less efficient but plausible formation process is the
gas-phase formation from precursors catalyzed by the presence of a transition metal such as
copper chloride. The ideal temperatures for de novo dioxin formation are between 200 and
400°C. Reducing temperatures to below 200°C, especially at the air pollution control device,
will minimize dioxin formation and releases from combustion sources. Chlorine sources present
in feeds are necessary for dioxin formation. Experiments suggest that a chlorine content of 1% in
the feed/fuel is the threshold for a direct relationship to dioxin formation from combustion
sources, i.e., a chlorine content > 1% is strongly correlated to the amount of dioxin formed, but a
chlorine content <1% is not. However, in well-designed, well-controlled, and well-operated full-
scale combustion systems there does not appear to be a direct relationship between the amount of
chlorine present in the waste and the amount of dioxin emissions from the stack.
1-43
-------
Controversy exists regarding the role of PVC in the formation of CDDs/CDFs during
municipal waste combustion. Experimental evidence suggests that PVC combustion generates
hydrogen chloride gas (HC1) and dioxin precursors such as chlorobenzenes and chlorophenols,
both of which may contribute to dioxin formation. HC1 is a progenitor of chlorine radicals that
then participate in the dioxin formation chemistry. Precursors are foundation molecules to dioxin
formation. If PVC is the only source of chlorine and dioxin precursors during the combustion of
MSW, then the removal of PVC may reduce the amount of dioxin formed and emitted.
However, the complex mixture of materials in MSW provides sufficient chlorine for de novo
synthesis, and dioxin precursors are formed as products of the incomplete combustion of the
waste constituents. Therefore, the elimination of PVC from the waste prior to combustion would
not necessarily eliminate the formation and emissions of CDDs/CDFs from municipal waste
combustion.
Current information strongly suggests that releases of CDDs/CDFs to the U.S.
environment occur principally from anthropogenic activities. However, scientific studies have
identified the possibility of natural formation of some CDDs/CDFs (e.g., in ball clay).
1.3.5. Congener Profiles of CDD/CDF Sources
This document presents congener profiles for a number of sources, as shown in Figure 1-
10. These profiles show the relative amounts of CDD/CDF congeners in environmental releases.
These profiles can be useful for (1) identifying source contributions to near-field air
measurements of CDDs/CDFs, (2) comparing sources, and (3) providing insights into the
formation of CDDs/CDFs in the releases. There are numerous procedures for deriving a
congener profile, and there is no single agreed-upon convention (Cleverly et al., 1997; Lorber et
al., 1996; Hagenmaier et al., 1994).
For this report, congener profiles were developed primarily by calculating the ratio of
specific 2,3,7,8-substituted CDDs/CDFs in the emissions or product to the total (C14 - C18)
CDDs/CDFs. With respect to combustion sources, the profiles were derived by dividing the
congener-specific emission factors by the total (C14 - C18) CDD/CDF emission factor for each
tested facility and then averaging the congener profiles developed for all tested facilities within
the combustor type. For chemical processes and commercial chemicals, CDD/CDF profiles were
typically generated by dividing average congener concentrations (ppt) in the chemical by the total
CDDs/CDFs present. Profiles for select source categories are presented in Figure 1-10.
1-44
-------
Municipal Waste Combustor with Hot ESP
Municipal Waste Combustor with DS/FF
_
.
1.
I.I
12%
10%
8%
JZ
' *N ^
4 4 ,<(*
•6^ ^>'
Medical/hospital Waste Incinerators
Hazardous Waste Incinerators
Figure 1-10. Congener profiles (as percent distributions to the sum of
CDDs and CDFs) of anthropogenic sources of chlorinated dibenzo-/?-
dioxins and chlorinated dibenzofurans in the United States.
On the basis of inspection and comparisons of the average CDD/CDF congener profiles
across combustion and noncombustion sources, the following observations were made (Cleverly
et al., 1997) (these generalizations are derived from this data set, and their application beyond
these data is uncertain):
• It appears that combustion sources emit all 2,3,7,8-substituted CDDs/CDFs, although
in varying percentages of total CDDs/CDFs.
• In combustion source emissions, 2,3,7,8-TCDD is usually 0.1 to 1% of total
CDDs/CDFs. The exception is stack emissions from industrial oil-fired boilers,
where the available but limited data indicate that 2,3,7,8-TCDD constitutes an
average of 7% of total CDD/CDF emissions.
1-45
-------
Backyard Refuse Barrel Burning
Industrial Wood Waste Combustion
y//////
Portland Cement Kilns
1
1
1 1 - •
*/f /////// f //////
Portland Cement Kilns Burning Hazardous Waste
.!• . .
III. ll_l
Figure 1-10. Congener profiles (as percent distributions to the sum of CDDs and
CDFs) of anthropogenic sources of chlorinated dibenzo-p-dioxins and chlorinated
dibenzofurans in the United States (continued).
• It cannot be concluded that OCDD is the dominant congener for all combustion-
generated emissions of CDDs/CDFs. OCDD dominates total emissions from mass-
burn MWCs that have DSs and FFs for dioxin control, industrial oil-fired boilers,
industrial wood-fired boilers, unleaded gasoline combustion, diesel fuel combustion
in trucks, and sewage sludge incinerators. The dominant congeners for other
combustion sources are 1,2,3,4,6,7,8-HpCDF in emissions from mass-burn MWCs
equipped with hot-sided electrostatic precipitators (ESPs), hazardous waste
incineration, and secondary aluminum smelters and 2,4-D salts and esters; OCDF in
emissions from medical waste incineration and industrial/utility coal-fired boilers;
2,3,4,7,8-PeCDF in cement kilns burning hazardous waste; and 2,3,7,8-TCDF in
cement kilns not burning hazardous waste.
1-46
-------
Coal-fired Electrical Generating Facilities
Oil-fired Electrical Generating Facilities
.l
1.1
i
Petroleum Refineries
Bleached Pulp and Paper Mills (effluent)
35%
30%
25%
20%
15%
10%
1
1-
• I
ill
Figure 1-10. Congener profiles (as percent distributions to the sum of CDDs and
CDFs) of anthropogenic sources of chlorinated dibenzo-p-dioxins and chlorinated
dibenzofurans in the United States (continued).
Evidence for a shift in the congener patterns potentially caused by the application of
different air pollution control systems within a combustion source type can be seen in
the case of mass-burn MWCs. For mass-burn MWCs equipped with hot-sided ESPs,
the most prevalent CDD/CDF congeners are 1,2,3,4,6,7,8-HxCDF; OCDD;
l,2,3,4,6,7,8-HpCDD/l,2,3,4,7,8-HxCDF;2,3,4,6,7,8-HxCDF/OCDF; 1,2,3,6,7,8-
HxCDF. The most prevalent congeners emitted from MWCs equipped with DS/FF
are OCDD; 1,2,3,4,6,7,8-HpCDD; 1,2,3,4,6,7,8-HpCDF; OCDF; and 2,3,7,8-
TCDF/l,2,3,4,7,8-HxCDD;2,3,4,6,7,8-HxCDF.
There is evidence of marked differences in the distribution of CDD/CDF congeners
between cement kilns that burn hazardous waste and those that do not. When not
burning hazardous waste as supplemental fuel, the dominant congeners appear to be
2,3,7,8-TCDF; OCDD; 1,2,3,4,6,7,8-HpCDD, and OCDF. When burning hazardous
waste, the dominant congeners are 2,3,7,8-PeCDF; 2,3,7,8-TCDF; 1,2,3,4,7,8-
1-47
-------
Secondary Aluminum Smelters
Secondary Lead Smelters
I777TT
i
nt
I
Secondary Copper Smelters
v>yx//
• I.
////
A
Iron Foundries
0.07%
0.06%
0.05%
0.04%
0.03%
0.02%
0.01%
I
H
Figure 1-10. Congener profiles (as percent distributions to the sum of CDDs and
CDFs) of anthropogenic sources of chlorinated dibenzo-p-dioxins and chlorinated
dibenzofurans in the United States (continued).
xCDF; and 1,2,3,4,6,7,8-HpCDD. When burning hazardous waste, OCDD and
OCDF are minor constituents of stack emissions.
The congener profile of 2,4-D salts and esters seems to mimic a combustion source
profile in the number of congeners represented and in the minimal amount of 2,3,7,£
TCDD relative to all 2,3,7,8-substituted congeners. A major difference is the
prevalence of 1,2,3,7,8-PeCDD in 2,4-D (14%), which is not seen in any other
combustion or noncombustion source presented here.
1-48
-------
There are similarities in the congener profiles of PCP, diesel truck emissions,
unleaded gasoline vehicle emissions, and emissions from industrial wood combustors.
In these sources, OCDD dominates total emissions, but the relative ratio of
1,2,3,4,6,7,8-HpCDD to OCDD is also quite similar.
The congener profiles for diesel truck exhaust and those for air measurements from a
tunnel study of diesel traffic are quite similar.
Automobiles Burning Leaded Gasoline
Diesel Truck Exhaust
t
45%
40%
35%
30%
25%
20%
15%
10%
i
Automobiles Burning UnLeaded Gasoline
1
id
O^ -^ ,/" J\T
-------
Forest Fires
20%
10%
i
Figure 1-10. Congener profiles (as percent distributions to the sum of CDDs
and CDFs) of anthropogenic sources of chlorinated dibenzo-p-dioxins and
chlorinated dibenzofurans in the United States (continued).
1-50
-------
2. MECHANISMS OF FORMATION OF DIOXIN-LIKE COMPOUNDS
DURING COMBUSTION OF ORGANIC MATERIALS
More than a decade of combustion research has contributed to a general understanding of
the central molecular mechanisms that form CDDs/CDFs emitted from combustion sources.
Current understanding of the conditions necessary to form CDDs/CDFs were derived primarily
from studies of full-scale municipal waste combustors (MWCs), augmented with observations
involving the experimental combustion of synthetic fuels and feeds in the laboratory. However,
the formation mechanisms elucidated by these studies are generally relevant to most combustion
systems in which organic material is burned with chlorine.
Intensive studies have examined MWCs from the perspective of identifying the specific
formation mechanism(s) that occurs within the system. This knowledge may lead to methods
that prevent the formation of CDDs/CDFs and their release into the environment. Although
much has been learned from such studies, a method that completely prevents CDDs/CDFs from
forming during the combustion of certain organic materials in the presence of a source of
chlorine and oxygen is still unknown. The wide variability of organic materials incinerated and
thermally processed by a wide range of combustion technologies that have varying temperatures,
residence times, and oxygen requirements adds to this complex problem. However, central
chemical events involved in the formation of CDDs/CDFs can be identified by evaluating
emission test results from MWCs in combination with results from laboratory experiments.
CDD/CDF emissions from combustion sources can potentially be explained by three
principal mechanisms that should not be regarded as being mutually exclusive. In the first
mechanism (referred to as "pass through"), CDDs/CDFs are present as contaminants in the
combusted organic material; they pass through the furnace and are emitted unaltered. This
mechanism is discussed in Section 2.1. In the second mechanism (referred to as "precursor"),
CDDs/CDFs ultimately form from the thermal breakdown and molecular rearrangement of
precursor ring compounds, which are defined as chlorinated aromatic hydrocarbons that have a
structural resemblance to the CDD/CDF molecules. Ringed precursors that emanate from the
combustion zone are a result of the incomplete oxidation of the constituents of the feed (i.e.,
products of incomplete combustion). The precursor mechanism is discussed in Section 2.2. The
third mechanism (referred to as "de novo synthesis") is similar to the precursor mechanism and is
described in Section 2.3. De novo synthesis describes a pathway of CDD/CDF formation from
heterogeneous reactions on fly ash (particulate matter[PM]) involving carbon, oxygen, hydrogen,
chlorine, and a transition metal catalyst. With these reactions, intermediate compounds that have
an aromatic ring structure are formed.
2-1
-------
Studies in this area suggest that aliphatic compounds, which arise as products of
incomplete combustion, may play a critical role in initially forming simple ring molecules, which
later evolve into complex aromatic precursors. CDDs/CDFs are then formed from the
intermediate compounds. In both the second and the third mechanism, formation occurs outside
the furnace, in the so-called post-combustion zone. Particulate-bound carbon is suggested as the
primary reagent in the de novo synthesis pathway.
Section 2.4 presents an overview of studies that have investigated the role that chlorine
plays in forming CDDs/CDFs. Although chlorine is an essential component for the formation of
CDDs/CDFs in combustion systems, the empirical evidence indicates that for commercial-scale
incinerators, chlorine levels in feed are not the dominant controlling factor for rates of CDD/CDF
stack emissions. There are complexities related to the combustion process itself, and some types
of air pollution control equipment tend to mask any direct association. Therefore, the chlorine
content of fuel and feeds to a combustion source is not a good indicator of levels of CDDs/CDFs
emitted from the stack of that source.
Section 2.6 discusses the generation and formation of coplanar polychlorinated biphenyls
(PCBs). The presence of coplanar PCBs in stack emissions from combustors is an area in need
of further research. Evidence to date suggests that PCB emissions are mostly attributable to PCB
contamination in waste feeds and that emissions are related to the first mechanism described
above. However, newly published research has also indicated that it is possible that PCBs form
in much the same way as described in the second and third mechanisms identified in the
formation of CDDs/CDFs within the post-combustion zone.
Section 2.7 provides a closing summary of the three principal formation mechanisms and
the role of chlorine. From the discussions in this chapter, it should be evident that no clear
distinction exists between the precursor and the de novo synthesis mechanisms of CDD/CDF
formation. Both formation pathways depend on the evolution of precursors within combustion
gases, the interaction of reactive fly ashes, a generally oxidative environment, the presence of a
transition metal catalyst, the presence of gaseous chlorine, and a favorable range of temperatures.
The temperature of the combustion gases (i.e., flue gases) is perhaps the single most important
factor in forming dioxin-like compounds. Temperatures between 200 and 450°C are most
conducive to the formation of CDDs/CDFs, with maximum formation occurring at around
350°C. If the temperature falls outside this range, the amount of CDDs/CDFs formed is
minimized.
2-2
-------
2.1. MECHANISM 1 (PASS THROUGH): CDD/CDF CONTAMINATION IN FUEL AS
A SOURCE OF COMBUSTION STACK EMISSIONS
The first mechanism involved in stack emissions of CDDs/CDFs is the incomplete
destruction of CDD/CDF contaminants present in the fuel or feeds delivered to the combustion
chamber. Not all of these molecules are destroyed by the combustion system, thus allowing
trace amounts to be emitted from the stack. Most work in this area has involved the study of
incineration of municipal solid waste (MSW), where CDDs/CDFs were analytically measured in
the raw refuse fed into the incinerator. CDDs/CDFs are ubiquitous in the environment (air,
water, and soil) and in foods and paper; therefore, they clearly are present in municipal waste
(Tosine et al., 1983; Ozvacic, 1985; Clement et al., 1988; Federal Register, 1991a; Abad et al.,
2002).
Abad et al. (2002) provided contemporary measurements of CDDs/CDFs in raw MSW.
Twenty-two samples were collected and analyzed for CDDs/CDFs over a 1-year period, from
September 1998 through September 1999. The congeners that dominated the total mass of
CDDs/CDFs were OCDD and 1,2,3,4,7,8,9-HpCDD. Figure 2-1 displays the mean CDD and
CDF congener distribution from this study. Abad et al. found that the I-TEQ concentration in the
MSW was highly variable and ranged from 1.55 to 45.16 ng I-TEQ/kg MSW.
350
300
>- 250
0)
Q.
(/)
E
5
O)
o
o
200
150
100
50
Q Q Q Q
O O O O
X X X X
I I I I
Q. Q. LL
I I Q
O
O
oo oo
oo
OJ OJ OJ
oo
O
oo o>
r«." oo"
to
OJ OJ
Congener
Figure 2-1. Typical mean distribution of CDD and CDF congeners in
contemporary municipal solid waste.
Source: Adapted from Abad et al. (2002).
2-3
-------
A number of studies have provided evidence that most of the CDDs/CDFs present in
MSW are destroyed during combustion (Abad et al., 2002; Clement et al., 1988; Commoner et
al., 1984, 1985, 1987; Hay et al., 1986; Environment Canada, 1985). These studies involved a
mass balance of the input versus output of CDDs/CDFs at two operational MWCs. The mass of
CDDs/CDFs outside the incinerator furnace was found to be much greater than the mass of
CDDs/CDFs in the raw MSW fed into the incinerator, and the profiles of the distributions of
CDD/CDF congeners were strikingly different. Primarily, the more highly chlorinated congeners
were detected as contaminants in the waste, whereas the total array of tetra- through octa-
CDDs/CDFs could be detected in the stack gases. Moreover, the ratio of the total CDD
concentration to the total CDF concentration in the MSW was greater than 1, whereas in typical
incinerator stack emissions this ratio is less than 1 (meaning more dibenzofurans than dioxins are
emitted). From such evidence it can be concluded that CDDs/CDFs are being synthesized after
the contaminated feed has been combusted (Abad et al., 2002). It is also expected that the
conditions of thermal stress imposed by high temperatures reached in typical combustion would
destroy and reduce the CDDs/CDFs present as contaminants in the waste feed to levels that are
0.0001 to 10% of the initial concentration, depending on the performance of the combustion
source and the level of combustion efficiency. Stehl et al. (1973) demonstrated that the moderate
temperature of 800°C enhances the decomposition of CDDs at a rate of about 99.95%, but lower
temperatures result in a higher survival rate.
Theoretical modeling has shown that unimolecular destruction of CDDs/CDFs at 99.99%
can occur at the following temperatures and retention times within the combustion zone: 977°C
with a retention time of 1 sec, 1,000°C at a retention time of 0.5 sec, 1,227°C at a retention time
of 4 msec, and 1,727°C at a retention time of 5 |isec (Schaub and Tsang, 1983). Thus,
CDDs/CDFs would have to be in concentrations of parts per million in the feed in the combustor
to be found in the parts-per-billion or parts-per-trillion level in the stack gas emissions (Shaub
and Tsang, 1983). However, it cannot be ruled out that CDDs/CDFs in the waste or fuel may
contribute (up to some percentage) to the overall concentration leaving the stack. The only other
possible explanation for CDD/CDF emissions from high-temperature combustion of organic
material is formation outside and downstream of the furnace.
The above studies point to formation mechanisms other than simple pass through of
noncombusted feed contamination. These formation mechanisms are discussed and reviewed in
the following sections.
2.2. MECHANISM 2 (PRECURSOR): FORMATION OF CDDs/CDFs FROM
PRECURSOR COMPOUNDS
The second mechanism involves the formation of CDDs/CDFs from aromatic precursor
compounds in the presence of a chlorine donor. This mechanism has been elucidated by
2-4
-------
laboratory experiments involving the combustion of known precursors in quartz ampules under
starved-air conditions and in experiments that investigated the role of combustion fly ash in
promoting the formation of CDDs/CDFs from precursor compounds. The general reaction in this
formation pathway is an interaction between an aromatic precursor compound and chlorine
promoted by a transition metal catalyst on a reactive fly ash surface (Stanmore, 2004; Dickson
and Karasek, 1987; Liberti and Brocco, 1982). Examples of well-studied precursor compounds
include chlorobenzenes, chlorophenols, phenol, and benzene (Esposito et al., 1980). Gaseous
hydrogen chloride (HC1), free chlorine (C12), and chlorine radicals (C1-) are the chlorinating
agents within the combustion gases. CDD/CDF formation results from heterogeneous gas-phase
reactions involving chlorinated precursor compounds and a source of chlorine. Chlorophenol
and chlorobenzene compounds have been measured in flue gases from MWCs (Dickson and
Karasek, 1987).
Precursors are carried from the furnace to the flue duct as products of incomplete
combustion. These compounds can adsorb on the surface of combustion fly ash or entrain in the
gas phase within the flue gases. Thus, there are two formation pathways from precursor
compounds: heterogeneous solid-phase reactions and homogeneous gas-phase reactions. In the
post-combustion region outside the furnace, heterogeneous reactions on the surface of reactive
fly ash can ensue to form CDDs/CDFs from the precursor compounds. This occurs at the cool-
down temperatures of 200 to 400°C. The heterogeneous gas-phase reactions occur from the
breakdown and molecular rearrangement of precursor compounds followed by condensation and
chlorination at the higher temperatures of 500 to 800°C. Both reaction pathways are catalyzed by
copper chloride (CuCl2) or another transition metal.
Laboratory experiments involving the controlled combustion of precursor compounds
have caused the breakdown of the precursor reagent and the subsequent appearance of
CDDs/CDFs as products of the reaction. For example, Jansson et al. (1977) produced CDDs
through the pyrolysis of wood chips treated with tri-, tetra-, and pentachlorophenol (PCP) in a
bench-scale furnace operated at 500 to 600°C. Stehl and Lamparski (1977) combusted grass and
paper treated with the herbicide 2,4,5-trichlorophenoxyacetic acid in a bench-scale furnace at 600
to 800°C and generated ppmv levels of TCDD. Ahling and Lindskog (1982) reported CDD
formation during the combustion of tri- and tetrachlorophenol formulations at temperatures of
500 to 600°C. Decreases in oxygen during combustion generally increased the CDD yield.
Ahling and Lindskog (1982) noted that adding copper salts to the tetrachlorophenol
formulation significantly enhanced the yield of CDDs. This may have been an early indication
of copper's role in catalyzing the condensation of chlorophenol to dioxin. Combustion of PCP
resulted in low yields of CDDs. However, when PCP was burned with an insufficient supply of
oxygen in the presence of copper, the investigators noted the formation of tetra- through
2-5
-------
octachlorinated congeners. Buser (1979) generated CDDs/CDFs on the order of 0.001 to 0.08%
(by weight) by heating tri-, tetra-, and pentachlorobenzenes at 620°C in quartz ampules in the
presence of oxygen. It was noted that chlorophenols formed as combustion by-products; Buser
speculated that these chlorophenols were acting as reaction intermediates in the formation of
CDDs/CDFs.
The second condition postulated to regulate the synthesis of CDDs/CDFs from the
aromatic precursor compound is the adsorption and interaction with the reactive surface of
combustion-generated fly ash (PM) entrained in the combustion plasma and the presence of a
transition metal catalyst (Stanmore, 2004; Dickson et al., 1992; Bruce et al., 1991; Cleverly et al.,
1991; Gullet et al., 1990a; Commoner et al., 1987; Dickson and Karasek, 1987; Vogg et al.,
1987). These are heterogeneous solid-phase reactions that occur at temperatures below 450°C.
The molecular precursor leaves the gas phase and condenses onto the fly ash particle. This
condition, which places greater emphasis on heterogeneous surface reactions and less emphasis
on homogeneous gas-phase reactions, was first postulated by Shaub and Tsang (1983) using
thermal-kinetic models based on the temperature of the heat of formation, adsorption, and
desorption. Shaub and Tsang modeled CDD production from chlorophenols and concluded that
solid-phase formation of CDDs/CDFs was of greater importance than gas-phase formation within
an incineration system.
The temperature of the combustion gases is a critical factor in the formation of
CDDs/CDFs from aromatic precursor compounds (Weber and Hagenmaier, 1999; Fangmark et
al., 1994; Vogg et al., 1987, 1992; Oberg et al.,1989). Vogg et al. (1987) found that formation
probably occurs outside of and downstream from the combustion zone of a furnace, in regions
where the temperature of the combustion offgases has cooled within a range of 200 to 450°C.
After carefully removing organic contaminants from MWC fly ash, Vogg et al. (1987)
added known concentrations of isotopically labeled CDDs/CDFs to the matrix. The MWC fly
ash was then heated for 2 hr in a laboratory furnace at varying temperatures. The treated fly ash
was exposed to temperatures increasing in 50°C increments within a temperature range of 150 to
500°C. Table 2-1 summarizes these data. Because the relative concentration of CDDs/CDFs
increased while exposed to varying temperatures, it was concluded that the temperature of the
combustion gas is crucial to promoting the formation of CDDs/CDFs on the surface of fly ash.
Within a temperature range of 200 to 450°C, the concentration of CDDs/CDFs increases to some
maxima; outside this range, the concentration diminishes.
2-6
-------
Table 2-1. Concentration of CDDs/CDFs on municipal incinerator fly ash at
varying temperatures
Congener
CDD
Tetra
Penta
Hexa
Hepta
Octa
CDF
Tetra
Penta
Hexa
Hepta
Octa
CDD/CDF concentration on fly ash (ng/g) by temperature
200°C
15
40
65
100
90
122
129
61
48
12
250°C
26
110
217
208
147
560
367
236
195
74
300°C
188
517
1,029
1,103
483
1,379
1,256
944
689
171
350°C
220
590
550
430
200
1,185
1,010
680
428
72
400°C
50
135
110
60
15
530
687
260
112
12
Source: Adapted from Vogg et al. (1987).
The region of cooler gas temperature is often referred to as the "post-combustion region."
This region extends from near the exit of the furnace to the point of release of the combustion
gases at stack tip. The heat loss may be inherent in the conduction and transfer through the
combustion gas metal ducting system or related to the adsorption/exchange of heat to water in
boiler tubes.
Fangmark et al. (1994) found that CDDs/CDFs exhibit a similar dependence at a
temperature range of 260 to 430°C, with maximum formation occurring around 340°C. Using a
pilot-scale combustor, Behrooz and Altwicker (1996) found that the formation of CDDs/CDFs
from the precursor 1,2-dichlorobenzene rapidly occurred within the post-combustion region in a
temperature range of 390 to 400°C, with residence times of only 4 to 5 sec. On the other hand,
CDD/CDF formation from 1,2-dichlorophenol seemed to require higher temperatures.
Oberg et al. (1989) examined the role of temperature in the formation kinetics using a
full-scale hazardous waste incinerator (HWI) operating in Sweden. The investigators observed
that maximum CDD/CDF formation transpired in the boiler used to extract heat for cogeneration
of energy. In this study, significant increases in total concentration of I-TEQDF occurred between
280 and 400°C, and concentrations declined at temperatures above 400°C. Weber and
Hagenmaier (1999) showed that in gas-phase reactions, chlorophenols react in the presence of
2-7
-------
oxygen at temperatures above 340°C to form CDDs/CDFs. Phenoxyradicals were formed, which
in turn caused the formation of CDDs. Polychlorinated dihydroxybiphenyls were identified as
reaction intermediates in the gas-phase dimerization of chlorophenols, and these intermediates
could form CDFs.
Konduri and Altwicker (1994) proposed that rate-limiting factors were the nature and the
concentrations of the precursors, the reactivity and availability of the fly ash surface, and the
residence time in the post-combustion zone. Dickson and Karasek (1987) investigated fly ash
reactivity with 13C6-chlorophenol compounds. Several samples of fly ash from MWCs and
copper smelters and a variety of combustion fuels were heated at 300°C in quartz tubes under
conditions known to catalyze the conversion of chlorophenols to CDDs/CDFs. The MSW fly ash
included a sample from a poorly operated mass burn refractory incinerator and a sample from a
well-operated fluidized-bed combustor. The MWC fly ash proved to be the most active catalytic
medium, despite similarities among the samples with respect to specific surface area and average
pore diameter. The fly ash from the refractory MWC generated about seven times more mass of
dioxin-like compounds than did the fluidized-bed MWC. In the MSW fly ashes, all CDD/CDF
congener groups were formed from labeled chlorophenols; however, only trace amounts of
heptachloro- and octachlorodioxin were formed with the copper smelter/refiner. X-ray
photoelectron spectroscopy revealed the presence of chlorine adsorbed to the surface of the
MWC fly ash but an absence of chlorine sorbed to the copper smelter fly ash.
CDD congener groups have been postulated to form from the labeled PCP precursors by
(1) first forming octachlorodioxin by the condensation of two PCP molecules, and (2) forming
other less-chlorinated dioxins through dechlorination of the more highly chlorinated isomers.
These steps seemed to proceed by an increased reactivity of the chemisorbed precursor molecule
caused by the removal of one or more hydrogen or chlorine atoms along the ring structure
(Dickson and Karasek, 1987), an observation consistent with the kinetic model of Shaub and
Tsang(1983).
In related experiments, Dickson and Karasek (1987) more specifically reported on
forming CDDs/CDFs from condensation reactions of chlorophenols on the surface of MWC fly
ash heated in a bench-scale furnace. Their experiment was designed to mimic conditions of
MSW incineration, to identify the step-wise chemical reactions involved in converting a
precursor compound into dioxin, and to determine whether MWC fly ash could promote these
reactions. MWC fly ash was obtained from facilities in Canada and Japan. The fly ash was
rinsed with solvent to remove any organic constituents prior to initiating the experiment. Twenty
grams of fly ash were introduced into a bench-scale furnace (consisting of a simple flow-tube
combustion apparatus) and heated at 340°C overnight to desorb any remaining organic
compounds from the matrix. 13C12-labeled PCP and two trichlorophenol isotopes (13C12-2,3,5-
2-8
-------
trichlorophenol and 3,4,5-trichlorophenol) were added to the surface of the clean fly ash matrix
and placed in the oven for 1 hr at 300°C. Pure inert nitrogen gas (flow rate of 10 mL/min) was
passed through the flow tube and a constant temperature was maintained.
Tetra- through octa-CDDs were formed from the labeled PCP experiment; more than 100
|ig/g of total CDDs were produced. The congener pattern was similar to that found in MWC
emissions. The 2,3,5-trichlorophenol experiment primarily produced HxCDDs and very small
amounts of tetra- through octa-CDDs. The 3,4,5-trichlorophenol experiment mainly produced
OCDD and 1,2,3,4,6,7,8-HpCDD.
Dickson and Karasek (1987) proposed that CDDs on fly ash surfaces may result from
chlorophenol undergoing molecular rearrangement or isomerization as a result of dechlorination,
dehydrogenation, and transchlorination before condensation occurs. These reactions were
proposed as controlling the types and amounts of CDDs that are ultimately formed. Born et al.
(1993) conducted experiments on the oxidation of chlorophenols with fly ash in a quartz tube
reactor heated to about 300°C. The MWC fly ash mediated the oxidation of chlorophenols to
produce carbon dioxide (CO2) and carbon monoxide (CO) as major products and polychlorinated
benzenes, monobenzofurans, and nonhalogenated dibenzo-/>-dioxins as trace species. Formation
of these trace aromatic species occurred after residence times of only 7 to 8 sec, which was
consistent with the later experimental result of Behrooz and Altwicker (1995), which showed the
potential for rapid formation from a precursor.
Milligan and Altwicker (1996) fitted experimental flow-tube reactor data to classical
catalytic reaction models to empirically explain the interaction of 2,3,4,6-tetrachlorophenol (as a
model precursor) with reactive MWC fly ash during MSW incineration. The precursor was
found to be highly adsorptive on the surface of fly ash, with a first-order dependence on gas-
phase precursor concentration to CDD formation. The investigators concluded that
chlorophenol's dependence on gas-phase concentration to form CDDs on fly ash reflects the
highly heterogeneous nature of the fly ash surface. Moreover, the estimated 6 x 1018 adsorption
sites per gram of fly ash suggested the presence of highly energetic sites, which may be important
in the surface-catalyzed reactions forming CDDs. An interesting observation by Milligan and
Altwicker was that precursor molecules appeared to compete with oxygen molecules for the
reactive sites; therefore, chlorophenols are expected to adsorb less readily to the fly ash surface
in the presence of oxygen.
Experimental evidence suggests that condensation to CDD of chlorophenol compounds
via isomerization and the Smiles rearrangement on reactive MWC fly ash surfaces is a proven
pathway for the formation of dioxins from a precursor compound (Addink and Olie, 1995).
However, no detailed mechanisms have been presented for CDD/CDF formation from other
precursors such as chlorobenzenes under conditions simulating incineration.
2-9
-------
A condition in the synthesis of CDDs/CDFs from aromatic precursor compounds is that
the presence of a transition metal catalyst promotes the chemical reaction on the surface of fly
ash. CuCl2 is a strong catalyst for promoting surface reactions on PM to convert aromatic
precursor compounds to CDDs/CDFs (Vogg et al., 1987). CuCl2 promotes ring condensation
reactions (of the chlorophenols) on fly ash to form CDDs/CDFs (Addink and Olie, 1995) via the
Ullman reaction (Born et al., 1993). In the Ullman reaction, copper catalyzes the formation of
diphenyl ethers by the reaction of halogenated benzenes with alkali metal phenolates (Born et al.,
1993), with copper participating in a nucleophilic aromatic substitution reaction. Thus, Born et
al. proposed a similar mechanism in catalyzing the formation of dioxin-like compounds. Using
the Ullman reaction as a model, the authors proposed that the copper-catalyzed condensation of
two ortho-substituted chlorophenol molecules form chlorine-free dibenzo-p-dioxins.
Vogg et al. (1987) proposed an oxidation reaction pathway, giving rise to the formation of
CDDs/CDFs in the post-furnace regions of the incinerator in the following order: (1) HC1 is
thermolytically derived as a product of the combustion of heterogeneous fuels containing
abundant chlorinated organic chemicals and chlorides; (2) oxidation of HC1, with CuCl2 as a
catalyst, yields free gaseous chlorine via the Deacon reaction; (3) phenolic compounds (present
from combustion of lignin in the waste or other sources) entrained in the combustion plasma are
substituted on the ring structure by contact with the C12; and (4) a chlorinated precursor to dioxin
(e.g., chlorophenol) is further oxidized (with CuCl2 as a catalyst) to yield CDDs/CDFs and
chlorine.
Gullett et al. (1990a, b, 1991a, b, 1992) studied the formation mechanisms through
extensive combustion research at EPA and verified the observations of Vogg et al. (1987). It was
proven that CDDs/CDFs could ultimately be produced from low-temperature (i.e., 350°C)
reactions between chlorine (Cl) and a phenolic precursor combining to form a chlorinated
precursor, followed by oxidation of the chlorinated precursors (catalyzed by a copper catalyst
such as CuCl2), as shown below.
1. The initial step in dioxin formation is the formation of chlorine from HC1 in the
presence of oxygen (the Deacon reaction), as follows (Bruce et al., 1991; Vogg et al., 1987):
Heat
2HC1 + '/2 O2 > H2O + C12
2. Phenolic compounds adsorbed on the surface of fly ash are chlorinated to form the
dioxin precursor, and the dioxin is formed as a product of the breakdown and molecular
2-10
-------
rearrangement of the precursor. The reaction is promoted by CuCl2 acting as a catalyst (Vogg et
al., 1987; Dickson and Karasek, 1987; Gullett et al., 1992):
(a) phenol + C12 > chlorophenol (dioxin precursor)
CuCl2
(b) 2-chlorophenol + V2 O2 > dioxin + C12
Eklund et al. (1986) observed the high-temperature formation of a large variety of
chlorinated toxic compounds, including CDDs/CDFs, from precursors during a simple
experiment in which phenol was oxidized with HC1 at 550°C. One milligram of phenol was
placed in a quartz tube reactor with an aqueous solution (10 |j,L) of HC1 and heated at a
temperature of 550°C for 5 min. Trichlorobenzene, dichlorophenol, dichlorobenzofuran,
tetrachlorobenzene, trichlorophenol, and tetrachlorophenol were identified as major products
formed. Monochlorobenzene, chlorophenol, dichlorobenzene, tetrachloropropene,
pentachloropropene, trichlorobenzofuran, TCDF, TrCDD, TCDD, HxCDD, HxCDF,
pentachlorobenzene, pentachlorobiphenyl, and pentachlorodihydroxycylohexane were observed
as minor products. Trace species formed included MCDF, PeCDF, PeCDD, OCDF, and OCDD.
Eklund et al. (1986) hypothesized that chlorinated organic compounds can be produced
from phenols, acids, and any chlorine source in the hot post-combustion region (just beyond the
exit to the furnace). The reaction was seen as very sensitive to HC1 concentration. No
chlorinated compounds could be detected when HC1 concentrations were <10"3 mol.
Nestrick et al. (1987) reported that the thermolytic reaction between benzene (an
unsubstituted precursor) and iron (HI) chloride on a silicate surface yielded CDDs/CDFs at
temperatures > 150°C. The experimental protocol introduced 100 to 700 mg benzene and 13C6-
benzene into a macroreactor system consisting of a benzene volatilization chamber connected to
a glass tube furnace. The investigators noted the relevance of this experiment to generalizations
about combustion processes because benzene is the usual combustion by-product of organic
fuels. Inert nitrogen gas carried the benzene vapor to the furnace area. The exit from the glass
tubing to the furnace was plugged with glass wool, and silica gel was introduced from the
entrance end to give a bed depth of 7 cm to which ferric trichloride (FeCl3) was added to form an
FeCl3/silica reagent. The thermolytic reaction took place in a temperature range of 150 to 400°C
at a residence time of 20 min. Although di- through octa-CDDs/CDFs were formed by this
reaction at all temperatures studied, the percent yields were extremely small. Table 2-2
summarizes these data.
2-11
-------
Table 2-2. CDDs/CDFs formed from the thermolytic reaction of 690 mg
benzene + FeCl3 silica complex
Congener
DiCDD
TrCDD
TCDD
PeCDD
HxCDD
HpCDD
OCDD
Total CDDs
DiCDF
TriCDF
TCDF
PeCDF
HxCDF
HpCDF
OCDF
Total CDFs
Mass produced (ng)
4.9
54.0
130.0
220.0
170.0
98.0
20.0
696.9
990
7,800
12,000
20,000
33,000
40,000
74,000
187,000
Number of mols produced
0.019
0.019
0.400
0.620
0.440
0.230
0.040
1.940
4.2
29.0
39.0
59.0
88.0
98.0
167.0
484.2
Percent yield"
4.3 e-7
4.3 e-6
9.0 e-6
1.4 e-5
9.9 e-6
5.2 e-6
9.0 e-7
4.4 e-5
9.5 e-5
6.6 e-4
8.8e-4
1.3 e-3
2.0 e-3
1.1 e-3
3. 8 e-3
1.1 e-2
a Number of mols of CDD or CDF/mols benzene x 100.
FeCl3 = ferric chloride
Source: Nestricketal. (1987).
2.3. MECHANISM 3 (DE NOVO SYNTHESIS): SYNTHESIS OF CDDs/CDFs DURING
COMBUSTION OF ORGANIC MATERIALS
The third mechanism promotes CDD/CDF formation in combustion processes from the
oxidation of carbon particulate catalyzed by a transition metal in the presence of chlorine. As in
the precursor mechanism (mechanism 2), synthesis is believed to occur in regions outside of the
furnace zone of the combustion process, where the combustion gases have cooled to a range of
temperatures considered favorable to formation chemistry. A key component to de novo
synthesis is the production of intermediate compounds (either halogenated or nonhalogenated)
that are precursors to CDD/CDF formation. Research in this area has produced CDDs/CDFs
directly by heating carbonaceous fly ash in the presence of a transition metal catalyst without the
apparent generation of reactive intermediates. Thus, the specific steps involved in the de novo
process have not been fully and succinctly delineated. However, laboratory experimentation has
proven that MWC fly ash itself is a reactive substrate, and the matrix can actually catalyze the de
2-12
-------
novo formation chemistry. Typically, fly ash is composed of an alumina-silicate construct, with
5 to 10% concentrations of silicon, chlorine (as inorganic chlorides), sulfur, and potassium
(NATO, 1988). Twenty percent of the weight of fly ash particles is carbon, and the particles
have specific surface areas in the range of 200 to 400 m2/kg (NATO, 1988).
The de novo synthesis essentially is the oxidative breakdown of macromolecular carbon
structures, and CDDs/CDFs are formed partially from the aromatic carbon-oxygen functional
groups embedded in the carbon skeleton (Huang et al., 1999). The distinguishing feature of the
de novo synthesis over the precursor synthesis is the oxidation of carbon in particulate at the start
of the process to yield precursor compounds. In mechanism 2, the precursor compound is the
starting molecule of the condensation reactions forming CDDs/CDFs (Dickson et al., 1992). By
this distinction, however, one could argue that mechanism 3 is really an augmentation of
mechanism 2 because the production of CDDs/CDFs may still require the formation of a
CDD/CDF precursor as an intermediate species. Nevertheless, a distinction is presented here to
describe additional pathways suggested for the thermal formation of these compounds.
To delineate the de novo synthesis of CDDs/CDFs, Stieglitz et al. (1989) conducted
experiments that involved heating particulate carbon containing adsorbed mixtures of
magnesium-aluminum (Mg-Al) silicate in the presence of CuCl2 (as a catalyst to the reaction).
The authors described heating mixtures of Mg-Al silicate with activated charcoal (4% by
weight), chloride as potassium chloride (7% by weight), and CuCl2 (1% in water) in a quartz
flow tube reactor at 300°C. The retention time was varied at 15 min, 30 min, and 1, 2, and 4 hr
to obtain differences in the amounts of CDDs/CDFs that could be formed. The results are
summarized in Table 2-3. In addition to the CDDs/CDFs formed as primary products of the de
novo synthesis, the investigators observed precursors formed at the varying retention times
during the experiment. In particular, similar yields of tri- through hexachlorobenzenes, tri-
through heptachloronaphthalenes, and tetra- through heptachlorobiphenyls were quantified; this
was seen as highly suggestive of the role these compounds may play as intermediates in the
continued formation of CDDs/CDFs.
Stieglitz et al. (1989) made the following observations:
• The de novo synthesis of CDDs/CDFs via the oxidation of carbonaceous PM occurred
at a temperature of 300°C. Additionally, the experiment yielded parts-per-billion to
parts-per-million concentrations of chlorinated benzenes, chlorinated biphenyls, and
chlorinated naphthalenes through a similar mechanism. When potassium bromide
was substituted for potassium chloride as a source of halogen for the organic
compounds in the reaction, polybrominated dibenzo-^-dioxins and dibenzofurans
formed as reaction products.
2-13
-------
Table 2-3. De novo formation of CDDs/CDFs after heating Mg-Al silicate,
4% charcoal, 7% Cl, 1% CuCl2 in H2O at 300°C
Congener
TCDD
PeCDD
HxCDD
HpCDD
OCDD
Total CDDs
TCDF
PeCDF
HxCDF
HpCDF
OCDF
Total CDFs
Concentration of CDD/CDF (ng/g) by reaction time (hr)
0.25
2
110
730
1,700
800
3,342
240
1,360
2,500
3,000
1,260
8,360
0.5
4
120
780
1,840
1,000
3,744
280
1,670
3,350
3,600
1,450
10,350
1
14
250
1,600
3,500
2,000
7,364
670
3,720
6,240
5,500
1,840
17,970
2
30
490
2,200
4,100
2,250
9,070
1,170
5,550
8,900
6,700
1,840
24,160
4
100
820
3,800
6,300
6,000
17,020
1,960
8,300
14,000
9,800
4,330
38,390
Cl = chlorine
CuCl2 = copper chloride
Mg-Al = magnesium-aluminum
Source: Stieglitz et al. (1989).
• The transition metal compound CuCl2 catalyzed the de novo synthesis of CDDs/CDFs
on the surface of particulate carbon in the presence of oxygen, yielding CO2 and
chlorinated/brominated aromatic compounds.
• Particulate carbon, which is characteristic of combustion processes, may act as the
source for the direct formation of CDDs/CDFs as well as other chlorinated organics.
Stieglitz et al. (1991) investigated the role that parti culate carbon plays in the de novo
formation of CDDs/CDFs from fly ash containing appreciable quantities of organic chlorine.
The investigators found that the fly ash contained 900 ng/g of bound organic chlorine, of which
only 1% was extractable. Heating the fly ash at 300 to 400°C for several hours caused the carbon
to oxidize, leading to a reduction in the total organic chlorine in the matrix and a corresponding
increase in the total extractable organic chlorine (5% extractable total organic chlorine at 300°C
and 25 to 30% at 400°C). From this, the authors concluded that the oxidation and degradation of
carbon in fly ash are the sources of the formation of CDDs/CDFs; therefore, they are essential in
the de novo synthesis of these compounds.
2-14
-------
Addink et al. (1991) conducted a series of experiments to observe the de novo synthesis of
CDDs/CDFs in a carbon fly ash system. In this experiment, 4 g of carbon-free MWC fly ash
were combined with 0.1 g of activated carbon and placed into a glass tube between two glass
wool plugs. The glass tube was then placed into a furnace at specific temperatures ranging from
200 to 400°C. This protocol was repeated for a series of retention times and temperatures. The
investigators observed that CDD/CDF formation was optimized at 300°C and at the furnace
retention times of 4 to 6 hr. Figure 2-2 displays the relationship between retention time and
temperature in CDD/CDF production from the heating of carbon particulate.
1000
Total CDDs
Total CDFs
Rete
ntion time (hr)
Figure 2-2. The de novo synthesis of CDDs/CDFs from heating carbon
particulate at 300°C at varying retention times.
Source: Addink et al. (1991).
Addink et al. (1991) also investigated the relationship between furnace temperature and
CDD/CDF production from the heating of carbonaceous fly ash. Figure 2-3 displays this
relationship. In general, the concentration began to increase at 250°C and crested at 350°C, with
a sharp decrease in concentration above 350°C. The authors also noted a relationship between
2-15
-------
1000
Total CDD
Total CDF
1 00
200 300 400
Temperature (°C)
500
Figure 2-3. Temperature effects on CDD/CDF formation.
Source: Addink et al. (1991).
temperature and the CDD/CDF congener profile: at 300 to 350°C, the less-chlorinated tetra- and
penta-CDD/CDF congeners increased in concentration, whereas hexa-, hepta-, and octa-
CDD/CDF congeners either remained the same or decreased in concentration. The congener
profile of the original MWC fly ash (not subject to de novo experimentation) was investigated
with respect to changes caused by either temperature or residence time in the furnace. No
significant changes occurred, leading the authors to propose an interesting hypothesis for further
testing: after formation of CDDs/CDFs occurs on the surface of fly ash, the congener profile
remains fixed and insensitive to changes in temperature or residence time, indicating that some
form of equilibrium is reached in the formation kinetics.
Gullett and Lemieux (1994) used a pilot-scale combustor to study the effect of varying
combustion gas composition, temperature, residence time, quench rate, and sorbent (Ca[OH]2)
injection on CDD/CDF formation. The fly ash loading was simulated by injecting fly ash
collected from a full-scale MWC. Sampling and analysis indicated that CDDs/CDFs formed on
the injected fly ash at levels representative of those observed at full-scale MWCs. A statistical
2-16
-------
analysis of the results showed that, although the effect of combustor operating parameters on
CDD/CDF formation is interactive and very complicated, substantial reduction in CDD/CDF
formation can be realized with high-temperature sorbent injection to reduce HC1 or C12
concentrations, control excess air (which also affects the ratio of CDDs to CDFs formed), and
increase quench rate.
Milligan and Altwicker (1995) found that increases in the carbon gasification rate caused
increases in the amounts of CDDs/CDFs formed and gave further evidence linking the oxidation
of carbon to the formation of CDDs/CDFs. Neither the gas-phase CO2 or CO (products of
carbon oxidation) act as precursors to chlorobenzenes or CDDs/CDFs from reactions with carbon
particulate (Milligan and Altwicker, 1995). Activated carbon, with its high surface area and
excellent adsorptive characteristics, also has the highest gasification rate of all residual carbon
(Addink and Olie, 1995).
Experimental evidence suggests the following factors for the de novo synthesis of
CDDs/CDFs from carbon: (a) carbon consisting of imperfect and degenerated layers of graphite,
(b) the presence of oxygen, (c) the presence of chlorine, (d) catylization of the reactions by CuCl2
or some other transition metal, and (e) temperatures in the range of 200 to 350°C (Huang and
Buekens, 1995). The oxidation of carbon in fly ash is apparently inhibited at temperatures below
200°C, thus indicating the lower temperature limit for the thermal inertization ofde novo
synthesis (Lasagni et al., 2000).
Lasagni et al. (2000) determined that at a temperature of 250°C, the primary product of
the gasification of carbon in fly ash is CO2, but in a temperature range of 250 to 325°C, organic
compounds are formed as products of the oxidation of the carbon. Addink and Olie (1995)
raised the possibility that the molecular backbone of CDDs/CDFs may be present in carbon. If
this is the case, the generation of dioxins and furans from the oxidation of carbon would not
require the formation of intermediate aromatic ring structures. More work is needed to confirm
these possibilities.
The de novo synthesis of CDDs/CDFs also involves the possibility that aromatic
precursors are formed within the post-combustion zone in the following manner: (1) fuel
molecules are broken down into smaller molecular species (e.g., Cx and C2 molecules) during
primary combustion, and (2) these simple molecules recombine in the post-combustion zone to
form larger-molecular aromatic species (i.e., chlorobenzenes and chlorophenols) (Altwicker et
al., 1993). Thus, small molecular products that evolve in the hot zone of the furnace as a
consequence of incomplete fuel or feed material combustion may be important foundation
molecules to the subsequent formation of precursor compounds in the cooler, post-combustion
region.
2-17
-------
Eklund et al. (1988) reported formation of a wide range of chlorinated organic
compounds, including CDDs, CDFs, and PCBs, from the oxidation of methane with HC1 at
temperatures of 400 to 950°C in a quartz flow tube reactor. No active catalysts or reactive fly
ashes were added to the combustion system. From these experimental results, the authors
hypothesized that chlorocarbons, including CDDs/CDFs, are formed at high temperatures via a
series of reversible reactions starting with chloromethyl radicals. The chloromethyl radicals can
be formed from the reaction of methyl radicals and HC1 in a sooting flame. Methane is
chlorinated by HC1 in the presence of oxygen at high temperatures, forming chlorinated
methanes, which react with methyl radicals at higher temperatures (e.g., 800°C) to form aromatic
compounds. In an oxidative atmosphere, chlorinated phenols are formed, but alkanes and
alkenes are the primary products. The chlorinated phenols then act as precursors for the
subsequent formation of CDDs/CDFs.
Aliphatic compounds are common products of incomplete combustion, and they may be
critical to the formation of simple ring structures in the post-combustion zone (Weber et al.,
1999; Sidhu, 1999; Froese and Hutzinger, 1996a, b; Jarmohamed and Mulder, 1994). The
aromatic precursor compounds may be formed in a potentially rich reaction environment of
aliphatic compounds, reactive fly ash particles, HC1, and oxygen. Sidhu (1999) noted that
combustion of acetylene on carbon (a common combustion effluent) in the presence of gaseous
HC1 and CuCl2 (as a catalyst) at 300°C led to the formation of intermediate precursors and,
subsequently, CDDs/CDFs.
Propene oxidized at 350 to 550°C when in contact with reactive MWC fly ash in a flow
tube reactor formed a wide range of chlorinated aromatic compounds when the resulting
combustion gases were mixed with HC1 (Jarmohamed and Mulder, 1994). Although the
conversion was low (1 to 3%), the oxidation of propene on fly ash in the presence of HC1 can
yield chlorinated benzenes and monobenzofurans. Incorporating an oxygen atom into the
monobenzofuran structure then leads to the formation of monodibenzofuran. The HC1
contributes chlorine to the aromatic ring through the Deacon reaction, and cyclization on the fly
ash surface can yield cyclohexadienyl-substituted benzenes, which in turn can be further oxidized
into CDFs.
Froese and Hutzinger (1996a) investigated the heterogeneous combustion reactions of the
nonchlorinated C2 aliphatics. Acetylene, as a model aliphatic compound, was allowed to react
with precleaned MWC fly ash in a tube flow reactor at approximately 600°C. Metal oxides
(silicon dioxide [SiO2], iron oxide [Fe2O3], and copper oxide [CuO])—rather than the metal
chlorides used in other precursor experiments—were added separately as catalysts. The reactants
were put into contact with HC1 vapor, which was introduced at a constant flow rate. The
2-18
-------
acetylene flow was set at 1.1 mL/min and constantly fell to near 0.9 mL/min over 30 min.
Regulated air flow maintained homeostatic oxidation conditions.
Chlorobenzenes and chlorophenols were formed, with isomer patterns generally
resembling isomer patterns of chlorobenzene and chlorophenol emissions from MWCs. CuO
was seen as catalyzing condensation and chlorination reactions under heterogeneous conditions
to form the chlorinated CDD/CDF precursor compounds. Other more volatile compounds
formed were short-chain aliphatic products, such as chloromethane, dichloromethane, and
chloro- and dichloroacetylene. Chlorobenzene congeners were not the major products formed;
perchlorinated aliphatic compounds dominated as gas-phase reaction products.
Froese and Hutzinger (1997) noted that perchlorinated aliphatic compounds (e.g.,
hexachloropropene, hexachloro-1,3-butadiene, and hexachlorocyclopentadiene) are important
intermediates in aromatic ring formation; they concluded that the catalytic reaction of C2
aliphatic compounds at 600°C dramatically contributes to the formation of chlorinated and
nonchlorinated aromatic compounds during combustion. Thus, aliphatic compounds can form
CDD/CDF precursor compounds. Variable temperature effects were observed in the formation
of CDDs/CDFs in the same reactions. Maximal OCDD formation occurred at 400°C, and the
tetra through hepta homologue groups were maximally formed at 600°C. For CDFs, production
of more highly chlorinated homologues occurred at 400°C, and the formation of TCDFs occurred
at 500°C. Froese and Hutzinger (1996a) noted a 100-fold increase in TCDF formation at 500°C
when compared with formation at 400°C. An explanation for this increase is that the higher
temperature maximized the formation of the CDD/CDF precursor (chlorophenol) from the
aliphatic starting compound.
Froese and Hutzinger (1996b) produced polychlorinated benzene and phenol compounds
at a temperature range of 300 to 600°C, caused by the heterogeneous combustion reactions of
ethylene and ethane over fly ash in the presence of HC1, oxygen, and a metal catalyst. No
chlorobenzene congener precursors were formed from ethylene and ethane at 300°C; however,
the formation rate increased with temperature until a maximum production was achieved at
600°C. No definitive temperature dependence was observed for the formation of chlorophenols
from the aliphatic starting compounds. However, at 500°C, 2,4,6-trichlorophenol dominated the
reaction products; at 300°C, PCP was initially produced.
Froese and Hutzinger (1996b) also investigated the effects of elemental catalysts on
potentiating the heterogeneous combustion reactions by measuring the amount of chlorobenzene
and chlorophenol product formed from the reactions of ethylene/HCl over each catalyst at
600°C. The reaction with SiO2 did not have a catalytic effect. Aluminum oxide (A12O3) catalytic
action showed high intensity for the dichlorobenzene isomers and decreasing intensity for the
higher-chlorinated isomers. Comparison of the amount of dichlorobenzene product formed
2-19
-------
indicated that an equal quantity was produced with either A12O3 or fly ash; however, A12O3
formed four to five times more product than did the CuO catalyst. For tri- to hexachlorobenzene
congeners, MWC fly ash reactions produced 5 to 10 times more product than did the metal
catalysts. However, the presence of the CuO catalyst in these reactions produced a
chlorobenzene congener pattern comparable to that of the fly ash reactions. With regard to
chlorophenol production, A12O3 also produced a unique dichlorophenol pattern, suggesting that
A12O3 has a unique catalytic effect in the high-temperature reactions of C2 aliphatic compounds.
Reactions with CuO produced additional products, including chlorinated methyl
compounds, chlorinated C2 aliphatics, and perchlorinated C3-C5 alkyl compounds. Froese and
Hutzinger noted that these perchlorinated alkyl groups, formed by reacting ethylene and ethane
over fly ash in the presence of the CuO catalyst, were key intermediate compounds to the
formation of first aromatic rings in typical combustion systems. This emphasizes the importance
of copper's catalytic effects in a combustion fly ash system. A12O3 catalyzed reactions produced
nonchlorinated naphthalene and alkylbiphenyl compounds. Furthermore, the organic chlorine in
aliphatic compounds may also act as a direct source of chlorine for the formation of CDDs/CDFs
in a carbon fly ash system (Weber et al., 1999).
In an earlier experiment using a similar flow tube apparatus, Froese and Hutzinger (1994)
formed chlorinated benzenes and phenols in fly ash catalyzed reactions with trichloroethylene at
temperatures of 400 to 500°C. In this case, metal oxides (CuO, FeO3, and A12O3) were used as
catalysts, but no HC1 was added for oxychlorination of product compounds. Under combustion
conditions, temperature-dependent formation of chlorinated aromatics occurred from the
trichloroethylene starting compound. Reaction with fly ash at 600°C formed hexachlorobenzene
in concentrations that were about 1,000 times greater than those at 400 and 500°C, with similar
results for chlorophenols. The authors hypothesized that key aromatic precursors for
CDDs/CDFs are formed in the higher-temperature region of a post-combustion zone (about
600°C) and are then carried to the cooler post-combustion region (about 300°C), where the
precursors form CDDs/CDFs.
2.4. THE ROLE OF CHLORINE IN THE FORMATION OF CDDs/CDFs IN
COMBUSTION SYSTEMS
The formation of CDDs/CDFs in the post-combustion region of combustion systems via
either the precursor or de novo synthesis mechanisms requires the availability of a source of
chlorine (Luijk et al., 1994; Addink et al., 1995; Stanmore, 2004; Wikstrom et al., 2003 ).
Chlorine concentration in this region is somehow related to the chlorine content of combustion
fuels and feed materials in incineration/combustion systems because there can be no other source.
The main question regarding the role of chlorine in forming CDDs/CDFs is whether a positive
2-20
-------
and direct correlation exists between the amount of chlorine in feeds and the amount of
CDDs/CDFs formed and emitted from the stack of a combustion system. If a direct relationship
appears to exist, then reductions in the chlorine content of fuels/feeds prior to combustion should
result in a corresponding reduction in the concentrations of CDDs/CDFs formed after
combustion. If the oxychlorination reactions require a number of steps, then the relationship
between chlorine in uncombusted fuels and CDDs/CDFs formed after combustion may not be
linear, although it may still be dependent in some nonlinear association. The main question can
best be addressed by examining both formation mechanisms revealed in laboratory-scale
combustion experiments and correlations between chlorine inputs with CDD/CDF outputs in
commercial-scale combustors.
2.4.1. Review of Laboratory-Scale Studies
A wide body of experimental evidence has elucidated the direct and indirect associations
between chlorine in feeds and fuels and the potential formation of CDDs/CDFs during
combustion. The de novo synthesis of CDDs/CDFs requires two basic reactions: (1) the transfer
of chlorine to residual carbon particulate, with subsequent formation of carbon-chlorine bonds,
and (2) the oxidation of this macromolecular complex to yield CO2 and volatile and semivolatile
organic compounds as side products (Weber et al., 1999). Transition metal compounds such as
CuCl2 catalyze these reactions. Gaseous HC1, C12, and Cl- are the most abundant sources of
chlorine available for participation in the formation of CDDs/CDFs, and they are initially formed
as a combustion by-product from the inorganic and organic chlorine contained in the fuel
(Wikstrom et al., 2003; Rigo, 1998; Addink et al., 1995; Rigo et al., 1995; Halonen et al., 1994;
Luijk et al., 1994; Altwicker et al., 1993; Wagner and Green, 1993; Dickson et al., 1992; Bruce
et al., 1991; Gullet et al., 1990b; Commoner et al., 1987; Vogg et al., 1987).
MSW contains approximately 0.45 to 0.90% (w/w) chlorine (Domalski et al., 1986). The
most predominant chlorine species formed from MSW combustion is gaseous HC1, which
averages between 400 and 600 ppm in the combustion gas (Wikstrom et al., 2003; U.S. EPA,
1987a). Chlorine is initially released from the chlorine in the MSW and is rapidly transformed to
HC1 by the abstraction of hydrogen from reaction with hydrocarbons present in the fuel
(Wikstrom et al., 2003). HC1 may oxidize to yield C12 gas by the Deacon reaction, and the C12
directly chlorinates a CDD/CDF precursor along the aromatic ring structure. Further oxidation of
the chlorinated precursor in the presence of a transition metal catalyst (of which CuCl2 was found
to be the most active) yields CDDs/CDFs (Altwicker et al., 1993). Increasing the yield of
chlorine in vapor phase from HC1 oxidation generally increases the rate of CDD/CDF formation.
Formation kinetics are most favored at temperatures ranging from 200 to 450°C. However HC1
2-21
-------
is considered a weak chlorinating agent because of the tenacity of the hydrogen-to-carbon bond
of aromatic compounds (Wikstrom et al., 2003).
Chlorine production from gaseous HC1 can be reduced either by limiting initial HC1
concentration or by shortening the residence time (Bruce et al., 1991; Gullett et al., 1990b;
Commoner et al., 1987). Bruce et al. (1991) observed a general increase in CDD/CDF formation
with increases in the vapor-phase concentration of chlorine and verified a dependence of the
formation of CDDs/CDFs in the post-combustion zone on the concentration and availability of
gaseous chlorine. This latter finding is in agreement with the results of a simple experiment by
Eklund et al. (1986) in which unsubstituted phenol was mixed with HC1 at 550°C in a quartz
tube reactor. A wide range of toxic chlorinated hydrocarbons were formed, including
CDDs/CDFs. Eklund et al. (1988) also found a dependence of the amounts of chlorinated phenol
product formed from the nonchlorinated starting material on the increased amount of HC1
introduced into the reaction. Under the conditions of this experiment, no chlorinated compounds
were formed at an HC1 concentration of less than 10"3 mol, and maximum chlorophenol
concentration occurred at around 108 mol.
Born et al. (1993) also observed that increasing levels of HC1 gave rise to increasing rates
of oxychlorination of precursors, with increasing chances for the post-combustion formation of
CDDs/CDFs. However, Addink et al. (1995) observed that an HC1 atmosphere and/or chlorine
produced approximately equal quantities of CDDs/CDFs during the de novo synthesis from
oxidation of particulate carbon. Such results suggest that chlorine production via the Deacon
reaction in the de novo synthesis may not be the only chlorination pathway, and they may
indicate that the HC1 molecule can be a direct chlorinating agent. In addition, some chlorine is
expected to be formed from the oxidation of metal chlorides (e.g., CuCl2), but C12 formation from
the Deacon reaction is greater because of the continuous supply of HC1 delivered from the
combustion chamber (Bruce et al., 1991). In this case, a first-order dependence of HC1 to Cl2is
observed.
However, Wikstrom et al. (2003) reported on the importance of chlorine species on the de
novo formation of CDDs/CDFs. HC1 can react with oxidizing radicals (e.g., hydroxyl radical, or
OH) to produce C1-. Cl- are highly reactive and can replace hydrogen atoms with chlorine atoms
in the H-C bond of the aromatic structure. Thus, HC1 is most likely an indirect chlorinating agent
via the formation of C1-.
Experimentally, about 18% of the total chlorine content in fuels can be thermally
converted to Cl- in the post-combustion region (Procaccini et al., 2003). Although HC1 is the
primary chlorine-containing product formed from the combustion of chlorine-rich fuels, it may
not be the major chlorinating agent in the formation of chloro-organics in the cooled-down
region of the combustor. The experiments by Procaccini et al. (2003) indicate that the major role
2-22
-------
of HC1 in the formation of chloro-organic compounds at cooler temperatures may be that of a
chemical progenitor of C1-. HC1 reacts with the oxidizing radicals OH and O that are abundantly
present in combustion off-gases to reform C1-. Cl- readily abstract hydrogen atoms from the H-C
bond of aromatic compounds formed as combustion by-products of organic fuels. By this means,
unsubstituted aromatic compounds, e.g., benzene, undergo oxy-chlorination reactions with the
Cl- to form chlorobenzenes and chlorophenols. These products are well-defined precursor
compounds for the synthesis of CDDs/CDFs.
Wagner and Green (1993) investigated the correlation of chlorine content in feed to stack
emissions of chlorinated organic compounds in a pilot-scale incinerator using HC1 flue gas
measurements as a surrogate for fuel-bound organic chlorine. In addition to MSW as a fuel,
variable amounts of polyvinyl chloride (PVC) resin were added during 6 of 18 stack test runs.
The resulting data were regressed to determine the coefficient of correlation between HC1
measurements and total chlorobenzene compound emission measurements. In nearly all of the
regression analyses performed, the relationship between HC1 emissions and emissions of
chlorinated organic compounds was positive and well defined. In addition, the investigators
found a direct dependence of HC1 emission levels on the level of PVC in the waste, with
generally increasing amounts of HC1 formed as increasing amounts of PVC were added. From
these experiments, they concluded that decreased levels of organically bound chlorine in the
waste incinerated led to decreased levels of chlorinated organic compounds in stack emissions.
Kanters and Louw (1994) investigated a possible relationship between chlorine content in
waste feed and chlorophenol emissions in a bench-scale thermal reactor. MSW incineration with
a higher content of chlorine in the feed caused higher emissions of chlorophenols via the de novo
synthesis pathway. The investigators lowered the chlorine content of the prototype MWC by
replacing chlorine-containing fractions with cellulose. They observed appreciable decreases in
the amounts of chlorophenol formed from combustion, and concluded that reductions in the
chlorine content of waste feeds or elimination of PVC prior to municipal waste combustion
should result in a corresponding reduction in chlorophenol and CDD/CDF emissions.
In a similar experiment, Wikstrom et al. (1996) investigated the influence of chlorine in
feed materials on the formation of CDDs/CDFs and benzenes in a laboratory-scale fluidized-bed
reactor. Seven artificial fuels (composed of 34% paper, 30% wheat flour, 14% saw dust, 7%
polyethylene (PE), and 2% metals), to which varying amounts of organic chlorine and inorganic
chlorine (CaCl2 • 6H2O) were added, were combusted. The chlorine content of these fuels varied
from 0.12 to 2%. All combustion was performed with a high degree of combustion efficiency
(99.999%) to avoid the formation of polyvinylidene chloride and naphthalenes as products of
incomplete combustion of pure PVC. With the combustion conditions held constant, only the
2-23
-------
chlorine content of the fuel was varied. Flue gases were sampled for CDDs/CDFs and
chlorobenzenes.
In these experiments, concentrations of PCB isomers were approximately 1,000-fold
higher than CDDs/CDFs (expressed as concentration of I-TEQDF). Moreover, a correlation was
found between I-TEQDF and PCB levels in the flue gases and the chlorine content of the fuel. A
fivefold increase in both I-TEQDF and PCB concentrations was observed in the flue gases from
combustion of fuels containing 0.5 and 1.7% total chlorine. Furthermore, no differences were
observed in the amount of chlorinated product produced or when the source of chlorine in the
fuel was organic or inorganic. No correlation was observed between total CDD/CDF and PCB
formation and total chlorine in the feed when chlorine levels in feed were 0.5% or lower. The
highest amounts of CDDs/CDFs and PCBs were formed from the fuel with the highest total
chlorine content (1.7%).
Under the conditions of this experiment, Wikstrom et al. (1996) observed that a chlorine
fuel content of 1% was a threshold for formation of excess CDDs/CDFs and PCBs during
combustion. The authors noted that MSW in Sweden contained about 0.7% chlorine, of which
approximately 40% was organic chlorine. They concluded that MSW was below the observed
threshold value of 1% chlorine content associated with a general increase in CDD/CDF and PCB
formation in the post-combustion region. They also stated that their study did not support the
hypothesis that elimination of only PVC from waste prior to combustion will cause a significant
reduction in CDD/CDF emissions if the combustion process is well controlled (high combustion
efficiency). Wang et al. (2003) verified the existence of a theoretical chlorine-in-fuel threshold
when they demonstrated de novo synthesis when combusting fuels with 0.8 to 1.1% chlorine.
A primary by-product of PVC combustion is HC1. Paciorek et al. (1974) thermally
degraded pure PVC resin at 400°C and produced 550 mg/g HC1 vapor as a primary thermolysis
product, which was observed as being 94% of the theoretical amount, based on the percent
weight of chlorine on the molecule. Ahling et al. (1978) concluded that HC1 can act as a chlorine
donor to ultimately yield chlorinated aromatic hydrocarbons from the thermolytic degradation of
pure PVC and that these yields are a function of transit time, percent oxygen, and temperature.
They observed data from 11 separate experiments conducted with temperatures ranging from 570
to 1,130°C. These data indicated that significant quantities of various isomers of dichloro-,
trichloro-, tetrachloro-, and hexachlorobenzenes could be produced. Choudhry and Hutzinger
(1983) proposed that the radical species Cl- and FT generated in the incineration process may
attack the chlorinated benzenes and abstract hydrogen atoms to produce orthochlorine-substituted
chlorophenol radicals. These intermediate radical species then react with molecular oxygen to
yield ortho-substituted chlorophenols. As a final step, the ortho-substituted chlorophenols act as
ideal precursors to yield CDDs/CDFs with heat and oxygen. The chlorine in aliphatic
2-24
-------
compounds has been observed as both yielding high amounts of HC1 during combustion and
acting as a direct chlorine source for the de novo synthesis of CDDs/CDFs (Weber et al., 1999).
Kim et al. (2004) determined that the combustion of pure PVC yielded appreciable
amounts of polycyclic aromatic hydrocarbons (PAHs), PCBs, chlorobenzenes, and
chlorophenols. They suggested that the gas-phase production of PCBs and chlorobenzenes
contributed to the gas-phase formation of CDDs/CDFs through the precursor mechanism.
Chlorophenols, however, contributed to the de novo formation. Kim et al. (2004) reported that
the de novo synthesis of CDDs/CDFs from chlorophenols was approximately 100 times greater
than their formation from PCB and chlorobenzene precursors.
Katami et al. (2002) found a clear correlation between dioxin formation and the chlorine
content of mixed plastics combusted in a laboratory-scale incinerator. PVC, PE, polystyrene
(PS), polyethylene terephthalate (PET), and their various mixtures were burned at temperatures
greater than 600°C. Average CO concentrations in the exhaust gases were varied from 2 to 880
ppm as a general indication of the quality of the fire in the combustion chamber. When
incinerated, each type of plastic formed CDDs/CDFs in the exhaust gases. Of the total CDDs
formed, HxCDD and TCDD formed in the greatest amounts when PE was combusted. Mono-
CDF was the most abundant CDF formed from PE combustion. Mono-ortho coplanar PCBs
were preferentially formed over nonortho-PCBs. The combustion of PS caused TCDD to be
formed in the greatest abundance of all possible CDDs, whereas TCDF was the most abundant
dibenzofuran. Mono-ortho PCBs formed more than nonortho coplanar PCBs when PS was
combusted. The combustion of PET mostly formed MCDD and MCDF among the CDDs/CDFs
formed.
When PVC was combusted with the conditions of high temperature and low CO (good
combustion), a total of 53.5 ng/g of total CDD was formed, with the HxCDD predominating. In
addition, good combustion conditions formed a total of 771 ng/g of CDFs, with C12 and C13 CDF
congeners dominating. When PVC was combusted with the conditions of low temperature and
high CO (poor combustion), the total CDDs and CDFs formed increased significantly to 429 ng/g
and 8,492 ng/g, respectively. TrCDD and DiCDF dominated the congener distributions,
suggesting that poor combustion of PVC tends to form high levels of lower-chlorinated
CDDs/CDFs. The investigators observed that maintaining good combustion tended to minimize
the formation of CDDs/CDFs from the combustion of chlorinated plastics.
Shibata et al. (2003) reported on the formation of CDDs/CDFs from the combustion of
PVC in quartz ampules. Synthesis of CDDs/CDFs proceeded de novo in a temperature range of
200 to 400°C, with the reaction catalyzed by CuO. Maximum formation occurred at 300°C.
HpCDDs and OCDD were the dominant CDDs observed in the flue gases, whereas TCDFs,
2-25
-------
PeCDFs, and HxCDFs dominated the CDFs. The ratio of CDFs to CDDs from PVC combustion
was greater than 1, which is typical of MSW combustion (Shabata et al., 2003).
Addink and Altwicker (1999) reported on the role of the inorganic chloride ion in the
formation of CDDs/CDFs using the labeled compound Na37Cl. The inorganic chloride ion forms
carbon-chlorine bonds on soot particles during combustion. The chlorine in the soot can be
directly inserted into a CDD/CDF molecule during formation, or it can exchange with the
chloride ions in the transitional metal catalyst, which promotes CDD/CDF formation. Thus, the
inorganic chlorine ion participates as a chlorine donor to CDD/CDF formation.
De Fre and Rymen (1989) reported on the formation of CDDs/CDFs from hydrocarbon
combustion in a domestic gas/oil burner in the presence of 15 and 300 ppm concentrations of
HC1. More than 100 chlorinated organic compounds were detected in the flue gases whenever
HC1 was injected into the system. The investigators observed formation of CDDs and CDFs in
all experiments where HC1 was injected in a hydrocarbon flame. In this case, CDFs were always
more abundant than CDDs. It was concluded that the relationship between the HC1 concentration
and the emitted concentration of CDDs/CDFs under fixed combustion conditions appeared to be
exponential for a wide range of temperatures (240 to 900°C).
2.4.2. Review of Full-Scale Combustion Systems
The review of experimental data clearly indicates an association between chlorine content
of feed/fuels and the potential synthesis of CDDs/CDFs. Paradoxically, the review of full-scale
operating incineration processes does not yield such unequivocal results, indicating that complex
kinetic events make strong associations difficult in full-scale systems. The following is a review
of studies of the association between chlorine in feeds and stack releases of CDDs/CDFs in full-
scale incineration systems.
In the stack testing of a variety of industrial stationary combustion sources during the
National Dioxin Study in 1987, EPA made a series of qualitative observations about the
relationship between total chlorine present in the fuel/waste and the magnitude of emissions of
CDDs/CDFs from the stack of the tested full-scale combustion facilities (U.S. EPA, 1987a). In
general, combustion units with the highest CDD emission concentrations had greater quantities
of chlorine in the fuel; conversely, sites with the lowest CDD emission concentrations contained
only trace quantities of chlorine in the feed. The typical chlorine content of various combustion
fuels was reported by Lustenhouwer et al. (1980) as coal, 1,300 |ig/g; MSW, 2,500 |ig/g; leaded
gasoline, 300 to 1,600 |ig/g; and unleaded gasoline, 1 to 6 |ig/g.
Thomas and Spiro (1995) also analyzed the relationship between CDD/CDF emissions
from combustion and the chlorine content of feed materials. Thomas and Spiro (1996) plotted
average CDD/CDF emission factors for a variety of combustion systems and processes (black
2-26
-------
liquor boilers, unleaded gasoline combustion, leaded gasoline combustion, wire incineration,
cigarette combustion, sewage sludge incineration, MWC, PCP-treated wood combustion,
hazardous waste incineration, and hospital waste incineration) against the average chlorine
concentration of the combusted material. The plot showed that average CDD/CDF emissions of
combustion source categories tended to increase with the average chlorine content of the
combusted fuel. This analysis indicated that combustion sources with relatively high combustion
efficiency and adequate air pollution controls tended to have emissions two orders of magnitude
lower than those of poorly operated sources. This suggests that the magnitude of CDD/CDF
emissions is strongly dependent on chlorine concentration in fuels in the context of the more
poorly controlled and operated combustion sources, and the association becomes less apparent in
the well-controlled facilities operating with good combustion practices. The slope of the log-log
plot was between 1 and 2 for the poorly controlled and operated facilities, indicating that the
relationship between chlorine content and CDD/CDF emissions was more than proportional.
Costner (1998) reported finding a positive correlation between chlorine content of feed
material and CDD/CDF emissions at a full-scale hospital waste incinerator. Costner concluded
that emissions at this facility were dependent on chlorine input at a concentration as low as
0.031% and that there was no evidence of a threshold in the relationship between chlorine in feed
and CDD/CDF emissions.
Rigo et al. (1995) summarized the results of a study commissioned by the American
Society of Mechanical Engineers (ASME, 1995). The study was a statistical evaluation of the
relationship between HC1 concentration in flue gases and various combustion systems (MWCs,
hospital waste incinerators, HWIs, biomass combustors, laboratory combustors, and bench-scale
combustors) and stack emissions of total CDDs/CDFs. In this study, HC1 was used as a
surrogate for total chlorine content in the fuel. The data analysis was sufficient for 92 facilities in
the database that showed both HC1 and CDD/CDF emissions. Of the 92 facilities, 72 did not
show a statistically significant relationship between chlorine input and CDD/CDF output in
emissions streams, 2 showed increasing CDD/CDF concentrations with increasing chlorine, and
8 showed decreasing CDD/CDF concentrations with increasing chlorine. ASME (1995) reports
the following conclusion:
The failure to find simultaneous increases in most cases and finding inverse
relationships in a few indicates that any effect chlorine has on CDD/CDF
emissions is smaller than the variability of other causative factors. Whatever
effect chlorine has on CDD/CDF emissions in commercial-scale systems is
masked by the effect of APCS (air pollution control systems) temperature, ash
chemistry, combustion conditions, measurement imprecision, and localized flow
stratification.
2-27
-------
Liberson and Belanger (1995) reported the results of an analysis of the formation and
emission of CDDs/CDFs as a function of total chlorine in combustion feed materials at a rotary
kiln HWI. The data were generated from multiple test series conducted over a 13-month period
at the HWI while operating a carbon injection system specifically designed to control and reduce
CDD/CDF stack emissions. The chlorine feed rates ranged from 0 to 3,300 Ib/hr, and the
CDD/CDF emission rates ranged from 0.7 to 39 ng/dscm. The authors noted that multiple series
of CDD/CDF control systems were used on this HWI (a high-temperature secondary combustion
chamber, a spray dryer-evaporative quench that further cools the combustion gases, activated
carbon injection to adsorb semivolatile organics, and a cool-side electrostatic precipitator
followed by an acid gas scrubber to collect HC1 and C12). From analyses of the data, the authors
concluded that no correlation exists between CDD/CDF emissions and chlorine feed in a modern
MWC using carbon injection for CDD/CDF control.
More recently, Wang et al. (2003) investigated the association between chlorine content
of waste feeds and CDD/CDF emissions from full-scale combustion systems. Previously,
Wikstrom et al. (1996) had discerned a chlorine content in feeds of 1% as being a threshold
concentration for the formation of CDDs/CDFs, i.e., an association with the magnitude of
CDDs/CDFs formed occurred only when chlorine content in the feed was > 1%. Wang et al.
confirmed the apparent existence of a chlorine threshold for emissions of total CDDs/CDFs after
statistically reviewing input of chlorine in feed versus output of CDDs/CDFs in emissions at two
tested medical incinerators and two tested MWCs. Additionally, the authors examined second-
hand data from 13 other dioxin sources obtained from the literature and found that the formation
of CDFs was greater than the formation of CDDs when the chlorine content of the waste feed
exceeded the threshold. However, when the chlorine content was below the approximate 1%
threshold, the formation of CDDs was greater than the formation of CDFs. The authors proposed
that chlorine content below the threshold formed chlorinated precursors to CDDs rather than
forming the dibenzofuran molecule. Chlorine content above the threshold contributed to
deterioration of combustion conditions, causing the formation of PAHs, which, in turn,
contributed to the formation of CDFs.
2.5. POTENTIAL PREVENTION OF CDD/CDF FORMATION IN COMBUSTION
SYSTEMS
Given what is currently understood about oxychlorination reactions in the synthesis of
CDDs/CDFs, researchers have identified certain interventions that could be taken to reduce or
impede formation in combustion systems. Raghunathan and Gullett (1996) demonstrated in a
pilot-scale incinerator that sulfur compounds can combine with the metal catalyst necessary to
stimulate the Deacon reaction of HC1 and oxygen to yield C12, thereby neutralizing the catalyzing
2-28
-------
agent and reducing the formation of CDDs/CDFs. The Deacon reaction, which forms C12 in the
combustion plasma, is seen as occurring only in the presence of a catalyst. Thus, the sulfur
dioxide (SO2) molecule (formed when sulfur in the fuel combines with oxygen) can inhibit the
catalytic activity of the fly ash by either combining with a metal-based Deacon catalyst in the fly
ash or depleting the C12 formed. The authors observed that the principal action of sulfur in
inhibiting the formation of CDDs/CDFs in combustion systems is through SO2 depletion of C12,
as follows:
C12 + SO2 + H2O « 2HC1 + SO2
The relevance of this finding is that the co-combustion of MSW with coal (that contains
sulfur) should lead to dramatic reductions in the amount of CDDs/CDFs formed and emitted, and
it may explain why, in the United States, coal combustion at power plants results in CDD/CDF
emission rates more than a magnitude lower than those at MWCs.
Naikwadi and Karasek (1989) investigated the addition of calcium oxide (CaO) and
triethylamine (TEA) to the flue gases of a combustion system as an inhibitor of the catalytic
activity of fly ash. They placed 500 |ig 13C-labeled PCP (a dioxin precursor) in a combustion
flow tube and allowed it to react with organic-extracted MWC fly ash at 300°C under an air
stream. Under these conditions, CDDs/CDFs were formed at concentrations ranging from 1,660
to 2,200 ng/100 |ig 13C-PCP. The experimental method was then modified by mixing reactive
MWC fly ash with either CaO or TEA. The results showed that the amount of CDDs/CDFs
formed could be reduced by an order of magnitude from the reaction of PCP with fly ash and the
addition of TEA as an inhibitor. When CaO was mixed with fly ash, the amount of CDDs/CDFs
formed decreased more than 20-fold.
2.6. THEORY ON THE EMISSION OF PCBs
Air emissions of PCBs from MSW incineration is less well studied. Probably the
formation mechanisms that apply to CDDs/CDFs would also apply to PCBs. Mechanism 1 (pass
through) is implicit in the Toxic Substances Control Act rule, which requires 99.9999%
destruction in HWIs. When this occurs, 0.0001% of the initial amount of PCBs fed into the HWI
may be emitted from the stack. This may indicate that some small fraction of the PCBs present
in the fuel fed into an incineration process may result in PCB emissions from the stack of the
process.
PCBs have been measured as contaminants in raw refuse prior to incineration in an MWC
(Choudhry and Hutzinger, 1983; Federal Register, 199la). Using this information, it is possible
to test mechanism 1 for CDD/CDF emissions: that the PCB contamination present in the fuel is
2-29
-------
mainly responsible for emissions from the stack. The mass balance of total PCBs, beginning
with measurement in the raw refuse and ending with measurement at the stack of a refuse-
derived fuel (RDF) MWC (Federal Register, 199 la), can be used to calculate the destruction
rated efficiency (DRE) of incineration of the PCB-contaminated MSW. Using results from test
number 1 1 at the RDF facility (Federal Register, 199 la), a computation of DRE can be made
using the following equation (B runner, 1984):
DRE= - x 100%
Wj
where:
Wj = mass rate of contaminant fed into the incinerator system
W0 = mass rate of contaminant exiting the incinerator system
In test 11, 811 ng total PCB/g refuse were measured in the MSW fed into the incineration
system and 9.52 ng/g were measured at the inlet to the pollution control device (i.e., outside the
furnace region but preceding emission control). From these measurements, a DRE of 98.8% can
be calculated. Therefore, it appears that PCB contamination in the raw MSW fed into this
particular incinerator may have accounted for the PCB emissions from the stack of the MWC.
PCBs can be thermolytically converted into CDFs (Choudhry and Hutzinger, 1983;
U.S. EPA, 1984). This process occurs at temperatures somewhat lower than those typically
measured inside the firebox of an MWC. Laboratory experiments conducted by EPA indicate
that the optimum conditions for CDF formation from PCBs are near a temperature of 675°C in
the presence of 8% oxygen and a residence time of 0.8 sec (U.S. EPA, 1984). This resulted in a
3 to 4% efficiency of conversion of PCBs into CDFs. Because 1 to 2% of the PCBs present in
the raw refuse may survive the thermal stress imposed in the combustion zone of the incinerator
(Federal Register, 199 la), it is reasonable to presume that PCBs in the MSW may contribute to
the total mass of CDF emissions released from the stack of the incinerator.
Although it appears that contamination of waste feed with PCBs may be an important
factor in detecting PCBs in stack emissions from combustion processes, recent research has
indicated that these compounds may also be formed in the post-combustion region, either from
de novo synthesis or from precursor compounds. Zheng et al. (1999) observed the formation of
PCBs in the post-combustion region from the pyrolysis of chlorobenzenes using a laboratory-
scale furnace. The investigators observed that PCBs were optimally formed from less-
chlorinated chlorobenzenes (e.g., 1,3-dichlorobenzene) catalyzed by CuCl2. In this experiment,
maximum PCB production occurred at a temperature of 350°C. Wikstrom et al. (1998) reported
2-30
-------
secondary formation of PCBs in the post-combustion region similar to the de novo synthesis of
CDDs/CDFs, albeit PCBs were formed in only small amounts relative to CDDs/CDFs.
Fangmark et al. (1994) postulated that formation of PCBs and CDDs/CDFs in the post-
combustion region may occur through the same mechanisms. On the other hand, Blumenstock et
al. (1998) produced results in a pilot-scale furnace that were inconsistent with the de novo
formation of CDDs/CDFs in the post-combustion region (i.e., PCBs seemed to be optimally
formed at high temperatures in oxygen-deficient atmospheres). Shin and Chang (1999) noted a
positive correlation between PCB concentrations on MSW incineration fly ash and fly ash
concentrations of CDDs/CDFs, suggesting that high PCB levels in fly ash may be a contributory
cause of the post-combustion formation of CDDs/CDFs (i.e., PCBs are precursors to
CDDs/CDFs). Nito et al. (1997) noted the formation of CDDs/CDFs from the pyrolysis of PCBs
in a fluidized-bed system, indicating that PCBs in feeds may account for CDFs formed in MSW
incineration. More combustion-related research needs to be conducted to firmly establish
whether PCB contamination in feeds or post-combustion formation (or both) may explain the
presence of PCBs in combustion flue gases.
2.7. SUMMARY AND CONCLUSIONS
2.7.1. Mechanisms of Formation of Dioxin-Like Compounds
There are three primary mechanisms for CDD/CDF emissions from combustion sources.
Mechanism 1 (pass through). This mechanism involves CDDs/CDFs contained in the
feed passing through the combustor intact and being subsequently released into the environment.
For most systems, this is not thought to be a major contributor to CDD/CDF emissions for three
reasons. First, for commercial systems with good combustion controls, the temperatures and
residence times should result in the destruction of most CDDs/CDFs in the feed. Second, mass
balance studies of a number of combustion systems show that more CDDs/CDFs can be detected
in the cool-down region downstream of the furnace than in the feed. Third, the CDD/CDF
congener profile in the feed differs from the congener profile in the stack emissions.
Consequently, synthesis appears to be a more important mechanism than is pass through. The
concentration of CDDs/CDFs in the flue gases of any particular combustion system will
ultimately be derived as a result of the balance between reactions leading to formation and
reactions leading to destruction of these compounds.
Mechanism 2 (precursor). This mechanism involves the formation of CDDs/CDFs
from the thermal breakdown and molecular rearrangement of aromatic precursors either
originating in the feed or forming as a product of incomplete combustion. Actual synthesis of
CDDs/CDFs occurs in the post-combustor environment. Gaseous benzene is the most abundant
aromatic compound associated with products of incomplete combustion of waste. Benzene
2-31
-------
reacts with Cl- within the combustion gas plasma, causing aromatic H abstraction and the
subsequent formation of chlorobenzenes and chlorophenols. Homogeneous gas-phase formation
of CDDs/CDFs occurs from these precursor compounds at temperatures >500°C, catalyzed by
the presence of copper compounds. In addition, the CDDs/CDFs can form from gas-phase
precursors as heterogeneous, catalytic reactions with reactive fly ash surfaces. This reaction has
been observed to be catalyzed by the presence of a transition metal sorbed to the fly ash. The
most potent catalyst is CuCl2. Relatively low temperatures—in the range of 200 to 450°C—have
been identified as a necessary condition for these heterogeneous reactions to occur, with either
lower or higher temperatures inhibiting the process. Because these reactions involve
homogeneous gas-phase and heterogeneous solid-phase chemistry, the rate of emissions is less
dependent on reactant concentration than on conditions that are favorable to formation, such as
temperature, retention time, source and species of chlorine, and the presence of a catalyst.
Mechanism 3 (de novo synthesis). This mechanism involves the heterogeneous solid-
phase formation of CDDs/CDFs in the post-combustion environment on the surface of fly ash.
Such heterogeneous chemistry occurs in two ways: (1) directly from the oxidation of carbon
within the fly ash and subsequent reactions with organic and inorganic chlorine, and (2) the
oxidative breakdown of macromolecular carbon structures (e.g., graphite) and oxychlorination
reactions of aromatic precursors (such as chlorobenzenes and chlorophenols) on fly ash surfaces,
leading to CDD/CDF formation. In either case, formation kinetics is most favored at
temperatures in the range of 200 to 450°C and is promoted by the catalytic properties of either
the fly ash or the presence of a transition metal compound.
Mechanisms 2 and 3 can occur simultaneously, share a number of common reaction
pathways, and occur in the same physical environment, and they are controlled by many of the
same physical conditions. In well-designed and well-operated combustion systems, the precursor
species needed for mechanism 2 are reduced; consequently de novo synthesis (mechanism 3) can
become the dominant pathway for formation. In systems with incomplete combustion, it is
difficult to sort out the relative contribution of these two mechanisms to total emissions. Both
mechanisms, however, can be curtailed if steps are taken to minimize the physical conditions
needed to support formation (i.e., time, temperature, and reactive surface).
The combustion formation chemistry of PCBs is less well studied than that of
CDDs/CDFs, but it is reasonable to assume that these same three mechanisms would apply. For
waste incineration, PCBs can exist in significantly higher concentrations in the feed than do
CDDs/CDFs. Consequently, mechanism 1 may play a more prominent role in the origin of PCB
emissions than of CDD/CDF emissions.
2-32
-------
2.7.2. Role of Chlorine
From the various analyses on the role and relationship of chlorine to CDD/CDF formation
and emissions, the following observations and conclusions are made.
1. Although chlorine is an essential component in the formation of CDDs/CDFs in
combustion systems, the empirical evidence indicates that, for commercial-scale incinerators,
chlorine levels in feed are not the dominant controlling factor for the amount of CDDs/CDFs
released in stack emissions. Important factors that can affect the rate of CDD/CDF formation
include overall combustion efficiency, post-combustion flue gas temperatures and residence
times, and the types and designs of air pollution control devices employed on combustion
systems. Data from bench-, pilot-, and commercial-scale combustors indicate that CDD/CDF
formation can occur by three principal mechanisms. Some of these data, primarily from bench-
and pilot-scale combustors, have shown direct correlation between chlorine content in fuels and
rates of CDD/CDF formation. Other data, primarily from commercial-scale combustors, show a
weaker relationship between the presence of chlorine in feed and fuels and rates of CDD/CDF
released from the stacks of combustion systems. The conclusion that the amount of chlorine in
feed is not a strong determinant of the magnitude of CDD/CDF stack emissions applies to the
overall population of commercial-scale combustors. For any individual commercial-scale
combustor, circumstances may exist in which changes in chlorine content in feed could affect
CDD/CDF emissions. For uncontrolled combustion, such as open burning of household waste,
chlorine content of wastes may play a more significant role in levels of CDD/CDF emissions
than the one observed in commercial-scale combustors.
2. Both organic and inorganic forms of chlorine in combustion fuels yield abundant
gaseous HC1 in the post-combustion region. It has been shown that Cl- are the most potent
chlorinating agent in the formation of chloro-organic compounds from combustion. HC1 may be
the dominant chemical progenitor of Cl- participating in oxychlorination reactions to CDD/CDF
synthesis. Formation of Cl- from HC1 occurs in the cool-down zone via the oxidation of HC1 in
the presence of a transition metal catalyst (the Deacon reaction). Although the preponderance of
scientific evidence suggests that this is an important pathway for producing chlorinated
compounds in emissions, it is still unclear whether HC1 can also directly chlorinate aromatics or
whether it must first be oxidized to yield C12.
3. Laboratory-scale experiments have examined correlations between chlorine content of
feeds and total CDDs/CDFs formed in combustion systems. These experiments suggest that for
feeds containing <1% chlorine, the amount of CDDs/CDFs formed is independent of the chlorine
2-33
-------
content of the feed. For feeds with a chlorine content >1%, a direct correlation was observed.
The existence of an apparent threshold to the chlorine content of waste has been verified in full-
scale combustion systems. It has not been determined, however, whether these relationships are
relevant to poorly controlled combustion of wastes and biomass, such as backyard barrel burning,
landfill fires, and agricultural burning.
4. The combustion of PVC can contribute to the formation of CDDs/CDFs in two ways.
First, gaseous HC1 is a primary product formed from the combustion of PVC. We have seen that
HC1 is a major contributor of Cl- necessary for the formation of CDDs/CDFs. Thus, PVC
indirectly contributes to dioxin synthesis. Second, the combustion of PVC directly forms
benzene, which is followed by oxychlorination reactions that further form chlorinated benzenes
and chlorinated phenols; these compounds then act as precursors to CDD/CDF formation.
5. The most critical factors associated with minimizing CDD/CDF formation in
combustion systems are (a) achieving nearly complete combustion of the feed through the
application of good combustion practice (i.e., time, temperature, and turbulence), and (b)
ensuring that combustion gases are quenched to below the temperature range for heterogeneous
solid-phase formation chemistry in the post-combustion region of the system, i.e., reducing the
temperature to below 200°C.
2.7.3. General Conclusion
Although the formation chemistry of CDDs/CDFs is more complicated and less
understood than the relatively simple constructs described in this review, the current weight of
evidence suggests that the formation mechanisms outlined above describe the principal pathways
of most CDD/CDF formation and emission from combustion sources.
2-34
-------
3. COMBUSTION SOURCES OF CDDs/CDFs: WASTE INCINERATION
Incineration is the destruction of solid, liquid, or gaseous wastes through the application
of heat within a controlled combustion system. The purposes of incineration are to reduce the
volume of waste that needs land disposal and to reduce the toxicity of the waste. In keeping with
this definition, incinerator systems can be classified by the types of wastes incinerated:
municipal solid waste (MSW) incineration (commonly referred to as municipal waste
combustion), medical and pathological waste incineration, hazardous waste incineration, sewage
sludge incineration, tire incineration, and biogas flaring. Each of these types of incineration is
discussed in this chapter. The purpose of this chapter is to characterize and describe waste
incineration technologies in the United States and to derive estimates of annual releases of CDDs
and CDFs into the atmosphere from waste incineration facilities for reference years 1987, 1995,
and 2000.
As discussed in Chapter 2, combustion research has developed three theories on the
mechanisms involved in the emission of CDDs/CDFs from combustion systems: (1)
CDDs/CDFs can be introduced into the combustor with the feed and pass through the system
unchanged (pass through mechanism), (2) CDDs/CDFs can be formed during combustion
(precursor mechanism), and/or (3) CDDs/CDFs can be formed via chemical reactions in the post-
combustion portion of the system (de novo synthesis). Total CDD/CDF emissions are likely to
be the net result of all three mechanisms; however, the relative importance of each mechanism is
often uncertain.
To the extent practical with the available data, the combustors in each source category
were divided into classes according to similarity of emission factors. This classification effort
attempted to reflect the emission mechanisms described above. The emission mechanisms
suggest that the aspects of combustor design and operation that could affect CDD/CDF emissions
are (a) furnace design, (b) composition of the waste feed, (c) temperature in the post-combustion
region of the system, and (d) the type of air pollution control device (APCD) used to remove
contaminants from the flue gases. Therefore, incineration systems that are similar in terms of
these factors should have similar CDD/CDF emissions. Accordingly, this chapter proposes
classification schemes that divide combustors into a variety of design classes based on these
factors. Design class, as used here, refers to the combination of furnace type and accompanying
APCD.
3.1. MUNICIPAL WASTE COMBUSTION
As noted above, emissions can be related to several factors, including furnace design,
composition of the waste feed, temperature in the post-combustion region of the system, and
5-1
-------
type of APCD used to remove contaminants from the flue gases. This section proposes a
classification scheme that divides municipal waste combustors (MWCs) into a variety of design
classes based on those factors. Because different APCDs are operated at different temperatures,
operating temperature is used to define some design classes. To account for the influence of the
waste feed, the proposed classification system distinguishes between refuse-derived fuel (RDF)
and normal MSW. This section begins with a description of the MWC technology and then
proposes the design classification scheme. Using this scheme, the municipal waste combustion
industry is characterized for the reference years 1987, 1995, and 2000. Finally, the procedures
for estimating emissions are explained and the results summarized.
3.1.1. Description of Municipal Waste Combustion Technologies
For the purposes of this report, municipal waste combustion furnace types are divided
into three major categories: mass burn, modular, and RDF. Mass burn and RDF technologies
dominate the large MWC category and modular technology dominates the small MWC category.
Each of these furnace types is described below, followed by a description of the APCDs used
with the system.
3.1.1.1. Furnace Types
Mass burn. This furnace type was so named because it burned MSW as received (i.e.,
no preprocessing of the waste was conducted other than removal of items too large to go through
the feed system). Today, a number of other furnace types also burn unprocessed waste, as
described below. Mass burn furnaces are distinguished from the other types because they burn
the waste in a single stationary chamber. In a typical mass burn facility, MSW is placed on a
grate that moves through the combustor. Three subcategories of mass burn technologies are
described below.
• Mass burn refractory-walled systems represent an older class of MWCs (generally
built in the late 1970s to early 1980s) that were designed only to reduce the volume of
waste disposed of by 70 to 90%. These facilities usually lack boilers to recover the
combustion heat for energy purposes. In the mass burn refractory-walled design, the
MSW is delivered to the combustion chamber by a traveling grate or a ram feeding
system. Combustion air in excess of stoichiometric amounts (i.e., more oxygen is
supplied than is needed for complete combustion) is supplied both below and above
the grate. As of 2000, few mass burn refractory-walled MWCs remain; almost all
have closed or been dismantled.
• Mass burn waterwall (MB-WW) facilities represent enhanced combustion efficiency,
as compared with mass burn refractory-walled incinerators. Although it achieves
similar volume reductions, the MB-WW incinerator design provides a more efficient
delivery of combustion air, resulting in higher sustained temperatures. Figure 3-1 is a
O O
3-2
-------
Belt Total
Conveyor Ash
Discharge
Figure 3-1. Typical mass burn waterwall municipal solid waste combustor.
Source: U.S. EPA(1997a).
schematic of a typical MB-WW MWC. The term "waterwall" refers to a series of
steel tubes that run vertically along the walls of the furnace and contain water. Heat
from combustion produces steam, which is then used to drive an electrical turbine
generator or for other industrial needs. This transfer of energy is called energy
recovery. MB-WW incinerators are the dominant form of incinerator found at large
municipal waste combustion facilities.
Mass burn rotary kilns use a water-cooled rotary combustor that consists of a rotating
combustion barrel configuration mounted at a 15- to 20-degree angle of decline. The
refuse is charged at the top of the rotating kiln by a hydraulic ram (Donnelly, 1992).
Preheated combustion air is delivered to the kiln through various portals. The slow
rotation of the kiln (10 to 20 rotations/hr) causes the MSW to tumble, thereby
exposing more surface area for complete burnout of the waste. These systems are
also equipped with boilers for energy recovery. Figure 3-2 is a schematic of a typical
rotary kiln combustor.
5-3
-------
Figure 3-2. Typical mass burn rotary kiln combustor.
Source: U.S. EPA(1997a).
Modular. This is a second general type of municipal waste combustion furnace used in
the United States. As with the mass burn type, modular incinerators burn waste without
preprocessing. Modular MWCs consist of two vertically mounted combustion chambers (a
primary and a secondary chamber). In 1995, the combustion capacity of modular combustors
ranged from 4 to 270 metric tons per day, i.e., they are predominately small MWCs. The two
major types of modular systems, excess air and starved air, are described below.
The modular excess-air system consists of a primary and a secondary combustion
chamber, both of which operate with air levels in excess of stoichiometric
requirements (i.e., 100 to 250% excess air). Figure 3-3 illustrates a typical modular
starved-air MWC.
5-4
-------
To Stack or
Waste Heat Boiler
Primary Air
Secondary
Gas Burner
Ash
Quench
Figure 3-3. Typical modular starved-air combustor with transfer rams.
Source: U.S. EPA(1997a).
In the starved (or controlled) air type of modular system, air is supplied to the primary
chamber at substoichiometric levels. The products of incomplete combustion entrain
in the combustion gases that are formed in the primary combustion chamber and then
pass into a secondary combustion chamber. Excess air is added to the secondary
chamber, and combustion is completed by elevated temperatures sustained with
auxiliary fuel (usually natural gas). The high, uniform temperature of the secondary
chamber, combined with the turbulent mixing of the combustion gases, results in low
levels of particulate matter (PM) and organic contaminants being formed and emitted.
Therefore, many existing modular units lack post-combustion APCDs. Figure 3-4 is
a schematic view of a modular excess-air MWC.
Refuse-derived fuel (RDF). The third major type of MWC furnace technology is
designed to combust RDF; this technology is generally used at very large MWC facilities. RDF
is a general term that describes MSW from which relatively noncombustible items are removed,
5-5
-------
Figure 3-4. Typical modular excess-air combustor.
Source: U.S. EPA(1997a).
thereby enhancing the combustibility of the waste. RDF is commonly prepared by shredding,
sorting, and separating out metals to create a dense MSW fuel in a pelletized form having a
uniform size. Three types of RDF systems are described below.
The dedicated RDF system burns RDF exclusively. Figure 3-5 shows a typical
dedicated RDF furnace using a spreader-stoker boiler. Pelletized RDF is fed into the
combustor through a feed chute using air-swept distributors; this allows a portion of
the feed to burn in suspension and the remainder to burn out after falling on a
horizontal traveling grate. The traveling grate moves from the rear to the front of the
furnace, and distributor settings are adjusted so that most of the waste lands on the
rear two-thirds of the grate. This allows more time to complete combustion on the
grate. Underfire and overfire air are introduced to enhance combustion, and these
incinerators typically operate at 80 to 100% excess air. Waterwall tubes, a
-------
Superheater
Figure 3-5. Typical dedicated refuse-derived fuel-fired spreader stoker boiler.
Source: U.S. EPA(1997a).
superheater, and an economizer are used to recover heat for production of steam or
electricity. The 1995 inventory indicated that dedicated RDF facilities range from
227 to 2,720 metric tons per day total combustion capacity.
Cofired RDF furnaces burn either RDF or normal MSW along with another fuel.
The fluidized-bed RDF furnace burns the waste in a turbulent and semisuspended bed
of sand. The MSW may be fed into the incinerator either as unprocessed waste or as
a form of RDF. The RDF may be injected into or above the bed through ports in the
combustor wall. The sand bed is suspended during combustion by introducing
underfire air at a high velocity, hence the term "fluidized." Overfire air at 100% of
stoichiometric requirements is injected above the sand suspension. Waste-fired
fluidized-bed RDF furnaces typically operate at 30 to 100% excess air levels and at
5-7
-------
bed temperatures around 815°C. A typical fluidized-bed RDF furnace is represented
in Figure 3-6. The technology has two basic designs: a bubbling-bed incineration
unit and a circulating-bed incineration unit. The 1995 inventory indicated that
fluidized-bed MWCs have capacities ranging from 184 to 920 metric tons per day.
These systems are usually equipped with boilers to produce steam.
Exhaust and Ash
Thermocouple
Sludge
Inlet
Fluidizing
Air Inlet
Pressure Tap
Sight
Glass
Burner
Tuyeres
Fuel Gun
Pressure Tap
Startup
Preheat
Burner
for Hot
Windbox
Figure 3-6. Fluidized-bed refuse-derived fuel furnace.
Source: U.S. EPA(1997a).
-------
3.1.1.2. Air Pollution Control Devices
MWCs are commonly equipped with one or more post-combustion APCDs to remove
various pollutants such as PM, heavy metals, acid gases, and organic contaminants prior to
release from the stack (U.S. EPA, 1992a). Types of APCDs include
• Electrostatic precipitator
• Fabric filter
• Spray dry scrubbing system
• Dry sorbent injection
• Wet scrubber
Electrostatic precipitator (ESP). The ESP is generally used to collect and control PM
that evolves during MSW combustion by introducing a strong electrical field into the flue gas
stream; this in turn charges the particles entrained in the combustion gases (Donnelly, 1992).
Large collection plates receive an opposite charge to attract and collect the particles. CDD/CDF
formation can occur within the ESP at temperatures in the range of 150°C to about 350°C. As
temperatures at the inlet to the ESP increase from 150 to 300°C, CDD/CDF concentrations have
been observed to increase by approximately a factor of 2 for each 30°C increase in temperature
(U.S. EPA, 1994a). As the temperature increases beyond 300°C, formation rates decline.
Although ESPs in this temperature range efficiently remove most particulates and the
associated CDDs/CDFs, the CDD/CDF formation that does occur can result in a net increase in
CDD/CDF emissions. This temperature-related formation of CDDs/CDFs within the ESP can be
applied, for purposes of this report, to distinguish cold-sided ESPs, which operate at or below
230°C, from hot-sided ESPs, which operate at an inlet temperature greater than 230°C. Most
ESPs have been replaced with better-performing and lower-cost fabric filter technology.
Fabric filter (FF). FFs are also PM control devices that remove dioxins associated with
particles and any vapors that adsorb to the particles. The filters are usually 6- to 8-inch-diameter
bags, 30 feet long, made from woven fiberglass material, and arranged in series. An induction
fan forces the combustion gases through the tightly woven fabric. The porosity of the fabric
allows the bags to act as filter media and retain a broad range of particle sizes (down to less than
1 |im in diameter). The FF is sensitive to acid gas; therefore, it is usually operated in
combination with spray dryer (SD) adsorption of acid gases.
Spray dry scrubbing system (SDSS). Spray dry scrubbing, also called SD adsorption,
involves the removal of both acid gas and PM from the post-combustion gases. By themselves,
the units probably have little effect on dioxin emissions. In a typical SDSS, hot combustion
gases enter a scrubber reactor vessel. An atomized hydrated lime slurry (water plus lime) is
injected into the reactor at a controlled velocity (Donnelly, 1992). The slurry rapidly mixes with
5-9
-------
the combustion gases within the reactor. The water in the slurry quickly evaporates, and the heat
of evaporation causes the combustion gas temperature to rapidly decrease. The neutralizing
capacity of hydrated lime reduces the acid gas constituents of the combustion gas (e.g., HC1 and
SO2) by greater than 70%. A dry product consisting of PM and hydrated lime settles to the
bottom of the reactor vessel.
SDSS technology is used in combination with ESPs or FFs. SDSSs reduce ESP inlet
temperatures to make a cold-sided ESP. In addition to acid gas, particulate, and metals control,
SDSSs with FFs or ESPs achieve greater than 90% dioxin control (U.S. EPA, 1992a), and they
typically achieve greater than 90% SO2 and HC1 control.
Dry sorbent injection (DSI). DSI is used to reduce acid gas emissions. As with SDSSs,
these units by themselves probably have little effect on dioxin emissions. In this system, dry
hydrated lime or soda ash is injected directly into the combustion chamber or into the flue ductof
the hot post-combustion gases. In either case, the reagent reacts with and neutralizes the acid gas
constituents (Donnelly, 1992).
Wet scrubber (WS). WS devices are designed for acid gas removal and are more
common to MWC facilities in Europe than in the United States. They should help reduce
emissions of dioxin in both vapor and particle forms. The devices consist of two-stage
scrubbers. The first stage removes HC1, and the second stage removes SO2 (Donnelly, 1992).
Water is used to remove HC1, and caustic or hydrated lime is added to remove SO2 from the
combustion gases.
Other types of APCDs. In addition to the APCDs described above, some less common
types are also used in some MWCs. An example is activated carbon injection (CI) technology.
Activated carbon is injected into the flue gas prior to the gas reaching SDSSs with FFs (or an
ESP). Dioxin (and mercury) are absorbed onto the activated carbon, which is then captured by
the FFs or ESP. CI technology improves dioxin control technologies by an additional 75% and
is commonly referred to as flue gas polishing. Many APCDs have been retrofitted to include CI,
including more than 120 large MWCs.
3.1.1.3. Classification Scheme
Based on the array of municipal waste combustion technologies described above, a
classification system for deriving CDD/CDF emission estimates was developed. Assuming that
facilities with common design and operating characteristics have a similar potential for
CDD/CDF emissions, the MWCs operating in 1987 and 1995 were divided into categories
according to the eight furnace types and six APCDs described above. This resulted in 17 design
classes in 1987 and 40 design classes in 1995. Because fewer types of APCDs were used in 1987
3-10
-------
than in 1995, fewer design classes are needed for estimating emissions. The MWCs operating in
2000 were divided into three furnace types and 12 APCDs, resulting in 36 design classes.
Design classes for all three reference years are summarized in Figures 3-7 through 3-9.
3.1.2. Characterization of MWCs in Reference Years 2000,1995, and 1987
Table 3-1 lists, by design/APCD type, the number of facilities and activity level (kg
MSW incinerated/yr) for MWCs in reference year 2000. Similar inventories are provided for
reference years 1995 and 1987 in Tables 3-2 and 3-3, respectively. This information was
derived from five reports: U.S. EPA (1987b), SAIC (1994), Taylor and Zannes (1996), Solid
Waste Technologies (1994), and a memorandum dated March 27, 2003, from Jason Huckaby,
ERG, to Walt Stevenson, U.S. EPA. In general, the information was collected via telephone
interviews with the plant operators.
1
Mass Burn
Muncipal Solid Waste
Incinerator Design
Classes for 1987
I
i i
Refuse-Derived Fuel Modular
Rotary Kiln
Combustor
-
H-ESP
FF
Excess-air
-
Uncontrc
EGB
WS
Figure 3-7. Municipal waste combustor design classes for 1987.
DS/FF = Dry scrubber combined with a fabric filter
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator (temperature at control device is >230°C)
WS = Wet scrubber
3-11
-------
DS/FF
DSI/FF
DS/CI/FF
DS/C-ESP
DSI/H-ESP
DSI/CI/H-ESP
C-ESP
H-ESP
Muncipal Solid Waste
Incinerator Design
Classes for 1995
ws
C-ESP
DS/FF
DSI/FF
Rotary Kiln
Combustor
-
C-ESP
DSI/C-ES
DSI/FF
DS/FF
DS/FF
DSI/FF
DS/C-ESP
DSI/H-ESP
C-ESP
H-ESP
DS/FF/C-ESP
Fluidized-bed
-
DS/FF
DSI/FF
DSI/EGB
Uncontrolled
C-ESP
H-ESP
WS
WS/FF
DSI/FF
DS/DSI/C-ESP
Uncontrolled
C-ESP
WS/C-ESP
DS/FF
DSI/FF
DSI/H-ESP
H-ESP
Figure 3-8. Municipal waste combustor design classes for 1995.
C-ESP = Cold-sided electrostatic precipitator (temperature at control device is <230°C)
DS/CI/FF = Dry scrubber with carbon injection and fabric filter
DS/FF = Dry scrubber combined with a fabric filter
DSI/FF = Dry sorbent injection coupled with a fabric filter
EGB = Electro gravel bed
H-ESP = Hot-sided electrostatic precipitator (temperature at control device is >230°C)
WS = Wet scrubber
Using Tables 3-1, 3-2, and 3-3, a number of comparisons can be made between the
reference years:
• The number of facilities stayed about the same (105 in 1987, 130 in 1995, and 104 in
2000), but the amount of MSW incinerated more than doubled from 1987 to 1995
(from 13.4 billion kg in 1987 to 29 billion kg in 1995); it remained constant from
1995 to 2000 (30.6 billion kg in 2000).
3-12
-------
Municipal Solid Waste Incinerator Design Classes for 2000
1
Mass Burn
1
-\ SD/FF/CI
-\ SD/FF/CI/SNCR
-\ SD/FF/SNCR
SD/ESP/CI
-| SD/FF/SNCR
-| SD°/FF/CI/SNCR
-| SD/FF
-| DSI/FF
-| ESP
H FF
1
Modular, Excess air
1
_l r~CD
H
H SD/FF -1
i
H DSI/FF U WS/FSP
1
1
H K
flodular, Starved air
i
-| DSI/FF
-1 ESP
1
H FF
i
-1 SD/FF/CI
U WS
1
Refuse-Derived Fuel
DSI/FF
SD/ESP
SD/FF
SD/FF/CI/SNCR
DSI/FF/hhO/SNCR
SD/ESP/FF/CI
SD/FF7SNCR
SD/FF/SNCR
ESP
Figure 3-9. Municipal waste combustor design classes for 2000.
aAlso equipped with furnace dry sorbent injection system.
bAlso equipped with flue gas cooling (280-290°F).
"Also equipped with compact hybrid paniculate collector system.
CI = Carbon injection
DSI = Dry sorbent injection
ESP = Electrostatic precipitator
FF = Fabric filter
H2O = Water scrubber
SD = Spray dryer
SNCR = Selective noncatalytic reduction
WS = Wet scrubber
• In terms of number of facilities, the dominant furnace technology shifted from
modular in 1987 (49 units and 1.1 billion kg) to MB-WW facilities in 1995 (57 units
and 16.8 billion kg) and 2000 (140 units and 19 billion kg).
• The dominant APCD technology shifted from hot-sided ESPs in 1987 (46 units and
11 billion kg) to SDs with FFs (SDs/FFs) in 1995 (41 units and 12.8 billion kg) and
SDs/FFs with CI and selective noncatalytic reduction (88 units and 4.6 billion kg),
and SDs/FFs (27 units and 3.1 billion kg) in 2000.
• The use of hot-sided ESPs dropped from 46 facilities in 1987 (11 billion kg) to 12
facilities in 1995 (1.3 billion kg). No hot-sided ESPs were used in 2000.
• The number of uncontrolled facilities dropped from 38 in 1987 (0.6 billion kg) to 10
in 1995 (0.2 billion kg) and 6 in 2000 (0.08 billion kg).
3-13
-------
Table 3-1. Inventory of municipal waste combustors (MWCs) in 2000 by technology, air pollution control
device (APCD), size", and annual activity level (kg/yr)
APCD"
DSI/FFC
DSI/ESP
DSI/FF/
H20/SNCR
ESP
FF
WS
WS/ESP
SDd/FF/CI/
SNCR
SD/ESP
SD/ESP/CI
SD/ESP/
CI/SNCR
SD/ESP/
FFVCI
SD/FF
SD/FFV
SNCR
SD/FF/CI
Unc
TOTAL
MWC type
MB/RC
Size
(N)
S(4)
S(4)
L(3)
L(6)
L(9)
S(8)
Activity
level
2.78e+08
2.61e+08
3.97e+08
l.lle+09
1.51e+09
4.94e+08
MB
Size
(N)
S(2)
S(2)
Activity
level
6.27e+07
6.27e+07
MB/WW/RC
Size
(N)
S(l)
S(l)
Activity
level
1.76e+07
1.76e+07
MB/REF
Size
(N)
S(2)
S(2)
Activity
level
1.25e+08
1.25e+08
MB/WW
Size
(N)
S(2)
S(6)
1475)
L(4)
L(15)
S(4)
1413)
S(2)
145)
S(2)
L(124)
S(16)
Activity
level
6.27e+07
2.79e+08
1.34e+10
3.74e+08
2.79e+09
1.25e+08
2.99e+09
1.25e+08
9.51e+08
3.14e+07
2.05e+10
6.24e+08
MOD/EA
Size
(N)
S(3)
S(6)
S(3)
S(3)
S(15)
Activity
level
1.13e+08
9.22e+07
1.13e+08
1.32e+08
4.50e+08
MOD/SA
Size
(N)
S(3)
S(7)
S(8)
S(2)
S(4)
S(4)
S(4)
S(31)
Activity
level
4.7e+07
1.49e+08
1.14e+08
3. 14e+07
5.46e+07
6.27e+07
5.17e+07
5.10e+08
FB/RDF
Size
(N)
S(2)
S(2)
Activity
level
9.41e+07
9.41e+07
RDF
Size
(N)
142)
S(2)
144)
144)
U2)
1412)
US)
1434)
S(2)
Activity
level
4.44e+07
1.56e+08
6.68e+08
1.08e+09
7.41e+08
1.69e+09
1.57e+09
5.97e+09
1.56e+08
TOTAL
Size
(N)
S(14)
S(9)
U2)
S(26)
S(3)
S(4)
S(3)
L(82)
L(4)
L(4)
L(15)
U2)
S(9)
L(18)
L(21)
S(6)
US)
S(6)
L(167)
S(79)
Activity
level
5.95e+08
2.75e+08
4.44e+07
9.10e+08
4.90e+07
5.46e+07
1.13e+08
1.45e+10
1.08e+06
3.74e+08
2.79e+09
7.41e+08
3.20e+08
2.80e+09
4.56e+09
1.88e+08
9.51e+08
8.31e+07
2.80e+10
2.60e+09
-------
Table 3-1. Inventory of municipal waste combustors (MWCs) in 2000 by technology, air pollution control
device (APCD), size", and annual activity level (kg/yr) (continued)
Tor size, S = small; L = large.
bSlash(es) indicates devices used in conjunction.
°Also equipped with flue gas cooling (138 to 143°C).
dAlso equipped with furnace dry sorbent injection system.
eAlso equipped with compact hybrid paniculate collector system.
APCD:
CI = Carbon injection
DSI = Dry sorbent injection
ESP = Electrostatic precipitator
FF = Fabric filter
jj 2O = Water scrubber
SD = Spray dryer
SNCR = Selective noncatalytic reduction
Unc = Uncontrolled
WS = Wet scrubber
MWCtype:
FB/RDF = Fluidized-bed refuse-derived fuel
MB = Mass burn
MB/RC = Mass burn rotary kiln
MB/REF = Mass burn refractory walled
MB/WW/RC = Mass burn waterwalled/refractory walled
MOD/EA = Modular excess air
RDF = Refuse-derived fuel
-------
Table 3-2. Inventory of municipal waste combustors (MWCs) in 1995 by technology, air pollution control device
(APCD), and annual activity level (kg/yr)
APCD"
Unc
H-ESP
C-ESP
DSI/H-ESP
DS/FF
DS/CI/FF
DS/FF/C-ESP
WS/FF
WS/C-ESP
DS/C-ESP
DS/DSI/C-ESP
DSI/CI/H-ESP
DSI/C-ESP
DSI/FF
DSI/EGB
WS
TOTAL
MWC type
MB/RC
N
2
2
6
2
12
Activity
level
2.00e+08
1.14e+09
5.07e+08
2.59e+08
2.10e+09
MB/REF
N
1
2
1
1
2
7
Activity level
1.69e+08
2.68e+08
4.22e+08
1.13e+08
2.04e+08
1.18e+09
MBAVW
N
6
8
1
28
3
8
1
2
57
Activity level
1.04e+09
2.81e+09
4.22e+08
8.57e+09
1.17e+09
2.31e+09
2.75e+08
1.97e+08
1.68e+10
FB/RDF
N
1
1
1
3
Activity level
1.69e+08
8.45e+07
1.13e+08
3.67e+08
RDF/ded
N
1
4
1
7
1
4
1
19
Activity level
4.22e+07
1.81e+09
2.00e+08
2.51e+09
5.63e+08
1.75e+09
4.22e+08
7.30e+09
MOD/SA
N
9
4
4
1
1
1
3
23
Activity level
1.87e+08
1.82e+08
1.25e+08
2.82e+07
7.60e+07
3.42e+07
4.90e+07
6.81e+07
MOD/EA
N
1
1
3
1
1
1
1
9
Activity level
1.41e+07
1.97e+07
8.28e+07
1.41e+07
1.18e+08
6.76e+07
l.Ole+08
4.17e+08
TOTAL
N
10
12
22
3
41
3
1
1
1
13
1
1
6
9
1
5
130
Activity
level
2.01e+08
1.28e+09
5.20e+09
6.36e+08
1.28e+10
1.17e+09
5.63e+08
2.82e+07
6.76e+07
4.48e+09
7.60e+07
2.75e+08
5.07e+08
1.21e+09
1.13e+08
6.94e+08
2.93e+10
-------
Table 3-2. Inventory of municipal waste combustors (MWCs) in 1995 by technology, air pollution control device
(APCD), and annual activity level (kg/yr) (continued)
aSlash(es) indicates devices used in conjunction.
APCD:
C-ESP = Cold-sided electrostatic precipitator
CI = Carbon injection
DS = Dry scrubber
DSI = Dry sorbent injection
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator
SD = Spray dryer
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
FB/RDF = Fluidized-bed refuse-derived fuel
MB/RC = Mass burn rotary kiln
MB/REF = Mass burn refractory walled
MB/WW = Mass burn waterwalled
MOD/EA = Modular excess air
MOD/SA = Modular starved air
RDF/ded = Refuse-derived fuel/dedicated
-------
Table 3-3. Inventory of municipal waste combustors (MWCs) in 1987 by technology, air pollution control
device (APCD), and annual activity level (kg/yr)
APCD'
Unc
H-ESP
DS/FF
FF
EGB
WS
TOTAL
MWC type
MB/RC
N
3
1
4
Activity
level
3.94e+08
1.58e+07
4.10e+08
MB/REF
N
12
1
7
20
Activity level
2.00e+09
1.41e+07
9.01e+08
2.92e+09
MB/WW
N
19
1
20
Activity
level
5.20e+09
1.55e+08
5.36e+09
RDF/ded
N
7
2
9
Activity
level
3.01e+09
3.38e+08
3.35e+09
RDF/coflred
N
3
3
\ctivity level
2.53e+08
2.53e+08
MOD/SA
N
36
2
3
4
45
\ctivity level
5.73e+08
1.17e+08
1.43e+08
5.30e+07
8.9e+08
MOD/EA
N
2
1
1
4
\ctivity level
4.17e+07
6.76e+07
1.27e+08
2.36e+08
TOTAL
N
38
46
2
4
1
14
105
Activity
level
6.15e+08
1.10e+10
1.69e+08
1.59e+08
6.76e+07
1.42e+09
1.34e+10
oo
aSlash indicates devices used in conjunction.
APCD:
DS = Dry scrubber
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
MB/RC = Mass burn rotary kiln
MB/REF = Mass burn refractory walled
MB/WW = Mass burn waterwalled
MOD/EA = Modular excess air
MOD/SA = Modular starved air
RDF/cofired = Refuse-derived fuel/cofired
RDF/ded = Refuse-derived fuel/dedicated
-------
3.1.3. Estimation of CDD/CDF Emissions from MWCs
Compared with other CDD/CDF source categories, MWCs have been more extensively
evaluated for CDD/CDF emissions. In 2000, due to new regulations, EPA's Office of Air
Quality Planning and Standards (OAQPS) obtained emission test reports for all large MWCs.
3.1.3.1. Estimating CDD/CDF Emissions from MWCs in Reference Year 2000
EPA regulates dioxin emissions from MWCs on the basis of whether a facility is large or
small (see Section 3.1.7). Large facilities are those with a total design combustion capacity >250
tons/day; small facilities are those with a combustion capacity <250 tons/day). Combustion
capacity is determined on the basis of a single incineration unit. Facilities may comprise more
than one incinerator unit. In 2000, all large MWC facilities were source tested for stack
emissions of dioxin, as required by EPA regulations. Therefore, actual emissions from large
facilities were used to represent facility-specific dioxin emissions in 2000. More than 70% of the
small MWCs have been tested for dioxin emissions, and the available data were used to represent
facility-specific dioxin emissions in 2000. For small MWCs that were not source tested, dioxin
emissions were estimated on the basis of emission factors.
Using the test reports, concentrations and emissions were calculated for each of the 17
named dioxin/furan congeners and the remainder of the congener groups (homologues), making
up total dioxin/furan emissions (for 27 congeners/groups) for each of the MWC units
(Memorandum dated March 27, 2003, from Jason Huckaby, ERG, to Walt Stevenson, U.S.
EPA). The calculations were based on the individual CDD/CDF congener/group concentrations
for the MWC, the flue gas flow rate and MWC steam generation rate during the test, and the
annual steam generation at the MWC. Table 3-4 presents the average CDD/CDF congener-
specific stack gas concentrations used to derive the emission factors for the nontested facilities.
Table 3-4 shows concentrations for three detection limit (DL) assumptions: (1) a value of zero
for concentrations below the DL, (2) a value of one-half the DL for concentrations below theDL,
and (3) a value of the DL for concentrations below the DL.
Table 3-5 shows facility-specific dioxin emissions for all MWCs operating in 2000
(because of its size, Table 3-5 is placed at the end of this chapter). Note that the facilities are
divided into large and small MWCs. In total, 83.8 g TEQDF-WHO98 (76.3 g I-TEQ) were emitted
from all 104 MWCs in 2000. Of this total amount, 13.8 g TEQDF-WHO98 (12.7 g I-TEQ) were
emitted from large MWCs and 69.9 g TEQDF-WHO98 (63.6 g I-TEQ) were emitted from small
MWCs. Although 91% of the MSW combusted in 2000 was incinerated in large MWCs, the
large MWCs accounted for only 17% of total dioxin emissions.
3-19
-------
Table 3-4. National average CDD/CDF congener concentrations for
large municipal waste combustors (ng/dscm @ 7% O2)
Congener
TrCDD
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
Other TCDDs
Other PeCDDs
Other HxCDDs
Other HpCDDs
TrCDF
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Other TCDF
Other PeCDF
Other HxCDF
Other HpCDF
TOTAL
Nondetect set to zero3
0.031
0.005
0.016
0.016
0.037
0.032
0.219
0.345
0.232
0.323
0.494
0.220
0.037
0.072
0.050
0.069
0.082
0.059
0.013
0.066
0.156
0.024
0.090
1.081
0.747
0.326
0.079
4.92
Nondetect set to
l/2 detection limit3
0.031
0.006
0.016
0.016
0.036
0.032
0.219
0.345
0.239
0.334
0.502
0.220
0.037
0.072
0.051
0.069
0.083
0.059
0.013
0.066
0.157
0.024
0.092
1.083
0.758
0.329
0.079
4.97
Nondetect set to
detection limit3
0.031
0.006
0.017
0.016
0.037
0.032
0.219
0.345
0.246
0.344
0.510
0.220
0.037
0.073
0.052
0.069
0.083
0.060
0.014
0.067
0.159
0.024
0.094
1.085
0.770
0.332
0.079
5.02
aValues incorporating use of the detection limit when the laboratory report indicated "not detected" for individual
CDD/CDF congeners.
Source: Memorandum dated March 27, 2003, from Jason Huckaby, ERG, to Walt Stevenson, U.S. EPA.
3-20
-------
3.1.3.2. Estimating CDD/CDF Emissions from MWCs in Reference Years 1995 and 1987
Within the context of this report, adequate emissions testing for CDDs/CDFs was
available for 11 of the 113 facilities in the 1987 inventory and 27 of the 130 facilities in the 1995
inventory. Nationwide CDD/CDF air emissions from MWCs for reference years 1987 and 1995
were estimated using the three-step process described below.
Step 1. Estimation of emissions from all stack-tested facilities. The EPA stack testing
method (EPA Method 23) produces a measurement of CDDs/CDFs in units of mass
concentration of CDD/CDF (nanograms per dry standard cubic meter [ng/dscm] of combustion
gas) at standard temperature and pressure (20°C and 1 atmosphere [atm]) and adjusted to a
measurement of 7% oxygen in the flue gas (U.S. EPA, 1995a). This concentration is assumed to
represent conditions at the point of release from the stack into the air. Equation 3-1 was used to
derive annual emission estimates for each tested facility:
C x V x CF x H
ETEQ= (3-1)
W9ng/g
where:
E^ = annual TEQ emissions (g/yr)
C = combustion flue gas TEQ concentration (ng/dscm) (20°C, 1 atm; adjusted to 7% O2)
V = volumetric flow rate of combustion flue gas (dscm/hr) (20°C, 1 atm; adjusted to
7% O2)
CF = capacity factor; fraction of time that the MWC operates (0.85)
H = total hours in a year (8,760 hr)
After calculating annual emissions for each tested facility, the emissions were summed
across all tested facilities for each reference year. (Many of the emission tests do not correspond
exactly to these two years. In these cases, the equipment conditions present at the time of the test
were compared with those during the reference year to determine their applicability.)
Step 2. Estimation of emissions from all nonstack-testedfacilities. This step involves
multiplying the emission factor and the annual activity level for each MWC design class and then
summing across classes. The activity levels for reference years 1995 and 1987 are summarized
in Tables 3-2 and 3-3, respectively. The emission factors were derived by averaging the emission
factors across each tested facility in a design class. The emission factor for each facility was
calculated using the following equation:
3-21
-------
C xFv
EFMWC = (3-2)
where:
EFMWC = emission factor; average ng TEQ/kg of waste burned
C = TEQ or CDD/CDF concentration in flue gases (ng TEQ/dscm) (20°C,
1 atm; adjusted to 7% O2)
Fv = volumetric flue gas flow rate (dscm/hr) (20°C, 1 atm; adjusted to 7% O2)
Iw = average waste incineration rate (kg/hr)
Using an MB-WW MWC equipped with a cold-sided ESP as an example, and given:
C = 10 ng TEQ/dscm (20°C, 1 atm; adjusted to 7% O2)
Fv = 40,000 dscm/hr (20°C, 1 atm; adjusted to 7% O2)
Iw = 10,000 kg MSW/hr, then
10 ng 40,000 dscm hr
EFMB_WW = x x
dscm hr 10,000 kg
40 ng TEQ
kg MSW burned
EPA was not able to obtain engineering test reports of CDD/CDF emissions for a number
of design classes. In these cases, the above procedure could not be used to derive emission
factors. Instead, the emission factors of the tested design class that was judged most similar in
terms of dioxin control was assumed to apply to the untested class. The following logic was used
to make this decision:
1. The tested APCDs for the furnace type of the untested class were reviewed to see
whether any operated at a similar temperature.
2. If any operated at similar temperatures, the one with the most similar technology was
assumed to apply.
3-22
-------
3. If none operated at a similar temperature, then the most similar furnace type with the
same control device was assumed to apply.
Table 3-6 lists all design categories with no tested facilities and shows the class with
tested facilities that was judged to be most similar. The emission factors for each design class
are the same for both reference years because the emission factor is determined only by the
design and operating conditions and is independent of the year of the test.
Step 3. Summation of emissions from tested and untested facilities. This step involves
summing emissions from all tested and untested facilities. This process is shown in Tables 3-7a
and 3-7b and 3-8a and 3-8b for the reference years 1995 and 1987, respectively. The tables are
organized by design class and show the emission estimates for the tested and untested facilities
separately. The calculation of emissions from untested facilities is broken out to show the
activity level and emission factor for each design class.
3.1.4. Summary of CDD/CDF (TEQ) Emissions from MWCs for 2000,1995, and 1987
The activity level estimates (i.e., the amount of MSW that is annually combusted by the
various municipal waste combustion technologies) are given a high confidence rating for 1987
(13.4 billion kg of waste), 1995 (29.3 billion kg of waste), and 2000 (30.6 billion kg of waste).
For all three years, independent sources conducted comprehensive surveys of activity levels for
virtually all facilities (U.S. EPA, 1987b; Solid Waste Technologies, 1994; SAIC, 1994; Taylor
and Zannes, 1996; Memorandum dated March 27, 2003, from Jason Huckaby, ERG, to Walt
Stevenson, U.S. EPA).
The emission factor estimates are given a high confidence rating for 2000 and a medium
confidence rating for both 1995 and 1987. A large number of MWC facilities were tested in
2000, whereas a moderate fraction of the facilities were tested in 1995 and 1987: 27 of 130
facilities (21%) in 1995 and 11 of 104 facilities in 1987 (10%). The tested facilities represented
27 and 21% of the total activity level of operating MWCs in 1995 and 1987, respectively. These
tests represent most of the design categories identified in this report. The emission factors were
developed from emission tests that followed standard EPA protocols, used strict QA/QC
procedures, and were well documented in engineering reports. Because all tests were conducted
under normal operating conditions, some uncertainty exists about the magnitude of emissions
that may have occurred during other conditions (e.g., upset conditions, start up, and shut down).
In summary, this report estimates the following dioxin releases to air from MWCs operating in
2000, 1995 and 1987.
3-23
-------
Table 3-6. CDD/CDF TEQ emission factors (ng TEQ/kg waste) for
municipal solid waste incineration
Incinerator
design
MOD/SA
MOD/EA
FB/RDF
Air pollution
control device
(APCD)a
C-ESP
DS/DSI/
C-ESP
DSI/FF
FF
H-ESP
UNC
ws
WS/FF
C-ESP
DS/FF
DSI/FF
DSI/H-ESP
EGB
H-ESP
Unc
WS
WS/C-ESP
DS/FF
DSI/EGB
DSI/FF
Average I-
TEQDF
emission factor
16.2
16.2
0.025
16.2
79
0.025
16.2
16.2
16.2
16.2
0.025
118
0.025
118
0.025
16.2
16.2
0.63
0.63
0.63
Average
TEQDF-WH098
emission factor
17
17
0.024
17
85.7
0.024
17
17
17
17
0.024
119
0.024
119
0.024
17
17
0.72
0.72
0.72
Basis and rationale
Based on MOD/EA; C-ESP, similar furnace
(modular design) and same APCD
Based on MOD/EA; C-ESP, similar furnace
(modular design) and similar emission control
Based on direct tests
Based on MOD/EA; C-ESP, similar furnace
(modular design) and similar emission control
Based on direct tests
Based on MOD/SA; DSI/FF, same furnace and
most similar expected emissions
Based on MOD/EA; C-ESP, similar furnace
(modular design) and similar APCD
temperature
Based on MOD/EA; C-ESP, similar furnace
(modular design) and similar APCD
temperature
Based on direct tests
Based on MOD/EA; C-ESP, same furnace and
similar temperature in APCD; may
overestimate emissions
Based on MOD/SA; DSI/FF, similar (modular
design) furnace and same APCD
Based on MOD/EA; H-ESP, same furnace and
similar emissions
Based on MOD/SA; DSI/FF, same furnace and
most similar expected emissions
Based on direct tests
Based on MOD/SA; DSI/FF, same furnace and
most similar expected emissions
Based on MOD/EA; C-ESP, same furnace and
similar APCD temperature
Based on MOD/EA; C-ESP, same furnace and
similar APCD
Based on MB/WW; DS/FF similar furnace and
same APCD
Based on MB/WW; DS/FF similar furnace;
may underestimate emissions
Based on MB/WW; DS/FF similar furnace;
may underestimate emissions
aSlash indicates devices used in conjunction.
APCD:
C-ESP = Cold-sided electrostatic precipitator
DS = Dry scrubber
DSI = Dry sorbent injection
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
FB/RDF = Fluidized-bed refuse-derived fuel
MB/WW = Mass burn waterwalled
MOD/EA = Modular excess air
MOD/SA = Modular starved air
3-24
-------
Table 3-7a. Annual I-TEQDF emissions from municipal waste combustors
(MWCs) operating in 1995
MWC
type
MB/WW
MB/REF
MB/RC
RDF/ded
MOD/SA
Air pollution
control device
(APCD)a
C-ESP
DS/C-ESP
DS/CI/FF
DS/FF
DSI/CI/H-ESP
DSI/FF
DSI/H-ESP
H-ESP
Subtotal
C-ESP
DS/C-ESP
DS/FF
DSI/FF
WS
Subtotal
C-ESP
DS/FF
DSI/C-ESP
DSI/FF
Subtotal
C-ESP
DS/C-ESP
DS/FF
DSI/FF
DSI/H-ESP
H-ESP
DS/FF/C-ESP
Subtotal
C-ESP
DSI/FF
H-ESP
Unc
WS
WS/FF
DS/DSI/C-ESP
Subtotal
I-TEQDF
emissions from
tested facilities
(gTEQ/yr)
0
2.1
0.64
2
2.1
0.3
0
163
170.1
39.8
21.6
0
0
0
61.4
0
0.25
0
5.3
5.6
32.5
0.3
0.1
0
0
0
0
32.9
0
0
8
0
0
0
0
8
Average
I-TEQDF
emission factor
(ng/kg)
6.1
6.1
1.5
0.63
-
-
7.74
473
—
-
0.63
1.91
236
47.0
0.65
47.0
47.0
231
0.53
0.24
231
231
1,492
0.24
16.2
-
79
0.025
16.2
16.2
16.2
Activity
level
nontested
facilities
(kg/yr)
2.81e+09
1.88e+09
7.44e+08
5.98e+09
0
0
4.22e+08
1.79e+08
1.20e+10
0
0
2.68e+08
1.13e+08
2.04e+08
5.85e+08
2.00e+08
7.57e+08
5.07e+08
1.46e+08
1.61e+09
1.67e+09
1.14e+09
1.58e+09
4.22e+08
2.00e+08
4.22e+07
5.63e+08
5.62e+09
1.25e+08
0
8.03e+07
1.87e+08
4.90e+07
2.82e+07
7.60e+07
5.46e+08
I-TEQDF
emissions
from
nontested
facilities
(gTEQ/yr)
17.1
11.5
1.1
3.8
0.0
0.0
3.3
84.7
121.4
0
0
0.2
0.2
48.1
48.5
9.4
0.5
23.8
6.9
40.6
385.8
0.6
0.4
97.5
46.2
63.0
0.1
593.5
2.0
0.0
6.3
0.0
0.8
0.5
1.2
10.9
Total
I-TEQDF
emissions
from all
facilities
(gTEQ/yr)
17.1
13.6
1.7
5.8
2.1
0.3
o o
J.J
247.6
291.5
39.8
21.6
0.2
0.2
48.1
109.9
9.4
0.8
23.8
12.2
46.2
418.3
0.9
0.5
97.5
46.2
63
0.1
626.4
2
0
14.3
0.005
0.8
0.5
1.2
18.9
3-25
-------
Table 3-7a. Annual I-TEQDF emissions from municipal waste combustors
(MWCs) operating in 1995 (continued)
MWC
type
MOD/EA
FB/RDF
TOTAL
Air pollution
control device
(APCD)a
C-ESP
DS/FF
DSI/FF
DSI/H-ESP
H-ESP
Unc
WS/C-ESP
Subtotal
DS/FF
DSI/EGB
DSI/FF
Subtotal
I-TEQDF
emissions from
tested facilities
(g TEQ/yr)
0.1
0
0
0
2.3
0
0
2.4
0
0
0
0
280.4
Average
I-TEQDF
emission factor
(ng/kg)
16.2
16.2
0.025
118
b
0.025
16.2
0.63
0.63
0.63
Activity
level
nontested
facilities
(kg/yr)
6.25e+07
1.18e+08
l.Ole+08
1.41e+07
0
1.41e+07
6.76e+07
3.77e+08
1.69e+08
1.13e+08
8.45e+07
3.67e+08
2.11e+10
I-TEQDF
emissions
from
nontested
facilities
(g TEQ/yr)
1
1.9
0.003
1.7
0
0.0003
1.1
5.7
0.1
0.1
0.1
0.3
820.9
Total
I-TEQDF
emissions
from all
facilities
(g TEQ/yr)
1.1
1.9
0.003
1.7
2.3
0.0003
1.1
8.1
0.1
0.1
0.1
0.3
1,101.3
aSlash indicates devices used in conjunction.
bValue could not be calculated.
- = Emissions not developed
APCD:
C-ESP = Cold-sided electrostatic precipitator
CI = Carbon injection
DS = Dry scrubber
DSI = Dry sorbent injection
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
FB/RDF = Fluidized-bed refuse-derived fuel
MB/RC = Mass burn rotary kiln
MB/REF = Mass burn refractory walled
MB/WW = Mass burn waterwalled
MOD/EA = Modular excess air
MOD/SA = Modular starved air
RDF/ded = Refuse-derived fuel/dedicated
3-26
-------
Table 3-7b. Annual TEQDF-WHO98 emissions from municipal waste
combustors (MWCs) operating in 1995
MWC
type
MB/WW
MB/REF
MB/RC
RDF/ded
MOD/SA
Air pollution
control device
(APCD)a
C-ESP
DS/C-ESP
DS/CI/FF
DS/FF
DSI/CI/H-ESP
DSI/FF
DSI/H-ESP
H-ESP
Subtotal
C-ESP
DS/C-ESP
DS/FF
DSI/FF
WS
Subtotal
C-ESP
DS/FF
DSI/C-ESP
DSI/FF
Subtotal
C-ESP
DS/C-ESP
DS/FF
DSI/FF
DSI/H-ESP
H-ESP
DS/FF/C-ESP
Subtotal
C-ESP
DSI/FF
H-ESP
Unc
WS
WS/FF
DS/DSI/C-ESP
Subtotal
TEQDF-
WHO98
emissions
from tested
facilities
(gTEQ/yr)
0
2.24
0.68
2.1
2.26
0.3
0
183
190.6
43
22.5
0
0
0
65.5
0
0.265
0
10.5
10.8
35.6
0.34
0.1
0
0
0
0
36
0
0.0008
8.69
0
0
0
0
8.7
Average
TEQDF-
WH098
emission
factor
(ng/kg)
6.54
6.54
1.61
0.72
-
-
8.22
535.0
—
-
0.72
2.07
254.0
93.1
0.68
93.1
93.1
253.0
0.56
0.26
253.0
253.0
1,679.0
253.0
17.0
-
85.7
0.024
17.0
17.0
17.0
Activity level
nontested
facilities
(kg/yr)
2.81e+09
1.88e+09
7.44e+08
5.98e+09
0.0
0.0
4.22e+08
1.79e+08
1.20e+10
0.0
0.0
2.68e+08
1.13e+08
2.04e+08
5.85e+08
2.00e+08
7.57e+08
5.07e+08
1.46e+08
1.61e+09
1.67e+09
1.14e+09
1.58e+09
4.22e+08
2.00e+08
4.22e+07
5.63e+08
5.62e+09
1.25e+08
0.0
8.03e+07
1.87e+08
4.90e+07
2.82e+07
7.60e+07
5.46e+08
TEQDF-
WH098
emissions
from
nontested
facilities
(gTEQ/yr)
18.4
12.3
1.2
4.3
0.0
0.0
3.5
95.8
135.4
0.0
0.0
0.2
0.2
51.8
52.2
18.6
0.5
47.2
13.6
79.9
422.5
0.6
0.4
106.8
50.6
70.9
142.4
794.2
2.1
0.0
6.9
0.0
0.8
0.5
1.3
11.6
Total TEQDF-
WHO98
emissions
from all
facilities
(gTEQ/yr)
18.4
14.54
1.88
6.4
2.26
0.3
3.5
278.8
326.0
43.0
22.5
0.2
0.2
51.8
117.7
18.6
0.8
47.2
24.1
90.7
458.1
1.0
0.5
106.8
50.6
70.9
142.4
830.2
2.1
0.001
15.6
0.005
0.8
0.5
1.3
20.3
3-27
-------
Table 3-7b. Annual TEQDF-WHO98 emissions from municipal waste
combustors (MWCs) operating in 1995 (continued)
MWC
type
MOD/EA
FB/RDF
TOTAL
Air pollution
control device
(APCD)a
C-ESP
DS/FF
DSI/FF
DSI/H-ESP
H-ESP
Unc
WS/C-ESP
Subtotal
DS/FF
DSI/EGB
DSI/FF
Subtotal
TEQDF-
WHO98
emissions
from tested
facilities
(gTEQ/yr)
0.068
0
0
0
2.35
0
0
2.4
0
0
0
0
314
Average
TEQDF-
WH098
emission
factor
(ng/kg)
17.0
17.0
0.024
119.0
—
0.024
17.0
0.72
0.72
0.72
Activity level
nontested
facilities
(kg/yr)
6.25e+07
1.18e+08
l.Ole+08
1.41e+07
0.0
1.41e+07
6.76e+07
3.77e+08
1.69e+08
1.13e+08
8.45e+07
3.67e+08
2.11e+10
TEQDF-
WH098
emissions
from
nontested
facilities
(gTEQ/yr)
1.06
2.01
0.002
1.68
0.0
0.003
1.15
5.9
0.114
0.076
0.057
0.3
1,079.5
Total TEQDF-
WHO98
emissions
from all
facilities
(gTEQ/yr)
1.1
2.0
0.002
1.7
2.4
0.003
1.2
8.3
0.1
0.1
0.1
0.3
1,393.5
aSlash indicates devices used in conjunction.
- = Emissions not developed
APCD:
C-ESP = Cold-sided electrostatic precipitator
CI = Carbon injection
DS = Dry scrubber
DSI = Dry sorbent injection
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
FB/RDF = Fluidized-bed refuse-derived fuel
MB/RC = Mass burn rotary kiln
MB/REF = Mass burn refractory walled
MB/WW = Mass burn waterwalled
MOD/EA = Modular excess air
MOD/SA = Modular starved air
RDF/ded = Refuse-derived fuel/dedicated
3-28
-------
Table 3-8a. Annual I-TEQDF emissions to the air from municipal waste
combustors (MWCs) operating in 1987
MWC type
MB/WW
MB/REF
MB/RC
RDF/ded
RDF/cofired
MOD/SA
MOD/EA
TOTAL
Air pollution
control
device
(APCD)a
DS/FF
H-ESP
Subtotal
DS/FF
H-ESP
WS
Subtotal
FF
H-ESP
Subtotal
H-ESP
WS
Subtotal
H-ESP
FF
H-ESP
Unc
WS
Subtotal
EGB
Unc
WS
Subtotal
I-TEQDF
emissions
from tested
facilities
(gTEQ/yr)
0.0373
433.0
433.0
0.0
0.0
0.0
0.0
0.0
48.2
48.2
840.0
0.0
840.0
0.0
0.0
0.0643
0.0
0.0
0.0643
0.0
0.0
0.0
0.0
1,321.3
Average
I-TEQDF
emission
factor
(ng/kg)
—
473.0
0.63
473.0
236.0
47.0
285.0
1,492.0
231.0
231.0
16.2
79.0
0.025
16.2
0.025
0.025
16.2
Activity level
nontested
facilities
(kg/yr)
1.55e+08
3.27e+09
3.43e+09
1.41e+08
2.00e+09
9.01e+08
3.04e+09
1.58e+07
2.25e+08
2.41e+08
2.45e+09
3.38e+08
2.79e+09
2.53e+08
1.43e+08
3.61e+08
5.73e+08
5.30e+07
1.13e+09
6.76e+07
4.17e+07
1.27e+08
2.36e+08
l.lle+10
I-TEQDF
emissions
from
nontested
facilities
(g TEQ/yr)
0.0
1,546.7
1,546.7
0.09
946.0
212.6
1,158.7
0.7
64.1
64.8
3,655.4
78.1
3,733.5
58.4
2.3
28.5
0.01
0.86
31.7
0.002
0.001
2.057
2.1
6,537.5
Total I-
TEQDF
emissions
from all
facilities
(g TEQ/yr)
0.0373
1,979.7
1,979.7
0.09
946.0
212.6
1,158.7
0.7
112.3
113.0
4,495.4
78.1
4,573.5
58.4
2.3
28.6
0.01
0.86
31.8
0.002
0.001
2.057
2.1
7,858.8
aSlash indicates devices used in conjunction.
APCD:
DS = Dry scrubber
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
MB/RC = Mass burn rotary kiln
MB/REF = Mass burn refractory walled
MB/WW = Mass burn waterwalled
MOD/EA = Modular excess air
MOD/SA = Modular starved air
RDF/cofired = Refuse-derived fuel/cofired
RDF/ded = Refuse-derived fuel/dedicated
3-29
-------
Table 3-8b. Annual TEQDF-WHO98 emissions to the air from municipal
waste combustors (MWCs) operating in 1987
MWC type
MB/WW
MB/REF
MB/RC
RDF/ded
RDF/cofired
MOD/SA
MOD/EA
TOTAL
Air pollution
control device
(APCD)a
DS/FF
H-ESP
Subtotal
DS/FF
H-ESP
WS
Subtotal
FF
H-ESP
Subtotal
H-ESP
WS
Subtotal
H-ESP
FF
H-ESP
Unc
WS
Subtotal
EGB
Unc
WS
Subtotal
TEQDF-
WH098
emissions
from tested
facilities
(g TEQ/yr)
0.039
485.0
485.0
0.0
0.0
0.0
0.0
0.0
53.4
53.4
946.0
0.0
946.0
0.0
0.0
0.068
0.0
0.0
0.068
0.0
0.0
0.0
0.0
1,484.5
Average
TEQDF-
WH098
emission
factor
(ng/kg)
535.0
0.72
535.0
254.0
93.1
316.0
1,679.0
253.0
253.0
17.0
85.7
0.024
17.0
0.024
0.024
17.0
Activity level
nontested
facilities
(kg/yr)
0
3.27e+09
1.41e+08
2.00e+09
9.01e+08
1.58e+07
2.25e+08
2.45e+09
3.38e+08
2.53e+08
1.43e+08
3.61e+08
5.73e+08
5.30e+07
6.76e+07
4.17e+07
1.27e+08
3.04e+09
TEQDF-
WHO98
emissions
from
nontested
facilities
(g TEQ/yr)
0.0
1,749.5
1,749.5
0.1
1,070.0
228.9
1,299.0
1.47
71.1
72.6
4,113.6
85.5
4,199.1
64.0
2.4
30.9
0.01
0.9
34.2
0.0016
0.001
2.16
2 2
7,420.6
Total TEQDF-
WH098
emissions
from all
facilities
(g TEQ/yr)
0.039
2,234.5
2,234.5
0.1
1,070.0
228.9
1,299.0
1.47
124.5
126.0
5,059.6
85.5
5,145.1
64.0
2.4
31.0
0.01
0.9
34.3
0.0016
0.001
2.16
2.2
8,905.1
aSlash indicates devices used in conjunction.
APCD:
DS = Dry scrubber
EGB = Electro gravel bed
FF = Fabric filter
H-ESP = Hot-sided electrostatic precipitator
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
MB/RC = Mass bum rotary kiln
MB/REF = Mass bum refractory walled
MB/WW = Mass bum waterwalled
MOD/EA = Modular excess air
MOD/SA = Modular starved air
RDF/cofired = Refuse-derived fuel/cofired
RDF/ded = Refuse-derived fuel/dedicated
3-30
-------
Reference year g TEQPF-WHOQg g I-TEQ Confidence rating
2000 83.8 76.3 A
1995 1,393.5 1,101.3 B
1987 8,905.1 7,858.8 B
As noted, a high confidence rating (A) is assigned to the estimate of dioxin emissions
from MWCs in 2000 because a large number of facilities were tested, providing a highly certain
and representative emissions estimate. Table 3-5 displays the CDD/CDF TEQ emissions by each
MWC facility operating in 2000. Moreover, the activity level of MWCs in 2000 is known and is
very representative of this source. A confidence rating B is assigned to the overall dioxin air
releases for years 1995 and 1987. Although the activity level for those years is known and is
considered very representative, fewer facilities were stack tested in those years relative to the
total number of MWCs in operation. Therefore, there was a reliance on the use of emission
factors to estimate releases in 1995 and 1987. However, estimates for these years are considered
representative of MWCs operating in those years.
3.1.5. Congener Profiles of Municipal Waste Combustion Facilities
The air emissions from MWCs contain a mixture of CDD and CDF congeners. These
mixtures can be translated into what are called "congener profiles," which represent the
distribution of total CDDs and total CDFs present in the mixture. A congener profile may serve
as a signature of the types of CDDs/CDFs associated with a particular MWC technology and
APCD. Figure 3-10 is a congener profile of an MB-WW MWC equipped with an SDSS and an
FF (the most common type of MWC and APCD design in use today). This congener profile
indicates that OCDD dominates CDD/CDF emissions and that every toxic CDD/CDF congener
is detected in the emissions. Figures 3-11 and 3-12 present 2,3,7,8-TCDD frequency distribution
and 1,2,3,7,8-PeCDD frequency distribution, respectively. According to a memorandum dated
March 27, 2003, from Jason Huckaby, ERG, to Walt Stevenson, U.S. EPA, the distribution of
these two congeners varies little from MWC to MWC. Although these two congeners represent
less than 1% of total dioxin/furan emissions, they contribute approximately 13 to 23% of the I-
TEQDF emissions, depending on which TEF system is used.
3.1.6. Estimated CDDs/CDFs in MWC Ash
Ash from MWCs is required to be disposed of in permitted landfills from which releases
to the general environment are controlled. For background purposes, however, some information
is presented below about the quantities of CDDs/CDFs in ash from MWCs.
3-31
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.02 0.04 0.06 0.08 0.1 0.12
to conOOPP
cacotvcoE-OQh
Ratio (congener group emission factor/total CDD/CDF emission factor)
0 0.08 0.1 0.15 0.2 0.25
Figure 3-10. Congener and congener group profiles for air
emissions from a mass burn waterwall municipal waste
combustor equipped with a dry scrubber and fabric filter.
3-32
-------
CO
do
CO
ND = 0 ND = 1/2 detection limit ND = full detection limit
Arithmetic average (ng/dscm @ 7%O2) 0.00545
Arithmetic standard deviation 0.01542
0.00578
0.01535
0.0061
0.0153
o
o
30
25
20
15
10
0-.5 .5-1 1-1.5 1.5-2 2-2.5 2.5-3 3-3.5 3.5-4 4-4.5 4.5-5 5-5.5 5.5-6 6-6.5 6.5-7 7-7.5 7.5-8 8-8.5 8.5-9 9-9.5 9.5-10
Bin
• ND = 0
• ND = 1/2 detection limit
D ND = full detection limit
Figure 3-11. 2,3,7,8-TCDD frequency distribution (negative natural log concentration).
-------
CO
do
ND = 0 ND = 1/2 detection limit ND = full detection limit
Arithmetic average (ng/dscm @ 7%O2) 0.01589 0.0163 0.01669
Arithmetic standard deviation 0.03375 0.03364 0.0336
35
30
25
•£ 20
3
O 15
10
5
i
m
•n •
0-.5 .5-1 1-1.5 1.5-2 2-2.5 2.5-3 3-3.5 3.5-4 4-4.5 4.5-5 5-5.5 5.5-6 6-6.5 6.5-7 7-7.5 7.5-8 8-8.5 8.5-9 9-9.5 9.5-10
Bin
• ND = 0
• ND = 1/2 detection limit
HI ND = full detection limit
Figure 3-12. 1,2,3,7,8-PeCDD frequency distribution (negative natural log concentration).
-------
An estimated 7 million metric tons of total ash (bottom ash plus fly ash) were generated by
MWCs in 1992 (telephone conversation between J. Loundsberry, U.S. EPA Office of Solid
Waste, and L. Brown, Versar, Inc., February 24, 1993). EPA indicated that 2 to 5 million metric
tons of total ash were produced annually in the late 1980s from MWCs, with fly ash comprising
5 to 15% of the total (U.S. EPA, 199 la).
EPA reported the results of analyses of MWC ash samples for CDDs/CDFs (U.S. EPA,
1990a). Ashes from five state-of-the-art facilities located in different regions of the United States
were analyzed for all 2,3,7,8-substituted CDDs/CDFs. The TEQ levels in the ash (fly ash mixed
with bottom ash) ranged from 106 to 466 ng I-TEQDF/kg, with a mean value of 258 ng I-
TEQDF/kg. CDD/CDF levels are generally much higher in fly ash than in bottom ash. For
example, Fiedler and Hutzinger (1992) reported levels of 13,000 ng I-TEQDF/kg in fly ash.
In another study (Washington State Department of Ecology, 1998), CDD/CDF congener
data were reported for ash and other solid residuals from three municipal incinerators (Fort
Lewis, Bellingham [municipal plus medical wastes], and Spokane). The data were compiled and
evaluated to determine a total I-TEQ concentration and loading. Nondetect values were included
as either zero or one-half the DL or at the DL. The results were as follows, assuming that
nondetect values were at zero concentration:
Location Type of Residual I-TEQ (|J.g/kg) I-TEQ (mg/day)
Ft. Lewis Bottom ash 0.0 0.0
Fly ash 4.98 0.76
Bellingham Mixed ash
(average of three tests) 0.038 1.14
Spokane Mixed ash 0.163 38.0
Fly ash 0.510 24.3
Bottom ash 0.0001 0.02
In Shane et al. (1990), ash from five municipal incinerators was analyzed for a number of
constituents, including CDDs (but not CDFs) and PCBs. For dioxins, three of the incinerators
were at nondetectable levels (DL of 1 (J-g/kg). The other two incinerators had detectable levels of
five CDD congener groups (no analyses were reported for individual congeners), and the
averages for the two units were 26, 59, 53, 25, and 12 jig/kg for TCDD, PeCDD, HxCDD,
HpCDD, and OCDD, respectively. These levels were much higher that those reported by EPA
(U.S. EPA, 1990a).
For PCBs, the five sets of ashes were analyzed for 10 congener groups. All groups were
detected for one of the incinerators. However, the other four incinerators contained little or no
3-35
-------
octa, nona, or deca congeners. The average PCB concentration (all congener groups) for the five
incinerators was 216 i-ig/kg, with a range of 99 to 322 i-ig/kg.
No generation rates of the ashes were given (Shane et al., 1990); therefore, the measured
concentrations cannot be readily converted to quantities of CDDs or PCBs. The ashes from each
of the five incinerators were disposed of in various ways. For two of the incinerators, the ash
was sent to metal recovery and also landfilled. For a third, the fly ash was sold. For a fourth, the
ashes were landfilled only. For the fifth, the ashes were used in road building and also landfilled.
For those incinerators with more than one ash disposition, no breakdown was given of how much
went to each location. Fifteen other incinerators were discussed in Shane et al. (1990). Thirteen
of them disposed of their ash exclusively in landfills, and the other two partially disposed of their
ash in landfills.
Table 7 in Clement et al. (1988) presents 13 data sets for CDD/CDF congener groups for
municipal incinerator ash. The average data for each congener group and the ranges of each
group are given in Table 3-9. No data were presented for individual congeners or for ash
quantities.
Table 3-9. Average and range of CDD/CDF congener groups in fly ash from
a municipal incinerator (|-ig/kg)
Congener group
TCDD
PeCDD
HxCDD
HpCDD
OCDD
TOTAL CDDs
TCDF
PeCDF
HxCDF
HpCDF
OCDF
TOTAL CDFs
Average concentration
3.7
6.4
9.1
2.3
1.5
23.0
12.0
17.0
14.0
2.9
1.2
47.1
Concentration range
1.6-12
2-25
1.5-42
0.5-9.2
0.2-6
6.2-94
5.1-36
8.3-40
3.9-40
0.8-9.2
ND-2.1
22-110
Source: Clement etal. (1988).
3-36
-------
Ash from three incinerators (one in North America, one in Europe, and one in Japan) had
mean CDD concentrations of 363, 588, and 2.6 i-ig/kg, respectively (Table 3-3 in U.S. EPA,
1987a). The values ranged from less than 0.5 to 3.537 i-ig/kg. For CDFs, the respective mean
concentrations for the first two incinerators were 923 and 288 i-ig/kg. Data for the third
incinerator were not reported. The CDF range for the two incinerators was from less than 0.5 to
1,770 i-ig/kg. No data were given for individual congeners or for quantities of ashes.
In Table 1 in Lahl et al. (1991), data are presented for concentrations of total CDDs and
total CDFs in the ash from an ESP from a municipal incinerator. Total CDDs were 140.46 |ig/kg
in the summer samples and 86 |ig/kg in the winter samples. Total CDFs were 54.97 |ig/kg in the
summer samples and 73.85 |ig/kg in the winter samples. No data were given for individual
congeners, nor was there information about the quantity of precipitator ash generated. It was
assumed that the data were not for TEQs.
A wire reclamation incinerator was reported to have 0.41 jig/kg of CDDs and 11.6 jig/kg
of CDFs in fly ash from its stack emissions (Table 3-11 in U.S. EPA, 1987a). For the same
incinerator, the furnace ash concentrations were reported as 0.58 |ig/kg CDDs and 0.73 |ig/kg
CDFs. Again, no data were given for individual congeners or for quantities of the ashes.
Data from the aforementioned sources are compiled in Table 3-10 of this document for
comparison purposes. Annual TEQ amounts were estimated by multiplying the mean TEQ total
ash concentration by the estimated amount of MWC ash generated annually (approximately 7
million metric tons in 1995 and 5 million metric tons in 1987). Where possible, ash quantities
were broken down into fly ash or bottom ash. Fly ash was assumed to be 10% of the total ash,
and bottom ash was assumed to be 90% of the total ash.
Imagawa and Lee (2001) analyzed samples collected from eight Japanese MSW
incinerators to determine dioxin levels in the fly ash (Table 3-11). Specific congener data were
not available, so TEQ calculations could not be performed.
Kobylecki et al. (2001) analyzed the reduction of dioxins in fly ash by pelletizing the ash
and reburning the pellets in a laboratory-scale bubbling fluidized-bed furnace. Fly ash for the
test input material was collected from a fly ash filter vessel during 4 days of MWC operation.
The concentrations of the dioxin collected and composited congeners are shown in Table 3-12.
The total TEQ value derived by Kobylecki was 862 ng I-TEQDF/kg of fly ash.
Sakai et al. (2001) analyzed the levels of dioxins and PCBs in fly ash and bottom ash
from a newly constructed MWC in Japan (Table 3-13). TEQ values derived from the data give a
total of 423 ng I-TEQDF/kg for fly ash and 10.5 ng I-TEQDF/kg for bottom ash for dioxins and
31.6 ng I-TEQDF/kg for fly ash and 0.85 ng I-TEQDF/kg for bottom ash for PCBs.
3-37
-------
Table 3-10. Comparison of the amount of TEQs generated annually in
municipal waste combustor ash
Data source
U.S. EPA (1990a)
Washington State
Department of Ecology
(1998)
Ft. Lewis
Bellingham
Spokane
Shane etal. (1990)
Clement etal. (1988)
U.S. EPA (1987a)
North America
Europe
Japan
Wire reclamation
Lahl etal. (1991)
Type of
ash
Mixed
Bottom
Fly
Mixed
Mixed
Fly
Bottom
Fly
Fly
Fly
Fly
Fly
Fly
Bottom
Mixed
Mean total
CDD/CDF
concentrati
on
(ng/kg)
12,383
0
71,280
1,884
1,414
10,320
100
175,000
70,000
1,286,000
876,000
2,600
12,010
1,310
177,640
Mean
I-TEQDF
(ng/kg)
258
0
4,980
38
163
510
0.1
-
-
-
-
-
-
-
Annual
TEQ
amount
1995 value3
(gl-
TEQDF/yr)
1,806
0
3,486
266
1,141
357
1
-
-
-
-
-
-
-
Annual TEQ
amount
1987 value3
(gl-
TEQDF/yr)
1,290
0
2,490
190
815
255
0.05
-
-
-
-
-
-
-
aln calculating the annual TEQ amounts, fly ash and bottom ash were considered to be 10% and 90% of the total ash,
respectively.
- = Value could not be calculated
Each of the five facilities sampled by EPA had companion ash disposal facilities
equipped with leachate collection systems or some means of collecting leachate samples (U.S.
EPA, 1990a). Leachate samples were collected and analyzed for each of these systems.
Detectable levels were found in the leachate at only one facility (3 ng I-TEQDF/L); the only
detectable congeners were HpCDDs, OCDD, and HpCDFs.
3-38
-------
Table 3-11. Concentration of CDD/CDF congener groups (ng/kg) in fly ash
samples from combustion of municipal solid waste in eight Japanese
incinerators"
Congener
group
TCDD
PeCDD
HxCDD
HpCDD
OCDD
TCDF
PeCDF
HxCDF
HpCDF
OCDF
Stoker incinerators
B
5,000
20,000
45,000
70,000
125,000
25,000
50,000
65,000
75,000
40,000
c
200,000
340,000
440,000
340,000
110,000
210,000
410,000
400,000
230,000
20,000
D
80,000
200,000
250,000
230,000
160,000
330,000
320,000
300,000
200,000
40,000
E
75,000
105,000
90,000
37,000
15,000
50,000
45,000
22,000
10,000
1,000
Fluidized-bed incinerators
F
6,000
10,000
12,000
8,000
7,000
13,000
14,000
21,000
17,000
10,000
I
10,000
28,000
41,000
40,000
25,000
18,000
32,000
34,000
33,000
13,000
J
10,000
37,000
100,000
200,000
187,000
50,000
125,000
210,000
225,000
150,000
L
5,000
10,000
30,000
40,000
50,000
70,000
120,000
200,000
270,000
120,000
a Incinerators are designated by letters because they remained anonymous.
Source: Imagawa and Lee (2001) (numbers estimated from Figure 2 of report).
Table 3-12. Concentration of CDD/CDF congener groups in fly ash samples
from municipal solid waste
Congener group
TCDD
PeCDD
HxCDD
HpCDD
OCDD
TCDF
PeCDF
HxCDF
HpCDF
OCDF
TOTAL
Concentration (ng/kg)
8,000
9,000
40,000
10,800
8,000
8,000
10,000
9,500
8,500
8,000
119,800
TEQ (ng/kg)
15
45
100
50
1
10
300
300
40
1
862
Source: Kobylecki et al. (2001) (estimated from values in Figure 4 of "Before Incineration").
3-39
-------
Table 3-13. CDD/CDF concentrations in municipal solid waste ash from a
newly constructed municipal waste combustor in Japan
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
TOTAL
Concentration (ng/kg)
Fly ash
19
78
92
210
130
1,300
2,800
150
290
320
310
310
21
400
1,100
110
320
7,960
Bottom ash
1.6
3.1
2.6
5.6
3.6
33.0
110.0
4.8
5.3
5.9
4.4
4.9
0.36
6.7
23.0
1.6
9.3
226.0
I-TEQs (ng/kg)
Fly ash
19.0
39.0
9.2
21.0
13.0
13.0
2.8
15.0
14.5
160.0
31.0
31.0
2.1
40.0
11.0
1.1
0.32
423.0
Bottom ash
1.6
1.65
0.26
0.56
0.36
0.33
0.11
0.48
0.265
2.95
0.44
0.49
0.036
0.67
0.23
0.016
0.0093
10.5
Source: Sakaietal. (2001).
3.1.7. Recent EPA Regulatory Activities
As part of the 1990 Clean Air Act mandates, EPA promulgated CDD/CDF emission
standards for all existing and new MWC units at facilities with aggregate combustion capacities
greater than 35 metric tons per day (Federal Register, 1995a). These standards, established under
Section 129 of the Clean Air Act, required facilities to use "maximum achievable control
technology" (MACT) at MWC units and emission control retrofit for large MWC units (units
with capacities greater than 225 metric tons per day) by December 2000. In response to a court
remand, the regulations were subsequently amended to remove small MWC units (units with
capacities ranging from 35 to 225 metric tons per day) (Federal Register, 1995a).
The specific emission standards for large MWCs (expressed as ng/dscm of total
CDD/CDF, based on standard dry gas corrected to 7% oxygen) are a function of the size, APCD
configuration, and age of the facility, as listed below.
3-40
-------
1995 emission standards
(ng total CDD/CDF/dscnO Facility age, size, and APCD
60 Existing; >225 metric tons/day; ESP-
based APCD
30 Existing; >225 metric tons/day; non-
ESP-based APCD
13 New; >225 metric tons/day
EPA reestablished emission standards for small MWCs in December 2000. These
standards contain two dioxin emission limits: one for small MWCs at plants with an aggregate
capacity greater than 250 tons/day (Class I MWCs) and another for small MWCs at plants with
an aggregate capacity less than 250 tons/day (Class II MWCs). The limits for the Class I MWCs
were the same as the 1995 limits for large MWCs. The limit for the smaller Class II MWCs is
125 ng/dscm. These small MWCs were on schedule to comply with the standards by December
2005. Small MWC emissions were estimated to be 63 g/yr I-TEQ in 2000 and should be less
than 2 g/yr in 2005, when all control retrofits are completed (Federal Register, 2003).
3.2. HAZARDOUS WASTE INCINERATION
Hazardous waste incineration is the controlled pyrolysis and/or oxidation of potentially
dangerous liquid, gaseous, and solid waste. It is one of the technologies used to manage
hazardous waste under the Resource Conservation and Recovery Act (RCRA) and the
Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA, or
Superfund).
Hazardous wastes are burned in a variety of situations and are covered in a number of
different sections in this report.
• Much hazardous waste is burned in facilities dedicated to burning this type of waste.
Most of these dedicated facilities are located on-site at chemical manufacturing
facilities and burn only the waste associated with their on-site industrial operations.
Hazardous waste is also burned at dedicated facilities located off-site. These facilities
accept waste from multiple sources. On- and off-site hazardous waste burning
facilities are addressed in Sections 3.2.1 through 3.2.4.
• Hazardous waste is also burned in industrial boilers and furnaces that are permitted to
burn the waste as supplemental fuel. These facilities have significantly different
furnace designs and operations than those of dedicated hazardous waste incinerators
(HWIs). They are discussed in Section 3.2.6.
• Hazardous waste is also burned in halogen acid furnaces (HAFs), in which halogen
acids (such as HC1) may be produced from halogenated secondary materials. These
facilities are discussed in Section 3.2.7.
3-41
-------
A number of cement kilns and lightweight aggregate kilns are also permitted to burn
hazardous waste as auxiliary fuel. These are discussed separately in Section 5.1.
Mobile HWIs are typically used for site cleanup at Superfund sites. These units can
be transported from one location to another and operate for a limited duration at any
given location. Because these facilities are transitory, they are not included in this
inventory at this time.
The following sections review the types of hazardous waste incineration technologies
commonly in use in the United States and present the CDD/CDF emission estimates from all
facilities operating in 1987, 1995, and 2000.
3.2.1. Furnace Designs for HWIs
The four principal furnace designs employed for the combustion of hazardous waste in
the United States are rotary kiln, liquid injection, fixed-hearth, and fluidized-bed (Dempsey and
Oppelt, 1993). The majority of commercial operations use rotary kiln incinerators. On-site
(noncommercial) hazardous waste incineration technologies use an equal mix of rotary kiln and
liquid injection furnaces, along with some fixed-hearth and fluidized-bed operations (U.S. EPA,
1996a). These HWI technologies are discussed below.
Rotary kiln. Rotary kiln incinerators consist of a rotary kiln coupled with a high-
temperature afterburner. Because rotary kilns are excess-air units designed to combust hazardous
waste in any physical form (i.e., liquid, semisolid, or solid), they are the most common type of
HWI used by commercial off-site operators. The rotary kiln is a horizontal cylinder lined with
refractory material. Rotation of the cylinder on a slight slope provides for gravitational transport
of the hazardous waste through the kiln (Buonicore, 1992a). The tumbling action of the rotating
kiln causes mixing and exposure of the waste to the heat of combustion, thereby enhancing
burnout.
Solid and semisolid wastes are loaded into the top of the kiln by an auger or rotating
screw. Fluid and pumpable sludges and wastes are typically introduced into the kiln through a
water-cooled tube. Liquid hazardous waste is fed directly into the kiln through a burner nozzle.
Auxiliary fuel (natural gas or oil) is burned in the kiln chamber at startup to reach elevated
temperatures. The typical heating value of hazardous waste (8,000 British thermal units
[Btu]/kg) is sufficient to sustain combustion without auxiliary fuel (U.S. EPA, 1996a). The
combustion gases emanating from the kiln are passed through a high-temperature afterburner
chamber to more completely destroy organic pollutants entrained in the flue gases. Rotary kilns
can be designed to operate at temperatures as high as 2,580°C, but they more commonly operate
at about 1,100°C.
3-42
-------
Liquid injection. Liquid injection incinerators are designed to burn liquid hazardous
waste. These wastes must be sufficiently fluid to pass through an atomizer for injection as
droplets into the combustion chamber. The incinerator consists of a refractory-lined steel
cylinder mounted in either a horizontal or a vertical alignment. The combustion chamber is
equipped with one or more waste burners. Because of the rather large surface area of the
atomized droplets of liquid hazardous waste, the droplets quickly vaporize. The moisture
evaporates, leaving a highly combustible mix of waste fumes and combustion air (U.S. EPA,
1996a). Secondary air is added to the combustion chamber to complete the oxidation of the
fume and air mixture.
Fixed-hearth. Fixed-hearth incinerators are starved-air or pyrolytic incinerators. Waste
is ram-fed into the primary chamber and incinerated at about 50 to 80% of stoichiometric
requirements. The resulting smoke and pyrolytic combustion products are then passed through a
secondary combustion chamber where relatively high temperatures are maintained by the
combustion of auxiliary fuel. Oxygen is introduced into the secondary chamber to promote
complete thermal oxidation of the organic molecules entrained in the gases. Other types of
hearths include roller hearths and rotary hearths. Roller hearths use a conveyor system to move
waste from the kiln entrance to the exit. In rotary hearths, waste enters and exits through the
same gate, and the hearth rotates inside a circular tunnel kiln.
Fluidized-bed. The fluidized-bed incinerator is similar in design to the incinerators used
in MSW incineration (see Section 3.1). In fluidized-bed HWIs, a layer of sand is placed on the
bottom of the combustion chamber. The bed is preheated by underfire auxiliary fuel at startup.
The hot gases channel through the sand at relatively high velocity, and the turbulent mixing of
combustion gases and combustion air causes the sand to become suspended (Buonicore, 1992a)
and take on the appearance of a fluid medium; hence the term "fluidized-bed" combustor. The
incinerator is operated at temperatures below the melting point of the bed material (typical
temperatures are within a range of 650 to 940°C). A constraint on the types of waste burned is
that the solid waste particles must be capable of being suspended within a furnace. When the
liquid or solid waste is combusted in the fluid medium, the exothermic reaction causes heat to be
released into the upper portion of the combustion chamber. The upper portion typically has
much larger volume than the lower portion, and temperatures can reach 1,000°C (Buonicore,
1992a). This high temperature is sufficient to combust volatilized pollutants emanating from the
combustion bed.
3-43
-------
3.2.2. APCDsforHWIs
Most HWIs use APCDs to remove undesirable components from the flue gases that
evolve during the combustion of the hazardous waste. These unwanted pollutants include
suspended ash particles (PM), acid gases, metals, and organic pollutants. The APCD controls
collect these pollutants and reduce the amount discharged from the incinerator stack to the
atmosphere. The levels and types of these combustion byproducts are highly site specific,
depending on factors such as waste composition and incinerator system design and operating
parameters (e.g., temperature and exhaust gas velocity). The APCD typically comprises a series
of different devices that work together to clean the combustion exhaust flue gas. Unit operations
usually include exhaust gas cooling followed by PM and acid gas control.
Exhaust gas cooling may be achieved by using a waste heat boiler or heat exchanger,
mixing with cool ambient air, or injecting a water spray into the exhaust gas. A variety of types
of APCDs are used to remove PM and acid gases. Such devices include WSs (such as venturi,
packed bed, and ionizing systems), ESPs, and FFs (sometimes used in combination with dry acid
gas scrubbing). In general, the control systems can be grouped into the following three
categories: wet, dry, and hybrid wet/dry systems. The controls for acid gases (either dry or wet
systems) cause temperatures to be reduced before the gases reach the control device. This
impedes the formation of CDDs/CDFs in the post-combustion area of the typical FIWI. It is not
unusual for stack concentrations of CDDs/CDFs at a particular HWI to be in the range of 1 to
100 ng/dscm (Helble, 1993), which is low when compared with concentrations from other waste
incineration systems. However, the range of total CDD/CDF flue gas concentrations measured
in the stack emissions of HWIs during trial burns across the class of HWI facilities spans four
orders of magnitude, ranging from 0.1 to 1,600 ng/dscm (Helble).
The three categories of APCD systems are described below:
• Wet system. A WS is used for both particulate and acid gas control. Typically, a
venturi scrubber and a packed-bed scrubber are used in a back-to-back arrangement.
Ionizing WSs, wet ESPs, and innovative venturi-type scrubbers may be used for more
efficient particulate control. WSs generate a wet effluent liquid wastestream
(scrubber blowdown). They are relatively inefficient at fine particulate control when
compared with dry control techniques, and they have equipment corrosion concerns.
However, WSs provide efficient control of acid gases and have lower operating
temperatures (compared with dry systems), which may help control the emissions of
volatile metals and organic pollutants.
• Dry system. In SDSSs, an FF or ESP is used for particulate control, frequently in
combination with dry scrubbing for acid gas control. Compared with WSs, SDSSs
are inefficient in controlling acid gases.
3-44
-------
Hybrid system. In hybrid systems, a dry technique (ESP or FF) is used for
particulate control, followed by a wet technique (WS) for acid gas control. Hybrid
systems have the advantages of both wet and dry systems (lower operating
temperature for capture of volatile metals, efficient collection of fine particulates,
efficient capture of acid gases) while avoiding many of the disadvantages. In some
hybrid systems, known as "zero discharge systems," the WS liquid is used in the dry
scrubbing operation, thus minimizing the amount of liquid byproduct waste.
Facilities that do not use any APCDs fall under a separate and unique category. These are
primarily liquid waste injection facilities, which burn wastes with low ash and low chlorine
content; therefore, they are low emitters of PM and acid gases.
3.2.3. Estimation of CDD/CDF Emission Factors for HWIs
To estimate emission factors, EPA's Office of Research and Development (ORD)
generally subdivides the combustors in each source category into design classes judged to have
similar potential for CDD/CDF emissions. However, as explained below, dedicated HWIs have
not been subdivided.
Total CDD/CDF emissions are likely the net result of all three of the mechanisms
described above (pass through, precursor, and de novo synthesis); however, the relative
importance of each mechanism can vary among source categories. In the case of HWIs, the third
mechanism (post-combustion formation) is likely to dominate, because HWIs are typically
operated at high temperatures and with long residence times, and most have sophisticated real-
time monitoring and controls to manage the combustion process. Therefore, any CDDs/CDFs
present in the feed or formed during combustion are likely to be destroyed before exiting the
combustion chamber. Consequently, for purposes of generating emission factors, it was decided
not to subdivide this class on the basis of furnace type.
Emissions resulting from the post-combustion formation of CDDs/CDFs in HWIs can be
minimized using a variety of technologies:
• Rapid flue gas quenching. The use of wet and dry scrubbing devices to remove acid
gases usually results in the rapid reduction of flue gas temperatures at the inlet to the
APCD. If the temperature is reduced below 200°C, the low-temperature catalytic
formation of CDDs/CDFs is substantially retarded.
• Use of PM APCDs. PM control devices can effectively capture condensed and
adsorbed CDDs/CDFs that are associated with the entrained PM (in particular, those
adsorbed on unburned carbon-containing particulates).
3-45
-------
Use of activated carbon. Activated CI is used at some HWIs to collect (sorb)
CDDs/CDFs from the flue gas. This may be achieved using carbon beds or by
injecting carbon and collecting it in a downstream PM APCD.
All of these approaches appear to be very effective in controlling dioxin emissions at
dedicated HWIs; emissions data are insufficient to generalize about any minor differences.
Consequently, for purposes of generating emission factors, ORD decided not to subdivide this
class on the basis of APCD type.
EPA's Office of Solid Waste (OSW) compiled a database summarizing the results of
stack testing for CDDs/CDFs at a number of HWIs between 1993 and 2000 (U.S. EPA, 2002b).
The CDD/CDF emission factors for HWIs in 1995 are based on data from 17 HWIs tested
between 1993 and 1996; emissions of HWIs in 2000 are based on data from 22 HWIs tested in
2000. The furnaces at the 22 HWI facilities tested in 2000 were 1 1 rotary kiln incinerators, 6
liquid injection incinerators, 2 rotary hearth units, 1 fluidized-bed incinerator, and 1 roller hearth.
Rather than classifying the dedicated HWI designs to derive an emission factor, ORD
decided to derive the emission factor as an average across all tested facilities. First, an average
emission factor was calculated using eq 3-3.
CxFv
EFHWI= - (3-3)
where:
EFjjw! = emission factor (average ng TEQ per kg of waste burned)
C = TEQ or CDD/CDF concentration in flue gases (ng TEQ/dscm) (20°C,
1 atm; adjusted to 7% O2)
Fv = volumetric flue gas flow rate (dscm/hr) (20°C, 1 atm; adjusted to 7% O2)
Iw = average waste incineration rate (kg/hr)
Although 22 HWIs were tested in 2000, the OSW database contained values for flue gas
flow rates for only 12 of these incinerators. Therefore, only 12 HWIs could be used to develop
an emission factor. After developing an average emission factor for each HWI, the overall
average congener-specific emission factor was derived using eq 3-4.
EFawi + EFawi + ........ + EF) /N (3-4)
3-46
-------
where:
EFavgHWI= average emission factor for the tested HWIs (ng/kg)
N = number of tested facilities
Tables 3-14a and 3-14b present the average emission factors developed for specific
congeners, total CDDs and total CDFs, and TEQs for the HWIs tested from 1993 to 1996 and in
2000, respectively. The average congener emission profile for the 17 HWIs tested from 1993 to
1996 are presented in Figure 3-13. The average emission factor for the 17 HWIs was 3.88 ng
TEQDF-WHO98/kg (3.83 ng I-TEQDF/kg) of waste feed (assuming nondetect values were zero).
The average emission factor for the 22 HWIs tested in 2000 was 2.13 ng TEQDF-WHO98/kg (2.12
ng I-TEQDF/kg) of waste feed (assuming nondetect values were zero). The emission factor
developed for reference year 1995 was used as a surrogate for reference year 1987.
3.2.4. Emission Estimates for HWIs
Although emissions data were available for 10% of the HWIs operating in 1995 and 17%
of the HWIs operating in 2000 in the United States (i.e., 22 of the 132 HWIs operating in 2000
have been tested), the emission factor estimates are assigned a medium confidence rating because
of uncertainties resulting from the following:
• Variability of the waste feeds. The physical and chemical composition of the waste
can vary from facility to facility and even within a facility. Consequently, CDD/CDF
emissions measured for one feed may not be representative of those of other feeds.
• Trial burns. Much of the CDD/CDF emissions data were collected during trial
burns, which are required as part of the RCRA permitting process and are used to
establish the destruction rate efficiency of principal hazardous organic constituents in
the waste. During trial burns, a prototype waste is burned that is intended to
maximize the difficulty in achieving good combustion. For example, chlorine,
metals, and organics may be added to the waste. The HWI may also be operated
outside normal operating conditions. The temperature of both the furnace and the
APCD may vary by a wide margin (high and low temperatures), and the waste feed
system may be increased to maximum design load. Accordingly, it is uncertain how
representative the CDD/CDF emissions measured during the trial burn will be of
emissions during normal operating conditions.
Dempsey and Oppelt (1993) estimated that up to 1.3 million metric tons of hazardous
waste were combusted in HWIs during 1987. A confidence rating of medium is assigned to this
estimate. EPA estimated that 1.5 million metric tons of hazardous waste were combusted in
HWIs each year in the early 1990s (Federal Register, 1996a). The activity level estimate for
3-47
-------
Table 3-14a. CDD/CDF emission factors for hazardous waste incinerators
and boilers tested from 1993 to 1996
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total CDD/CDF
Incinerator average mean
emission factor (17 facilities)
(ng/kg feed)
Nondetect set
to
Vz detection
limit
0.44
0.18
0.22
0.32
0.49
1.77
4.13
2.96
2.36
2.56
9.71
3.95
0.31
2.7
16.87
1.74
13.79
4.22
4.29
153
Nondetect set to
zero
0.14
0.14
0.18
0.28
0.48
1.74
3.74
2.69
2.33
2.51
9.71
3.95
0.29
2.7
16.68
1.71
13.46
3.83
3.88
153
Hot-sided ESP boilers mean
emission factor (2 facilities)
(ng/kg feed)
Nondetect set to
Vz detection
limit
0.1
0.11
0.15
0.2
0.22
1.17
5.24
0.81
0.38
0.52
0.83
0.37
0.08
0.56
1.04
0.18
0.7
0.78
0.83
28.83
Nondetect set to
zero
0.0
0.04
0.08
0.18
0.2
1.17
5.24
0.81
0.38
0.52
0.83
0.37
0.02
0.56
0.93
0.16
0.7
0.64
0.65
28.39
ESP = Electrostatic precipitator
3-48
-------
Table 3-14b. CDD/CDF emission factors for hazardous waste incinerators
and boilers tested in 2000
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDF
Total TEQnF-WHO9S
Total CDD/CDF
Incinerator average mean
emission factor (12 facilities)
(ng/kg feed)3
Nondetect set to
Vz detection limit
0.0615
0.6141
0.2347
0.5408
0.3037
2.729
5.211
0.6931
0.9406
0.88
4.085
3.031
2.667
1.218
28.74
5.056
36.270
2.54
2.809
195.70
Nondetect set
to zero
0.036
0.0907
0.1395
0.4351
0.2178
2.699
5.17
0.6399
0.8375
0.735
4.045
3.001
2.637
1.121
28.71
5.021
36.23
2.119
2.127
194.10
Hot-sided ESP boilers mean
emission factor (1 facility)
(ng/kg feed)3
Nondetect set to
Vz detection limit
0.0346
0.0488
0.1149
0.1715
0.3361
1.406
1.554
0.9531
0.4599
0.8836
3.611
0.69
0.038
1.3272
4.6345
0.1895
0.7841
1.313
1.335
17.24
Nondetect set to
zero
0.0
0.0
0.0789
0.1228
0.231
1.4055
1.5541
0.9531
0.3862
0.8836
3.6108
0.561
0.0
1.3272
4.6345
0.1257
0.7841
1.214
16.66
aValues incorporating use of the detection limit when the laboratory report indicated "not detected" for individual
CDD/CDF congeners.
ESP = Electrostatic precipitator
NR = Not reported
Source: U.S. EPA(2002a).
3-49
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Figure 3-13. Congener profile for air emissions from 17 hazardous waste
incinerators tested from 1993 through 1996.
1995 is assigned a high confidence rating because it is based on a review by EPA of the various
studies and surveys conducted in the 1990s to assess the quantity and types of hazardous wastes
being managed by various treatment, storage, and disposal facilities. Because of a lack of data
regarding the amount of waste burned in 2000, the 1995 estimate (1.5 million metric tons) was
also used for determining TEQ emissions for 2000.
The annual TEQ emissions for reference years 1987, 1995, and 2000 were estimated
using eq 3-5.
= FF
^r
HWI
(3-5)
where:
annual emissions from all HWIs, tested and nontested (g TEQ/yr)
= mean emission factor for HWIs (ng TEQ/kg of waste burned)
annual activity level of all operating HWIs (million metric tons/yr)
3-50
-------
Applying the average TEQ emission factor for dedicated HWIs (3.88 ng TEQDF-
WHO98/kg waste [3.83 ng I-TEQDF/kg waste]) to these production estimates yields estimated
emissions of 5 g TEQ (TEQDF-WHO98 or I-TEQDF) in 1987 and 5.8 g TEQDF-WHO98 (5.7 g I-
TEQDF) in 1995. For 2000, applying the average TEQ emission factors for dedicated HWIs (2.13
ng TEQDF-WHO98/kg waste [2.12 ng I-TEQDF/kg waste]) to a production estimate of 1.5 million
metric tons yields estimated emissions of 3.2 g TEQDF-WHO98 (3.18 g I-TEQDF). Medium
confidence rating is assigned to these estimates because the emission factor was given a medium
confidence rating.
3.2.5. Recent EPA Regulatory Activities
CDD/CDF emissions from HWIs are regulated by EPA (Federal Register, 1999a, 2004).
The regulations are specific to the I-TEQ concentration in the combustion gases leaving the
stack. Existing HWIs equipped with waste heat boilers and dry scrubbers (as APCDs) cannot
emit more than 0.28 ng I-TEQ/dscm. All other existing HWIs are limited to 0.4 ng I-TEQ/dscm
of stack gas. Regulatory requirements are more strict for newly built HWIs: those equipped with
waste heat boilers and dry scrubbers (as APCDs) cannot emit more than 0.11 ng I-TEQ/dscm,
and all others are limited to 0.2 ng I-TEQ/dscm of stack gas.
3.2.6. Industrial Boilers and Furnaces Burning Hazardous Waste
In 1991, EPA established rules that allow the combustion of some liquid hazardous waste
in industrial boilers and furnaces (Federal Register, 1991b). These facilities typically burn oil or
coal for the primary purpose of generating electricity. Liquid hazardous waste can be burned
only as supplemental (auxiliary) fuel, and the rule limits use to no more than 5% of the primary
fuels. These facilities typically use an atomizer to inject the waste as droplets into the
combustion chamber. They are equipped with particulate and acid gas emission controls and in
general are sophisticated, well-controlled facilities that achieve good combustion.
The national OSW database contains congener-specific emission concentrations for two
boilers burning liquid hazardous waste as supplemental fuel tested from 1993 to 1996. The
average congener and congener group emission profiles for the industrial boiler data set are
presented in Figure 3-14. The database also contains congener-specific emission concentrations
for four boilers tested in 2000. Of the boilers tested in 2000, sufficient data to calculate average
TEQ emissions were available for only one boiler. The average congener and TEQ emission
factors are presented in Tables 3-14a and 3-14b. The limited set of emissions data prevented
subdividing this class to derive an emission factor. The equation used to derive the emission
factor is the same as eq 3-4. The TEQ emission factors for the industrial boiler are 0.65 ng
3-51
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2 0.25
Figure 3-14. Congener and congener group profiles for air emissions from
boilers and industrial furnaces burning hazardous waste.
3-52
-------
TEQDF-WHO98/kg (0.64 ng I-TEQDF/kg) of waste feed for 1993 to 1996 and 1.212 ng TEQDF-
WHO98/kg (1.214 ng I-TEQDF/kg) of waste feed for 2000. These emission factors are assigned a
low confidence rating because they reflect testing at only 2 of 136 hazardous waste boilers and
furnaces operating from 1993 to 1996 and only 1 of the 114 hazardous waste boilers and furnaces
operating in 2000.
Dempsey and Oppelt (1993) estimated that approximately 1.2 billion kg of hazardous
waste were combusted in industrial boilers/furnaces in 1987. EPA estimated that in each year in
the early 1990s approximately 0.6 billion kg of hazardous waste were combusted in industrial
boilers/furnaces (Federal Register, 1996a). It is possible that cement kilns and light-weight
aggregate kilns burning hazardous waste were included in the estimate by Dempsey and Oppelt
for 1987; the estimate for 1995 does not appear to include these hazardous waste-burning kilns.
A confidence rating of low is assigned to the estimated activity level for 1987, which was largely
based on a review of state permits (Dempsey and Oppelt, 1993). The activity level estimate for
1995 is assigned a medium confidence rating because it was based on a review by EPA of the
various studies and surveys conducted in the 1990s to assess the quantity and types of hazardous
wastes being managed by various treatment, storage, and disposal facilities. Because of a lack of
data regarding the amount of waste burned in 2000, the 1995 estimate (1.5 million metric tons)
was used as a surrogate for 2000.
Equation 3-5, which was used to calculate annual TEQ emissions for dedicated HWIs,
was also used to calculate annual TEQ emissions for industrial boilers/furnaces. Multiplying the
average TEQ emission factors by the total estimated kg of liquid hazardous waste burned in
1987, 1995, and 2000 yields annual emissions in g-TEQ/yr. From this procedure, the emissions
from all industrial boilers/furnaces burning hazardous waste as supplemental fuel are estimated
as 0.78 g TEQDF-WHO98 (0.77 g I-TEQDF) in 1987, 0.39 g TEQDF-WHO98 (0.38 g I-TEQDF) in
1995, and 1.82 g TEQ (TEQDF-WHO98 or I-TEQDF) in 2000. Because of the low confidence
rating for the emission factor, the overall confidence rating is low for the emission estimates for
all three reference years.
3.2.7. Halogen Acid Furnaces Burning Hazardous Waste
Hazardous waste can be used in the production of halogen acids using an HAF.
According to EPA rules, products that qualify as hazardous waste, as defined in 40 CFR 261.2
must be regulated as such, even if the products are used in the production of halogen acids using
an HAF (Federal Register, 1991c).
The national OSW database contains congener-specific emission concentrations for two
HAFs burning liquid hazardous waste as supplemental fuel tested in 2000. Data from these two
facilities were used to calculate an emission factor for HAFs. The average congener and TEQ
3-53
-------
emission factors are presented in Table 3-15. The equation used to derive the emission factor is
the same as eq 3-4. The average TEQ emission factor for HAFs is 0.836 ng TEQDF-WHO98/kg
(0.803 ng I-TEQDF/kg) of waste feed for reference year 2000. This emission factor is assigned a
low confidence rating because it reflects testing at only 12.5% of all HAFs operating in 2000 (2
out of 16).
Table 3-15. CDD/CDF emission factors for halogen acid furnaces tested in
2000
Congener/congener
group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total CDD/CDF
Incinerator average mean emission factor (12 facilities) (ng/kg feed)
Nondetect set to
Vz detection limit
0.0274
0.1164
0.0979
0.1663
0.1686
0.9868
1.4944
0.3821
0.583
0.5689
1.1244
0.7172
0.4412
0.2685
3.4914
1.0429
25.015
0.8176
0.8519
62.4773
Nondetect set to zero
0.0208
0.112
0.0913
0.1594
0.1293
0.9868
1.4944
0.3821
0.583
0.5689
1.1244
0.7172
0.4412
0.2685
3.4914
1.0429
25.015
0.8034
0.8356
62.4607
Source: U.S. EPA(2002a).
The amount of hazardous waste combusted using HAFs in 2000 was conservatively
estimated to be 375,600 metric tons. This estimate is based on data provided by OSW that
described activity levels for each individual HAF in 2000. Activity data were available for 14 of
the 16 facilities. By assuming that plants operate continuously throughout the year, that they are
always running at 80% of maximum capacity, and that the activity levels represent the maximum
capacity, a conservative estimate for the annual quantity burned per HAF was derived (23,480
3-54
-------
kg/yr). This quantity multiplied by the total universe of 16 facilities yields the final estimate of
375,600 metric tons. This was assigned a low confidence rating because the data was possibly
nonrepresentative.
Equation 3-5, which was used to calculate annual TEQ emissions from dedicated HWIs,
was also used to calculate annual TEQ emissions from HAFs. Multiplying the average TEQ
emission factors by the total estimated kilograms of liquid hazardous waste burned in 2000 yields
annual emissions in g I-TEQDF. From this procedure, the emissions from all industrial
boilers/furnaces burning hazardous waste as supplemental fuel are estimated as 0.31 g TEQDF-
WHO98 (0.3 g I-TEQDF). Because of the low confidence rating for the emission factor, the
overall confidence rating is low for the emission estimates.
3.2.8. Solid Waste from Hazardous Waste Combustion
U.S. EPA (1987a) contains limited data on ash generated from hazardous waste
incineration. The study indicates that the mean concentrations of CDDs and CDFs from an HWI
with an afterburner were 538 |ig/kg and 2,853 i-ig/kg, respectively (Table 3-8 in U.S. EPA,
1987a). Specific data for congeners and for ash quantities were not provided.
3.3. MEDICAL WASTE INCINERATION
Medical waste incineration is the controlled burning of solid wastes generated primarily
by hospitals, veterinary facilities, and medical research facilities. EPA defines medical waste as
any solid waste generated in the treatment, diagnosis, or immunization of humans or animals or
research pertaining thereto or in the production or testing of biologicals (Federal Register,
1997a). The primary purposes of medical waste incineration are to reduce the volume and mass
of waste in need of land disposal and to sterilize the infectious materials. The following sections
review the basic types of medical waste incinerator (MWI) designs used to incinerate medical
waste and the distribution of APCDs used on MWIs and summarize the derivation of dioxin TEQ
emission factors for MWIs and the national dioxin TEQ emission estimates for reference years
1987, 1995, and 2000.
3.3.1. Design Types of MWIs Operating in the United States
For purposes of this document, EPA has classified MWIs into three broad technology
categories: modular furnaces using controlled air, modular furnaces using excess air, and rotary
kilns. Of the MWIs in use today, the vast majority are believed to be modular furnaces using
controlled air. EPA has estimated that 97% are modular furnaces using controlled air, 2% are
modular furnaces using excess air, and 1% are rotary kiln combustors (U.S. EPA, 1997a).
3-55
-------
Modular furnaces using controlled air. Modular furnaces have two separate
combustion chambers mounted in series (one on top of the other). The lower chamber is where
the primary combustion of the medical waste occurs. Medical waste is ram-fed into the primary
chamber and underfire air is delivered beneath the incinerator hearth to sustain good burning of
the waste. The primary combustion chamber is operated at below stoichiometric levels, hence
the terms "controlled air" or "starved air." With substoichiometric conditions, combustion
occurs at relatively low temperatures (760 to 985°C). Under the conditions of low oxygen and
low temperatures, partial pyrolysis of the waste occurs and volatile compounds are released.
The combustion gases pass into a second chamber. Auxiliary fuel (such as natural gas) is
burned to sustain elevated temperatures (985 to 1,095°C) in this secondary chamber. The net
effect of exposing the combustion gases to an elevated temperature is more complete destruction
of the organic contaminants entrained in the combustion gases emanating from the primary
combustion chamber. Combustion air at 100 to 300% in excess of stoichiometric requirements is
usually added to the secondary chamber. Gases exiting the secondary chamber are directed to an
incinerator stack (U.S. EPA, 1991b, 1997a; Buonicore, 1992b). Because of its low cost and good
combustion performance, this design has been the most popular choice for MWIs and has
accounted for more than 95% of systems installed over the past two decades (U.S. EPA, 1990b,
1991b; Buonicore, 1992b).
Modular furnaces using excess air. These systems use the same modular furnace
configuration as described above for the controlled-air systems. The difference is that the
primary combustion chamber is operated at air levels of 100 to 300% in excess of stoichiometric
requirements, hence the name "excess air." A secondary chamber is located on top of the
primary unit. Auxiliary fuel is added to sustain high temperatures in an excess-air environment.
Excess-air MWIs typically have smaller capacity than do controlled-air units, and they are
usually batch-fed operations. This means that the medical waste is ram-fed into the unit and
allowed to burn completely before another batch of medical waste is added to the primary
combustion chamber. Figure 3-4 shows a schematic of a typical modular furnace using excess
air.
Rotary kiln. In terms of design and operational features, the rotary kiln technology used
in medical waste incineration is similar to that employed in both municipal and hazardous waste
incineration (see description in Section 3.1). Because of their relatively high capital and
operating costs, few rotary kiln incinerators are in operation for medical waste treatment (U.S.
EPA, 1990b, 1991b; Buonicore, 1992b).
MWIs can be operated in three modes: batch, intermittent, and continuous. Batch
incinerators burn a single load of waste, typically only once per day. Waste is loaded and ashes
are removed manually. Intermittent incinerators, which are loaded continuously and frequently
3-56
-------
with small waste batches, operate less than 24 hr/day, usually on a shift basis. Either manual or
automated charging systems can be used, but the incinerator must be shut down for ash removal.
Continuous incinerators are operated 24 hr/day and use automatic charging systems to charge
waste into the unit in small, frequent batches. All continuous incinerators operate using a
mechanism to automatically remove the ash from the incinerator (U.S. EPA, 1990b, 1991b).
3.3.2. Characterization of MWIs for Reference Years 1987,1995, and 2000
Medical waste incineration remains a poorly characterized industry in the United States
in terms of knowing the exact number of facilities in operation over time, the types of APCDs
installed on these units, and the aggregate volume and weight of medical waste that is combusted
in any given year (U.S. EPA, 1997a). The primary reason for this lack of information is that
permits were not generally required for the control of pollutant stack emissions from MWIs until
the early 1990s, when state regulatory agencies began setting limits on emissions of PM and
other contaminants (Federal Register, 1997a). Prior to that, only opacity was controlled.
The information available to characterize MWIs from 1987 and 1995 comes from
national telephone surveys, stack emission permits, and data gathered by EPA during public
hearings (Federal Register, 1997a). For 2000, information was also provided by a memorandum
on emissions from MWIs (Strong and Hanks, 1999) and a limited telephone survey (McAloon,
2003). Strong and Hanks provided information on MWIs in the United States, including the
APCD being used by each facility. A telephone survey was conducted with the state agencies in
each of these six states to obtain the number of MWIs that were operating in 2000. EPA was
able to obtain an updated list from four of the six states, which are shown below, along with the
dates they were contacted, the number of MWIs operating in 1999, the updated number of MWIs
for that state in 2000, and the percent of facilities closed over this time period for each state.
No. of MWIs Percentage of facilities
State Date contacted 1999 2000 closed from 1999 to 2000
Illinois Jan. 16,2003 97 13 86.6
Louisiana Jan. 16,2003 92 24 73.91
Maryland Dec. 2,2002 36 30 16.67
Michigan Nov. 26,2002 228 45 80.26
The geometric mean of the closure percentages for the four states was determined to be
54.09 and the arithmetic mean was 64.36. Maryland had the lowest closure percent from 1999 to
2000; however, through discussions with representatives of Maryland state agencies, it was
3-57
-------
determined that close to 70% of the facilities operating in 1999 would be shut down as of 2003.
It was therefore assumed that the average closure percent of 64.36 was a fairly good estimate for
all states. This average was applied to the total number of facilities operating in 1999 from the
Strong and Hanks (1999) memorandum to estimate the number of facilities operating in 2000.
The information obtained from these sources suggests the following:
• The number of MWIs in operation for each reference year was approximately 5,000 in
1987 (U.S. EPA, 1987c), 2,375 in 1995 (Federal Register, 1997a), and 1,065 in 2000
(Strong and Hanks, 1999; McAloon, 2003).
The amount of medical waste combusted annually in the United States was
approximately 1.43 billion kg in 1987 (U.S. EPA, 1987c) and 0.77 billion kg in 1995
(Federal Register, 1997a).
These estimates indicate that between 1987 and 1995 the total number of operating MWIs
and the total amount of waste combusted decreased by more than 50%. From 1995 to 2000, the
total number of operating MWIs decreased by approximately 55%. A variety of factors probably
contributed to the reduction in the number of operating facilities, including federal and state
regulations and air pollution control requirements. In 1997, EPA adopted emission guidelines
for existing MWIs (incinerators constructed on or before June 20, 1996) and New Source
Performance Standards for new MWIs (incinerators constructed after June 20, 1996). The Clean
Air Act requires that states implement the emission guidelines according to a state plan and that
they submit the state plan to EPA within one year of EPA's promulgation of the guidelines (i.e.,
by September 15, 1998). The compliance schedule, however, allows up to three years from EPA
approval of the state plan for MWIs to comply, provided the plan includes enforceable
increments of progress. All MWIs were required to be in compliance within three years of
approval of their state plan or by September 15, 2002, whichever was earlier.
Compliance is stated to be either completion of retrofit of air pollution controls or
shutdown of the facility. As a result, many facilities have closed down and hospitals have
switched to less expensive medical waste treatment technologies, such as autoclaving (Federal
Register, 1997a). Autoclaving, or steam sterilization, is one of the most common waste
management practices used today. This process involves placing bags of infectious waste into a
sealed chamber, sometimes pressurized, and then heating it by direct contact with steam to
sterilize the waste.
The actual controls used on MWIs on a facility-by-facility basis in 1987 are unknown,
and EPA generally assumes that MWIs were mostly uncontrolled (U.S. EPA, 1987c). However,
the modular design does cause some destruction of organic pollutants within the secondary
3-58
-------
combustion chamber. Residence time within the secondary chamber is key to inducing the
thermal destruction of the organic compounds. Residence time is the time that the organic
compounds entrained within the flue gases are exposed to elevated temperatures in the secondary
chamber. EPA has demonstrated with full-scale MWIs that increasing residence time from 1/4
sec to 2 sec in the secondary chamber can reduce organic pollutant emissions, including
CDDs/CDFs, by up to 90% (Federal Register, 1997a). In this regard, residence time can be
viewed as a method of air pollution control.
EPA estimates that about two-thirds of the medical waste burned in MWIs in 1995 went
to facilities that had some method of air pollution control (Federal Register, 1997a).The types of
APCDs installed and the methods used on MWIs include DSI, FFs, ESPs, WSs, and FFs
combined with packed-bed scrubbers (composed of granular activated carbon). Some organic
constituents in the flue gases can be adsorbed by the packed bed. Within the uncontrolled class
of MWIs, about 12% of the waste was combusted in facilities with design capacities of less than
200 Ib/hr, with the majority of waste burned at facilities with capacities greater than 200 Ib/hr. In
controlled facilities, an estimated 70% of the aggregate activity level is associated with facilities
equipped with either WSs, FFs, or ESPs; 29.9% is associated with facilities that use DSI
combined with FFs; and less than 1% is associated with facilities that have an FF/packed-bed
APCD (AHA, 1995; Federal Register, 1997a).
Strong and Hanks (1999) provided information on the types of APCDs used by facilities
operating in 1999. Ten types were included in the memorandum, which included residence time
as a type of control technology. The 10 types were 1/4-sec combustion, 1-sec combustion, 2-sec
combustion, low-efficiency WS, moderate-efficiency WS, high-efficiency WS, dry lime inject-
FF, dry lime inject-FF with CI, WS/dry lime inject-FF, and SD/FF with CI. Table 3-16 provides
an estimated breakdown of these APCDs.
3.3.3. Estimation of CDD/CDF Emissions from MWIs
Emission tests reported for 22 MWIs (about 3% of the existing facilities operating in
2000) were collected for use in this document; emission levels of dioxin-like compounds at most
facilities are unmeasured. Because so few facilities have been evaluated, the estimation of
annual air emissions of CDDs/CDFs from MWIs is quite dependent on extrapolations,
engineering judgment, and assumptions. In addition, the information about the activity levels of
these facilities is also quite limited.
The analysis divided MWIs into three design types on the basis of mode of daily
operation: batch, intermittent, or continuous. This was done using the information from the
inventory on design-rated annual incineration capacity of each facility. The smaller capacity
3-59
-------
Table 3-16. Estimated breakdown of facilities by air pollution control device
(APCD)
APCD
1/4-sec combustion
1-sec combustion
2-sec combustion
Low-efficiency wet scrubber
Moderate-efficiency wet scrubber
High-efficiency wet scrubber
Dry lime inject fabric filter
Dry lime inject fabric filter with carbon injection
Wet scrubber/dry lime inject fabric filter
Spray dryer fabric filter with carbon injection
Number of
facilities
229
259
455
208
75
16
44
7
14
1
Percent of
total
17.5
19.8
34.8
15.9
5.7
1.2
3.4
0.5
1.1
0.1
Source: Strong and Hanks (1999).
units were assumed to be batch operations, and the others were classified as either intermittent or
continuous, assuming a ratio of 3 to 1.
The activity level of each facility was estimated by multiplying the design-rated annual
incineration capacity of the MWI (kg/hr) by the hours of operation (hr/yr). The annual hours of
operation were determined by assuming a capacity factor (defined as the fraction of time that a
unit operates over the year) for each design type of MWI (Randall, 1995). Table 3-17 is a
summary of the estimated annual operating hours for each MWI design type.
In estimating dioxin emissions, the MWIs were divided into two classes: those having
APCDs (controlled) and those lacking any APCD (uncontrolled). These two classes of MWIs
are discussed below.
For 1987, it is assumed that every MWI was uncontrolled. An EPA study of MWIs
conducted at that time indicated that MWIs operating in 1987 did not need controls because they
were not subject to state or federal limits on either PM or organic pollutant emissions (U.S. EPA,
1987c). The activity level estimates were derived from data presented in that 1987 study. This
approach resulted in the following activity level assumptions for 1987: (a) 15% of the activity
level (0.22 billion kg) was incinerated annually by MWIs with capacities less than or equal to
200 Ib/hr, and (b) 85% of the activity level (1.21 billion kg) was incinerated annually by facilities
with capacities greater than 200 Ib/hr (see Table 3-18). For 1995, the activity levels were then
summed across facilities for each APCD subclass (see Table 3-19).
3-60
-------
Table 3-17. Summary of annual operating hours for each medical waste
incinerator (MWI) type
MWI type
Continuous commercial
Continuous onsite
Intermittent
Batch
Capacity
range
(Ib/hr)
>1,000
501-1,000
>1,000
<500
Case by case
Annual
charging
hours
(hr/yr)
7,776
1,826
2,174
1,250
Case by case
Maximum
annual
charging
hours
(hr/yr)
8,760
5,475
4,380
Capacity
factor
0.89
0.33
0.40
0.29
Case by case
Source: U.S. EPA(1990c).
In 1997, the amount of waste combusted by MWIs was estimated to be 0.8 million tons/yr
(0.7 billion kg/yr) (NRC, 2000). This number represents a 9% decrease from 1995. If we
assume that this decrease occurred every two years from 1997 to 2000, the estimated amount of
waste combusted by MWIs for 2000 would be 0.6 billion kg/yr. This is a conservative estimate,
considering the large number of facilities that have shut down or switched to less expensive
medical waste treatment technologies. For 2000 activity level estimates, the same distributions
among APCD classes were assumed as for 1995. These activity level estimates are presented in
Table 3-20. For all years, these activity levels were assigned a rating of low confidence because
the data were judged to be possibly nonrepresentative.
The stack test results showing the air emissions of dioxin from 24 MWIs were obtained
and used to calculate 1987 and 1995 emission estimates. After reviewing these test reports, EPA
determined that 20 met the criteria for acceptability (see Section 3.1.3). In some cases,
CDD/CDF congener-specific data were not reported or values were missing. In other cases, the
protocols used in the laboratory analysis were not described; therefore, no determination of the
adequacy of the laboratory methods could be made. For 2000, two additional test reports from
facilities operating in that year were obtained and were included with the previously obtained test
reports in order to calculate updated emission estimates. Each test report was included in its
respective MWI subclass according to its APCD and was also included in the overall emission
estimate.
3-61
-------
Table 3-18. TEQ emissions from medical waste incinerators (MWIs) for reference year 1987
MWI class3
<200 Ib/hr
>200 Ib/hr
TOTAL
No. of tested
facilities
3
5
8
Activity
level
(kg/yr)
2.19e+08
1.21e+09
1.43e+09
Total CDD/
CDF
emission
factorb
(g/kg)
9.25e-05
6.05e-05
I-TEQDF
emission
factor
(g/kg)
1.86e-06
1.68e-06
TEQDF-
WHO98
emission
factor
(g/kg)
1.98e-06
1.78e-06
Annual
CDD/
CDF
emissions
(g/yr)
2.02e+04
7.32e+04
9.34e+04
Annual
I-TEQDF
emissions
(g/yr)
4.08e+02
2.03e+03
2.44e+03
Annual
WHO98-
TEQDF
emissions
(g/yr)
4.34e+02
2.14e+03
2.57e+03
Table 3-19. TEQ emissions from medical waste incinerators (MWIs) for reference year 1995
MWI class
(air pollution
control device
[APCD])
Uncontrolled
Controlled
TOTAL
MWI
subclass
(capacity or
APCD3)
<200 Ib/hr
>200 Ib/hr
WS/FF/ESP
DSI/FF
FF/PBS
No. of
tested
facilities
3
5
9
2
1
Total
CDD/CDF
emission
factor
(ng/kg)
9.25e+04
6.05e+04
4.67e+04
2.85e+02
l.lle+05
I-TEQDF
emission
factor
(ng/kg)
1.86e+03
1.80e+03
7.22e+01
6.78
1.35e+03
TEQDF-
WHO98
emission
factor (ng/kg)
1.98e+03
1.78e+03
6.63e+01
4.61
1.49e+03
Activity
level
(kg/yr)
3.06e+07
2.23e+08
3.71e+08
1.46e+08
6.99e+05
7.71e+08
Annual
CDD/CDF
emissions
(g/yr)
2.83e+03
1.35e+04
1.73e+03
4.16e+01
7.76e+01
1.82e+04
Annual
I-TEQDF
emissions
(g/yr)
5.71e+01
3.75e+02
2.68e+01
9.90e-01
9.44e-01
4.59e+02
Annual
TEQDF-
WHO98
emissions
(g/yr)
6.06e+01
3.97E+02
2.76E+01
l.OOE+00
1.04e+00
4.87E+02
00
I
05
to
aSlash(es) indicates devices used in conjunction.
APCD (air pollution control device):
DSI = Dry sorbent injection
ESP = Electrostatic precipitator
FF = Fabric filter
PBS = Packed-bed scrubber
WS = Wet scrubber
-------
Table 3-20. TEQ emissions from medical waste incinerators (MWIs) for reference year 2000
MWI class
(air pollution
control device
[APCD])
Uncontrolled
Controlled
TOTAL
MWI
subclass
(capacity or
APCD3)
<200 Ib/hr
>200 Ib/hr
WS/FF/ESP
DSI/FF
FF/PBS
No. of
tested
facilities
3
5
9
2
1
Total
CDD/CDF
emission
factor
(ng/kg)
9.25e+04
6.05e+04
4.67e+04
2.85e+02
l.lle+05
I-TEQDF
emission
factor
(ng/kg)
1.86e+03
1.68e+03
6.44e+01
4.56
1.35e+03
TEQDF-
WH098
emission
factor
(ng/kg)
1.98e+03
1.78e+03
6.63e+01
4.61
1.49e+03
Activity
level
(kg/yr)
2.40e+07
1.74e+08
2.88e+08
1.14e+08
5.40e+05
6.01e+08
Annual
CDD/CDF
emissions
(g/yr)
2.22e+03
1.05e+04
1.34e+04
3.25e+01
5.99e+03
3.22e+04
Annual
I-TEQDF
emissions
(g/yr)
4.46e+01
3.13e+02
2.08e+01
7.73e-01
7.29e+01
3.57e+02
Annual
TEQDF-
WHO98
emissions
(g/yr)
4.75e+01
3.10e+02
1.91e+01
5.26e-01
8.05e+01
3.78e+02
00
I
05
00
aSlash(es) indicates devices used in conjunction.
APCD:
DSI = Dry sorbent injection
ESP = Electrostatic precipitator
FF = Fabric filter
PBS = Packed-bed scrubber
WS = Wet scrubber
-------
The EPA stack testing method (EPA Method 23) produces a measurement of CDDs/
CDFs in units of mass concentration (ng/dscm) at standard temperature and pressure and 1 atm
and adjusted to a measurement of 7% oxygen in the flue gas (U.S. EPA, 1995a). This
concentration is assumed to represent conditions at the point of release from the stack into the air
and to be representative of routine emissions. The emission factors were derived by averaging
the emission factors across each tested facility in a design class. The emission factor for each
tested MWI was calculated using the following equation:
C x Fv
EFMWI = (3-8)
4
where:
EFjvnvj = emission factor per MWI (average ng TEQ per kg medical waste burned)
C = average TEQ concentration in flue gases of tested MWIs (ng TEQ/dscm)
(20°C, 1 atm; adjusted to 7% O2)
Fv = average volumetric flue gas flow rate (dscm/hr) (20°C, 1 atm; adjusted
to 7% O2)
Iw = average medical waste incineration rate of the tested MWI (kg/hr)
3.3.4. Summary of CDD/CDF Emissions from MWIs
Annual dioxin emissions were estimated by multiplying the emission factor and activity
level developed for each design class and then summing the calculated emissions for all classes.
Tables 3-18, 3-19, and 3-20 summarize the resulting national TEQ air emissions for reference
years 1987, 1995, and 2000, respectively. These tables also indicate the activity level and the
TEQ emission factor used in estimating annual TEQ emissions.
In estimating annual TEQ emissions for each reference year, a low confidence rating was
assigned to the estimate of the activity level, primarily because very limited information is
available on a facility-level basis for characterizing MWIs in terms of the frequency and duration
of operation, the actual waste volume handled, and the level of pollution control. The 1987
inventory of facilities was based on very limited information. Although the 1995 OAQPS
inventory was more comprehensive than the 1987 inventory, it was still based on a fairly limited
survey of operating facilities (approximately 6%). The 2000 inventory included only two
additional facilities and estimated an activity level based on a 1997 value and the distribution
among APCDs from the 1995 estimates.
The emission factor estimates were given a low confidence rating because the reports of
only 20 tested MWI facilities could be used to derive emission factors representing the 2,375
facilities operating in 1995 (i.e., less than 1% of the estimated number of operating facilities) and
only two additional test reports were obtained for 2000. Even fewer tested facilities could be
3-64
-------
used to represent the larger number of facilities operating in 1987 (8 tested facilities were used to
represent 5,000 facilities). The limited emission tests available cover all design categories used
here to develop emission factors. However, because of the large number of facilities in each of
these classes, it is very uncertain whether the few tested facilities in each class capture the true
variability in emissions.
Table 3-20 shows the 2000 emissions estimate as being 378 g TEQDF-WHO98 (357 g I-
TEQDF). The TEQ emissions are estimated to have been 487 g TEQDF-WHO98 (459 g I-TEQDF)
in 1995 (Table 3-19) and 2,590 g TEQDF-WHO98 (2,440 g I-TEQDF) in 1987 (Table 3-18).
Because the activity level and emission factors had low confidence ratings, the emission
estimates for all years were assigned a low confidence rating, i.e., a Category C. Figures 3-15
and 3-16 display the congener and congener group profiles of MWIs without APCDs and those
equipped with WSs and FFs, respectively.
3.3.5. Recent EPA Regulatory Activities
In September 1997, EPA promulgated final regulations under the Clean Air Act
Amendments limiting CDD/CDF stack emissions from MWIs (Federal Register, 1997a). These
emission limits are specific to the sum of CDD and CDF emissions (the sum of tetra- through
octa-CDDs and CDFs). For either new or existing MWIs that were operational before or after
June 20, 1996, EPA limits the total CDD/CDF concentration in the stack gases to 2.3ng/dscm.
This would require the application of WSs, DSI of activated carbon combined with FFs and/or
SDs/FFs. EPA expects that many facilities that currently operate on-site incinerators will switch
to less expensive methods of treatment and disposal of medical and infectious waste when faced
with the compliance costs associated with the emission standards for MWIs. EPA projects that,
following full compliance with these standards, annual emissions from MWIs will be 5 to 7 g I-
TEQDF/yr.
3.4. CREMATORIA
3.4.1. Human Crematoria
3.4.1.1 Emissions Data
Bremmer et al. (1994) measured CDD/CDF emissions at two crematoria in the
Netherlands. The first, a "cold"-type furnace with direct, uncooled emissions, was calculated to
yield 2,400 ng I-TEQDF per body. In the cold-type furnaces, the coffin is placed inside at a
temperature of about 300°C. The temperature of the chamber is then increased to 800 to 900°C
using a burner and kept there for 2 to 2.5 hr. The second furnace, a "warm" type in which flue
gases are cooled to 220°C prior to discharge, was calculated to yield 4,900 ng I-TEQDF per body.
In the warm-type furnace, the coffin is placed in a chamber preheated to 800°C or higher for 1.2
to 1.5 hr. The chamber exhausts from both furnace types were incinerated in an afterburner at a
3-65
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
[,2,3,7,8,9-HxCDD
,3,4,6,7,8-HpCDD
1,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
!,3,4,6,7,8-HpCDF
!,3,4,7,8,9-HpCDF
3,4,6,7,8,9-OCDF
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2 0.25
Figure 3-15. Congener and congener group profiles for air emissions from
medical waste incinerators without air pollution control devices (nondetects
set equal to zero).
3-66
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2
Figure 3-16. Congener and congener group profiles for air emissions from
medical waste incinerators equipped with a wet scrubber and fabric filter.
3-67
-------
temperature of about 850°C. The higher emission rate for the warm-type furnace was attributed
by the authors to the formation of CDDs/CDFs during the intentional cooling of the flue gases to
220°C.
Jager et al. (1992) (as reported in Bremmer et al., 1994) measured an emission rate of
28,000 ng I-TEQDF per body for a crematorium in Berlin, Germany. No operating process
information was provided by Bremmer et al. for the facility.
Mitchell and Loader (1993) reported even higher emission factors for two crematoria in
the United Kingdom. The first facility tested was manually operated and had primary and
secondary combustion chambers preheated to 650°C and a residence time of 1 sec in the
secondary combustion chamber. The second tested facility was computer controlled and had
primary and secondary combustion chambers heated to 850°C and a residence time of 2 sec in
the secondary combustion chamber. The measured stack gas TEQ concentrations ranged from 42
to 71.3 ng I-TEQDF/m3 (at 11% oxygen) at the first facility and from 25.4 to 45.5 ng I-TEQDF/m3
(at 11% oxygen) at the second facility. Emission factors based on these test results and gas
generation rates reported by Bremmer et al. (1994) were calculated to range from 70,000 to
80,000 ng I-TEQDF/body (HMIP, 1995).
Takeda et al. (1998) measured CDD/CDF emissions at 10 crematoria in Japan. Although
there are more than 1,600 crematoria in Japan, the 10 tested facilities handle 4% of the
cremations carried out in Japan annually. A wide range of CDD/CDF emissions were observed.
When nondetect values were treated as zero, the emission factor range was 42 to 62,000 ng I-
TEQDF/body (mean of 9,200 ng I-TEQDF/body). When nondetect values were treated as one-half
the DL, the range was 450 to 63,000 ng I-TEQDF/body (mean of 11,000 ng I-TEQDF/body).
To obtain more data on CDD/CDF emissions from crematoria in Japan, Takeda et al.
(2001) measured CDD/CDF emissions at 17 additional crematoria. In that study, all the
crematoria except one had secondary combustion chambers. Additionally, one crematorium had
a secondary combustion chamber but did not use it. One to four main chambers were connected
to the secondary chambers, and the temperature of the main chambers ranged from
approximately 650 to 1,150°C. In most cases, only one body was cremated at time. However,
between two and four bodies were cremated at four sampling events. A coffin and any
accompanying materials were combusted along with the body. Emission factors ranged from 120
to 24,000 ng I-TEQDF/body. In general, as the average temperature in the main combustion
chamber increased, CDD/CDF emissions decreased. However, the crematorium that had a
secondary combustion chamber but did not use it had both high temperatures in the main
combustion chamber and high CDD/CDF emissions. Additionally, with the rise of the average
temperature in the secondary combustion chamber of the eight crematoria without dust
collectors, CDD/CDF emissions decreased. For crematoria with dust collectors, the relationship
3-68
-------
between the average temperature in the secondary combustion chamber and CDD/CDF emissions
was not clear.
EPA obtained test data from two crematoria for humans operating in the United States,
one at Camellia Memorial Lawn in California (CARB, 1990a) and one at Woodlawn Cemetery in
New York (U.S. EPA, 1999a). Additionally, EPA obtained test data from one crematorium for
animals operating in the United States: University of Georgia Veterinary School (U.S. EPA,
2000a); however, it is not appropriate to use the emission factors from this facility to characterize
emissions associated with human cremation.
Testing at the Camellia Memorial Lawn crematorium, which is classified as a warm-type
facility using the criteria of Bremmer et al. (1994), was conducted in 1990 (CARB, 1990a). The
combusted material at this facility consisted of the body, as well as 4 Ib of cardboard, up to 6 Ib
of wood, and an unquantified amount of unspecified plastic wrapping. The three emissions tests
conducted at this facility, which operates using an afterburner, yielded an average emissionfactor
of 543 ng TEQDF-WHO98/body (501 ng I-TEQDF/body). Table 3-21 presents the congener-
specific emission factors for this facility.
Testing at Woodlawn Cemetery, which has a crematorium with a primary combustion
chamber, a secondary combustion chamber, and a scrubber APCD, was conducted in 1995. Tests
were run at three secondary combustion chamber temperatures: 675, 870, and 980°C (U.S. EPA,
1999a). The combusted material consisted of the body, as well as a 10- to 100-lb casket
constructed of fiberboard, particle board, or wood and various body wrappings and articles such
as a plastic sheet, a cloth sheet, or clothes. For this facility, average emission factors of 362 and
709 ng TEQDF-WHO98/body cremated (348 and 638 ng I-TEQDF/body cremated) were calculated,
based on emissions collected at the scrubber inlet and outlet, respectively. The congener-specific
emission factors for this facility are shown in Table 3-22.
In 1995, 1,155 crematoria were reported to be operating in the United States; this number
had decreased to approximately 1,060 by 2000. To determine whether the emissions data
collected at the Woodlawn Cemetery facility are representative of a typical crematorium
operating in the United States, representatives from the Cremation Association of North America
(CANA) were contacted to identify the typical operating conditions at U.S. crematoria.
According to the CANA representatives, all crematoria operating in the United States have
primary and secondary combustion chambers. Additionally, crematoria with operating
conditions that indicate the presence of an afterburner are considered to contain secondary
combustion chambers. The primary and secondary combustion chambers at U.S. crematoria
typically operate at between 675 and 870°C, but many operate at 980°C, as required by their
respective states.
3-69
-------
Table 3-21. Congener-specific profile for Camellia Memorial Lawn
crematorium
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean facility emission factor (ng/body)
Assuming nondetect
set to zero
28.9
89.6
108
157
197
1,484
2,331
206
108
339
374
338
657
135
1,689
104
624
4,396
4,574
501.8
544.1
554
860
2,224
3,180
2,331
4,335
2,563
4,306
2,030
624
23,007
Assuming nondetect set to
1A detection limit
28.9
89.6
108
157
197
1,484
2,331
206
117
349
374
338
657
135
1,813
112
624
4,396
4,725
508.6
550.9
554
860
2,224
3,180
2,331
4,335
2,563
4,306
2,154
624
23,131
Source: CARB (1990a).
3-70
-------
Table 3-22. Congener-specific profile for the Woodlawn Cemetery
crematorium
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Mean emission factor,
scrubber inlet (ng/body)
Nondetect set
to zero
11
31
74
115
83
724
1,120
106
116
285
263
278
146
466
962
165
435
348
362
Nondetect set to 1A
detection limit
12
44
74
115
83
724
1,120
106
116
285
264
278
146
466
963
165
435
356
376
Mean emission factor,
scrubber outlet (ng/body)
Nondetect
set to zero
39
168
239
565
524
1,253
10,698
256
150
409
252
253
139
429
872
142
3,499
638
709
Nondetect set to 1A
detection limit
45
364
258
603
553
1,302
1,154
279
170
463
280
282
148
474
948
148
363
780
961
Source: U.S. EPA(1999a).
Only one or two facilities in the United States incorporate the use of an APCD, such as a
scrubber. Therefore, the inlet dioxin emission factors rather than the outlet dioxin emission
factors at the Woodlawn crematorium would be representative of a typical crematorium operating
in the United States (telephone conversation between Allen Krobath, CANA, and K. Riley,
Versar, Inc., February 12, 2003, and telephone conversation between Dale Walter, Matthews
Cremation, and K. Riley, Versar Inc., February 13, 2003).
In the previous inventory, an average emission factor of 17,000 ng I-TEQDF/body
(assuming nondetect values were zero) was developed, based on emission factors measured for
16 of the tested facilities, including the one at Camellia Memorial Lawn (CARB, 1990a), the 10
Japanese facilities (Takeda et al., 1998), the two Dutch facilities (Bremmer et al., 1994), the one
German facility (Jager et al., 1992), and the two British facilities (Mitchell and Loader, 1993).
The more recent data provided by Takeda et al. (2001) for the 17 Japanese facilities support the
emission factor of 17,000 ng I-TEQDF/body. However, an average emission factor developed
3-71
-------
using the data reported for the two U.S. crematoria (i.e., the outlet values for the Camellia
Memorial Lawn facility and the inlet values for the Woodlawn Cemetery facility) is 453 ng
TEQDF-WHO98/body (425 I-TEQDF/body cremated), assuming nondetect values were zero.
These values are two orders of magnitude less than the overall average calculated above. An
examination of the differences in U.S. and foreign operating practices may provide a rationale
for the large discrepancies.
Bremmer et al. (1994) reported an emission factor of 2,400 ng I-TEQDF/body for a Dutch
facility with a cold-type furnace and an emission factor of 4,900 ng I-TEQDF/body for another
Dutch facility with a warm-type furnace where flue gases were cooled to 220°C. Neither of the
U.S. facilities are considered to have cold-type furnaces. Additionally, the flue gases at the
Camellia Memorial Lawn crematorium were not cooled prior to exiting the furnace. At the
Woodlawn Cemetery facility, the flue gases were cooled from 681 to 860°C prior to entering the
scrubber to 271 to 354°C by the time they exited the scrubber and the furnace. The emissions
were higher at the scrubber outlet than at the inlet (961 vs. 325 ng TEQDF-WHO98/body [780 vs.
319 I-TEF/body]); however, the emissions were not of the same magnitude as those reported by
Bremmer for the warm-type facility (4,900 ng I-TEQDF/body). The Jager et al. (1992) report did
not include operating process information; therefore, the German facility could not be compared
with the U.S. facilities. Additionally, the emission values derived from the Mitchell and Loader
(1993) emission concentrations were calculated using gas generation rates from the Bremmer et
al. report and, as such, may not be indicative of crematoria in the United States.
In the Takeda et al. (1998, 2001) reports, the burn time for the cremations varied from 47
to 117 min. The average burn time in the U.S. studies was 120 min. This shorter burn time may
not be optimal for dioxin reduction, resulting in higher dioxin emissions. Also, the secondary
combustion chamber temperatures ranged from 250 to 950°C in the Takeda studies, again
resulting in higher emission rates. In fact, in Takeda et al. (2001) two of the three runs that had
the highest TEQ concentrations per body came from a crematorium that did not use a secondary
combustion chamber. Of the 31 crematoria sampled in Takeda et al. (2001), 26 had lower than
5,000 ng I-TEQDF/body.
Because the Woodlawn facility is unique in that it incorporates an APCD, the sample data
for the air stream entering the scrubber versus the stream exiting the scrubber should be analyzed.
A comparison of the dioxin concentrations of these air streams shows a significant increase in
dioxin concentrations in the stream exiting the scrubber. This increase can be attributed to the
decrease in temperature that occurred in the scrubber. Upon exiting the scrubber, the flue gas
temperatures were in the range of 271 to 354°C, compared with temperatures of between 681 and
860°C at the scrubber inlet. As discussed in Section 2, these exit flue gas temperatures lie in the
3-72
-------
optimum temperature range for dioxin formation; therefore, an increase in dioxin concentrations
would be expected.
An analysis of scrubber inlet dioxin data indicates that the average dioxin concentrations
increased with temperature (189, 445, and 503 ng TEQDF-WHO98/body at 681, 772, and 860°C,
respectively). Because the operating temperatures are outside the temperature range for the
formation of dioxin (200 to 400°C), dioxin concentrations should decrease as temperatures
increase. Further analysis of the data shows that as temperatures at the scrubber inlet increased,
so did concentrations of PM, HC1, and lead (Table 3-23). The data also indicate that oxygen
levels decreased as the temperature increased (U.S. EPA, 1999a). Given these data, one could
speculate that as the temperature increased, incomplete combustion conditions arose, leading to
an increase in dioxin formation.
Table 3-23. Operational data for the Woodlawn Cemetery crematorium,
scrubber inlet
Parameter
Paniculate matter (gr/dscf @ 7% O2)
Hydrochloric acid (Ib/hr)
Lead (g/hr)
Oxygen (%)
Mean value
675°C
0.015
0.053
0.1
9.9
870°C
0.033
0.14
0.32
8.6
980°C
0.068
0.26
0.59
7.5
Source: U.S. EPA (1999a).
Using data from U.S. crematoria, EPA recommends an average emission factor of 453 ng
TEQDF-WHO98/body (425 ng I-TEQDF/body). This is derived from the scrubber inlet dioxin
concentrations from the Woodlawn Cemetery study and the results from the Camellia Memorial
Lawn study. These average congener-specific emission are presented in Table 3-24, and the
CDD/CDF congener and congener group emission profiles are presented in Figure 3-17.
Because the emission factor was derived using emissions data from only 2 of 1,060 crematoria,
the average emission factor is assigned a low confidence rating.
3.4.1.2. Activity Level Information
A total of 323,371 cremations were performed in reference year 1987, 488,224 in 1995,
and 629,362 in 2000. A high confidence rating is assigned to these activity level estimates
because they are based on comprehensive data provided by CANA (CANA, 2006).
3-73
-------
Table 3-24. Congener-specific profile for the Camellia Memorial Lawn
crematorium and the Woodlawn Cemetery crematorium
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean facility emission factor (ng/body)
Nondetect set to zero
20
60.3
91
136
140
1,104
1,725.5
156
112
312
318.5
308
401.5
300.5
1,325.5
134.5
529.5
424.8
452.9
330
488
1,254.5
1,721.5
1,304.5
2,240.5
1,514.5
2,634
1,097.5
529.5
13,114.5
Nondetect set to 1A detection limit
20.5
66.8
91
136
140
1,104
1,725.5
156
116.5
317
319
308
401.5
300.5
1,388
138.5
529.5
431.9
463.3
330
488
1,254.5
1,721.5
1,304.5
2,240.5
1,514.5
2,634
1,097.5
529.5
13,114.5
Sources: CARB (1990a); U.S. EPA (1999a).
3-74
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
4D 2378
5D 12378
6D 123478
6D 123678
6D 123789
7D 1234678
8D
4F2378
5F 12378
5F 23478
6F 123478
6F 123678
6F 123789
6F 234678
7F 1234678
7F 1234789
8F
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2
Figure 3-17. Congener and congener group profiles for air emissions from
the Camellia Memorial Lawn crematorium and Woodlawn Cemetery
crematorium.
Sources: CARB (1990a); U.S. EPA (1999a).
3-75
-------
3.4.1.3. Emission Estimates
Combining the average emission rate of 453 ng WHO-TEQ98/body (425 ng I-
TEQDF/body) with the number of cremations in 1987, 1995, and 2000 (323,371; 488,224; and
629,362, respectively) yields an estimated annual release of 0.15 g TEQDF-WHO98 (0.14 g I-
TEQDF) in 1987, 0.22 g TEQDF-WHO98 (0.2 g I-TEQDF) in 1995, and 0.29 g TEQDF-WHO98 (0.27
g I-TEQDF) in 2000. An overall confidence rating of low was assigned to the emissions because
the emission factor had a low rating.
3.4.2. Animal Crematoria
3.4.2.1. Emissions Data
Only one study that measured CDD/CDF emissions from animal cremation could be
located. In 1999, CDD/CDF emissions from a newly installed animal incineration unit located at
the University of Georgia Veterinary School were measured (U.S. EPA, 2000a). The
incineration unit, which consists of a primary and a secondary combustion chamber, is used to
dispose of animals (mostly cows and horses) used in experimentation. Emissions are
uncontrolled, with the exception of an NFPA spark screen located at the stack outlet. Based on
four test runs, the averageTEQ emission factor was 0.12 TEQDF-WHO98/kg (0.11 ng I-TEQDF/kg)
of animal cremated. The average emission factors for these test runs are provided in Table 3-25
and a congener-specific profile based on these data is provided as Figure 3-18.
3.4.2.2. Activity Level Information
As part of the 2000 inventory, OAQPS calculated a national animal cremation activity
level estimate of 81.9 million kg/yr for reference year 2000. This estimate was scaled from the
1999 activity level estimate by applying the ratio of the 2000 national human population
(281,421,906) to the 1999 national human population (249,440,000). The 1999 national activity
level was based on 1990 data provided by OAQPS' Emission Standards Division. The 1999 and
2000 activity level estimates assume that animal mortality and cremation rates are constant and
that the animal population is directly proportional to human population.
3.4.2.3. Emission Estimates
Applying the TEQ emission factor of 0.12 ng TEQDF-WHO98/kg (0.11 ng I-TEQDF/kg) of
animal combusted to the activity level estimated by OAQPS (81.9 million kg/yr) yields estimated
annual emissions of 0.0098 g TEQDF-WHO98 (0.009 g I-TEQDF) in 2000. This estimate does not
include events such as the mass burning of animals affected by mad cow disease. These
estimates are based on extremely limited data and should be regarded as preliminary indications
of possible emissions from this source; further testing is needed to confirm the true magnitude of
the emissions.
3-76
-------
Table 3-25. Congener-specific profile for the University of Georgia
Veterinary School
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total CDD/CDF
Total I-TEQDF
Total TEQ™-WHOOS
Mean facility emission factor (ng/kg of animal)
Nondetect set to zero
7.51e-03
2.13e-02
4.46e-03
8.86e-03
7.17e-03
5.03e-03
l.Ole-03
1.79e-02
6.70e-03
1.41e-01
2.93e-02
1.85e-02
7.44e-02
2.35e-02
4.20e-03
3.16e-03
2.00e-04
0.37
0.11
0.12
Nondetect set to 1A detection limit
7.51e-03
2.13e-02
4.46e-03
8.86e-03
7.17e-03
5.03e-03
l.Ole-03
1.79e-02
6.70e-03
1.41e-01
2.93e-02
1.85e-02
7.44e-02
2.35e-02
4.20e-03
3.16e-03
2.00e-04
0.37
0.11
0.12
Source: U.S. EPA (2000a).
3.5. SEWAGE SLUDGE INCINERATION
The three principal combustion technologies used to incinerate sewage sludge in the
United States are multiple-hearth incineration, fluidized-bed incineration, and electric furnace
incineration (Brunner, 1992; U.S. EPA, 1995a). All of these technologies are excess-air
processes (i.e., they combust sewage sludge with oxygen in excess of theoretical requirements).
Approximately 80% of operating sludge incinerators are multiple-hearth design, about 20% are
fluidized-bed incinerators, and fewer than 1% are electric incinerators. Other types of furnaces
not widely used in the United States are single-hearth cyclones, rotary kilns, and high-pressure,
wet-air oxidation units (U.S. EPA, 1997a; e-mail dated July 13, 1998, from K. Maw, Pacific
Environmental Services, to G. Schweer, Versar, Inc.).
Multiple-hearth incinerators. These types of furnaces consist of refractory hearths
arranged vertically in series, one on top of the other. Dried sludge cake is fed to the top hearth of
the furnace. The sludge is mechanically moved from one hearth to another through the length of
3-77
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Figure 3-18. Congener profile for air emissions from the University of
Georgia animal crematorium.
Source: U.S. EPA (2000a).
the furnace. Moisture is evaporated from the sludge cake in the upper hearths. The center
hearths are the burning zone, where gas temperatures reach 871°C. The bottom hearths are the
burn-out zone, where the sludge solids become ash. A waste-heat boiler is usually included in
the burning zone, where steam is produced to provide supplemental energy at the sewage
treatment plant. Air pollution control measures typically include a venturi scrubber, an
impingement tray scrubber, or a combination of both. Wet cyclones and dry cyclones are also
used (U.S. EPA, 1995a).
Fluidized-bed incinerators. A fluidized-bed incinerator is a cylindrical refractory-lined
shell with a steel plate structure that supports a sand bed near the bottom of the furnace (Brunner,
1992). Air is introduced through openings in the bed plate supporting the sand. This causes the
sand bed to undulate in a turbulent air flow; hence, the sand appears to have a fluid motion when
observed through furnace portals. Sludge cake is added to the furnace at a position just above
this fluid motion of the sand bed. The fluid motion promotes mixing in the combustion zone.
Sludge ash exits the furnace with the combustion gases; therefore, air pollution control systems
3-78
-------
typically consist of high-energy venturi scrubbers or venturi/impingement tray combinations
(U.S. EPA, 1995a).
Electric furnaces. Also called infrared furnaces, these consist of a long, rectangular,
refractory-lined chamber. A belt conveyer system moves the sludge cake through the length of
the furnace. To promote combustion of the sludge, supplemental heat is added by electric
infrared heating elements located just above the traveling belt within the furnace. Electric power
is required to initiate and sustain combustion. Emissions are usually controlled with a venturi
scrubber or some other WS (Brunner, 1992; U.S. EPA, 1995a).
3.5.1. Emissions Estimates from Sewage Sludge Incinerators
EPA measured CDD/CDF emissions at three multiple-hearth incinerators as part of Tier 4
of the National Dioxin Survey (U.S. EPA, 1987a). During the pretest surveys, two of the
facilities were judged to have "average" potential and one facility was judged to have "high"
potential for CDD/CDF emissions with respect to other sewage sludge incinerators. The results
of these tests include congener group concentrations in stack gas but lack measurements for
specific congeners other than 2,3,7,8-TCDD and 2,3,7,8-TCDF. The results show a wide
variability in the emission factors at these three facilities; total CDD/CDF emission factors
ranged from 90 to 3,400 ng/kg (average of 1,266 ng/kg). Total TEQ emissions could not be
determined for these facilities because of the lack of congener-specific data.
In 1990, EPA measured CDD/CDF emissions (including all 17 toxic congeners) at
another multiple-hearth incinerator and also at a fluidized-bed incinerator (U.S. EPA, 1990d).
Assuming nondetects were zero, the total CDD/CDF emission factors for these two facilities
were 79 and 846 ng/kg, and the total average TEQ emission factors were 3.6 and 43.2 ng TEQDF-
WHO98/kg (2.4 and 43.3 ng I-TEQDF/kg) of dry sludge. In 1995, the Association of Metropolitan
Sewerage Agencies (AMSA) submitted to EPA the results of stack tests conducted at an
additional 13 sewage sludge incinerators (Green et al., 1995). Two of these data sets were
considered not usable by EPA because either DLs or feed rates and stack flow rates were not
provided. As with the EPA-tested facilities (U.S. EPA, 1987a, 1990d), wide variability was
observed in the emission factors for the 11 AMSA facilities. Assuming nondetects were zero,
total CDD/CDF emission factors ranged from 0 to 1,392 ng/kg (average of 217 ng/kg), and total
average TEQ emission factors ranged from 0 to 16 ng TEQDF-WHO98/kg (average, 3.47 ng) (3.46
ng I-TEQDF/kg) of dry sludge.
In 1999, stack tests were conducted at a multiple-hearth incinerator equipped with a
venturi scrubber and a three-tray impingement conditioning tower (U.S. EPA, 2000b). Four test
runs were conducted; however, the first test run was aborted, and the CDD/CDF results from the
fourth test run were determined to be statistical outliers (p>0.05). The back-half emission
3-79
-------
concentrations for test run 4 were 50 to 60% lower than those for test runs 2 and 3. Overall, total
CDD/CDF emissions measured during test run 4 were 48 ng/kg, whereas total CDD/CDF
emissions measured during test runs 2 and 3 were 120 and 116 ng/kg, respectively. It could not
be determined whether the lower concentrations associated with test run 4 were due to analyte
loss or whether they represented an accurate reflection of a change in incinerator emission
releases.
The average TEQ emission factor, excluding test run 4, was 3.28 ng TEQDF-WHO98/kg
(3.23 ng I-TEQDF/kg). The average TEQ emission factor based on the data for the 11 AMSA
facilities (Green et al., 1995) and the three facilities reported by EPA (U.S. EPA, 2000b, 1990d)
is 6.74 ng TEQDF-WHO98/kg (6.65 ng I-TEQDF/kg) of dry sludge, assuming nondetect values are
zero. Figure 3-19 presents the average congener and congener group profiles based on these
data. Additionally, Table 3-26 presents the average congener and congener-specific group
emission factors and the average TEQ emission factors for these facilities. Table 3-26 also
presents 2,3,7,8-TCDD, 2,3,7,8-TCDF, and congener-specific group emission factors for the
three facilities reported by EPA (U.S. EPA, 1987a).
Studies from other countries have reported similar results. Bremmer et al. (1994)
reported an emission rate of 5 ng I-TEQDF/kg for a fluidized-bed sewage sludge incinerator in the
Netherlands that was equipped with a cyclone and a WS. Cains and Dyke (1994) measured
CDD/CDF emissions at two sewage sludge incinerators in the United Kingdom. The emission
rate at an incinerator equipped with an ESP and a WS ranged from 2.75 to 28 ng I-TEQDF/kg.
The emission rate measured at a facility equipped with only an ESP was 43 ng I-TEQDF/kg.
In 1988, approximately 199 sewage sludge incineration facilities combusted abo 0.865
million metric tons of dry sewage sludge (Federal Register, 1993a). In 1995, approximately 257
sewage sludge incinerators (some of which were backup or alternate incinerators) combusted
about 2.11 million dry metric tons of sewage sludge (e-mail dated July 13, 1998, from K. Maw,
Pacific Environmental Services, to G. Schweer, Versar, Inc.). Using trends in wastewater flow
rates from the 1988 National Sewage Sludge Survey and the 1984 to 1996 Needs Surveys, EPA
estimated that in 2000 approximately 6.4 million metric tons of dry sewage sludge would be
generated (U.S. EPA, 1999b). Of this amount, EPA projected that 22% (1.42 million metric
tons) would be incinerated.
According to EPA, sewage sludge generation would increase to 6.9 million dry tons in
2005 and 7.4 million dry tons in 2010; however, the percentage of sewage sludge incinerated
will decrease slightly, to 20% in 2005 and 19% in 2010. EPA estimates that approximately 1.38
million metric tons of dry sewage sludge would be incinerated in 2005 and 1.41 million metric
tons will be incinerated in 2010. EPA believes that incineration as a disposal method for sewage
sludge will decrease as a result of increasing costs and public concerns about the environmental
and health impacts associated with incineration.
3-80
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2 0.25 0.3
4D2378
Ratio (congener group emission factor/total CDD/CDF emission factor)
0 0.1 0.2 0.3 0.4
Figure 3-19. Congener and congener group profiles for air emissions from
sewage sludge incinerators.
3-81
-------
Table 3-26. CDD/CDF emission factors for sewage sludge incinerators
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total CDD/CDF
Mean emission factor
(ng/kg)
for U.S. EPA (1987a)
(3 facilities)
Nondetect
set to zero
0.39
NR
NR
NR
NR
NR
46.2
179
NR
NR
NR
NR
NR
NR
NR
NR
109
37.6
2.66
16.6
53.9
46.2
528
253
75.4
144
109
NR
NR
1,266
Nondetect set to
1A detection
limit
0.44
NR
NR
NR
NR
NR
46.2
179
NR
NR
NR
NR
NR
NR
NR
NR
109
37.7
2.81
16.9
54
46.2
528
253
75.9
144
109
NR
NR
1,268
Mean emission factor (ng/kg) for
Green et al. (1995) (11 facilities)
U.S. EPA (1990d) (2 facilities)
U.S. EPA (1999b) (1 facility)
Nondetect set
to zero
0.16
0.22
0.04
0.12
0.29
2.46
12.78
25.41
1.92
6.47
2.11
0.77
0.03
1.22
1.46
0.17
1.17
35.8
1.11
1.74
4.39
12.78
123.85
59.94
12.69
2.63
1.17
6.65
6.74
256
Nondetect set to
1A detection limit
0.26
0.3
0.11
0.17
0.35
2.59
13.16
25.41
1.92
6.47
2.11
0.77
0.03
1.22
1.46
0.17
1.17
37.81
1.63
2.25
5.03
13.16
124.1
60.16
13.5
3.12
1.55
6.87
7.01
262
NR = Not reported
3-82
-------
A medium confidence rating is assigned to the average TEQ emission factor because it
was derived from stack testing at 14 U.S. sewage sludge incinerators. The 1988 activity level
estimate (used as a surrogate for the 1987 activity level) and the 2000 activity level estimate are
assigned a high confidence rating because they are based on extensive EPA surveys to support
rule-making activities. The 1995 activity level estimate is assigned a medium confidence rating
because assumptions were made for numerous facilities concerning hours of operation, operating
capacity, and design capacity.
Using the above estimated amounts of sewage sludge incinerated per year and the average
TEQ emission factor of 6.74 ng TEQDF-WHO98/kg (6.65 ng I-TEQDF/kg), the estimate of TEQ
emissions to air is 5.8 g TEQDF-WHO98 (5.8 g I-TEQDF) in 1987, 14.2 g TEQDF-WHO98 (14 g I-
TEQDF) in 1995, and 9.6 g TEQDF-WHO98/kg (9.4 g I-TEQDF/kg) in 2000. Because the emission
factor had a medium confidence rating, the overall emission estimates were assigned a medium
confidence rating for all years.
3.5.2. Solid Waste from Sewage Sludge Incinerators
In Table 5-16 of U.S. EPA (1987a), data are presented indicating that 2,3,7,8-TCDD was
not detected in the bottom ash or scrubber water filtrate from three sewage sludge incinerators.
However, total CDDs for the three incinerators and the filtrate were nondetects, 20 ng/kg, 10
ng/kg, and 0.3 ng/kg, respectively. For total CDFs, the respective values were nondetects, 70
ng/kg, 50 ng/kg, and 4 ng/kg. No data were given for any congeners (other than 2,3,7,8-TCDD),
nor were there any data on the quantities of ash or filtrate.
3.6. TIRE COMBUSTION
Most discarded tires are combusted in dedicated tire incinerators or cement kilns. Some
are combusted as auxiliary fuel in industrial boilers and in pulp and paper mill combustion
facilities. Additionally, tires may be unintentionally burned in an uncontrolled fashion at
landfills (open burning). This section addresses the total TEQ emissions that may result from the
combustion of tires in dedicated tire incinerators, industrial boilers, and pulp and paper mill
combustion facilities, but excludes cement kilns (addressed in Section 5.1). The open burning of
tires is not discussed in this report due to the lack of information.
Emissions of CDDs/CDFs from the incineration of discarded automobile tires were
measured at a dedicated tire incinerator tested by the California Air Resources Board (CARB,
1991). The facility consists of two excess air furnaces equipped with steam boilers to recoverthe
energy from the heat of combustion. Whole tires were fed to the incineration units at rates
ranging from 2,800 to 5,700 kg/hr during the three test days. The facility was equipped with a
DS and an FF for the control of emissions prior to exiting the stack. Table 3-27 presents the
congener-specific emission factors for this facility. Figure 3-20 presents CDD/CDF congener
3-83
-------
Table 3-27. CDD/CDF air emission factors for a tire combustion facility
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean facility emission factor (ng/kg)
Assuming nondetect set to zero
0.149
0.006
0.018
0.055
0.036
0.379
4.156
0.319
0.114
0.086
0.103
0.059
0.036
0.1
0.0
0.027
0.756
4.799
1.6
0.282
0.281
0.153
0.032
0.391
0.695
4.156
1.204
0.737
0.71
0.119
0.802
8.999
Assuming nondetect set to
1A detection limit
0.149
0.026
0.023
0.062
0.048
0.379
4.156
0.319
0.118
0.091
0.111
0.09
0.068
0.148
0.166
0.095
0.756
4.843
1.962
0.312
0.320
0.153
0.032
0.391
0.695
4.156
1.204
0.737
0.71
0.186
0.802
9.067
Source: CARB (1991).
3-84
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.1 0.2 0.3 0.4 0.5
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
,2,3,4,6,7,8,9-OCDF
Ratio (congener emission factor/total CDD/CDF emission factor)
0 0.1 0.2 0.3 0.4 0.5
Figure 3-20. Congener and congener group profiles for air emissions from a
tire combustor.
3-85
-------
and congener group profiles based on these TEQ emission factors. From these data, the average
emission factor is estimated to be 0.281 ng TEQDF-WHO98/kg (0.282 ng I-TEQDF/kg) of tires
incinerated (when all nondetect values are treated as zero). This emission factor was used to
estimate annual TEQ releases from the tire combustion source category for the years 1987, 1995,
and 2000.
EPA assigned a low confidence rating to the estimated TEQ emission factor because it is
possible that it is not representative of TEQ emissions from all tire combustion facilities. It is
also possible that this emission factor is an underestimation of emissions from this source
category because it was derived from the emissions of a facility equipped with very advanced air
pollution control technology specific for the control of dioxin emissions. These devices (DS/FF)
are capable of greater than 95% reduction and control of dioxin-like compounds prior to
discharge from the stack into the air. Because other facilities may not be equipped with similar
air pollution control systems, the TEQ emissions could be higher than the estimates shown
above. For example, Cains and Dyke (1994) reported much higher emission rates for two tire
incinerators in the United Kingdom that were equipped with only simple grit arresters. These
emissions produced emission factors of 188 and 228 ng I-TEQDF/kg of tires combusted.
EPA estimated that approximately 500 million kg of tires were combusted in 1990
(U.S. EPA, 1992b). Of this total, 23% (115 million kg) were combusted in cement kilns, and it
is assumed that the remaining 385 million kg were combusted in dedicated tire combustion
facilities, industrial boilers, and pulp and paper mill combustion facilities. This activity level
was adopted for the years 1987 and 1995 and is assigned a medium confidence rating.
The Rubber Manufacturers Association (2002) reported that 281 million scrap tires
weighing approximately 5.68 million metric tons were generated in the United States in 2001.
Approximately 115 million of these scrap tires were combusted as tire-derived fuel, or roughly
2.32 million metric tons (2.32 billion kg) of tires. Subtracting the 23% of the tires burned in
cement kilns yields a total of 1.8 billion kg of tires estimated to have been combusted in facilities
other than cement kilns in 2001. This figure is used to represent the activity level for tire
combustion in 2000. This activity level is assigned a medium confidence rating.
Annual emissions for the reference years were estimated by multiplying the activity level
times the TEQ emission factor. The TEQ emission factor of 0.281 ng TEQDF-WHO98/kg (0.282
ng I-TEQDF/kg) of tires combusted was used to estimate annual emissions for all years.
Multiplying the emission factor by the activity level (385 million kg of tires) yields an estimate
of 0.11 g TEQDF-WHO98/yr (0.11 g I-TEQDF/yr) emitted to the air in 1987 and 1995. Using the
same emission factor multiplied by the estimated activity level of 1.8 billion kg tires combusted
in 2000 gives an estimate of 0.51 g TEQDF-WHO98/yr (0.51 g I-TEQDF/yr). The estimated TEQ
3-86
-------
emissions to air from tire combustion for 1987, 1995, and 2000 are given a low confidence rating
because of the low confidence rating of the emission factor.
3.7. COMBUSTION OF WASTEWATER SLUDGE AT BLEACHED CHEMICAL PULP
MILLS
Approximately 20.5% of the wastewater sludges generated at bleached chemical pulp
mills are dewatered and burned in bark boilers at the mills. These sludges can contain
CDDs/CDFs and elevated levels of chloride. However, the level of heat input from sludge in the
mixed feed to bark boilers rarely exceeds 10% (NCASI, 1995).
NCASI (1995) provided congener-specific test results for four wood residue/sludge
boilers tested between 1987 and 1993. Sludge comprised 6 to 10% of the solids in the feed. The
average congener-specific emission factors derived from the stack test results obtained from
these facilities are presented in Table 3-28. The average TEQ emission factor derived from the
test results is 0.062 ng I-TEQDF-WHO98 (0.061 ng I-TEQDF/kg) of feed (i.e., sludge and wood
residue), assuming nondetect values were zero. The range in facility-specific emission factors
was wide (0.0004 to 0.118 ng I-TEQDF/kg, assuming nondetect values are zero).
NCASI (1995) also presented stack emission test results for five other bark boilers.
These boilers combusted only bark during the tests even though the boilers normally fire bark in
combination with sludge and coal. These boilers are discussed in Section 4.2.2 as industrial
facilities burning wood scrap/residues. The average TEQ emission factor for these facilities was
0.4 ng I-TEQDF/kg of feed. The emissions test data presented in NCASI (1995), and discussed
above, indicate that the CDD/CDF emission factors for bark/sludge combustors are similar to the
emission factor developed in Section 4.2.2 for industrial facilities burning only wood
residues/scrap. Based on the fact that wood residues comprise a far greater fraction of the feed to
these bark/sludge burners than does sludge, the national TEQ emission estimates derived in
Section 4.2.2 for industrial wood-burning facilities are assumed to include emissions from these
bark/sludge combustion units.
3.8. BIOGAS COMBUSTION
Using a specially developed sampling apparatus, Schreiner et al. (1992) measured the
CDD/CDF content of a flare combusting exhaust gases from an anaerobic sewage sludgedigestor
in Germany. The nozzle of the apparatus was moved through three cross-sections of the flame
and cooling zone. The CDD/CDF content was 1.4 pg I-TEQDF/standard cubic meter (Nm3) at the
bottom of the flare, 3.3. pg I-TEQDF/Nm3 at the top of the flare, and 13.1 pg I-TEQDF/Nm3 in the
middle of the flare. Congener-specific results were not reported. Using the theoretical ratio of
3-87
-------
Table 3-28. CDD/CDF emission factors for combustion of bleached-kraft
mill sludge in wood residue boilers
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total CDD/CDF
Mean emission factors
(ng/kg feed)
Nondetect set to zero
0.005
0.005
0.012
0.05
0.035
0.301
1.189
0.104
0.022
0.019
0.069
0.043
0.036
0.004
0.274
0.081
0.187
0.101
0.03
0.599
0.956
1.189
0.56
0.469
0.748
1.102
0.187
0.061
0.062
5.941
Nondetect set to
1A detection limit
0.013
0.012
0.022
0.056
0.043
0.302
1.192
0.107
0.029
0.027
0.071
0.046
0.041
0.012
0.275
0.083
0.188
0.108
0.109
0.6
0.958
1.192
0.56
0.47
0.748
1.102
0.188
0.082
0.087
6.035
Source: NCASI (1995).
flare gas volume to digestor gas volume combusted, 78.6:1, and the average CDD/CDF content
of the three measurements, 5.9 pg I-TEQDF/Nm3, yields an emission rate of 0.46 ng I-TEQDF/Nm3
of digestor gas combusted.
During 1996, publicly owned treatment works (POTWs) in the United States treated
approximately 122 billion L of wastewater daily (U.S. EPA, 1997b). Although reliable data are
-------
not readily available on the amount of sewage sludge generated by POTWs that is subjected to
stabilization by anaerobic digestion, a reasonable approximation is 25% of the total sludge
generated (i.e., the sludge generated from treatment of about 30 trillion L per day of wastewater).
An estimated 196 kg of sludge solids are generated for every 1 million L of wastewater subjected
to primary and secondary treatment (Water Pollution Control Federation, 1990). Thus,
multiplying 30 billion L/day (25% of 122 billion L) by 196 kg/million L and 365 days/yr yields
an annual estimate of 2 million metric tons of sludge solids that may be anaerobically digested in
POTWs annually.
The volume of sludge digestor gas combusted in flares annually can be estimated using
operation parameters for a "typical" anaerobic digestor system, as described in Water Pollution
Control Federation (1990). Multiplying the annual amount of sludge solids of 2 million metric
tons by the following parameters and appropriate conversion factors yields an annual flared
digestor gas volume of 467-million Nm3:
• Fraction of total solids that are volatile solids is 75%.
• Reduction of volatile solids during digestion is 50%.
• Specific gas production is 0.94 m3/kg volatile solids reduced.
• Fraction of produced gas that is flared is 66%.
Because there are no direct measurements of CDD/CDF emissions from U.S. anaerobic
sludge digestor flares and because of uncertainties about the activity level for biogas combustion,
no national emissions estimate has been developed for inclusion in the national inventory.
However, a preliminary estimate of the potential annual TEQ emissions from this source can be
obtained by multiplying the emission factor of 0.46 ng I-TEQDF/Nm3 of digestor gas flared by the
estimated volume of gas flared annually in the United States, 467 million Nm3. This calculation
yields an annual potential release in 2000 of 0.22 g. This estimate should be regarded as a
preliminary indication of possible emissions from this source category; further testing is needed
to confirm the true magnitude of these emissions.
3-89
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000
Facility name
State
County/city
MWC type
Unit
no.
MSW
capacity
(tons/day)
MSW
combusted
(tons/yr)
APCD
Emission
s(gl-
TEQ/yr)
Emissions
(gWHODF-
TEQ/yr)
I. Large MWCs (>250 tpd/unit)
Huntsville
Huntsville
Long Beach SERRF
Long Beach SERRF
Long Beach SERRF
Long Beach SERRF
Stanislaus (Modesto)
Stanislaus (Modesto)
Wheelabrator Bridgeport, L.P.
Wheelabrator Bridgeport, L.P.
Wheelabrator Bridgeport, L.P.
Bristol RRF
Bristol RRF
Mid-Connecticut RRF
Mid-Connecticut RRF
Mid-Connecticut RRF
Southeastern Connecticut RRF
Southeastern Connecticut RRF
Wheelabrator Lisbon Inc.
Wheelabrator Lisbon Inc.
Wheelabrator South Broward
Wheelabrator South Broward
Wheelabrator South Broward
AL
AL
CA
CA
CA
CA
CA
CA
CT
CT
CT
CT
CT
CT
CT
CT
CT
CT
CT
CT
FL
FL
FL
Madison
Madison
Los Angeles
Los Angeles
Los Angeles
Los Angeles
Stanislaus
Stanislaus
Fairfield
Fairfield
Fairfield
Hartford
Hartford
Hartford
Hartford
Hartford
New London
New London
New London
New London
Ft. Lauderdale
Ft. Lauderdale
Ft. Lauderdale
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
RDF
RDF
RDF
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
1
2
1
1
2
3
1
2
1
2
3
2
2
1
2
o
J
1
2
1
2
1
2
o
6
345
345
360
460
460
460
400
400
750
750
750
325
325
675
675
675
345
345
250
250
750
750
750
88,154.49
88,633.51
95,572.00
154,264.93
164,072.92
160,371.15
131,607.90
129,798.10
238,974.40
237,183.53
231,472.07
92,453.00
93,627.93
246,570.00
251,454.04
252,415.96
122,528.01
121,053.99
89,556.15
89,452.85
259,426.57
244,492.13
252,013.30
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI
SD/FF/CI
SD/FF/CI
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/CI
SD/FF/CI
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/SNCR
0.004
0.008
0.028
0.014
0.031
0.018
0.013
0.009
0.019
0.02
0.024
0.004
0.004
0.019
0.041
0.048
0.015
0.046
0.001
0.025
0.023
0.078
0.046
0.004
0.008
0.031
0.015
0.033
0.023
0.014
0.01
0.021
0.021
0.027
0.004
0.004
0.022
0.046
0.048
0.016
0.051
0.001
0.028
0.025
0.085
0.051
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Wheelabrator North Broward
Wheelabrator North Broward
Wheelabrator North Broward
Dade Co. RRF
Dade Co. RRF
Dade Co. RRF
Dade Co. RRF
Hillsborough Co. RRF
Hillsborough Co. RRF
Hillsborough Co. RRF
McKay Bay REF
McKay Bay REF
McKay Bay REF
McKay Bay REF
Lake Co. RRF
Lake Co. RRF
Lee County Solid Waste RRF
Lee County Solid Waste RRF
Pasco County Solid Waste RRF
Pasco County Solid Waste RRF
Pasco County Solid Waste RRF
Pinellas County RRF
Pinellas County RRF
Pinellas County RRF
State
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
County/city
Broward
Broward
Broward
Dade
Dade
Dade
Dade
Hillsborough
Hillsborough
Hillsborough
Hillsborough
Hillsborough
Hillsborough
Hillsborough
Lake
Lake
Lee
Lee
Pasco
Pasco
Pasco
Pinellas
Pinellas
Pinellas
MWC type
MB/WW
MB/WW
MB/WW
RDF
RDF
RDF
RDF
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
Unit
no.
1
2
o
5
i
2
3
4
1
2
3
1
2
3
4
1
2
1
2
1
2
3
1
2
3
MSW
capacity
(tons/day)
750
750
750
672
672
672
672
400
400
400
250
250
250
250
264
264
600
600
350
350
350
1000
1000
1000
MSW
combusted
(tons/yr)
260,456.43
255,371.04
259,802.53
172,792.98
171,880.39
167,673.72
155,352.91
116,426.35
120,265.74
121,674.90
90,232.04
90,232.04
90,413.44
90,050.47
82,586.03
83,262.97
197,620.93
197,203.07
98,891.97
102,727.33
108,759.70
299,275.82
289,763.18
301,717.00
APCD
SD/FF/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
Emission
s(gl-
TEQ/yr)
0.018
0.018
0.07
0.384
0.393
0.062
0.444
0.019
0.023
0.039
0.014
0.016
0.002
0.003
0.034
0.036
0.101
0.11
0.012
0.008
0.004
0.013
0.074
0.043
Emissions
(gWHODF-
TEQ/yr)
0.02
0.02
0.076
0.406
0.419
0.066
0.475
0.02
0.025
0.042
0.015
0.017
0.002
0.003
0.036
0.038
0.11
0.12
0.012
0.009
0.004
0.015
0.083
0.047
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
North County RRF
North County RRF
Savannah RRF
Savannah RRF
Honolulu RRF
Honolulu RRF
Indianapolis RRF
Indianapolis RRF
Indianapolis RRF
Haverhill RRF
Haverhill RRF
Wheelabrator North Andover
Wheelabrator North Andover
Wheelabrator Saugus
Wheelabrator Saugus
SEMASS RRF
SEMASS RRF
SEMASS RRF
Wheelabrator Millbury
Wheelabrator Millbury
Wheelabrator Baltimore
State
FL
FL
GA
GA
HI
HI
IN
IN
IN
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
County/city
West Palm
Beach
West Palm
Beach
Chatham
Chatham
Honolulu
Honolulu
Marion
Marion
Marion
Essex
Essex
Essex
Essex
Essex
Essex
Plymouth
Plymouth
Plymouth
Worcester
Worcester
Independent
City
MWC type
RDF
RDF
MB/WW
MB/WW
RDF
RDF
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
RDF
RDF
RDF
MB/WW
MB/WW
MB/WW
Unit
no.
1
2
1
2
1
2
1
2
o
J
1
2
1
2
1
2
1
2
3
1
2
1
MSW
capacity
(tons/day)
1000
1000
250
250
1080
1080
787
787
787
825
825
750
750
750
750
1000
1000
1000
750
750
750
MSW
combusted
(tons/yr)
273,939.00
288,988.00
61,886.44
58,830.56
258,726.76
256,019.24
209,507.33
220,779.62
224,095.05
281,121.60
286,516.40
202,456.92
180,415.08
217,656.46
214,057.54
368,821.00
372,224.92
364,595.08
230,829.25
233,033.75
242,224.16
APCD
SD/ESP
SD/ESP
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/ESP
SD/ESP
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/ESP/FF/CI
SD/ESP/FF/CI
SD/FF/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
Emission
s(gl-
TEQ/yr)
0.279
0.681
0.029
0.023
0.802
1.181
0.025
0.025
0.005
0.048
0.063
0.006
0.006
0.006
0.004
0.045
0.048
0.033
0.038
0.017
0.043
Emissions
(gWHODF-
TEQ/yr)
0.302
0.747
0.032
0.025
0.888
1.318
0.027
0.028
0.005
0.051
0.067
0.007
0.006
0.006
0.005
0.051
0.052
0.036
0.042
0.019
0.047
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Wheelabrator Baltimore
Wheelabrator Baltimore
Montgomery Co. RRF
Montgomery Co. RRF
Montgomery Co. RRF
Greater Portland Region RRF
Greater Portland Region RRF
Penobscot Energy Recovery
Penobscot Energy Recovery
Maine Energy Recovery
Maine Energy Recovery
Kent Co. WTE Facility
Kent Co. WTE Facility
Central Wayne Energy
Greater Detroit RRF
Greater Detroit RRF
Greater Detroit RRF
Wilmarth Plant
Wilmarth Plant
Red Wing Plant
Red Wing Plant
Covanta Hennepin
Covanta Hennepin
State
MD
MD
MD
MD
MD
ME
ME
ME
ME
ME
ME
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
County/city
Independent
City
Independent
City
Montgomery
Montgomery
Montgomery
Cumberland
Cumberland
Penobscot
Penobscot
York
York
Kent
Kent
Wayne
Wayne
Wayne
Wayne
Blue Earth
Blue Earth
Goodhue
Goodhue
Hennepin
Hennepin
MWC type
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
RDF
RDF
RDF
RDF
MB/WW
MB/WW
MB/WW
RDF
RDF
RDF
RDF
RDF
RDF
RDF
MB/WW
MB/WW
Unit
no.
2
3
1
2
3
1
2
1
2
1
2
1
2
3
1
2
o
6
i
2
1
2
1
2
MSW
capacity
(tons/day)
750
750
600
600
600
250
250
360
360
300
300
313
313
300
1100
1100
1100
360
360
360
360
600
600
MSW
combusted
(tons/yr)
241,753.99
231,232.85
178,986.24
170,879.15
170,131.61
87,323.81
85,676.19
108,975.08
110,899.92
123,654.50
123,654.50
90,813.23
88,925.77
61,800.00
279,883.11
212,094.32
201,288.58
94,983.00
108,339.00
92,993.00
89,211.00
182,903.45
182,190.55
APCD
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/FF
SD/FF
SD/FF/SNCR
SD/FF/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF
SD/FF
SD/FF
SD/FF/SNCR
SD/FF/SNCR
DSI/FF
DSI/FF
SD/FF/CI/SNCR
SD/FF/CI/SNCR
Emission
s(gl-
TEQ/yr)
0.176
0.177
0.096
0.09
0.099
0.051
0.108
0.008
0.008
0.018
0.019
0.006
0.013
0.005
0.259
0.01
0.193
0.028
0.032
0.017
0.066
0.009
0.049
Emissions
(gWHODF-
TEQ/yr)
0.179
0.186
0.107
0.091
0.108
0.055
0.115
0.008
0.008
0.02
0.02
0.007
0.013
0.005
0.287
0.011
0.218
0.03
0.034
0.017
0.069
0.01
0.052
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Great River Energy
Great River Energy
Great River Energy
New Hanover Co. WTE
Wheelabrator Concord
Wheelabrator Concord
Camden RRF
CamdenRRF
Camden RRF
Essex Co. RRF
Essex Co. RRF
Essex Co. RRF
Gloucester County
Gloucester County
Union Co. RRF
Union Co. RRF
Union Co. RRF
Hempstead
Hempstead
Hempstead
Niagara Falls
Niagara Falls
Onondaga Co. RRF
Onondaga Co. RRF
State
MN
MN
MN
NC
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NY
NY
NY
NY
NY
NY
NY
County/city
Sherburne
Sherburne
Sherburne
New Hanover
Merrimack
Merrimack
Camden
Camden
Camden
Essex
Essex
Essex
Gloucester
Gloucester
Union
Union
Union
Nassau
Nassau
Nassau
Niagara
Niagara
Onondaga
Onondaga
MWC type
RDF
RDF
RDF
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
Unit
no.
1
2
o
5
3
1
2
1
2
o
J
1
2
3
1
2
1
2
3
1
2
3
3
4
1
2
MSW
capacity
(tons/day)
250
250
500
301
250
250
350
350
350
900
900
900
288
288
480
480
480
835
835
835
1100
1100
330
330
MSW
combusted
(tons/yr)
75,878.19
75,122.55
133,529.26
126,709.00
91,065.04
92,550.96
97,654.44
93,330.56
85,658.00
322,862.32
327,332.25
335,288.43
90,774.74
90,397.26
169,630.18
171,047.55
167,844.27
294,685.61
299,122.26
293,532.12
358,183.15
355,338.85
117,389.13
99,956.76
APCD
SD/FF
SD/FF
SD/FF
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/ESP/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
Emission
s(gl-
TEQ/yr)
0.013
0.013
0.022
0.016
0.001
0.006
0.098
0.218
0.062
0.07
0.049
0.039
0.008
0.009
0.005
0.01
0.01
0.015
0.431
0.011
0.209
0.197
0.002
0.004
Emissions
(gWHODF-
TEQ/yr)
0.014
0.013
0.024
0.017
0.001
0.007
0.106
0.252
0.068
0.078
0.054
0.043
0.009
0.01
0.005
0.011
0.011
0.017
0.471
0.012
0.225
0.214
0.002
0.004
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Onondaga Co. RRF
Babylon RRF
Babylon RRF
Huntington RRF
Huntington RRF
Huntington RRF
Adirondack RRF
Adirondack RRF
Wheelabrator Westchester, L.P.
Wheelabrator Westchester, L.P.
Wheelabrator Westchester, L.P.
Walter B. Hall RRF (Tulsa)
Walter B. Hall RRF (Tulsa)
Walter B. Hall RRF (Tulsa)
Marion Co. WTE
Marion Co. WTE
Lancaster County
Lancaster County
Lancaster County
Wheelabrator Falls RRF
Wheelabrator Falls RRF
American Ref-fuel of Delaware
Valley
American Ref-fuel of Delaware
Valley
State
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OK
OK
OK
OR
OR
PA
PA
PA
PA
PA
PA
PA
County/city
Onondaga
Suffolk
Suffolk
Suffolk
Suffolk
Suffolk
Washington
Washington
Westchester
Westchester
Westchester
Tulsa
Tulsa
Tulsa
Marion
Marion
Bainbridge
Bainbridge
Bainbridge
Bucks
Bucks
Delaware
Delaware
MWC type
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/RC
MB/RC
Unit
no.
o
J
1
2
1
2
3
1
2
1
2
3
1
2
3
1
2
1
2
3
1
2
1
2
MSW
capacity
(tons/day)
330
375
375
250
250
250
250
250
750
750
750
375
375
375
275
275
400
400
400
750
750
448
448
MSW
combusted
(tons/yr)
117,794.12
112,323.03
107,951.97
102,526.89
102,744.63
110,853.48
82,458.60
79,860.40
226,329.03
217,185.24
206,044.73
111,998.10
113,501.52
113,849.38
92,091.75
92,288.25
125,595.02
127,240.06
128,049.91
264,839.58
259,729.42
179,845.19
190,795.79
APCD
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/ESP/CI
SD/ESP/CI
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF
SD/FF
Emission
s(gl-
TEQ/yr)
0.007
0.002
0.007
0.003
0.007
0.007
0.509
0.437
0.025
0.017
0.026
0.012
0.006
0.003
0.004
0.003
0.017
0.015
0.028
0.015
0.025
0.126
0.052
Emissions
(gWHODF-
TEQ/yr)
0.007
0.002
0.008
0.003
0.008
0.008
0.556
0.477
0.027
0.019
0.029
0.013
0.006
0.003
0.005
0.003
0.018
0.016
0.029
0.016
0.027
0.138
0.055
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
American Ref-fuel of Delaware
Valley
American Ref-fuel of Delaware
Valley
American Ref-fuel of Delaware
Valley
American Ref-fuel of Delaware
Valley
Montenay Montgomery
Montenay Montgomery
York County
York County
York County
Foster Wheeler Charleston RRF
Foster Wheeler Charleston RRF
Nashville Thermal Transfer Corp.
Nashville Thermal Transfer Corp.
Nashville Thermal Transfer Corp.
Alexandria/ Arlington RRF
Alexandria/ Arlington RRF
Alexandria/ Arlington RRF
1-95 Energy RRF
1-95 Energy RRF
1-95 Energy RRF
1-95 Energy RRF
State
PA
PA
PA
PA
PA
PA
PA
PA
PA
SC
SC
TN
TN
TN
VA
VA
VA
VA
VA
VA
VA
County/city
Delaware
Delaware
Delaware
Delaware
Montgomery
Montgomery
York
York
York
Charleston
Charleston
Davidson
Davidson
Davidson
Alexandria
Alexandria
Alexandria
Fairfax
Fairfax
Fairfax
Fairfax
MWC type
MB/RC
MB/RC
MB/RC
MB/RC
MB/WW
MB/WW
MB/RC
MB/RC
MB/RC
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
Unit
no.
3
4
5
6
1
2
1
2
3
1
2
1
2
o
J
1
2
3
1
2
3
4
MSW
capacity
(tons/day)
448
448
448
448
600
600
448
448
448
300
300
330
330
390
325
325
325
750
750
750
750
MSW
combusted
(tons/yr)
186,088.66
189,206.50
176,939.66
191,389.21
204,290.69
198,884.31
135,266.80
135,104.31
126,412.88
112,950.14
98,453.86
67,181.87
72,257.44
85,979.69
113,578.13
109,442.21
109,410.66
270,077.99
271,967.16
272,396.24
271,628.61
APCD
SD/FF
SD/FF
SD/FF
SD/FF
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/ESP/CI
SD/ESP/CI
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
SD/FF/CI/SNCR
Emission
s(gl-
TEQ/yr)
0.181
0.045
0.023
0.029
0.169
0.354
0.023
0.025
0.023
0.151
0.15
0
0
0
0.007
0.022
0.02
0.011
0.012
0.006
0.012
Emissions
(gWHODF-
TEQ/yr)
0.197
0.05
0.025
0.032
0.179
0.372
0.025
0.028
0.025
0.16
0.16
0
0
0
0.008
0.024
0.021
0.011
0.013
0.006
0.012
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
SPSA Waste To Energy
SPSA Waste To Energy
SPSA Waste To Energy
SPSA Waste To Energy
Spokane Regional Disposal Facility
Spokane Regional Disposal Facility
LaCrosse Co.
LaCrosse Co.
Large MWC totals
State
VA
VA
VA
VA
WA
WA
WI
WI
County/city
Portsmouth
Portsmouth
Portsmouth
Portsmouth
Spokane
Spokane
LaCrosse
LaCrosse
MWC type
RDF
RDF
RDF
RDF
MB/WW
MB/WW
RDF
RDF
Unit
no.
1
2
o
5
4
1
2
1
2
MSW
capacity
(tons/day)
500
500
500
500
400
400
288
288
MSW
combusted
(tons/yr)
122,153.75
122,153.75
122,153.75
122,153.75
141,392.12
143,927.88
22,727.86
21,718.14
2.80e+007
APCD
SD/FF
SD/FF
SD/FF
SD/FF
SD/FF/CI/SNCR
SD/FF/CI/SNCR
DSI/FF/H2O/SNCR
DSI/FF/H2O/SNCR
Emission
s(gl-
TEQ/yr)
0.123
0.111
0.103
0.13
0.009
0.008
0.344
0.344
12.73
Emissions
(gWHODF-
TEQ/yr)
0.133
0.12
0.111
0.141
0.01
0.009
0.36
0.36
13.815
II. Small MWCs (<250 tpd/unit)
Juneau RRF
Juneau RRF
Wallingford RRF
Wallingford RRF
Wallingford RRF
Bay Resource Management Center
Bay Resource Management Center
Miami International Airport
Southernmost WTE
Southernmost WTE
PittsfieldRRF
PittsfieldRRF
Pittsfield RRF
AK
AK
CT
CT
CT
FL
FL
FL
FL
FL
MA
MA
MA
Juneau
Borough
Juneau
Borough
New Haven
New Haven
New Haven
Bay
Bay
Dade
Monroe
Monroe
Berkshire
Berkshire
Berkshire
MOD/SA
MOD/SA
MOD/EA
MOD/EA
MOD/EA
MB/RC
MB/RC
MOD/SA
MB/WW
MB/WW
MOD/EA
MOD/EA
MOD/EA
1
2
1
2
o
J
1
2
1
1
2
1
2
3
35
35
140
140
140
245
245
60
75
75
120
120
120
10,975.85
10,975.85
43,903.39
43,903.39
43,903.39
76,830.94
76,830.94
18,815.74
23,519.67
23,519.67
37,631.48
37,631.48
37,631.48
ESP
ESP
SD/FF
SD/FF
SD/FF
ESP
ESP
Unc
ESP
ESP
WS/ESP
WS/ESP
WS/ESP
0.484
0.484
0.064
0.021
0.006
3.442
4.633
0.649
0.322
0.322
0.978
0.978
0.978
0.533
0.533
0.07
0.023
0.007
3.786
5.096
0.714
0.355
0.355
1.076
1.076
1.076
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Springfield RRF
Springfield RRF
Springfield RRF
Harford Co. WTE Fac.
Harford Co. WTE Fac.
Harford Co. WTE Fac.
Harford Co. WTE Fac.
Mid Maine Waste Action Corp.
Mid Maine Waste Action Corp.
Jackson Co. RRF
Jackson Co. RRF
Central Wayne Co.
Central Wayne Co.
Pope-Douglas Waste
Pope-Douglas Waste
Red Wing Solid Waste Boiler
Facility
Red Wing Solid Waste Boiler
Facility
Olmstead WTE Facility
Olmstead WTE Facility
Fergus Falls
Fergus Falls
Perham Renewable RF
Perham Renewable RF
State
MA
MA
MA
MD
MD
MD
MD
ME
ME
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
County/city
Hampden
Hampden
Hampden
Harford
Harford
Harford
Harford
Androscoggin
Androscoggin
Jackson
Jackson
Wayne
Wayne
Douglas
Douglas
Goodhue
Goodhue
Olmstead
Olmstead
Otter Tail
Otter Tail
Otter Tail
Otter Tail
MWC type
MOD/EA
MOD/EA
MOD/EA
MOD/SA
MOD/SA
MOD/SA
MOD/SA
MB
MB
MB/WW
MB/WW
RDF
RDF
MOD/EA
MOD/EA
MOD/EA
MOD/EA
MB/WW
MB/WW
MOD/SA
MOD/SA
MOD/SA
MOD/SA
Unit
no.
1
2
o
5
i
2
3
4
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
MSW
capacity
(tons/day)
120
120
120
90
90
90
90
100
100
100
100
249
249
36
36
36
36
100
100
47
47
57
57
MSW
combusted
(tons/yr)
37,631.48
37,631.48
37,631.48
28,223.61
28,223.61
28,223.61
28,223.61
31,359.57
31,359.57
31,359.57
31,359.57
78,085.32
78,085.32
11,289.44
11,289.44
11,289.44
11,289.44
31,359.57
31,359.57
14,739.00
14,739.00
17,874.95
17,874.95
APCD
DSI/FF
DSI/FF
DSI/FF
DSI/ESP
DSI/ESP
DSI/ESP
DSI/ESP
SD/FF
SD/FF
SD/FF
SD/FF
ESP
ESP
ESP
ESP
ESP
ESP
ESP
ESP
WS
WS
ESP
ESP
Emission
s(gl-
TEQ/yr)
0.028
0.028
0.028
1.352
1.352
1.352
1.352
0.024
0.024
1.135
1.135
0.061
0.061
0.293
0.293
0.287
0.287
1.449
1.449
0.621
0.621
0.37
0.37
Emissions
(gWHODF-
TEQ/yr)
0.03
0.03
0.03
1.488
1.488
1.488
1.488
0.027
0.027
1.249
1.249
0.067
0.067
0.323
0.323
0.316
0.316
1.594
1.594
0.683
0.683
0.407
0.407
U)
JD
oo
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Polk Co.
Polk Co.
Pascagoula
Pascagoula
Livingston/Park County
Livingston/Park County
New Hanover Co.
New Hanover Co.
SES Claremont
SES Claremont
Warren Energy RF
Warren Energy RF
Dutchess Co. RRF
Dutchess Co. RRF
Oswego Co. WTE
Oswego Co. WTE
Oswego Co. WTE
Oswego Co. WTE
MacArthur WTE
MacArthur WTE
Miami RRF
Miami RRF
Miami RRF
Coos Bay Incinerator
State
MN
MN
MS
MS
MT
MT
NC
NC
NH
NH
NJ
NJ
NY
NY
NY
NY
NY
NY
NY
NY
OK
OK
OK
OR
County/city
Polk
Polk
Jackson
Jackson
Park
Park
New Hanover
New Hanover
Sullivan
Sullivan
Warren
Warren
Dutchess
Dutchess
Oswego
Oswego
Oswego
Oswego
Suffolk
Suffolk
Ottawa
Ottawa
Ottawa
Coos
MWC type
MOD/SA
MOD/SA
MOD/EA
MOD/EA
MOD/SA
MOD/SA
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/WW
MB/RC
MB/RC
MOD/SA
MOD/SA
MOD/SA
MOD/SA
MB/RC
MB/RC
MOD/SA
MOD/SA
MOD/SA
MOD/SA
Unit
no.
1
2
1
2
1
2
1
2
1
2
1
2
1
2
o
J
4
1
2
1
2
1
2
o
J
1
MSW
capacity
(tons/day)
40
40
75
75
36
36
100
100
100
100
200
200
200
200
50
50
50
50
243
243
35
35
35
50
MSW
combusted
(tons/yr)
12,543.83
12,543.83
23,519.67
23,519.67
11,289.44
11,289.44
31,359.57
31,359.57
31,359.57
31,359.57
62,719.13
62,719.13
62,719.13
62,719.13
15,679.78
15,679.78
15,679.78
15,679.78
76,203.74
76,203.74
10,975.85
10,975.85
10,975.85
15,679.78
APCD
ESP
ESP
ESP
ESP
Unc
Unc
SD/FF
SD/FF
DSI/FF
DSI/FF
SD/FF/CI
SD/FF/CI
DSI/FF
DSI/FF
SD/FF/CI
SD/FF/CI
SD/FF/CI
SD/FF/CI
DSI/FF
DSI/FF
Unc
Unc
Unc
DSI/FF
Emission
s(gl-
TEQ/yr)
0.433
0.433
0.611
0.611
0.389
0.389
0.024
0.024
0.113
0.095
0.001
0.001
0.015
0.027
0.007
0.024
0.015
0.015
0.001
0.001
0.379
0.379
0.379
0
Emissions
(gWHODF-
TEQ/yr)
0.476
0.476
0.672
0.672
0.428
0.428
0.027
0.027
0.124
0.104
0.002
0.002
0.016
0.029
0.008
0.026
0.017
0.017
0.001
0.001
0.417
0.417
0.417
0
U)
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Coos Bay Incinerator
Coos Bay Incinerator
Harrisburg WTE
Harrisburg WTE
Sumner Co.
Suniner Co.
City of Cleburne
City of Cleburne
City of Cleburne
Panola Co. WTE
Center RRF
Davis/Wasatch
Davis/Wasatch
Arlington -pen tagon
Arlington -pen tagon
Galax City SW
NASA Refuse-fired Steam
Generator
NASA Refuse-fired Steam
Generator
Harrisonburg
Harrisonburg
Tacoma
Tacoma
Barren Co.
State
OR
OR
PA
PA
TN
TN
TX
TX
TX
TX
TX
UT
UT
VA
VA
VA
VA
VA
VA
VA
WA
WA
WI
County/city
Coos
Coos
Dauphin
Dauphin
Sumner
Sumner
Johnson
Johnson
Johnson
Panola
Shelby
Davis
Davis
Arlington
Arlington
Gray son
Hampton City
Hampton City
Rockingham
Rockingham
Pierce
Pierce
Barron
MWC type
MOD/SA
MOD/SA
MB/WW
MB/WW
MB/RC
MB/RC
MOD/SA
MOD/SA
MOD/SA
MOD/SA
MOD/SA
MB/REF
MB/REF
MOD/SA
MOD/SA
MB/WW/RC
MB/WW
MB/WW
MB/WW
MB/WW
FB/RDF
FB/RDF
MOD/SA
Unit
no.
2
o
J
1
2
1
2
1
2
3
1
1
1
2
1
2
1
1
2
1
2
1
2
1
MSW
capacity
(tons/day)
50
50
245
245
100
100
38
38
38
40
40
200
200
50
50
56
100
100
50
50
150
150
50
MSW
combusted
(tons/yr)
15,679.78
15,679.78
76,830.94
76,830.94
31,359.57
31,359.57
11,916.63
11,916.63
11,916.63
12,543.83
12,543.83
62,719.13
62,719.13
15,679.78
15,679.78
17,561.36
31,359.57
31,359.57
15,679.78
15,679.78
47,039.35
47,039.35
15,679.78
APCD
DSI/FF
DSI/FF
ESP
ESP
ESP
ESP
DSI/ESP
DSI/ESP
DSI/ESP
WS
WS
DSI/ESP
DSI/ESP
FF
FF
FF
ESP
ESP
Unc
Unc
DSI/FF
DSI/FF
ESP
Emission
s(gl-
TEQ/yr)
0
0
12.894
8.29
0.43
0.43
0.526
0.526
0.411
0.433
0.554
1.275
1.384
0.697
0.697
0.241
0.614
0.614
0.151
0.151
0.089
0.089
0.692
Emissions
(gWHODF-
TEQ/yr)
0
0
14.184
9.119
0.473
0.473
0.579
0.579
0.452
0.476
0.609
1.403
1.522
0.766
0.766
0.265
0.675
0.675
0.166
0.166
0.098
0.098
0.761
o
o
-------
Table 3-5. National CDD/CDF TEQ emissions (g/yr) for large and small municipal waste combustors (MWCs)
operating in 2000 (continued)
Facility name
Barren Co.
Small MWC totals
All MWCs operating in 2000
(totals)
State
WI
County/city
Barren
MWC type
MOD/SA
Unit
no.
2
MSW
capacity
(tons/day)
50
MSW
combusted
(tons/yr)
15,679.78
2.60e+006
3.06e+007
APCD
ESP
Emission
s(gl-
TEQ/yr)
0.692
63.583
76.3
Emissions
(gWHODF-
TEQ/yr)
0.761
69.941
83.8
RRF = Resource recovery facility (steam-generating facility)
MSW = Municipal solid waste
WTE = Waste-to-energy (facility)
APCD (air pollution control device):
CI = Carbon injection
DS = Dry scrubber
DSI = Dry sorbent injection
ESP = Electrostatic precipitator
FF = Fabric filter
SD = Spray dryer
SNRL = Selective noncatalytic reduction
Unc = Uncontrolled
WS = Wet scrubber
MWC type:
FB/RDF = Fluidized-bed refuse-derived fuel
MB/REF = Mass burn refractory
MB/WW = Mass burn waterwall
MB = Mass burn
MB/RC = Mass burn rotary kiln
MOD/EA = Modular excess air
MOD/SA = Modular starved air
RDF = Refuse-derived fuel
-------
4. COMBUSTION SOURCES OF CDDs/CDFs: POWER/ENERGY GENERATION
4.1. MOTOR VEHICLE FUEL COMBUSTION
Ballschmiter et al. (1986) reported detecting CDDs/CDFs in used motor oil, thus
providing some of the first evidence that CDDs/CDFs might be emitted by the combustion
processes in gasoline- and diesel-fueled engines. Incomplete combustion and the presence of a
chlorine source in the form of additives such as dichloroethane or pentachlorophenate in the oil
or the fuel were speculated to lead to the formation of CDDs/CDFs. The congener patterns found
in the used oil samples were characterized by Ballschmiter et al. as being similar to the patterns
found in fly ash and stack emissions from municipal waste incinerators.
Since 1986, several studies have been conducted to measure or estimate CDD/CDF
concentrations in emissions from vehicles. Although there is no standard approved protocol for
measuring CDDs/CDFs in vehicle exhaust, some researchers have developed and implemented
several approaches for collecting and analyzing tailpipe emissions. Other researchers have
estimated vehicle exhaust emissions of CDDs/CDFs indirectly from studies of tunnel air. The
results of these two types of studies are summarized in Sections 4.1.1 and 4.1.2.
Estimates of national annual CDD/CDF TEQ emissions from on-road and off-road motor
vehicles fueled with leaded gasoline, unleaded gasoline, and diesel fuel based on the results of
those studies are presented in Section 4.1.3. It should be noted, however, that relatively few tests
on emissions from diesel- and unleaded gasoline-fueled vehicles are available, considering the
variety and number of such vehicles currently in operation and the range of operational,
technical, and environmental conditions in which they are operated. As a result, the emission
factors developed in this report for on-road and off-road motor vehicles are quite uncertain.
4.1.1. Tailpipe Emission Studies
Marklund et al. (1987) provided the first direct evidence of the presence of CDDs/CDFs
in car exhaust by measuring emissions from tailpipes of Swedish cars. Approximately 20 to 220
pg I-TEQDF/km driven from tetra- and penta-CDDs/CDFs were reported for four cars running on
leaded gasoline. For this study, an unleaded gasoline was used, with tetramethyl lead (0.15 g/L
[0.57 g/gal]) and 1,2-dichloroethane (0.1 g/L as a scavenger) added. The fuel used may not have
accurately represented commercial fuels at that time, which typically contained a mixture of
chlorinated and brominated scavengers (Marklund et al., 1990). Also, the lead content of the fuel
used (0.15 g lead/L) was the normal content for Swedish fuels at the time, but it was higher than
that of leaded gasoline in the United States during the late 1980s (lowered to 0.1 g lead/gal
[0.026 g lead/L] effective January 1, 1986). The authors reported a striking similarity between
the TCDF and PeCDF congener profiles in the car exhausts and those found in emissions from
municipal waste incinerators. For two cars running on unleaded gasoline, CDD/CDF emissions
4-1
-------
were below the detection limit (DL), which corresponded to approximately 13 pg I-TEQDF/km
driven.
Table 4-1 presents a summary of the results of Marklund et al. (1987) and subsequent
studies, which are discussed below. Tables 4-2 and 4-3 present the results of tailpipe emission
studies reported for diesel-fueled cars and trucks, respectively. The results of studies of leaded
gasoline-fueled cars are shown in Table 4-4 and those for unleaded gasoline-fueled cars in Tables
4-5 and 4-6. Figures 4-1, 4-2, and 4-3 present congener and congener group profiles for
emissions from diesel-fueled vehicles, leaded gasoline-fueled vehicles, and unleaded gasoline-
fueled vehicles, respectively.
Virtually no testing of vehicle emissions for CDDs/CDFs in the United States has been
reported. In 1987, the California Air Resources Board (CARB) produced a draft report on the
testing of exhausts from four gasoline-fueled cars and three diesel-fueled vehicles (one truck, one
bus, and one car) (CARB, 1987). However, CARB indicated to EPA that the draft report should
not be cited or quoted to support general conclusions about CDDs/CDFs in motor vehicle
exhausts because of the small sample size of the study and because the use of low-resolution
rather than high-resolution mass spectrometry in the study resulted in high DLs and inadequate
selectivity in the presence of interferences (letter dated July 1993, from G. Lew, California Air
Resources Board, to J. Schaum, U.S. EPA).
CARB stated that the results of a single sample from the heavy-duty diesel truck could be
reported because congeners from most of the homologue groups were present in the sample at
levels that could be detected by the analytical method and there were no identified interferences
in this sample. This test was conducted under steady-state conditions (50 km/hr) for 6 hr with an
engine with a fuel economy of 5.5 km/L. The TEQ emission factor of this one sample was
equivalent to 7,190 pg TEQDF-WHO98/L (7,290 pg I-TEQDF/L) fuel burned. An assumed fuel
economy of 5.5 km/L yields an emission factor of 1,307 pg TEQDF-WHO98/km (1,325 pg I-
TEQDF/km). Assuming that nondetect values were zero, the TEQ emission factors were 3,280 pg
TEQDF-WHO98/L (3,720 pg I-TEQDF/L) fuel burned and 596 pg TEQDF-WHO98/km (676 pg I-
TEQDF/km) driven (letter dated January 11, 1996, from G. Lew, California Air Resources Board,
to G. Schweer, Versar, Inc.).
Haglund et al. (1988) sampled exhaust gases from three vehicles (one car fueled with
leaded gasoline and one with unleaded gasoline and a heavy-duty diesel truck) for the presence
of brominated dibenzo-p-dioxins (BDDs) and ethylene dibromide dibenzofurans (BDFs). The
authors concluded that the ethylene dibromide scavenger added to the tested gasoline probably
acted as a halogen source. Tetra-BDF emissions were measured as 23,000 pg/km in the car with
leaded gasoline and 240 pg/km in the car with unleaded gasoline. Tetra- and penta-emissions
were measured as 3,200 and 980 pg/km, respectively, in the car with leaded gasoline. All
BDDs/BDFs were below DLs in the diesel truck emissions.
4-2
-------
Table 4-1. Description and results of vehicle emission testing studies for CDDs and CDFs
Study
CARB (1987); letter
dated January 11, 1996,
from G. Lew,
California Air
Resources Board, to G.
Schweer, Versar, Inc.
Marklundetal. (1987)
Binghametal. (1989)
Marklundetal. (1990)
Hagenmaier et al.
(1990)
Oehmeetal. (1991)
(tunnel study)
Country
United States
Sweden
New Zealand
Sweden
Germany
Norway
Fuel type
Diesel (truck)
Unleaded
Leaded
Unleaded
Leaded
Unleaded
Leaded
Unleaded
Leaded
Diesel (truck)
Unleaded
Unleaded
Leaded
Diesel (car)
-
Scavenger3
No
No
Yes
No
Yes
No
Yes
No
Yes
No
No
No
Yes
No
-
Catalyst
equipped
NR
Yes
No
NR
NR
No
No
Yes
No
NR
No
Yes
No
NR
-
Number
of test
vehicles
1
2
4
1
4
2
2
1
2
1
1
1
1
1
e
TEQ emission factor"
(pg/km driven)
676-1325° [597-1307]
Not detected (<13)
Approx. 20-220
Not detected (<20)
1-39
0.36-0.39
2.4-6.3
0.36
l.l-2.6d
Not detected (<18)c
5.1C[6.0]
0.7C [0.8]
108C [129]
2.1C[2.5]
520f
38f
Avg = 280
9500f
720f
Avg =5, 100
Driving cycle; sampling location
6-hr dynamometer test at 50 km/hr
A10 (2 cycles); muffler exhaust
A10 (2 cycles); muffler exhaust
A10 (3 or 4 cycles); muffler exhaust
A10 (3 or 4 cycles); muffler exhaust
FTP-73 test cycle; before muffler
FTP-73 test cycle; before muffler
FTP-73 test cycle; in tailpipe
FTP-73 test cycle; in tailpipe
U.S. federal mode 13 cycle; before muffler
Comparable to FTP-73 test cycle; in tailpipe
Comparable to FTP-73 test cycle; in tailpipe
Comparable to FTP-73 test cycle; in tailpipe
Comparable to FTP-73 test cycle; in tailpipe
Cars moving uphill (3.5% incline) at
60 km/hr
Cars moving downhill (3 .5 % decline) at
70 km/hr
Car average
Trucks moving uphill (3 .5 % incline) at
60 km/hr
Trucks moving downhill (3.5% decline) at
70 km/hr
Truck average
-------
Table 4-1. Description and results of vehicle emission testing studies for CDDs and CDFs (continued)
Study
Schwindetal. (1991)
Hutzingeretal. (1992)
Gertleretal. (1996,
1998) (tunnel study)
Gullett and Ryan
(1997)
Country
Germany
United States
United States
Fuel type
Leaded
Unleaded
Unleaded
Diesel (car)
Diesel (truck)
Diesel (truck)
Diesel (truck)
Scavenger3
Yes
No
No
No
No
—
No
Catalyst
equipped
No
No
Yes
No
No
—
-
Number
of test
vehicles
1
TEQ emission factor"
(pg/km driven)
5.2-1 18C [7.2-142]
9.6-17.7c [10.2-18.1]
1-2.6C [1-2.8]
1-13C [1.2-14]
13-15C [14-15]
Mean = 172
Mean = 29
Driving cycle; sampling location
Various test conditions (loads and speeds)
Various test conditions (loads and speeds)
Various test conditions (loads and speeds)
Various test conditions (loads and speeds)
Various test conditions (loads and speeds)
Mean of seven 12-hr samples
Mean of five sample routes
aExcept in Marklund et al. (1987), dichloroethane and dibromoethane were used as scavengers.
Values are in units of I-TEQDF; values in brackets are in units of TEQDF-WHO98.
°Results reported were in units of pg TEQ/L of fuel. For purposes of this table, the fuel economy factor used by Marklund et al. (1990), 10 km/L (24
miles/gal), was used to convert the emission rates into units of pg TEQ/km driven for the cars. For the diesel-fueled truck, the fuel economy factor reported in
CARB (1987) for a 1984 heavy-duty diesel truck, 5.5 km/L (13.2 miles/gal), was used.
dTable reflects the range of summary results reported in Marklund et al. (1990); however, the congener-specific results for the single run reported indicate an
emission rate of about 7.3 pg I-TEQDF/km.
Tests were conducted over portions of 4 days, with traffic rates of 8,000-14,000 vehicles/day. Heavy-duty vehicles (defined as vehicles over 7 m in length)
ranged from 4 to 15% of total.
Emission factors are reported in units of pg Nordic TEQ/km driven; the values in units of I-TEQDF/km are expected to be about 3 to 6% higher.
BTests were conducted over 5 days, with heavy-duty vehicle rates of 1,800-8,700 vehicles per 12-hr sampling event. Heavy-duty vehicles accounted for 21
to 28% of all vehicles.
NR= Not reported
- = No data
-------
Table 4-2. CDD/CDF congener emission factors (pg/L) for diesel-fueled automobiles
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF (nondetect set to 0)
Total I-TEQDF (nondetect set to l/i DL)
Total TEQDF-WHO98 (nondetect set to 0)
Total TEQDF-WHO98 (nondetect set to 1A DL)
Automobile tailpipe emission study results
63 km/hra
7.9
9
ND(5.1)
ND(5.1)
ND(5.1)
44.1
440
20.5
ND(5.1)
7.1
6.5
6.7
ND(5.1)
ND(5.1)
40.7
8.5
94.4
501
184.4
20.8
22.2C
24.8
26.2
Idling
(test no. 25)"
13.1
6.3
21.4
36
28
107
635
79
171
58.7
121
75
17.1
52
159
11.9
214
846.8
958.7
100.7
100.7
103.1
103.1
57km/hr
(test no. 24)"
2.4
4.1
1
1.4
2
22.9
525
18.1
1.8
3.4
4.1
o
5
0.8
ND (0.4)
18.9
7.1
101
558.8
158.2
10.4
10.4
11.9
1.9
57km/hr
(full load)
(test no. 28)"
22
23
7.8
21
10
166
560
236
111
85
68
55
4.7
31
214
7.8
305
809.8
1117.5
129.6
129.6
140.4
140.4
Mean emission factors
Assuming
nondetect set
to zero
11.4
10.6
7.6
14.6
10
85
540
88.4
71
38.6
49.9
34.9
5.7
20.8
108.2
8.8
178.6
679.1
604.9
65.5
70.1
Assuming
nondetect set to
Vz detection
limit (DL)
11.4
10.6
8.2
15.2
10.6
85
540
88.4
71.6
38.6
49.9
34.9
6.3
21.4
108.2
8.8
178.6
681
606.7
65.8
70.5
-------
Table 4-2. CDD/CDF congener emission factors (pg/L) for diesel-fueled automobiles (continued)
Congener/congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF (nondetect set to 0)
Total CDD/CDF (nondetect set to l/i DL)
Automobile tailpipe emission study results
63 km/hra
37.4
19.7
23.6
88.5
440.5
76.7
39.3
25.6
80.6
94.4
926.3
926.3
Idling
(test no. 25)"
317
214
256
187
635
436
821
556
321
214
3,957
3,957
57km/hr
(test no. 24)"
31
22
20
77
525
58
36
26
72
101
968
968
57km/hr
(full load)
(test no. 28)"
394
228
164
356
560
3,093
1,205
472
241
305
7,018
7,018
Mean emission factors
Assuming
nondetect set
to zero
195
121
116
177
540
916
525
270
179
179
3,218
Assuming
nondetect set to
J/2 detection
limit (DL)
195
121
116
177
540
916
525
270
179
179
3,218
aSource: Hagenmaieretal. (1990).
bSource: Schwind et al. (1991); Hutzinger et al. (1992).
°AnI-TEQDF emission factor of 23.6 pg/L is reported in Hagenmaier et al. (1990); however, anI-TEQDF emission factor of 22.2 pg/L is calculated, based on
reported congener levels.
ND = Not detected (value in parenthesis is the detection limit)
-------
Table 4-3. CDD/CDF congener emission factors (pg/L) for diesel-fueled trucks
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF (nondetect set to 0)
Total I-TEQDF (nondetect set to Vi DL)
Total TEQDF-WHO98 (nondetect set to 0)
Total TEQDF-WHO98 (nondetect set to !/2 DL)
Truck tailpipe study results
50km/hr
(test no. 40)a
25
5
14
28
14
119
1355
87
45
18
56
84
4.7
63
375
40
397
1,560
1,169.7
81
81
82
82
90km/hr
(full load)
(test no. 42)a
16
18
5.7
6
6
74
353
53
34
51
29
31
5.1
23
71
5.4
104
478.7
406.5
70
70
79
79
50 km/hr "
ND (560)
ND (1,340)
ND(2,160)
ND (1,770)
ND (2,640)
116,000
344,400
ND (605)
ND (4,750)
ND(5,190)
ND (8,210)
ND (6,480)
13,400
ND (7,780)
73,460
ND (11,700)
140,400
460,400
227,260
3,720
7,290
3,280
7,190
Mean emission factors
Assuming
nondetect set to
zero
13.7
7.7
6.6
11.3
6.7
38,731
115,369
46.7
26.3
23
28.3
38.3
4,469
28.7
24,636
15.1
46,981
154,146
76,292.4
1,290
1,150
Assuming
nondetect set to Vz
detection limit (DL)
107
231
367
307
446
38,731
115,369
148
819
887
1,397
1,119
4,469
1,325
24,636
1,960
46,981
155,558
83,741
2,480
2,450
-------
Table 4-3. CDD/CDF congener emission factors (pg/L) for diesel-fueled trucks (continued)
Congener/congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF (nondetect set to 0)
Total CDD/CDF (nondetect set to 1A DL)
Truck tailpipe study results
50km/hr
(test no. 40)a
200
32
130
200
1,355
763
230
524
509
397
4,340
4,340
90km/hr
(full load)
(test no. 42)a
208
117
67
155
353
694
736
268
76
104
2,778
2,778
50 km/hr "
ND (3,760)
ND (3,020)
ND (45,300)
203,300
344,000
25,000
47,900
169,200
150,700
140,300
1,080,400
1,106,440
Mean emission factors
Assuming
nondetect set to
zero
136
49.7
65.7
67,892
115,252
8,831
16,294
56,670
50,414
46,932
362,536.4
Assuming
nondetect set to Vz
detection limit (DL)
762
553
7,620
67,892
115,252
8,831
16,294
56,670
50,414
46,932
371,220
oo "Source: Schwind et al. (1991); Hutzinger et al. (1992).
bSource: Letter dated July 1993, from G. Lew, California Air Resources Board, to J. Schaum, U.S. EPA; letter dated January 11, 1996, from G. Lew, California
Air Resources Board, to G. Schweer, Versar, Inc..
ND = Not detected (value in parenthesis is the detection limit)
-------
Table 4-4. CDD/CDF congener emission factors (pg/L) for leaded gasoline-fueled automobiles
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF ( nondetect set to 0)
Total I-TEQDF (nondetect set to
!/2 DL)
Total TEQDF-WHO98 (nondetect
set to 0)
Total TEQDF-WHO98 (nondetect
set to !/2DL)
Automotive tailpipe emission study results
FTP cycle3
ND(14.4)
ND(36)
ND(54)
ND(54)
ND(54)
ND(54)
ND(90)
432
21.6
43.2
ND(54)
ND(54)
ND(54)
ND(54)
ND(54)
ND(54)
ND(90)
ND
496.8
65.9
102
65.9
111
63 km/hr"
128
425
188
207
188
503
498
1,542
1,081
447
856
856
ND(76)
273
4,051
ND(76)
230
2,137
9,336
1,075
1,080
1,287
1,291
Idling
(test no.
12)c
NR
43
17
32
NR
119
380
NR
49
26
33
22
NR
NR
170
NR
1115
>591
>1,415
>52
>52
>72
>72
Full load
(test no.
13)c
60
106
15
35
NR
136
513
678
367
156
70
60
NR
25
NR
NR
NR
>865
> 1,356
>300
>300
>352
>352
64 km/hr
(test no.
14)<
141
468
206
228
206
554
549
1,697
1,190
492
942
942
NR
301
4,460
NR
253
2,352
> 10,277
> 1,184
> 1,184
> 1,417
> 1,417
Rated
power
(test no.
15)'
NR
40
16
30
NR
111
1,166
78
45
24
31
20
NR
NR
158
NR
447
> 1,363
>803
>56
>56
>75
>75
FTP cycle
(test no.
22)c
5
73
41
62
35
518
1,581
214
218
225
381
375
85
1,033
2,301
109
1,128
2,315
6,069
419
419
454
454
Mean emission factors
Assuming
nondetect
set to
zero
67
165
69
85
107
277
670
774
425
202
330
325
28
326
1,857
36
529
1,440
4,832
>450
>532
Assuming
nondetect
set to 1A
detection
limit (DL)
68
168
73
89
114
281
676
774
425
202
334
329
50
332
1,861
58
536
1,469
4,901
>456
>539
-------
Table 4-4. CDD/CDF congener emission factors (pg/L) for leaded gasoline-fueled automobiles (continued)
Congener/congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF (nondetect set
toO)
Total CDD/CDF (nondetect set to
!/2 DL)
Automotive tailpipe emission study results
FTP cycle3
5,220
ND (360)
ND (540)
ND(90)
ND(90)
15,300
2,430
ND (540)
ND (270)
ND(90)
22,950
23,940
63 km/hr"
4,555
3,338
1,868
1,164
498
50,743
11,591
6,308
5,642
230
85,937
85,937
Idling
(test no.
12)c
517
658
354
194
380
2,167
452
192
170
1,115
6,199
6,199
Full load
(test no.
13)c
8,134
2,161
623
297
513
20,513
3,608
477
NR
NR
>36,326
>36,326
64 km/hr
(test no.
14)c
5,012
3,675
2,056
1,281
549
55,857
12,757
6,947
6,210
253
94,597
94,597
Rated
power
(test no.
15)c
4,558
6,389
1,973
2,374
1,166
29,353
10,580
12,553
4,767
447
74,160
74,160
FTP cycle
(test no.
22)c
921
359
996
988
1,581
4,290
3,165
3,132
2,920
1,128
19,480
19,480
Mean emission factors
Assuming
nondetect
set to
zero
4,131
2,369
1,124
900
670
25,460
6,369
4,230
3,285
529
49,067
Assuming
nondetect
set to Vz
detection
limit (DL)
4,131
2,394
1,163
906
676
25,460
6,369
4,268
3,307
536
49,210
aSource: Marklund et al. (1990); values in the table were calculated from the reported units of pg/km to pg/L using a fuel economy of 9 km/L for leaded gas as
reported in Marklund et al. (1990).
bSource: Hagenmaier et al. (1990).
cSource: Schwind et al. (1991); Hutzinger et al. (1992).
ND = Not detected (value in parenthesis is the reported detection limit)
NR = Not reported
-------
Table 4-5. CDD/CDF congener emission factors (pg/L) for unleaded gasoline-fueled automobiles without catalytic
converters
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF (nondetect set to 0)
Total I-TEQDF (nondetect set to
!/2 DL)
Total TEQDF-WHO98 (nondetect set
toO)
Total TEQDF-WHO98 (nondetect set
to !/2 DL)
Automotive tailpipe emission study results
FTP cycle3
ND(5)
ND(3)
ND(40)
ND(40)
ND(40)
ND(40)
ND(50)
64
ND(7)
ND(7)
ND(40)
ND(40)
ND(40)
ND(40)
ND(40)
ND(40)
ND(70)
ND
64
6.4
26.2
6.4
26.9
63 km/hr"
2.6
19.1
16.6
17.1
17.6
40.4
176
44
44.5
20.7
41.9
21.2
37.8
54.3
27.9
16.6
119
289.4
427.9
50.9
50.9
60.2
60.2
FTP cycle
(test no.
21)c
24
14
24
84
15
192
868
70
40
30
68
62
47
55
278
ND(1)
374
1,221
1,024
96.4
96.4
102
102
64 km/hr
(test no.
17)c
44
31
26
28
29
66
280
71
72
34
68
34
61
88
45
27
194
504
694
122
122
138
138
64 km/hr
(test no.
20)c
7
11
25
42
23
121
685
77
69
184
88
35
ND(1)
42
22
24
288
914
829
144
144
148
148
64 km/hr
(test no.
31/2)c
8.9
14.1
16.3
60.1
17.1
197.8
2,634
295.2
161.8
135.2
129.1
113.2
36.9
82.1
418
54.5
991
2,948.3
2,417
177
177
181
181
Mean emission factors
Assuming
nondetect
set to zero
14.4
14.9
18
38.5
17
103
774
104
64.6
67.3
65.8
44.2
30.5
53.6
132
20.4
328
979
910.4
99.5
106
Assuming
nondetect
set to 1A
detection
limit (DL)
14.8
15.1
21.3
41.9
20.3
106
778
104
65.1
67.9
69.2
47.6
33.9
56.9
135
23.8
334
997.4
937.4
103
109
-------
Table 4-5. CDD/CDF congener emission factors (pg/L) for unleaded gasoline-fueled automobiles without catalytic
converters (continued)
Congener/congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF (nondetect set to 0)
Total CDD/CDF (nondetect set to
ViDL
Automotive tailpipe emission study results
FTP cycle3
13
ND(3)
ND(40)
ND(10)
ND(5)
170
ND(7)
ND(40)
ND(20)
ND(7)
183
249
63 km/hr"
435
481
305
93
176
569
931
378
476
119
3,963
3,963
FTP cycle
(test no.
21)'
429
837
484
392
868
718
531
165
278
374
5,076
5,076
64 km/hr
(test no.
17)c
706
784
496
147
280
923
1,513
615
773
194
6,431
6,431
64 km/hr
(test no.
20)c
500
542
563
225
685
478
437
258
445
288
4,421
4,421
64 km/hr
(test no.
31/2)c
304
170
114
301
2,634
6,379
1,969
1,226
1,088
991
15,176
15,176
Mean emission factors
Assuming
nondetect
set to zero
398
469
327
193
774
1,540
897
440
510
328
5,876
Assuming
nondetect
set to 1A
detection
limit (DL)
398
469
330
194
774
1,540
897
444
512
328
5,886
to
aSource: Marklund et al. (1990); the pg/L values in the table were calculated from the reported units of pg/km assuming a fuel economy of 10 km/L for
unleaded gas.
bSource: Hagenmaier et al. (1990).
cSource: Schwind et al. (1991); Hutzinger et al. (1992).
ND = Not detected (value in parenthesis is the reported detection limit)
-------
Table 4-6. CDD/CDF congener emission factors (pg/L) for unleaded gasoline-fueled automobiles with catalytic
converters
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF (nondetect set to 0)
Total I-TEQDF (nondetect set to !/2 DL)
Total TEQDF-WHO98 (nondetect set to 0)
Total TEQDF-WHO98 (nondetect set to l/i DL)
Automotive tailpipe emission study test results
63 km/hra
1.6
1.6
2.4
3.5
3.1
15.3
170
4.3
3.3
2.4
4.8
6.3
0.2
4.6
16.3
ND (0.2)
27.9
197.5
70.1
7.2
7.2
7.8
7.8
64 km/hr
(test no. 29)"
3
2.6
5.3
6
6
27.8
275
10.6
8.7
7.2
10.6
9.1
ND(3.8)
18.1
54.3
ND(3.8)
38
325.7
156.6
16
16.2
17.1
17.3
64 km/hr
(test no. 30)"
ND (7.9)
ND (7.9)
ND (7.9)
6.4
ND (7.9)
78.1
427
12.7
5.1
6.2
4.5
3.9
2.1
8.2
154.2
7.9
106
511.5
310.8
10.1
16.8
9.6
18.3
64 km/hr
(test no. 18)"
14
4
1
2
2
14
197
35
13
6
5
7
5
ND(1)
51
1
140
234
263
26.3
26.4
28
28.1
Mean emission factors
Assuming
nondetect
set to zero
4.7
2.1
2.2
4.5
2.8
33.8
267
15.7
7.5
5.5
6.2
6.6
1.8
7.7
69
2.2
78
317.1
200.2
14.9
15.6
Assuming
nondetect set
to Vz detection
limit (DL)
5.6
3
3.2
4.5
3.8
33.8
267
15.7
7.5
5.5
6.2
6.6
2.3
7.9
69
2.7
78
320.9
201.4
16.6
17.9
-------
Table 4-6. CDD/CDF congener emission factors (pg/L) for unleaded gasoline-fueled automobiles with catalytic
converters (continued)
Congener/congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF (nondetect set to 0)
Total CDD/CDF (nondetect set to l/i DL)
Automotive tailpipe emission study test results
63 km/hra
28.6
25.5
26.3
38.7
170
52.6
53.4
33.3
27.1
27.9
483.4
483.4
64 km/hr
(test no. 29)"
51
51
56
50
275
152
122
71
62
38
928
928
64 km/hr
(test no. 30)"
13
ND(15)
36
163
427
79
29
60
174
106
1,087
1,087
64 km/hr
(test no. 18)"
82
101
50
25
197
332
84
39
83
140
1,133
1,133
Mean emission factors
Assuming
nondetect
set to zero
43.7
44.4
42.1
69.2
267.3
153.9
72.1
50.8
86.5
78
908
Assuming
nondetect set
to Vz detection
limit (DL)
43.7
46.3
42.1
69.2
267.3
153.9
72.1
50.8
86.5
78
909.9
aSource: Hagenmaieretal. (1990).
bSource: Schwind et al. (1991); Hutzinger et al. (1992).
ND = Not detected (value in parenthesis is the reported detection limit)
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0.05 0.1 0.15 0.2 0.25 0.3
0.35
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Ratio (congener group emission factor/total CDD/CDF emission factor)
0.05 0.1 0.15 0.2 0.25 0.3
0.35
I Cars
Trucks
Figure 4-1. Congener and congener group profiles for air emissions from
diesel-fueled vehicles (based on profiles calculated from emission factors
[nondetects equal one-half the detection limit] from Tables 4-2 and 4-3).
4-15
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0.01 0.02 0.03 0.04
0.05
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Ratio (congener group emission factor/total CDD/CDF emission factor)
0.1 0.2 0.3 0.4 0.5
0.6
Figure 4-2. Congener and congener group profiles for air emissions from
leaded gas-fueled vehicles (based on profiles calculated from emission factors
[nondetects equal one-half the detection limit] from Table 4-4).
4-16
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0.05 0.1 0.15 0.2 0.25 0.3
0.35
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Ratio (congener group emission factor/total CDD/CDF emission
0 0.05 0.1 0.15 0.2 0.25 0.3
0.35
Figure 4-3. Congener and congener group profiles for air emissions from
unleaded gas-fueled vehicles (catalytic converter equipped vehicles; based on
data from Table 4-6).
4-17
-------
Bingham et al. (1989) analyzed the exhausts of four cars using leaded gasoline (uniform
lead and organics content of 0.45 g/L tetramethyl lead, 0.22 g/L dichloroethane, and 0.2 g/L
dibromoethane) and the exhaust of one car using unleaded gasoline. Analytical results and DLs
were reported for only 5 of the 17 toxic CDD/CDF congeners. TEQ emission rates for the cars
using leaded fuel, based on detected congeners only, ranged from 1 to 39 pg I-TEQDF/km.
CDDs/CDFs were not detected in the exhaust from the vehicle using unleaded fuel; the total TEQ
emission rate for this car, based on one-half the DLs for the five reported congeners, was 20 pg I-
TEQDF/km.
Marklund et al. (1990) tested Swedish cars fueled with commercial fuels, measuring
CDD/CDF emissions before and after the muffler. Both new and old vehicles were tested. The
tests were done on three cars using unleaded gasoline and two cars using leaded gasoline (0.15 g
Pb/L with 1,2-dichloroethane and ethylene dibromide scavengers). CDDs/CDFs were not
detected in the fuels at a DL of 2 pg I-TEQDF/L but were detected at a level of 1,200 pg I-TEQDF/L
in the new semisynthetic engine lube oil used in the engines. The test driving cycle used (31.7
km/hr as a mean speed, 91.2 km/hr as a maximum speed, and 17.9% of time spent idling) yielded
fuel economies ranging from approximately 9 to 10 km/L (22 to 24 miles/gal) in the various cars.
The reported ranges of emission factors were
• Leaded gas, measured before muffler: 2.4 to 6.3 pg I-TEQDF/km (21 to 60 pg I-
TEQDF/L fuel consumed)
• Leaded gas, measured in tailpipe: 1.1 to 2.6 pg I-TEQDF/km (10 to 23 pg I-TEQDF/L)
• Unleaded gas, catalyst-equipped, measured in tailpipe: 0.36 pg I-TEQDF/km (3.5 pg
I-TEQDF/L)
• Unleaded gas, measured before muffler: 0.36 to 0.39 pg I-TEQDF/km (3.5 pg I-
TEQDF/L)
The TEQ levels in exhaust gases from older cars using leaded gasoline were up to six
times greater when measured before the muffler than when measured after the muffler. No
muffler-related difference was observed in new cars running on leaded gasoline or in old or new
cars running on unleaded gasoline.
Marklund et al. (1990) also analyzed the emissions of a heavy-duty, diesel-fueled truck for
CDDs/CDFs. None were detected; however, the authors pointed out that the test fuel was a
reference fuel and may not have been representative of commercial diesel fuel. Also, due to
analytical problems, a much higher DL (about 100 pg I-TEQDF/L) was realized in this diesel fuel
test than in the gasoline tests conducted (5 pg I-TEQDF/L). Further uncertainty was introduced
because the diesel emission samples were collected only before the muffler.
4-18
-------
Hagenmaier et al. (1990) ran a set of tests using conditions comparable to the FTP-73 test
cycle on gasoline- and diesel-fueled engines for light-duty vehicles in Germany. The following
average TEQ emission rates per liter of fuel consumed were reported:
• Leaded fuel: 1,287 pg TEQDF-WHO98/L (1,080 pgI-TEQDF/L)
• Unleaded fuel (catalyst equipped): 7.9 pg TEQDF-WHO98/L (7.2 pg I-TEQDF/L)
• Unleaded fuel (not catalyst equipped): 60.2 pg TEQDF-WHO98/L (50.9 pg I-TEQDF/L)
• Diesel fuel: 24.8 pg TEQDF-WHO98/L (20.8 pg I-TEQDF/L)
Schwind et al. (1991) published the major findings of a German study of emissions of
halogenated dibenzo-p-dioxins and dibenzofurans from internal combustion engines running on
commercial fuels. The full report was published in 1992 (Hutzinger et al., 1992). The study was
conducted by the universities of Stuttgart, Tubingen, and Bayreuth for the Federal Ministry for
Research and Technology, the Research Association for Internal Combustion Engines, and the
German Association for the Petroleum Industry and Coal Chemistry. Tests were conducted using
engine test benches and rolling test benches under representative operating conditions. Tests were
performed on leaded gasoline engines, unleaded gasoline engines, diesel car engines, and diesel
truck engines. The reported range of CDD/CDF emission rates across the test conditions in units
of pg I-TEQ/L of fuel consumed are presented below.
• Leaded fuel: 72 to 1,417 pg TEQDF-WHO98/L (52 to 1,184 pg I-TEQDF/L)
• Unleaded fuel (not catalyst equipped): 102 to 181 pg TEQDF-WHO98/L (96 to 177 pg
I-TEQDF/L)
• Unleaded fuel (catalyst equipped): 9.6 to 28 pg TEQDF-WHO98/L (10 to 26 pg I-
TEQDF/L)
• Diesel fuel (cars): 12 to 140 pg TEQDF-WHO98/L (10 to 130 pg I-TEQDF/L)
• Diesel fuel (trucks): 79 to 82 pg TEQDF-WHO98/L (70 to 81 pg I-TEQDF/L
Tables 4-2 through 4-6 show the results from tests with commercial fuels that were not
conducted under normal operating conditions and for which congener-specific emission results
were presented in Hutzinger et al. (1992).
Although no specific details on the methodology used were provided, Hagenmaier (1994)
reported that analyses of emissions of a diesel-fueled bus run on either the steady-state or the
"Berlin cycle" showed no CDDs/CDFs present at a DL of 1 pg/L of fuel consumed for individual
congeners.
4-19
-------
Gullett and Ryan (1997) reported the results of the first program to sample diesel engine
emissions for CDDs/CDFs during actual highway and city driving. The exhaust emissions from a
1991 Freightliner diesel tractor with a 10.3 L, six-cylinder Caterpillar engine—representative of
the first generation of computerized fuel-controlled vehicles manufactured in the United
States—were sampled during both highway and city routes. The average emission factor for the
three highway tests conducted (15.1 pg I-TEQDF/km; range, 11.7 to 18.7 pg I-TEQDF/km; standard
deviation, 3.5 pg I-TEQDF/km) was below the average of the two city driving tests by a factor of 3
(49.9 pg I-TEQDF/kg; range, 3 to 96.8 pg I-TEQDF/km). DLs were considered to be zero in the
calculation of these emission factors. The average of all five tests was 29 pg I-TEQDF/km, with a
standard deviation of 38.3 pg I-TEQDF/km. This standard deviation reflects the 30-fold variation
in the two city driving route tests.
Geueke et al. (1999) analyzed dioxin emissions from heavy-duty vehicle diesel engines in
Germany. Table 4-7 shows the results of the analysis. I-TEQ values ranged from 2 to 18 pg I-
TEQ/m3, including one value so high that it could not be reproduced. Miyabara et al. (1999)
analyzed CDDs/CDFs found in vehicle exhaust particles from a gasoline engine and a diesel
engine in Japan. Table 4-8 presents the data from three tests conducted on the exhaust particles
deposited on the tailpipe of the gasoline engine. TEQ values ranged from 3.44 to 5.32 pg TEQDF-
WHO98/g (3.46 to 5.33 pg I-TEQ/g) exhaust particles. Suspended particulate matter (PM) was
also collected from an electrostatic precipitator (ESP) connected to a highway tunnel. The I-TEQ
for the suspended PM was 257.5 pg TEQDF-WHO98/g (241.6 pg I-TEQ/g), two orders of
magnitude higher than the TEQ for exhaust particles deposited on the tailpipes. Table 4-9 depicts
the data from three tests conducted on the exhaust particles deposited on the tailpipe of the diesel
engine. TEQ values ranged from 7.14 to 18.1 pg TEQDF-WHO98/g (7.13 to 14 pg I-TEQ/g) soot.
4.1.2. Tunnel Emission Studies
Several European studies and one U.S. study evaluated CDD/CDF emissions from
vehicles by measuring the presence of CDDs/CDFs in tunnel air. This approach has the advantage
of allowing the random sampling of exhaust from large numbers of cars with a range of ages and
maintenance levels. The disadvantage of this approach is that it relies on indirect measurements
(rather than tailpipe measurements), which may introduce bias and make interpretation of the
findings difficult. Concerns have been raised that tunnel monitors detect resuspended particulates
that have accumulated over time, leading to overestimates of emissions. Also, the driving patterns
encountered in these tunnel studies are more or less steady-state driving conditions, which may
produce emission levels different from those of the transient driving cycle and cold engine starts
that are typical of urban driving conditions. These studies are summarized below in chronological
order.
4-20
-------
Table 4-7. Total dioxin emission concentrations from heavy-duty diesel
engines in Germany
Sample
Stationary engine 1 (SI)
Stationary engine 2 (S2)
Stationary engine 3 (S3)
Stationary engine 4 (S4)
Stationary engine 5 (S5)
Truck engine 1 (VI)
Truck engine 2 (V2)
Truck engine 3 (V3)
Concentration in exhaust3
(pg I-TEQ/m3)
6.1
61b
18C
6.9
6.6
9.7
2.1
2
Sample volume
(m3)
32.89
10.35
10.73
10.06
10.06
10.03
10.07
9.99
"Detection limit for sampling: 4.1 pg/m3 for stationary samples, 4.5 pg/m3 for truck samples.
bAnalysis could not be confirmed.
"High analytical detection limit (11 pg/m3).
Source: Geueke et al. (1999).
Rappe et al. (1988) reported the CDD/CDF content of two air samples (60 m3/sample)
collected from a tunnel in Hamburg, Germany, in January 1986 to be 0.44 and 0.59 pg TEQDF-
WHO98/m3 (0.42 and 0.58 pg I-TEQDF/m3). Each sample was collected over a period of about
60 hr. The tunnel handled 65,000 vehicles per day, of which 17% were classified as "heavy
traffic." The congener-specific results of the two samples are presented in Table 4-10. Ambient
air measured in September 1986 at a nearby highway in Hamburg was reported to contain
CDD/CDF levels two to six times lower than those measured in the tunnel.
Larssen et al. (1990) and Oehme et al. (1991) reported the results of a tunnel study inOslo,
Norway, performed during April and May of 1988. Oehme et al. estimated total vehicle emissions
by measuring CDD/CDF concentrations in tunnel inlet and outlet air of both the uphill and the
downhill lanes. Emission rates for light-duty and heavy-duty vehicle classes were estimated by
counting the number of light-duty and heavy-duty vehicles passing through the tunnel on
workdays and a weekend and assuming a linear relationship between the percentage of the light-
er heavy-duty traffic and the overall emission rate. Thus, the linear relationship for each emission
rate was based on only two points (i.e., the weekday and the weekend measurements).
4-21
-------
Table 4-8. Levels of 2,3,7,8-chlorine-substituted congeners and total
CDDs/CDFs in vehicle exhaust particles (pg/g) for gasoline engines and
suspended particulate matter (SPM)
Congener
2,3,7,8-TCDD
Other TCDD
2,3,7,8-TCDF
Other TCDF
1,2,3,7,8-PeCDD
Other PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Other PeCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Other HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
Other HxCDF
1,2,3,4,6,7,8-HpCDD
Other HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Other HpCDF
OCDD
OCDF
Total CDD/CDF
I-TEQ
TEQDF-WH098
Sample 1
<4.4
6.21
3.98
36.8
<7.6
<7.6
5.58
2.87
24.4
4.3
<3.8
<3.8
4.14
6.85
4.94
<1.9
<1.9
47.2
<8.2
<8.2
<1.9
<1.9
<1.9
8.76
4.78
160.8
3.73
3.72
Gasoline
Sample 2
<2.1
19
5.17
68.8
<3.6
11.5
6.46
5.24
53.9
<1.8
2.66
<1.8
20.5
3.95
4.48
<0.9
4.94
23.7
11.4
11.3
12.7
1.06
8.36
13.8
5.09
294
5.33
5.32
Sample 3
<1.2
7.41
3.53
41.9
<2.1
4.25
3.07
3.66
38.3
0.86
1.36
0.63
10.5
2.26
2.35
<0.5
1.99
15.2
7.64
9
7.41
0.5
4.88
17
3.03
186.7
3.46
3.44
SPM
<5.2
4,580
108
2,830
40.8
1,240
184
107
29,700
42.3
96.7
71
1,100
243
231
38.6
387
1,600
1,700
1,360
1,330
143
778
3,650
1450
53,010.4
241.6
257.5
Source: Miyabaraetal. (1999).
The emission rates estimated in this study, in units of Nordic TEQ, are as follows:
• Light-duty vehicles using gasoline (approximately 70 to 75% using leaded gas):
uphill, 520 pg TEQ/km; downhill, 38 pg TEQ/km; mean, 280 pg TEQ/km
4-22
-------
Table 4-9. Levels of 2,3,7,8-chlorine-substituted congeners and total
CDDs/CDFs in vehicle exhaust particles (pg/g) for diesel engines
Congener
2,3,7,8-TCDD
Other TCDD
2,3,7,8-TCDF
Other TCDF
1,2,3,7,8-PeCDD
Other PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Other PeCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Other HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
Other HxCDF
1,2,3,4,6,7,8-HpCDD
Other HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Other HpCDF
OCDD
OCDF
Total CDD/CDF
I-TEQ
TEQDF-WH098
Sample 1
2.81
267
5.71
84.2
10.5
165
3.17
1.11
27.3
3.39
4.59
2.14
40.9
1.29
<1.2
<1.2
<1.2
3.7
8.78
10.1
<1
<1
<1
<2.8
<4.4
641.7
10.6
13.0
Diesel
Sample 2
<14.4
117
15.9
335
<28.8
73.5
16.6
<11.5
211
<17.3
<17.3
<17.3
28.1
15.9
31.3
<10.1
<10.1
182
<36
<36
<8.6
<8.6
<8.6
<23
<36
1,026.3
7.14
7.14
Sample 3
<2
86.9
7.5
313
8.15
83.6
15.1
9.52
243
4.01
4.6
<1.5
26.9
9.03
8.22
0.86
9.58
79
1.24
<1
4.69
<1
6.28
<0.5
4.25
925.4
14
18.1
Source: Miyabaraetal. (1999).
• Heavy-duty diesel trucks: uphill, 9,500 pg TEQ/km; downhill, 720 pg TEQ/km;
mean, 5,100 pg TEQ/km
The mean values are the averages of the emission rates corresponding to the two operating
modes: vehicles moving uphill on a 3.5% incline at an average speed of 37 mph and vehicles
4-23
-------
Table 4-10. CDD/CDF concentrations (pg/m3) from European tunnel air studies
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF (nondetect set to 0)
Total I-TEQDF (nondetect set to l/i DL)
Total TEQDF-WHO98 (nondetect set to 0)
Total TEQDF-WHO98 (nondetect set to
!/2DL)
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF (nondetect set to 0)
Total CDD/CDF (nondetect set to l/i DL)
Germany3
ND(O.Ol)
0.31
0.37
1.19
0.44
1.9
6.3
0.17
0.4
0.19
0.26
0.16
ND (0.04)
0.12
1.2
ND(0.16)
ND(1.3)
10.51
2.5
0.58
0.59
0.73
0.74
0.23
2.5
7.8
3.4
6.3
3.5
3.6
2.
1.9
ND(1.3)
31.2
31.9
Germany3
0.06
0.28
ND(0.17)
0.66
ND(0.17)
2
6.4
0.72
0.36
NR
0.13
0.15
ND (0.05)
ND (0.05)
0.98
ND(0.17)
ND(1)
9.40
2.34
0.42
0.44
0.55
0.58
0.22
1.3
2.7
3.4
6.4
6.2
4.1
1.1
1.2
ND(1)
26.6
27.1
Belgium"
0.002
0.025
0.025
0.042
0.03
0.468
2.19
0.013
0.143
0.039
0.073
0.093
0.143
0.004
0.499
0.074
0.25
2.78
1.33
0.096
0.096
0.106
0.106
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Norway
(workdays)0'11
0.02
0.18
0.06
0.29
0.25
1.41
0.1
0.58
0.83
0.78
0.79
0.62
0.04
0.74
1.78
0.22
1.62
2.31
8.0
0.91
0.91
1
1
0.26
1.78
1.32
1.31
0.1
13.20
10.17
6.42
2.62
1.62
38.8
38.8
Norway
(weekend)0'"
0.02
0.04
0.03
0.03
0.06
0.16
0.5
0.07
0.75
0.58
0.34
0.31
0.03
0.13
0.93
0.14
2.54
0.84
5.82
0.48
0.48
0.49
0.49
0.16
0.41
0.12
0.23
0.5
1.7
7.91
2.08
1.41
2.54
17.06
17.06
"Source: Rappe et al. (1988).
bSource: Wevers et al. (1992).
"Source: Oehme etal. (1991).
dListed values are the differences between the concentrations at the inlet and the outlet of the northbound tunnel lanes.
DL = Detection limit
ND = Not detected (value in parenthesis is the detection limit)
NR = Not reported
4-24
-------
moving downhill on a 3.5% decline at an average speed of 42 mph. Although Oehme et al.
reported results in units of Nordic TEQ, the results in I-TEQDF should be nearly identical (only
about 3 to 6% higher) because the only difference between the two TEQ schemes is the TEF
assigned to 1,2,3,7,8-PeCDF (0.1 in Nordic TEQ and 0.05 in I-TEQDF), a minor component of the
toxic CDDs/CDFs measured in the tunnel air. Table 4-10 presents the congener-specific
differences in concentrations between the tunnel inlet and outlet concentrations.
Wevers et al. (1992) measured the CDD/CDF content of air samples taken during the
winter of 1991 inside a tunnel in Antwerp, Belgium. Background concentrations outside the
tunnel were also determined. Two to four samples were collected from each location with two
devices: a standard high-volume sampler with a glass fiber filter and a modified two-phase, high-
volume sampler equipped with a glass fiber filter and a polyurethane foam (PUF) plug. The
I-TEQDF concentration in the air sampled with the filter with the PUF plug was 74 to 78% of the
value obtained with the standard high-volume sampler. However, the results obtained from both
sets of devices indicated that the tunnel air had a CDD/CDF TEQ concentration about twice as
high as that of the outside air (filter with PUF: 80.3 fg I-TEQDF/m3 for tunnel air vs. 35 fg I-
TEQDF/m3 for outside air; filter only: 100 fg I-TEQDF/m3 for tunnel air vs. 58 fg I-TEQDF/m3 for
outside air). The authors presented the congener-specific results for only one tunnel air
measurement; these results are presented in Table 4-10.
During October and November of 1995, Gertler et al. (1996, 1998) conducted a study at
the Fort McHenry Tunnel in Baltimore, MD. Their stated objective was to measure CDD/CDF
emission factors from in-use vehicles operating in the United States, with particular emphasis on
heavy-duty trucks. The air volume entering and leaving the tunnel bore (the area that the traffic
goes through) that is used by most of the heavy-duty trucks (i.e., approximately 25% of the
vehicles using the bore are heavy-duty trucks) was measured, and the air was sampled for
CDDs/CDFs during seven 12-hr sampling periods. Three of the samples were collected during
daytime (6 a.m. to 6 p.m.) and four samples were collected during the night (6 p.m. to 6 a.m.).
The air volume and concentration measurements were combined with information on vehicle
counts (obtained from videotapes) and tunnel length to determine average emission factors.
A total of 33,000 heavy-duty trucks passed through the tunnel during the seven sample
runs (21.2 to 28.8% of all vehicles). The emission factors, calculated on the assumption that all
CDDs/CDFs emitted in the tunnel were from heavy-duty trucks, are presented in Table 4-11. The
average TEQ emission factor was reported to be 181.8 pg TEQDF-WHO98/km (172 pg I-
TEQDF/km). The major uncertainties identified by the study authors were tunnel air volume
measurement, sampler flow volume control, and analytical measurement of CDDs/CDFs.
4-25
-------
Table 4-11. Baltimore Harbor tunnel study: estimated emission factors (pg/km) for heavy-duty diesel trucks"
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Run-specific emission factors
Run 2
24.5
40.2
18.2
37.5
53.6
0
0
0
0
24.5
15.4
0.3
27.7
15.2
12.6
0
0
174
95.7
73.8
93.9
Run 3
61.6
20.6
25.2
28.2
56.5
401
3,361
94.3
48.9
75.7
139
75.1
14.8
82.5
280
58.5
239
3,954
1,107.8
174.8
181.8
Run5
0
15.4
46.5
64.3
91.6
729
3,382
67.6
72.6
131
204
73.7
75.6
152
445
60.8
401
4,328.8
1,683.3
170.5
174.8
Run 6
21.2
5.6
8.3
19.6
48.4
111
1,120
152.8
23.6
46.6
93.8
51
0
55.7
154
31.1
175
4,328
1,684
170
175
Run 8
37.8
38.4
64.5
153
280
2,438
9,730
155.8
53.3
85
124
61.3
20.6
93
313
25
416
1,334.1
783.6
95.7
97.3
Run 9
40.1
0
0
71.1
126
963
5,829
73.4
0
63.9
164
54.4
37.2
86.8
354
2.3
534
7,029
1,370
152.9
147.2
Run 10
54.9
83
123
186
370
2,080
7,620
61.7
43.3
108
166
95.5
63.5
111
308
34.9
370
10,516.9
1,361.9
302.5
336.8
Mean
emission
factors
34.3
29
40.8
80
146.6
960.3
4,434.6
86.5
34.5
76.4
129.5
58.8
34.2
85.2
266.7
30.4
305
5,725.6
1,107.2
172.2
182.4
-1^
to
-------
Table 4-11. Baltimore Harbor tunnel study: estimated emission factors (pg/km) for heavy-duty diesel trucks"
(continued)
Congener/congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Heavy-duty vehicles as % of total
vehicles
Run-specific emission factors
Run 2
245
110
677
0
0
0
124
136
0
0
1,292
21.2
Run 3
0
21.9
0
802
3,361
901
119
319
223
239
5,985.9
22
Run5
140
83.3
753
1,498
3,382
1,314
1,152
852
814
401
10,389.3
22.6
Run 6
165
35.6
54.5
142
1,120
656
78.4
67.6
144
175
2,638.1
34
Run 8
311
174
2,009
5,696
9,730
2,416
1,055
444
513
416
22,764
28.8
Run 9
109
0
1,666
1,933
5,829
1,007
282
719
354
534
12,433
24.2
Run 10
97.3
165
2,971
4,377
7,620
687
626
619
637
370
18,169.3
27.4
Mean
emission
factors
152.5
84.3
1,161.5
2,064
4,434.6
997.3
490.9
450.9
383.6
305
10,524.5
25.7
J^.
K>
aValues are based on the difference between the calculated chemical mass entering the tunnel and the mass exiting the tunnel. All calculated negative
emission factors were set equal to zero. All CDD/CDF emissions were assumed to result from heavy-duty diesel-fueled vehicles.
Source: Gertleretal. (1996, 1998).
-------
EPA's Office of Transportation and Air Quality (OTAQ) reviewed the Gertler et al.(1996)
study (Lorang, 1996) and found it to be technologically well done; no major criticisms or
comments on the test methodology or protocol were offered, nor did OTAQ find any reason to
doubt the validity of the emission factor determined by the study. OTAQ noted that the
particulate emission rate for heavy-duty vehicles measured in the study (0.32 g/mile) was lower
than the general particulate emission rate used by EPA (about 1 g/mile) and, thus, may
underestimate CDD/CDF emissions under different driving conditions. OTAQ cautioned that the
reported emission factor should be regarded only as a conservative estimate of the mean emission
factor for the interstate trucking fleet under the driving conditions of the tunnel (i.e., speeds on the
order of 50 mph, with those of the entering traffic slightly higher and those of the exiting traffic
slightly lower).
Figure 4-4 graphically presents the results of the studies by Rappe et al. (1988), Oehme et
al. (1991), Wevers et al. (1992), and Gertler et al. (1996, 1998). The figure compares the
congener profiles (i.e., congener concentrations or emission factors normalized to total
concentration or emission factor of 2,3,7,8-substituted CDDs and CDFs) reported in the four
studies. The dominant congeners in the Rappe et al., Wevers et al., and Gertler et al. studies are
OCDD; 1,2,3,4,6,7,8-HpCDD; OCDF; and 1,2,3,4,6,7,8-HpCDF. With the exception of OCDD,
these congeners are also the major congeners reported by Oehme et al. The Oehme et al. study
also differs from the other tunnel studies in that the total of 2,3,7,8-substituted CDFs far exceeds
the total of 2,3,7,8-substituted CDDs (by a factor of 2), whereas the other three observed just the
opposite.
4.1.3. National Emission Estimates
Estimates of national CDD/CDF TEQ emissions for reference years 1987 and 1995 are
presented in this section only for on-road vehicles using gasoline and diesel fuel. For reference
year 2000, EPA's Office of Air Quality Planning and Standards (OAQPS) developed national
CDD/CDF TEQ emission estimates for on-highway gasoline and diesel vehicles, off-highway
gasoline and diesel equipment, diesel railroad equipment, and diesel commercial marine vessels.
4.1.3.1. Activity Information for On-Road Vehicles
Reference year 2000 activity information for on-highway gasoline and diesel vehicles was
estimated by OAQPS as county-level vehicle miles driven (VMD). The estimates include
calculations by month, road type, and vehicle type. To develop the VMD, OAQPS relied on data
supplied by the Federal Highway Administration (FHWA).
4-28
-------
Q
O
Q
Q
O
0.7
0.6
0.5
0.4
* 0.3
O
1 0.2
I
-------
For on-highway diesel-fueled vehicles, OAQPS estimated a national activity level of 359
billion km for 2000. The activity level for each vehicle type was
Vehicle type Billion kilometers
Light-duty vehicles 6.44
Light-duty trucks 1 6.44
2B-heavy diesel vehicles 33.80
Light heavy-duty vehicles 25.75
Medium heavy-duty vehicles 59.55
Heavy heavy-duty vehicles 217.26
Buses heavy-duty vehicles 9.66
For reference year 1995, FHWA reported that 1,448 billion total vehicle miles (2,330
billion km) were driven by automobiles and motorcycles in the United States. Trucks accounted
for 1,271 billion km (790 billion miles) and buses accounted for 10 billion km (6.2 billion miles)
(U.S. DOC, 1997). In 1992, diesel-fueled trucks accounted for 14.4% of total truck vehicle
kilometers driven (VKD); gasoline-fueled trucks accounted for the remaining 85.6% (U.S. DOC,
1995a). Applying this factor of 14.4% to the 1995 truck estimate of 1,271 billion km results in an
estimate of 183 billion km driven by diesel-fueled trucks in 1995.
All other VKD (2,947 billion km) are assumed to be by gasoline-fueled vehicles
(nondiesel trucks, all automobiles, all buses, and all motorcycles); although a fraction of buses
and automobiles use diesel fuel, the exact numbers are not known. It is further assumed that all of
these kilometers were driven by unleaded gasoline-fueled vehicles because in 1992 only 1.4% of
the gasoline supply was leaded fuel (EIA, 1993). Use of leaded fuel should have declined further
by 1995 because its use in motor vehicles for highway use in the United States was prohibited as
of December 31, 1995 (Federal Register, 1985a).
For reference year 1987, an estimated 3,092 billion km were driven in the United States, of
which trucks accounted for 887 billion km (U.S. DOC, 1995b). Diesel-fueled trucks accounted
for 17.2% of total truck kilometers driven (U.S. DOC, 1995a). Applying this factor of 17.2% to
the 1987 truck kilometer estimate of 887 billion results in an estimate of 153 billion km driven by
diesel-fueled trucks. All other VKD (2,939 billion) are assumed to have been by gasoline-fueled
vehicles. Leaded gasoline accounted for 24.1% of the gasoline supply in 1987 (EIA, 1993); thus,
708 billion km are estimated to have been driven by leaded gasoline-fueled vehicles. The
remaining 2,231 billion km are estimated to have been driven by unleaded gasoline-fueled
vehicles. These mileage estimates are given a high confidence rating because they are based on
U.S. Census Bureau transportation studies.
4-30
-------
4.1.3.2. Activity Information for Off-Road Uses
Although on-road vehicles are the largest users of gasoline and diesel fuel, certain sectors
of the economy account for significant amounts of farm, railroad, marine vessel, and other
off-highway uses. Reference year 2000 activity information for off-highway gasoline and diesel
equipment was estimated by OAQPs from NONROAD model runs prepared for the National
Emissions Inventory. For off-highway gasoline-driven equipment, OAQPS calculated a national
activity level of 23,091.01 million L for 2000. The activity level for each equipment type was
Vehicle type Million liters
Lawn and garden equipment, 4-stroke engines 8,100.78
Pleasure craft, 2-stroke engines 3,607.50
Commercial equipment, 4-stroke engines 3,255.45
Recreational equipment, 2-stroke engines 2,032.77
Recreational equipment, 4-stroke engines 1,782.93
Pleasure craft, 4-stroke engines 1,374.10
Lawn and garden equipment, 2-stroke engines 1,192.40
Industrial equipment, 4-stroke engines 579.16
Construction and mining equipment, 4-stroke engines 473.18
Agricultural equipment, 4-stroke engines 306.62
Industrial equipment, 4-stroke engines; other oil field equipment 124.92
Construction and mining equipment, 2-stroke engines 102.21
Commercial equipment, 2-stroke engines 79.49
Logging equipment, 4-stroke engines 37.85
Logging equipment, 2-stroke engines 26.50
Airport ground support equipment, 4-stroke engines 7.57
Railroad, 4-stroke engines 3.79
Agricultural equipment, 2-stroke engines 3.40
Industrial equipment, 2-stroke engines 1.14
For reference year 2000, OAQPS calculated national activity levels of 40,125.37 million
L for off-highway diesel-driven vehicles, 12,491.86 million L for diesel railroad equipment, and
7,684.39 million L for diesel commercial marine vessels. For diesel commercial marine vessels,
the national activity level comprises port emissions (5,905.24 million L) and underway emissions
(1,968.41 million L). The activity level for each type of diesel railroad equipment was:
Diesel locomotive type Million liters
Class I locomotives 10,561.30
Class Mil locomotives 700.30
Yard locomotives 794.94
Passenger trains 230.91
Commuter trains 215.77
4-31
-------
The following paragraphs define each of the off-road fuel uses listed at the beginning of
this section and present distillate fuel sales (in liters) in each sector for reference years 1987 and
1995 (EIA, 1992, 1997a). For these sectors, the majority of "distillate fuel" sales are diesel fuels;
a small fraction are fuel oils. The activity level information for reference years 1987 and 1995 is
provided for informational purposes only because emission estimates for these years could not be
calculated due to the lack of emission factors.
Farm fuel use includes sales for use in tractors, irrigation pumps, and other agricultural
machinery, as well as fuel used for crop drying, in smudge pots, and for space heating of
buildings. Sales were 11,352.45 million Lin 1987 and 13,158.1 million L in 1995.
Railroad fuel use includes sales to railroads for any use, including diesel fuel for use in
locomotives and fuel used for heating buildings operated by railroads. Sales were 10,788.42 L in
1987 and 12,980.18 Lin 1995.
Marine vessel fuel use includes sales for the fueling of commercial or private boats such
as pleasure craft, fishing boats, tug boats, and oceangoing vessels, including vessels operated by
oil companies. Excluded are sales to the U.S. Armed Forces. Sales were 7,059.79 L in 1987 and
8,854.08 Lin 1995.
Off-highway fuel use includes sales for use in construction equipment, including, e.g.,
earthmoving equipment, cranes, stationary generators, and air compressors, and sales for
nonconstruction off-highway uses such as logging. Sales were 5,905.24 L in 1987 and 8,225.7 L
in 1995.
4.1.3.3 Emission Estimates
Using the results of the studies discussed in Section 4.1.1, separate national annual
emission estimates were developed for vehicles burning leaded gasoline, unleaded gasoline, and
diesel fuel.
Leaded gasoline. The literature indicates that CDD/CDF emissions occur from full
combustion in vehicles using leaded gasoline, and that considerable variation occurs depending, at
least in part, on the types of scavengers used. Marklund et al. (1987) reported emissions ranging
from 20 to 220 pg I-TEQDF/km from four cars fueled with a reference unleaded fuel to which lead
(0.5 g/leaded gal) and a chlorinated scavenger were added. Marklund et al. (1990) reported much
lower emissions in the exhaust of cars using a commercial leaded fuel (0.5 g/L) containing both
1,2-dichloroethane and ethylene dibromide as scavengers (1.1 to 6.3 pg I-TEQDF/km). The
difference in the emission measurements in the 1987 and 1990 studies was attributed to the
different mix of scavengers used in the two studies, which may have resulted in preferential
formation of mixed chlorinated and brominated dioxins and furans.
4-32
-------
Hagenmaier et al. (1990) reported TEQ emissions of 1,080 pg I-TEQDF/L fuel
(approximately 129 pg TEQDF-WHO98/km [108 pg I-TEQDF/km]) from a car fueled with a
commercial leaded fuel (lead content not reported). Bingham et al. (1989) reported emissions
ranging from 1 to 39 pg I-TEQDF/km from four cars using gasoline with a lead content of 1.7 g/L
in New Zealand. The German study reported by Schwind et al. (1991) and Hutzinger et al. (1992)
measured emissions of 52 to 1,184 pg I-TEQDF/L (approximately 7.2 to 142 pg TEQDF-WHO98/km
[5.2 to 118 pg I-TEQDF/km]) for cars under various simulated driving conditions. The tunnel
study by Oehme et al. (1991) estimated that emissions from cars running primarily on leaded
gasoline (70 to 75% of the cars) ranged from 38 to 520 pg Nordic TEQ/km.
The average emission factor (see Table 4-4) was 532 pg TEQDF-WHO98/L (450 pg I-
TEQDF/L), as reported for the tailpipe emission studies performed using commercial leaded fuel
(Marklund et al., 1990; Hagenmaier et al., 1990; Schwind et al., 1991), which presented analytical
results for all 17 toxic CDD/CDF congeners. Assuming an average fuel economy of 10 km/L, this
emission factor was approximately 53 pg TEQDF-WHO98/km (45 pg I-TEQDF/km). A low
confidence rating is assigned to this emission factor because it is based on European fuels and
emission control technologies, which may have differed from U.S. leaded-fuel and engine
technologies, and because the factor is based on tests with only nine cars.
Combining this average emission factor (53 pg TEQDF-WHO98/km [45 pg I-TEQDF/km],
assuming nondetect values were zero) with the estimate for kilometers driven by leaded gasoline-
fueled vehicles in 1987 (708 billion km) suggests that 37.5 g TEQDF-WHO98 (31.9 g I-TEQDF)
were emitted from vehicles using leaded fuels in 1987. Although some on-road vehicles used
leaded fuel in 1995, further use of leaded fuel in motor vehicles for highway use in the United
States was prohibited as of December 31, 1995 (Federal Register, 1985a). In 1992, the last year
for which data are available on consumption of leaded gasoline by on-road vehicles, only 1.4% of
the gasoline supply was leaded gasoline (EIA, 1993). A conservative assumption that 1% of the
total VKD in 1995 (29.5 billion km of a total of 2,947 billion km) was by leaded gasoline-fueled
vehicles, in conjunction with the emission factor of 53 pg TEQDF-WHO98/km (45 pg I-TEQDF/km),
yields an annual emission of 1.6 g TEQDF-WHO98 (1.3 g I-TEQDF) in 1995. These emission
estimates are assigned a low confidence rating on the basis of the low rating for the emission
factor.
Unleaded gasoline. The literature documenting results of European studies indicates that
CDD/CDF emissions from vehicles burning unleaded fuels are lower than emissions from
vehicles burning leaded gas with chlorinated scavengers. It also appears, based on the limited
data available, that catalyst-equipped cars have lower emission factors than do noncatalyst-
equipped cars. Marklund et al. (1987) did not detect CDDs/CDFs in emissions from two catalyst-
equipped cars running on unleaded gasoline at a DL of 13 pg I-TEQDF/km. Marklund et al. (1990)
4-33
-------
reported emission factors of 0.36 and 0.39 pg I-TEQDF/km for two noncatalyst-equipped cars and
an emission factor of 0.36 pg I-TEQDF/km for one catalyst-equipped car. Hagenmaier et al.(1990)
reported an emission factor of 5.1 pg I-TEQDF/km for one noncatalyst-equipped car and 0.7 pg I-
TEQDF/km for one catalyst-equipped car. Schwind et al. (1991) and Hutzinger et al. (1992)
reported emission factors of 9.6 to 17.7 pg I-TEQDF/km for several noncatalyst-equipped cars
tested under various conditions; the reported emission factor range for catalyst-equipped cars was
1 to 2.6 pg I-TEQDF/km.
All automobiles running on unleaded gasoline in the United States are equipped with
catalysts. The average emission factor reported for the tailpipe emission studies performed on
catalyst-equipped cars (Hagenmaier et al. 1990; Schwind et al., 1991; Hutzinger et al., 1992) was
15.6 pg TEQDF-WHO98/L (14.9 pg I-TEQDF/L) (calculated from the mean emission factor in Table
4-6). A low confidence rating is assigned to this emission factor because the European fuels and
emission control technology used may have differed from U.S. fuels and technology and also
because the emission factor range is based on tests with only three catalyst-equipped cars.
OAQPS calculated emissions for reference year 2000 for dioxins and furans from
gasoline-fueled vehicles using the final version of the MOBILE6 model. On-road emissions were
calculated by converting the emission factor of 15.6 pg TEQDF-WHO98/L (14.9 pg I-TEQDF/L) to a
milligram-per-mile basis using a conversion factor of 3.78e-09 and assuming a fuel economy of
21.5 miles/gal. The new emission factor was then multiplied by the corresponding county-level
VMD in miles per year. The off-highway gasoline equipment emission estimates for reference
year 2000 were developed by multiplying the mean emission factor of 15.6 pg TEQDF-WHO98/L
(14.9 pg I-TEQDF/L) by 2000 activity estimates developed from NONROAD model runs prepared
for the National Emissions Inventory. The activity estimates represent county-level gasoline
consumption in gallons. The emission factor was converted from picograms per liter to
milligrams per gallon by multiplying by a conversion factor of 3.78e-09. The use of these
methodologies resulted in national estimates for reference year 2000 of 7 g TEQDF-WHO98 (6.7 g
I-TEQDF) for on-highway gasoline vehicles and 0.36 g TEQDF-WHO98 (0.35 g I-TEQDF) for off-
highway gasoline equipment.
Applying the same emission factors from Gertler et al. (1996, 1998) and assuming an
average fuel economy of 10 km/L yields an emission factor of 1.6 pg TEQDF-WHO98/km (1.5 pgl-
TEQDF/km). Applying this emission factor to the estimate derived for VKD in 1995 by all
gasoline-fueled vehicles (2,947 billion km) suggests that 4.7 g TEQDF-WHO98 (4.4 g I-TEQDF)
were emitted from vehicles using unleaded fuels in 1995. Applying the same emission factors to
the estimate derived above for VKD in 1987 by unleaded gasoline-fueled vehicles (2,231 billion
km) suggests that 3.6 g TEQDF-WHO98 (3.3 g I-TEQDF) may have been emitted in 1987. The
4-34
-------
emission estimates for all reference years were assigned a low confidence rating on the basis of
the low rating given to the emission factor.
Diesel fuel. Limited data are available upon which to base an evaluation of the extent of
CDD/CDF emissions resulting from diesel fuel combustion, and these data address only emissions
from on-road vehicles; no emissions data are available for off-road diesel uses (construction
vehicles, farm vehicles, and stationary equipment). Two U.S. tailpipe studies have been reported:
CARS (1987) and Gullett and Ryan (1997). CARB reported a relatively high emission factor of
676 pg I-TEQDF/km (nondetect values assumed to be zero) for one heavy-duty truck with a fuel
economy of 5.5 km/L at 50 km/hr. Gullett and Ryan reported a range of emission factors for one
diesel truck tested on six highway or city driving routes of 3 to 96.8 pg I-TEQDF/km (mean of 29
pg I-TEQDF/km).
The results of several tailpipe studies conducted in Europe have also been published.
Marklund et al. (1990) reported no emissions at a DL of 100 pg I-TEQDF/L (or 18 pg I-TEQDF/km,
assuming a fuel economy of 5.5 km/L) for one tested truck. Schwind et al. (1991) and Hutzinger
et al. (1992) reported emission factors of 32 to 81 pg I-TEQDF/L (or 6 to 15 pg I-TEQDF/km,
assuming a fuel economy of 5.5 km/L) for a truck engine run under various simulated driving
conditions. Hagenmaier (1994) reported no emissions from a bus at a DL of 1 pg/L fuel
consumed for individual congeners. For diesel-fueled cars, Hagenmaier et al. (1990) reported an
emission factor of 24 pg I-TEQDF/L (or approximately 2.4 pg I-TEQDF/km) for one tested car.
Schwind et al. and Hutzinger et al. reported emission factors of 5 to 13 pg I-TEQDF/km for a car
engine run under various simulated driving conditions.
The tunnel study by Oehme et al. (1991) generated an estimated mean emission factor of
5,100 pg TEQ/km and a range of 720 to 9,500 pg TEQ/km (in units of Nordic TEQ) for diesel-
fueled trucks. Insufficient information was provided in Oehme et al. to enable an exact
calculation of emissions in units of I-TEQDF or TEQDF-WHO98. However, based on the
information that was provided, the mean emission factor in units of TEQ is approximately 5,250
to 5,400 pg I-TEQDF/km. These indirectly estimated emission factors are considerably larger than
those reported in engine studies by Marklund et al. (1990), Schwind et al. (1991), and Hutzinger
et al. (1992); the CARB (1987) diesel truck emission factor falls at the low end of the range.
Although aggregate samples representing several thousand heavy-duty diesel vehicles
were collected in Oehme et al. (1991), several characteristics of the study introduce considerable
uncertainty with regard to the use of the study's results as a basis for estimating emissions in the
United States: (a) heavy-duty vehicles represented only 3 to 19% of total vehicle traffic in the
tunnel; (b) the majority of the light-duty vehicles were fueled with leaded gasoline, the
combustion of which, as noted in Table 4-4, can release considerable amounts of CDD/CDFs; and
4-35
-------
(c) technology differences likely existed between the 1988 Norwegian and the 1987 and 1995U.S.
vehicle fleets.
The tunnel study conducted in Baltimore, MD, by Gertler et al. (1996, 1998) shares the
disadvantages of all tunnel studies relative to studies that directly measured CDDs and CDFs in
tailpipe emissions. Specifically, tunnel studies rely on indirect measurements (rather than tailpipe
measurements), which may introduce bias, and the emission factors calculated from these studies
reflect driving conditions of only the vehicle fleet using the tunnel and not necessarily of the
overall vehicle fleet under other driving conditions.
However, the Gertler et al. study does have strengths that are lacking in the Oehme et al.
(1991) tunnel study, and it has advantages over the two U.S. diesel truck tailpipe studies,
including: (a) the study was conducted (fairly recently) in the United States and thus reflects
current U.S. fuels and technology, (b) virtually no vehicle using the tunnel used leaded gasoline,
(c) the tunnel walls and streets were cleaned 1 week prior to the start of sampling and, in addition,
the study analyzed road dust and determined that resuspended road dust contributed only about
4% of the estimated emission factors, (d) heavy-duty vehicles comprised, on average, a relatively
large proportion (25.7%) of vehicles using the tunnel, and (e) a large number of heavy-duty
vehicles—approximately 33,000—passed through the tunnel during the sampling period, which
generates confidence that the emission factor is representative of interstate trucks.
Considering the strengths and weaknesses of the available emission factor data from the
tailpipe and tunnel studies, the mean TEQ emission factor reported by Gertler et al. (1996,
1998)—182 pg TEQDF-WHO98/km (172 pg I-TEQDF/km)—is assumed to represent the bestcurrent
estimate of the average emission factor for on-road diesel-fueled trucks. This emission factor is
assigned a low confidence rating because it may not be representative of emission rates for the
entire fleet of diesel-fueled trucks under the wide array of driving conditions encountered on the
road.
For reference year 2000, OAQPS developed national CDD/CDF TEQ emission estimates
for on-highway diesel vehicles, off-highway diesel equipment, diesel railroad equipment, and
diesel commercial marine vessels. For on-highway diesel vehicles, OAQPS combined the
calculated mean emission factors from Gertler et al. (1996, 1998) with the OAQPS estimate for
VMD. The picogram-per-kilometer emission factors were first converted to a miligram-per-mile
basis using a conversion factor of 1.61e-09. OAQPS estimated national emissions of 65.4 g
TEQDF-WHO98 (61.7 g I-TEQDF) from on-highway diesel-fueled vehicles for reference year 2000.
For all years, the emissions from diesel vehicles were assigned a low confidence rating because
the emission factors were assigned a low confidence rating.
For off-highway diesel equipment, OTAQ developed the NONROAD emissions model to
estimate emissions from nonroad (off-road) equipment types. However, the NONROAD model
4-36
-------
does not contain emission factors for calculating CDD/CDF emissions. To calculate emissions
for 2000, OAQPS estimated fuel consumption, as reported by the May 2002 "Lockdown C" draft
version of NONROAD, and multiplied this estimate by an average fuel efficiency of 7 miles/gal
and the emission factor from Gertler et al. (1996, 1998). The NONROAD model does not contain
activity estimates for commercial marine vessels and railroad equipment.
OAQPS developed estimates for county-level diesel consumption, in gallons, for diesel
commercial marine vessels and diesel railroad equipment and multiplied these estimates by an
average fuel efficiency of 7 miles/gal and the emission factor from Gertler et al. (1996, 1998).
The results from using these methodologies suggest that 22 g TEQDF-WHO98 (21 g I-TEQDF), 4.3g
TEQDF-WHO98 (4 g I-TEQDF), and 6.8 g TEQDF-WHO98 (6.4 g I-TEQDF) were emitted from off-
highway diesel equipment, diesel commercial marine vessels, and diesel railroad equipment,
respectively, in reference year 2000.
The use of the same emission factors from Gertler et al. (1996, 1998) and an assumption
of an average fuel economy of 10 km/L results in an emission factor of 1.6 pg TEQDF-WHO98/km
(1.5 pg I-TEQDF/km). Applying this factor to the estimate for VKD in 1995 in the United States
by diesel-fueled trucks (183 billion km) suggests that 33.3 g TEQDF-WHO98 (31.5 g I-TEQDF)
were emitted from diesel-fueled trucks in 1995. Combining the same emission factors with the
estimate derived above for VKD in 1987 by diesel-fueled trucks (153 billion km) suggests that
27.8 g TEQDF-WHO98 (26.3 g I-TEQDF) were emitted from diesel-fueled trucks in 1987.
For 1987 and 1995 off-road diesel emissions, EPA used the emission factor from Gertler
et al. (1996, 1998) and multiplied it by an average fuel efficiency of 2.98 km/L (U.S. EPA, 2003b)
and a conversion factor of 1.61e-09 g-km/pg-mile to obtain emission factors of 0.51 ng I-
TEQDF/L and 0.54 ng TEQDF-WHO98/L. These emission factors are assigned a low confidence
rating because they possibly are nonrepresentative of the source. Multiplying these emission
factors by the 1987 activity factors for off-highway equipment (17,278.61 million L), marine
vessels (7,068.35 million L), and railroad use (10,801.5 million L), EPA estimated the following
emissions for 1987: 8.8 g I-TEQDF (9.4 g TEQDF-WHO98) for off-highway equipment, 3.6 g I-
TEQDF (3.8 g TEQDF-WHO98) for marine vessels, and 5.5 g I-TEQDF (5.8 g TEQDF-WHO98) for
railroad use. Similarly, using the 1995 activity factors for off-highway equipment (21,409.71
million L), marine vessels (8,864.81 million L), and railroad use (12,995.91 million L), EPA
estimated the following emissions for 1995: 11 g I-TEQDF (12 g TEQDF-WHO98) for off-highway
equipment, 4.5 g I-TEQDF (4.8 g TEQDF-WHO98) for marine vessels, and 6.6 g I-TEQDF (7 g
TEQDF-WHO98) for railroad use. These emission estimates are given a low confidence rating
because the emission factor may possibly be nonrepresentative of the source.
4-37
-------
4.2. WOOD COMBUSTION
For reference year 1987, wood energy consumption is estimated to have been 2,437trillion
British thermal units (Btu), or 3.2% of the total primary energy consumed in the United States. In
1995, wood fuel (including black liquor solids) provided about 2.6% (2,350 trillion Btu) of the
total primary energy consumed (EIA, 1997b). Wood energy consumption in 2000 is estimated to
have been 2,473 trillion Btu, or 2.5% of the total primary energy consumed (EIA, 2003a). The
industrial sector is the largest consumer of wood fuel, accounting for 65% of total consumption in
1987, 72% in 1995, and 80% in 2000; the residential sector accounted for 35% of total
consumption in 1987, 25% in 1995, and 18% in 2000; and the commercial sector accounted for
approximately 2% of total consumption in all three reference years (EIA, 2003a).
These energy consumption estimates appear to include the energy value of black liquor
solids, which are combusted in recovery boilers by wood pulp mills. In 1987, 1995, and 2000, the
energy values of combusted black liquor solids were 950, 1,078, and 998 trillion Btu, respectively
(American Paper Institute, 1992; American Forest and Paper Association, 1997; letter dated
August 5, 2002, from W. Gillespie, National Council of the Paper Industry for Air and Stream
Improvement, to C. Peck, Versar, Inc.). Subtracting the estimates of black liquor energy values
from the 1987, 1995, and 2000 national totals for wood fuel yields 1,487, 1,272, and 1,475 trillion
Btu, respectively. Assuming that 1 kg of oven-dried wood (2.15 kg of green wood) provides
approximately 19,000 Btu (EIA, 1994), an estimated 78.3, 66.9, and 77.6 million metric tons of
oven-dried wood equivalents were burned for energy purposes in 1987, 1995, and 2000,
respectively. Of these totals, an estimated 44.8, 31.4, and 23 million metric tons were consumed
by the residential sector and an estimated 33.2, 32.6, and 51.5 million metric tons were consumed
by the industrial sector in 1987, 1995, and 2000, respectively.
The following subsections discuss the results of relevant emission studies for the
residential and industrial sectors and present annual TEQ emission estimates for reference years
1987, 1995, and 2000.
4.2.1. Flue Emissions from Wood Combustion (Residential)
Several studies have provided direct measurement of CDDs/CDFs in flue gas emissions
from wood-burning stoves and fireplaces (Schatowitz et al., 1993; Vikelsoe et al., 1993; Bremmer
et al., 1994; Broker et al., 1992; Launhardt and Thoma, 2000; Environment Canada, 2000). The
findings of each of these studies are summarized below.
4.2.1.1. Emissions Data
Schatowitz et al. (1993) measured the CDD/CDF content of flue gas emissions from
several types of wood burners used in Switzerland: a household stove (6 kW), automatic chip
4-38
-------
furnaces (110 to 1,800 kW), and a wood stick boiler (35 kW). Emissions were measured from the
combustion of a variety of wood fuels (natural beech wood, natural wood chips, uncoated
chipboard chips, and waste wood chips from building demolition). The results from the testing of
the household stove are most relevant for assessing releases from residential combustion. The
household stove was tested with the stove door both open and closed. The open-door stove can be
assumed to be representative of fireplaces because both have an uncontrolled draft. Although the
congener and congener group analytical results were not reported, the following emission factors
(dry weight for wood, wet weight for household waste) and emission rates (corrected to 13%
oxygen) for the household stoves and furnaces were reported.
Stoves
• Open-door burning of beech wood sticks: 0.77 ng I-TEQDF/kg
(0.064 ng I-TEQDF/Nm3)
Closed-door burning of beech wood sticks: 1.25 ng I-TEQDF/kg
(0.104ng I-TEQDF/Nm3)
• Closed-door burning of household waste: 3,230 ng I-TEQDF/kg
(114.4ngI-TEQDF/Nm3)
Furnaces
• Natural wood chips: 0.79 to 2.57 ng I-TEQDF/kg
• Chipboard chips (uncoated): 0.29 to 0.91 ng I-TEQDF/kg
• Waste wood chips from building demolition: 26 to 173.3 ng I-TEQDF/kg
Vikelsoe et al. (1993) studied emissions of CDD/CDF congener groups from residential
wood stoves in Denmark. The wood fuels used in the experiments were seasoned birch, beech,
and spruce, equilibrated to 18% absolute moisture. Four different types of stoves (including one
experimental stove) were evaluated under both normal and optimal operating conditions (i.e., well
controlled, with carbon monoxide [CO] emissions as low as possible). Total CDD/CDF
emissions varied widely for the 24 fuel/stove type/operating condition combinations. Emissions
from spruce were about twice as high as those from birch and beech. Surprisingly, the optimal
operating condition led to significantly higher CDD/CDF emissions for two stove types but not
for the other stoves. The predominant congener group for all experiments was TCDF. The
weighted average emission factor and the flue gas concentration for wood stoves (considering
4-39
-------
wood and stove types) were reported to be 1.9 and 0.18 ng Nordic TEQ/Nm3, respectively.
Because Vickelsoe et al. did not measure congener levels, the reported emission factor and
emission rate were estimated by assuming the same congener distribution in each congener group
that had been found for municipal waste incinerators.
Bremmer et al. (1994) reported results of testing performed with a cast-iron wood-burning
stove with a combustion chamber lined with fire refractory clay. Measurements were conducted
at three loads (maximum, average, and minimum) using clean wood as fuel. The emission factors
ranged from 1 to 3.3 ng I-TEQDF/kg (average of about 2.2 ng I-TEQDF/kg). Bremmer et al. also
reported results of testing conducted with a fireplace of a type that is common in the Netherlands.
Measured emission factors from the burning of clean wood ranged from 13 to 28.5 ng I-TEQDF/kg
(average of about 20 ng I-TEQDF/kg). The authors noted that the measured emission factors for
fireplaces were considerably higher than those reported by others (see Broker et al., 1992, below)
and assigned "great uncertainty" to the emission factors.
Broker et al. (1992) reported results of a series of three tests with a wood stove and a
fireplace. The average, minimum, and maximum emission factors measured for the wood stove
tests ranged from 0.53 to 0.94 ng I-TEQDF/kg. The geometric mean of the two average values was
0.71 ng I-TEQDF/kg. The average of the minimum and maximum emission factors measured for
the fireplace tests ranged from 0.2 to 1.06 ng I-TEQDF/kg. The geometric mean of these two
average values is 0.46 ng I-TEQDF/kg.
Launhardt and Thoma (2000) conducted an investigation on organic pollutants from a
domestic heating system using various solid biofuels. Tests were conducted using a multifuel
furnace designed for domestic applications. Table 4-12 shows the average dioxin concentration in
the flue gas for the four fuels used (spruce wood, wheat straw, hay, and triticale). The
concentrations in the flue gas ranged from 52 to 891 pg TEQ/m3.
Table 4-12. Average CDD/CDF concentration in flue gas while burning wood and
crops
Fuel
Spruce wood
Wheat straw
Hay (set-aside land)
Triticale (whole crop)
Concentration (pg TEQ/m3)
52
656
891
52
Number of trials
7
5
4
5
Source: Launhardt and Thoma (2000).
4-40
-------
Environment Canada (2000) conducted a study on the release of dioxins and furans into
the atmosphere by residential wood combustors. The study analyzed two wood stoves believed to
be representative of stoves used in Canada: a conventional wood stove that was popular in the
early 1980s and an advanced combustion, noncatalytic, EPA-certified wood stove. Each stove
was tested using hard maple and black spruce wood. Results from the study ranged from 0.222 to
0.952 ng I-TEQ/kg wood (see Table 4-13). Because these tests took place in North America using
indigenous wood, and they included the analysis of an EPA-certified wood stove, the mean value
of the Environment Canada study (0.5 ng I-TEQ/kg wood) was used to determine the national
emissions estimate for residential burning of clean wood in fireplaces and stoves. This emission
factor is assigned a low confidence rating because it is judged to be nonrepresentative of all
residential wood combustion (e.g., home fireplaces).
Several studies have reported that combustion of treated or manufactured wood in stoves
and fireplaces can result in significantly higher CDD/CDF emission factors. A few researchers
(e.g., Vikelsoe et al., 1993) have reported high CDD/CDF emission rates when pentachlorophenol
(PCP)-contaminated wood is combusted in residential wood stoves and furnaces. The European
Inventory (Quab and Fermann, 1997) used the results of these studies to derive best estimates of
CDD/CDF emission factors for combustion of "slightly contaminated wood (excluding PCP)" and
"PCP-contaminated wood": 50 and 500 ng I-TEQDF/kg, respectively. Although it is likely that
there is some residential combustion of these types of wood in the United States, there are no
corresponding activity level data upon which to base a national annual estimate of emissions.
4.2.1.2. Activity Level Information
In 1987, 22.5 million households in the United States burned wood (EIA, 1991). Wood
was used as the primary heating fuel in 5 million of those households and as a secondary source
for aesthetic purposes (i.e., in fireplaces) in 17.4 million (EIA, 1991, 1997b). Lower numbers
were reported for 1995; wood was reported to be used as the primary fuel in only 3.53 million
households (EIA, 1997b). More rural, low-income households consumed wood as a primary
heating fuel than did other sectors of the population. The majority of these households used
wood-burning stoves as the primary heating appliance. Although fireplaces were the most
common type of wood-burning equipment in the residential sector, only 7% of fireplace users
reported using fireplaces for heating an entire home (EIA, 1991, 1994).
Residential wood consumption was 852 trillion Btu (44.8 million metric tons), or 35% of
total U.S. wood energy consumption, in 1987 and 596 trillion Btu (31.4 million metric tons), or
25% of total U.S. wood energy consumption, in 1995 (EIA, 1997b). An estimated 433 trillion Btu
(23 million metric tons) of wood were consumed in residences in 2000 (EIA, 2003 a). These
production estimates are given high confidence ratings because they are based on recent
government survey data.
4-41
-------
Table 4-13. CDD/CDF concentrations (pg TEQ/kg wood) in emissions from residential wood stoves in Canada
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
TOTAL
Mean Emission Factor
U.S. EPA-certified
Maple
Runl
212
108
21
21
21
2
1
129
22
243
22
15
17
10
2
2
0
848
Run 2
214
138
16
16
16
2
0
134
24
371
31
18
17
10
1
1
0
1,009
Run 3
256
117
17
17
17
2
0
127
23
350
23
20
14
11
2
1
0
997
951
Spruce
Runl
82
41
10
10
10
1
0
95
12
186
12
8
8
8
1
1
0
485
Run 2
110
66
18
18
18
2
0
47
13
149
13
13
13
13
2
2
0
497
Run 3
91
57
14
14
14
2
0
55
17
302
23
15
10
10
2
2
0
628
537
Conventional
Maple
Run
1
68
34
8
8
8
1
0
28
6
85
27
10
7
4
2
1
0
297
Run 2
75
56
13
13
13
1
1
38
4
78
11
8
8
8
2
1
1
331
Run 3
56
47
9
9
9
2
1
36
4
66
10
6
6
6
1
1
0
269
299
Spruce
Runl
63
27
7
7
7
1
0
27
5
54
16
4
4
4
1
0
0
227
Run 2
70
39
7
7
7
2
1
16
2
17
7
7
7
7
1
1
0
198
Run 3
66
41
10
10
10
1
0
18
5
33
18
8
8
8
1
1
0
238
221
J^.
to
Source: Environment Canada (2000).
-------
OAQPS developed emission estimates for residential wood combustion from the results of
a study by EPA's Emission Factor and Inventory Group (U.S. EPA, 2001a). The activity data for
residential wood combustion were based on the type of combustion unit, and the activity data for
wood stoves and fireplaces with inserts were estimated on the basis of total amount of wood
consumed in a year. OAQPS used 1997 national activity data to extrapolate an estimate for 1999
by applying a growth rate factor based on wood energy consumption data from the Energy
Information Administration (EIA). Activity data for fireplaces were estimated on the basis of
number of homes in the U.S. with usable fireplaces, as reported by the U.S. Census Bureau.
OAQPS assumed that the extent of wood consumption in residential combustion units is
directly related to ambient temperature (with more wood consumption in colder climates).
Historical climate data were used to assign each U.S. county to one of five climate zones, as
defined by the National Climatic Data Center. Each climate zone was then assigned a percentage
of total national wood consumption on the basis of information contained in the EIA's Residential
Energy Consumption database.
The consumption in each climate zone was then allocated to individual counties in that
zone. Each county was designated as urban or rural to reflect unit location preferences reported in
the 1999 American Housing Survey, which estimated that 68% of fireplaces are found in urban
areas, compared with 32% in rural areas. An estimated 69% of wood stoves are found in rural
areas, compared with 31% in urban areas. Fireplaces with inserts were evenly split between urban
and rural areas. In each zone, the total urban and rural county wood consumption was summed
and an adjustment was made within the zone for each county's consumption if the urban and rural
totals did not match the expected percentage. These steps resulted in final cordwood consumption
by county, which was converted to tons of wood consumed using a conversion factor of one cord
of wood equaling 1.163 tons.
Wood consumption estimates for stoves and fireplaces with inserts were further
categorized to account for the different designs of units that exist in the marketplace. Different
designs of stoves and inserts have been found to have different levels of emissions. According to
data received from the Hearth Products Association, the three primary types of units currently in
use are noncertified (92% of the stoves manufactured), certified noncatalytic (5.7%), and certified
catalytic (2.3%). These proportions were applied to the national, state, and county cordwood
consumption estimates prior to the application of emission factors.
4-43
-------
Activity levels were estimated to be as follows:
Activity level
Wood combustion category (million metric tons/yr)
Fireplaces 2.79
Fireplaces with inserts, certified catalytic 0.92
Fireplaces with inserts, certified noncatalytic 0.47
Fireplaces with inserts, noncertified 7.64
Noncatalytic wood stoves 0.26
Catalytic wood stoves 0.64
Conventional wood stoves 10.60
4.2.1.3. Emission Estimates
The emission factor used to determine national emission estimates (0.5 ng I-TEQ/kg
wood) was obtained from Environment Canada (2000) because it was the most comprehensive
and recent study. Combining this emission factor with the mass of wood consumed in residences
in 1987, 1995, and 2000 yields annual TEQ air emissions from this source of approximately 22,
15.7, and 11.3 g I-TEQDF, respectively. These estimates are given a low confidence rating for all
years because the emission factor was judged to be of low confidence.
4.2.2. Stack Emissions from Wood Combustion (Industrial)
4.2.2.1. Emissions Data
Congener-specific measurements of CDDs/CDFs in stack emissions from industrial wood-
burning furnaces were measured by CARB at four facilities in 1988 (CARB, 1990b, c, d, e).
Measurements of CDD/CDF congener groups and 2,3,7,8-TCDD and 2,3,7,8-TCDF were
reported for one facility by EPA (U.S. EPA, 1987a). The National Council of the Paper Industry
for Air and Stream Improvement (NCASI) presented congener-specific emission factors for five
boilers tested during burns of bark and wood residue (NCASI, 1995). The average congener-
specific emission factors derived from the four CARB and five NCASI studies are presented in
Table 4-14. Average congener and congener group profiles are presented in Figure 4-5a for the
four CARB studies and in Figure 4-5b for the five NCASI studies.
CARB (1990b) measured CDDs/CDFs in the emissions from a quad-cell wood-fired boiler
used to generate electricity. The fuel consisted of coarse wood waste and sawdust from
nonindustrial logging operations. The exhaust gases passed through a multicyclone before
entering the stack. From this study, the average TEQ emission factor for total CDDs/CDFs was
calculated to be 0.64 ng I-TEQDF/kg of wood burned.
4-44
-------
Table 4-14. CDD/CDF mean emission factors (ng/kg wood) for industrial wood combustors
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total I-TEQDF
Total TEQDF-WH098
Total CDD/CDF
Four facilities tested by CARB
Nondetect set to
zero
0.007
0.044
0.042
0.086
0.079
0.902
6.026
0.673
0.79
0.741
0.761
0.941
0.343
0.45
2.508
0.26
1.587
0.151
1.039
1.748
2.936
6.026
4.275
9.75
7.428
3.747
1.588
0.82
0.84
38.69
Nondetect set to
Vi detection limit
0.016
0.054
0.055
0.096
0.132
0.905
6.026
0.673
0.79
0.741
0.768
0.941
0.35
0.491
2.749
0.344
1.59
0.154
1.039
1.748
2.936
6.026
4.275
9.75
7.428
3.988
1.59
0.85
0.87
38.93
Five facilities tested by NCASI
Nondetect
set to zero
0.066
0.11
0.179
0.191
0.522
0.635
1.317
0.707
0.145
0.159
0.108
0.071
0.064
0.015
0.072
0.017
0.049
1.628
1.958
1.792
1.12
1.317
4.532
1.548
0.536
0.111
0.049
0.4
0.46
14.59
Nondetect set to 1A
detection limit
0.068
0.112
0.183
0.193
0.524
0.637
1.317
0.719
0.149
0.164
0.111
0.073
0.067
0.017
0.074
0.02
0.06
1.629
1.98
1.796
1.132
1.317
4.552
1.549
0.543
0.116
0.06
0.41
0.47
14.67
Nine facilities tested by
CARB and NCASI
Nondetect set to
zero
0.04
0.079
0.115
0.138
0.321
0.745
3.329
0.684
0.406
0.389
0.375
0.418
0.178
0.192
1.062
0.113
0.674
0.969
1.521
1.663
1.821
3.329
4.353
4.93
3.316
1.58
0.674
0.56
0.6
24.16
Nondetect set to
Vi detection limit
0.046
0.084
0.123
0.143
0.342
0.748
0.329
0.69
0.409
0.392
0.379
0.419
0.183
0.209
1.155
0.152
0.681
0.97
1.533
1.665
1.823
0.329
4.364
4.93
3.32
1.674
0.681
0.58
0.62
21.29
CARB = California Air Resources Board
NCASI = National Council for Air and Stream Improvement
Sources: CARB (1990b, c, d, e); NCASI (1995).
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0.05 0.1
0.15
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Ratio (mean congener group emission factor/total CDD/CDF emission factor)
0.05 0.1 0.15 0.2 0.25
0.3
Figure 4-5a. Congener and congener group profiles for air emissions from
industrial wood combustors (nondetects set equal to zero).
Sources: CARB (1990b, c, d, e).
4-46
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0.1 0.2 0.3
0.4
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Ratio (mean congener group emission factor/total CDD/CDF emission factor)
0.05 0.1 0.15 0.2 0.25 0.3
Figure 4-5b. Congener and congener group profiles for air emissions from bleached
Kraft mill bark combustors (nondetects set equal to zero).
Source: NCASI (1995).
4-47
-------
In the second study (CARB, 1990d), CDDs/CDFs in the emissions from two spreader-
stoker wood-fired boilers operated in parallel by an electric utility for generating electricity were
measured. The exhaust gas stream from each boiler was passed through a dedicated ESP, after
which the gas streams were combined and emitted to the atmosphere through a common stack.
Stack tests were conducted when the facility burned fuels allowed by existing permits and when it
burned a mixture of permitted fuel supplemented by urban wood waste at a ratio of 7:3. From this
study, the average TEQ emission factor for total CDDs/CDFs was calculated to be 0.82 ng I-
TEQDF/kg of wood burned.
In the third study (CARB, 1990e), CDDs/CDFs in the emissions from twin fluidized-bed
combustors designed to burn wood chips for the generation of electricity were measured. The air
pollution control device (APCD) system consisted of ammonia injection for controlling nitrogen
oxides and a multicyclone and ESP for controlling PM. During testing, the facility burned wood
wastes and agricultural wastes allowed by existing permits. From this study, the average TEQ
emission factor for total CDDs/CDFs was calculated to be 1.32 ng I-TEQDF/kg of wood burned.
In the fourth study (CARB, 1990f), CDDs/CDFs in the emissions from a quad-cell wood-
fired boiler were measured. During testing, the fuel consisted of wood chips and bark. The flue
gases passed through a multicyclone and an ESP before entering the stack. From this study, the
average TEQ emission factor for total CDDs/CDFs was calculated to be 0.5 ng I-TEQDF/kg of
wood burned.
NCASI (1995) presented stack emission test results for five boilers burning bark or wood
residues. One of these facilities, which was equipped with a multicyclone, normally burned bark
in combination with sludge and coal. Another facility, which was equipped with an ESP,
normally fired pulverized coal. The other three facilities were spreader-stokers equipped with
multicyclones or ESPs. Although stack gas flow rates were obtained during these tests, accurate
measurements of the amounts of bark and wood fired were not made and had to be estimated from
steam production rates. The average TEQ emission factor for these facilities was 0.46 ng TEQDF-
WHO98/kg (0.4 ng I-TEQDF/kg of feed).
The mean of the emission factors derived from the four CARB studies and five NCASI
studies—0.6 ng TEQDF-WHO98/kg wood (0.56 ng I-TEQDF/kg wood), assuming nondetect values
were zero—is used in this document as the most representative of industrial wood combustion.
This emission factor was assigned a medium confidence rating. However, these mean emission
factors may not be appropriate for the combustion of waste wood containing elevated chlorine
content. NCASI (1995) concluded that CDD/CDF emissions from facilities burning salt-laden
wood residue may be considerably higher than those from facilities burning salt-free wood.
Umweltbundesamt (1996) reported the results of stack gas testing at approximately 30
facilities of varying design types burning various types of wood fuel. The author noted that
4-48
-------
CDD/CDF emissions were elevated when the combustion conditions were poor, as evidenced by
elevated CO emissions, or when the fuel contained elevated chlorine levels. Umweltbundesamt
attributed the correlation between elevated CDD/CDF emissions and elevated chlorine content of
the fuel to the fire-retardant effects of chlorine, which may have inhibited complete combustion.
The chlorine content of untreated wood and bark were reported as 0.001 to 0.01% by weight and
0.01 to 0.02% by weight, respectively. Chipboard can contain up to 0.2% chlorine by weight
because of the binding agents used to manufacture the chipboard. Preservative-treated wood and
PVC-coated wood were reported to contain chlorine contents as high as 1.2 and 0.3% by weight,
respectively.
The facility tested by EPA in 1987 (U.S. EPA, 1987a) was located at a lumber products
plant that manufactured overlay panels and other lumber wood products. Nearly all the wood fed
to the lumber plant had been stored in sea water adjacent to the facility and therefore had a
significant concentration of inorganic chloride. The wood-fired boiler tested was a three-cell
dutch oven equipped with a waste heat boiler. The feed wood was a mixture of bark, hogged
wood, and green and dry planer shavings. The exhaust gases from the boiler passed through a
cyclone and fabric filter (FF) prior to discharge from the stack. From this study, an average
emission factor for total CDDs/CDFs of 1,020 ng/kg wood burned (range, 552 to 1,410 ng/kg)
was reported for the three collected samples. An average TEQ emission factor of 17.1 ng
I-TEQDF/kg wood burned (range, 7.34 to 22.8 ng/kg) was estimated by EPA using measured
congener group concentrations and concentrations of 2,3,7,8-TCDD and 2,3,7,8-TCDF. Similar
emission factors were reported by Luthe et al. (1998) from testing conducted during the 1990s at
four Canadian coastal, salt-laden wood-fueled boilers—1.4, 2.6, 17.4, and 27.6 ng I-TEQDF/kg
wood combusted.
The overall average of the five tested facilities in Canada and the United States was 13.2
ng I-TEQDF/kg of wood combusted. The confidence rating assigned to this emission factor is low
because it is based on reporting of limited congener data at one U.S. facility and testing at four
non-U.S. sources and because the fraction of salt-laden wood combusted across facilities is likely
to have been highly variable.
For reference year 2000, NCASI provided congener-specific estimates of CDD/CDF
releases from the pulp and paper industry, including emissions from wood residue-fired boilers
(letter dated August 5, 2002, from W. Gillespie, National Council of the Paper Industry for Air
and Stream Improvement, to C. Peck, Versar, Inc.). The emission factors were taken from
"NCASI Handbook of Chemical Specific Information for SARA (Superfund Amendments and
Reauthorization Act) Section 313 Form R Reporting." The factors provided in the handbook were
compiled from valid test data supplied to NCASI by a variety of sources, including NCASI
member companies that had performed the tests in response to a regulatory program. Data from
4-49
-------
11 bark and wood residue-fired boilers used by the forest products industry were used to calculate
an emissions estimate. Concentrations of emissions from the wood residue-fired boilers were
0.017 jig TEQDF-WHO98/ton wood (see Table 4-15).
Table 4-15. NCASI CDD/CDF TEQ concentrations and emissions for wood
residue-fired boilers
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
TOTAL
Wood-fired boiler emissions
TEQDF-WH098
concentrations
(median
Ug/ton)
0
0
0
4.00e-04
5.00e-04
1.05e-03
5.69e-04
4.40e-03
7.50e-04
5.00e-03
9.00e-04
7.00e-04
2.10e-03
9.00e-04
2.80e-04
1.10e-04
2.10e-05
1.72e-02
Emissions
(ng/yr)
0
0
0
1.68e+07
2.10e+07
4.41e+07
2.39e+07
1.85e+08
3.15e+07
2.10e+08
3.78e+07
2.94e+07
8.82e+07
3.78e+07
1.18e+07
4.62e+06
8.82e+05
7.42e+08
Wood-fired boiler ash not landfilled
(72% of total ash landfilled)
TEQ
concentrations
(ng/kg)
1.84e+00
1.73e+00
3.25e-01
4.28e-01
2.60e-01
4.01e-01
1.90e-02
4.20e+00
3.35e-01
3.23e+00
2.21e-01
1.60e-01
5.40e-02
3.80e-02
4.10e-03
1.30e-03
5.40e-04
1.32e+01
Emissions
(ng/yr)
3.06e+08
2.88e+08
5.41e+07
7.12e+07
4.33e+07
6.71e+07
3.96e+06
7.06e+08
5.56e+07
5.37e+08
3.68e+07
2.66e+07
8.98e+06
6.32e+06
6.82e+05
2.16e+05
1.64e+05
6.19e+08
(ash not landfilled)
2.21e+09
(ash landfilled)
NCASI = National Council for Air and Stream Improvement
Source: Letter dated August 5, 2002, from W. Gillespie, National Council of the Paper Industry for Air and Stream
Improvement, to C. Peck, Versar, Inc.
4.2.2.2. Activity Level Information
In 1987, 33.2 million metric tons of wood were burned for fuel in industrial furnaces. In
1995, industrial wood consumption totaled 32.6 million metric tons. EIA (2003b) estimated that
industrial wood consumption totaled 1988 trillion Btu (104.6 million metric tons) in 2000. This
4-50
-------
total becomes 51.5 million metric tons with the removal of kraft black liquor combustion. The
majority of wood fuel consumed in the industrial sector consists of wood waste (chips, bark,
sawdust, and hogged fuel). Consumption in the industrial sector is dominated by two industries:
paper and allied products and lumber and wood products (EIA, 1994). These activity level
estimates are assigned a high confidence rating because they are based on recent government
survey data.
Activity level data on combustion of salt-laden wood are not normally collected, even
though the associated emission factor is greater than the factor associated with nonsalt-laden
wood. Nonetheless, attempts have been made to estimate this activity level. NCASI combined
the results from a 1995 survey of combustion units in the pulp and paper industry with those from
an ad hoc telephone survey of mills in the Pacific Northwest (Oregon and Washington) to produce
a conservative estimate of the amount of salt-laden wood burned at U.S. pulp and paper mills in
1995: 254,000 metric tons (0.8% of the estimated 32.6 million metric tons of industrial wood
consumed that year). NCASI suspected that a similar fraction of industrial wood combusted in
1987 by pulp and paper mills was salt laden (letter dated October 8, 1998, from W. Gillespie,
National Council of the Paper Industry for Air and Stream Improvement, to G. Schweer, Versar,
Inc.).
For purposes of the NCASI survey, salt-laden wood was defined as wood that had been
transported, stored, or otherwise exposed to saltwater prior to being processed as fuel. None of
the three responding mills in Oregon reported the use of salt-laden wood. Eight of the 13
responding mills in Washington reported some combustion of salt-laden wood. Of the total wood
consumed in the Washington mills, 17% was estimated to be salt-laden wood.
As noted above, the majority of industrial wood combustion (97%) occurs in two
industries: the paper and allied products industry and the lumber and wood products industry.
The relative amount of wood combusted by each of these two industries was the same in 1990 and
1992, the only years for which these statistics are readily available (EIA, 1991, 1994). It can be
assumed that the percentage of total wood combusted nationally by the lumber and wood products
industry that is salt laden is the same percentage as for the paper and allied products industry,
0.8%; therefore, the total percentage of wood combusted by industry that is salt laden is 1.6%.
For reference years 1987, 1995, and 2000, this equates to 0.5, 0.5, and 0.8 million metric tons,
respectively. These activity level estimates are assigned a low confidence rating because they are
possibly nonrepresentative of the activity levels for the source category combusting salt-laden
wood.
4.2.2.3. Emission Estimates
Applying the average TEQ emission factor from the four CARB and five NCASI studies
(0.6 ng TEQDF-WHO98/kg wood [0.56 ng I-TEQDF/kg wood]) to the estimated quantities of
nonsalt-laden wood burned by industrial facilities in 1987 (33.2 million metric tons), 1995 (32.6
4-51
-------
million metric tons), and 2000 (51.5 million metric tons) yields estimated TEQ emissions to air of
19.9 g TEQDF-WH098 (18.6 g I-TEQDF) in 1987, 19.6 g TEQDF-WHO98 (18.3 g I-TEQDF) in 1995,
and 30.9 g TEQDF-WHO98 (28.8 g I-TEQ) in 2000.
Applying the average TEQ emission factor from the five studies on boilers combusting
salt-laden wood (13.2 ng I-TEQDF/kg wood) to the estimated quantities of salt-laden wood burned
by industrial facilities in 1987 (0.5 million metric tons), 1995 (0.5 million metric tons), and 2000
(0.8 million metric tons) yields estimated TEQ emissions to air of 6.6 g I-TEQDF in both 1987 and
1995 and 10.6 g I-TEQDF in 2000.
Total emissions for 1987, 1995, and 2000 are estimated to have been 26.5, 26.2, and 41.5
g TEQDF-WHO98 (25.2, 24.9, and 39.4 g I-TEQDF), respectively. Of the 2000 estimate, NCASI
estimated that 0.74 g TEQDF-WHO98/yr of dioxins were emitted from pulp and paper wood-fired
boilers (letter dated August 5, 2002, from W. Gillespie, National Council of the Paper Industry for
Air and Stream Improvement, to C. Peck, Versar, Inc.). As noted above, the total emissions are
based on tests conducted at nine facilities in two industries that account for 97% of total industrial
wood fuel combustion. The remaining 3% of industrial combustion and the combustion of wood
by the commercial sector (for which no reliable activity level estimates are available) may not be
well represented by the emission factors used above, particularly if poorly controlled combustors
or treated wood (e.g., treated with PCP or plastics) are burned. The emission estimates for 1987,
1995, and 2000 are given a low confidence level because the activity level estimates were
assigned a low confidence rating.
4.2.3. Solid Waste from Wood Combustion (Residential and Industrial)
The measurement of CDDs/CDFs in chimney soot and bottom ash from wood-burning
stoves and fireplaces has been reported by several researchers (Bumb et al., 1980; Nestrick and
Lamparski, 1982, 1983; Clement et al., 1985; Bacher et al., 1992; Van Oostam and Ward, 1995;
and Dumler-Gradl et al., 1995).
Bumb et al. (1980) detected TCDDs (nondetect to 0.4 |ig/kg), HxCDDs (0.2 to 3 |ig/kg),
HpCDDs (0.7 to 16 jig/kg), and OCDD (0.9 to 25 jig/kg) in residues from the wall of a home
fireplace and from the firebrick of another home fireplace; for lack of a suitable analytical method,
analysis was not performed for PeCDDs. Neither of the fireplaces sampled by Bumb et al. had
burned preservative-treated wood.
Nestrick and Lamparski (1982, 1983) expanded the research of Bumb et al. by conducting
a survey of CDD concentrations in chimney soot from residential wood-burning units in three
rural areas of the United States. Samples were collected from the base of six chimneys in each of
the three study areas. Samples were not collected from units where any type of treated or
manufactured wood had been burned. For lack of a suitable analytical method, analysis was not
performed for PeCDDs. The results of this survey are summarized in Table 4-16. There was
4-52
-------
Table 4-16. CDD/CDF concentrations (ng/kg) in residential chimney soot from
wood stoves and fireplaces
Congener/
congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQcf
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
U.S. east
region3
66
NR
250e
250e
208
1,143
2,033
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
3,950
NR
>150
>98
1,987
NR
2,183
2,104
2,033
NR
NR
NR
NR
NR
8,307
U.S. west
region3
13.3
NR
522e
522e
282
1,653
2,227
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
5,219
NR
>165
>163
269
NR
4,273
3,243
2,227
NR
NR
NR
NR
NR
10,012
U.S.
central
region3
66
NR
l,831e
l,831e
1,450
6,160
13,761
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
21,437
NR
>286
>81
1,511
NR
14,243
12,603
13,761
NR
NR
NR
NR
NR
42,118
German
farmhouseb
150
70
35
60
30
90
90
930
560
590
330
400
70
200
490
40
70
525
3,680
720
355
3,900
880
600
200
90
13,400
6,100
3,200
720
70
29,160
Canadian
wood
stovec
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
ND(10)
ND(10)
ND(50)
100
200
ND(10)
ND(10)
ND(50)
ND(50)
ND(50)
300
Canadian
fireplace0
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
ND(10)
500
1,700
500
400
300
1,400
1,700
400
100
7,000
Canadian
wood
stove11
ND(12)
70
ND(10)
625
281
948
530
235
58
68
51
57
8
24
97
20
41
2,454
659
211
246
11
608
3,450
1,550
530
1,010
948
482
154
41
8,784
^Source: Nestrick and Lamparski (1982, 1983); mean values listed, six samples collected in each region.
bSource: Bacher et al. (1992).
°Source: Clement et al. (1985).
dSource: Van Oostdam and Ward (1995); mean of two samples, nondetect values assumed to be zero.
eAnalytical method could not distinguish between congeners; listed value is the sum of both congeners.
ND = Not detected (value in parenthesis is the reported detection limit)
NR = Not reported
4-53
-------
wide variation in the results across soot samples, with standard deviations for congeners and
congener groups often equal to or exceeding the mean value; however, CDDs in each congener group
were detected in the soot from almost all sampled units. The authors concluded that the results did
not appear to present any easily discernible patterns with respect to geographic region, furnace
operational parameters, or wood fuel type. They attributed the wide variability observed to
differences in design of the units, which affected the sampling point or the conditions at the sampling
point, and possible contamination of the fuel wood.
Clement et al. (1985) analyzed chimney soot and bottom ash from residential wood stoves
and fireplaces in Canada. The CDD/CDF congener concentrations are presented in Table 4-16 (soot)
and Table 4-17 (bottom ash). CDD/CDF congeners were detected in all samples analyzed, although
the relative amounts of the different congener groups varied considerably and inconsistently between
wood-burning unit types and between ash and soot samples from the same unit.
Table 4-17. CDD/CDF concentrations (ng/kg) in bottom ash from residential
wood stoves and fireplaces
Congener/congener
group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Canadian wood
stove ash
ND (10)
ND (10)
ND (50)
300
2,600
9,100
2,200
1,000
700
ND (50)
15,900
Canadian wood
stove ash
100
3,000
10,000
1,200
900
400
4,600
9,300
1,000
100
30,600
Canadian
wood stove ash
100
200
700
500
100
100
200
500
300
ND (50)
2,700
Canadian
fireplace ash
ND (10)
ND (10)
300
2,000
3,100
ND (10)
ND (10)
100
400
100
6,000
"No values were reported for individual congeners or for total 2,3,7,8-CDD, 2,3,7,8-CDF, or total TEQ.
ND = Not detected (value in parenthesis is the reported detection limit)
Source: Clement etal. (1985).
Clement et al. also presented total CDD/CDF concentration data for bottom ash from
open-air burning of wood. No analyses were reported for individual congeners. The results for
the congener groups are shown below. The quantity of ash produced by the open-air burning test
was not provided; hence, it is not possible to readily determine the quantities of CDDs/CDFs
disposed of.
4-54
-------
Congener group Concentration (|ig/kg)
TCDDs 0.8
PeCDDs 4.2
HxCDDs 7.2
HpCDDs 11
OCDD 10
TCDFs 2.2
PeCDFs 7.6
HxCDFs 8.2
HpCDFs 11
OCDF 1.7
Bacher et al. (1992) characterized the full spectrum (mono through octa substitution) of
CDD/CDF and BDD/BDF congeners in the soot from an old farmhouse in southern Germany.
The chimney carried smoke from an oven that had used untreated wood at the rate of about
5 m3/yr for more than 10 yr. The sample was taken during the annual cleaning by a chimney
sweep. The only BDF detected was mono-BDF (230 ng/kg). No BDDs, BCDDs, or BCDFs were
detected at a DL of 20 ng/kg. The results for the tetra- through octa-CDDs/CDFs are presented in
Table 4-16. The results indicate that CDFs exceeded the CDDs in each congener group except
octa. Also, the lower-chlorinated congener groups exceeded the higher-chlorinated congener
groups for both the CDDs and the CDFs. The TEQ content of the chimney soot was 755 ng
TEQDF-WHO98/kg (720 ng I-TEQDF/kg), of which less than 30% was due to CDDs.
Van Oostdam and Ward (1995) analyzed soot from two wood stoves in British Columbia,
Canada. The average TEQ concentration was 246 ng TEQDF-WHO98/kg (211 ng I-TEQDF/kg).
The congener-specific results are presented in Table 4-16. The soot from a wood stove burning
salt-laden wood in a coastal area was found to have an I-TEQDF content of 7,706 ng I-TEQDF/kg,
or 20 to 90 times more than the concentrations found in the soot from the other two tested stoves.
Dumler-Gradl et al. (1995) analyzed chimney soot samples collected by chimney sweeps
from 188 residences in Bavaria, Germany. The summary results of the survey, the largest
published survey of its kind to date, are presented in Table 4-18. As in Nestrick and Lamparski
(1982, 1983) and Clement et al. (1985), CDDs/CDFs were detected in all samples; however, there
was wide variability in total TEQ concentrations within and across unit type/fuel type
combinations.
The Washington State Department of Ecology (1998) reported CDD/CDF congener data
for ash from hog fuel boilers at three paper mills. The data were compiled and evaluated to
determine total I-TEQ concentrations and loading. Nondetect values were included as zero, one-
half the DL, or at the DL. The results, assuming nondetect values are at zero, are shown below.
4-55
-------
Table 4-18. CDD/CDF concentrations in chimney soot (Bavaria, Germany)
Unit type
Oven
Tiled stove
Heating system
Oven
Tiled stove
Oven
Fuel type
Wood
Wood
Wood
Wood/coal
Wood/coal
Wood, wood/coal,
waste
Number of
samples
33
39
9
27
5
5
Concentration
(ngI-TEQDF/kg)
Minimum
10.4
4
16.9
77.3
53.1
116.3
Mean
2,015
3,453
1,438
2,772
549
6,587
Maximum
15,849
42,048
20,450
10,065
4,911
10,652
Source: Dumler-Gradletal. (1995).
Location
Daishowa America,
Port Angeles
Ft. James
Rayonier
I-TEQDF
0.31
35.4
12,640
1,150
2,299
225
I-TEQDF
(mg/day)
0.012
0.544
68.9
6.27
12.5
1.23
Type of residual
Mixed ash
Fly ash
Filter ash
Vacuum filter and grate
Filter ash
Fly ash
Pohlandt and Marutzky (1994) presented CDD/CDF concentration data for various types
of ash (bottom, furnace, boiler, and fly) from 12 wood-burning boilers. The fly ash samples from
two wood-working industry boilers appeared to have the greatest concentrations of CDDs/CDFs.
Table 4-19 lists the average congener concentration for the two boilers. Three boiler bottom ash
samples contained detectable amounts of only total HpCDDs/HpCDFs and OCDD/OCDF. All
the other boiler samples were from boilers that burned copper/chrome/boron-impregnated woods.
These samples had total TEQs (assumed to be I-TEQs) ranging from 0.07 to 89 ppt, the highest
being for the fly ash samples (52 and 89 ppt). The quantities produced by the boilers that were
tested were not reported; hence, it is not possible to readily determine the quantities of
CDDs/CDFs disposed of.
The results of analyses of two ash samples from wood-burning facilities in New
Hampshire were reported in a facsimile dated January 23, 2001, from Andrew Carpenter,
Resource Management, Inc., to Stephen Schwartz, Versar, Inc. Both samples were from the
4-56
-------
Table 4-19. CDD/CDF concentrations (ng/kg) in fly ash from wood-working
industry
Congener/congener group
2,3,7,8-TCDD
Total TCDD
1,2,3,7,8-PeCDD
Total PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Total HxCDD
1,2,3,4,6,7,8-HpCDD
Total HpCDD
Total OCCD
2,3,7,8-TCDF
Total TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Total PeCDF
,2,3,4,7,8-HxCDF
,2,3,7,8,9-HxCDF
,2,3,6,7,8-HxCDF
Total HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
Total HpCDF
Total OCDF
Estimated TEQ
Average
concentration
<15
1,730
100
1,250
130
150
140
750
280
470
300
130
1,300
100
120
790
40
40
<10
150
320
<10
570
60
I-TEQDF
<15
50
13
15
14
3
0.3
13
5
60
4
4
<1
3
<0.1
0.06
89-90
TEQDF-WH098
<15
100
13
15
14
3
0.03
13
5
60
4
4
<1
3
<0.1
0.006
89-90
Source: Pohlandt and Marutzky (1994).
burning of clean (i.e., untreated) wood chips, sawdust, and bark. The first sample was a
combination of fly ash and bottom ash. The second sample was fly ash only, but it was a
combination of fly ash from two wood-burning boilers. For the first sample, none of the 2,3,7,8-
substituted congeners were detected at DLs that ranged from 0.98 ng/kg for 2,3,7,8-TCDD and
2,3,7,8-TCDF to 9.8 ng/kg for OCDD and OCDF. (All other congeners had a DL of 4.9 ng/kg.)
For the second sample, all but two congeners were below DLs (which ranged from 0.379 to 0.831
ng/kg). The two congeners that exceeded DLs were OCDD, at 1.261 ng/kg, and 1,2,3,4,6,7,8-
HpCDF, at 1.022 ng/kg. For this sample, assuming that the nondetected congeners were not
4-57
-------
present, the I-TEQDF concentration was 0.011 ng/kg. The quantities of the ash produced were not
reported.
In a CARB report of emissions from a wood waste-fired incinerator (CARB, 1990b), data
are given for CDDs and CDFs for four ash samples. The concentrations of 2,3,7,8-substituted
CDD/CDF congeners for each of those four tests were all below the DLs except for OCDD, which
was detected in three samples at concentrations of 14, 18, and 32 ng/kg, and 2,3,7,8-TCDF, which
was detected in one sample at a concentration of 2.2 ng/kg. The DLs for each CDD and CDF
congener ranged from 0.63 ppt (for 2,3,7,8-TCDD) to 9.5 ppt (for HpCDF congeners). Total
CDD and CDF values were given for each of the four samples. However, those values assumed
that nondetected congeners were at the DL level. Consequently, the total CDD and total CDF
values were biased high. The average of the four total CDD values was 28.8 ng/kg (range, 20.3 to
44 ng/kg). The average of the four total CDF values was 21.9 ng/kg (range, 16 to 26.9 ng/kg).
In CARB (1990d), data are presented for CDDs/CDFs for several samples of ESP waste
ash from a wood-fired boiler. The report provides sample results for 2 weeks of sampling
conducted at the facility. During the first week, the boiler burned fuels that were allowed by the
facility permit; during the second week, the boiler burned a mixture containing 70% permitted
fuel and 30% urban wood wastes. For the six samples collected over the 3 days of the first week,
many of the concentrations of CDD/CDF congeners in the ESP ash were below the DLs. The
reported CDD concentrations in ESP waste ash ranged from 24 to 264 ng/kg, and the CDF
concentrations ranged from 12 to 151 ng/kg. However, those values assumed that nondetected
congeners were present at the detection level. One sample did not have any nondetect values for
CDDs. The total CDD concentration for this sample was 264 ng/kg, or about 11.4 ng/kg TEQDF-
WHO98 (8.3 ng/kg I-TEQDF). The TEQDF-WHO98 and I-TEQDF CDF concentrations for this
sample were both less than 1.5 ng/kg. These values were less than 1 ng/kg for the other five
samples. All of the samples had some nondetects for the CDF analysis.
Six samples were also collected over 3 days during the second week of sampling, when the
70/30 permitted/urban wood waste mix was burned. For the samples from the second week, the
CDD concentrations in ESP waste ash ranged from 1,365 to 3,190 ng/kg, and the CDF
concentrations ranged from 2,866 to 11,282 ng/kg. The study authors assumed that nondetected
congeners were present at the detection level; however, this is a reasonable estimate for this data
set because there was only one nondetect value. Table 4-20 presents the average congener
concentrations for these samples. The report did not present quantities of ESP ash produced by
the boiler; therefore, it is not possible to readily determine the quantities of CDDs/CDFs disposed
of.
4-58
-------
Table 4-20. CDD/CDF concentrations (ng/kg) in electrostatic precipitator waste ash
from wood-fired industrial boiler
Congener/congener group
2,3,7,8-TCDD
Total TCDD
1,2,3,7,8-PeCDD
Total PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Total HxCDD
1,2,3,4,6,7,8-HpCDD
Total HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
Total TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Total PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
Total HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Total HpCDF
Total OCDF
Total CDF
Estimated TEQ
Average
concentration
17.85
239
30.67
226.83
20.33
26.33
23.33
300
325
706.67
786.67
2,439.17
285
2,713.33
154.5
641.67
2,666.67
244.83
179.67
296.67
7.28
1,520
147.67
21.33
248.33
48.33
7,196.67
I-TEQDF
17.85
15.33
2.03
2.63
2.33
3.25
0.79
44.22
28.5
7.73
320.83
24.48
17.97
29.67
0.73
1.48
0.21
0.05
431.64
475.64
TEQDF-WH098
17.85
30.67
2.03
2.63
2.33
3.25
0.08
58.85
28.5
7.73
320.83
24.48
17.97
29.67
0.73
1.48
0.21
0
431.6
490.44
Source: CARB (1990d).
Appendix n in Luthe et al. (1998) shows TEQ concentrations (assumed to be I-TEQDF) in
ashes collected from APCDs from "salt-laden" wood steam boilers. The I-TEQDF content of ashes
from three of the primary multiclone hoppers varied significantly: 0.0978, 0.186, and 9.375
4-59
-------
Hg/kg. Two samples of ash were taken from the secondary multiclone hoppers. The secondary
multiclone removes dust from the primary multiclone emissions; therefore, the ash is finer than
primary dust. The I-TEQDF values for the ash samples were 1.073 and 20.879 |ig/kg. The I-
TEQDF values for two samples taken from the ESP that collected dust from the secondary
multiclone emissions, which therefore was finer than multiclone dust, were 3.926 and 8.044
jig/kg. No data were given for individual congeners. In fact, because the reference discusses only
"dioxins," it is unclear whether the TEQ data are for CDDs or for CDDs plus CDFs. Quantities of
collected ash were not given.
Table II in a report by Luthe et al. (1996) presents data for the "TEQs" (assumed to be I-
TEQs) on particulates from a secondary collection device for boilers at four paper mills burning
salt-laden wood. Eight data points were given (two for each mill), the average of which was 3.6
jig/kg. The range of values was 1.3 to 8 |ig/kg. As in Luthe et al. (1998), no data were given for
individual congeners. It is also unclear whether the TEQ data were for CDDs or for CDDs plus
CDFs. Quantities of collected ash were not given.
Table 5-16 in the National Dioxin Survey (U.S. EPA, 1987a) contains data indicating that
the bottom ash from wood combustion from one source (it is not indicated whether it was a boiler)
contained 140 ng/kg of 2,3,7,8-TCDD, 138,200 ng/kg of CDDs, and 7,400 ng/kg of CDFs. For a
second wood combustion source, the ash contained no detectable 2,3,7,8-TCDD, but it did contain
about 125 ng/kg of CDDs and nondetectable levels of CDFs. The FF dust from the second source
contained 100 ng/kg of 2,3,7,8-TCDD, 1,143,600 ng/kg of CDDs, and 315,600 ng/kg of CDFs.
Specific data for congeners and for ash/dust quantities were not given.
NCASI also provided information on emissions from wood residue boiler ash for reference
year 2000 (letter dated August 5, 2002, from W. Gillespie, National Council of the Paper Industry
for Air and Stream Improvement, to C. Peck, Versar, Inc.). As with the boiler emissions,
emission factors for the boiler ash were taken from "NCASI Handbook of Chemical Specific
Information for SARA Section 313 Form R Reporting." Total TEQ concentrations were
estimated to be 13.2 ng/kg. Because 72% of the total ash produced is landfilled, emission
estimates were 2.21 g TEQDF-WHO98/yr for ash landfilled and 0.62 g/yr for ash not landfilled (see
Table 4-15). It is not known at this time whether the amount of dioxin in nonlandfilled ash results
in an environmental release. Therefore, this value was not included in the inventory.
4.3. OIL COMBUSTION
The two major categories of fuel oils that are burned by combustion sources are distillate
oils and residual oils. These oils are further distinguished by grade: numbers 1 and 2 are distillate
4-60
-------
oils, 5 and 6 are residual oils, and 4 is either distillate oil or a mixture of distillate and residual
oils. Number 6 fuel oil is sometimes referred to as Bunker C. Distillate oils are more volatile and
less viscous than residual oils. They have negligible nitrogen and ash content and usually contain
less than 0.3% sulfur (by weight). Distillate oils are used mainly in domestic and small
commercial applications. The heavier residual oils (5 and 6), being more viscous and less volatile
than distillate oils, must be heated for ease of handling and to facilitate proper atomization.
Because residual oils are produced from the residue after the lighter fractions (gasoline, kerosene,
and distillate oils) are removed from the crude oil, they may contain significant quantities of ash,
nitrogen, and sulfur. Residual oils are used mainly in utility, industrial, and large commercial
applications (U.S. EPA, 1995a).
4.3.1. Institutional/Commercial and Residential Oil Combustion
No testing information on the CDD/CDF content of air emissions from institutional/
commercial or residential oil-fired combustion units in the United States could be located.
However, EPA (U.S. EPA, 1997a) estimated CDD/CDF congener group and TEQ emission
factors using average CDD/CDF concentrations reported for soot samples from 21 distillate fuel
oil-fired furnaces used for central heating in Canada and a particulate emission factor for distillate
fuel oil combustors (300 mg/L oil) obtained from AP-42 (U.S. EPA, 1995a). The TEQ emission
factor estimate in U.S. EPA (1997a) was derived using the calculated emission factors for 2,3,7,8-
TCDD, 2,3,7,8-TCDF, and the 10 congener groups. These emission factors are presented in Table
4-21 and the congener group profile is presented in Figure 4-6.
For reference year 1987, assuming a barrel of oil contains 42 gallons, distillate fuel oil
sales to the residential sector and the commercial sector totaled 28.1 billion L (177 million
barrels) and 16.2 billion L (102 million barrels), respectively (EIA, 1999). Residual oil sales to
the commercial sector in 1987 totaled 6.7 billion L (42 million barrels) (EIA, 1999). Using the
emission factor presented in Table 4-21 (150 pg I-TEQDF/L oil combusted [190 pg TEQDF-
WHO98/L oil combusted]), EPA estimated that 4.22 g I-TEQDF (5.35 g TEQDF-WHO98) were
emitted in 1987 for the residential sector. For the institutional/commercial sector, EPA estimated
TEQ emissions of 1.34 g I-TEQDF (1.54 g TEQDF-WHO98) for residual oil and 3.24 g I-TEQDF
(3.73 g TEQDF-WHO98) for distillate oil for 1987, using an emission factor of 200 pg I-TEQDF/L
oil combusted (230 pg TEQDF-WHO98/L oil combusted) (see Section 4.3.2). Because the
representativeness of the emission factor to 1987 emissions is uncertain and may not be
representative, this estimate is assigned a low confidence rating.
4-61
-------
Table 4-21. Estimated CDD/CDF emission factors for oil-fired residential
furnaces
Congener/
congener group
2,3,7,8-TCDD
Total PeCDD
Total HxCDD
Total HpCDD
OCDD
2,3,7,8-TCDF
Total PeCDF
Total HxCDF
Total HpCDF
OCDF
TOTAL
Mean
facility
emission
factor
(PS/L oil)
56
82
66
63
66
53
420
170
73
30
WHO-TEF
1
1
0.1
0.01
0.0001
0.1
0.05
0.1
0.01
0.0001
Emission
factor (pg
TEQDF-
WHO98/L
oil)
56
82
7
1
0
5
21
17
1
0
190
I-TEF
1
0.5
0.1
0.01
0.001
0.1
0.05
0.1
0.01
0.001
Emission
factor (pg
I-TEQDF/L
oil)
56
41
7
1
0
5
21
17
1
0
149
Source: U.S. EPA(1997a).
For reference year 1995, a low confidence estimate of potential national TEQ emissions
from this source category was made using the same emission factors as for the 1987 estimates.
Distillate fuel oil sales to the residential and commercial sector totaled 26.2 and 13.5 billion L,
respectively, in 1995 (EIA, 1997a). Applying the respective emission factors to these fuel oil
sales estimates results in estimated emissions of 3.93 g I-TEQDF (4.98 g TEQDF-WHO98) for the
residential sector and 2.7 g I-TEQDF (3.11 g TEQDF-WHO98) for the institutional/commercial
sector in 1995. Residual oil sales to the commercial sector in 1995 totaled 3.7 billion L (23
million barrels) (EIA, 1999). Applying the emission factor of 200 pg I-TEQDF/L oil combusted
(230 pg TEQDF-WHO98/L oil combusted) (see Section 4.3.2) yields TEQ emissions of 0.73 g I-
TEQDF (0.84 g TEQDF-WHO98) for residual oil in 1995.
For reference year 2000, EPA/OAQPS developed national emission estimates for residual
oil and distillate oil consumed in institutional/commercial heating and distillate oil consumed in
residential heating. EPA used state-level 2000 activity data (EIA, 1999), which were allocated to
4-62
-------
Ratio (congener group emission factor/total CDD/CDF emission factor)
0.1 0.2 0.3
0.4
Figure 4-6. Congener group profile for air emissions from residential
oil-fueled furnaces.
Source: U.S. EPA(1995b).
counties by the 1999 year county-to-state proportion of employment for numerous SIC codes, as
identified in the U.S. Census Bureau's 2000 census. EPA estimated that 2.82 billion L of residual
oil and 12.7 billion L of distillate oil were consumed in institutional/commercial heating in 2000.
Applying the emission factor of 200 pg I-TEQDF/L oil combusted (230 pg TEQDF-WHO98/L oil
combusted) (see Section 4.3.2) to these activity levels yields TEQ emissions of 0.56 g I-TEQDF
(0.65 g TEQDF-WHO98) for residual oil and 2.53 g I-TEQDF (2.92 g TEQDF-WHO98) for distillate
oil for 2000. EPA/OAQPS estimated that 23.9 billion L of distillate oil were consumed for
residential heating in 2000. Using the emission factors discussed above from U.S. EPA (1997a),
EPA/OAQPS estimated emissions of 3.59 g I-TEQDF (4.54 g TEQDF-WHO98) from distillate oil
used for residential heating in 2000.
4-63
-------
4.3.2. Utility Sector and Industrial Oil Combustion
Preliminary CDD/CDF emission factors were reported for oil-fired utility boilers using the
results of boiler tests conducted over several years (U.S. EPA, 1997a). The data are for a
composite of various furnace configurations and APCD systems. Table 4-22 lists the median
emission factors presented by EPA. The congener and congener group profiles based on these
data are presented in Figure 4-7. The median I-TEQDF emission factor was reported to be 366 pg
TEQDF-WHO98/L (314 pg I-TEQDF/L) oil burned.
In 1993, the Electric Power Research Institute (EPRI) sponsored a project to gather
information of consistent quality on power plant emissions. The Field Chemical Emissions
Measurement (FCEM) project included testing of two cold-sided, ESP-equipped, oil-fired power
plants for CDD/CDF emissions (EPRI, 1994). The averages of the congener and congener group
emission factors reported for these two facilities are presented in Table 4-22. The average TEQ
emission factor was 93.6 pg TEQDF-WHO98/L (83.1 pg I-TEQDF/L) oil burned when nondetect
values were treated as zero.
The TEQ emission factors reported by EPRI (1994) were less than the median TEQ
emission factor reported by EPA by a factor of 3 to 4 (U.S. EPA, 1997a). For purposes of this
assessment, the EPA median and EPRI mean emission factors were averaged, for an emission
factor of 230 pg TEQDF-WHO98/L (200 pg I-TEQDF/L). Although the estimated emission factors
are assumed to be the current best estimates for utility/industrial oil burning, they are assigned a
low confidence rating.
Residual fuel oil sales totaled 77.3 billion L in 1987 and 46.6 billion L in 1995 (EIA,
1992, 1997a). Vessel bunkering was the largest consumer (48% of sales), followed by electric
utilities and the industrial sector. A high confidence rating is assigned to these production
estimates. Application of the TEQ emission factor of 230 pg TEQDF-WHO98/L (200 pg I-
TEQDF/L) to these residual fuel oil sales results in estimated TEQ emissions of 17.8 g TEQDF-
WHO98 (15.5 g I-TEQDF) for 1987 and 10.7 g TEQDF-WHO98 (9.3 g I-TEQDF) for 1995.
For reference year 2000, OAQPS developed national emission estimates for residual and
distillate oil consumption for the industrial sector. OAQPS used state-level 2000 activity data
(EIA, 2003a), which were allocated to counties by the 1999 county-to-state proportion of
employment for numerous SIC codes, as identified by the 2000 census. OAQPS estimated that
7.33 billion L of residual oil and 31.5 billion L of distillate oil were consumed in the industrial
sector in 2000. OAQPS combined these national activity levels with the emission factor of 230
pg TEQDF-WHO98/L (200 pg I-TEQDF/L) to estimate 2000 TEQ emissions of 1.69 g TEQDF-
WHO98 (1.47 g I-TEQDF) from residual oil consumption and 7.25 g TEQDF-WHO98
4-64
-------
Table 4-22. CDD/CDF emission factors (pg/L oil) for oil-fired utility/industrial
boilers
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
U.S. EPA (1997a)
median
emission factora'b
117
104
215
97
149
359
413
83
77
86
109
68
104
86
169
179
179
1,454
1,140
314.6
366.1
102
104
145
359
413
90
131
172
27
179
1,722
EPRI (1994) mea
Non detect set to
zero
0
24.7
63.3
65.8
79.7
477
2,055
0
64.1
49.3
76.5
35.4
0
23.8
164
0
0
2,765.5
413.1
83.1
93.6
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
n emission factor3'0
Nondetect set to Vz
detection limit
26.6
43.1
108
79.3
102
546
2,141
35.7
73.9
59.6
94.9
45.2
37.7
42.2
218
137
139
3,046
883.2
147.4
166.9
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
a Assumes a density for residual fuel oil of 0.87 kg/L.
bNumber of facilities not reported.
"Based on two cold-sided power plants equipped with electrostatic precipitators.
NR = Not reported
4-65
-------
Ratio Factor (congener emission factor/total CDD/CDF emission factor)
0 0.05 0.1 0.15 0.2
0.25
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Ratio (congener group emission factor/total CDD/CDF emission factor)
0.05 0.1 0.15 0.2
0.25
Figure 4-7. Congener and congener group profiles for air emissions from
industrial oil-fueled boilers.
Source: U.S. EPA(1995b; 1997a).
4-66
-------
(6.3 g I-TEQDF) from distillate oil combustion. Emission estimates for all reference years are
assigned a low confidence rating on the basis of the low rating for the emission factor.
4.3.3. Used Oil Combustion
The emission factors derived by EPA (U.S. EPA, 1997a) and EPRI (1994) were based on
combustion of virgin oil by utility boilers. Significantly greater emission factors have been
reported by Bremmer et al. (1994) for combustion of used oil by smaller combustion units in the
Netherlands. Flue gases from a garage stove consisting of an atomizer fueled by spent lubricating
oil from diesel engines (35 mg ClVkg) were reported to contain 0.1 ng I-TEQDF/Nm3 (2,000 pg I-
TEQDF/kg) oil burned. The flue gases from a hot water boiler consisting of a rotary cup burner
fueled with the organic phase of rinse water from oil tanks (340 mg ClVkg) contained 0.2 ng
I-TEQDF/Nm3 (4,800 pg I-TEQDF/kg) oil burned. The flue gases from a steam boiler consisting of
a rotary cup burner fueled by processed spent oil (240 mg ClVkg) contained 0.3 ng I-TEQDF/Nm3
(6,000 pg I-TEQDF/kg) oil burned. The emission factor for a ferry burning heavy fuel oil
containing 11 ng/kg organic chlorine was 3,200 to 6,500 pg I-TEQDF/kg oil burned. From these
data, Bremmer et al. derived an average emission factor for combustion of used oil of 4,000 pg I-
TEQDF/kg oil burned.
Bremmer et al. (1994) also reported measuring CDD/CDF emissions from a river barge
and a container ship fueled with gas oil (less than 2 ng/kg organic chlorine). The exhaust gases
contained from 0.002 to 0.2 ng I-TEQDF/Nm3. From these data, Bremmer et al. derived an average
emission factor for inland oil-fueled vessels of 1,000 pg I-TEQDF/kg oil burned.
The applicability of these emission factors to used oil combustors in the United States is
uncertain. Therefore, estimates of potential emissions from used oil combustion in the United
States are not being developed at this time.
4.4. COAL COMBUSTION
During 2000, coal consumption accounted for approximately 18.9% of the energy
consumed in the United States from all sources (EIA, 1999). Of the 980 million metric tons of
coal consumed in 2000, 891 million metric tons (90.9%) were consumed by electric utilities,
including independent power producers; 85.4 million metric tons (8.7%) were consumed by the
industrial sector, including 26.2 million metric tons consumed by coke plants; and 3.7 million
metric tons (0.4%) were consumed by residential and commercial sources (EIA, 1999).
In 1995, coal consumption (872 million metric tons) accounted for approximately 22% of
the energy consumed from all sources in the United States (U.S. DOC, 1997). Of this total,
4-67
-------
88.4% (771 million metric tons) were consumed by electric utilities; 11% (96 million metric
tons) were consumed by the industrial sector, including consumption of 30 million metric tons by
coke plants; and 0.6% (5.3 million metric tons) were consumed by residential and commercial
sources (EIA, 1997b).
In 1987 a total of 759 million metric tons of coal were consumed by electric utilities (651
million metric tons), coke plants (33.5 million metric tons), other industries (68.2 million metric
tons), and the residential and commercial sectors (6.3 million metric tons) (EIA, 1995).
All of the above estimates are assigned a high confidence rating because they are based on
detailed studies specific to the United States.
The following two sections discuss the results of relevant emission studies for the
utility/industrial and residential sectors and present annual TEQ emission estimates for reference
years 1987, 1995, and 2000.
4.4.1. Utilities and Industrial Boilers
Few studies have been performed to measure CDD/CDF concentrations in emissions from
coal-fired plants. Those studies did not have the congener specificity or DLs necessary to fully
characterize this potential source (U.S. EPA, 1987a; NATO, 1988; Wienecke et al., 1992). The
results of more recent testing of coal-fired utility and industrial boilers in the Netherlands
(Bremmer et al., 1994), the United Kingdom (Cains and Dyke, 1994; CRE, 1994), Germany
(Umweltbundesamt, 1996), and the United States (Riggs et al., 1995; EPRI, 1994) have achieved
lower DLs.
Bremmer et al. (1994) reported the results of emission measurements at two coal-fired
facilities in the Netherlands. The emission factor reported for a pulverized coal electric power
plant equipped with an ESP and a wet scrubber for sulfur removal was 0.35 ng I-TEQDF/kg coal
combusted (0.02 ng I-TEQDF/Nm3 at 11% oxygen). The emission factor reported for a grass-
drying chain grate stoker equipped with a cyclone APCD was 1.6 ng I-TEQDF/kg coal fired (0.16
ng I-TEQDF/Nm3 at 11% oxygen).
Cains and Dyke (1994) reported an emission factor of 102 to 109 ng I-TEQDF/kg of coal at
a small-scale facility in the United Kingdom that was equipped with an APCD consisting of only
a grit arrester. CRE (1994) reported results of testing at 13 commercial and industrial coal-fired
boilers in the United Kingdom, with TEQ emission factors ranging from 0.04 to 4.8 ng I-
TEQDF/kg coal combusted (mean value, 0.6 ng I-TEQDF/kg). CRE also reported testing results for
one coal-fired power plant: 0.06 ng I-TEQDF/kg of coal combusted. Umweltbundesamt (1996)
reported that the I-TEQDF content of stack gases from 16 coal-burning facilities in Germany
4-68
-------
ranged from 0.0001 to 0.04 ng I-TEQDF/m3; however, the data provided in that report did not
enable emission factors to be calculated.
In 1993, the U.S. Department of Energy (DOE) sponsored a project to assess emissions of
hazardous air pollutants at coal-fired power plants. As part of the project, CDD/CDF stack
emissions were measured at seven U.S. coal-fired power plants. The preliminary results of the
project, concentrations in stack emissions, were reported by Riggs et al. (1995) and are
summarized in Table 4-23. The levels reported for individual 2,3,7,8-substituted congeners were
typically very low (less than or equal to 0.033 ng/Nm3) or not detected. In general, CDF levels
were higher than CDD levels. OCDF and 2,3,7,8-TCDF were the most frequently detected
congeners (at four of the seven plants). Table 4-24 presents characteristics of the fuel used and
the APCD employed at each plant. Riggs et al. could not attribute variations in emissions
between plants to any specific fuel or operational characteristic.
As mentioned in Section 4.3.2, EPRI sponsored the FCEM project to gather information
of consistent quality on power plant emissions. Testing for CDD/CDF emissions was performed
on four coal-fired power plants equipped with cold-sided ESPs. Two plants burned bituminous
coal and two burned sub-bituminous coal. The results of the testing were integrated into the final
results of the DOE project discussed above (Riggs et al., 1995) and published in 1994 (EPRI,
1994). The average congener and congener group emission factors derived from this 11-facility
data set, as reported in EPRI (1994), are presented in Table 4-25. Congener and congener group
profiles for the data set are presented in Figure 4-8. Assuming nondetect values were zero, the
average emission factor was 0.078 ng TEQDF-WHO98/kg (0.079 ng I-TEQDF/kg) coal combusted.
A medium confidence rating is assigned to the emission factors derived from the DOE and EPRI
studies because they were based on recent testing at U.S. power plants.
Because the EPRI and DOE data characterized emissions from units with only cold-sided
ESPs, there has been uncertainty regarding the applicability of the emission factors derived from
these data to units with hot-sided ESPs. In July 1999, EPA conducted testing of stack emissions
at a coal-fired utility equipped with a hot-sided ESP. The preliminary results of this testing
indicated that the TEQ emission factor for hot-sided ESPs is of the same order of magnitude as
the average TEQ emission factors derived above.
4-69
-------
Table 4-23. CDD/CDF concentrations (pg/Nm3) in stack emissions from U.S. coal-fired power plants
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Plant 1
ND(3.5)
ND (0.56)
ND (0.56)
ND (0.44)
ND (0.56)
ND(1.7)
ND (1)2.4
3.3
1.1
ND (0.44)
ND(2)2
ND (O.g&
0
14.4
1.8
ND (1)1.3
3.4
ND (12)
ND (5.314
7.6
4.3
5.6
29.4
Plant 2
ND(3.5)
ND (4.8)
ND(5.75)
4.9
29
32
8.1
NjSi|®)
16
ND(5J!
ND(4.®
ND (6.313
70.9
97.1
12
4.4
18
45
32
29
33
39
34
33
279.4
Plant 3
1
ND(1.8)
ND (3.6)
ND(1.8)
ND(1.8)
flB f H?
7.8
7.2
6.6
8.4
2.9
ND (1.83
6
ND (3.62.4
1
44.3
12
6
2.7
ND (2.4)
ND (14^g
61
29
9
2.4
200.1
Plant 4
NTJJTO)
ND(10)
ND (10)
ND (10)
ND(10)
ATT-» /—>n\
ND(21
ND (10)
ND(10)
ND(10)
ND(10)
ND(10)
ND(10)
ND(10)
ND(10)
>JD r^n^
0
0
XTD
ND(10)
ND (10)
ND (10)
ND(20)
NB(TO)
ND(10)
ND(10)
NT) I""1 0s!
0
Plant 5
ND(3.3)
Nj^l^.l)
mm
ND(3.3)
ND (3.2)
NJ5D^f$.4)
ND (5.8)
Nj®i^.®.4)
W(15.4)
ND(131)
0
0
6.7
ND (26.4)
ND(131)
ND(3.3)
NJ5t($.6) 4)
ND (29.5)
ND(131)
6.7
Plant 6
ND (2.6)
ND(3 2)
V • /
ND (2.7)
ND (4.2)
ND (4.3)1.3
20
m> (5.7)
ND (4.8)
ND(5.1)
ND(4)
ND (6.9)
ND (2.5)
NiygO^
>JD c\ o^
24.3
13
ND (2.6)
ND (3.2)
ND(14J0
88
14
ND(5A
ND (20)
>rn t^^)^
122
Plant 7
ND(1.7)
ND(1.8)
ND(2)
ND(1.4)
ND (1.2^.4
21.6
0.7
ND(l.l)
ND (1.4)
ND(1.8)
ND(1.3)
ND(1.5)
ND(2)
ND (2.2)
ND (2J}.4
24
12.1
ND (55)
NT) n?^
1NJ-/ I JZ, }
ND(24)
ND(8J)1.6
ND (37)3
ND (27^ 9
11.4
38.9
-1^
o
ND = Not detected; value in parenthesis is the detection limit
NR = Not reported; suspected contamination problem
Source: Riggs et al. (1995).
-------
Table 4-24. Characteristics of U.S. coal-fired power plants tested by the U.S.
Department of Energy
Plant no.
1
2
3
4
5
6
7
Coal type
Bituminous
Bituminous
Sub-bituminous
Sub-bituminous
Bituminous
Lignite
Bituminous
Coal chlorine
content (mg/kg)
800
1,400
300
390
1,400
400
1,000
Temperature (°C)a
ESP
160
130
-
-
130
170
150
FF
-
-
150
70
-
-
-
FGD
-
-
-
130
120
170
-
Stack
160
130
150
75
40
110
150
Temperature at pollution control device and stack.
ESP = Electrostatic precipitator
FF = Fabric filter
FGD = Flue gas desulfurization system
Source: Riggs et al. (1995).
Applying the TEQ emission factor of 0.078 ng TEQDF-WHO98/kg (0.079 ng I-TEQDF/kg)
coal combusted to the consumption totals of 651, 771, and 891 million metric tons of coal
consumed by U.S. utility sectors in 1987, 1995, and 2000, respectively, yields estimated annual
emissions by the utility sector of 50.89 g TEQDF-WHO98 (51.4 g I-TEQDF) in 1987, 60.1 g
TEQDF.WHO98 (60.9 g I-TEQDF) in 1995, and 69.5 g TEQDF-WHO98 (70.4 g I-TEQDF) in 2000.
These emission estimates are assigned a medium confidence rating because the emission factor
rating for this category was judged to be medium.
No testing results could be located for CDD/CDF content in air emissions from
commercial and industrial coal-fired combustion units in the United States. It is uncertain
whether the data collected in the European studies (Bremmer et al., 1994; CRE, 1994) accurately
represent U.S. sources, but the data suggest that emission factors for commercial/industrial sources
might be higher than those reported for U.S. coal-fired utilities. Therefore, no national emission
estimate has been derived for this category. However, preliminary estimates of potential national
TEQ emissions from this source category can be derived for 1987, 1995, and 2000 using the total
coal consumption for each of those reference years, excluding consumption by coke plants, and
the average emission factor, 0.6 ng I-TEQDF/kg coal combusted.
4-71
-------
Table 4-25. CDD/CDF emission factors (ng/kg coal) for coal-fired utility/industrial
power plantsa
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Non detect set to zero
0.005
0
0
0.004
0.004
0.216
0.513
0.109
0.007
0.074
0.098
0.014
0.013
0.043
0.354
0.087
0.158
0.078
0.078
0.051
0.014
0.03
0.063
0.513
0.154
0.18
0.104
0.064
0.158
1.331
Non detect set to 1A
detection limit
0.018
0.016
0.034
0.028
0.035
0.241
0.644
0.117
0.021
0.084
0.12
0.03
0.038
0.06
0.385
0.112
0.281
0.124
0.131
0.052
0.015
0.03
0.074
0.644
0.158
0.18
0.104
0.064
0.281
1.602
"Eleven-facility data set.
Source: EPRI (1994).
4-72
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
0.09
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Ratio (congener group emission factor/total CDD/CDF emission factor)
0.05 0.1 0.15 0.2 0.25 0.3
0.35
Figure 4-8. Congener and congener group profiles for air emissions from
industrial/utility coal-fueled combustors (nondetects set equal to zero).
Source: EPRI (1994).
4-73
-------
Applying the emission factor to the estimated combustion for 1987, 1995, and 2000 (68.2,
66, and 59.14 million metric tons, respectively) yields 40.9 g I-TEQDF/kg for 1987, 39.6 g I-
TEQDF/kg for 1995, and 35.4 g I-TEQDF/kg coal combusted for 2000. These estimates should be
regarded as preliminary indications of possible emissions from commercial/industrial coal-fired
boilers; further testing is needed to confirm the true magnitude of these emissions. These
emission estimates are assigned a Category D uncertainty rating because the emission factor has
been judged to be clearly nonrepresentative of commercial and industrial coal-fired combustion
units.
4.4.2. Residential Coal Combustion
In the residential sector, coal is usually combusted in underfed or hand-stoked furnaces.
Other coal-fired heating units include hand-fed room heaters, metal stoves, and metal and
masonry fireplaces. Stoker-fed units are the most common design for warm-air furnaces and for
boilers used for steam or hot water production. Most coal combusted in these units is either
bituminous or anthracite. These units operate at relatively low temperatures and do not efficiently
combust the coal. Coal generally contains small quantities of chlorine; therefore, the potential for
CDD/CDF formation exists. Typically, coal-fired residential furnaces are not equipped with PM
or gaseous pollutant control devices that may limit emissions of any CDDs/CDFs formed (U.S.
EPA, 1997b). No testing results for CDD/CDF content in air emissions from
residential/commercial coal-fired combustion units in the United States could be located;
however, several relevant studies have been performed in European countries.
Thub et al. (1995) measured flue gas concentrations of CDDs/CDFs from a household
heating system in Germany that was fired with either salt lignite coal (total chlorine content of
2,000 ppm) or normal lignite coal (total chlorine content of 300 ppm). CDDs/CDFs were detected
in the flue gases generated by combustion of both fuel types (see Table 4-26). The congener
profiles and patterns were similar for both fuel types, with OCDD the dominant congener and
TCDF the dominant congener group. However, the emissions were higher by a factor of 8 for the
"salt" coal (0.109 ng I-TEQDF/m3 [2.74 ng I-TEQDF/kg]) than for the "normal" coal (0.015 ng I-
TEQDF/m3 [0.34 ng I-TEQDF/kg]).
Using the results of testing performed by the Coal Research Establishment in the United
Kingdom, Eduljee and Dyke (1996) estimated emission factors for residential coal combustion
units of 2.1 ng I-TEQDF/kg for anthracite coal and 5.7 to 9.3 ng I-TEQDF/kg (midpoint, 7.5 ng
I-TEQDF/kg) for bituminous coal.
4-74
-------
Table 4-26. CDD/CDF emission factors (ng/kg coal) for residential coal
combustors
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDDc
Total 2,3,7,8-CDFc
Total I-TEQDFC
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
"Salt"
lignite"
0.58
0.73
0.63
0.6
0.4
3.24
16.19
2.49
2.24
2.09
0.38
1.86
0.07
1.01
2.59
0.25
0.63
22.37
13.6
2.74
14.23
14.15
11.14
7.06
16.19
80.34
29.21
12.72
3.87
0.63
189.5
"Normal"
lignite"
0.06
0.08
0.06
0.09
0.06
0.59
2.42
0.5
0.43
0.31
0.13
0.36
0.02
0.12
0.95
0.06
0.3
3.38
3.2
0.34
9
2.22
1.81
0.82
2.42
20.33
8.98
3.78
1.27
0.3
50.9
Anthracite15
1.6
NR
NR
NR
NR
NR
77
42
NR
NR
NR
NR
NR
NR
NR
NR
4.2
NR
NR
60
61.6
31
60
57
77
412
340
130
32
4.2
1,205
Bituminous15
2.4
NR
NR
NR
NR
NR
120
63
NR
NR
NR
NR
NR
NR
NR
NR
6.3
NR
NR
98.5
92.4
46
90
86
120
613
550
190
47
6.3
1,841
aSource: Thub et al. (1995); listed results represent means of three flue gas samples.
bSource: U.S. EPA (1997a); based on average paniculate CDD/CDF concentrations from chimney soot samples
collected from seven coal ovens and paniculate emission factors for anthracite and bituminous coal combustion.
°Values as reported in sources.
NR = Not reported
4-75
-------
CDD/CDF emission factors were estimated (U.S. EPA, 1997a) for coal-fired residential
furnaces using average paniculate CDD/CDF concentrations from chimney soot samples collected
from seven coal ovens and PM emission factors obtained from AP-42 that are specific to
anthracite and bituminous coal combustion (U.S. EPA, 1995a). The TEQ emission factors
estimated (60 and 98.5 ng I-TEQDF/kg anthracite and bituminous coal, respectively) were derived
using the calculated emission factors for 2,3,7,8-TCDD, 2,3,7,8-TCDF, and the 10 congener
groups (U.S. EPA, 1997a). EPA stated that the estimated factors should be considered
representative of maximum emission factors because soot may not be representative of the PM
actually emitted to the atmosphere. These emission factors are presented in Table 4-26; congener
group profiles are presented in Figure 4-9.
Although the congener group profiles of the measurements by Thub et al. (1995) and the
estimates by EPA (U.S. EPA, 1997a) are similar, the TEQ emission factors of the two studies
differ by factors of 175 to 289. The emission factors used by Eduljee and Dyke (1996) to estimate
national annual TEQ emissions from residential coal combustion in the United Kingdom fall in
between the other two sets of estimates but are still about one to two orders of magnitude greater
than the estimated emission factor for industrial and utility coal combustors (see Section 4.4.1).
For 1987 and 1995, preliminary estimates of potential national TEQs were derived using
the emission factors from Eduljee and Dyke (1996). U.S. EPA (1997a) reported that 72.5% of the
coal consumed by the residential sector in 1990 was bituminous and 27.5% was anthracite.
Assuming that these relative proportions reflect the actual usage in 1987 and 1995, then
application of the emission factors from Eduljee and Dyke (2.1 ng I-TEQDF/kg anthracite coal and
7.5 ng I-TEQDF bituminous coal) to the consumption values of 6.3 and 5.3 million metric tons in
1987 and 1995, respectively (U.S. DOC, 1997), results in estimated TEQ emissions of 37.9 and
32.0 g I-TEQDF in 1987 and 1995, respectively. These estimates should be regarded as
preliminary indications of possible emissions from this source category because the emission
factor is judged to be clearly nonrepresentative of the sources. Further testing is needed to
confirm the true magnitude of these emissions.
For 2000, OAQPS developed national emission estimates for coal combustion for
residential heating. The activity level for residential coal combustion was taken from state-level
2000 activity data (EIA, 2003a). Because EIA no longer disaggregates coal consumption into
anthracite versus bituminous/lignite, OAQPS estimated each state's coal consumption using the
state's 1999 proportion of anthracite versus bituminous/lignite to total coal consumption.
Emissions were allocated to the county level as a proportion of state population in states that
consume anthracite coal and bituminous and lignite coal for residential heating. OAQPS
estimated that in 2000, 67,400 metric tons of anthracite coal and 343,000 metric tons of
4-76
-------
Ratio (congener group emission factor/total CDD/CDF emission factor)
0.1 0.2 0.3
0.4
I Anthracite
I I Bituminous
Figure 4-9. Congener group profile for air emissions from residential
coal-fueled combustors.
Source: U.S. EPA(1997a).
bituminous and lignite coal were consumed for residential heating. Applying the TEQ emission
factors of 2.1 ng I-TEQDF/kg anthracite coal combusted and 7.5 ng I-TEQDF/kg bituminous coal
combusted (Eduljee and Dyke, 1996) to these production factors yields preliminary estimates of
annual emissions of 0.14 g I-TEQDF of anthracite coal and 2.6 g I-TEQDF of bituminous/sub-
bituminous coal in 2000. These emission estimates are assigned a Category D uncertainty rating
because the emission factor has been judged to be clearly nonrepresentative of residential coal
combustion.
4.4.3. Solid Wastes from Coal Combustion
A limited amount of CDD/CDF concentration data have been developed for utility
industry solid wastes (U.S. EPA, 1999c), and these data are for wastes that are comanaged (i.e.,
combinations of fly ash, bottom ash, boiler slag, and flue gas desulfurization wastes). A total of
15 samples were taken from 11 disposal sites. The average concentration for each of the CDD
4-77
-------
and CDF congeners is presented in the second column of Table 4-27. It should be noted that most
of the concentration values shown in Table 4-27 represent DLs. Consequently, the values
overestimate actual concentrations.
U.S. EPA (1999d) indicates that approximately 63 million tons (assumed to be short tons,
i.e., 2,000 pounds) of large-volume utility coal combustion solid wastes were produced in 1995.
Of this amount, about 67% was landfilled and the balance was disposed of in surface
impoundments. The concentration data presented in Table 4-27 are for only the 53 million tons
that were comanaged (about 84% of the total wastes). For purposes of this analysis it is assumed
that the CDD/CDF concentrations in the comanaged wastes are the same as for the entire waste
quantity. Combining the concentration data with the 63 million tons of total waste yields the total
quantities of each congener disposed of in 1995. These data are presented in the fourth column of
Table 4-27. As indicated in Section 4.4 of this document, total consumption of coal for electric
utility boilers in 1987 was 98.4% of 1995 consumption. Consequently, the quantities of
CDDs/CDFs disposed of in 1987 is assumed to be 98.4% of the 1995 values. These values are
presented in column 3 of Table 4-27. The 1995 congener quantities are converted into I-TEQDF
and TEQDF-WHO98 values in columns 5 and 6. Because disposal of these wastes does not
constitute an environmental release, the values are not included in the inventory.
4-78
-------
Table 4-27. CDD/CDF concentrations and amounts disposed of in solid wastes
from coal-fired utilities
Congener
2,3,7,8-TCDDd
l,2,3,7,8-PeCDDd
l,2,3,4,7,8-HxCDDd
l,2,3,6,7,8-HxCDDd
l,2,3,7,8,9-HxCDDe
l,2,3,4,6,7,8-HpCDDf
OCDD8
2,3,7,8-TCDFh
l,2,3,7,8-PeCDFd
2,3,4,7,8-PeCDFd
l,2,3,4,7,8-HxCDFe
l,2,3,6,7,8-HxCDFd
2,3,4,6,7,8-HxCDFd
l,2,3,7,8,9-HxCDFd
l,2,3,4,6,7,8-HpCDFe
l,2,3,4,7,8,9-HpCDFd
OCDF1
TOTAL
Mean
concentration"
(ng/kg)
0.17
0.25
0.35
0.28
0.3
0.59
10.54
0.19
0.17
0.17
0.25
0.18
0.28
0.24
0.29
0.35
0.59
Disposed of
in solid
waste in
1987 (g/yr)b
10
14
20
16
17
33
593
11
10
10
14
10
16
14
16
20
33
Disposed of
in solid waste
in 1995 (g/yr)c
10
14
20
16
17
34
603
11
10
10
14
10
16
14
17
20
34
I-TEQDF
/yr 1995
(g)
9.72
7.15
2
1.6
1.72
0.34
0.6
1.09
0.49
4.86
1.43
1.03
1.6
1.37
0.17
0.2
0.03
35.41
TEQDF-
WHO98/yr
1995 (g)
9.72
14.3
2
1.6
1.72
0.34
0.6
1.09
0.49
4.86
1.43
1.03
1.6
1.37
0.17
0.2
<0.01
41.98
"Source: U.S. EPA (1999c, Table 2-9).
bAssumes that solid waste quantity for 1987 was 98.4% of 1995 quantity, based on total utility coal use in those years
(see Section 4.4).
"Based on EPRI estimate of 63 million tons/yr of large-volume utility coal combustion solid wastes. See Section
3.3 of U.S. EPA (1999d). Assumes all waste characteristics were the same as for comanaged wastes.
dAll 17 analyses were nondetects.
eSixteen of the 17 analyses were nondetects.
Eleven of the 17 analyses were nondetects.
gFive of the 17 analyses were nondetects.
fourteen of the 17 analyses were nondetects.
'Fifteen of the 17 analyses were nondetects.
4-79
-------
5. COMBUSTION SOURCES OF CDDs/CDFs: OTHER
HIGH-TEMPERATURE SOURCES
5.1. CEMENT KILNS
This section addresses CDD/CDF emissions from Portland cement kilns. These facilities
use high temperatures to convert mineral feedstocks into Portland cement and other types of
construction materials. For purposes of this analysis, cement kilns are divided into two
categories: those that burn hazardous waste as a supplementary fuel and those that do not. For
the 1987 and 1995 estimates, the hazardous waste-burning cement kiln category was further
divided into kilns with inlet air pollution control device (APCD) temperatures above 232°C and
those with APCD temperatures below 232°C. Cement kiln technology, the derivation of TEQ
emission factors for cement kilns that burn hazardous waste as supplemental fuel and those that
do not, and the derivation of annual TEQ air emissions (g/yr) for reference years 1987, 1995, and
2000 are discussed below.
5.1.1. Process Description of Portland Cement Kilns
In the United States, the primary cement product is Portland cement. Portland cement is
a fine, gray powder consisting of a mixture of four basic materials: lime, silica, alumina, and iron
compounds. Cement production involves heating (pyroprocessing) the raw materials to a very
high temperature in a rotary (rotating) kiln to induce chemical reactions that produce a fused
material called clinker. The cement clinker is then ground into a fine powder and mixed with
gypsum to form the Portland cement.
The cement kiln is a large, steel, rotating cylindrical furnace lined with refractory
material. The kiln is aligned on a slight angle, usually a slope of 3 to 6 degrees, which allows the
materials to pass through the kiln by gravity. The kiln rotates at about 50 to 70 revs/hr, and the
rotation induces mixing and the downward movement of mixed materials. The upper end of the
kiln, known as the cold end, is generally where the raw materials, or meal, are fed into the kiln.
Midpoint injection is practiced at some facilities. The lower end of the kiln, known as the hot
end, is where the combustion of primary fuels (usually coal and petroleum coke) occurs,
producing a high temperature. The cement kiln operates using counter currents: hot combustion
gases are convected up through the kiln while the raw materials pass down toward the lower end.
As the meal moves through the cement kiln and is heated by hot combustion gases, water is
vaporized and pyroprocessing of materials occurs.
The cement kiln consists of three thermal zones to produce cement clinker. Zone 1 is at
the upper end of the kiln where the raw meal is added. Temperatures in this zone typically range
from ambient up to 600°C. In this area of the kiln, moisture is evaporated from the raw meal.
5-1
-------
Zone 2 is known as the calcining zone. Calcining occurs when the hot gases from the
combustion of primary fuels dissociate calcium dioxide from the limestone and form calcium
oxide. In this zone, temperatures range from 600 to 900°C. Zone 3, the burning or sintering
zone, is the lowest and hottest region of the kiln. Here, temperatures in excess of 1,500°C induce
the calcium oxide to react with silicates, iron, and aluminum in the raw materials to form the
cement clinker. The formation of clinker actually occurs close to the combustion of primary fuel.
The chemical reactions that occur in zone 3 are referred to as pyroprocessing.
The cement clinker, which leaves the kiln at the hot end, is a gray, glass-hard material
consisting of dicalcium silicate, tricalcium silicate, calcium aluminate, and tetracalcium
aluminoferrite. At this point, the temperature of the clinker is about 1,100°C. The hot clinker is
then dumped onto a moving grate, where it cools as it passes under a series of cool-air blowers.
After it is cooled to ambient temperature, the clinker is ground into a fine powder and mixed with
gypsum to produce the Portland cement product.
Cement kilns can be either wet process or dry process. In the wet process, the raw
materials are ground and mixed with water to form a slurry, which is fed into the kiln through a
pump. This is an older process. A greater amount of heat energy is needed in the wet process
kiln than in other types of kilns. These kilns consume about 5 to 7 trillion Btu per ton of clinker
product to evaporate the additional water. In the dry process, a preheater is used to dry the raw
meal before it enters the kiln. A typical preheater consists of a vertical tower containing a series
of cyclone-type vessels. Raw meal is added at the top of the tower and hot exhaust gases from
the kiln operation preheat the meal, thus lowering the fuel consumption of the kiln. Dry kilns are
now the most popular type of cement kiln.
Portland cement clinker production in the United States is estimated to have been 52
billion kg in 1987 (U.S. DOC, 1996), 67.6 billion kg in 1995 (U.S. DOC, 1996), and 75.2 billion
kg in 2000 (PC A, 2001). The 2000 estimate is based on the assumption that of the annual
maximum clinker capacity reported for that reference year (PC A, 2001), only 90% was actually
produced (e-mail dated January 31, 2003, from Garth Hawkins, Portland Cement Association, to
Karie Riley, Versar, Inc.).
5.1.2. Cement Kilns That Burn Hazardous Waste
The high temperatures achieved in cement kilns make the kilns an attractive technology
for combusting hazardous waste as supplemental fuel. Sustaining the relatively high combustion
temperatures that are needed to form cement clinker (1,100 to 1,500°C) requires the burning of a
fuel with a high energy output. Therefore, coal or petroleum coke is typically used as the primary
fuel source. Because much of the cost of operating the cement kiln at high temperatures is
associated with the consumption of fossil fuels, some cement kiln operators burn hazardous
5-2
-------
liquid and solid waste as supplemental fuel. In 2000, approximately 60% of all facilities burned
hazardous waste as the primary fuel to offset the amount of coal/coke purchased and burned by
the kiln (PCA, 2001). Organic hazardous waste may have a heating value similar to that of coal
(9,000 to 12,000 Btu/lb for coal). The kiln operator may charge the waste generator a disposal
fee to combust the hazardous waste; this fee also offsets the cost of kiln operation. The high-
energy and ignitable wastes include diverse substances, such as waste oils, spent organic
solvents, sludges from the paint and coatings industry, waste paints and coatings from auto and
truck assembly plants, and sludges from the petroleum refining industry (Greer et al., 1992).
The conditions in the cement kiln mimic conditions of hazardous waste incineration. For
example, the gas residence time in the burning zone is typically 3 sec at temperatures in excess of
1,500°C (Greer et al., 1992). The method of introducing liquid and solid hazardous waste into
the kiln is a key factor in the complete consumption of the waste during the combustion of the
primary fuel. Liquid hazardous waste is either injected separately or blended with the primary
fuel (coal). Solid waste is mixed and burned along with the primary fuel.
Trial burns have consistently shown that destruction and removal efficiencies of 99.99 to
99.9999% can be achieved for very stable organic wastes using cement kilns (Greer et al., 1992).
Hazardous waste was combusted at 34 of the 212 kilns operating in 1995 (Federal Register,
1996a) and at 33 of the 201 kilns operating in 2000 (e-mail correspondence dated February 24,
2003, between M. Benoit, Cement Kiln Recycling Coalition, and K. Riley, Versar, Inc.; PCA,
2001). Other types of supplemental fuel used by these facilities include natural gas, fuel oil,
automobile tires, used motor oil, sawdust, and scrap wood chips.
5.1.3. Air Pollution Control Devices
The pyroprocessing of raw meal in a cement kiln also produces fine particulates, referred
to as cement kiln dust (CKD). CKD is collected and controlled with fabric filters (FFs),
electrostatic precipitators (ESPs), or both. Acid gases such as sulfur dioxide (SO2) can be formed
during pyroprocessing of the sulfur-laden minerals and fuels, but the minerals have high
alkalinity, which partially neutralizes SO2 gases. Most APCDs used at cement kilns in 1987 and
1995 were considered to be hot-sided control devices. A hot-sided control device is one that
operates at kiln exhaust gas temperatures above 232°C (some EPA rules use different definitions
for hot-sided control devices for different industries). Most APCDs currently used at cement
kilns are cold-sided devices (i.e., they operate at kiln exhaust gas temperatures below 232°C.
Reducing the temperature at the inlet of the APCD is one factor that has been shown to
have a significant impact on limiting dioxin formation and emissions at cement kilns (U.S. EPA,
1997c). Emissions testing at a Portland cement kiln showed that CDDs/CDFs were almost
entirely absent at the inlet to a hot-sided ESP, but measurements taken at the exit showed
5-3
-------
conclusively that dioxins were formed within the hot-sided ESP (U.S. EPA, 1997c). Reducing
the kiln exhaust gas temperature in the APCD to below 232°C has been shown to substantially
limit CDD/CDF formation. Lower temperatures are believed to prevent the post-combustion
catalytic formation of CDDs/CDFs. Consequently, a number of cement kilns have added exhaust
gas-quenching units upstream of the APCD to reduce the inlet APCD temperature, thereby
reducing CDD/CDF stack concentrations. A quenching unit usually consists of a water spray
system within the flue duct.
5.1.4. CDD/CDF Emissions Data
The general strategy used to derive emission factors for this report was to divide each
source category on the basis of design and operation. However, because cement kilns are
relatively uniform in terms of kiln design, raw feed material, operating temperatures, and
APCDs, they have been categorized, as noted above, only on the basis of whether or not
hazardous waste is burned as a supplementary fuel.
CDD/CDF emissions data from tests conducted between 1989 and 1996 were obtained
for 16 cement kilns burning hazardous waste and 15 cement kilns burning nonhazardous waste
(U.S. EPA, 1996b). More recent CDD/CDF emissions data were also obtained from tests
conducted in 2000 at 3 cement kilns burning hazardous waste (U.S. EPA, 2002a) and from tests
conducted in June and July of 1999 at one facility burning nonhazardous waste (Bell, 1999). The
majority of stack emissions data from cement kilns burning hazardous waste were derived during
trial burns and may overestimate the CDD/CDF emissions that most kilns achieve during normal
operations. Stack emissions data from kilns burning nonhazardous waste were derived from
testing during normal operations.
5.1.4.1. Emissions Data for 1989 Through 1996 (U.S. EPA, 1996b)
The average TEQ emission factors for this period, based on the data reported by EPA in
1996 (U.S. EPA, 1996b), were 0.000941 to 232 ng TEQDF-WHO98/kg (average of 22.48 ng
TEQDF-WHO98/kg [20.91 ng I-TEQDF/kg]) clinker produced for cement kilns burning hazardous
waste and 0.000012 to 2.76 ng TEQDF-WHO98/kg (average of 0.29 ng TEQDF-WHO98/kg [0.27 ng
I-TEQDF/kg]) clinker produced for cement kilns burning nonhazardous waste.
These data show that the average emission factor for kilns burning hazardous waste was
about 90 times greater than that for kilns burning nonhazardous waste. However, it should be
noted that the average emission factor for kilns burning hazardous waste was derived from "near
worst case" testing of hazardous waste-burning kilns. As discussed in Section 5.1.8, a
comparison of CDD/CDF concentrations in CKD samples shows a similar relationship (i.e., the
CDD/CDF TEQ concentration of the CKD from kilns burning hazardous waste was about 100
5-4
-------
times higher than that of the dust from kilns burning nonhazardous waste). Although the average
emission factors for the two groups of kilns differ substantially, the emission factors for
individual kilns in the two groups overlap. Therefore, other aspects of the design and operation
of the kilns—in particular, the temperature of the APCD equipment (as discussed in Section
5.1.3)—were likely affecting CDD/CDF emissions.
Previous attempts to understand these differences using parametric testing of cement
kilns yielded mixed results. EPA (U.S. EPA, 1997c) conducted a limited comparison of
CDD/CDF TEQ stack gas concentrations (ng TEQ/dscm) between cement kilns burning
hazardous wastes and those not burning hazardous wastes. Those comparisons were made at 14
cement kilns. With the exception of the fuel being burned, operating conditions (e.g., APCD
temperature) were the same or similar for each set of comparisons. Baseline conditions used coal
as the only primary fuel. The results of these comparisons found:
• seven kilns in which the baseline (i.e., no combustion of hazardous waste) CDD/CDF
TEQ stack gas concentrations were about the same as those for the burning of
hazardous wastes,
• two kilns in which the baseline CDD/CDF I-TEQDF stack gas concentrations were
about double those for the burning of hazardous wastes, and
• five kilns in which the hazardous waste CDD/CDF I-TEQDF stack gas concentrations
were substantially greater (3- to 29-fold greater) than those for the baseline operating
conditions.
Subsequently, EPA's Office of Research and Development (ORD) conducted analyses of
the available emissions data to evaluate, on a congener-specific basis, whether there were
significant differences in emission factors between (a) kilns burning hazardous waste and those
burning nonhazardous waste, (b) kilns with APCD inlet temperatures greater than 232°C and
those with temperatures less than 232°C, (c) hazardous waste-burning and nonhazardous waste-
burning facilities with APCD inlet temperatures greater than 232°C, (d) hazardous waste-burning
and nonhazardous waste-burning facilities with APCD inlet temperatures less than 232°C, (e)
hazardous waste-burning facilities with APCD inlet temperatures less than and greater than
232°C, and (f) nonhazardous waste-burning facilities with APCD inlet temperatures less than and
greater than 232°C. The results of all analyses showed significant differences in the sample
mean values (p<0.05).
Given the strong empirical evidence that real differences existed, ORD decided to address
the kilns burning hazardous waste separately from those burning nonhazardous waste to develop
a CDD/CDF emissions inventory and to subdivide the hazardous waste-burning category into
5-5
-------
subcategories by APCD inlet temperature (i.e., less than 232°C and greater than 232°C). APCD
inlet temperature data were available for 88 test runs at 14 cement kilns. The number of test runs
conducted at individual kilns ranged from 1 to 26. Each test run was treated as an individual
facility and each was classified according to APCD inlet temperature and whether or not
hazardous waste was burned. The emission factor for each cement kiln test run was calculated
using eq 5-1.
EFCK = CxFv (5-1)
where:
EFCK = cement kiln emission factor (burning or not burning hazardous waste)
(ng TEQ/kg of clinker produced)
C = TEQ or CDD/CDF concentration in kiln exhaust gases (ng TEQ/dscm)
(20°C, 1 atm; adjusted to 7% O2)
Fv = volumetric kiln exhaust gas flow rate (dscm/hr) (20°C, 1 atm; adjusted to 7% O2)
Icl = average cement kiln clinker production rate (kg/hr)
After developing the emission factor for each cement kiln test run, the overall average congener-
specific emission factor was derived for all test runs in each subcategory using eq 5-2.
CK3
EF = (5-2)
L^± avgCK » r
where:
EFavgCK = average emission factor of tested cement kilns burning hazardous
waste as supplemental fuel and with APCD inlet temperatures
either greater than or less than 232°C (ng TEQ/kg clinker)
N = number of cement kiln test runs
TEQ emission values for hazardous waste-burning cement kilns with APCD inlet temperatures
greater than 232°C and less than 232°C were 30.7 and 1.11 ng TEQDF-WHO98/kg clinker
produced, respectively.
5-6
-------
5.1.4.2. Emissions Data for 1999 and 2000 (Bell, 1999; U.S. EPA, 2002a)
The results of a test conducted in 1999 for a cement kiln burning nonhazardous waste
(Bell, 1999) showed average TEQDF-WHO98 and I-TEQ emission factors of 0.14 ng/kg clinker
produced. This value is within the range of emission factor values developed using the 1989
through 1996 data (U.S. EPA, 1996b).
In 1999, EPA's Office of Solid Waste and Emergency Response (OSWER), under the
National Emission Standards for Hazardous Air Pollutants (NESHAPs), promulgated Final
Standards for Hazardous Air Pollutants for Hazardous Waste Combustors (Federal Register,
1999a). During the development of this rulemaking, a database of national emission estimates
for hazardous waste combustors (HWCs), including hazardous waste-burning cement kilns, was
established. In this database EPA identified 30 hazardous waste cement kilns operating in 1997.
Twenty-four of the 30 kilns had provided emissions sampling data to EPA, including dioxin
concentrations. Using the most recent sampling data for each kiln, along with imputed release
estimates for the missing six kilns and the assumption that cement kilns were operating 24
hr/day, 365 days/yr, OSWER developed a national emissions estimate of 13 g TEQ/yr for 1997.
A number of parties sought judicial review of the 1999 rule, and in 2001 the standards
were vacated. Interim standards were issued in 2002 until EPA could issue revised final
standards in 2004. The existing national emissions database was updated to include new
sampling data and data for newly operational HWCs and to remove eight of the cement kilns that
were identified as no longer burning hazardous waste. From this updated database (U.S. EPA,
2002a), EPA was able to develop new emission estimates for 2000.
5.1.4.3. Emission Factor Estimates for Cement Kilns Burning Hazardous Waste
For reference years 1987 and 1995, EPA estimated the TEQ emission factor by
subdividing the emissions data reported in 1996 (i.e., for 1989 through 1996) (U.S. EPA, 1996b)
by inlet APCD temperature above and below 232°C. For cement kilns operating at temperatures
above 232°C, the TEQ emission factor was 30.7 ng TEQDF-WHO98/kg clinker produced, and for
cement kilns operating at temperatures below 232°C, the TEQ emission factor was 1.11 ng
TEQDF-WHO98/kg clinker produced. These emission factors are presented in Table 5-1 and the
average congener profile is presented in Figure 5-1.
Because a vast majority of the facilities had reduced their APCD inlet temperature to
below 232°C by 2000, and because only a few new test reports applicable to reference year 2000
were available, EPA removed the 232°C divider and combined the emission factor results (i.e.,
1989 through 1996 data) (U.S. EPA, 1996b) for facilities that were still operating in 2000 with
5-7
-------
Table 5-1. CDD/CDF emission factors (ng/kg clinker produced) for cement
kilns burning hazardous waste for reference years 1987 and 1995
Congener/
congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean emission factor
(nondetect values set equal to zero)
APCD inlet temperature
>232°C
3.38
4.28
4.85
6.93
9.55
27.05
18.61
36.26
13.36
23.48
22.24
8.46
0.96
13.33
7.73
2.16
2.51
28.58
30.7
406.76
608.65
845.99
192.99
18.61
295.72
127.99
50.75
8.36
2.51
2,558.33
APCD inlet temperature
<232°C
0.02
0.13
0.29
0.42
0.4
3.16
1.08
3.24
0.23
0.65
0.55
0.27
0.06
0.52
0.34
0.16
0.37
1.04
1.10
1.78
0.89
0.69
0.42
1.08
11.52
3.83
1.88
0.47
0.37
22.93
APCD = Air pollution control device
Source: U.S. EPA (1996b).
the newer data reported (U.S. EPA, 2002a). Therefore, emission tests from five facilities (U.S.
EPA, 1996b) were not used to estimate the 2000 emission factor because the facilities no longer
burned hazardous waste in 2000. Using this approach, a conservative TEQ emission estimate of
-------
Ratio (congener emission factor/total CDD/CDF emission factor)
Oa 0.005 0.01 0.015 0.02
Figure 5-1. Congener profile for air emissions from cement kilns burning
hazardous waste for reference years 1987 and 1995 (nondetect set equal to
zero).
5.95 ng TEQDF-WHO98/kg (5.49 ng I-TEQ/kg) clinker produced was developed for reference year
2000. The congener-specific emission factors are presented in Table 5-2 and the average
congener and congener group profiles are presented in Figure 5-2.
5.1.4.4. Emission Factor Estimates for Cement Kilns Burning Nonhazardous Waste
Because only one test report applicable to reference year 2000 was located for a cement
kiln burning nonhazardous waste (Bell, 1999), and the results from the tests were similar to the
results reported by EPA in 1996 (U.S. EPA, 1996b), EPA combined the results from the two data
sets to obtain a TEQ emission factor estimate of 0.27 ng TEQDF-WHO98/kg (0.26 ng I-TEQ/kg)
clinker produced for reference years 1987, 1995, and 2000. The congener-specific emission
factors are presented in Table 5-3 and the average congener and congener group profiles are
presented in Figure 5-3.
5-9
-------
Table 5-2. CDD/CDF emission factors (ng/hr) for cement kilns burning
hazardous waste for reference year 2000
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDF (g/yr)
Total TEO™-WHOOS fe/yr)
Mean emission factor
9,533
32,510
48,483
78,043
66,369
446,105
142,108
63,866
37,470
77,981
59,340
48,179
33,967
47,458
47,762
15,098
12,418
16.6
18.8
Sources: U.S. EPA (1996b, 2002a).
5.1.4.5. Confidence Ratings of Emission Factor Estimates
The TEQ emission factors are given a low confidence rating for all subcategories and all
years. The emission factor for nonhazardous waste-burning kilns was given a low rating because
test data were available for only 16 facilities. The tested facilities may not be representative of
routine CDD/CDF emissions from all kilns burning nonhazardous waste. Although a higher
percentage of the kilns burning hazardous waste (with reported APCD temperature data) had
been tested, greater uncertainty exists about whether the emissions are representative of normal
operations because the tests used trial burn procedures and because a greater majority of the
operating facilities had reduced their APCD temperatures to below 232°C. Accordingly, a low
confidence rating is also assigned to the estimated emission factors for kilns burning hazardous
waste.
5.1.5. Activity Level Information
In 1987, approximately 52 billion kg of cement clinker were produced in the United
States (U.S. DOC, 1996). In 1995, approximately 67.6 billion kg of clinker were produced in the
5-10
-------
Ratio (congener ends s ion factor / 2378-CDD/CDF emis s ion factor)
0.000
4D2378
0.020
0.030
Ratio (congener group em ission factor/total C D D/C D F em ission factor)
0 0.1 0.2 0.3 0.4 0.5
Figure 5-2. Congener profile for air emissions from cement kilns burning
hazardous waste for reference year 2000.
5-11
-------
Table 5-3. CDD/CDF emission factors (ng/kg clinker produced) for cement kilns not
burning hazardous waste for reference years 1987,1995, and 2000a
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean emission factor
Nondetect set to zero
0.01
0.03
0.03
0.04
0.04
0.39
0.64
0.73
0.1
0.22
0.17
0.05
0.01
0.08
0.13
0
0.22
0.26
0.27
1.89
1.92
5.51
0.78
0.64
7.72
2.06
0.56
0.23
0.22
21.53
Nondetect set to Vz detection
limit
0.02
0.04
0.04
0.05
0.06
0.39
0.64
0.73
0.11
0.23
0.18
0.06
0.02
0.08
0.14
0.02
0.24
0.29
0.31
1.89
1.92
5.51
0.78
0.64
7.72
2.06
0.56
0.23
0.24
21.55
The same CDD/CDF emission factor was assumed for all three years.
Sources: U.S. EPA (1996b); Bell (1999).
United States (U.S. DOC, 1996), and of this amount, 61.3 billion kg were produced by cement
kilns burning nonhazardous waste (Memorandum dated August 23, 1995, from E. Heath,
Research Triangle Institute, to J. Wood, U.S. EPA); therefore, approximately 6.3 billion kg were
produced by cement kilns burning hazardous waste. Based on the fact that 9.3% of the clinker
5-12
-------
Ratio (congener emission factor/2378-CDD/CDF emission factor)
0.000 0.050 0.100 0.150 0.200 0.250
0.300
0.350
Ratio (congener group emission factor/total CDD/CDF emission factor)
0.000 0.050 0.100 0.150 0.200 0.250 0.300
4D Total
5D Total
6D Total
7D Total
8D 12346789
4F Total
5F Total
6F Total
7F Total
8F 12346789
Figure 5-3. Congener profile for air emissions from cement kilns
burning nonhazardous waste for reference years 1987,1995, and 2000.
5-13
-------
produced in 1995 was from cement kilns burning hazardous waste, it is assumed that
approximately 4.8 billion kg of the clinker produced in 1987 were from cement kilns burning
hazardous waste.
In 2000, cement kilns produced approximately 75.2 billion kg of clinker. This amount
is based on the assumption that cement kilns operated at 90% of the maximum annual clinker
capacity of 83.6 billion kg (e-mail dated January 31, 2003, from Garth Hawkins, Portland
Cement Association, to Karie Riley, Versar, Inc.). Based on the annual clinker capacities of
individual cement kilns, approximately 11.5 billion kg of clinker (15%) were produced by
cement kilns burning hazardous waste and approximately 63.7 billion kg of clinker (85%) were
produced by cement kilns not burning hazardous waste (PCA, 2001). The activity level estimates
for 1995 and 2000 are given a high confidence rating because they are based on comprehensive
survey data, but the rating for 1987 is medium because of uncertainty concerning the proportion
produced by hazardous waste-burning kilns (U.S. EPA, 1996b).
5.1.6. National CDD/CDF Emission Estimates
5.1.6.1. Estimates for Reference Years 1987 and 1995
National estimates of CDD/CDF air emissions (g TEQ/yr) from all Portland cement kilns
for reference years 1987 and 1995 were made by multiplying the average TEQ emission factors
by an estimate of the annual activity level (cement clinker produced) for each of the three
subcategories (hazardous waste-burning kilns with APCD inlet temperatures greater than 232°C
and less than 232°C and kilns burning nonhazardous waste). Of the 10 hazardous waste-burning
kilns with APCD temperature data, 8 facilities (80%) had APCD inlet temperatures greater than
232°C and 2 facilities (20%) had APCD inlet temperatures less than 232°C. The percentages of
hazardous waste-burning kilns with input temperatures less than and greater than 232°C were
assumed to represent the actual distribution of activity level in the industry. These percentages,
coupled with the TEQ emission factors presented in Table 5-1 and Table 5-3 (hazardous waste
cement kilns and nonhazardous waste cement kilns, respectively) and the activity levels
established in Section 5.1.5, were used to calculate the annual national TEQ emission estimates
shown in Table 5-4.
Overall, 131 g TEQDF-WHO98 (122 g I-TEQDF) were produced by cement kilns in 1987.
Of this amount, 116.7 g TEQDF-WHO98 (108.6 g I-TEQDF) were produced by hazardous waste-
burning cement kilns with inlet APCD temperatures greater than 232°C, 1.1 g TEQDF-WHO98
(1 g I-TEQDF) were produced by cement kilns burning hazardous waste with inlet APCD
temperatures less than 232°C, and 12.7 g TEQDF-WHO98 (12.3 g I-TEQDF) were produced by
cement kilns burning nonhazardous waste.
5-14
-------
Table 5-4. National emission estimates for cement kilns for reference years
1987 and 1995
Category
TEQ emission factor
(ng/kg clinker)
I-TEQDF
TEQDF-
WHO98
Activity
level
(billion kg
clinker/yr)
Annual TEQ emission
(g/yr)
I-TEQDF
TEQDF-
WHO98
Reference year 1987
Hazardous waste >232°C
Hazardous waste <232°C
Nonhazardous waste
28.58
1.04
0.26
30.7
1.11
0.27
TOTAL
3.8
1
47.2
52
108.6
1
12.3
122
116.7
1.1
12.7
131
Reference year 1995
Hazardous waste >232°C
Hazardous waste <232°C
Nonhazardous waste
28.58
1.04
0.26
30.7
1.11
0.27
TOTAL
5.04
1.26
61.3
67.6
144
1.3
15.9
161
154.7
1.4
16.6
173
In 1995, a total of 173 g TEQDF-WHO98 (161 g I-TEQDF) were produced by cement kilns.
Of this amount, 154.7 g TEQDF-WHO98 (144 g I-TEQDF) were produced by hazardous waste-
burning cement kilns with inlet APCD temperatures greater than 232°C, 1.4 g TEQDF-WHO98
(1.3 g I-TEQDF) were produced by cement kilns burning hazardous waste with inlet APCD
temperatures less than 232°C, and 16.6 g TEQDF-WHO98 (15.9 g I-TEQDF) were produced by
cement kilns burning nonhazardous waste.
The overall rating for these emission estimates is low because the emission factors had a
low confidence rating.
5.1.6.2. Estimates for Reference Year 2000
National estimates of CDD/CDF air emissions (g TEQ/yr) from all Portland cement kilns
for reference year 2000 were made by multiplying the average TEQ emission factors by an
estimate of the annual activity level (cement clinker produced) for the nonhazardous waste
category. The TEQ emission factors presented in Table 5-3 (above) and the activity levels
established in Section 5.1.5 were used to calculate the annual national TEQ emission estimates
shown in Table 5-5. For 2000, 17.2 g TEQDF-WHO98 (16.6 g I-TEQDF) were emitted by cement
kilns not burning hazardous waste. The emission estimates for cement kilns not burning
hazardous waste in 2000 have an overall low confidence rating because of the small number of
tested facilities providing the basis for the emission factor. The emission factor may not be
5-15
-------
representative of emissions from cement kilns not burning hazardous waste. However, the
activity level is given a high confidence rating because it is based on measured data.
In reference year 2000, 22 cement kilns burning hazardous waste as supplemental fuel
were operational in the United States, and most had been stack tested and evaluated for dioxin
emissions. Using the measured data from the stack tests, EPA estimated that 18.8 g TEQDF-
WHO98 (16.6 g I-TEQDF) were released from all cement kilns burning hazardous waste in 2000.
The emission estimates for cement kilns burning hazardous waste in 2000 have an overall high
confidence rating because the emission factors have a high confidence rating due to the large
number of tested facilities providing the basis for the emission factor. The activity level is given
a high confidence rating because the amount of clinker produced by cement kilns burning
hazardous waste is known.
5.1.7. EPA Regulatory Activities
In 1999 EPA promulgated national emission standards for new and existing cement kilns
burning and not burning hazardous waste (Federal Register, 1999a, b). A number of parties,
representing both industrial and environmental communities, requested judicial review of the rule
affecting cement kilns and incinerators burning hazardous waste and challenged EPA's emission
standards and several implementation provisions. On July 24, 2001, the U.S. Court of Appeals
for the District of Columbia Circuit vacated the emission standards. In response to this action,
EPA proposed new standards affecting cement kilns and incinerators burning hazardous waste on
April 20, 2004 (Federal Register, 2004), which were made final in December 2005 (Federal
Register, 2005).
With regard to dioxin emissions, the regulations are specific to the I-TEQ concentration
in the combustion gases leaving the stack. Existing and new cement kilns either combusting or
not combusting hazardous waste as supplemental fuel cannot emit more than 0.2 ng I-TEQ/dscm.
In addition, the temperature of the combustion gases measured at the inlet to the APCD cannot
exceed 232°C. The rule required owners or operators of facilities to test for CDDs/CDFs every
2V2 years. EPA expects this rule to reduce I-TEQDF emissions from existing and new facilities by
more than 30% over the next several years (Federal Register, 1999a, b, 2004, 2005).
5.1.8. Solid Waste from Cement Manufacturing: Cement Kiln Dust
EPA characterized CKD (the solid residual material generated during the manufacturing
of cement) in a report to Congress (U.S. EPA, 1993a) that was based in part on a 1991 survey of
cement manufacturers conducted by the Portland Cement Association (PCA). Survey responses
were received from 64% of the active cement kilns in the United States. On the basis of the
5-16
-------
Table 5-5. National emission estimates for cement kilns for reference
year 2000
Tested facility ID
Hazardous waste
burning kilns
200
201
203
204
205
206
207
208
228
300
302
303
318
319
322
323
403
404
473
491
680
681
Total Hazardous
waste burning kilns
Total Non-hazardous
waste burning kilns
All cement kilns
Estimate in g I-TEQ/yr
0.61
0.01
3.77
3.15
0.14
1.12
0.05
0.88
0.06
3.45
0.23
0.13
0.27
0.50
0.04
0.06
0.32
0.06
0.11
1.59
0.01
0.06
16.6
16.6
33.2
Estimate in g WHO-TEQ/yr
0.67
0.01
4.24
3.63
0.16
1.39
0.06
1.26
0.07
3.53
0.27
0.16
0.30
0.59
0.04
0.07
0.38
0.07
0.12
1.75
0.01
0.07
18.8
17.2
36.0
Source: U.S. EPA(2002a).
5-17
-------
survey responses, EPA estimated that in 1990 the U.S. cement industry generated about 12.9
million metric tons of gross CKD and 4.6 million metric tons of net CKD, of which 4.2 million
metric tons were land-disposed. The material collected by the APCD system is called gross CKD
(or as-generated CKD); it is either recycled back into the kiln system or removed from the system
for disposal (becoming net CKD or as-managed CKD). As discussed below, low levels of dioxin
have been measured in CKD. This material is disposed of in permitted landfills and therefore is
not considered to be an environmental release and is not included in the inventory of dioxin
releases presented in this report. However, for informational purposes only, estimates of the
amount of dioxin in CKD were developed for the reference years 1987, 1995, and 2000.
The PCA recently provided current estimates of the amount of CKD removed from the
manufacturing process for beneficial reuse and long-term management units (i.e., landfill
disposal) in 1990, 1995, and 2000 (e-mails dated January 31 and March 18, 2003, from Garth
Hawkins, Portland Cement Association, to Karie Riley, Versor, Inc.). Possible beneficial reuses
include municipal waste daily cover material, municipal waste landfill final cover material, soil
stabilization for roadways or other structures, waste neutralization/stabilization/solidification
(food wastes, hazardous wastes, etc.), and agricultural soil amendment. The PCA estimated that
the amount of CKD beneficially reused on or off site was 752 million kg in 1990, 652 million kg
in 1995, and 575 million kg in 2000. The amount of CKD disposed of annually in landfills was
estimated to be 2.7 billion kg in 1990, 3.1 billion kg in 1995, and 2.2 billion kg in 2000.
In its report to Congress (U.S. EPA, 1993a), EPA also included the results of sampling
and analysis of CKD and clinker conducted in 1992 and 1993. The purposes of the sampling and
analysis efforts were to (a) characterize the CDD/CDF content of clinker and CKD, (b) determine
the relationship, if any, between the CDD/CDF content of CKD and the use of hazardous waste
as fuel, and (c) determine the relationship, if any, between the CDD/CDF content of CKD and
the use of wet-process and dry-process cement kilns.
Clinker samples were collected from five cement kilns burning nonhazardous waste and
six kilns burning hazardous waste. CDDs/CDFs were not detected in any of the samples. Tetra-
through octa-chlorinated CDDs/CDFs were detected in the gross CKD samples obtained from 10
of the 11 kilns and in the net CKD samples obtained from 8 of the 11 kilns. The CDD/CDF
content ranged from 0.008 to 247 ng I-TEQDF/kg for gross CKD and from 0.045 to 195 ng I-
TEQDF/kg for net CKD. Analyses for seven PCB congeners were also conducted, but no
congeners were detected in any clinker or CKD sample.
Mean CDD/CDF concentrations in net CKD generated by the kilns burning hazardous
waste were higher (35 ng I-TEQDF/kg) than in net CKD generated by the facilities burning
nonhazardous waste (0.003 ng I-TEQDF/kg). These calculations of mean values treated nondetect
values as zero. If the nondetects had been excluded from the calculation of the means, the mean
5-18
-------
for net CKD from kilns burning hazardous waste would increase by a factor of 1.2 and the mean
for net CKD from kilns burning nonhazardous waste would increase by a factor of 1.7. One
sampled kiln had a net CKD TEQ concentration of more than two orders of magnitude greater
than the TEQ levels found in samples from any other kiln. If this kiln were considered atypical
of the industry (U.S. EPA, 1993 a) and were not included in the calculation, then the mean net
CKD concentration for hazardous waste-burning kilns would decrease to 2.9 ng I-TEQDF/kg.
CDD/CDF congener data for CKD from Holnam, Inc., Seattle, WA, were presented in a
report by the Washington State Department of Ecology (1998). The data were compiled and
evaluated to determine total I-TEQ concentrations and loadings. Nondetect values were included
as either zero, one-half of the detection limit (DL), or at the DL. The results of three separate
tests of CKD were as follows, assuming that nondetect values were zero:
I-TEQ I-TEQ
Date (ng/kg) (mg/day)
05/15/96 0.038 0.0038
10/21/97 0.67 0.0674
10/21/97 0.95 0.0948
EPA provided data for ashes from an ESP connected to a cement kiln and an FF
connected to a lightweight aggregate (LWA) kiln (U.S. EPA, 1999e). The average congener
concentrations for the ash samples are listed in Table 5-6. The average concentrations for the
cement kiln were determined from two different waste streams, each with five sample burns.
The average concentrations for the LWA kiln were determined using one waste stream with three
sample burns.
The amount of CDDs/CDFs associated with CKD was calculated for informational
purposes only. National estimates were divided among cement kilns burning hazardous waste
and those burning nonhazardous waste for both CKD that was beneficially reused and CKD that
was sent landfills. The activity levels used in the estimates were those provided by the PCA (e-
mail dated March 18, 2003, from Garth Hawkins, Portland Cement Association, to Karie Riley,
Versar, Inc.). The 1990 activity levels provided by PCA were used for reference year 1987. The
CDD/CDF concentrations in CKD used in the estimates were 35 ng I-TEQDF/kg for cement kilns
burning hazardous waste (which includes the high value discussed above) and 0.003 ng I-
TEQDF/kg for cement kilns burning nonhazardous waste.
As shown in Table 5-7, by combining the appropriate activity levels and CDD/CDF
concentrations, national estimates of CDDs/CDFs in CKD were developed for reference years
1987, 1995, and 2000. For cement kilns burning hazardous waste, approximately 4.2 g I-TEQDF
5-19
-------
Table 5-6. CDD/CDF concentrations (ng/kg) in ash samples from cement
kiln electrostatic precipitator and lightweight aggregate (LWA) kiln fabric
filter
Congener
2,3,7,8-TCDD
Total TCDD
1,2,3,7,8-PeCDD
Total PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Total HxCDD
12,2,3,4,6,7,8-HpCDD
Total HpCDD
OCDD
Cement
kiln
Avg.
cone.
0.429
36.1
0.886
54.9
1.03
2.36
2.47
173
17.7
55.2
21
LWA
kiln
Avg.
cone.
3.97
333
17.3
467
15.4
35.6
56.6
500
133
300
133
Total TCDD TEQs
2,3,7,8-TCDF
Total TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Total PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
Total HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Total HpCDF
OCDF
4.65
18.1
1.04
2.59
31.8
2.13
0.869
0.523
2.14
9.26
1.84
0.739
3.06
1.43
833
4630
100
267
2930
267
100
7.8
133
1230
167
22.6
2670
39.2
Total TCDF TEQs
Cement kiln
I-TEQ
0.429
-
0.443
-
0.103
0.236
0.247
-
0.177
-
0.021
1.66
0.465
-
0.0518
1.3
-
0.213
0.0869
0.0523
0.214
-
0.0184
0.00739
-
0.00143
2.41
WHO-
TEQ
0.429
-
0.886
-
0.103
0.236
0.247
-
0.177
-
0.0021
2.08
0.465
-
0.0518
1.3
-
0.213
0.869
0.0523
0.214
-
0.0184
0.00739
-
0.000143
3.19
LWA kiln
I-TEQ
3.97
-
8.65
-
1.54
3.56
5.66
-
1.33
-
0.133
24.8
83.3
-
5
133
-
26.7
10
0.780
13.3
-
1.67
0.226
-
0.0392
274
WHO-TEQ
3.97
-
17.3
-
1.54
3.56
5.66
-
1.33
-
0.133
33.5
83.3
-
5
133
-
26.7
10
0.780
13.3
-
1.67
0.226
-
0.00392
274
- = No data available
in 1987, 3.6 g I-TEQDF in 1995, and 3.3 g I-TEQDF in 2000 were produced from CKD that was
beneficially reused, and approximately 14.9 g I-TEQDF in 1987, 17.7 g I-TEQDF in 1995, and 12.8
g I-TEQDF in 2000 were produced from CKD that was disposed of in a landfill. For cement kilns
burning nonhazardous waste, approximately 0.0019 g I-TEQDF in 1987, 0.0016 g I-TEQDF in
1995, and 0.0014 g I-TEQDF in 2000 were produced from CKD that was beneficially reused, and
5-20
-------
Table 5-7. CDD/CDF estimates in cement kiln dust (CKD) for reference
years 1987,1995, and 2000
Category
CDD/CDF
concentration
(ng I-TEQDF/kg
of CKD)
CKD beneficially reused on or
off site
Activity level
(million kg
tons CKD/yr)
Annual TEQ
CDD/CDF
concentration
(g/yr)
CKD sent to a landfill for
disposal
Activity
level
(million kg
tons
CKD/yr)
Annual TEQ
CDD/CDF
concentration
(g/yr)
Reference year 1987
HW kilns
NHW kilns
35
0.003
120
632
4.2
0.0019
426
2,230
14.9
0.0067
Reference year 1995
HW kilns
NHW kilns
35
0.003
104
547
3.6
0.0016
505
2,642
17.7
0.0079
Reference year 2000
HW kilns
NHW kilns
35
0.003
94
480
3.3
0.0014
365
1,858
12.8
0.0056
HW = Hazardous waste
NHW = Nonhazardous waste
approximately 0.0067 g I-TEQDF in 1987, 0.0079 g I-TEQDF in 1995, and 0.0056 g I-TEQDF in
2000 were produced from CKD that was disposed of in a landfill.
EPA is currently developing CKD storage and disposal requirements. In 1999, a
proposed rule for the standards for the management of CKD was developed by EPA (Federal
Register, 1999a). Under the rule, CKD would remain a nonhazardous waste, provided that
proposed management standards are met, which would protect groundwater and control releases
of fugitive dust. Additionally, the rule proposes concentration limits on various pollutants in
CKD used for agricultural purposes (Federal Register, 1999c).
5.2. LIGHTWEIGHT AGGREGATE KILNS
LWA kilns heat raw materials such as clay, shale, or slate to expand the particles to form
lightweight materials for use in concrete products. In 1995, only 5 of the more than 36 LWA
kilns in the United States were burning hazardous waste; in 2000, 9 LWA kilns were burning
hazardous waste. LWA kilns are estimated to have emitted 3.3 g I-TEQDF to air in 1990 (Federal
Register, 1998a) and 2.4 g I-TEQDF in 1997 (Federal Register, 1999a); these estimates are used in
this report for reference years 1987 and 1995, respectively.
5-21
-------
The CDD/CDF emission factors for 2000 are based on the data for five LWA kilns tested
in 2000 (U.S. EPA, 2002a). They were calculated using the process described in Section 3.2.3.
The average emission factor for the LWA kilns was 1.986 ng TEQDF-WHO98/kg (2.063 ng
I-TEQDF/kg) of waste feed, assuming nondetect values of zero. These are assigned a low
confidence rating because the emission factor may not be representative of emissions from the
source category.
The amount of hazardous waste combusted using LWA kilns in 2000 was conservatively
estimated to be 903,000 metric tons, based on estimated activity levels derived for each halogen
acid furnace (HAF) in 2000. Data were available for all of the nine facilities operating in 2000.
A conservative estimate for the average annual quantity burned per HAF (100,280 metric tons/yr)
was derived by assuming that plants operate continuously throughout the year and are always
running at 80% of capacity. This quantity, multiplied by the total universe of nine facilities,
yielded the final estimate. Because the activity level was not derived from a survey but was
estimated, it is given a low confidence rating.
Equation 3-5 (Chapter 3, Section 3.2.4), used to calculate annual TEQ emissions for
dedicated hazardous waste incinerators, was also used to calculate annual TEQ emissions for
LWA kilns. Multiplying the average TEQ emission factors by the total estimated amount of
liquid hazardous waste burned in 2000 yields an annual emissions estimate. From this
procedure, the emissions from all LWA kilns burning hazardous waste as supplemental fuel were
estimated as 1.86 g TEQDF-WHO98 (1.79 g I-TEQDF) for 2000. Because of the low confidence
rating for the emission factor, the overall confidence rating for the emission estimates is low.
5.3. ASPHALT MIXING PLANTS
Asphalt consists of an aggregate of gravel, sand, and filler mixed with liquid asphalt
cement or bitumen. Filler typically consists of limestone, mineral stone powder, and sometimes
ash from power plants and municipal waste combustors. The exact composition of an asphalt
formulation depends on how it will be used. The aggregate typically constitutes more than 92%
by weight of the total asphalt mixture. The components of the aggregate are dried, heated to a
temperature ranging from 135 to 163°C, and then mixed and coated with the bitumen at an
asphalt mixing installation. "Old" asphalt (i.e., asphalt from dismantled bridges and roads) can
be heated and disaggregated to its original components and reused in the manufacture of new
asphalt (U.S. EPA, 1996c). "Hot mix" asphalt paving materials can be manufactured by batch
mix plants, continuous mix plants, parallel-flow drum mix plants, and counterflow drum mix
plants (U.S. EPA, 1998c).
Bremmer et al. (1994) reported the CDD/CDF emissions factor for an asphalt mixing
plant in the Netherlands as 47 ng I-TEQDF per metric ton of produced asphalt. No congener-
5-22
-------
specific emission factors were reported. The mixing plant heated old asphalt to about 150°C in
an individual recycling drum with kiln exhaust gases that were mixed with ambient air and
heated to a temperature of 300 to 400°C. Parallel to this recycling drum was the main drum,
which dried and heated the aggregate (sand and gravel/granite chippings) to a temperature of
about 220°C. The kiln exhaust gases leaving the recycling drum were led along the main burner
of the main drum for incineration. The old asphalt, the minerals from the main drum, and new
bitumen from a hot storage tank (about 180°C) were mixed in a mixer to form new asphalt.
Natural gas fueled the plant during the sample collection period, and 46% of the feed was old
asphalt. The plant's APCD system consisted of cyclones and an FF.
Umweltbundesamt (1996) reported lower emission factors for three tested facilities in
Germany that were also equipped with FFs. These three facilities were fueled by oil or butane
gas and used old asphalt at rates ranging from 30 to 60% of the feed. The emission factors
calculated from the stack gas concentrations, gas flow rates, and hourly throughputs for these
three facilities were 0.2, 3.5, and 3.8 ng I-TEQDF/metric ton of asphalt produced, respectively.
EPA conducted stack emissions testing at two hot-mix asphalt production plants in the
United States (U.S. EPA, 2000g). The Midwest Research Institute performed emission tests on
outlet of FFs that controlled emissions from the counter-flow rotary dryer process used at the
asphalt plant in Clayton, NC, and from the parallel-flow rotary dryer process used at the asphalt
plant in Gary, NC. In both processes, virgin aggregate of various sizes was fed to the drum by
cold-feed controls in proportions dictated by the final mix specifications. Aggregate was
delivered at the opposite end of the burner in the counter-flow continuous drum mix process and
at the same end as the burner in the parallel-flow continuous drum mix process. EPA developed
average TEQ emission factors from the tested facilities to represent hot-mix asphalt plants in
general (U.S. EPA, 2000g). The average TEQ emission factor for hot-mix asphalt plants was
1.40e-03 ng I-TEQ (1.46e-3 ng TEQDF-WHO98/kg) asphalt produced. Table 5-8 summarizes the
TEQ emission factors for hot-mix asphalt plants.
Because only two U.S. facilities have been tested, these emission factors are given a low
confidence rating. It is likely that these emission factors are not representative of CDD/CDF
emissions from all types of asphalt production facilities in the United States; nevertheless, they
are sufficient to derive a preliminary estimate of annual emissions from asphalt production
facilities in the United States for 2000.
Approximately 500 million tons of hot-mix asphalt paving materials were produced at
approximately 3,600 active asphalt plants in the United States in 1996 (U.S. EPA, 1998c). This
activity level was used to represent reference year 2000. The activity level is given a high
confidence level because it was based on a comprehensive survey. A preliminary estimate of
annual TEQ air releases for 2000 was calculated by multiplying the activity level (in kilograms)
5-23
-------
Table 5-8. Congener-specific emission factors (ng/kg asphalt produced)
derived from the testing of air emissions at two hot-mix asphalt plants in the
United States
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total TEQ
Mean <
I-TEQDF
9.53e-05
7.03e-05
1.91e-05
5.90e-05
4.45e-05
2.18e-05
1.13e-05
4.40e-05
9.75e-05
1.91e-04
1.81e-04
5.44e-05
8.62e-05
3.81e-04
2.95e-05
1.22e-05
2.18e-06
1.40e-03
emission factor
TEQDF-WH098
9.53e-05
1.41e-04
1.91e-05
5.90e-05
4.45e-05
2.18e-05
1.13e-06
4.40e-05
9.75e-05
1.91e-04
1.81e-04
5.44e-05
8.62e-05
3.81e-04
2.95e-05
1.22e-05
2.18e-07
1.50e-03
Source: U.S. EPA(2000g).
by the average emission factor of 1.40e-03 ng I-TEQ (1.46e-3 ng TEQDF-WHO98/kg) asphalt
produced. Approximately 0.70 g I-TEQ (0.73 g TEQDF-WHO98) were emitted into the air from
asphalt plants in the United States in 2000. This estimate is assigned a low confidence level
because the emission factors are given a low confidence level rating.
5.4. PETROLEUM REFINING CATALYST REGENERATION
Regeneration of spent catalyst from the reforming process at petroleum refineries is a
potential source of CDDs/CDFs, according to limited testing conducted in the United States
(Amendola and Barna, 1989; Kirby, 1994), Canada (Maniff and Lewis, 1988; Thompson et al.,
1990), and the Netherlands (Bremmer et al., 1994). This section summarizes the catalyst
regeneration process, relevant studies performed to date, and the status of EPA regulatory
investigations of this source.
5-24
-------
Catalytic reforming is the process used to produce high-octane reformates from lower-
octane reformates for blending of high-octane gasolines and aviation fuels. The reforming
process occurs at high temperature and pressure and requires the use of a platinum or
platinum/rhenium catalyst. During the reforming process, a complex mixture of aromatic
compounds, known as coke, is formed and deposited onto the catalyst. As coke deposits onto the
catalyst, its activity is decreased. The high cost of the catalyst necessitates its regeneration.
Catalyst regeneration is achieved by removing the coke deposits via burning at temperatures of
399 to 454°C and then reactivating the catalyst at elevated temperatures (454 to 538°C) using
chlorine or chlorinated compounds (e.g., methylene chloride, 1,1,1-trichloroethane, and ethylene
dichloride; most refineries use chlorine of perchloroethylene). Burning of the coke produces kiln
exhaust gases that can contain CDDs and CDFs along with other combustion products. Because
kiln exhaust gases, if not vented directly to the atmosphere, may be scrubbed with caustic or
water, internal effluents may become contaminated with CDDs/CDFs (Kirby, 1994; SAIC,
1994).
Three basic catalyst regeneration processes are used: semi-regenerative, cyclic, and
continuous. During the semi-regenerative process, the entire catalytic reformer is taken offline.
In the cyclic process, one of two (or more) reforming reactors is taken offline for catalyst
regeneration; the remaining reactor(s) remains on line so that reforming operations continue. In
the continuous process, aged catalyst is continuously removed from one or more on-line stacked
or side-by-side reactors, regenerated in an external regenerator, and then returned to the system;
the reforming system, consequently, never shuts down (SAIC, 1994).
In 1988, a study by the Canadian Ministry of the Environment (Maniff and Lewis, 1988)
detected concentrations of CDDs ranging from 1.8 to 22.2 |ig/L and CDFs ranging from 4.4 to
27.6 |ig/L in an internal waste stream of spent caustic at a petroleum refinery. The highest
concentration of 2,3,7,8-TCDD was 0.0054 |ig/L. CDDs were also observed in the refinery's
biological sludge at a maximum concentration of 74.5 |ig/kg, and CDFs were observed at a
maximum concentration of 125 |ig/kg. The concentration of CDDs/CDFs in the final combined
refinery plant effluent was below the DLs.
Amendola and Barna (1989) reported detecting trace levels of hexa- to octa-CDDs and
CDFs in untreated wastewaters (up to 2.9 pg I-TEQDF/L) and wastewater sludges (0.26 to 2.4 ng
I-TEQDF/kg) at a refinery in Ohio. The levels of detected total CDDs/CDFs in the wastewater
and sludge were much lower (<3 ng/L and <1 |ig/kg, respectively) than the levels reported by
Maniff and Lewis (1988). No CDDs/CDFs were detected in the final treated effluent (less than
0.2 ng I-TEQDF/L). The data collected in the study were acknowledged to be too limited to
enable identifying the source(s) of the CDDs/CDFs within the refinery. The study authors also
presented in an appendix to their report the results of analyses of wastewater from the catalyst
5-25
-------
regeneration processes at two other U.S. refineries. In both cases, untreated wastewaters
contained CDDs/CDFs at levels ranging from high picograms to low nanograms per liter (results
were reported for congener group totals, not specific congeners). However, CDDs/CDFs were
not detected in the only treated effluent sample collected at one refinery.
Thompson et al. (1990) reported total CDD and CDF concentrations of 8.9 ng/m3 and 210
ng/m3, respectively, in stack gas samples from a Canadian petroleum refinery's reforming
operation. They also observed CDDs/CDFs in the picogram- to nanogram-per-liter range in the
internal washwater from a scrubber of a periodic/cyclic regenerator.
Beard et al. (1993) conducted a series of benchtop experiments to investigate the
mechanism(s) of CDD/CDF formation in the catalytic reforming process. A possible pathway
for the formation of CDFs was found, but the results could not explain the formation of CDDs.
Analyses of the kiln exhaust gas from burning coked catalysts revealed the presence of
unchlorinated dibenzofuran in quantities up to 220 |ig/kg of catalyst. Chlorination experiments
indicated that dibenzofuran and, possibly, biphenyl and similar hydrocarbons act as CDF
precursors and can become chlorinated in the catalyst regeneration process. Corrosion products
on the steel piping of the process plant seemed to be the most likely chlorinating agent.
In May 1994, EPA's Office of Water conducted a sampling and analytical study of
catalyst regeneration wastewater for CDDs/CDFs at three petroleum refining plants (Kirby,
1994). The study objectives were to determine the analytical method best suited for determining
CDDs/CDFs in refinery wastewater and to screen and characterize wastewater discharges from
several types of reforming operations for CDDs/CDFs. The report for this study (Kirby, 1994)
also presented results submitted voluntarily to EPA by two other facilities. The sampled internal
untreated wastewaters and spent caustics were found to contain a wide range of CDD/CDF
concentrations, 0.1 pg I-TEQDF/L to 57.2 ng I-TEQDF/L. The study results also showed that 90%
of the TEQ was contained in the wastewater treatment sludges generated during the treatment of
wastewater and caustic from the regeneration process.
In 1995, EPA issued a notice of its proposed intent to not designate spent reformer
catalysts as a listed hazardous waste under RCRA (Federal Register, 1995b). The final rule was
issued in August 1998 (Federal Register, 1998b). The Agency's assessment of current
management practices associated with recycling of reforming catalyst found no significant risks
to human health or the environment. The Agency estimated that 94% of the approximately 3,600
metric tons of spent reformer catalyst sent off site by refineries were being recycled for their
precious metal content. EPA made no determination of the "listability" of spent caustic residuals
formed during regeneration of spent reforming catalyst, but it did identify the potential air
releases from the combustion of the reforming catalyst prior to reclamation as possibly being of
concern. The Agency requested comments on (a) opportunities for removing dioxin prior to
5-26
-------
discharge of scrubber water into the wastewater treatment system, (b) opportunities to segregate
this wastestream, and (c) potential health risks associated with insertion of dioxin-contaminated
media back into the refinery process (such as the coker). In this proposed rulemaking, EPA also
noted the possibility of dioxin releases to air during regeneration operations.
As part of its regulatory investigation under RCRA, EPA's Office of Solid Waste
commissioned a study to analyze and discuss existing data and information concerning
CDD/CDF formation in the treatment of catalytic reformer wastes. This report (SAIC, 1994)
also identified potential process modifications that may prevent the formation of CDDs/CDFs.
The report's authors concluded that, although the available data indicate that CDDs/CDFs can be
generated during the catalyst regeneration process, the available data indicate that CDD/CDF
concentrations in treated wastewater and in solid waste are minimal. Releases to air could result
from vented kiln exhaust gases at some facilities. In addition, the CDDs/CDFs formed could
possibly be reintroduced into other refining operations (e.g., the coker) and resulting products.
In 1998, emissions from the caustic scrubber used to treat gases from the external
regeneration unit of a refinery in California were tested (CARB, 1999). This facility uses a
continuous regeneration process. The reactor is not taken offline during regeneration; rather,
small amounts of catalyst are continuously withdrawn from the reactor and are regenerated. The
emissions from the regeneration unit are neutralized by a caustic scrubber before being vented to
the atmosphere. The catalyst recirculation rate during the three tests ranged from 733 to 1,000
Ib/hr.
All 2,3,7,8-substituted CDDs/CDFs were detected in each of the three samples collected.
The average emission factors in units of nanograms per barrel of reformer feed are presented in
Table 5-9. The congener profile is presented in Figure 5-4. The samples showed a wide range in
concentrations of the CDD/CDF congeners (up to a fivefold difference); however, the congener
profile was consistent in all samples. The concentrations of the individual furan congener groups
were always higher than the concentrations of the corresponding dioxin congener group. The
average TEQDF-WHO98 emission factor for these three tests was 3.18 ng TEQ/barrel and the
average I-TEQDF was 3.04 ng TEQ/barrel.
In 1991, stack testing was performed on the exhaust from one of the three semi-
regenerative catalytic reforming units of a refinery in California (Radian Corporation, 1991). At
these units, a caustic solution is introduced to the exhaust to neutralize HC1 emissions from the
catalyst beds prior to release to the atmosphere. The tested unit was considered to be
representative of the other units. Each unit is periodically taken offline (approximately once a
year) so the catalyst beds can be regenerated. The tested unit has a feed capacity of 7,000
barrels/day. Approximately 59,500 pounds of catalyst were regenerated during the tested
regeneration cycle, which lasted for 62 hr.
5-27
-------
The average emission factors for this facility (in units of nanogams per barrel of reformer
feed) are presented in Table 5-9 and the congener profile is presented in Figure 5-4. The
majority of the 2,3,7,8-substituted CDD congeners were not detected during testing. In contrast,
the majority of the 2,3,7,8-substituted CDF congeners were detected. The average TEQDF-
WHO98 emission factor (assuming nondetect values were zero) was 1.04e-03 ng TEQ/barrel and
the average I-TEQDF emission factor was l.Ole-03 ng TEQ^arrel. These values are three orders
of magnitude less than the emission factor reported in CARB (1999). The calculation of these
emission factors involved several assumptions: the unit is regenerated once per year, the unit
operates at capacity (7,000 barrels/day), and the facility operates 362 days/yr.
The average of the two facility emission factors, 1.59 ng TEQDF-WHO98/barrel (1.52 ng I-
TEQDF/barrel) of reformer feed, is assumed to apply to all reference years (1987, 1995, and 2000)
and is assigned a low confidence rating. Only one continuous and one semiregenerative unit in
the United States have been tested. Combined, these two facilities represent less than 1% of the
catalytic reforming capacity in U.S. petroleum refineries in 1987 (3.805 million barrels/day),
1995 (3.867 million barrels/day), and 2000 (3.770 million barrels/day) (EIA, 2002a). The
average emission factor developed above assumes that emissions are proportional to reforming
capacity; however, they may be more related to the amount of coke burned, the APCD equipment
present, or other process parameters.
The national daily average catalytic reforming capacities in the United States were 3.805,
3.867, and 3.770 million barrels per day for 1987, 1995, and 2000, respectively (EIA, 2002).
These were assigned a high confidence rating because they are based on comprehensive surveys
of industry. If it is conservatively assumed that all units operated at full capacity in all three
years, then applying the average emission factors of TEQ/barrel yields annual emissions of 2.21 g
TEQDF-WHO98 (2.11 g I-TEQDF) in 1987, 2.24 g TEQDF-WHO98 (2.14 g I-TEQDF) in 1995, and
2.19 g TEQDF-WHO98 (2.09 g I-TEQDF) in 2000. These emissions have a low confidence rating
because they are based on an emission factor with a low confidence rating.
5.5. CIGARETTE SMOKING
Bumb et al. (1980) were the first to report that cigarette smoking is a source of CDD
emissions. Subsequent studies by Muto and Takizawa (1989), Ball et al. (1990), and Lofroth and
Zebiihr (1992) also reported the presence of CDDs as well as CDFs in cigarette smoke. A study
by Matsueda et al. (1994) reported on the CDD/CDF content of tobacco from 20 brands of
cigarettes from seven countries. Although a wide range in the concentrations of total
CDDs/CDFs and total TEQs were reported in these studies, similar congener profiles and
patterns were reported. The findings of each of these studies are described in this section.
5-28
-------
Table 5-9. CDD/CDF emission factors (ng/barrel)a for petroleum catalytic
reforming units
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Semiregenerative unit
Nondetect set
to zero
ND
5.69e-05
4.22e-05
ND
ND
7.02e-04
2.55e-03
2.32e-04
4.68e-04
1.09e-03
1.06e-03
1.07e-03
ND
1.24e-03
2.94e-03
8.32e-04
l.Ole-03
3.35e-03
9.94e-03
l.Ole-03
1.03e-03
ND
3.56e-04
1.28e-03
1.39e-03
2.55e-03
2.70e-03
5.12e-03
7.85e-03
4.88e-03
l.Ole-03
2.71e-02
Nondetect set to
Vz detection limit
2.35e-05
9.58e-05
8.09e-05
5.52e-05
5.10e-05
7.02e-04
2.55e-03
2.32e-04
4.68e-04
1.09e-03
1.06e-03
1.07e-03
6.82e-05
1.24e-03
2.94e-03
8.32e-04
l.Ole-03
3.56e-03
l.OOe-02
1.07e-03
1.12e-03
2.35e-05
3.56e-04
1.28e-03
1.39e-03
2.55e-03
2.70e-03
5.12e-03
7.85e-03
4.88e-03
l.Ole-03
2.72e-02
Continuous regeneration unit
Nondetect set
to zero
1.61e-02
2.87e-01
3.47e-01
8.45e-01
5.56e-01
3.02e+00
1.71e+00
6.10e-01
1.72e+00
2.33e+00
4.70e+00
3.58e+00
4.34e-01
3.10e+00
1.59e+01
1.45e+00
3.75e+00
6.78e+00
3.76e+01
3.04e+00
3.18e+00
6.84e+00
5.61e+00
8.18e+00
6.58e+00
1.71e+00
4.68e+01
3.30e+01
2.96e+01
2.11e+01
3.75e+00
1.63e+02
Nondetect set to
Vz detection limit
1.61e-02
2.87e-01
3.47e-01
8.45e-01
5.56e-01
3.02e+00
1.71e+00
6.10e-01
1.72e+00
2.33e+00
4.70e+00
3.58e+00
4.34e-01
3.10e+00
1.59e+01
1.45e+00
3.75e+00
6.78e+00
2.18e+01
3.04e+00
3.18e+00
6.84e+00
5.61e+00
8.18e+00
6.58e+00
1.71e+00
4.68e+01
3.30e+01
2.96e+01
2.11e+01
3.75e+00
1.63e+02
aOne barrel assumed to be equivalent to 139 kg.
ND = Not detected
Sources: Radian Corporation (1991); CARB (1999).
5-29
-------
Ratio (mean congener emission factor/total CDD/CDF emission factor)
Oa 0.02 0.04 0.06 0.08 0.1 0.12
Ratio (mean congener emission factor/total CDD/CDF emission factor)
Oa 0.05 0.1 0.15 0.2 0.25 0.3
0.35
3OOOOOOOOOOOOOOOOOOOQ
Figure 5-4. Congener and congener group profiles for air emissions from
petroleum catalytic reforming units (nondetect set equal to zero).
Sources: CARB (1999); Radian Corporation (1991).
5-30
-------
No studies published to date have demonstrated a mass balance, and it is not known
whether the CDDs/CDFs measured in cigarette smoke are the result of formation during tobacco
combustion, volatilization of CDDs/CDFs present in the unburned tobacco, or a combination of
these two source mechanisms. The combustion processes operating during cigarette smoking are
complex and could be used to justify both source mechanisms. As reported by Guerin et al.
(1992), during a puff on a cigarette, gas-phase temperatures reach 850°C at the core of the
firecone, and solid-phase temperatures reach 800°C at the core and 900°C or greater at the char
line. Thus, temperatures are sufficient to cause at least some destruction of CDDs/CDFs initially
present in the tobacco. Both solid- and gas-phase temperatures rapidly decline to 200 to 400°C
within 2 mm of the char line.
Formation of CDDs/CDFs has been reported in combustion studies with other media in
this temperature range of 200 to 900°C. However, it is known that a process likened by Guerin
et al. (1992) to steam distillation takes place in the region behind the char line because of high,
localized concentrations of water and temperatures of 200 to 400°C. At least 1,200 tobacco
constituents (e.g., nicotine, n-paraffin, some terpenes) are transferred intact from the tobacco into
the smoke stream by distillation in this region, and it is plausible that CDDs/CDFs present in the
unburned tobacco would be subject to similar distillation.
Bumb et al. (1980), using low-resolution mass spectrometry, analyzed the CDD content
of mainstream smoke from the burning of a U.S. brand of unfiltered cigarette. A package of 20
cigarettes was combusted in each of two experiments. Approximately 20 to 30 puffs of 2 to 3 sec
duration were collected from each cigarette on a silica column. Hexa-, hepta-, and octa-CDDs
were detected at levels of 0.004 to 0.008, 0.009, and 0.02 to 0.05 ng/g, respectively.
Muto and Takizawa (1989) employed a continuous smoking apparatus to measure CDD
congener concentrations in the mainstream smoke generated from the combustion of one kind of
filtered cigarette (brand not reported). The apparatus pulled air at a constant continuous rate
(rather than a pulsed rate) through a burning cigarette and collected the smoke on a series of traps
(glass fiber filter, polyurethane foam, and XAD-n resin). The CDD content of the smoke as well
as the CDD content of the unburned cigarette and the ash from the burned cigarettes were also
analyzed using low-resolution mass spectrometry. The results are presented in Table 5-10, and
the congener group profiles are presented in Figure 5-5. Table 5-11 and Figure 5-6 present the
mainstream smoke results on a mass-per-cigarette basis to enable comparison with the results of
other studies.
The major CDD congener group found was HpCDD, which accounted for 84% of total
CDDs found in the cigarette, 94% of total CDDs found in smoke, and 99% of total CDDs found
in the ash. The 2,3,7,8-HpCDDs also accounted for the majority of the measured TEQ in the
5-31
-------
Table 5-10. CDD concentrations in Japanese cigarettes, smoke, and ash
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Cigarette
(Pg/g)
ND (0.5)
ND (0.5)
2.0P
a
a
1343
257
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
1602.01
NR
13.9
13.7
44.9
ND (0.5)
13.41
1629
257
NR
NR
NR
NR
NR
1944
Concentrations
Mainstream smoke
(ng/m3)
ND (0.22)
0.43
2.153
a
a
783
240
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
1025.58
NR
8.5
8.3
68
1.51
7.51
4939
240
NR
NR
NR
NR
NR
5256
Ash
(Pg/g)
ND (0.5)
ND (0.5)
0.563
a
a
ND (0.5)
ND (0.5)
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
0.56
NR
0.06
0.06
4.63
ND (0.5)
5.01
3211
ND (0.5)
NR
NR
NR
NR
NR
3221
"Value reported only for total 2,3,7,8-substituted HxCDDs.
ND = Not detected (value in parenthesis is the detection limit)
NR = Not reported
Source: Muto and Takizawa (1989).
5-32
-------
Figure 5-5. CDD profiles for Japanese cigarettes, smoke, and ash.
Source: Matsueda et al. (1994).
cigarettes and smoke; however, none were measured in the ash. Although no PeCDDs were
detected in the cigarette, they were detected at low levels in the smoke, indicating probable
formation during combustion. On the basis of the similarities in the congener group profiles for
the three media, the study authors concluded that most of the CDDs found in the cigarette smoke
result from volatilization of CDDs/CDFs present in the unburned cigarette rather than being
formed during combustion.
Ball et al. (1990) measured the CDD/CDF content of mainstream smoke for the 10 best-
selling German cigarette brands. The international test approach (1 puff/min; puff flow rate of
35 mL/2 sec) was employed with an apparatus that smoked 20 cigarettes at a time in three
successive batches and had a large collection device. The average TEQ content (on both an I-
TEQDF and a TEQDF-WHO98 basis) in mainstream smoke for the 10 brands tested, normalized to
a mass-per-cigarette basis, was 0.09 pg/cigarette (i.e., 16.5 times less than the value reported by
Muto and Takizawa, 1989, for a Japanese cigarette brand). However, the congener group
5-33
-------
Table 5-11. CDD/CDF concentrations (pg/cig) in cigarette smoke,
normalized to a per-cigarette basisa
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Muto and
Takizawa (1989)
(1 Japanese brand)
(mainstream
smoke)
ND (0.04)
0.075
0.376
b
b
137
42
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
179.45
NR
1.49
1.49
11.9
0.264
1.31
864
42
NR
NR
NR
NR
NR
919.47
Ball et al. (1990)
(avg. of 10 German
brands)
(mainstream
smoke)
ND (0.03)
ND (0.03)
0.06
0.05
0.04
1.3
3.4
0.19
0.13
0.04
ND (0.03)
0.03
0.03
0.05
0.16
0.03
0.11
4.85
0.77
0.09
0.09
0.51
0.14
0.53
2.9
3.4
1.41
0.83
0.35
0.27
0.11
10.45
Lofroth and
Zebiihr (1992)
(1 Swedish brand)
(mainstream
smoke)
0.028
0.15
0.1
0.34
0.25
6.05
22.1
1.2°
0.34C
0.34
1.3C
0.48
0.14
0.21
10
2.6
3.2
29.02
19.81
0.9
0.96
0.61
1.07
2.52
12.3
22.1
4.5
3.23
5.3
19.8
3.2
74.63
Lofroth and
Zebiihr (1992)
(1 Swedish brand)
(sidestream smoke)
0.07
0.32
0.19
0.6
0.55
12.2
38.8
2.1C
0.8C
0.6
3.8C
1.2
0.39
0.5
23.5
5
10.7
52.7
48.6
1.96
2.08
0.67
2.14
5.2
21.3
38.8
5.75
6.35
12.9
47.8
10.7
151.6
"Emissions calculated assuming 0.0035 m3 of smoke are inhaled per 20 cigarettes smoked (Muto and Takizawa,
1992).
bMuto and Takizawa (1989) reported a value only for total 2,3,7,8-HxCDDs (0.38 pg/cig).
"Concentrations listed include the contribution of a coeluting non-2,3,7,8-substituted congener.
ND = Not detected (value in parenthesis is the detection limit)
NR = Not reported
5-34
-------
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
TCDD PeCDD HxCDDHpCDD OCDD TCDF PeCDF HxCDF HpCDF OCDF
| Japanese Mainstream (row 1)
| Swedish Mainstream (row 3)
| German Mainstream (row 2)
| Swedish Sidestream (row 4)
Figure 5-6. Congener group profiles for mainstream and sidestream
cigarette smoke.
Source: Matsueda et al. (1994).
profiles were similar to those reported by Muto and Takizawa, with HpCDD and OCDD the
dominant congener groups found.
Lofroth and Zebiihr (1992) measured the CDD/CDF content of mainstream and
sidestream smoke from one common Swedish cigarette brand. The cigarette brand was labeled
as giving 17 mg carbon monoxide, 21 mg tar, and 1.6 mg nicotine. The international test
approach was used, and the smoke was collected on glass fiber filters followed by two
polyurethane plugs. The analytical results for mainstream and sidestream smoke are presented in
Table 5-11. The TEQ content in mainstream smoke, normalized to a mass-per-cigarette basis,
was 0.96 pg TEQDF-WHO98/cigarette (0.9 pg I-TEQDF/cigarette) (i.e., about two times less than
the value reported by Muto and Takizawa, 1989, and 10 times greater than the average value
reported by Ball et al., 1990). As in the Muto and Takizawa and Ball et al. studies, the dominant
congener groups were HpCDDs and OCDD; however, HpCDFs were also relatively high in
5-35
-------
comparison with the other congener group totals. The sidestream smoke contained 2.08 pg
TEQDF-WHO98/cigarette (1.96 pg I-TEQDF/cigarette), or twice that of mainstream smoke.
Using high-resolution mass spectrometry, Matsueda et al. (1994) analyzed the CDD/CDF
content of tobacco from 20 brands of commercially available cigarettes collected in 1992 from
Japan, the United States, Taiwan, China, the United Kingdom, Germany, and Denmark. Table 5-
12 presents the study results. The total CDD/CDF content ranged from 109 to 1,136 pg/pack,
and total TEQDF-WHO98 content ranged from 1.9 to 14 pg/pack (1.4 to 12.6 pg/pack on an I-
TEQDF basis). The Chinese cigarette brand contained significantly lower CDDs/CDFs and TEQs
than did any other brand of cigarette. Figure 5-7 depicts the congener group profiles for the
average results for each country. A high degree of similarity is seen among the CDF congener
group profiles of the tested cigarette brands. The Japanese and Taiwanese cigarettes show CDD
congener group profiles different from those of the other countries' cigarettes.
Brown (2002) estimated that 440 billion cigarettes were consumed in the United States in
2000. In 1995, approximately 487 billion cigarettes were consumed in the United States and by
U.S. overseas armed forces personnel. In 1987, approximately 575 billion cigarettes were
consumed. According to The Tobacco Institute (1995), per capita U.S. cigarette consumption,
based on the total U.S. population aged 16 and over, was a record high of 4,345 in 1963,
declining to 2,415 in 1995 and 1,563 in 2000 (USDA, 1997; U.S. Census Bureau, 2000). The
activity level estimates by Brown (2002) were adopted, and a high confidence rating is assigned
because they are based on known consumption rates.
The available emission factor data presented above provide the basis for two methods of
estimating the amount of TEQs that may have been released to the air in the United States in
2000, 1995, and 1987 from the combustion of cigarettes. The confidence rating assigned to the
emission factor is low because of the very limited amount of testing performed to date. First, an
annual emission estimate for 2000 of 0.19 g TEQ (on a TEQDF-WHO98 or I-TEQDF basis) is
obtained if it is assumed that (a) the average TEQ content of seven brands of U.S. cigarettes
reported by Matsueda et al. (1994)—6.3 pg TEQDF-WHO98/pack (5.5 pg I-TEQDF/pack)—is
representative of cigarettes smoked in the United States, (b) CDDs/CDFs are not formed and the
congener profile reported by Matsueda et al. (1994) is not altered during combustion of
cigarettes, and (c) all CDDs/CDFs contributing to the TEQ are released from the tobacco during
smoking.
The available emission factor data presented above provide the basis for two methods of
estimating the amount of TEQs that may have been released to the air in the United States in
2000, 1995, and 1987 from the combustion of cigarettes. The confidence rating assigned to the
emission factor is low because of the very limited amount of testing performed to date. First, an
5-36
-------
Table 5-12. CDD/CDF concentrations (pg/pack) in tobacco cigarette brands
from various countries
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCD
^
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
U.S.
(avg. of 7
brands)
1.2
1.6
6.9
a
a
52.7
589.3
18.2
8.7
b
8.1
C
C
C
17.6
d
24.6
651.7
77.2
7.1
7.3
47.1
27.6
40.6
108.7
589.3
183.8
57.7
29.1
27.3
24.6
1,135.8
Japan
(avg. of 6
brands)
0.5
1.4
4.8
a
a
17.8
244
4.8
5.3
b
8.1
C
C
C
11.1
d
10.5
268.5
39.8
3.8
4.3
296.3
33.6
29.2
40
244
102.1
45.9
26.4
16.6
10.5
844.6
United
Kingdom
(avg. of 3
brands)
1.7
3.1
6.1
a
a
23.9
189.5
15.6
21.2
b
17
C
C
C
13.6
d
8.3
224.3
75.7
8.8
10.1
85.1
62.9
49.2
47.7
189.5
348.9
134.5
51.3
19
8.3
996.4
Taiwan
(1 brand)
1
3.3
12.2
a
a
26.4
272.7
11
16
b
12.9
C
C
C
13.2
d
13.9
315.6
67
7.7
9.1
329
150.5
99.4
62
272.7
372.1
149.1
45.8
18.5
13.9
1513
China
(1 brand)
ND
1.1
1.1
a
a
2.2
28.2
1.2
1.5
b
2.2
C
C
C
1.5
d
0.5
32.6
6.9
1.1
1.7
9.7
5.2
5.4
3.8
28.2
35.4
11.2
7.8
1.7
0.5
108.9
Denmark
(1 brand)
0.5
0.8
6.2
a
a
53.3
354.3
2.2
4.3
b
4.3
C
C
C
7
d
10.5
415.1
28.3
3.4
3.4
17
9.8
26.7
93.1
354.3
97.8
35.5
18.1
11.1
10.5
673.9
Germany
(1 brand)
1.1
3.3
5.7
a
a
32.7
288.6
7.9
14.4
b
13.2
C
C
C
12.9
d
13.9
331.4
62.3
6.9
8.3
49.5
40.8
40.6
60.2
288.6
233.4
97.5
40.8
21.2
13.9
886.5
"Value reported only for total 2,3,7,8-substituted HxCDDs.
bValue reported only for total 2,3,7,8-substituted PeCDFs.
"Value reported only for total 2,3,7,8-substituted HxCDFs.
dValue reported only for total 2,3,7,8-substituted HpCDFs.
Source: Matsueda et al. (1994).
annual emission estimate for 2000 of 0.19 g TEQ (on a TEQDF-WHO98 or I-TEQDF basis) is
obtained if it is assumed that (a) the average TEQ content of seven brands of U.S. cigarettes
5-37
-------
I
b.
C
a
B
o
a*
0.6
0.5
0.4
0.3
0.2
0.1
/A
/A
TCDD PeCDD HxCDD HpCDD OCDD TCDF PeCDF HxCDF HpCDF OCDF
^| China (column 1) | Denmark (column 2) ^| Japan (column 3)
Germany (column 4) ^^H United Kingdom (column 5)
I United States (column 6)
Figure 5-7. Congener group profiles for cigarette tobacco from various
countries.
Source: Matsuedaetal. (1994).
reported by Matsueda et al. (1994)—6.3 pg TEQDF-WHO98/pack (5.5 pg I-TEQDF/pack)—is
representative of cigarettes smoked in the United States, (b) CDDs/CDFs are not formed and the
congener profile reported by Matsueda et al. (1994) is not altered during combustion of
cigarettes, and (c) all CDDs/CDFs contributing to the TEQ are released from the tobacco during
smoking.
The second method of estimating is based on the assumption that the TEQ emission rates
for a common Swedish brand of cigarette reported by Lofroth and Zebiihr (1992) for mainstream
smoke (0.96 pg TEQDF-WHO98/cigarette [0.9 pg I-TEQDF/cigarette]) and sidestream smoke (2.08
pg TEQDF-WHO98/cigarette [1.96 pg I-TEQDF/cigarette]) are representative of the emission rates
for U.S. cigarettes. For 2000, the two methods yield estimates of 0.11 g TEQDF-WHO98 (0.1 g I-
TEQ) and 0.67 g TEQDF-WHO98 (0.63 g I-TEQDF). For 1995, the two methods yield estimates of
0.2 g (on a TEQDF-WHO98 or I-TEQDF basis) and 1.48 g TEQDF-WHO98 (1.41 g I-TEQDF). For
5-38
-------
1987, the two methods yield estimates of 0.14 g TEQDF-WHO98 (0.12 g I-TEQDF basis) and 1.75
g TEQDF-WH098 (1.67 g I-TEQDF).
For purposes of this report, the best estimates of annual emissions are assumed to be the
average of the annual emissions estimated by the two methods for 2000, 1995, and 1987 (0.4 g,
0.8 g, and 1 g TEQDF-WHO98 or I-TEQDF, respectively). These emissions are assigned a low
confidence rating because the emission factor has a low confidence rating. Although these
emission quantities are relatively small when compared with the emission quantities estimated
for various industrial combustion source categories, they are significant because humans are
directly exposed to cigarette smoke.
5.6. PYROLYSIS OF BROMINATED FLAME RETARDANTS
The pyrolysis and photolysis of brominated phenolic derivatives and polybrominated
biphenyl ethers used as flame retardants in plastics (especially those used in electronic devices),
textiles, and paints can generate considerable amounts of polybrominated dibenzo-^-dioxins
(BDDs) and dibenzofurans (BDFs) (Watanabe and Tatsukawa, 1987; Thoma and Hutzinger,
1989; Luijk et al., 1992). Watanabe and Tatsukawa (1987) observed the formation of BDFs from
the photolysis of decabromobiphenyl ether. Approximately 20% of the decabromobiphenyl ether
was converted to BDFs in samples that were irradiated with ultraviolet light for 16 hr.
Thoma and Hutzinger (1989) observed the formation of BDFs during combustion
experiments with polybutylene-terephthalate polymers containing 9 to 11% decabromodiphenyl
ether. Maximum formation of BDFs occurred at 400 to 600°C, with a BDF yield of 16%.
Although the authors did not provide specific quantitative results for similar experiments
conducted with octabromodiphenyl ether and l,2-bis(tri-bromophenoxy)ethane, they did report
that BDDs and BDFs were formed.
Luijk et al. (1992) studied the formation of BDDs/BDFs during the compounding and
extrusion of decabromodiphenyl ether into high-impact polystyrene polymer at 275°C. Hepta-
and octa-BDF were formed during repeated extrusion cycles, and the yield of BDFs increased as
a function of the number of extrusion cycles. HpBDF increased from 1.5 to 9 ppm (in the
polymer matrix), and OBDF increased from 4.5 to 45 ppm after four extrusion cycles.
Insufficient data are available at this time from which to derive annual BDD/BDF
emission estimates for this source.
5.7. CARBON REACTIVATION FURNACES
Granular activated carbon (GAC) is an adsorbent that is widely used to remove organic
pollutants from wastewater and to treat finished drinking water at water treatment plants.
Activated carbon is manufactured from the pyrolytic treatment of nut shells and coal (Buonicore,
5-39
-------
1992a). The properties of GAC make it ideal for adsorbing and controlling vaporous organic and
inorganic chemicals entrained in combustion plasmas as well as soluble organic contaminants in
industrial effluents and drinking water. The high ratio of surface area to particle weight
(600:1,600 m2/g), combined with the extremely small pore diameter of the particles (15 to 25
angstroms), increases the adsorption characteristics (Buonicore, 1992a). GAC eventually
becomes saturated, and the adsorption properties significantly degrade. When saturation occurs,
GAC usually must be discarded and replaced, which significantly increases the costs of pollution
control.
The introduction of carbon reactivation furnace technology in the mid-1980s created a
method involving the thermal treatment of used GAC to thermolytically desorb the synthetic
compounds and restore the adsorption properties for reuse (Lykins et al., 1987). Large-scale
regeneration operations, such as those used in industrial water treatment operations, typically use
multiple-hearth furnaces. For smaller-scale operations, such as those used in municipal water
treatment operations, fluidized-bed and infrared furnaces are used. Emissions are typically
controlled by afterburners followed by water scrubbers (U.S. EPA, 1997a).
The used GAC can contain compounds that are precursors to the formation of
CDDs/CDFs during the thermal treatment process. A study by EPA (U.S. EPA, 1987a)
measured precursor compounds in spent GAC that was used as a feed material to a carbon
reactivation furnace tested during the National Dioxin Study. The total chlorobenzene content of
the GAC ranged from 150 to 6,630 ppb. Trichlorobenzene was the most prevalent species
present, with smaller quantities of di- and tetra-chlorobenzenes detected. Total halogenated
organics were measured to be about 150 ppm.
EPA has stack-tested two GAC reactivation furnaces for the emission of dioxin (U.S.
EPA, 1987a; Lykins et al., 1987). One facility was an industrial carbon reactivation plant, and
the second facility was used to restore GAC at a municipal drinking water plant. EPA (U.S.
EPA, 1997a) reported results of other testing performed at a county water facility in California
during 1990.
The industrial carbon reactivation plant processed 36,000 kg/day of spent GAC used in
the treatment of industrial wastewater effluents. This facility was chosen for testing because it
was considered to be representative of other facilities in the source category (U.S. EPA, 1987a).
Spent carbon was reactivated in a multiple-hearth furnace, cooled in a water quench, and shipped
back to primary chemical manufacturing facilities for reuse. The furnace was fired by natural gas
and consisted of seven hearths arranged vertically in series. The hearth temperatures ranged from
480 to 1,000°C. Air pollutant emissions were controlled by an afterburner, a sodium spray
cooler, and an FF. Temperatures in the afterburner were about 930°C. The estimated I-TEQDF
emission factor (treating nondetect values as zero) was 0.76 ng TEQDF-WHO98/kg (0.64 ng
5-40
-------
I-TEQDF/kg) carbon processed. The emission factor for total CDDs/CDFs was 58.6 ng/kg.
Because analyses were performed only for 2,3,7,8-TCDD; 2,3,7,8-TCDF; OCDD; and OCDF
and the congener groups, equivalent concentrations were assumed for all toxic and nontoxic
congeners in each of the penta, hexa, and hepta congener groups.
The second GAC reactivation facility tested by EPA consisted of a fluidized-bed furnace
located at a municipal drinking water treatment plant (Lykins et al., 1987). The furnace was
divided into three sections: a combustion chamber, a reactivation section, and a dryer section.
The combustion chamber was fired by natural gas and consisted of a stoichiometrically balanced
stream of fuel and oxygen. Combustion temperatures were about 1,038°C. Gases from the
reactivation section and combustion chamber were directed through an acid gas scrubber and
high-temperature afterburner prior to discharge from a stack. Although measurable
concentrations of dioxin-like compounds were detected in the stack emissions, measurements of
the individual CDD/CDF congeners were not performed; therefore, it was not possible to derive
TEQ emission factors for this facility. With the afterburner operating, no CDD congeners below
HpCDD were detected in the stack emissions. Concentrations of HpCDDs and OCDD ranged
from 0.001 to 0.05 ppt/volume basis (ppt/v) and 0.006 to 0.28 ppt/v, respectively. All CDF
congener groups were detected in the stack emissions even with the afterburner operating. Total
CDFs emitted from the stack averaged 0.023 ppt/v.
EPA (U.S. EPA, 1997a) reported a TEQ emission factor of 1.73 ng I-TEQDF/kg of carbon
processed for the reactivation unit at a county water facility in California in 1990. The emission
factor for total CDDs/CDFs was reported to be 47 ng/kg (i.e., similar to the total CDD/CDF
emission factor of 58.6 ng/kg at the industrial GAC facility). Because congener-specific results
were not reported, it was not possible to calculate the TEQDF-WHO98 emission factor. The report
also did not provide the configuration and type of furnace tested; however, it did state that the
emissions from the furnace were controlled by an afterburner and a scrubber.
The industrial GAC reaction furnace test data indicate that an average of 0.64 ng
I-TEQDF/kg of GAC may be released. The I-TEQDF emission rate for the reactivation unit at the
county water treatment facility was 1.73 ng I-TEQDF/kg carbon. Low confidence ratings are
given to these emission factors because only two GAC reactivation furnaces were stack tested,
and not all congeners were analyzed at the industrial GAC facility.
The mass of GAC that is reactivated annually in carbon reactivation furnaces is not
known. However, a rough estimate, to which a low confidence rating is assigned, is the mass of
virgin GAC shipped each year by GAC manufacturers. According to the U.S. Department of
Commerce (U.S. DOC, 1990c), 48,000 metric tons of GAC were shipped in 1987. EPA reported
that in 1990, water and wastewater treatment operations consumed 65,000 metric tons of GAC
5-41
-------
(U.S. EPA, 1995b, 1997a). The 1990 activity level was used in this document as a surrogate for
the 1995 and 2000 activity levels.
Applying the average TEQ emission factor of 1.2 ng (TEQDF-WHO98 or I-TEQDF) per kg
of reactivated carbon for the two tested facilities to the estimates of potential GAC reactivation
volumes yields annual release estimates of 0.06 g (TEQDF-WHO98 or I-TEQDF) in 1987 and
0.08 g (TEQDF-WHO98 or I-TEQDF) in 1995 and 2000 (assuming that the activity level for 1990 is
representative of the 1995 and 2000 activity levels). These emission estimates are assigned a low
confidence rating because both the activity and emission factor estimates had low confidence
ratings.
5.8. KRAFT BLACK LIQUOR RECOVERY BOILERS
Kraft black liquor recovery boilers are associated with the production of pulp in the
making of paper using the Kraft process. In this process, wood chips are cooked in large vertical
vessels called digesters at elevated temperatures and pressures in an aqueous solution of sodium
hydroxide and sodium sulfide. Wood is broken down into two phases: a soluble phase
containing primarily lignin, and an insoluble phase containing the pulp. The spent liquor (called
black liquor) from the digester contains sodium sulfate and sodium sulfide, which the industry
recovers for reuse in the Kraft process.
In the recovery of black liquor chemicals, weak black liquor is first concentrated in
multiple-effect evaporators to about 65% solids. The concentrated black liquor also contains 0.5
to 4% chlorides by weight, which are recovered through combustion. The concentrated black
liquor is sprayed into a Kraft black liquor recovery furnace equipped with a heat recovery boiler.
The bulk of the inorganic molten smelt that forms in the bottom of the furnace contains sodium
carbonate and sodium sulfide in a ratio of about 3:1. The combustion gas is usually passed
through an ESP that collects PM prior to being vented out the stack. The PM can be processed to
further recover and recycle sodium sulfate (Someshwar and Pinkerton, 1992).
In 1987, EPA stack-tested three Kraft black liquor recovery boilers for the emission of
dioxin in conjunction with the National Dioxin Study (U.S. EPA, 1987a). The three sites tested
by EPA were judged to be typical of Kraft black liquor recovery boilers at that time. During
pretest surveys, two facilities were judged to have average potential for CDD/CDF emissions and
one was judged to have high potential, based on the amount of chlorine found in the feed to these
units. Dry-bottom ESPs controlled emissions from two of the boilers; a wet-bottom ESP
controlled emissions from the third. The results of these tests included congener group
concentrations but lacked measurement results for specific congeners other than 2,3,7,8-TCDD
and 2,3,7,8-TCDF.
5-42
-------
NCASI (1995) provided congener-specific emission test results for six additional boilers
tested during 1990 to 1993. Three boilers were of the direct contact type, and three were
noncontact type. All were equipped with ESPs. The average congener and congener group
emission factors are presented in Table 5-13 for the three facilities reported by EPA (U.S. EPA,
1987a) and the six facilities reported by NCASI (1995). Figure 5-8 presents the average
congener and congener group profiles based on the test results presented by NCASI (1995).
The average TEQ emission factor, based on the data for the six NCASI facilities with
complete congener data, was 0.028 ng TEQDF-WHO98/kg (0.029 ng I-TEQDF/kg) of black liquor
solids, assuming nondetect values were zero, and 0.078 ng TEQDF-WHO98/kg (0.068 ng I-
TEQDF/kg), assuming nondetect values were present at one-half the DL. This value is assumed to
apply to all three reference years (1987, 1995, and 2000). The results for the three facilities
reported by EPA were not used in the derivation of the TEQ emission factor because congener-
specific measurements for most 2,3,7,8-substituted congeners were not made in the study (U.S.
EPA, 1987a). A medium confidence rating is assigned to those emission factors because they
were derived from the stack testing of six Kraft black liquor recovery boilers that were judged to
be fairly representative of technologies used at Kraft pulp mills in the United States.
A 1995 survey of the industry indicated that 215 black liquor recovery boilers were in
operation at U.S. pulp and paper mills. All but one of these boilers used ESPs for control of
particulate emissions; the one unique facility used dual scrubbers. In addition, ESPs were
reported to have been the predominant means of particulate control at recovery boilers for the
past 20 years (letter dated October 8, 1998, from W. Gillespie, National Council of the Paper
Industry for Air and Stream Improvement, Inc., to G. Schweer, Versar, Inc.).
The amounts of black liquor solids burned in Kraft black liquor recovery boilers in the
United States during 1987 and 1995 were 69.8 million metric tons and 80.8 million metric tons,
respectively (American Paper Institute, 1992; American Forest and Paper Association, 1997).
These activity level estimates are assigned a high confidence rating because they are based on
comprehensive industry survey data. Combining the emission factors derived above with the
activity level estimates for 1987 and 1995 yields estimated annual emissions from this source of
approximately 2 g (TEQDF-WHO98 or I-TEQDF) in 1987 and 2.3 g (TEQDF-WHO98 or I-TEQDF) in
1995. These emission estimates are assigned a medium confidence rating because the emission
factor have a medium confidence rating.
For 2000, NCASI provided estimates of activity levels for Kraft recovery furnaces and
Kraft lime kilns and CDD/CDF releases, including emissions from 11 Kraft recovery furnaces
and four Kraft lime kilns (Gillespie, 2002). The activity levels were reported to be 90.7 million
5-43
-------
Table 5-13. CDD/CDF mean emission factors (ng/kg feed) for black liquor
recovery boilers
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total I-TEQDF
Total TEQDF-WHO98
Total CDD/CDF
U.S. EPA (1987a)
(3 facilities)
Nondetect set to
zero
0
NR
NR
NR
NR
NR
4.24
0.04
NR
NR
NR
NR
NR
NR
NR
NR
0.35
0.21
0.27
0.8
2.05
4.24
0.95
0.64
1.16
1.05
0.35
0.10a
0.10a
11.72
Nondetect set to Vz
detection limit
0.04
NR
NR
NR
NR
NR
4.24
0.06
NR
NR
NR
NR
NR
NR
NR
NR
0.35
0.36
0.35
1.02
2.05
4.24
1
0.77
1.2
1.05
0.35
0.15a
0.16a
12.39
NCASI (1995)
(6 facilities)
Nondetect set to
zero
0
0
0.001
0.003
0.006
0.108
1.033
0.04
0.03
0.033
0.007
0.012
0.005
0.01
0.024
0
0.113
0.106
0.013
0.104
0.252
1.033
1.27
0.37
0.102
0.024
0.113
0.029
0.028
3.39
Nondetect set to Vz
detection limit
0.016
0.016
0.018
0.015
0.019
0.135
1.054
0.049
0.036
0.037
0.022
0.021
0.016
0.021
0.035
0.014
0.13
0.123
0.059
0.122
0.279
1.054
1.275
0.376
0.109
0.038
0.13
0.065
0.072
3.57
"Estimate based on the measured data for 2,3,7,8-TCDD; 2,3,7,8-TCDF; OCDD; and OCDF and congener group
emissions (i.e., for the penta-, hexa-, and hepta-CDD and CDFs, it was assumed that the measured emission factor
within a congener group was the sum of equal emission factors for all congeners in that group, including non-
2,3,7,8-substituted congeners).
NR = Not reported
metric tons for Kraft recovery furnaces and 13 million metric tons for Kraft lime kilns. These
activity level estimates are assigned a high confidence rating because they are based on
comprehensive industry survey data. Emission factors were taken from "NCASI Handbook of
Chemical Specific Information for SARA Section 313 Form R Reporting." The factors provided
5-44
-------
) I
II. I ^ II.
Ratio (mean congener group emission factor / total CDD/CDF emission factor)
0' 0.1 0.2 0.3 0.4 0.5
Figure 5-8. Congener and congener group profiles for air emissions from
Kraft black liquor recovery boilers (nondetect set equal to zero).
Source: NCASI (1995).
5-45
-------
in this handbook were compiled from valid test data supplied to NCASI by a variety of sources,
including NCASI member companies who had performed the tests in response to a regulatory
program. They are assigned a high confidence rating because they are based on a comprehensive
survey of stack emissions. Congener-specific CDD/CDF TEQ emission factors were provided
for both source categories (Table 5-14). Using the congener-specific emission factors and the
activity levels provided above, NCASI estimated CDD/CDF TEQDF-WHO98 emissions for each
congener (Table 5-14) and reported total emissions of 0.75 g TEQDF-WHO98/yr and 6.9e-5 g
TEQDF-WHO98/yr for Kraft recovery furnaces and Kraft lime kilns, respectively. This 2000
emissions estimate has a high confidence rating because both the emission factor and activity
level are rated as high confidence.
5.9. OTHER IDENTIFIED SOURCES
Several manufacturing processes are identified as potential sources of CDD/CDF
formation because the processes use chlorine-containing components or involve application of
high temperatures. However, no testing of emissions from these processes has been performed in
the United States, and only minimal emission rate information has been reported for these
processes in other countries. Therefore, these sources are rated as Category E sources, meaning
their emissions cannot be quantified.
Burning of candles. Schwind et al. (1995) analyzed the wicks and waxes of uncolored
candles as well as the fumes of burning candles for CDDs/CDFs, total chlorophenol, and total
chlorobenzene content. The results, presented in Table 5-15, show that beeswax contained the
highest levels of CDDs/CDFs and total chlorophenols. In contrast, the concentration of total
chlorobenzenes in stearin wax was higher than that in paraffin or beeswax by a factor of 2 to 3.
The concentrations of the three analyte groups were significantly lower in the wicks than in the
waxes. Emissions of CDDs/CDFs from all three types of candles were very low during burning.
In fact, comparison of the emission factor with the original CDD/CDF concentrations in the wax
indicates a net destruction of the CDDs/CDFs originally present in the wax. Information on the
activity level is lacking, therefore, no estimate of environmental release can be made at this time.
Glass manufacturing. Annual emissions of less than 1 g I-TEQDF/yr have been
estimated for glass manufacturing facilities in the Netherlands (Bremmer et al., 1994) and the
United Kingdom (Douben et al., 1995). Glass is manufactured by heating a mixture of sand and,
depending on the type of glass, lime, sodium carbonate, dolomite, clay, or feldspar to a
temperature of 1,400 to 1,650°C. In addition, various coloring and clarifying agents may be
added. Chlorine enters the process as a contaminant (NaCl) in sodium carbonate (Bremmer et al.
1994). However, the emission factors used by Bremmer et al. and Douben et al. were not
reported. Umweltbundesamt (1996) reported relatively low emission factors (approximately
0.002 and 0.007 ng I-TEQDF/kg) for two glass manufacturing facilities in Germany.
5-46
-------
Table 5-14. CDD/CDF TEQ emission factors and emission estimates from Kraft
recovery furnaces and Kraft lime kilns
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Kraft recovery furnaces
TEQDF-WH098
(ng/lb BLS)
O.OOe+00
O.OOe+00
O.OOe+00
2.00e-04
5.00e-04
4.90e-04
1.42e-04
5.00e-04
l.OOe-04
1.50e-03
4.00e-04
2.00e-04
O.OOe+00
4.00e-04
6.00e-05
O.OOe+00
2.60e-05
Emissions
(ng/yr)
O.OOe+00
O.OOe+00
O.OOe+00
3.33e+07
8.31e+07
8.15e+07
2.36e+07
8.31e+07
1.66e+07
2.49e+08
6.65e+07
3.33e+07
O.OOe+00
6.65e+07
9.98e+06
O.OOe+00
4.32e+06
Kraft lime kilns
TEQDF-WH098
(ng/lb CaO)
O.OOe+00
O.OOe+00
O.OOe+00
l.OOe-04
O.OOe+00
2.80e-04
2.56e-04
8.00e-04
l.OOe-04
O.OOe+00
9.00e-04
2.00e-04
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
Emissions
(ng/yr)
O.OOe+00
O.OOe+00
O.OOe+00
2.60e+03
O.OOe+00
7.27e+03
6.65e+03
2.08e+04
2.60e+03
O.OOe+00
2.34e+04
5.20e+03
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
BLS = Black liquor solids
CaO = Calcium oxide
Table 5-15. CDD/CDF concentrations in candle materials and emissions
Wax
material
Paraffin
Stearin
Beeswax
Paraffin
Stearin
Beeswax
Candle
component
Wax
Wax
Wax
Wick
Wick
Wick
Concentration
CDD/CDF
(ngI-TEQDF/kg)
0.59
1.62
10.99
0.18
0.12
0.08
Total
chlorophenols
(H
-------
Lime kilns. Annual emissions from lime kilns in Belgium and the United Kingdom have
been reported by Wevers and De Fre (1995) and Douben et al. (1995), respectively. However,
the emission factors used to generate those estimates were not provided. Umweltbundesamt
(1996) reported low emissions (0.016 to 0.028 ng I-TEQDF/kg) during tests at two lime kilns in
Germany.
Ceramics and rubber manufacturers. Douben et al. (1995) estimated annual emissions
from ceramic manufacturers and rubber manufacturers in the United Kingdom. Lexen et al.
(1993) had previously detected high concentrations of CDDs/CDFs in emissions from a ceramic
manufacturer in Sweden that occasionally glazed ceramics by volatilization of sodium chloride in
a coal-fired oven. Lexen et al. (1993) also detected high pg/L levels of I-TEQDF in the scrubber
water from the vulcanization process at a Swedish rubber manufacturer.
5-48
-------
6. COMBUSTION SOURCES OF CDDs/CDFs: MINIMALLY CONTROLLED
AND UNCONTROLLED COMBUSTION SOURCES
This chapter discusses combustion sources of CDDs/CDFs that have some (in the case of
combustion of landfill gas) or no post-combustion pollution control equipment for conventional
pollutant emissions.
6.1. COMBUSTION OF LANDFILL GAS
6.1.1. Emissions Data
Although no data could be located on levels in untreated landfill gas, several studies have
reported detecting CDDs/CDFs in the emissions resulting from the combustion of landfill gas.
Only one study of CDD/CDF emissions from a landfill flare has been reported for a U.S. landfill
(CARB, 1990c). The TEQDF-WHO98 and I-TEQDF emission factor calculated from the results of
this study is approximately 2.4 ng TEQ/m3 landfill gas combusted. The congener-specific results
of this study are presented in Table 6-1. Figure 6-1 presents the CDD/CDF congener emission
profile based on these emission factors. Bremmer et al. (1994) reported a lower emission factor,
0.4 ng I-TEQDF/m3, from the incineration of untreated landfill gas in a flare at a facility located in
the Netherlands. No congener-specific emission factors were provided. The average TEQ
emission factor for the CARB and Bremmer et al. studies is 1.4 ng I-TEQDF/m3 landfill gas
combusted.
Umweltbundesamt (1996) reported even lower TEQ emission factors for landfill gas
burned in engines or boiler mufflers rather than in a flare. The reported results for 30 engines
and mufflers tested in Germany ranged from 0.001 to 0.28 ng I-TEQDF/m3, with most values
below 0.1 ng I-TEQDF/m3. However, Bremmer et al. (1994) reported an emission factor of 0.5 ng
I-TEQDF/m3 from a landfill gas-fired engine in the Netherlands.
6.1.2. Activity Level Information
In 1996 EPA promulgated emission standards and guidelines to control emissions of
landfill gas from existing and future landfills under the Clean Air Act (Federal Register, 1996b).
Those regulations require the largest landfills in the United States (on the basis of design
capacity) to periodically measure and determine their annual emissions of landfill gas. Landfills
that emit more than 50 metric tons of nonmethane organic compounds (NMOC) annually must
collect landfill gas and reduce its NMOC content by 98% weight through the use of a control
device.
6-1
-------
Table 6-1. CDD/CDF emission factors for a landfill flare
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean facility emission factor"
(ng/m3 gas combusted)
0.02
0.09
0.07
0.07
0.26
0.76
4.41
14.07
0.39
1.14
1.46
0.42
0.11
0.68
1.22
0.07
0.64
5.68
20.20
2.39
2.43
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
aAssumes heat content of 1.86e+07 J/m for landfill gas (Federal Register, 1996a).
NR = Not reported
Source: CARB (1990c).
6-2
-------
Ratio (congener emission factor/total 2378-CDD/CDF emission factor)
].2A4.7.8-II\CI)I)
1.2A(!.7.8-M\U)I)
1.2.x7.X/)-|]\U)l)
.2.3.4.6.7.8-1 IpCDI)
.2.3.4.6.7.X.C>-OC'I)I)
:.3.7.8-TCI)r
l.2.3.7.8-Pen)l;
2.v4.7.8-PcU)l;
1.2. 3.4.7.8-1 I\U)];
l.2.3.6.7.8-II\U)r
2.3.4.6.7.8-1 I\U)I-'
.4. 6.7.8. ')-()( 4 )!•'
0.1
0.2
0.3
0.4
0.5
0.6
Figure 6-1. Congener profile for landfill flare air emissions.
Source: CARB (1990c).
6.1.2.1. Activity Levels for 1987 and 1995
EPA estimated that when the 1996 regulations were implemented, the controls would
reduce annual NMOC emissions from existing landfills by 77,600 metric tons. The cost analysis
supporting this rulemaking based control device costs on open flares because flares are
applicable to all the regulated facilities. Assuming that the mass reduction would be achieved by
the use of flares, the corresponding volume of landfill gas burned would be approximately 14
billion m3/yr. The calculation was based on an assumed default NMOC concentration in landfill
gas of 1,532 ppmv and a conversion factor of 3.545 mg/m3 NMOC per 1 ppmv NMOC (Federal
Register, 1993b).
6-3
-------
Of the approximately 312 landfills that were affected by the promulgation of the emission
standards and guidelines in 1996, EPA estimated that more than 100 had some form of collection
or control system (or both) in place in 1991 (Federal Register, 1991c). Thus, a rough
approximation of the volume of landfill gas combusted in 1995 was 4.7 billion m3/yr (or 33% of
the future expected 14 billion m3/yr reduction). This estimate is similar to the 2 to 4 billion m3
landfill gas estimated by the Energy Information Administration (EIA, 1994) as collected and
consumed for energy recovery purposes in 1992.
EIA (1992) estimated that between 0.9 and 1.8 billion m3 of landfill gas were collected
and burned in 1990 for energy recovery purposes. Because there were no specific data available
for 1987, EPA assumed that the mean of this range, 1.35 billion m3, would serve as an
approximate estimate of the volume of landfill gas combusted in 1987.
6.1.2.2. Activity Level for 2000
According to the EPA 2001 inventory of greenhouse gas emissions, approximately 7.7
billion m3 of landfill gas were combusted in 2000 through 477 landfill flares (average of 16.5
million m3 of landfill gas per flare). As of 2003, there were more than 1,000 landfill flares in the
United States (e-mails dated February 28, 2003, and March 7, 2003, from B. Guzzone, U.S. EPA,
to K. Riley, Versar). Assuming that the amount of landfill gas combusted through the 477
landfill flares inventoried is representative of the landfill gas combusted through these 1,000+
flares, approximately 16 billion m3 of landfill gas were combusted in the United States through
flares in 2000.
6.1.3. Emission Estimates
The limited emission factor data that are available were judged inadequate for developing
national emission estimates that could be included in the national inventory. However, a
preliminary estimate of the potential annual TEQ releases from landfills can be obtained using
the estimated volume of combusted gas and the available emission factors. Combining the
estimates of landfill gas volume that was combusted (1.35 billion m3 in 1987, 4.7 billion m3 in
1995, and 16 billion m3 in 2000) with the emission factor of 1.4 ng I-TEQDF/m3 of flare-
combusted gas yields annual emission estimates of 1.9, 6.6, and 22 g I-TEQDF/m3 for 1987, 1995,
and 2000, respectively. These estimates should be regarded as preliminary indications of
possible emissions from this source; further testing is needed to confirm the true magnitude of
those emissions.
6-4
-------
6.2. ACCIDENTAL FIRES
Accidental fires in buildings and vehicles are uncontrolled combustion processes that,
because of poor combustion conditions, typically result in relatively high emissions of
incomplete combustion products (Bremmer et al., 1994), which can include CDDs and CDFs.
Polyvinyl chloride (PVC) building materials and furnishings, chloroparaffin-containing textiles
and paints, and other chlorinated organic compound-containing materials appear to be the
primary sources of the chlorine (Retard, 1993). Although the results of several studies have
demonstrated the presence of CDD/CDF concentrations in soot deposits and residual ash from
such fires, few direct measurements of CDDs/CDFs in the fumes or smoke of fires have been
reported. The results of some of those studies are described below, and an evaluation of the
available data follows.
6.2.1. Soot and Ash Studies
Christmann et al. (1989a) analyzed the soot formed during combustion and pyrolysis of
pure PVC and PVC cable sheathings in simple laboratory experiments designed to mimic the
conditions of fires. For the combustion experiments, 2 g of a PVC sample were incinerated with
a laboratory gas burner. The combustion products were collected on the inner walls of a cooled
gas funnel placed above the sample. For the pyrolysis experiments, about 50 mg of the sample
were placed in a quartz tube and heated to about 950°C for 10 min in either an air atmosphere or
a nitrogen atmosphere. The combustion experiments yielded CDD/CDF concentrations in soot
of 110 |ig I-TEQDF/kg for a low-molecular-weight PVC, 450 |ig I-TEQDF/kg for a high-
molecular-weight PVC, and 270 jig I-TEQDF/kg for PVC cable. The pyrolysis experiments in the
air atmosphere yielded lower CDD/CDF concentrations in soot: 24.4 jig I-TEQDF/kg for a low-
molecular-weight PVC, 18.7 jig I-TEQDF/kg for a high-molecular-weight PVC, and up to 41 jig I-
TEQDF/kg for PVC cable.
In general, more CDFs than CDDs were formed. The lower-chlorinated CDF congeners
were dominant in the combustion experiments; however, the HpCDF and OCDF congeners were
dominant in the pyrolysis experiments. No CDDs/CDFs were detected in pyrolysis experiments
under a nitrogen atmosphere. Also, no CDDs/CDFs were detected when chlorine-free
polyethylene samples were subjected to the same combustion and pyrolysis conditions.
Deutsch and Goldfarb (1988) reported finding CDD/CDF concentrations ranging from
0.04 to 6.6 |ig/kg in soot samples collected after a 1986 fire in a State University of New York
lecture hall. The fire consumed or melted plastic furnishings and cleaning products containing
chlorine, wood, and paper.
In a study that analyzed 200 ash and soot samples from sites of accidental fires in which
PVC was involved (Funcke et al., 1988, as reported in Bremmer et al., 1994; Retard, 1993),
6-5
-------
CDDs/CDFs were detected in more than 90% of the samples at concentrations in the ng I-
TEQDF/kg to jig I-TEQDF/kg range. Fires involving the combustion of materials containing
relatively large amounts of PVC and other chlorinated organic substances resulted in the highest
levels of CDDs/CDFs, with concentrations ranging from 0.2 to 110 jig I-TEQDF/kg residue.
Thiesen et al. (1989) analyzed residues from surfaces of PVC-containing materials that
were partially burned during accidental fires at sites in Germany that manufactured or stored
plastics. CDD/CDF concentrations in residues were reported as 0.5 jig I-TEQDF/kg for soft PVC,
4.6 |ig I-TEQDF/kg for PVC fibers, and 28.3 |ig I-TEQDF/kg for a hard PVC. The ratio of total
CDFs to total CDDs in the three samples ranged from 4:1 to 7:1. The dominant 2,3,7,8-
substituted CDF and CDD congeners in all three samples were 1,2,3,4,6,7,8-HpCDF and
1,2,3,4,6,7,8-HpCDD.
In an accidental fire at a Swedish carpet factory in 1987, 200 metric tons of PVC and 500
metric tons of PVC-containing carpet were burned. Marklund et al. (1989) analyzed snow
samples up to 1,500 m downwind from the fire site and found CDD/CDF concentrations in the
top 2 cm ranging from 0.32 jig I-TEQDF/m2 at 10 m from the site to 0.01 jig I-TEQDF/m2 at 1,500
m. Because of an atmospheric inversion and very light wind at the time of the fire, the smoke
from the fire remained close to the ground. The soot deposited onto the snow was thus assumed
to be representative of the soot generated and released from the fire. Wipe samples of soot from
interior posts of the plant that were 5 and 20 m from the fire contained Eadon TEQ
concentrations of 0.18 and 0.05 |ig/m2, respectively. On the basis of these deposition
measurements, the investigators estimated total CDD/CDF emissions from the fire to be less than
3mgI-TEQDF.
Carroll (1996) estimated a soot-associated CDD/CDF emission factor for the Swedish
carpet factory fire (i.e., not including volatile emissions) of 28 to 138 ng I-TEQDF/kg PVC burned
using the following assumptions: the PVC carpet backing was one-half the weight of the carpet,
the carpet backing contained 30% by weight PVC resin, and 20 to 100% of the PVC and PVC
carpet backing present in the warehouse actually burned. Using the results of wipe samples
collected at downwind distances of up to 6,300 m, Carroll (1996) also estimated a similar soot-
associated emission factor (48 to 240 ng I-TEQDF/kg of PVC burned) for a fire at a plastics
recycling facility in Lengerich, Germany.
Fiedler et al. (1993) presented a case study of CDD/CDF contamination and associated
remedial actions taken at a kindergarten in Germany following a fire that destroyed parts of the
roof, windows, and furnishings. Soot collected from the building contained CDDs/CDFs at a
concentration of 45 jig I-TEQDF/kg (15 jig I-TEQDF/m2). The study authors attributed the
CDDs/CDFs detected to the combustion of plastic and wooden toys, floors, and furnishings;
however, no information was provided on the quantities of those materials.
6-6
-------
Fiedler and Lindert (1998) presented results of soot sampling following a serious fire at
the airport in Diisseldorf, Germany. Polystyrene sheets and PVC-coated cables were involved in
the fire, together with PCB-containing condensers (bulbs). Surface wipe samples contained up to
0.33 jig I-TEQDF/m2. Concentrations in soot ranged from 7 to 130 jig I-TEQDF/kg.
Concentrations of polybrominated dibenzo-^-dioxins and dibenzofurans were detected in soot at
concentrations as high as 0.9 mg/kg soot.
Wichmann et al. (1993, 1995) measured the CDD/CDF content of ash and debris and
deposited surface residues that resulted from experimental test burns of two cars (a 1974 Ford
Taurus [old car] and a 1988 Renault Espace [new car]), one subway car, and one railway coach in
a tunnel in Germany. On the basis of measurements obtained from the sampled ash and debris
and from soot collectors placed at regular intervals up to 420 m downwind of the burn site, the
total estimated amount of CDDs/CDFs in the ash/debris and tunnel surface residues from each
vehicle burn experiment was 0.044 mg I-TEQDF for the 1974 model car, 0.052 mg I-TEQDF for
the 1988 model car, 2.6 mg I-TEQDF for the subway car, and 10.3 mg I-TEQDF for the railway
coach. For each vehicle burn experiment, the mass of TEQ in tunnel surface residue exceeded
the mass in ash and debris: 73 to 89% were accounted for by the tunnel surface residues and 11
to 27% by ash and debris. The average CDD/CDF content of the ash and debris from each
experimental burn was: 0.14 jig I-TEQDF/kg for the new car, 0.3 jig I-TEQDF/kg for the old car,
3.1 jig I-TEQDF/kg for the subway car, and 5.1 jig I-TEQDF/kg for the railway coach.
6.2.2. Fume and Smoke Studies
Merk et al. (1995) collected fume and smoke generated during the burning of 400 kg of
wood and 40 kg of PVC in a building (4,500 m3 volume) over a 45-min period. The sampling
device consisted of dual glass fiber filters to collect particles greater than 0.5 jam followed by a
polyurethane foam filter to collect vapor-phase CDDs/CDFs. The paniculate phase (particles
greater than 0.5 jim diameter) and the gas phase showed the same congener pattern: decreasing
concentration with increasing degree of chlorination, thus indicating no preferential sorption of
higher-chlorinated congeners to smoke particulates. However, the CDDs/CDFs found in the gas
phase (about 5 ng I-TEQDF/m3) accounted for more than 90% of the detected CDDs/CDFs. The
authors also reported that the soot deposited from this fire onto aim2 aluminum sheet resulted in
surface contamination of 0.05 jig I-TEQDF/m2.
Although it was stated in Merk et al. (1995) that the building was "closed," subsequent
communication with one of the coauthors (telephone conversation on September 29, 1998,
between Karl-Werner Schramm, GSF-Institute of Ecological Chemistry, and Greg Schweer,
Versar, Inc.) clarified that a "gas cleaning" system was in operation. Because a ventilation
system was in operation, there was likely some loss of vapor-phase CDDs/CDFs from the hall;
6-7
-------
therefore, the deposits (from paniculate deposition and vapor-phase condensation) on the test
aluminum plate may not have reflected total CDD/CDF formation during the fire.
Dyke and Coleman (1995) reported a fourfold increase in CDD/CDF TEQ concentrations
in the ambient air during "bonfire night" (an annual event held on November 5 during which it is
customary to set off fireworks and have bonfires) in Oxford, England. Air concentrations before
and after bonfire night ranged from 0.15 to 0.17 pg I-TEQDF/m3. The air concentration during
bonfire night was 0.65 pg I-TEQDF/m3. The dominant congeners in all samples were the hepta-
and octa-CDDs. The study was not designed to collect data that would enable calculation of an
emission rate or to differentiate the relative importance of the various materials combusted.
However, the results do indicate that open burning of materials likely to be combusted in
accidental fires (with the exception of fireworks) results in the release of CDDs and CDFs.
6.2.3. Data Evaluation
6.2.3.1. Structural Fires
6.2.3.1.1. Emissions data. Only limited emissions data for structural fires were located. Most
of the studies obtained involved situations (field and laboratory) where relatively high loadings
of PVC or plastics were combusted. The effects of different mixes of combusted materials,
oxygen supplies, building configurations, and durations of burn, for example, that are likely to
occur or be found in accidental fires cannot be accounted for by the factors that can be derived
from these studies. Also, most of the studies addressed only soot or ash residues and did not
address potential volatile emissions of CDDs/CDFs, which, according to Merk et al. (1995), may
represent 90% of the CDDs/CDFs generated during the burning of PVC.
Two reports (Carroll, 1996; Thomas and Spiro, 1995) attempted to quantify CDD/CDF
emissions from U.S. structural fires, and Lorenz et al. (1996) estimated emissions from structural
fires in Germany.
Carroll (1996) estimated the total CDD/CDF content of soot and ash generated from the
358,000 residential fires in the United States in 1993 (as reported in U.S. DOC, 1995b). Detailed
estimates were developed of the PVC content of items in typical homes, including plumbing,
wiring, siding and windows, wallpaper, blinds and shades, and upholstery. The typical
percentage of PVC burned in household fires was assumed to be 9.5%. Extrapolating to all
358,000 one- to two-family unit fires yielded an annual mass of 2,470 metric tons PVC burned.
Carroll then developed TEQ emission factors from the results of Thiesen et al. (1989) and
Marklund et al. (1989). The estimated CDD/CDF content ranged from 0.47 to 22.8 g I-TEQDF,
with 0.07 to 8.6 g I-TEQDF in soot and 0.4 to 14.2 g I-TEQDF in ash. A soot emission factor (i.e.,
grams of soot produced per gram of PVC combusted) was derived from the investigator's
assumptions regarding the surface area of the soot collection funnel used by Christmann et al.
6-8
-------
(1989b) and the soot deposition rate on that funnel. These I-TEQDF emission factors were then
applied to the estimated 2,470 metric tons of PVC burned annually in one- to two-family unit
residential fires to obtain estimates of the annual mass of TEQ that would be found in the soot
and ash of residential fires (0.48 to 22.8 g I-TEQDF/yr). The average emission per fire is thus 1.3
to 64 |ig I-TEQDF.
Thomas and Spiro (1995) estimated that 20 g I-TEQDF may be released annually to air
from structural fires. This estimate assumes an emission factor of 4 ng I-TEQDF/kg material
combusted (i.e., the emission rate for "poorly" controlled wood combustion), a material
combustion factor of 6,800 kg per fire, and 688,000 structural fires per year. The average
emission per fire is thus 29 jig I-TEQDF.
Lorenz et al. (1996) estimated annual generation of CDDs/CDFs in Germany using data
on the number of residential and industrial/commercial structural fires coupled with data on
CDD/CDF content in soot and ash residues remaining after fires. The potential annual I-TEQDF
generation was estimated to be 78 to 212 g.
Using the emissions data estimated by Carroll (1996) and Thomas and Spiro (1995)
provides an average emission factor of 32 jig I-TEQ/fire.
6.2.3.1.2. Activity level information. In 1987, there were approximately 2,330,000 fires in the
Unites States, of which approximately 745,600 (32%) were structural fires (FEMA, 1997). In
1995, approximately 574,000 structural fires were reported in the United States. Of these,
426,000 were reported to be in residential structures, including 320,000 in one- to two-family
units, 94,000 in apartments, and 12,000 in other residential settings. The types of structures for
the remaining 148,000 fires were public assembly, 15,000; educational, 9,000; institutional,
9,000; stores and offices, 29,000; special structures, 29,000; storage, 39,000; and industry, utility,
and defense, 18,000. The latter two categories may be underreported, as some incidents were not
recorded because they were handled by private fire brigades or fixed suppression systems (U.S.
DOC, 1997). For 2000, the National Fire Data Center estimated that approximately 1,708,000
fires occurred in the United States, of which approximately 512,400 (30%) were structural fires
(FEMA, 2001).
6.2.3.1.3. Emission estimates. The limited data available on structural fires were judged
inadequate for developing national emission estimates. This conclusion was also reached for
national emission inventories developed for the Netherlands (Bremmer et al., 1994) and the
United Kingdom (U.K. Department of the Environment, 1995). However, preliminary estimates
were calculated by combining the average emission factor of 32 jig I-TEQ/fire and the number of
structural fires in the Unites States (745,600 in 1987; 426,000 in 1995; and 512,400 in 2000),
6-9
-------
yielding an annual release of 24 g I-TEQDF in 1987, 14 g I-TEQDF in 1995, and 16 g I-TEQDF in
2000. Confidence in these estimated emissions is very low because of the numerous assumptions
employed in their derivation. If the conclusion by Merk et al. (1995) is assumed to be correct,
that 90% of the CDDs/CDFs formed in fires are in the gaseous phase rather than particulate
phase, and it is also assumed that the estimates by Carroll (1996) and Thomas and Spiro (1995)
did not totally account for volatile emissions, then the total CDD/CDF emissions estimated by
Carroll and Thomas and Spiro may be underestimates. Further testing is needed to confirm the
true magnitude of these releases.
6.2.3.2. Vehicle Fires
As with structural fires, the limited data available on vehicle fires were judged inadequate
for developing national emission estimates that could be included in the national inventory.
However, a preliminary estimate of the range of potential CDD/CDF emissions that may result
from vehicle fires can be calculated using the results reported by Wichmann et al. (1993, 1995)
for controlled vehicle fires in a tunnel (0.044 mg I-TEQDF for an old car to 2.6 mg I-TEQDF for a
subway car). Although Wichmann et al. did not measure volatile CDDs/CDFs (which were
reported by Merk et al., 1995, to account for the majority of CDDs/CDFs formed during a fire),
the study was conducted in a tunnel, and it is likely that a significant fraction of the volatile
CDDs/CDFs sorbed to tunnel and collector surfaces and were thus measured as surface residues.
The number of vehicle fires reported in the United States was approximately 561,530 in
1987 (FEMA, 1997), 406,000 in 1995 (U.S. DOC, 1997), and 341,600 in 2000 (FEMA, 2001).
If it is assumed that 99% of those fires involved cars and trucks (the approximate percentage of
all U.S. motor vehicles that are in-service cars and trucks; U.S. DOC, 1995b) and that the
applicable emission rate is 0.044 mg I-TEQDF per incident, then the annual TEQ formation was
24.4 g I-TEQDF for 1987, 17.7 g I-TEQDF for 1995, and 14.9 g I-TEQDF for 2000. The emission
factor of 2.6 mg I-TEQDF/fire is assumed to be applicable to the remaining 1% of vehicle fires,
thus yielding emissions of 14.6 g I-TEQDF/yr for 1987, 10.6 g I-TEQDF/yr for 1995, and 8.9 g I-
TEQDF/yr for 2000. Total TEQ annual emissions for 1987, 1995, and 2000 were roughly
estimated to have been 39, 28.3, and 23.8 g I-TEQDF/yr, respectively. These estimates should be
regarded as preliminary indications of possible emissions from this source category; further
testing is needed to confirm the true magnitude of these emissions.
6-10
-------
6.3. LANDFILL FIRES
6.3.1. Emissions Data
In the late 1980s, two serious fires occurred in landfills near Stockholm, Sweden. The
first fire was in a large pile of refuse-derived fuel. Using measurements of chlorobenzenes in the
air emissions, it was estimated that 50 to 100 kg of chlorobenzenes were released. CDD/CDF
emissions were estimated to be several tens of grams, on the assumption that the ratio of
CDDs/CDFs to chlorobenzenes in landfill fire emissions is similar to the ratio observed in stack
gases of municipal waste combustors. To measure releases in connection with the second fire,
which occurred at a large conventional landfill, birch leaves were collected from trees close to
the fire and at distances up to 2 km downwind of the fire, as well as from nearby areas not
affected by smoke from the fire. The discharge of CDDs/CDFs necessary to cause the
concentrations measured on the leaves was estimated to be several tens of grams (Persson and
Bergstrom, 1991).
In response to these incidents, Persson and Bergstrom (1991) also measured CDD/CDF
emissions from experimental fires designed to simulate surface landfill fires and deep landfill
fires. The experiments used 9-month-old domestic waste. The tests showed no significant
difference in CDD/CDF content of the fire gases produced by the simulated surface and that of
the deep fires. The average CDD/CDF emission rate was reported to be 1 jig Nordic TEQ/kg
waste burned.
Persson and Bergstrom (1991) and Bergstrom and Bjorner (1992) estimated annual
CDD/CDF Nordic TEQ emissions in Sweden from landfill fires to be 35 g. The estimate was
based on the emission rate of 1 jig Nordic TEQ/kg waste burned, an assumed average density of
landfill waste of 700 kg/m3, an assumed waste burn of 150 m3 for each surface landfill fire (167
fires in Sweden per year), and an assumed waste burn of 500 m3 for each deep landfill fire (50
fires in Sweden per year). The estimates of waste burn mass for each type of fire were the
average values obtained from a survey of 62 surface fires and 25 deep fires. The estimated
number of fires per year was based on the results of a survey of all Swedish municipalities for
fires reported during 1988 and 1989. In 1991, Sweden had an estimated 400 municipal landfills
(Persson and Bergstrom, 1991).
Ruokojarvi et al. (1995) measured ambient air concentrations of CDDs/CDFs in the
vicinity of real and experimental landfill fires in Finland. The most abundant toxic congeners
were the hepta- and octa-CDDs and the penta-, hepta-, and octa-CDFs. The highest contributors
to the measured TEQ were 1,2,3,7,8-PeCDD and 2,3,4,7,8-PeCDF. In Finland, annual
CDD/CDF emissions from landfill fires are estimated to be 50 to 70 g Nordic TEQ (Aittola,
1993, as reported in Ruokojarvi et al., 1995).
6-11
-------
6.3.2. Activity Level Information and Emission Estimates
Although no U.S. monitoring studies are available, an emission factor similar to the
Swedish emission factor would be expected for the United States because the content of
municipal waste in the United States and Sweden are expected to be similar. Because no data
could be located on characterization of landfill fires in the United States (i.e., number, type, mass
of waste involved), the limited data available were judged inadequate for developing national
emission estimates that could be included in the national inventory. However, a preliminary
estimate of the potential magnitude of TEQ emissions associated with landfill fires in the United
States can be obtained by assuming a direct correlation of emissions to population size for the
United States and Sweden or by assuming a direct correlation between emissions and the number
of landfills in each country.
Both the United States and Sweden are industrialized countries. Although the per capita
waste generation rate in the United States is nearly 1.5 times that of Sweden, the composition of
municipal waste and the fraction of municipal waste disposed of in landfills in the two countries
are nearly identical (U.S. EPA, 1996d). The population of Sweden was 8,825,417 in 1995 (U.S.
DOC, 1995b) and 8,873,052 in 2000 (U.S. DOC, 2002). Based on these population estimates
and the estimated annual Nordic TEQ emission factor of 35 g, the per capita landfill
fire-associated Nordic TEQ emission factor was 4 jig TEQ per person per year for both 1995 and
2000. Because congener-specific results were not provided in Persson and Bergstrom (1991) or
Bergstrom and Bjorner (1992), it was not possible to derive emission factors in units of TEQDF-
WHO98 or I-TEQDF. Applying this factor to the U.S. population of 263,814,000 in 1995 (U.S.
DOC, 1995b) and 281,421,906 in 2000 (U.S. DOC, 2003) results in an estimated annual
emission of 1,050 g TEQ for 1995 and 1,126 g TEQ for 2000. These estimates should be
regarded as preliminary indications of possible emissions from this source category; further
testing is needed to confirm the true magnitude of these emissions.
6.4. FOREST AND BRUSH FIRES
6.4.1. Emissions Data
Because CDDs/CDFs have been detected both in the soot from residential wood burning
(Bumb et al., 1980; Nestrick and Lamparski, 1982, 1983; Bacher et al., 1992) and in the flue
gases from residential wood burning (Schatowitz et al., 1993; Vickelsoe et al., 1993; Launhardt
and Thoma, 2000; Environment Canada, 2000), it is reasonable to assume that wood burned in
forest and brush fires may also be a source of CDDs/CDFs (Section 4.2 contains details on these
studies).
Only one study (Tashiro et al., 1990) could be found that reported direct measurements of
CDDs/CDFs in emissions from forest fires. This study reported detection of total CDDs/CDFs in
6-12
-------
air at levels ranging from about 15 to 400 pg/m3. The samples were taken from fixed collectors
located 10m above the ground and from aircraft flying through the smoke. Background samples
collected before and after the tests indicated negligible levels in the atmosphere. These results
were presented in a preliminary report; however, no firm conclusions were drawn about whether
forest fires are a CDD/CDF source. The final report on this study, Clement and Tashiro (1991),
showed total CDD/CDF levels in the smoke of about 20 pg/m3. The authors concluded that
CDDs/CDFs are emitted during forest fires but recognized that some portion of these emissions
could represent resuspension from residues deposited on leaves rather than newly formed
CDDs/CDFs.
Although not designed to directly assess whether CDDs/CDFs are formed during brush
fires, Buckland et al. (1994) measured CDD/CDF levels in soil samples from both burnt and
unburnt areas in national parks in New Zealand 6 weeks after large-scale brush fires. Four
surface soil cores (2 cm depth) were collected and composited from each of three burnt and three
unburnt areas. Survey results indicated that brush fires did not have a major impact on
CDD/CDF levels in soil. The I-TEQDF content in the soil sample composites was 3, 8.7, and 10
ng/kg for the three unburnt areas and 2.2, 3.1, and 36.8 ng/kg for the three burnt areas. Total
CDD/CDF content ranged from 1,050 to 7,700 ng/kg in the unburnt area soil samples and from
1,310 to 27,800 ng/kg in the burnt area soil samples. OCDD accounted for 94 to 97% of total
CDD/CDF content in all samples.
Similarly, a survey of controlled straw-field burning in the United Kingdom (Walsh et al.,
1994) indicated that the straw burning did not increase the CDD/CDF burden in the soil;
however, a change in congener distribution was observed. Soils from three fields were sampled
immediately before and after burning, along with ash from the fire. The mean I-TEQDF
concentrations in the preburn soil, postburn soil, and ash were 1.79, 1.72, and 1.81 ng/kg,
respectively. Concentrations of 2,3,7,8-TCDF were lower in the postburn soils than in the
preburn soils. Conversely, concentrations of OCDD were higher in the postburn soils, indicating
possible formation of OCDD during the combustion process.
Van Oostdam and Ward (1995) reported finding no detectable levels of 2,3,7,8-
substituted CDDs/CDFs in three soil samples and four ash samples following a forest fire in
British Columbia. The detection limits (DLs), on a congener-specific basis (unweighted for
TEQ), ranged from 1 to 2 ng/kg. Nondetect values were also reported for ashes at a slash and
burn site: the soil contained about 0.05 ng I-TEQDF/kg, whereas background soil contained about
0.02ngI-TEQDF/kg.
The concentrations presented by Clement and Tashiro (1991) cannot accurately be
converted to an emission factor because the corresponding rates of combustion gas production
6-13
-------
and wood consumption are not known. As a result, the following four alternative approaches
were considered for developing an emission factor.
Soot-based approach. This approach assumes that the levels of CDDs/CDFs in chimney
soot are representative of the CDDs/CDFs in emissions. The CDD/CDF emission factor is
calculated as the product of the CDD/CDF concentration in soot and the total particulate
emission factor. This calculation involves first assuming that the CDD/CDF levels measured in
chimney soot by Bacher et al. (1992) (720 ng I-TEQDF/kg) are representative of the CDD/CDF
concentrations of particles emitted during forest fires. Second, the total parti culate generation
factor must be estimated. Using primarily data for head fires, Ward et al. (1976) estimated the
national average particulate emission factor for wildfires as 150 Ib/ton biomass dry weight.
Ward et al. (1993) estimated the national average particulate emission factor for prescribed
burning to be 50 Ib/ton biomass dry weight. Combining the total particulate generation rates
with the I-TEQDF level in soot results in emission factor estimates of 54 ng of I-TEQDF and 18 ng
of I-TEQDF/kg of biomass burned in wildfires and prescribed burns, respectively. These
estimated factors are likely to be overestimates because the levels of CDDs/CDFs measured in
chimney soot by Bacher et al. (1992) may represent the accumulation and enrichment of
CDDs/CDFs measured in chimney soot over time, leading to much higher assumed levels than
the actual levels on emitted particles.
Carbon monoxide (CO) approach. CO is a general indicator of the efficiency of
combustion, and the emission factors of many emission products can be correlated with the CO
emission factor. Data from Schatowitz et al. (1993) for emissions during natural wood burning
in open stoves suggest an emission factor of 10 jig I-TEQDF/kg CO. Combining this factor with
the CO emission factor during forest fires (roughly 0.1 kg CO/kg of biomass [Ward et al., 1993])
yields an emission factor of 1,000 ng I-TEQDF/kg biomass. This factor is higher than the soot-
based factor discussed above, which is itself considered to be an overestimate. In addition,
although the formation kinetics of CDDs/CDFs during combustion are not well understood,
CDD/CDF emissions have not been shown to correlate well with CO emissions from other
combustion sources.
Wood stove approach. This approach assumes that the emission factor for residential
wood burning (using natural wood and open door, i.e., uncontrolled draft) applies to forest fires.
As discussed in Section 4.2.1, this approach suggests an emission factor of about 0.5 ng I-
TEQ/kg wood combusted. This value appears more reasonable than the factors suggested by the
soot and CO approaches because it is based on direct measurement of CDDs/CDFs from
combustion of wood rather than on indirect techniques. However, forest fire conditions differ
significantly from combustion conditions in wood stoves. For example, forest fire combustion
6-14
-------
does not occur in an enclosed chamber, and the biomass consumed in forest fires is usually green
and includes underbrush, leaves, and grass.
Forest fire simulation approach. This approach quantifies CDD/CDF emissions
through the combustion of forest biomass in a controlled-burn facility. Using this approach,
Gullet and Touati (2003) estimated CDD/CDF emissions through the testing of three biomass
samples collected from the Oregon coast near Seal Rock and from four biomass samples
collected from the North Carolina Piedmont region, approximately 200 km from the Atlantic
coast. The samples generally consisted of equal portions of live shoots (needles cut from tree
branches) and needle litter gathered from the forest floor. The Oregon samples were composed
of pine needles (Pinus contorta and Pinus monticold) and hemlock needles (Tyuga heterophylla);
the North Carolina samples were composed entirely of lobolly pine (Pinus taeda). The
combustion of these seven samples, piled approximately 10 cm high, took place on top of an
open, flat combustion platform. CDD/CDF emissions were measured using a Graseby PS-1
sampler and EPA's ambient TO-9 method.
As shown in Table 6-2, the overall average total TEQ emission factor for the seven
samples was 19.9 ng TEQDF-WHO98/kg (18.6 ng I-TEQDF), assuming nondetects were zero.
Separately, the average total TEQ emission factors for the three Oregon samples and the four
North Carolina samples were 15 ng and 25 ng TEQDF-WHO98/kg, respectively. Even though the
average TEQ emission factors for the Oregon and North Carolina runs were similar, CDF
congeners were dominant in the Oregon samples, whereas CDD congeners were dominant in the
North Carolina samples. Figure 6-2 shows the congener profile for the Oregon and North
Carolina samples combined.
To test an alternative CDD/CDF sampling method, CDD/CDF emissions from one of the
Oregon samples were also measured using a "Nomad" (a prototype portable sampler designed for
mobile, in-field sampling). The results from both sampling methods showed very similar
CDD/CDF TEQ values, total values, and ratio values. An additional Oregon sample was also
combusted to test influences of fuel configuration on emissions. In this experiment, the biomass
was placed in a metal barrel with air holes cut into the bottom. The results showed the highest
total TEQ emission value calculated in this study (47 ng TEQDF-WHO98/kg); however, this value
is similar to the next highest total TEQ value (46 ng TEQDF-WHO98/kg).
Because the waxy cuticle layer on pine needles has been demonstrated to absorb
lipophilic compounds from the atmosphere, Gullet and Touati (2003) also extracted a raw, as-
received Oregon biomass sample to determine whether the observed emissions were due to
simple vaporization of existing CDDs/CDFs or the formation of new CDDs/CDFs in the
combustion process. The CDD/CDF concentration in the sample measured 1.3 ng TEQDF-
WHO98/kg, which is approximately 20 times lower than the Oregon CDD/CDF emission
6-15
-------
Table 6-2. CDD/CDF mean emission factors (ng/kg) for forest firesa
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total CDD/CDF
Total TEQDF-WHO98
Total I-TEQDF
Nondetect set to zero
1.15
3.83
5.68
10.70
17.34
166.27
663.67
6.98
6.34
10.09
16.72
7.14
1.11
9.81
25.39
3.06
10.27
965.55
19.90
18.60
Nondetect set to 1A detection
limit
1.28
3.83
5.68
10.70
17.34
166.27
663.67
6.98
6.35
10.11
16.74
7.16
1.20
9.85
25.39
3.12
10.32
965.99
20.06
18.75
"Values were derived from a total of seven biomass samples from Oregon and North Carolina.
Source: Gullet and Touati (2003).
concentrations (average, 25 ng TEQDF-WHO98/kg; range, 14 to 46 ng TEQDF-WHO98/kg). The
CDD/CDF isomer patterns of the extracted biomass samples and the emission samples were
similar. Therefore, this preliminary evidence suggests CDD/CDF emissions are not due solely to
vaporization of cuticle-bound CDDs/CDFs but are formed anew during forest fires.
Additionally, the new CDDs/CDFs formed may be adsorbed to the waxy cuticle layer in such a
manner that the isomer pattern reflects the ambient CDD/CDF concentrations.
6-16
-------
Ratio (congener emission factor / 2378-CDD/CDFemission factor)
0.000 0.100 0.200 0.300 0.400 0.500 0.600 0.700 0.800
2,3,7,8,-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-HxCDF
1,2,3,4,7,8-PeCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
Figure 6-2. Congener profile for forest fire simulation approach emissions.
Source: Gullet and Touati (2003).
Many factors may affect forest fire CDD/CDF emissions, such as the type of fire (crown
vs. understory and duff), types of species combusted, and location of the fire (near coastal vs.
inland). Additionally, combustion conditions such as wind speed and fuel moisture content may
also result in variations in emissions. These variables, therefore, yield uncertainties in the
calculation of a representative emission factor through forest fire simulations. However, the
emission factor of 19.9 ng TEQDF-WHO98/kg (18.6 ng I-TEQDF) calculated through this approach
appears to be more reasonable than the factors suggested by the soot, CO, and wood stove
approaches because the forest fire simulation approach directly measures CDD/CDF emissions
from forest biomass combusted in an open pile. Additionally, the forest biomass samples
consisted of both live shoots and needle litter of representative species from two distinct
locations.
6-17
-------
6.4.2. Activity Level Information
6.4.2.1. Approach for Reference Year 2000 (Office of Air Quality Planning and Standards
fOAQPSJ)
As part of the 2000 National Emissions Inventory, OAQPS developed activity levels of
wildfires and prescribed burning on a county-level basis for reference year 2000. The number of
acres burned by wildfires and prescribed burning was obtained from the U.S. Forest Service
(USFS) and four U.S. Department of Interior agencies: Bureau of Land Management, National
Park Service, U.S. Fish and Wildlife Service, and Bureau of Indian Affairs (BIA). USFS
provided data for federal, state, and private lands. All data were provided on a state level except
for the BIA wildfire data and the USFS prescribed burning data, which were provided on a
regional level.
Prior to allocating the forest fire activity to the county level, the BIA and USFS regional
data were first allocated to the state level. The BIA data were allocated using the number of
acres of tribal land in each state. The USFS data were allocated using factors developed from
landcover data in the Biogenic Emissions Landcover Database (BELD2) within EPA's Biogenic
Emissions Inventory System; however, the BELD2 data for California were replaced with data
from the 1996 National Toxics Inventory because USFS's Region 5 contains both Hawaii and
California.
For each of the forest fire categories, the activity from all the agencies were then totaled
by state and allocated to the county level using state-to-county land cover factors developed from
BELD2. These BELD2 factors were based on the acreage of rural forest, brush, and grass in each
county. This procedure was used for all states except Alaska and Hawaii, for which BELD2 does
not contain land cover data. For Alaska and Hawaii, state-to-county factors were derived from
data contained in the allocation factor file used for the 1996 National Emissions Inventory.
Using this approach, OAQPS estimated that approximately 8,357,958 acres were burned
by wildfires in 2000 and approximately 1,261,607 acres were burned by prescribed fires in 2000.
To obtain the amount of biomass consumed by wildfires and prescribed burning, the acres of
forest burned were combined with region-specific fuel loading factors, as shown in Table 6-3.
Nationally, approximately 228 million tons of biomass were consumed by wildfires and 15.8
million metric tons of biomass were consumed by prescribed burning in 2000.
6.4.2.2. Approach for Reference Years 1987 and 1995
According to the Council on Environmental Quality's 25th annual report (CEQ, 1997), 5
million acres of forest were lost to wildfires in 1987 and 7 million acres were lost in 1995.
Estimates of the acreage consumed annually during prescribed burns are not readily available for
6-18
-------
Table 6-3. Forest fire fuel loading factors (tons/acre)
Region
Alaska
California
Intermountain
North Central
Northern
Pacific Northwest
Rocky Mountain
Southern
Southwestern
Wildfires
6
18
8
11
60
60
30
9
10
Prescribed burning
12.6
14.2
6.3
8.7
47.3
47.3
23.7
7.1
7.9
Source: U.S. EPA(2002b).
1995 and 1997. An estimated 5.1 million acres of biomass were burned in 1989 during
prescribed burns (Ward et al., 1993). This value was assumed to be appropriate for use for
reference years 1987 and 1995.
To obtain the amount of biomass consumed by wildfires and prescribed burning, the acres
of forest burned were combined with biomass consumption rates of 9.43 metric tons/acre in areas
consumed by wildfires (Ward et al., 1976) and 7.44 metric tons/acre in areas consumed in
prescribed burns. For 1987 and 1995, approximately 38 million tons were consumed by
prescribed burns. Approximately 47 million metric tons of biomass were consumed by wild fires
in 1987 and approximately 66 million metric tons of biomass were consumed in 1995.
6.4.3. Emission Estimates
Combining the emission factor developed using the forest fire simulation approach (19.9
ng TEQDF-WHO98/kg biomass [18.6 ng I-TEQDF/kg biomass]) with the amount of biomass
consumed annually in wildfires and prescribed fires (total of 85 million metric tons in 1987, 104
million metric tons in 1995, and 244 million metric tons in 2000) yields annual emission
estimates of 1,700 g TEQDF-WHO98 (1,581 g I-TEQDF) for 1987; 2,080 g TEQDF-WHO98 (1,934 g
I-TEQDF) for 1995; and 4,880 g TEQDF-WHO98 (4,538 g I-TEQDF) for 2000. For wildfires
specifically, annual reference year emission estimates are 940 g TEQDF-WHO98 (874.2 g I-
TEQDF) for 1987; 1,320 g TEQDF-WHO98 (1,228 g I-TEQDF) for 1995; and 4,560 g TEQDF-
WHO98 (4,241 g I-TEQDF) for 2000. For prescribed fires specifically, annual emission estimates
6-19
-------
are 760 g TEQDF-WHO98 (706.8 g I-TEQDF) for reference years 1987 and 1995 and 320 g TEQDF-
WHO98 (297 g I-TEQDF) for reference year 2000.
These estimates should be regarded as preliminary indications of possible emissions from
this source; further testing is needed to confirm the true magnitude of emissions. The activity
level for both forest fires and biomass combustion is given a low confidence rating because these
values were estimated and may not be representative. The emission factor is highly variable and
dependent on type of biomass burned, therefore, it is judged to be clearly nonrepresentative.
6.5. BACKYARD BARREL BURNING
6.5.1. Emissions Data
In many rural and nonurban areas of the United States, residences may dispose of
household refuse through open backyard burning. This practice usually consists of burning
refuse in a 208-L capacity steel drum. Holes are punched near the bottom of the drum to allow
combustion air to enter. The fire is ignited with a petroleum fuel, e.g., kerosene. The low
combustion temperatures and oxygen-starved conditions associated with household refuse
burning in these "burn barrels" results in poor and uncontrolled combustion conditions. Under
such conditions, products of incomplete combustion are formed and visible smoke is emitted into
the air.
The practice of open burning in burn barrels causes CDDs and CDFs to be formed and
released as toxic air contaminants. In 1997, EPA's Control Technology Center, in cooperation
with the New York State's Department of Health and Department of Environmental
Conservation, conducted an initial study that examined, characterized, and quantified emissions
from simulated open burnings of household waste materials in barrels (Lemieux, 1997). The
representative waste was prepared on the basis of the typical percentages of various waste
materials disposed of by New York State residents; hazardous wastes (e.g., chemicals, paints,
oils) were not included in the test waste. A variety of compounds, including CDDs/CDFs, were
measured in the emissions from two simulated open burnings of this "baseline" waste.
Combustion studies were subsequently performed by EPA to provide additional baseline
waste tests and an initial indication of the impact of limited variation in waste composition and
combustion conditions on CDD/CDF emissions from a simulated domestic backyard barrel burn
of 6.8 kg of unshredded household waste (Gullet et al., 1999, 2000a, b; Lemieux et al., 2000; e-
mail dated September 7, 2000, from P. Lemieux, U.S. EPA, to D. Cleverly, U.S. EPA). The
results of seven baseline open burning waste tests were reported in these EPA studies. These
tests exhibited variation in the emissions of CDDs/CDFs, with a one to two order-of-magnitude
spread between the lowest and highest values for individual congeners, congener groups, total
CDDs/CDFs, and TEQ values. The average TEQ emission factor for the seven baseline tests was
6-20
-------
72.8 ng I-TEQDF/kg waste burned (setting nondetect values equal to zero) and 73.7 ng I-
TEQDF/kg waste burned (setting nondetect values equal to one-half the DL). The corresponding
TEQDF-WHO98 values were 76.8 and 77.7 ng/kg. Table 6-4 presents the average congener and
congener group results for these tests.
In addition to the baseline tests, the combustion experiments included testing at three
different PVC levels: 0, 1, and 7.5% by weight PVC. The average emissions were 14, 201, and
4,916 ng I-TEQDF/kg waste burned, respectively. Two tests using waste impregnated with
inorganic chloride (CaCl2) at a concentration of 7.5% by weight (and no PVC) averaged 734 ng I-
TEQDF/kg. Qualitative comparisons suggest that the tests conducted with higher chlorine, via
PVC or CaCl2, resulted in substantial increases in TEQ emissions.
Other variations in baseline waste composition included conducting one test with
compressed waste, one test with a double load of waste, and one test in which some of the waste
paper was wetted to simulate high-moisture burns. These tests resulted in a higher mean TEQ
emission factor (534 ng I-TEQDF/kg) than that of the baseline runs.
Several waste combustion variables were evaluated, such as average temperatures at
prescribed barrel heights, length of time temperatures (favorable temperature ranges) for
CDD/CDF formation, and measurement of CO, CO2, O2, particulate matter, and HC1. Statistical
analyses of the results indicated that CO emissions and temperature measured in the uppermost
portion of the barrel were the best predictors of TEQ variation. However, the wide variability in
test results (from less than 10 to more than 6,000 ng I-TEQDF/kg) also indicates that a high degree
of CDD/CDF emission variation can be expected due to factors that are not wholly related to
waste composition or burning practice, such as waste orientation. A mean emission factor of the
baseline tests (without PVC added) was developed from the data. This mean emission factor was
78.6 mg TEQDF-WHO/kg, and it was used to estimate releases from barrel burning. The
emission factor is given a low confidence rating because it is possibly nonrepresentative of
barrel-burning emissions.
6.5.2. Activity Level Information
The amount of refuse that is combusted annually in the United States in residential
backyard burn barrels is largely unknown. Although no national statistics are available, a limited
number of telephone surveys have attempted to measure the prevalence of backyard barrel
burning in a few geographical areas. This limited number of surveys, combined with census data
on the rural and nonurban population of the United States, were used to estimate annual activity
level in terms of the quantity of refuse combusted in burn barrels per reference year. The
following is a summary of this estimation procedure.
6-21
-------
Table 6-4. CDD/CDF average air emission factors (ng/kg waste burned)
from barrel burning of household wastea
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Nondetect set to 1A detection limit
3.4
8.2
6.6
9.9
19.1
39.8
49.7
45.6
37.2
65.2
113.8
38.5
61.9
3
128.6
14.6
37.5
136.7
545.9
73.7
77.7
413
281
221
105
43
1,880
1,021
492
169
32
4,657
Nondetect set to zero
2.7
8.1
6.4
9.7
19
39.8
49.7
45.6
37.2
65.2
113.8
38.5
61.9
2.5
124.4
15
36.4
135.4
540.5
72.8
76.8
413
281
221
105
43
1,880
1,021
492
169
30
4,655
aListed values are the arithmetic averages of seven tests for the congeners and the averages of five tests for the
congener groups.
Sources: E-mail dated September 7, 2000, from P. Lemieux, U.S. EPA, to D. Cleverly, U.S. EPA; Gullett et al.
(1999, 2000, 2001).
6-22
-------
6.5.2.1. Summary of Barrel Burn Surveys
A total of seven surveys of the prevalence of backyard combustion of domestic refuse in
burn barrels were identified in the literature. For the most part, these surveys were an attempt to
estimate the barrel-burning activity in a specific state, county, or region in support of regulatory
determinations on barrel burning. In general, the results of the surveys showed a prevalence of
barrel burning within the rural population to range from 12 to 40%, with a mean of 28%. The
following is a review of the surveys.
The Two Rivers Region Council of Public Officials (TRRCPO) and Patrick Engineering
conducted a telephone survey in the early 1990s of residents of five central Illinois counties.
They found that about 40% of the residents in a typical rural county burned household waste.
The survey also indicated that, on average, those households that burned waste disposed of
approximately 63% of their household waste by burning it in barrels (TRRCPO, 1994).
Similar results were obtained in a survey conducted by Zenith Research Group, Inc.,
(2000) for the Western Lake Superior Sanitary District of Minnesota. This survey of 760
residents of selected portions of northwestern Wisconsin and northeastern Minnesota addressed,
in part, the use of burn barrels or other devices to burn household garbage or other materials.
Among all survey respondents, 27.5% said they used burn barrels or other devices to burn
household garbage or other materials.
Environics Research Group conducted a household garbage disposal and burning survey
of 1,516 residents of Ontario, Canada. All respondents resided in detached single-family homes.
Approximately 24% of all respondents reported burning their household refuse in burn barrels
(Environics Research Group, 2001).
E.H. Pechan and Associates conducted a residential municipal solid waste survey for the
Mid-Atlantic/Northeast Visibility Union (MANE-VU) states and tribes (Pechan and Associates,
2002). The MANE-VU entities include Connecticut, Delaware, the District of Columbia, Maine,
Maryland, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, the Penobscot
Indian Nation, Rhode Island, the St. Regis Mohawk Tribe, and Vermont. Household waste
burning surveys were conducted by telephone for 72 residents of rural, suburban, and urban
jurisdictions, as classified by the 1990 census. The residents were asked to estimate the number
of households in their jurisdiction that burned household waste or trash. In general, the survey
estimated that 11.9% of the rural population burned refuse in backyard burn barrels.
The State of California Air Resources Board (CARB) undertook a study of the prevalence
of backyard refuse burning in rural areas of 21 air management districts in California (CARB,
2002). From this study, CARB estimated that approximately 18% of the rural population in
California combusted their household refuse in backyard burn barrels.
6-23
-------
In 1993, the St. Lawrence County Planning Office in Canton, NY, conducted a survey of
open burning of domestic refuse (St. Lawrence County, 1993). From the survey, it was
concluded that 48.2% of 9,926 households in rural areas of the county burned household refuse
in burn barrels.
In 1997, the State of Maine Department of Conservation Forestry Bureau surveyed rural
town fire wardens and state fire rangers regarding the prevalence of backyard burning of
household waste. It was revealed through the survey that each day approximately 19,147 kg of
domestic refuse was being combusted state wide in approximately 8,510 burn barrels. In relation
to the total population, it was noted that one burn barrel existed for every 144 individuals.
6.5.2.2. Estimates of Activity Level
The following chart summarizes the steps taken to estimate the quantity of household
refuse combusted in backyard burn barrels in!987, 1995, and 2000.
Step Assumption 2000 1995 1987
1 U.S. population 281,400,000 260,600,000 242,300,000
2 Population in rural and
nonurban areas 59,000,000 52,700,000 50,700,000
3 Percent nonurban
population burning
household refuse 28 40 40
4 Adjusted population
burning household refuse in
barrels 16,726,500 21,080,000 20,280,000
5 Per capita household refuse
generati on rate (kg/yr) 616 616 616
6 Percent of household refuse
generated burned at homes 63 63 63
7 Adjusted per capita
household refuse burned
(kg/yr) 388 388 388
8 Total refuse generated by
rural and nonurban
population burning
household refuse (kg/yr) 6,491,220,120 8,180,726,400 7,870,262,400
For steps 1 and 2, the U.S. Census Bureau Statistical Abstract of the United States was
used to determine population size. For step 3, the assumption of the percent of rural population
6-24
-------
combusting refuse in burn barrels in 2000 was derived from the overall mean prevalence rate
from six surveys (CARB, 2002; Zenith Research Group, 2001; Environics Research Group,
2001: Pechan and Associates, 2004; St. Lawrence County, 1993; and TRRCPO, 1994). This
produced a mean prevalence of 28%, which should reflect the impact of state bans or restrictions
on the practice of open burning of refuse. The 1995 and 1987 population was estimated using
the survey by TRRCPO (1994). This produced a prevalence of 40%, which should reflect the
fact that the practice of open burning of refuse was not banned or restricted by the majority of the
states.
The above mean prevalence rates were used in step 4 to calculate the number of people
residing in rural areas assumed to have burned household refuse in burn barrels in each reference
year (assumption in step 2 multiplied by the percentage in step 3). The annual per capita
household refuse generation rate in step 5 is from Municipal Solid Waste Fact Book (2000). The
figure of 616 kg/person/yr is the result of subtracting out weight of yard waste from the per capita
generation rate. The assumption in step 6 is that 63% of the household refuse generated in rural
areas is burned in backyard burn barrels. This percentage was derived from the survey conducted
in rural counties of Illinois by TRRCPO (1994).
These activity levels are adopted and assigned a confidence rating of low because they are
derived from limited surveys that are possibly nonrepresentative of the national activity level.
6.5.2.3. Alternative Approach to Estimating Activity Level
The actual amount of refuse combusted in burn barrels in the United States is unknown
and must be estimated. OAQPS developed activity levels of residential MSW combusted in
backyard barrels for reference year 2000. The activity levels were determined by first estimating
the amount of waste generated for each county in the United States. The amount of waste
generated was estimated by using a national average per capita waste generation factor, which is
1.5 kg/person/day. This value was calculated using population data from the 2000 census and
2000 waste generation data (U.S. EPA, 2002c). To better reflect the actual amount of household
residential waste subject to being burned, noncombustibles (glass and metals) and yard waste
were excluded. This factor was then applied to the portion of the county's total population that is
considered rural, since open burning is generally not practiced in urban areas.
Using data from TRRCPO (1994), it was estimated that for rural populations, 25 to 32%
of generated MSW is burned. A median value of 28% was assumed for the nation, and this
correction factor was applied to the total amount of waste generated. Controls (or burning bans)
were accounted for by assuming that no burning takes place in counties where the urban
population is at least 80% of the total population (i.e., urban plus rural). Zero emissions from
open burning were attributed to these counties. This technique produced an estimated annual
6-25
-------
activity level of 7.79 billion kg of residential household waste combusted in burn barrels in 2000.
This estimate is approximately 16.5% greater than the estimate used in this report.
6.5.3. Emission Estimates
CDD/CDF emissions from burn barrels for reference years 1987, 1995, and 2000 were
calculated by multiplying the estimated annual total weight of household refuse combusted in
burn barrels (see Section 6.5.2.2) by the dioxin emission factor. The emission factor was 76.8 ng
TEQDF-WHO98/kg (72.8 ng I-TEQDF/kg) waste burned.
Annual nationwide TEQ emissions for 1987, 1995, and 2000 were calculated using
the following equation:
F = £77 x AT
^TEQ 1-'± TEQ x ^1-tj
where:
ETEQ = Annual TEQDF emissions (g/yr)
EFTEQ = TEQDF emission factor (76.8 ng TEQDF-WHO98/kg waste burned)
AL = Annual activity level (7,870,262,400 kg in 1987; 8,180,726,400 kg in 1995; and
6,491,220,120 kg in 2000).
Using this equation, estimated nationwide TEQDF emissions were 604 g TEQDF-WHO98
(573 g I-TEQDF) in 1987, 628 g TEQDF-WHO98 (595 g I-TEQDF) in 1995, and 498.53 g TEQDF-
WHO98 (472.56 g I-TEQDF) in 2000. A low confidence rating is given to both the emission factor
and the estimate of activity level, therefore, the confidence rating is low for the estimate of TEQ
emissions from backyard barrel burning of refuse.
6.5.4. Composition of Ash from Barrel Burning
Ash samples were collected from open barrel burning (Lemieux, 1997) and analyzed for
CDDs/CDFs and PCBs. Ash samples from the experiments were combined, resulting in two
composite samples, one for recyclers and one for nonrecyclers. The results for PCBs depict only
the data for specific PCB congeners. The remaining PCB data reported in Lemieux (1997) could
not be related to a particular congener. The results are presented in Tables 6-5 and 6-6.
6.6. RESIDENTIAL YARD WASTE BURNING
6.6.1. Emissions Data
It is reasonable to assume that residential yard waste burning may be a source of
CDDs/CDFs, as they have been detected in forest and brush fires. No direct measurements of
CDD/CDF emissions from residential yard waste burning have been performed; however, Gullet
6-26
-------
Table 6-5. CDD/CDF analysis for composite ash samples from barrel
burning (ng/kg ash)
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD
Total CDF
Total CDD/CDF
Average
concentration in
composite ash sample
Avid
recycler
31
230
270
420
300
4,000
9,600
830
1,000
2,500
2,300
2,100
2,900
810
12,000
1,400
8,200
2,500
4,100
5,600
7,600
9,600
25,000
21,000
19,000
17,000
8,200
29,400
90,200
119,600
Non-
recycler
9
53
44
74
56
630
690
220
270
690
480
490
670
150
2,100
170
560
490
740
1,300
1,300
690
8,200
6,600
4,600
2,900
560
4,520
22,860
27,380
I-TEQDF
Avid
recycler
31
115
27
42
30
40
9.6
83
50
1,250
230
210
290
81
120
14
8.2
-
-
-
-
-
-
-
-
-
-
-
-
-
Non-
recycler
9
26.5
4.4
7.4
5.6
6.3
0.69
22
13.5
345
48
49
67
15
21
1.7
0.56
-
-
-
-
-
-
-
-
-
-
-
-
-
TEQDF-WH098
Avid
recycler
31
230
27
42
30
40
0.96
83
50
1,250
230
210
290
81
120
14
0.82
-
-
-
-
-
-
-
-
-
-
-
-
-
Non-
recycler
9
53
4.4
7.4
5.6
6.3
0.069
22
13.5
345
48
67
15
21
1.7
0.056
-
-
-
-
-
-
-
-
-
-
-
-
-
Source: Lemieux (1997).
6-27
-------
Table 6-6. PCB analysis for composite ash samples from barrel burning
(ng/kg ash)
Compound
2-Chlorobiphenyl
2,3'-Dichlorobiphenyl
2,2',6-Trichlorobiphenyl
2,2',5-Trichlorobiphenyl
2,3',5-Trichlorobiphenyl
2,3',4-Trichlorobiphenyl
2,4',5-Trichlorobiphenyl
2,4,4'-Trichlorobiphenyl
2,2',4,6'-Tetrachlorobiphenyl
2,2',3,6'-Tetrachlorobiphenyl
2,2',5,5'-Tetrachlorobiphenyl
2,2',3,5'-Tetrachlorobiphenyl
2,2',4,4',5-Pentachlorobiphenyl
2,2',3,3',5-Pentachlorobiphenyl
2,2',3,4,5,5'-Hexachlorobiphenyl
Avid recycler
<2,500
3,700
<500
32,000
800
<500
1,500
<500
<500
5,300
3,100
2,600
3,400
400
1,200
Nonrecycler
4,900
4,700
5,600
6,300
800
700
900
500
1,500
1,300
1,800
1,200
1,300
<500
<500
Source: Lemieux (1997).
and Touati (2003) measured an average CDD/CDF emission factor of 20 ng TEQDF-WHO98/kg
during forest fire simulations where biomass samples from Oregon and North Carolina were
burned on an open platform (see Section 6.4). Therefore, the emission factor of 20 ng TEQDF-
WHO98/kg developed by Gullet and Touati (2003) was used for residential yard waste burning.
6.6.2. Activity Level Information
Pechan and Associates (2002) estimated that approximately 233 Ib of yard waste per
household per year (based on a four-person household) were burned in 2000. This is similar to
the estimate of 106 to 319 Ib yard waste per household per year (based on a four-person
household) that Pechan calculated using the results of a household yard waste burning survey.
Pechan's telephone survey was completed by 181 rural, suburban, and urban jurisdictions of the
MANE-VU entities. The results indicated that approximately 28% of the population in rural
areas burned household yard waste and that households typically conducted two to three burns
6-28
-------
per year. Additionally, as indicated by information provided by three respondents, 1 to 3 cubic
yards of yard waste was typically burned at a time.
As part of the 2000 National Emissions Inventory, OAQPS determined on a county-level
basis the amount of yard waste burned in 2000. The activity level estimates were based on the
assumption that yard waste was generated at a rate of 0.54 Ib/person/day in 2000, which in turn
was derived using population data for 2000 and the assumption that 27.7 million tons of yard
waste were generated in 2000 (U.S. EPA, 2002c). Of the total amount of yard waste generated,
the composition was assumed to be 25% leaves, 25% brush, and 50% grass by weight (U.S. EPA,
200 Ib). Because open burning of grass clippings is not typically practiced by homeowners, only
50% of the yard waste generated was assumed to be burnable. Additionally, OAQPS assumed
that burning primarily occurs in rural areas (i.e., the per capita yard waste generation factor was
applied to only the rural population in each county) and that only 28% of the total yard waste
generated was actually burned (see Section 6.5).
The amount of yard waste assumed to be generated in each county was then adjusted for
variation in vegetation using BELD2. For counties with 10 to 50% forested land, the amount of
yard waste generated was reduced to 50% and for counties with less than 10% forested land, to
zero (i.e., no yard waste was generated). Adjustments for variation in vegetation were not made
to counties where the percentage of forested acres was greater than or equal to 50%. Before
calculating the percentage of forested acres per county, the acreage of agricultural lands was
subtracted from the acreage of forested lands to better account for the native vegetation that
would likely be occurring in the residential yards of farming states. Controls (or burning bans)
were accounted for by assuming that no burning took place in counties where the urban
population exceeded 80% of the total population (i.e., urban plus rural). Using this method,
OAQPS estimated that 255,000 metric tons of leaf and 255,000 metric tons of brush (total of
510,000 metric tons of yard waste) were burned in 2000.
6.6.3. Emission Estimates
Using the emission factor of 20 ng TEQDF-WHO98/kg (18.6 ng I-TEQDF/kg) and the
activity level of 510,000 metric tons yard waste burned in 2000, CDD/CDF emissions from open
burning of yard waste were 10.2 g TEQDF-WHO98 (9.5 g I-TEQDF) in 2000. Assuming 772 and
754 million kg of yard waste were burned in 1987 and 1995, respectively, then 15.4 g TEQDF-
WHO98 (14.4 g I-TEQDF) and 15.19 g TEQDF-WHO98 (14 g I-TEQDF) were emitted in 1987 and
1995, respectively. These numbers should be regarded as preliminary estimates of possible
emissions from this source; further testing is needed to confirm the true magnitude of emissions
because both the emission factor and activity levels are judged to be clearly nonrepresentative of
the source category.
6-29
-------
6.7. LAND-CLEARING DEBRIS BURNING
6.7.1. Emissions Data
During the clearing of land for the construction of new buildings (residential and
nonresidential) and highways, trees, shrubs, and brush are often torn out, collected in piles, and
burned. As with residential yard waste burning, it is assumed that the burning of land-clearing
debris may generate CDDs/CDFs because emissions have been detected from forest and brush
fires. No direct measurements of CDD/CDF emissions from the burning of land-clearing debris
have been performed, so the average emission factor of 20 ng TEQDF-WHO98/kg, which was used
for both forest fires and residential yard waste burning, is also used for burning of land clearing
debris (see Sections 6.4 and 6.6).
6.7.2. Activity Level Information
Activity levels associated with land-clearing debris were calculated by OAQPS on a
county-level basis using the number of acres disturbed through residential, nonresidential, and
roadway construction.
6.7.2.1. Residential Construction
In 2000, approximately 330,551 acres were disturbed by residential construction. This
number is based on county-level housing permit data and regional housing start data obtained
from the U.S. Census Bureau for single-family units, two-family units, and apartment buildings.
The county permit data were first adjusted to equal regional housing-start data, and then the
number of buildings in each housing category was estimated. The total number of acres
disturbed by residential construction was then determined by applying the following conversion
factors to the housing-start data for each category:
Unit type Acres per building
Single-family unit 1/4
Two-family unit 1/3
Apartments 1/2
6.7.2.2. Nonresidential Construction
In 2000, approximately 336,224 acres were disturbed by nonresidential construction.
This number is based on the national value of construction put in place, as reported by the U.S.
Census Bureau. The national value was allocated to counties using construction employment
data from the Bureau of Labor Statistics and Dun & Bradstreet. A conversion factor of 1.6 acres
disturbed per $100,000 spent was applied to the county-level estimates of the value of
6-30
-------
construction put in place to obtain the acres disturbed by nonresidential construction per county.
The conversion factor was developed using the Price and Cost Indices for Construction by
adjusting the 1992 value of 2 acres per $100,000 for 2000.
6.7.2.3. Roadway Construction
In 2000, approximately 190,367 acres were disturbed by roadway construction. This
number is based on 1999 Federal Highway Administration state expenditure data for capital
outlay within the following six road classifications: interstate (urban and rural), other arterial
(urban and rural), and collectors (urban and rural). The expenditure data were converted to miles
of road constructed based on data from the North Carolina Department of Transportation
(NCDOT). According to NCDOT, approximately $4 million per mile is spent for freeway and
interstate projects, and approximately $1.9 million per mile is spent for arterial and collector
projects. The number of miles was then converted to acres disturbed using the following
conversion factors for each road classification category:
Road type Acres per mile
Interstate, urban 15.2
Interstate, rural 15.2
Other arteri al, urb an 15.2
Other arterial, rural 12.7
Collectors, urban 9.8
Collectors, rural 7.9
For 1995, state expenditure for capital outlay was assumed to be 74% of total funding.
This percentage was derived using 2000 data (U.S. DOT, 2002). For 1987, 74% of the total
capital outlay of the average of 1985 and 1989 was used (capital outlays for 1985 and 1989 are
reported in U.S. DOT, 2002). Therefore, approximately 83,110 and 123,140 acres were
disturbed as the result of roadway construction in 1987 and 1995, respectively.
6.7.2.4. Fuel Loading Factors
To obtain the amount of biomass consumed by the burning of land-clearing debris, the
total acreage of land disturbed in each county by residential, nonresidential, and roadway
construction was distributed according to vegetation type (hardwood, softwood, and grass) and
then combined with vegetation-specific fuel loading factors. The percentage of vegetation type
within each county was determined using BELD2. The average loading factors used for each
fuel type were 99 tons/acre for hardwood, 57 tons/acre for softwood, and 4.5 tons/acre for grass.
6-31
-------
Using this method, OAQPS estimated that 28.4 million metric tons of biomass were
burned through land clearing activities in 2000. EPA developed a national average biomass
loading factor of 33 metric tons burned per acre in 2000. Using this loading factor combined
with total acreage disturbed, EPA estimated that approximately 27.7 and 26.4 million metric tons
of biomass were burned by land clearing in 1987 and 1995, respectively.
6.7.3. Emission Estimates
Using the emission factor of 20 ng TEQDF-WHO98/kg (18.6 ng I-TEQDF/kg) and the
activity level estimates in Section 6.7.2.4, CDD/CDF emissions from land clearing burning were
568 g TEQDF-WHO98 (528 g I-TEQDF) in 2000, 528 g TEQDF-WHO98 (491 g I-TEQDF) in 1995,
and 553 g TEQDF-WHO98 (515 g I-TEQDF) in 1987. These should be regarded as preliminary
estimates of possible emissions from this source because the emission factor is clearly
nonrepresentative; further testing is needed to confirm the true magnitude of emissions.
6.8. UNCONTROLLED COMBUSTION OF POLYCHLORINATED BIPHENYLS
The accidental combustion of PCB-containing electrical equipment or intentional
combustion of PCBs in incinerators and boilers not approved for PCB burning (40 CFR 761)
may produce CDDs/CDFs. At elevated temperatures, such as in transformer fires, PCBs can
undergo reactions to form CDFs and other by-products. More than 30 accidental fires and
explosions involving PCB transformers and capacitors in the United States and Scandinavia that
involved the combustion of PCBs and the generation of CDDs/CDFs have been documented
(Hutzinger and Fiedler, 1991a; O'Keefe and Smith, 1989; Williams et al., 1985). For example,
analyses of soot samples from a Binghamton, NY, office building fire detected 20 jig/g total
CDDs (0.6 to 2.8 |ig/g 2,3,7,8-TCDD) and 765 to 2,160 |ig/g total CDFs (12 to 270 |ig/g 2,3,7,8-
TCDF). At that site, the fire involved a mixture containing PCBs (65%) and chlorobenzene
(35%). Laboratory analyses of soot samples from a PCB transformer fire that occurred in Reims,
France, indicated total CDD and CDF levels in the range of 4 to 58,000 ng/g and 45 to 81,000
ng/g, respectively.
Using a bench-scale thermal destruction system, Erickson et al. (1984) determined the
optimum conditions for CDF formation to be a temperature of 675°C, an excess oxygen
concentration of 8%, and a residence time of 0.8 sec (or longer). Combusting mineral oil and
silicone oil containing 5, 50, and 500 ppm of Aroclor 1254 at these conditions yielded PCB to
CDF conversion efficiencies as high as 4%. Up to 3% conversion efficiency was observed when
an Askarel (70% Aroclor 1260) was combusted under the same conditions.
6-32
-------
The use of PCBs in new transformers in the United States is banned, and their use in
existing transformers and capacitors is being phased out under regulations promulgated under the
Toxic Substances Control Act.
Because of the accidental nature of these incidents, the variation in duration and intensity
of elevated temperatures, the variation in CDD/CDF content of residues, and uncertainty
regarding the amount of PCBs still in service in electrical equipment, EPA judged the available
data inadequate for developing any quantifiable emission estimates. However, Thomas and
Spiro (1995) conservatively estimated that about 15 g of TEQ may be generated annually from
fires in commercial and residential buildings each year. This estimate is based on the following
assumptions: (a) the I-TEQDF emission rate is 20 |ig/kg of PCB burned, (b) 74,000 metric tons of
PCB are still in use in various electrical equipment, and (c) 1% of the in-use PCBs are burned
during the course of structural fires annually.
6.9. VOLCANOES
To date, no studies demonstrating the formation of CDDs/CDFs by volcanoes have been
published. Given the available information from the studies discussed below, volcanoes do not
appear to be sources of CDD/CDF release to the environment.
Gribble (1994) summarized some of the existing information on the formation of
chlorinated compounds by natural sources, including volcanoes. Gribble reported that several
studies had demonstrated the presence of chlorofluorocarbons and simple halogenated aliphatic
compounds (one and two carbon chain length) in volcanic gases. In addition, several chlorinated
monoaromatic compounds as well as three PeCB congeners were detected in the ash from the
1980 eruption of Mount St. Helens. Gribble hypothesized that the formation of these PCB
compounds was the result of rapid, incomplete, high-temperature combustion of chloride-
containing plant material in the eruption zone. However, no information was presented to
indicate the formation of CDDs/CDFs by volcanoes.
Lamparski et al. (1990) analyzed groundfall ash samples collected at various distances
and locations from Mount St. Helens following the eruption in 1980. The findings of this study
indicate that volcanic paniculate emissions were free of detectable PCBs and nearly free of
detectable CDDs (0.8 ng/kg HpCDD detected) upon exiting the volcano and remained so
throughout their period of deposition in the blast zone. However, upon transport through the
atmosphere, measurable and increasing levels of CDDs and PCBs were detected in deposited ash
as it passed from rural to urban environments. The authors hypothesized that CDDs and PCBs in
the atmosphere became associated with the volcanic ash particulates through gas-phase sorption
or particulate agglomeration.
6-33
-------
Takizawa et al. (1994) investigated the CDD/CDF content of volcanic dust fall from two
active volcanoes in Japan (Mt. Fugendake and Sakurajima). The study was not designed to
determine whether the CDDs/CDFs observed were formed by the volcanoes or were scavenged
from the atmosphere by the falling dust and ash. The dust fall was collected for one-month
periods during July and October 1992; two samples of the volcanic ash were collected in 1992.
The results of the sample analyses for 2,3,7,8-substituted CDDs and CDFs, presented in Table
6-7, show that no 2,3,7,8-substituted congeners with less than seven chlorines were detected;
however, the authors reported that non-2,3,7,8-substituted congeners in the lower-chlorinated
congener groups were detected.
Table 6-7. CDDs/CDFs in dust fall and ashes from volcanoes
2,3,7,8-substituted
congener group
TCDD
PeCDD
FIxCDD
HpCDD
OCDD
TCDF
PeCDF
FIxCDF
HpCDF
OCDF
Dust fall (mg/km2/month)a
July 1992
<0.5
<0.5
<0.5
9.2
14
<0.5
<0.5
<0.5
1.9
4.2
Oct. 1992
<0.5
<0.5
<0.5
5.2
11
<0.5
<0.5
<0.5
2.8
1.8
Volcanic ash (ng/kg)b
Ash no. 1
<0.1
<0.1
<0.1
2.5
1.7
<0.1
<0.1
<0.1
1.2
<0.5
Ash no. 2
<0.1
<0.1
<0.1
1.8
2.2
<0.1
<0.1
<0.1
1.2
<0.5
aDust fall measured from the active volcano Fugendake.
Volcanic ash measured from the active volcano Sakurajima.
Source: Takizawa etal. (1994).
6.10. FIREWORKS
In order to produce various effects and illuminations, modern fireworks contain black
powder and substances such as chlorine-based oxidizers, flame-coloring copper salts, and
pulverized polyvinylchloride, which are known to be involved in dioxin-forming processes.
During deflagration of pyrotechnics, core temperatures reach as high as 2,500°C, which would
most likely inhibit the formation of organic pollutants. However, CDDs/CDFs may be generated
in the areas adjacent to the combustion zone, where temperatures are lower and dwell times are
longer. Therefore, CDDs/CDFs may be generated during the cooling period of the deflagration
6-34
-------
products because the temperatures of the smoke and ash are within the possible temperature
range of dioxin formation (Fleischer et al., 1999).
During a celebration in Oxford, England, that was accompanied by fireworks and bonfires,
Dyke and Coleman (1995) reported a fourfold increase in CDD/CDF TEQ concentrations in the
ambient air (see Section 6.2.2). Fleischer et al. (1999) conducted an experiment to measure the
air emissions resulting specifically from the following seven types of fireworks: firecracker,
cone fountain, jumping jack, whistler, sparkling rocket, roman candle, and four-color fountain.
The paper cartridges and charges were separated from each firework and deflagrated separately in
a steel chamber. CDD/CDF concentrations were measured both in air samples and in paper and
ash samples. The results indicated that dioxins were not present in significant quantities in the
air samples collected. Therefore, Fleisher et al. suspected that the increased background
concentrations of CDDs/CDFs detected by Dyke and Coleman (1995) were due mainly to the
bonfires and not the fireworks. However, concentrations of HpCDD and OCDD/OCDF were
present in the paper and ash collected after the fireworks were detonated at concentrations
ranging from less than the DL (10 ng/kg) to 1,200 ng/kg. Table 6-8 depicts the results of
Fleischer's tests.
Table 6-8. Residue of HpCDD/HpCDF and OCDD/OCDF (ng/kg) in paper
cartridges and charges of select pyrotechnic products
Product
Firecracker
Cone fountain
Jumping Jack
Whistler
Sparkling rocket
Roman candle
Four-color fountain
Paper Cartridges
HpCDD
16
111
<10
22
30
<10
<10
OCDD
322
384
33
353
129
426
18
OCDF
79
22
24
121
12
39
<10
Charge
HpCDD
<10
<10
<10
<10
<10
<10
<10
OCDD
535
<10
28
35
13
<10
<10
OCDF
26
<10
<10
1200
<10
22
<10
Source: Fleischer etal. (1999).
Given the lack of information on the potential for CDD/CDF emissions from fireworks,
emissions cannot be quantified.
6-35
-------
6.11. OPEN BURNING AND OPEN DETONATION OF ENERGETIC MATERIALS
Open burning and open detonation (OB/OD) practices are routinely used to destroy surplus
or unserviceable energetic materials. Mitchell and Suggs (1998) conducted a study to determine
emission factors from OB/OD. Air samples were collected for CDD/CDF analysis during four
burns and after three detonations. The results of the study indicated that emission levels of
CDDs/CDFs as a result of disposal of energetic materials by OB/OD were nondetectable.
6-36
-------
7. METAL SMELTING AND REFINING SOURCES OF CDDs/CDFs
7.1. PRIMARY NONFERROUS METAL SMELTING/REFINING
Little information has been published on the potential for the formation and
environmental release of CDDs/CDFs from primary nonferrous metal smelting facilities.
CDD/CDF releases from these facilities were first reported in the wastewater of a magnesium
refining facility and in the receiving water sediments downstream of a nickel refining facility in
Norway (Oehme et al., 1989). This study resulted in the evaluation of the potential for
CDD/CDF releases from primary nonferrous metal smelting operations in the United States. Air
emissions from several U.S. smelting operation facilities have been sampled. The findings of
these studies are reviewed in the following sections.
7.1.1. Primary Copper Smelting and Refining
Environmental Risk Sciences (1995) prepared an analysis for the National Mining
Association on the potential for CDD/CDF emissions from operations in the primary copper
smelting industry. The analysis included reviewing the process chemistry and technology of
primary copper smelting, identifying operating conditions, and comparing process stream
compositions from seven of the eight U.S. primary copper smelters that are members of the
National Mining Association. The analysis also included stack testing for CDDs/CDFs at two
facilities. The stack testing involved the principal off-gas streams for copper smelters: the main
stack stream, the plant tail gas stack stream, and vent fume exhaust (Secor International, Inc.,
1995b). The two facilities that were tested (Phelps Dodge Mining Co. in Playas, NM, and
Cyprus Miami Mining Co. in Claypool, AZ) were selected as representative of the industry
because of their similarity to other facilities in terms of process chemistry, process stream
composition, and process stream temperatures. CDDs/CDFs were not detected in the air
emissions from either facility.
The results of the analysis indicate that although there is some potential for CDD/CDF
formation in this industry's operations, several factors lessen the probability, including the
following: (a) most of the energy used to melt copper is derived from oxidation of copper sulfide
ore minerals (CuFeS2) rather than carbon (fossil fuels), (b) low concentrations of organic carbon
and chloride are present in raw materials and reagents, (c) high concentrations of SO2 are present
in process gases (6 to 40% by volume), (d) high temperatures are maintained in the furnaces and
converters (1,100 to 1,500°C), and (e) copper (II) chloride is apparently absent in process
emissions.
Although CDDs/CDFs have not been detected in U.S. facilities, CDD/CDF emissions
have been measured in the stack of a primary copper smelter in Canada. In 2001, emission
7-1
-------
measurements for various persistent, bioaccumulative, and toxic substances, including
CDDs/CDFs, were collected in Canada as a voluntary initiative under the Great Lakes Binational
Toxics Strategy (Cianciarelli, 2001). One of the facilities tested was the Falconbridge Kidd
Metallurgical plant in Timmins, Ontario, a copper smelting plant. Emission summaries are
provided in Table 7-1 as TEQ concentrations corrected for 11% oxygen for the average of three
runs. The total concentrations for the three runs were 3.8, 1.7, and 0.7 pg TEQ/m3. Annual
CDD/CDF emission rates were estimated to be 0.002 g I-TEQ/yr.
In 2002, Environment Canada began developing a generic dioxin/furan emissions testing
protocol for use by the base metals smelting sector (Charles E. Napier Company, Ltd., 2002).
Several base metals smelting and refining complexes were identified, and a summary of readily
available published information on dioxin/furan emissions from the base metals smelter
processes was compiled. A summary of this information is provided in Table 7-2. Four facilities
were identified as primary copper smelters and had CDD/CDF emission concentrations ranging
from less than 1 to 559 pg I-TEQ/dscm.
Table 7-1. CDD/CDF emission concentrations (pg TEQ/m3
primary copper smelters
11% oxygen) for
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
TOTAL
Runl
0
0
0
0.1
0.1
0.1
0
0.7
0.1
2.2
0
0.1
0.3
0
0.1
0
0
3.8
Run 2
0
0
0
0
0
0.1
0
0.2
0
0.9
0.2
0.1
0.2
0
0.1
0
0
1.8
Run3
0
0.1
0
0
0
0
0
0.1
0
0
0.1
0
0.1
0
0
0
0
0.4
Source: Cianciarelli (2001).
7-2
-------
Table 7-2. CDD/CDF emissions data from primary and secondary copper
and secondary lead smelters
Company, location
Process units
Emission control
technology
CDD/CDF emission
concentration
(pg I-TEQ/dscm)
Primary copper smelters
Norddeutsche Affinerie,
Germany
Falconbridge, Sudbury,
Ontario (nickel and copper)
Noranda, Home smelter,
Noranda, Quebec
Noranda, Gaspe smelter,
Murdockville, Quebec
Outokumpu flash smelting
furnace
Peirce-Smith converter
Roasting
Electric smelting
Peirce-Smith converters
Noranda reactor
Noranda continuous converter
Reverberatory furnace
Peirce-Smith converter
Waste heat boiler, ESP
ESP
Cyclone/ESP
ESP
ESP
<20
NA
559
<1
NA
82
Secondary copper smelters
Norddeutsche Affinerie,
Germany
Huttenwerke Kayser,
Germany
Mansfelder Kupfer und
Messing, Germany
Unknown company,
Germany
Unknown company,
Germany
Unknown company,
Germany
Unknown company,
Germany
Peirce-Smith converter
Blast furnace
Peirce-Smith converters
Hearth furnace (for tin/lead)
Reverberatory anode furnace
Blast furnace
Shaft furnace
Rotary furnace
Rotary furnace
Rotary furnace
FF
Post-combustion, waste
heat boiler, FF
FF
Waste heat boiler, FF
Post-combustion, waste
heat boiler, cooler, FF, FF
with lime/coke injection
Post-combustion, dry
quench with secondary
off-gas, FF
FF
Gas cooling, FF
Gas cooling, FF, activated
carbonized lignite
adsorbent boxes
<500
<500
<100
<500
<100
<100-1,000
<100
<100
ESP = Electrostatic precipitator
FF = Fabric filter
NA = Not available
Source: Charles E. Napier Company, Ltd. (2002).
7-3
-------
In 1995, eight primary smelters were in operation in the United States, one of which
closed at the end of that year (Edelstein, 1995). Total refinery production was 1.60 million
metric tons in 1995, including 0.36 million metric tons from scrap material (Edelstein, 1995),
and 1.13 million metric tons in 1987 (USGS, 1997a). In 2000, four primary smelters of copper
were in operation in the United States, producing 1.61 million metric tons of copper (USGS,
2002).
CDD/CDF emission estimates for primary copper smelters were developed using the
stack test data from the two tested facilities in the United States. Conservatively assuming that
all nondetect values were present at one-half the detection limit, Environmental Risk Sciences,
Inc. (1995) calculated the annual TEQ emission to air to be less than 0.5 g I-TEQDF in 1995 for
the seven facilities (out of a total of eight) belonging to the National Mining Association.
Assuming that 1987 feed and processing materials were similar, 1987 releases can be estimated
at less than 0.5 g I-TEQDF as well. Because the number of facilities was reduced in 2000, the
national emissions estimate was reduced proportionally to 0.29 g I-TEQ. The activity level
estimates are assigned a high confidence rating and the emission factor estimates a medium
rating; therefore, the emission estimates are assigned a medium confidence rating. The activity
levels are based on comprehensive surveys. The emission factors are reasonably representative
of emissions from the source category.
7.1.2. Primary Magnesium Smelting and Refining
Oehme et al. (1989) reported that the production of magnesium can lead to the formation
of CDDs and CDFs. They estimated that 500 g of I-TEQDF were released to the environment in
wastewater and 6 g I-TEQDF were released to air annually from a magnesium production facility
in Norway; CDFs predominated, with a CDF-to-CDD concentration ratio of 10:1. At the time of
sampling, the magnesium production process involved formation of magnesium oxide (MgO)
from calcinated dolomite followed by a step in which magnesium chloride (MgCl2) was produced
by heating MgO/coke pellets in a shaft furnace in a pure chlorine atmosphere to about 700 to
800°C. The MgCl2 was then electrolyzed to form metallic magnesium and chloride. The
chloride excess from the MgCl2 process and the chloride formed during electrolysis were
collected by water scrubbers and directly discharged to the environment. The discharged
wastewater contained 200 to 500 ppm of suspended particulate matter. All but trace quantities of
the hexa through octa congeners were associated with the particulates; up to 10% of the tetra and
penta congeners were present in the water phase.
A study by the firm operating the facility (Musdalslien et al., 1998) indicated that
installation of a water treatment system had reduced annual emissions to water to less than 1 g
Nordic TEQ, and emissions to air had been reduced to less than 2 g Nordic TEQ. This study also
7-4
-------
presented results demonstrating that the carbon-reducing agent used in the MgCl2 production step
and the operating conditions of the shaft furnace greatly affected the formation of CDDs/CDFs.
Gases from the furnace were measured nine times over sampling periods of 6 to 8 hr. The
calculated emission factor to air (i.e., before any air pollution control device [APCD] controls)
ranged from 468 to 3,860 ng Nordic TEQ per kg of MgCl2 produced. The APCD controls
consisted of three water scrubbers, a wet electrostatic precipitator (ESP), and an incinerator.
From 1950 to 2000, the United States was the world's largest producer of metallic
magnesium (Kramer, 1995). In 1995, three magnesium production facilities were operating in
the United States. As in the Norwegian plant, an electrolytic process (electrolysis of MgCl2) was
used at the plants in Texas (capacity of 65,000 metric tons/yr) and Utah (capacity of 40,000
metric tons/yr) to recover metallic magnesium from MgCl2. However, these two facilities
reportedly used seawater and lake brines as the source of magnesium, and the procedures to
obtain and purify MgCl2 did not involve chlorinating furnaces and carbonized pellets (Lockwood
et al., 1981). A thermic process was used to recover magnesium from dolomite at the facility in
Washington (capacity of 40,000 metric tons/yr) (Kramer, 1995). In thermic processes, MgO, a
component of calcinated dolomite, is reacted with a metal such as silicon (usually alloyed with
iron) to produce metallic magnesium. In 2000, the Magnesium Corporation of America facility
near Rowley, UT, was the only operational magnesium smelting facility in the United States.
Monitoring of wastewater discharges from U.S. magnesium production facilities for
CDD/CDF content has not been reported. Wastewater discharges of CDDs/CDFs reported for
the Norwegian facility (Oehme et al., 1989), discussed in the previous paragraphs, are not
adequate to support development of wastewater emission factors for U.S. facilities because of
possible differences in the processes used to manufacture MgCl2 and pollution control
equipment.
Monitoring of air emissions for CDD/CDF content has been reported for the Magnesium
Corporation of America facility near Rowley, UT (Western Environmental Services and Testing,
Inc., 2000). The average emission rates (for three tests) reported for the melt reactor stack and
the cathode stack were 0.31 mg I-TEQDF/hr and 0.16 mg I-TEQDF/hr, respectively.
Emissions data were judged inadequate for developing national emission estimates for
1987 that could be included in the national inventory. The confidence in the degree to which the
one tested facility represents emissions from the other two U.S. facilities is low. However, an
estimate of the potential TEQ annual emissions for 1995 from U.S. primary magnesium
production facilities can be made by assuming that the average total emission factor for the Utah
facility measured in May 2000 (0.47 mg I-TEQDF/hr ) is representative of the other two facilities
for magnesium production. Specifically, if it is assumed that this facility operated for 24 hr/day
for 365 days in 1995, then the annual release in 1995 would have been 4.1 g I-TEQDF. If it is
7-5
-------
further assumed that this facility operated at 98% of its rated capacity of 40,000 metric tons/yr,
then the production-based emission factor would be 105 ng I-TEQDF/kg of magnesium produced.
Applying this emission factor to 98% of the industry's production capacity in 1995 (142,000
metric tons) yields a preliminary annual emissions estimate of 14.6 g I-TEQDF in 1995.
In 2000, the Magnesium Corporation of America facility near Rowley, UT, was the only
magnesium smelting facility operating in the United States. Production of primary magnesium at
this facility was 41,000 metric tons in 2000. Using the emission factor of 105 ng I-TEQDF/kg of
magnesium produced, the national estimate for dioxin releases in 2000 due to primary
magnesium smelting is 4.3 g I-TEQ/yr. The emission factor has a high confidence rating because
it was developed using data from this facility; therefore, the emissions estimate is assigned a high
confidence level.
7.1.3. Primary Nickel Smelting and Refining
Oehme et al. (1989) reported that certain primary nickel refining processes generate
CDDs and CDFs, primarily CDFs. Although the current low-temperature process used at a
Norwegian facility was estimated to result in releases to water of only 1 g I-TEQDF/yr, a high-
temperature (800°C) process to convert nickel chloride to nickel oxide that had been used for
17 yr at the facility is believed to have resulted in significant releases in earlier years, based on
the parts-per-billion levels of CDFs detected in aquatic sediments downstream of the facility.
According to Kuck (1995), the only nickel mining and smelting complex in the United
States (located in Oregon) had a capacity of 16,000 metric tons/yr. The facility had been on
standby since August 1993 and had no production in 1994. The facility restarted operations in
April 1995 and produced 8,290 metric tons of nickel that year. In 1998, the smelter closed
because of low nickel prices (USGS, 2002). Monitoring for discharges of CDDs/CDFs at this
facility has not been reported. Emissions of CDDs/CDFs were reported for a Norwegian facility
in the late 1980s, as discussed above (Oehme et al., 1989). The emissions information contained
in the Norwegian study is not adequate to support development of emission factors for the U.S.
facility for 1987 and 1995. Because the facility closed in 1998, emission estimates for 2000 for
primary nickel smelting are zero.
7.1.4. Primary Aluminum Smelting and Refining
No sampling of air emissions for the presence of CDDs/CDFs has been reported for this
industry. Lexen et al. (1993) reported that samples of filter powder and sludge from a lagoon at
the only primary aluminum production plant in Sweden showed no or little CDDs/CDFs.
Because the primary smelting process does not use chlorine, there is widespread belief that
7-6
-------
dioxin emissions from primary aluminum smelting facilities do not exist; therefore, no sampling
has been done.
In the primary aluminum smelting process, bauxite ore, a hydrated oxide of aluminum
consisting of 30 to 56% alumina (A12O3), is refined into alumina by the Bayer process. The
alumina is then shipped to a primary aluminum smelter for electrolytic reduction to aluminum.
Electrolytic reduction of alumina occurs in shallow rectangular cells, or pots, which are steel
shells lined with carbon. Carbon electrodes (petroleum coke mixed with a pitch binder)
extending into the pot serve as the anodes, and the carbon lining serves as the cathode. Three
types of pots are used: prebaked anode cell, horizontal stud Soderberg anode cell, and vertical
stud Soderberg anode cell. Most of the aluminum produced in the United States is produced
using the prebaked cells. Molten cryolite (Na3AlF6) functions as both the electrolyte and the
solvent for the aluminum. Aluminum is deposited on the cathode as molten metal (U.S. EPA,
1998a).
Prior to casting, the molten aluminum may be batch treated in reverberatory furnaces
(such as those used in secondary aluminum smelting) to remove oxides, gaseous impurities, and
active metals such as sodium and magnesium. One process consists of adding a flux of chloride
and fluoride salts and then bubbling chlorine gas through the molten mixture (U.S. EPA, 1998a).
U.S. production of primary aluminum was 3.343 million metric tons in 1987 and 3.375
million metric tons in 1995. In 1995, 13 companies operated 22 primary aluminum reduction
plants (USGS, 1997b, c). In 2000, 12 companies operated 23 primary aluminum reduction
plants, and primary aluminum smelters produced 3.7 million metric tons of aluminum (USGS,
2002). Because emission factors have not been developed for this sector, there are no emission
estimates for this category.
7.1.5. Primary Titanium Smelting and Refining
It has been suggested that carbochlorination processes used in this industry may be a
source of CDDs/CDFs (Bramley,1998; ERG, 1998). As discussed below, CDDs/CDFs have
been measured in titanium dioxide production sludges. A brief summary of the processes used in
this industry is presented in the following paragraphs.
In primary titanium smelting, titanium oxide ores and concentrates are chlorinated in
fluidized-bed reactors in the presence of coke at 925 to 1,010°C to form titanium tetrachloride
(TiCl4). The TiCl4 is separated from other chlorides by double distillation. The TiCl4 is then
either oxidized at 985°C to form pigment-grade titanium dioxide or reduced using sodium or
magnesium to form titanium sponge (i.e., metallic titanium) (Knittel, 1983). Titanium ingot is
produced by melting titanium sponge or scrap or a combination of both using electron beam,
7-7
-------
plasma, and vacuum arc methods. Scrap currently supplies about 50% of ingot feedstock
(Gambogi, 1996).
Titanium sponge is produced at two facilities in the United States, one in Albany, OR,
and the other in Henderson, NV. In 1995, the U.S. production volume of titanium sponge was
withheld to avoid disclosing proprietary data; domestic sponge capacity was 29,500 metric
tons/yr. In 1987, U.S. production of titanium sponge was 17,849 metric tons.
More than 90% of titanium dioxide is produced using the process described above.
Titanium dioxide pigment is used in paints, plastics, and paper products. In 1995, titanium
dioxide was produced at nine facilities in the United States. Production volumes in 1987 and
1995 were 821,000 and 1.8 million metric tons, respectively (Gambogi, 1996; USGS, 1997d). In
2000, four companies at eight facilities in seven states produced 1.44 million metric tons of
titanium dioxide (USGS, 2002).
Titanium dioxide production creates a sludge waste, and CDDs/CDFs have been
measured in these sludges (U.S. EPA, 2001c). For the most part, these sludges have been
disposed of in either on-site or off-site RCRA Subtitle D solid waste disposal facilities.
However, given the potential for leaching of the heavy metals from the sludge in the Subtitle D
landfill, EPA has listed this waste as hazardous waste under Subtitle C. These sludges are now
considered a hazardous waste under RCRA and must be disposed of in permitted landfills (U.S.
EPA, 200Ic). Therefore, they are not considered to cause environmental releases under the
definition in this document and are not included in the inventory.
7.2. SECONDARY NONFERROUS METAL SMELTING
Secondary smelters primarily engage in the recovery of nonferrous metals and alloys from
new and used scrap and dross. The principal metals of this industry, both in terms of volume and
value of product shipments, are aluminum, copper, lead, zinc, and precious metals (U.S. DOC,
1990a). Scrap metal and metal wastes may contain organic impurities such as plastics, paints,
and solvents. Secondary smelting and refining processes for some metals (e.g., aluminum,
copper, and magnesium) use chemicals such as sodium chloride, potassium chloride, and other
salts. The combustion of these impurities and chlorine salts in the presence of various types of
metal during reclamation processes can result in the formation of CDDs/CDFs, as evidenced by
their detection in the stack emissions of secondary aluminum, copper, and lead smelters (Aittola
et al., 1992; U.S. EPA, 1987a, 1997a).
7.2.1. Secondary Aluminum Smelters
Secondary aluminum smelters reclaim aluminum from scrap using two processes:
precleaning and smelting. Both processes may produce CDD/CDF emissions.
7-8
-------
Precleaning processes involve sorting and cleaning scrap to prepare it for smelting. Cleaning
processes that may produce CDD/CDF emissions use heat to separate aluminum from
contaminants and other metals. These techniques are "roasting" and "sweating." Roasting uses
rotary dryers with a temperature high enough to vaporize organic contaminants but not high
enough to melt aluminum. An example of roasting is the delacquering and processing of used
beverage cans. Sweating involves heating aluminum-containing scrap metal to a temperature
above the melting point of aluminum but below the melting temperature of other metals such as
iron and brass. The melted aluminum trickles down and accumulates in the bottom of the sweat
furnace and is periodically removed (U.S. EPA, 1997a).
After precleaning, the treated aluminum scrap is smelted and refined. This usually takes
place in a reverberatory furnace. Once smelted, flux is added to remove impurities. The melt is
demagged to reduce the magnesium content of the molten aluminum by adding chlorine gas. The
molten aluminum is then transferred to a holding furnace and alloyed to final specifications (U.S.
EPA, 1997a).
CDD/CDF emissions to air have been measured at seven U.S. secondary aluminum
operations. Five facilities were tested in 1995 and 1996 and two facilities were tested in 1992.
Four of the 1995 tests were conducted by EPA in conjunction with The Aluminum Association,
Inc., to identify emission rates from facilities with potentially maximum-achievable-control-
technology-grade operations and APCD equipment. The fourth test was performed by EPA (U.S.
EPA, 1995c). Results from two facilities tested by the California Air Resources Board (CARB)
in 1992 were presented in two confidential reports.
The first facility tested in 1995 was a top-charge melt furnace (Advanced Technology
Systems, Inc., 1995). During testing, the charge material to the furnace was specially formulated
to contain no oil, paint, coatings, rubber, or plastics other than incidental amounts. The TEQ
emission factor from such a clean charge, 0.27 ng TEQDF-WHO98/kg (0.26 ng I-TEQDF/kg) charge
material, would be expected to represent the low end of the normal industry range.
The second facility operated a sweat furnace to preclean the scrap and a reverberatory
furnace to smelt the precleaned aluminum (U.S. EPA, 1995c). Stack emissions were controlled
by an afterburner operated at 788°C. The TEQ emission factor for this facility was 3.37 ng
TEQDF-WHO98/kg (3.22 ng I-TEQDF/kg) aluminum produced.
The third facility employed a crusher/roasting dryer as a precleaning step followed by a
reverberatory furnace (Galson Corporation, 1995). The emissions from the two units were
vented separately. The exhaust from the crusher/dryer was treated with an afterburner and a
fabric filter (FF). The exhaust from the furnace passed through an FF with lime injection. Both
stack exhausts were tested, and the combined TEQ emission factor was 13.55 ng TEQDF-
WHO98/kg (12.95 ng I-TEQDF/kg) aluminum produced. Because the activity level of the facility
7-9
-------
at the time of sampling was treated as confidential business information, the calculated emission
factor was based on the reported typical production rates of the two operations: 26,000 Ib/hr for
the crusher/dryer and 6,700 Ib/hr for the furnace.
The fourth facility operated a scrap roasting dryer followed by a sidewell reverberatory
furnace (Weston,1996). The emissions from the two units were vented separately. Exhaust from
the dryer passed through an afterburner and a lime-coated FF. The exhaust from the furnace
passed through a lime-coated FF. Both stack exhausts were tested, and the combined TEQ
emission factor was 8.52 ng TEQDF-WHO98/kg (7.93 ng I-TEQDF/kg) of charge material.
The two facilities tested by CARB in 1992, which were reported in two confidential
reports (CARB, 1992a, b, as reported in U.S. EPA, 1997a) had TEQ emission factors of 55.68
and 23.44 ng TEQDF-WHO98/kg (52.21 and 21.67 ng I-TEQDF/kg) of scrap aluminum consumed.
One facility was equipped with a venturi scrubber; the other was assumed to be uncontrolled
(U.S. EPA, 1997a).
The seventh facility was tested in 1995 by Commonwealth Aluminum Corp. (1995). The
test involved sampling air emissions from a delaquering kiln. The facility was equipped with an
FF as the primary APCD.
The CDD/CDF congener and congener group emission factors derived from these stack
tests were used to represent emissions from secondary aluminum facilities operating in 2000,
1995, and 1987. Table 7-3 shows the TEQ emission factors for facilities operating in 2000, and
Table 7-4 shows those for 1995 and 1987. The 2000 emission factors do not include the results
from the two facilities tested by CARB (1992a, b) because it was assumed that all facilities
operating in 2000 were equipped with APCDs on all vents and stacks. The average congener and
congener group profiles are presented in Figure 7-1. The average of the TEQ emission factors
measured at the five tested facilities representative of the year 2000 is 5.17 ng TEQDF-WHO98/kg
(4.90 ng I-TEQDF/kg) of scrap feed. A low confidence rating is assigned to the average emission
factor for 2000 because it is based on the results of testing at only five facilities, and they may
not be representative of all facilities operating in the United States in that year.
The emission factors for facilities operating in 1995 and 1987 do include the results from
the two facilities tested by CARB in 1992. The average of the TEQ emission factor measured at
the seven tested facilities representative of the years 1995 and 1987 is 15.0 ng TEQDF-WHO98/kg
(14.05 ng I-TEQDF/kg) of scrap feed. A low confidence rating is assigned to the average
emission factor for 1995 and 1987 because it is based on the results of testing at only seven
facilities, and that may not be representative of all facilities operating in the United States in
those reference years.
7-10
-------
Table 7-3. CDD/CDF emission factors (ng/kg scrap feed) for secondary
aluminum smelters for 2000
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total I-TEQDFf
Total TEQDF-WHO98f
Mean facility emission factor
a
ND (0.01)
0.02
0.05
0.13
0.15
0.51
0.42
0.44
0.06
0.17
0.32
0.11
0.02
0.3
0.07
0.03
0.3
0.26
0.27
b
0.13
0.39
0.24
0.86
1.26
7.67
14.97
0.74
1.51
2.44
2.44
2.69
1.02
3.82
11.39
5.5
30.4
3.22
3.37
c
0.51
1.19
1.35
1.52
2.51
2.6
1.01
14.20
10.47
11.06
21.84
7.1
0.47
7.09
14.61
1.21
3.15
12.95
13.55
d
0.40
1.19
0.72
0.94
1.62
3.49
NR
10.46
12.52
6.98
8.14
2.74
0.14
4.30
3.18
0.45
1.23
8.09
8.68
e
0.01
0.02
0.02
0.03
0.05
0.10
NR
0.07
0.08
0.12
0.16
0.06
0.01
0.08
0.17
0.04
0.06
0.14
0.15
Overall
mean
emission
factor
0.21
0.56
0.48
0.70
1.12
2.87
5.47
5.18
4.93
4.15
6.58
2.54
0.33
3.12
5.88
1.45
7.03
4.93
5.20
aSource: Advanced Technology Systems, Inc. (1995).
bSource: U.S. EPA (1995c).
°Source: Galson Corporation (1995).
dSource: Weston(1996).
eSource: Commonwealth Aluminum Corp. (1995).
fTEQ calculations assume nondetect values were zero.
NR = Not reported
ND = Not detected (value in parenthesis is the detection limit)
For comparison purposes, the European Commission uses 22 ng I-TEQDF/kg scrap
aluminum as the typical emission factor for the European Dioxin Inventory (Quab and Fermann,
1997). Umweltbundesamt (1996) reported stack testing results for 25 aluminum smelters and
foundries in Germany. This study provided sufficient data to enable calculation of TEQ emission
factors for 11 of the tested facilities. The calculated emission factors ranged from 0.01 to 167 ng
7-11
-------
Table 7-4. CDD/CDF emission factors (ng/kg scrap feed) for secondary
aluminum smelters for 1995 and 1987
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDF8
Total TEQDF-WHO988
Mean facility emission factor
a
ND (0.01)
0.02
0.05
0.13
0.15
0.51
0.42
0.44
0.06
0.17
0.32
0.11
0.02
0.3
0.07
0.03
0.3
0.26
0.27
b
0.13
0.39
0.24
0.86
1.26
7.67
14.97
0.74
1.51
2.44
2.44
2.69
1.02
3.82
11.39
5.5
30.4
3.22
3.37
c
0.51
1.19
1.35
1.52
2.51
2.6
1.01
14.20
10.47
11.06
21.84
7.1
0.47
7.09
14.61
1.21
3.15
12.95
13.55
d
0.40
1.19
0.72
0.94
1.62
3.49
NR
10.46
12.52
6.98
8.14
2.74
0.14
4.30
3.18
0.45
1.23
8.09
8.68
e
0.01
0.02
0.02
0.03
0.05
0.10
NR
0.07
0.08
0.12
0.16
0.06
0.01
0.08
0.17
0.04
0.06
0.14
0.15
f
1.97
7.1
4.26
5.3
5.3
28.9
33.2
23.2
33.8
48
46.1
46.1
22
39
122
27.1
60.5
52.21
55.68
f
0.85
3.64
2.82
4.12
2.02
19.3
24.3
4.84
1.18
23.3
17.6
16.9
1.35
16
42.6
6.2
29.5
21.68
23.45
Overall
mean
emission
factor
0.54
1.94
1.35
1.84
1.84
8.94
14.78
7.71
8.52
13.15
13.80
10.81
3.57
10.08
27.71
5.79
17.88
14.08
15.02
aSource: Advanced Technology Systems, Inc. (1995).
bSource: U.S. EPA (1995c).
°Source: Galson Corporation (1995).
dSource: Weston(1996).
eSource: Commonwealth Aluminum Corp. (1995).
fSource: CARB (1992a, b), as reported in U.S. EPA (1997a).
BTEQ calculations assume nondetects were zero.
NR = Not reported
ND = Not detected (value in parenthesis is the detection limit)
I-TEQDF/kg of scrap feed. Three facilities had emission factors exceeding 100 ng I-TEQDF/kg,
and two facilities had emission factors of less than 1 ng I-TEQDF/kg. The mean emission factor
for the 11 facilities was 42 ng I-TEQDF/kg.
Approximately 727,000 metric tons of scrap aluminum were consumed by 67 secondary
aluminum smelters in the United States in 1987 (U.S. DOC, 1995c). In 1995, consumption of
7-12
-------
Figure 7-1. Congener and congener group profiles for air emissions from
secondary aluminum smelters.
Sources: U.S. EPA (1995c); Galson Corporation (1995).
scrap aluminum by the 76 facilities that composed the secondary aluminum smelting industry
had nearly doubled, to 1.3 million metric tons (USGS, 1997e; The Aluminum Association, Inc.
1997). In 2000, secondary aluminum smelters consumed 1.6 million metric tons of scrap
aluminum (USGS, 2002). A high confidence rating is assigned to these production estimates
7-13
-------
because they are based on government survey data. Applying the 15.0 ng TEQDF-WHO98/kg
(14.05 ng I-TEQDF/kg) emission factor to the metric tons of scrap aluminum processed in 1987
and 1995 yields estimated annual emissions of 10.9 g TEQDF-WHO98 (10.21 g I-TEQ) in 1987
and 19.5 TEQDF-WHO98 (18.27 g I-TEQDF) in reference year 1995. Applying the emission factor
of 5.17 ng TEQDF-WHO98/kg (4.9 ng I-TEQDF/kg) to the metric tons of scrap aluminum processed
in 2000 yields estimated annual emissions of 8.27 g TEQDF-WHO98 (7.84 g I-TEQ) in reference
year 2000. These emission estimates are assigned a low confidence rating because the rating
given to the emission factor was low.
It should be noted that a significant amount of scrap aluminum is also consumed by other
segments of the aluminum industry. However, this scrap is generally from metal manufacturing
processes, including metal and alloy production (e.g., borings, turnings, and dross), rather than
old scrap that results from recycling of consumer products (e.g., cans, radiators, auto shredders).
In 1995, integrated aluminum companies consumed 1.4 million metric tons of scrap aluminum,
and independent mill fabricators consumed 0.68 million metric tons (USGS, 1997e).
7.2.2. Secondary Copper Smelters
Secondary copper smelting is part of the scrap copper, brass, and bronze reprocessing
industry. Brass is an alloy of copper and zinc; bronze is an alloy of copper and tin. Facilities in
this industry fall into three general classifications: secondary smelting, ingot making, and
remelting. Similar processing equipment may be used at all three types of facilities, so the
distinguishing features are not immediately apparent (U.S. EPA, 1994b).
The feature that distinguishes secondary smelters from ingot makers and remelters is the
extent to which pyrometallurgical purification is performed. A typical charge at a secondary
smelter may contain from 30 to 98% copper. The secondary smelter upgrades the material by
reducing the quantity of impurities and alloying materials, thereby increasing the relative
concentration of copper. This degree of purification and separation of the alloy constituents does
not occur at ingot makers and remelters. Feed material to a secondary copper smelter is a
mixture of copper-bearing scrap such as tubing, valves, motors, windings, wire, radiators,
turnings, mill scrap, printed circuit boards, telephone switching gear, and ammunition casings.
Nonscrap items such as blast furnace slags and drosses from ingot makers or remelters may
represent a portion of the charge. Secondary smelter operators use a variety of processes to
separate the alloy constituents. Some purify the scrap in the reductive atmosphere of a blast
furnace and then purify the charge in the oxidizing atmosphere of a converter; others perform all
purification by oxidation in top-blown rotary converters or in reverberatory furnaces (U.S. EPA,
1994b).
7-14
-------
Ingot maker operators blend and melt scrap copper, brass, and bronze of various
compositions to produce a specification brass or bronze ingot. When necessary, they add ingots
of other metals (e.g., zinc or tin) to adjust the metallurgy of the final product. The feed materials
for ingot makers contain relatively high amounts of copper. Examples of feed materials include
copper tubing, valves, brass and bronze castings, ammunition shell casings, and automobile
radiators. "Fire-refined" anode copper or cathode copper may also be charged. Items such as
motors, telephone switchboard scrap, circuit board scrap, and purchased slags are not used by
ingot makers. The reductive step (melting in a reducing atmosphere, as in a blast furnace) that
some secondary smelters employ is not used by ingot makers. Ingot makers do, however, use
some of the other types of furnaces used by secondary smelters, including direct-fired converters,
reverberatory furnaces, and electric induction furnaces (U.S. EPA, 1994b).
Remelting facilities do not conduct substantial purification of the incoming feeds. These
facilities typically melt the charge and then cast or extrude a product. The feeds to a remelter are
generally alloy material of approximately the desired composition of the product (U.S. EPA,
1994b).
7.2.2.1. Emissions Data
Stack emissions of CDDs/CDFs from a secondary copper smelter were measured by EPA
during 1984 and 1985 as part of the National Dioxin Tier 4 Study (U.S. EPA, 1987a). The
facility chosen for testing was estimated to have a high potential for CDD/CDF emissions
because of the abundance of chlorinated plastics in the feed. This facility ceased operations in
1986. The facility was chosen for testing by EPA because the process technology and APCD
equipment in place were considered typical for the source category.
During operations, copper and iron-bearing scrap were fed in batches to a cupola blast
furnace, which produced a mixture of slag and black copper. Approximately 4 to 5 tons of
metal-bearing scrap were fed to the furnace per charge, with materials typically being charged 10
to 12 times per hour. Coke fueled the furnace and represented approximately 14% (by weight) of
the total feed. During the stack tests, the feed consisted of electronic telephone scrap and other
plastic scrap, brass and copper shot, iron-bearing copper scrap, precious metals, copper-bearing
residues, refinery by-products, converter furnace slag, anode furnace slag, and metallic floor-
cleaning material. The telephone scrap made up 22% (by weight) of the feed and was the only
scrap component that contained plastic materials. Oxygen-enriched combustion air for
combustion of the coke was blown through tuyeres (nozzles) at the bottom of the furnace. At the
top of the blast furnace were four natural gas-fired afterburners to aid in completing combustion
of the exhaust gases. Particulate emissions were controlled by FFs, and the flue gas was then
discharged into a common stack. The estimated emission factors derived for this site are
7-15
-------
presented in Table 7-5. The emission factors are based on the total weight of scrap fed to the
furnace. Based on the measured congener and congener group emission factors, the TEQ
emission factor is 779 ng I-TEQDF/kg (810 ng TEQDF.WHO98/kg) of scrap metal smelted. Figure
7-2a presents the congener group profile based on these emission factors.
In 1992, stack testing of the blast furnace emissions of a secondary smelter located in
Philadelphia, PA (Franklin Smelting and Refining Co.), was conducted by Applied Geotechnical
& Environmental Services Corporation (AGES, 1992). Like the facility tested by EPA during
1984 and 1985, this facility processed low-purity copper-bearing scrap, telephone switch gear,
and slags, as well as higher-copper-content materials (U.S. EPA, 1994b). The facility used a
blast (cupola-type) furnace coupled with a pair of rotary converters to produce blister copper.
The blast furnace used coke as both the fuel and the agent to maintain a reducing atmosphere.
The black copper-slag mixture from the blast furnace was charged to the rotary converters for
further refining with the aid of oxygen, sand, and oak logs (AGES, 1992; U.S. EPA, 1994b). The
APCD equipment installed on the blast furnace included an afterburner, a cooling tower, and an
FF. During testing, the afterburner was reported to be operating erratically and was particularly
low during one of the two sampling episodes. Stack gas flow was also low during both sampling
episodes because one or more FF compartments were inoperable (AGES, 1992). The estimated
emission factors derived for this site from the AGES results are presented in Table 7-5. The
emission factors were based on the total weight of scrap fed to the blast furnace. The TEQ
emission factor was 16,618 ng I-TEQDF/kg (16,917 ng TEQDF-WHO98/kg) of scrap. Figure 7-2b
presents the congener and congener group profiles based on these emission factors.
In 1991, stack testing of the rotary furnace stack emissions of a secondary smelter
(Chemetco, Inc.) located in Alton, IL, was conducted by Sverdrup Corp. (1991). The Chemetco
facility used four tap-down rotary (i.e., oxidizing) furnaces. Furnace-processed gas emissions
were controlled by a primary quencher and a venturi scrubber. The feed was relatively high-
purity copper scrap containing minimal, if any, plastics. The same manufacturing process and
APCD equipment were in place in 1987 and 1995 (U.S. EPA, 1994b). Because this facility
operated under oxidizing rather than reducing conditions and processed relatively high-purity
scrap, the potential for CDD/CDF formation and release was expected to be dramatically
different from that of the two tested facilities reported above. The estimated emission factors
derived for this site from the results of Sverdrup Corp. (1991) are presented in Table 7-5. The
emission factors were based on the total weight of scrap feed going to the furnace. The TEQ
emission factor was 3.60 ng I-TEQDF/kg (3.66 ng TEQDF-WHO98/kg) of scrap.
7-16
-------
Table 7-5. CDD/CDF mean emission factors (ng/kg scrap feed) for secondary
copper smelters
Congener/
congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total I-TEQDFe
Total TEQDF-WHO98e
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
EPA Tier 4
facility3'"
127
NR
NR
NR
NR
NR
1,350
2,720
NR
NR
NR
NR
NR
NR
NR
NR
2,520
779f
810f
736
970
1,260
2,080
1,350
13,720
8,640
4,240
3,420
2,520
38,936
Franklin smelting
facility0
227
846
1,476
1,746
2,132
17,065
55,668
4,457
9,455
5,773
70,742
20,524
5,362
12,082
37,251
7,570
82,192
16,618
16,917
14,503
30,248
55,765
38,994
55,668
108,546
71,136
164,834
66,253
82,192
688,139
Chemetco smelting
facility"
ND (0.05)
0.21
0.39
0.70
1.26
8.95
22.45
2.11
1.47
2.63
7.30
2.15
4.06
0.27
11.48
2.74
21.61
3.59
3.66
3.05
5.19
9.62
16.71
22.45
46.42
27.99
27.96
23.38
21.61
204.38
aNo nondetect values were reported for 2,3,7,8-TCDD, 2,3,7,8-TCDF, or any congener group in the three test runs.
bSource: U.S. EPA(1987a).
cSource: AGES (1992).
dSource: Sverdrup Corp. (1991).
eTEQ calculations assume nondetect values were zero.
Estimated using the measured data for 2,3,7,8-TCDD, 2,3,7,8-TCDF, OCDD, and OCDF and congener group
emissions (i.e., for the penta-, hexa-, and hepta-CDDs and CDFs, it was assumed that the measured emission factor
within a congener group was the sum of equal emission factors for all congeners in that group, including non-
2,3,7,8-substiruted congeners).
NR = Not reported
ND = Not detected (value in parenthesis is the detection limit)
7-17
-------
Figure 7-2a. Congener group profile for air emissions from a secondary
copper smelter.
Source: U.S. EPA(1987d).
Only limited data on emissions from secondary copper smelters are reported in the
European Dioxin Inventory (LUA, 1997). TEQ emission factors reported for German shaft
furnaces/converters and reverberatory furnaces range from 5.6 to 110 ng I-TEQDF/kg and from
0.005 to 1.56 ng I-TEQDF/kg, respectively. Emission factors reported for two smelter and casting
furnaces in Sweden in which relatively clean scrap is used as input are 0.024 and 0.04 ng I-
TEQDF/kg. A smelter in Austria is reported to have a TEQ emission factor of 4 ng I-TEQDF/kg.
The minimum, typical, and maximum default emission factors selected in LUA (1997) are 5, 50,
and 400 ng I-TEQDF/kg, respectively.
In the 2002 Environment Canada report on CDD/CDF emissions from the base metals
smelting sector (Charles E. Napier Company, Ltd., 2002), three secondary copper smelters were
identified (see Table 7-2). CDD/CDF emission concentrations were reported as ranging from
less than 100 to less than 500 pg I-TEQ/dscm.
7-18
-------
I! n4
Figure 7-2b. Congener and congener group profiles for a closed secondary
copper smelter.
Source: AGES (1992).
7-19
-------
7.2.2.2. Activity Level Information
In 1987, four secondary copper smelters were in operation: Franklin Smelting and
Refining Co. (Philadelphia, PA), Chemetco, Inc. (Alton, IL), Southwire Co. (Carrollton, GA),
and a facility located in Gaston, SC, that was owned by American Telephone and Telegraph
(AT&T) until 1990, when it was purchased by Southwire Co. In 1987, estimated smelter
capacities were 13,600 metric tons for the Franklin facility, 120,000 metric tons for the
Chemetco facility, 48,000 metric tons for the Southwire facility, and 85,000 metric tons for the
AT&T facility (telephone conversation on August 17, 1999, between D. Edelstein, U.S.
Geological Survey, and G. Schweer, Versar, Inc.). In 1995, only three of these four facilities
were still in operation. The Southwire facility in Gaston was closed in January 1995. The
Franklin facility ceased operations in August 1997. Estimated smelter capacities in 1995 were
16,000 metric tons for the Franklin facility, 135,000 metric tons for the Chemetco facility, and
92,000 metric tons for the Southwire Georgia facility (Edelstein, 1999). In May 2000, the
Southwire Co. closed its Georgia facility and ceased operations (Edelstein, 2000). In November
2001, Chemetco closed its facility and ceased operations (Edelstein, 2001).
According to Edelstein (2001), smelters and refineries consumed 255,000 metric tons of
purchased copper-based scrap in 2000 and 196,000 metric tons in 2001. Assuming Chemetco
was the sole smelter facility operating in 2001, and that it operated for 10 of 12 months in 2001,
its estimated annual consumption of copper-based scrap would be 235,000 metric tons per year.
Assuming Chemetco's annual consumption rate did not change from 2000 to 2001, the estimated
consumption of copper-based scrap for the Southwire Co. in 2000 was 20,000 metric tons.
7.2.2.3. Emission Estimates
Although little research has been done to define the CDD/CDF formation mechanisms in
secondary copper smelting operations, two general observations have been made (Buekens et al.,
1997). First, the presence of chlorinated plastics in copper scraps used as feed for smelters is
believed to increase CDD/CDF formation. Second, the reducing or pyrolytic conditions in blast
furnaces can lead to high CDD/CDF concentrations in the furnace process gases. As noted in
Section 7.2.2.1, two of the U.S. facilities that have been tested (U.S. EPA, 1987a; AGES, 1992)
had the following characteristics: both processed low-purity copper-bearing scrap containing
significant quantities of plastics and telephone switch gears, and both used blast furnaces. The
APCD equipment at both facilities consisted of an afterburner, a cooling tower (Franklin facility
only), and an FF (U.S. EPA, 1994b). The other tested U.S. facility used oxidizing rather than
reducing conditions and processed relatively high-purity scrap (Sverdrup Corp., 1991).
7-20
-------
Annual TEQ emissions for 1987, 1995, and 2000 were derived as the sum of the TEQ
emissions for each secondary copper facility in operation during the reference years. The
following discussion summarizes the procedure used to estimate annual TEQ air emissions.
The Franklin Smelting facility operated in 1987 and 1995 but not in 2000. The TEQ
emission factor measured at this facility in 1992 is assumed to be representative of the TEQ
emission factors in 1987 and 1995. Combining this emission factor (16,618 ng I-TEQDF/kg
[16,917 ng TEQDF-WHO98/kg] of scrap feed) with the estimated smelter capacities (data are not
available on the amount of scrap processed) for this facility in 1987 (13,600,000 kg) and 1995
(16,000,000 kg) yields TEQ emission estimates of 226 g I-TEQDF (230 g TEQDF-WHO98) in 1987
and 266 g I-TEQDF (271 g TEQDF-WHO98) in 1995. This facility ceased operations in 1997.
The Chemetco facility operated in 1987, 1995, and 2000. Similarly, for purposes of this
report, the TEQ emission factor for the Chemetco facility is considered to be representative of the
TEQ emission factor for this facility for 1987, 1995, and 2000. Combining this emission factor
(3.60 ng I-TEQDF/kg [3.66 ng TEQDF-WHO98/kg] of scrap feed) with the estimated smelter
capacities of 120,000,000 kg in 1987 and 135,000,000 kg in 1995 yields TEQ estimates of 0.43 g
I-TEQDF (0.44 g TEQDF-WHO98) in 1987 and 0.49 g I-TEQDF (0.49 g TEQDF-WHO98) in 1995.
Combining the same emission factor with the scrap consumption for this facility in 2000
(235,000,000 kg) yields a TEQ estimate of 0.85 g I-TEQDF (0.86 g TEQDF-WHO98) for 2000.
The facility in Gaston, SC, was in operation during 1987 but ceased operations in 1995.
Prior to 1990, when this facility was owned by AT&T, the plant processed a great deal of high-
plastics-content scrap (such as whole telephones). This scrap was fed to a pyrolysis unit prior to
entering the blast furnace. In addition to a blast furnace, the facility also had an oxidizing
reverberatory furnace for processing higher-purity scrap. The facility had separate FFs for the
blast furnace, the converters, and the reverberatory furnace (U.S. EPA, 1994b). Because this
facility processed low-purity, high-plastics-content scrap in 1987, and presumably processed
much of this in the reducing atmosphere of a pyrolysis unit and blast furnace, the average of the
TEQ emission factors for the Tier 4 EPA-tested facility (U.S. EPA, 1987a) and the Franklin
facility (8,700 ng I-TEQDF/kg [8,860 ng TEQDF-WHO98/kg]) was used to estimate potential
emissions in 1987 of 740 g I-TEQDF (753 g TEQDF-WHO98) (assuming an activity level of
85,000,000 kg). This activity level is the estimated capacity of the facility; data were not
available on the amount of scrap processed.
The Southwire facility had both a blast furnace and a reverberatory furnace. In 1992,
approximately 50% of incoming scrap was processed in each furnace (U.S. EPA, 1994b). Unlike
the Franklin, Chemetco, and Gaston secondary copper smelters, the Southwire facility stopped
processing plastic-coated scrap in the 1970s. In addition, this facility had a more complex APCD
system, which may have reduced the formation and release of CDDs/CDFs. The blast furnace-
7-21
-------
processed gases passed through an afterburner (871°C), U-tube coolers, and an evaporative spray
system before entering the FF at a temperature of 107 to 191°C. For these reasons, EPA has
determined that the existing emissions data for secondary smelters cannot reliably be used to
generate a quantitative estimate of potential emissions during 1987, 1995, or 2000 for this
facility.
Total secondary copper smelter emissions for 1987 are the sum of the Franklin smelting
facility emissions (271 g TEQDF-WHO98 [266 g I-TEQDF]), the Chemetco smelter facility (0.44 g
TEQDF-WHO98 [0.43 g I-TEQDF]) and the Gaston, SC, facility (753 g TEQDF-WHO98 [740 g I-
TEQDF]). Total secondary copper smelter emissions for 1987 are 983.44 g TEQDF-WHO98
(966.43 gI-TEQDF).
Total secondary copper smelter emissions for 1995 are the sum of the Franklin smelting
facility emissions (271 g TEQDF-WHO98 [266 g I-TEQDF]) and the Chemetco smelter facility
(0.49 g TEQDF-WHO98 [0.49 g TEQDF-WHO98]). Total secondary copper smelter emissions for
1995 are 271.49 g TEQDF-WHO98 (266.49 I-TEQDF).
The Chemetco smelter provides the TEQ emissions estimate for the year 2000. Total
secondary copper smelter emissions for 2000 are 0.86 g TEQDF-WHO98 (0.85 g I-TEQDF).
A high confidence rating is assigned to the production and consumption estimates
because they are based on government survey data. A low confidence rating is assigned to the
TEQ emission estimates because they are based on limited measurements made at three smelters,
one of which was not in operation in 1987 or 1995.
It should be noted that a significant amount of scrap copper is consumed by other
segments of the copper industry. In 1995 and 2000, brass mills and wire-rod mills consumed
886,000 and 1,070,000 metric tons of copper-based scrap, respectively; foundries and
iscellaneous manufacturers consumed 71,500 and 96,200 metric tons, respectively (USGS,
1997e; Edelstein, 2001). As noted above, however, these facilities generally do not conduct any
significant purification of the scrap. Rather, the scrap consumed is already of alloy quality, and
processes employed typically involve only melting, casting, and extruding. Thus, the potential
for formation of CDDs/CDFs is expected to be much less than the potential during secondary
smelting operations.
7.2.3. Secondary Lead Smelters
The secondary lead smelting industry produces elemental lead through the chemical
reduction of lead compounds in a high-temperature furnace (1,200 to 1,260°C). Smelting is
performed in reverberatory, blast, rotary, or electric furnaces. Blast and reverberatory furnaces
are the most common types of smelting furnaces used by the 23 facilities that make up the
current secondary lead smelting industry in the United States. Of the 45 furnaces at these 23
7-22
-------
facilities, 15 are reverberatory furnaces, 24 are blast furnaces, 5 are rotary furnaces, and 1 is an
electric furnace. The electric furnace and 11 of the 24 blast furnaces are co-located with
reverberatory furnaces, and most share a common exhaust and emissions control system (U.S.
EPA, 1994c).
Furnace charge materials consist of lead-bearing raw materials, lead-bearing slag and
drosses, fluxing agents (blast and rotary furnaces only), and coke. Scrap motor vehicle lead-acid
batteries represent about 90% of the lead-bearing raw materials at a typical lead smelter. Fluxing
agents consist of iron, silica sand, and limestone or soda ash. Coke is used as fuel in blast
furnaces and as a reducing agent in reverberatory and rotary furnaces. Organic emissions from
co-located blast and reverberatory furnaces are more similar to the emissions of a reverberatory
furnace than to those of a blast furnace (U.S. EPA, 1994c).
In 1987, the lead smelting industry consisted of 24 facilities producing 0.72 million
metric tons of lead (U.S. EPA, 1994c). In 1995, there were 23 companies producing 0.97 million
metric tons (USGS, 1997e), and in 2000 there were 27 secondary lead smelters in operation in
the United States producing 1.02 million metric tons (USGS, 2002). In 1995, the total annual
production capacity of the 23 companies that made up the U.S. lead smelting industry was 1.36
million metric tons. Blast furnaces not co-located with reverberatory furnaces accounted for 21%
of capacity (0.28 million metric tons). Reverberatory furnaces and blast and electric furnaces co-
located with reverberatory furnaces accounted for 74% of capacity (1.01 million metric tons).
Rotary furnaces accounted for the remaining 5% of capacity (0.07 million metric tons) (U.S.
EPA, 1994c).
Actual production volume statistics by furnace type were not available. However, if it is
assumed that the total actual production volume of the industry reflects the production capacity
breakdown by furnace type, then the estimated actual production volumes of blast furnaces (not
co-located), reverberatory and co-located blast/electric and reverberatory furnaces, and rotary
furnaces were 0.15, 0.53, and 0.04 million metric tons, respectively, in 1987; 0.2, 0.72, and 0.05
million metric tons, respectively, in 1995; and 0.29, 1, and 0.07 million metric tons, respectively,
in 2000.
A report commissioned by Environment Canada (Charles E. Napier Company, Ltd.,
2000) reviewed published literature and other information on the dioxin/furan formation
mechanisms; dioxin/furan emissions; emission control technology, including cost; and
dioxin/furan published emission standards pertinent to steel production processes of plants
in Canada. The report included four facilities identified as primary lead smelters. CDD/CDF
emission concentrations were reported to range from less than 100 to less than 1,000 pg I-
TEQ/dscm.
7-23
-------
CDD/CDF emission factors were estimated for secondary lead smelters using the results
of emission tests performed by EPA at three smelters (a blast furnace [U.S. EPA, 1995d], a co-
located blast/reverberatory furnace [U.S. EPA, 1992c], and a rotary kiln furnace [U.S. EPA,
1995e]). The air pollution control systems at the three tested facilities consisted of both FFs and
scrubbers. Congener-specific measurements were made at both APCD exit points at each
facility. Table 7-6 presents the congener and congener group emission factors for the FF and the
scrubber for each site. Figure 7-3 presents the corresponding profiles for the FF emissions from
the tested blast furnace and reverberatory furnace. For the facilities in operation in 1995, all 23
smelters employed FFs, with only 9 employing scrubber technology. Facilities with scrubbers
accounted for 14% of the blast furnace (not co-located) production capacity, 52% of the
reverberatory and co-located furnace production capacity, and 57% of the rotary furnace
production capacity. TEQ emission factors (ng TEQ/kg lead produced when nondetect values
are set equal to zero) from the reported data for each of the three furnace configurations,
presented as a range reflecting the presence or absence of a scrubber, are
• Blast furnace: 0.64 to 8.81 ng TEQDF-WHO98/kg (0.63 to 8.31 ng I-TEQDF/kg)
• Reverberatory/co-located furnace: 0.05 to 0.42 ng TEQDF-WHO98/kg (0.05 to 0.41 ng
I-TEQDF/kg)
• Rotary furnace: 0.24 to 0.66 ng TEQDF-WHO98/kg (0.24 to 0.66 ng I-TEQDF/kg)
If these ranges of emission rates are assumed to be representative of those at nontested
facilities with the same basic furnace configuration, with and without scrubbers, then combining
these emission rates with the estimated production volumes derived above and the percentage of
each configuration type that have scrubbers yields the estimated air emissions shown in Table 7-7
for reference years 1987, 1995, and 2000.
A medium confidence rating is assigned to the emission factors because stack test data
were available for 3 of the 27 smelters operating in the United States (of which only 16 were in
operation as of December 1993), and the stack test data used represent the three major furnace
configurations. The activity level estimate has been assigned a medium confidence rating
because, although it is based on a U.S. Department of Commerce estimate of total U.S.
production, no production data were available on a furnace type or furnace configuration basis.
Therefore, a medium confidence rating is assigned to the emission estimates.
7-24
-------
Table 7-6. CDD/CDF emission factors (ng/kg lead produced) for secondary
lead smelters"
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
(nondetect set to zero)
Total TEQDF-WHO98
(nondetect set to zero)
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
(nondetect set to zero)
Blast furnaceb
Before
scrubber/
FF
2.11
0.99
0.43
0.99
1.55
2.06
1.4
8.73
3.88
6.65
5.83
1.67
0.11
2.06
2.34
0.63
1.39
9.53
33.29
8.31
8.81
74.33
39.29
20.05
4.2
1.39
145.71
69.59
19.73
4.74
1.39
380.42
After
scrubber/
FF
0.25
0.03
0
0.03
0.03
0.08
0.39
0.93
0.43
0.36
0.37
0.11
0
0.11
0.19
0.06
0.18
0.81
2.74
0.63
0.64
7.39
1.73
0.81
9.72
0.18
17.34
3.45
1.02
0.11
0.18
41.93
Blast/reverberatory
furnace0
Before
scrubber/
FF
0
0
0
0
0
0.1
0.57
1.46
0.24
0.31
0.63
0.19
0
0.15
0.48
0
0.29
0.67
3.75
0.42
0.42
0.97
0.15
0.14
0.09
0.57
8.21
3.07
1.14
0.72
0.29
15.35
After
scrubber/
FF
0
0
0
0
0
0.06
0.55
0.49
0.02
0
0
0
0
0
0
0
0
0.61
0.51
0.05
0.05
1.58
0.16
0.02
0.09
0.55
4.71
0.36
0.19
0.01
0.00
7.67
Rotary kiln"
Before
scrubber/
FF
0.1
0.01
0
0
0
0
0.24
0.40
0.14
0.14
0.11
0.02
0.04
0
0.03
0
0
0.35
0.88
0.24
0.24
3.4
0.29
0.1
0.01
0.24
10.82
1.69
0.15
0.05
0
16.75
After
scrubber/
FF
0.24
0
0
0
0
0.22
2.41
1.2
0.4
0.46
0.27
0.1
0.13
0
0.13
0
0
2.87
2.69
0.67
0.66
7.9
0.27
0.23
0.29
2.41
28.57
5.04
0.73
0.14
0
45.58
aExcept where noted, emission factors were calculated assuming nondetect values were zero.
bSource: U.S. EPA(1995d).
cSource: U.S. EPA (1992c).
dSource: U.S. EPA (1995e).
FF = Fabric filter
7-25
-------
Figure 7-3. Congener and congener group profiles for air emissions from
secondary lead smelters. Profiles are for emissions from fabric filters; nondetect
values set equal to zero.
Sources: U.S. EPA(1992c, 1995d, e).
7-26
-------
Table 7-7. Estimated annual TEQ emissions (g TEQ)a
Configuration
Blast furnaces w/scrubbers
Blast furnaces w/o
scrubbers
Reverberatory furnaces
w/scrubbers
Reverberatory furnaces w/o
scrubbers
Rotary furnaces
w/scrubbers
Rotary furnaces w/o
scrubbers
TOTAL
Ref. year 1987
TEQDF-
WHO98
0.013
1.136
0.014
0.106
0.015
0.004
1.288
I-TEQDF
0.013
1.072
0.014
0.104
0.015
0.004
1.222
Ref. year 1995
TEQDF-
WHO98
0.018
1.515
0.019
0.145
0.019
0.005
1.721
I-TEQDF
0.018
1.429
0.019
0.142
0.019
0.005
1.632
Ref. year 2000
TEQDF-
WH098
0.026
2.197
0.026
0.202
0.026
0.007
2.484
I-TEQDF
0.026
2.073
0.026
0.197
0.026
0.007
2.355
Calculated using emission factors based on nondetect values set equal to zero.
7.3. PRIMARY FERROUS METAL SMELTING/REFINING
Iron is manufactured from its ores (magnetic pyrites, magnetite, hematite, and carbonates
of iron) in a blast furnace, and the iron obtained from this process is further refined in steel plants
to make steel. The primary production of iron and steel involves two operations identified by
European researchers as potential emission sources of CDDs/CDFs: iron ore sinter production
and coke production. Each of these potential sources is discussed in the following subsections.
7.3.1. Sinter Production
At some iron manufacturing facilities, iron ores and waste iron-bearing materials undergo
sintering to convert the materials to usable feed for the blast furnace. In the sintering process,
iron ore fines and waste materials are mixed with coke fines, and the mixture is placed on a grate
that is then heated to a temperature of 1,000 to 1,400°C. The heat generated during combustion
sinters the small particles. Iron-bearing dusts and slags from processes in the steel plant are the
types of iron-bearing waste materials used as a feed mix for the sintering plant (Knepper, 1981;
Capes, 1983; U.S. EPA, 1995b).
7-27
-------
Several European investigators have reported that iron ore sintering plants are major
sources of airborne emissions of CDDs/CDFs (Rappe, 1992a; Lexen et al., 1993; Lahl, 1993,
1994). Lahl reported that the practice of recycling dusts and scraps from other processes in the
steel plant for use in the sintering plant introduces traces of chlorine and organic compounds that
generate the CDDs/CDFs found in these plants.
Organic compounds that are potential precursors to CDD/CDF formation come primarily
from oil, which is found in mill scale, as well as from some blast furnace sludges that are used as
part of the sinter feed mixture. Most U.S. plants limit the amount of oil because it increases
emissions of volatile organic compounds and may create a fire hazard. In addition, plants with
FFs must limit the oil content because the oil tends to blind the FFs. Typical oil content of the
feed at U.S. sintering plants ranges from 0.1 to 0.75% (Calcagni et al., 1998).
Sintering plants in Sweden have been reported to emit up to 3 ng I-TEQDF/Nm3 stack gas,
or 2 to 4 g I-TEQDF/yr (Rappe, 1992a; Lexen et al., 1993). Bremmer et al. (1994) reported the
results of stack testing at three iron ore sintering plants in the Netherlands. One facility equipped
with wet scrubbers (WSs) had an emission factor of 1.8 ng I-TEQDF/dscm (at 11% oxygen). The
other two facilities, both equipped with cyclones, had emission factors of 6.3 and 9.6 ng I-
TEQDF/dscm (at 7% oxygen). Lahl (1993, 1994) reported stack emissions for sintering plants in
Germany (after passage through mechanical filters and ESPs) ranging from 3 to 10 ng I-
TEQDF/Nm3. A compilation of emission measurements by the German Federal Environmental
Agency indicated stack emission concentrations ranging from 1.2 to 60.6 ng I-TEQDF/m3 (at 7%
oxygen); the majority of emissions in 1996 were around 3 ng I-TEQDF/m3 (Umweltbundesamt,
1996).
The report commissioned by Environment Canada in 2000 to review steel production
processes in Canadian plants (Charles E. Napier Company, Ltd., 2000) included information on
emissions from iron sintering. For iron sintering, the CDD/CDF emissions from one facility, the
Stelco Hilton Works sintering plant, were assumed to be representative of the 1998 sinter
production. The average emission rate was 19.9 ng I-TEQ/day. Applying a production rate of
1,143 metric tons/day yields a mass emission factor of 17.4 ng I-TEQ/kg of sinter.
EPA conducted tests at two of the nine U.S. sintering plants operating in 1997 in order to
quantify emissions of CDDs/CDFs (Calcagni et al., 1998). In choosing representative plants for
testing, EPA considered a variety of issues, including the types and quantities of feed materials,
the types of emission controls, and the oil content of the sinter feed. EPA decided to test a plant
with an FF and a plant with a venturi (or wet) scrubber. FFs and WSs are the principal APCDs
used to control emissions from the sintering plant windbox. Four plants used an FF and five
plants used a WS. The types of feed materials and oil content at the two selected plants were
7-28
-------
determined to be representative of other plants in the industry. Sampling was performed over 3
days (4 hr/day) at each plant.
The average CDD/CDF TEQ concentrations measured in the stack emissions were 0.19
ng I-TEQDF/Nm3 and 0.81 ng I-TEQDF/Nm3 for the WS and the FF, respectively. The
corresponding TEQ emission factors are 0.62 ng TEQDF-WHO98/kg (0.55 ng I-TEQDF/kg) sinter
and 4.61 ng TEQDF-WHO98/kg (4.14 ng I-TEQDF/kg) sinter, respectively, for WSs and FFs.
These emission factors are assigned a high rating because they are based on EPA testing at two
facilities considered by EPA to be representative of both current and 1995 standard industry
practices.
Congener-specific emission factors for these two facilities are presented in Table 7-8.
Figure 7-4 presents the congener profiles for these facilities. Although concentrations were
higher from the FF than from the WS, both concentrations were low relative to what had been
reported from testing at German, Dutch, Swedish, and Canadian sintering plants. This disparity
may be due to differences in the operation or APCDs of U.S. sintering plants and the tested
European plants.
Most of the U.S. integrated iron and steel plants, including those with sintering plants,
have eliminated the purchase and use of chlorinated organics in their facilities, and their rolling
mill oils (lubricants and hydraulic fluids) do not contain chlorinated compounds. In addition,
routine analyses of waste materials going to the sintering plant have not detected any chlorinated
solvents.
Finally, none of the U.S. plants use an ESP to control emissions from the sinter windbox
(Calcagni et al., 1998).
In 1996 (data were not readily available for 1995), 11 sintering plants were operating in
the United States, with a total annual production capacity of about 17.6 million metric tons
(Metal Producing, 1996). Since the 1980s, the size of this industry has decreased dramatically.
In 1982, 33 facilities were in operation, with a combined total capacity of 48.3 million metric
tons (U.S. EPA, 1982a). The nine U.S. sintering plants operating in 1995 had a combined
capacity of 15.6 million metric tons (Calcagni et al., 1998). In 1987, sinter consumption by iron
and steel plants was 14.5 million metric tons (AISI, 1990); in 1995, consumption was 12.4
million metric tons (Fenton, 1996), or approximately 70% of production capacity, assuming that
production capacity in 1995 was the same as in 1996. These activity level estimates are assigned
a confidence rating of medium.
Based on the production capabilities shown in Table 7-9, 59% of 1998 sinter production
capacity was at facilities with WSs and 41% was at facilities with FFs. If it is assumed that these
proportions of APCD-to-production capacity existed in 1995 and that actual production in 1995
was equal to sinter consumption at iron and steel plants (12.4 million metric tons), then estimated
7-29
-------
Table 7-8. CDD/CDF emission factors (ng/kg sinter) for sintering plants
Congener/congener
group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total 2,3,7,8 CDD/CDF3
Wet scrubber
Nondetect set
to zero
0.049
0.138
0.03
0.612
0.288
0.696
0.496
0.602
0.343
0.349
0.421
0.164
0.011
0.142
0.247
0.036
0.103
2.309
2.418
0.55
0.62
NR
NR
NR
NR
0.496
NR
NR
NR
NR
0.103
4.73
Nondetect set to 1A
detection limit
0.049
0.138
0.03
0.612
0.288
0.696
0.496
0.602
0.343
0.349
0.421
0.164
0.014
0.142
0.247
0.036
0.103
2.309
2.421
0.55
0.62
NR
NR
NR
NR
0.496
NR
NR
NR
NR
0.103
4.73
Fabric filter
Nondetect set
to zero
0.406
0.937
0.135
1.469
0.609
0.698
0.695
10.232
3.518
3.228
1.382
0.495
0.029
0.285
0.316
0
0.05
4.949
19.820
4.14
4.61
NR
NR
NR
NR
0.695
NR
NR
NR
NR
0.050
24.77
Nondetect set to 1A
detection limit
0.406
0.937
0.135
1.469
0.609
0.698
0.695
10.232
3.518
3.228
1.382
0.495
0.057
0.285
0.316
0.115
0.192
4.949
19.82
4.14
4.61
NR
NR
NR
NR
0.695
NR
NR
NR
NR
0.192
24.77
The listed values for total CDD/CDF include only the 17 toxic congeners.
Source: Calcagnietal. (1998).
7-30
-------
I.2A7.8-KX 1)1)
1.2..".4.7.8-11\( 1)1)
l.2Ao.7.8-ll\( DI)
I.2A7.8.0-I l\( 1)1)
A4.o./.8-1 Ip('l)l)
v4.d.7.8.°-< K'DD
2.0.7.8-1 ( 1)1
l.2.o.4.o.7.8-l]p('l)l
l.2.o.4.~.8.0-| |p('|)|
l.2A4.(>.7.X.O-(M'l)r
Figure 7-4. Congener profiles for air emissions from U.S. iron ore sintering
plants.
Source: Calcagnietal. (1998).
TEQ emissions from WS-equipped facilities were 4.5 g TEQDF-WHO98 (4 g I-TEQDF) and
emissions from FF-equipped facilities were 23.4 g TEQDF-WHO98 (21 g I-TEQDF), for a total of
27.9 g TEQDF-WHO98 (25.1 g I-TEQDF). These emission estimates are assigned an overall
medium confidence rating on the basis of the medium rating for the activity level estimates.
If these same assumptions are applied to the 1987 sinter consumption rate of 14.5 million
metric tons, then estimated TEQ emissions from WS-equipped facilities were 5.3 g TEQDF-
WHO98 (4.7 g I-TEQDF) and emissions from FF-equipped facilities were 27.4 g TEQDF-WHO98
(24.6 g I-TEQDF), for a total of 32.7 g TEQDF-WHO98 (29.3 g I-TEQDF). These emission
estimates are less certain than the estimates for 1995 because of uncertainties concerning actual
APCDs in place in 1987 and the content of waste feed (i.e., oil content and presence of
7-31
-------
Table 7-9. Operating parameters for U.S. iron ore sintering plants
Company
AK Steel
AK Steel3
Bethlehem Steel
Bethlehem Steel
Geneva Steel
Inland Steel
LTV Steel
U.S. Steel
Weirton Steel3
Wheeling-Pittsburgh Steel
WCI Steel
Location
Middletown, OH
Ashland, KY
Burns Harbor, IN
Sparrows Point, MD
Provo, UT
East Chicago, IN
East Chicago, IN
Gary, IN
Weirton, WV
East Steubenville, WV
Warren, OH
TOTAL
1998 capacity
(1,000 metric
tons/yr)
907
8163
2,676
3,856
816
1,089
1,270
3,992
1,1793
519
477
17,597b
Current air
pollution control
device
Wet scrubber
NA
Wet scrubber
Wet scrubber
Fabric filter
Fabric filter
Wet scrubber
Fabric filter
NA
Wet scrubber
Fabric filter
total 1998 capacity was 15,600,000 metric tons.
aNot in operation during 1998 (Calcagni et al., 1998).
bWhen the Ashland, KY, and Weirton, WV, facilities are excluded,
NA = Not available
Sources: Metal Producing (1991, 1996); Calcagni et al. (1998).
chlorinated organics in the oil) at that time. Consequently, a low confidence rating is assigned to
the emission factor and the emissions estimate.
In 2000, a total of 10,600 million metric tons of sinter were consumed in blast furnaces
(Fenton, 2001). This activity level has a high confidence rating because it is based on a
comprehensive survey. Assuming the same proportions for facilities with WSs and facilities
with FFs as in 1995 and 1987, then estimated TEQ emissions from WS-equipped facilities were
3.9 g TEQDF-WHO98 (3.4 g I-TEQDF) and emissions from FF-equipped facilities were 23.7 g
TEQDF-WHO98 (21.3 g I-TEQDF), for a total of 27.6 TEQDF-WHO98 (24.4 g I-TEQDF) for 2000.
This emissions estimate is assigned a high confidence rating on the basis of the high ratings
given to the activity level and emission factor for reference year 2000.
7.3.2. Coke Production
Coke is the principal fuel used in the manufacture of iron and steel. It is the solid
carbonaceous material produced by the destructive distillation of coal in high-temperature ovens.
7-32
-------
No testing of CDD/CDF emissions from U.S. coke facilities has been reported. However, at a
facility in the Netherlands, Bremmer et al. (1994) measured a CDD/CDF emission rate to air
during the water quenching of hot coke of 0.23 ng I-TEQDF/kg of coal consumed. Bremmer et al.
estimated minimal CDD/CDF air emissions (0.002 ng I-TEQDF/kg of coal) for flue gases
generated during the charging and emptying of the coke ovens.
The report commissioned by Environment Canada in 2000 to review steel production
processes in Canadian plants (Charles E. Napier Company, Ltd., 2000) also provided information
on emissions from coke ovens. Mean emission factors (ng I-TEQ/kg) of the four Canadian coke
oven facilities (as indicated in Table 7-10) were 0.3 ng I-TEQ/kg coke produced.
Although there are no testing data on which to base an estimate of CDD/CDF emissions
in the United States, a preliminary estimate of potential TEQ annual emissions from U.S. coke
plants can be made by combining the estimated consumption values of 33.5 million metric tons
in 1987, 29.9 million metric tons in 1995, and 26.2 million metric tons in 2000 (EIA, 2002) with
the emission factor reported by Bremmer et al. (1994) for a Dutch coke plant (0.23 ng I-
TEQDF/kg of coal consumed). These calculations yield annual emissions of 7'.7, 6.9, and 6.03 g I-
TEQDF for 1987, 1995, and 2000, respectively. These estimates should be regarded as
preliminary indications of possible emissions from this source category; further testing is needed
to confirm the true magnitude of these emissions.
7.4. SECONDARY FERROUS METAL SMELTING/REFINING
Electric arc furnaces in Europe have been reported to be sources of CDD/CDF emissions;
no testing has been reported at U.S. facilities. Electric arc furnaces are used to produce carbon
and steel alloys, primarily from scrap material, using a batch process. The input material is
typically 100% scrap. Scrap, alloying agents, and fluxing materials are loaded into the
cylindrical, refractory-lined furnace, and then carbon electrodes are lowered into the mix. The
current of the opposite-polarity electrodes generates heat through the scrap. Processing time of a
batch ranges from about 1.5 to 5 hr to produce carbon steel and from 5 to 10 hr to produce alloy
steel (U.S. EPA, 1995b).
The melting of scrap ferrous material contaminated with metalworking fluids and plastics
that contain chlorine provides the conditions conducive to formation of CDDs/CDFs. Tysklind
et al. (1989) studied the formation and release of CDDs/CDFs at a pilot 10-ton electric furnace in
Sweden. Scrap ferrous metal feedstocks containing varying amounts of chlorinated compounds
(PVC plastics, cutting oils, or calcium chloride) were charged into the furnace under different
operating conditions (continuous feed, batch feed into the open furnace, or batch feed through the
furnace lid). During continuous charging operations, the highest emissions, 1.5 ng Nordic
7-33
-------
Table 7-10. CDD/CDF emission estimates for Canadian coke oven facilities,
blast furnace facilities, and electric arc furnaces
Company/facility
Location
Plant
capacity
(1,000 net
tonnes/yr)
Estimated
production
(1,000 net
tonnes)
Estimated
CDD/CDF
emissions
(gl-TEQ)
Estimated
CDD/CDF
emission
factor
(ngl-
TEQ/kg)
Coke oven facilities
Algoma Steel Inc.
Dofasco Inc.
Stelco Inc., Lake Erie Steel
Stelco Inc., Hilton Works
Sault Ste. Marie, Ontario
Hamilton, Ontario
Nanticoke, Ontario
Hamilton, Ontario
TOTAL
1,021
1,656
563
1,035
4,275
979
1,588
540
993
4,100
0.29
0.48
0.16
0.30
1.23
0.296
0.302
0.296
0.302
Blast furnace facilities
Algoma Steel Inc.
Dofasco Inc.
Stelco Inc., Lake Erie Steel
Stelco Inc., Hilton Works
Sault Ste. Marie, Ontario
Hamilton, Ontario
Nanticoke, Ontario
Hamilton, Ontario
TOTAL
2,270
2,725
1,680
2,720
9,395
2,177
2,613
1,611
2,608
9,009
0.01
O.01
0.01
O.01
0.01
NA
NA
NA
NA
Electric arc furnaces
AltaSteel Ltd.
Atlas Specialty Steels
Atlas Stainless Steels
Co-Steel Lasco
Dofasco Inc.
Gerdau MRM Steel Inc.
Gerdan MRM Steel Inc.
IPSCO Inc.
Ispat Sidbec Inc.
Ivanco Rolling Mills Inc.
Slater Steels, Hamilton
Specialty Bar Div.
Stelco-McMaster Ltee
Sydney Corp.
Edmonton, Alberta
Welland, Ontario
Tracy, Quebec
Whitby, Ontario
Hamilton, Ontario
Cambridge, Ontario
Selkirk, Manitoba
Regina, Saskatchewan
Contrecoeur, Quebec
L'Original, Ontario
Hamilton, Ontario
Contrecoeur, Quebec
Sydney, Nova Scotia
TOTAL
295
218
118
907
1,225
290
281
907
1,633
408
363
499
454
7,598
256
189
103
788
1,065
252
244
788
1,419
355
315
434
395
6,603
0.67
0.49
0.27
0.79
0.50
0.66
0.63
1.13
3.69
0.92
0.82
1.13
0.40
12.10
2.62
2.59
2.62
1.00
0.469
2.62
2.58
1.43
2.60
2.59
2.60
2.60
1.01
NA = Not available
Source: Charles E. Napier Company, Ltd. (2000).
7-34
-------
TEQ/dry Nm3 (after an FF), were observed with a feedstock consisting of scrap metal with PVC
plastics (1.3 g chlorine/kg feedstock). This emission rate equates to 7.7 ng Nordic TEQ/kg of
feedstock.
The highest emissions during batch charging also occurred when the scrap metal with
PVC plastic was combusted (0.3 ng Nordic TEQ/dry Nm3 [1.7 ng Nordic TEQ/kg] feedstock).
Much lower emissions (0.1 ng Nordic TEQ/dry Nm3 [0.6 ng Nordic TEQ/kg] feedstock) were
observed when scrap metal with cutting oils that contained chlorinated additives (0.4 g
chlorine/kg feedstock) was melted. Although these cutting oil-related emissions were not
significantly different from the emissions observed from the melting of no-chlorine scrap metal,
relatively high levels of CDDs/CDFs (110 ng Nordic TEQ/dry Nm3) were detected in flue gases
prior to the FF.
The congener profiles of raw flue gas samples (prior to the APCD) showed that CDFs
rather than CDDs were predominant in all three feedstock types. The congener profile from the
test burn with PVC-containing feedstock showed a higher chlorinated congener content than was
observed with the other feedstocks.
Eduljee and Dyke (1996) used a range of 0.7 to 10 ng I-TEQDF/kg of scrap feed to
estimate national emissions for the United Kingdom. This range was assumed to be
representative of no-chlorine and high-chlorine operations. However, the study authors provided
little information on the supporting emission test studies (i.e., tested facility operational
materials, feed rates, congener-specific emission rates).
Umweltbundesamt (1996) reported stack testing results for a variety of electric arc
furnaces in Germany. Sufficient data were provided in the report to enable calculation of TEQ
emission factors for six of the tested facilities. Two facilities had emission factors exceeding 1
ng I-TEQDF/kg of scrap processed, and two facilities had emission factors of less than 0.1 ng I-
TEQDF/kg of scrap. The mean emission factor was 1.15 ng I-TEQDF/kg of scrap. The TEQ
concentrations in the stack gases at these facilities (corrected to 7% oxygen) ranged from less
than 0.1 to 1.3 ng I-TEQDF/m3.
The report commissioned by Environment Canada in 2000 to review steel production
processes at Canadian plants (Charles E. Napier Company, Ltd., 2000) included information on
emissions from iron sintering and provided information on emissions from electric arc furnaces,
which were estimated on the basis of plant capacity and estimated production. Mean emission
factors (ng I-TEQ/kg) of the 13 Canadian electric arc furnace facilities (as indicated in Table
7-10) were 2.1 ng I-TEQDF/kg steel produced.
In March 2000, Environment Canada reported on source testing to determine CDD/CDF
emissions from a facility in Ontario (Cianciarelli, 2000). Sampling was conducted on the
exhaust stack of the electric arc furnace of Dofasco Inc., and both concentrations and emission
7-35
-------
rates were provided (see Table 7-11). Total CDD/CDF concentrations were reported to be 51.15
pg TEQ/m3, and the total emission rate was reported to be 0.47 ng TEQ/kg steel produced. In
August 2000, the Emissions Research and Measurement Division of Environment Canada
conducted source testing to determine CDDs/CDFs from the electric arc furnace of another
facility, Gerdau Courtice Steel Inc. (Cianciarelli, 2001). These results (presented in Table 7-11)
are being used to support the Canadian dioxin/furan inventory for electric arc furnaces. The total
CDD/CDF concentrations were reported to be 125.5 pg TEQ/m3, and the total emission rate was
reported to be 1.1 ng TEQ/kg steel produced.
In 1987, electric arc furnaces accounted for 38.1% of U.S. steel production, or 30.8
billion kg of raw steel produced (Peters, 1988). In 1995, electric arc furnaces accounted for
40.4% of U.S. steel production, or 38.4 of the total 95.2 million metric tons of raw steel produced
(Fenton, 1996). In 2000, electric arc furnaces accounted for 46.2% of U.S. steel production, or
49 of the 106 million metric tons of raw steel produced (USGS, 2002).
No testing of CDD/CDF emissions from U.S. electric arc furnaces on which to base an
estimate of national emissions has been reported. A preliminary estimate of potential TEQ
annual emissions from U.S. electric arc furnaces can be made by combining the production
estimate of steel and an average emission factor of 1.21 ng I-TEQDF/kg steel derived from the
data reported in Umweltbundesamt (1996) and the three Environment Canada reports (Charles E.
Napier Company, Ltd., 2000; Cianciarelli, 2000, 2001). This calculation yields an annual
emissions estimate of 37.3 g I-TEQDF in 1987, 46.5 g I-TEQDF in 1995, and 59.3 g I-TEQDF in
2000. These estimates should be regarded as preliminary indications of possible emissions from
this source category; further testing is needed to confirm the true magnitude of these emissions.
7.5. FERROUS FOUNDRIES
Ferrous foundries produce high-strength iron and steel castings used in industrial
machinery, pipes, and heavy transportation equipment. Iron and steel castings are solid solutions
of iron, carbon, and various alloying materials. Castings are produced by injecting or pouring
molten metal into cavities of a mold made of sand, metal, or ceramic material. Metallic raw
materials are pig iron, iron and steel scrap, foundry returns, and metal turnings (U.S. EPA,
1995b, 1997a).
The melting process takes place primarily in cupola (or blast) furnaces and to a lesser
extent in electric arc furnaces. About 70% of all iron castings are produced using cupolas,
although steel foundries rely almost exclusively on electric arc furnaces or induction furnaces for
melting. The cupola is typically a vertical, cylindrical steel shell with either a refractory-lined or
a water-cooled inner wall. Charges are loaded at the top of the unit; the iron is melted as it flows
7-36
-------
Table 7-11. CDD/CDF emission concentrations and rates for Canadian
electric arc furnaces
Congener
Mean facility concentration
(pgTEQ/m3)
Runl
Run 2
Run 3
Avg.
Mean facility emission rate
(ng TEQ/tonne steel)
Runl
Run 2
Run 3
Avg.
Dofasco Inc.
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
TOTAL
0
1.96
0
0
0.25
0.11
0.04
9.96
0.48
5.88
1.12
0.51
0
0
0.11
0
0.01
20.43
1.99
4.09
0
0.33
0
0
0.01
37.11
1.23
14.9
2.23
0.81
0.52
0
0.05
0
0
63.27
0
6.44
0
1.13
0.63
0.05
0
29.45
1.60
22.98
3.86
2.02
1.43
0
0.16
0
0
69.75
0.66
4.16
0
0.49
0.29
0.05
0.02
25.51
1.1
14.59
2.4
1.11
0.65
0
0.11
0
0
51.14
0
20.5
0
0
2.7
1.2
0.4
104.2
5
61.6
11.7
5.3
0
0
1.1
0
0.1
213.8
17.2
35.3
0
2.8
0
0
0
320.9
10.6
128.8
19.3
7
4.5
0
0.4
0
0
546.8
0
60.5
0
10.6
5.9
0.5
0
276.5
15
215.7
36.3
19
13.4
0
1.5
0
0
654.9
5.7
38.8
0
4.5
2.9
0.6
0.1
233.9
10.2
135.4
22.4
10.4
6
0
1
0
0
471.9
Gerdau Courtice Steel Inc.
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
TOTAL
6.3
8.3
0.6
1
0.8
0.2
0
65
5.5
95.5
12.5
6.9
5.8
0.5
0.7
0.1
0
209.7
2.7
4.7
0.4
0.7
0.5
0.1
0
29.4
2.9
46.2
7.9
4.6
4.3
0.5
0.5
0.1
0
105.5
2.6
3.1
0.2
0.4
0.3
0.1
0
18
1.7
26
4.3
2.4
1.7
0.2
0.2
0
0
61.2
3.9
5.4
0.4
0.7
0.5
0.1
0
37.5
3.4
55.9
8.2
4.6
3.9
0.4
0.5
0.1
0
125.5
57
75
5
9
8
1
0
588
50
864
113
62
52
5
6
1
0
1,896
21
37
o
5
6
4
1
0
232
23
364
63
36
34
4
4
1
0
833
22
27
2
3
3
1
0
154
15
222
37
20
14
2
2
0
0
524
o o o
55.5
46.3
o o
5.5
6
5
1
0
324.7
29.3
483.3
71
39.3
33.3
3.7
4
0.7
0
1,084.2
Source: Cianciarelli (2000).
7-37
-------
down the cupola and is removed at the bottom. Electric induction furnaces are batch-type
furnaces in which the charge is melted by a fluctuating electromagnetic charge produced by
electrical coils surrounding the unit (U.S. EPA, 1995b, 1997a).
Iron and steel foundries, particularly those using electric arc furnaces, are highly
dependent on iron and steel scrap. Of the estimated 72 million metric tons of iron and steel scrap
consumed by the iron and steel industry in 1995, 25% (18 million metric tons) were used by
ferrous foundries. The other 75% were used by primary ferrous metal smelters (principally those
using electric arc furnaces) (USGS, 1997 f). In 2000, 20% (12.4 million metric tons) were used
by ferrous foundries; the remaining 80% were used by primary ferrous smelters (USGS, 2000).
In 2000, there were approximately 1,100 ferrous foundries in the United States producing
1.3 million metric tons of steel castings and 10 metric tons of iron castings. Thus, foundries face
the same potential for CDD/CDF emissions as do electric arc furnaces because of their use of
scrap that contains chlorinated solvents, plastics, and cutting oils (see Section 7.4) The potential
for formation and release of CDDs/CDFs during the casting process is not known.
In 1993, emissions testing was conducted at a U.S. ferrous foundry (CARB, 1993, as
reported in U.S. EPA, 1997a). The tested facility consisted of a batch-operated, coke-fired
cupola furnace charged with pig iron, scrap iron, scrap steel, coke, and limestone. Emission
control devices operating during the testing were an oil-fired afterburner and an FF. The
congener and congener group emission factors derived from the testing are presented in Table
7-12. The calculated TEQ emission factor for this set of tests is 0.42 ng TEQDF-WHO98 (0.37 ng
I-TEQDF/kg) of metal charged to the furnace.
Umweltbundesamt (1996) reported stack testing results for a variety of ferrous foundries
in Germany. Sufficient data were provided to enable calculation of TEQ emission factors for
eight of the tested facilities. Three facilities had emission factors exceeding 1 ng I-TEQDF/kg of
metal charge, and four facilities had emission factors less than 0.1 ng I-TEQDF/kg of metal
charge. The emission factors span more than four orders of magnitude. The mean emission
factor was 1.26 ng I-TEQDF/kg of metal feed.
In 1997, EPA conducted testing for emissions of dioxins at two ferrous foundries (U.S.
EPA, 1999c, f). One study was conducted on the cupola's WS; the second study was performed
on the cupola's FF. A summary of the results is presented in Table 7-13. The emission factor
developed from these tests is 2.05 ng I-TEQ/kg of metal processed.
Because of the wide range of emissions for the tested German foundries reported in
Umweltbundesamt (1996), the confidence in the degree to which the three tested U.S. facilities
represent the mean emission factor for the approximately 1,100 U.S. foundries is considered very
low. Therefore, the limited data available were judged inadequate for developing national
emission estimates that could be included in the national inventory. However, a preliminary
7-38
-------
Table 7-12. CDD/CDF emission factors for a U.S. ferrous foundry
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF (for reported congeners)
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF (not including OCDD)
Mean facility emission factor
(ng/kg scrap feed)
0.033
0.086
NR
0.051
NR
0.093
NR
0.52
0.305
0.35
0.19
0.17
NR
0.101
0.193
NR
0.059
0.263
1.888
0.372
0.415
3.96
1.76
0.55
0.19
NR
25.8
850
1.74
0.24
0.06
884.3
NR = Not reported
Source: CARB (1993), as reported in U.S. EPA (1997a).
estimate of potential TEQ annual emissions from U.S. ferrous foundries can be made by
combining the mean emission factor (1.23 ng I-TEQDF/kg of metal feed) derived from the data
reported in Umweltbundesamt (1996), CARB (1993), and U.S. EPA (1997a) with an estimated
activity level for U.S. foundries.
7-39
-------
Table 7-13. Congener-specific profile for ferrous foundries
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQ
Mean emission factor (2 facilities)
(ng I-TEQ/kg)
Nondetect set to zero
0.11
0.15
0.012
0.023
0.028
0.0033
0.16
0.084
1.08
0.21
0.1
0.0079
0.075
0.0082
0.0014
0.00009
0.00007
2.05
Nondetect set to Vz detection limit
0.11
0.15
0.012
0.023
0.028
0.0033
0.16
0.084
1.08
0.21
0.1
0.0079
0.075
0.0082
0.0014
0.00009
0.00007
2.05
Sources: U.S. EPA(1999c,f).
In 1987, U.S. shipments from ferrous foundries were 9.19 million metric tons, of which
about 90% were iron castings and 10% were steel castings (Houck, 1991). In 1995, U.S.
shipments from the approximately 1,000 U.S. ferrous foundries were 13.9 million metric tons, of
which about 90% were iron castings and 10% were steel castings (Fenton, 1996). In 2000, U.S.
shipments from the approximate 1,100 U.S. ferrous foundries were 11.3 million metric tons, of
which about 89% were iron castings and 11% were steel castings (USGS, 2001). Using the mean
emission factors and these activity levels yields annual emission estimates of 11.3 g I-TEQDF,
17.1 g I-TEQDF, and 13.9 g I-TEQDF for 1987, 1995, and 2000, respectively. These estimates
should be regarded as preliminary indications of possible emissions from this source category;
further testing is needed to confirm the true magnitude of these emissions.
7.6. SCRAP ELECTRIC WIRE RECOVERY
The objective of wire recovery is to reclaim the metal (copper, lead, silver, and gold) in
the electric wire by removing the insulating material. The recovery facility then sells the
reclaimed metal to a secondary metal smelter. Wire insulation commonly consists of a variety of
7-40
-------
plastics, asphalt-impregnated fabrics, or burlap. Chlorinated organics are used to preserve the
cable casing in below-ground cables. The combustion of chlorinated organic compounds in the
cable insulation, catalyzed by the presence of wire metals such as copper and iron, can lead to the
formation of CDDs and CDFs (Van Wijnen et al., 1992).
Although, in the past, scrap electric wire was commonly recovered using thermal
processing to burn off the insulating material, current recovery operations typically no longer
involve thermal treatment, according to industry and trade association representatives. Instead,
scrap electric wire is mechanically chopped into fine particles. The insulating material is then
removed by mixing, followed by settling of the heavier metal (telephone conversations between
T. Leighton, Versar, Inc., and R. Garino, Institute of Scrap Recycling Industries, March 2, 1993,
and T. Leighton and J. Sullivan, Triple F. Dynamics, March 8, 1993).
EPA measured dioxin-like compounds emitted to the air from a scrap wire reclamation
incinerator during its 1986 National Dioxin Study of combustion sources (U.S. EPA, 1987a).
EPA determined that the tested facility was typical of this industrial source category at that time.
Insulated wire and other metal-bearing scrap material were fed to the incinerator on a steel pallet.
The incinerator operated in a batch mode, with the combustion cycles for each batch of scrap
feed lasting between 1 and 3 hr. Natural gas was used to incinerate the material. Although most
of the wire had a tar-based insulation, PVC-coated wire was also fed to the incinerator.
Temperatures during combustion in the primary chamber furnace were about 570°C. The tested
facility was equipped with a high-temperature, natural gas-fired afterburner (980 to 1090°C).
Emission factors estimated for this facility are presented in Table 7-14. The estimated TEQ
emission factor (based only on 2,3,7,8-TCDD, 2,3,7,8-TCDF, OCDD, and OCDF) is 15.8 ng
TEQDF-WHO98 (16.9 ng I-TEQDF/kg) of scrap feed. Figure 7-5 presents a congener group profile
based on these emission factors.
Bremmer et al. (1994) reported emission factors for three facilities in the Netherlands that
subsequently ceased operations. Emission rates at a facility burning underground cables and
cables containing PVC ranged from 3.7 ng I-TEQDF/kg to 14 ng I-TEQDF/kg. The emission rates
at a second facility ranged from 21 ng I-TEQDF/kg of scrap (when burning copper core coated
with greasy paper) to 2,280 ng I-TEQDF/kg of scrap (when burning lead cable). The third facility,
which burned motors, was reported to have an emission rate of 3,300 ng I-TEQDF/kg of scrap.
On the basis of these measurements, Bremmer et al. used emission rates of 40 ng I-TEQDF/kg of
scrap and 3,300 ng I-TEQDF/kg of scrap for estimating national emissions in the Netherlands for
facilities burning wires and cables and those burning motors, respectively.
Although limited emissions testing has been conducted at one U.S. facility, the activity
level for this industry sector in reference years 1987, 1995, and 2000 is unknown; therefore, an
7-41
-------
Table 7-14. CDD/CDF emission factors for a scrap wire incinerator
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean facility emission factor" (ng/kg scrap feed)
0.374
NR
NR
NR
NR
NR
1,000
2.67
NR
NR
NR
NR
NR
NR
NR
NR
807
NR
NR
16.9b
15.8
1,000
4.42
13.7
71.1
347
107
97.4
203
623
807
3,274
aNo nondetect values were reported for 2,3,7,8-TCDD, 2,3,7,8-TCDF, or any congener group in the three test runs.
Estimated on the basis of the measured data for 2,3,7,8-TCDD, 2,3,7,8-TCDF, OCDD, and OCDF and congener
group emissions (i.e., for the penta-, hexa-, and hepta-CDDs and CDFs, it was assumed that the measured
emission factor within a congener group was the sum of equal emission factors for all congeners in that group,
including non-2,3,7,8-substiruted congeners).
NR = Not reported
Source: U.S. EPA(1987a).
7-42
-------
Figure 7-5. Congener group profile for air emissions from a scrap wire
incinerator.
Source: U.S. EPA(1987a).
estimate of national emissions cannot be made. It is uncertain how many facilities in the United
States still combust scrap wire. Trade association and industry representatives state that U.S.
scrap wire recovery facilities now burn only minimal quantities of scrap wire. However, an
inventory of CDD/CDF sources in the San Francisco Bay area noted that two facilities thermally
treated electric motors to recover electrical windings (BAAQMD, 1996).
In addition to releases from regulated recovery facilities, CDD/CDF releases from small-
scale burning of wire at unregulated facilities and open air sites have occurred; however, the
current magnitude of these types of activity in the United States is not known. Harnly et al.
(1995) analyzed soil/ash mixtures from three closed metal recovery facilities and from three
closed sites using open burning for copper recovery near a California desert town. The geometric
means of the total CDD/CDF concentrations at the facility sites and the open burning sites were
86,000 and 48,500 ng/kg, respectively. The geometric mean TEQ concentrations were 2,900 and
1,300 ng I-TEQDF/kg, respectively. A significantly higher geometric mean concentration (19,000
ng I-TEQDF/kg) was found in fly ash located at two of the facility sites. The congener-specific
7-43
-------
and congener group results from this study are presented in Table 7-15. The results show that the
four dominant congeners in the soil samples at both the facility and the open burning sites were
OCDF, 1,2,3,4,6,7,8-HpCDF, 1,2,3,4,7,8-HxCDF, and 2,3,7,8-TCDF. A slightly different profile
was observed in the fly ash samples, with 1,2,3,7,8-PeCDF and 1,2,3,4,7,8,9-HpCDF replacing
OCDD and 2,3,7,8-TCDF as the dominant congeners.
Table 7-15. CDD/CDF concentrations in fly ash and ash/soil at metal recovery
sites
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total I-TEQDF
Total CDD/CDF
Metal recovery facilities
Fly ash (2 sites)
Geometric
mean
(Hg/kg)
a
400
1,200
2,300
1,700
12,000
18,000
15,000
35,000
10,000
46,000
12,000
5,000
5,000
71,000
25,000
100,000
a
2,000
4,000
24,000
18,000
23,000
110,000
88,000
110,000
100,000
16,968
479,000
Relative
percent of
total
CDD/CDF
0.1
0.2
0.5
0.3
2.4
3.5
2.9
6.9
2
9
2.4
1
1
13.9
4.9
19.6
a
0.4
0.8
4.7
3.5
4.5
21.6
17.3
21.6
19.6
Ash/soil (3 sites)
Geometric
mean
(Hg/kg)
a
0.24
0.25
0.49
1.3
2.6
7.2
6.4
2.9
1.4
5.9
1.8
0.92
1.6
12
o
5
14
a
1.4
2.7
4.1
7.2
14
12
12
17
14
3
84.4
Relative
percent of
total
CDD/CDF
0.3
0.3
0.6
1.5
3.1
8.5
7.5
3.4
1.6
6.9
2.1
1.1
1.9
14.1
3.5
16.5
a
1.6
3.2
4.8
8.5
16.5
14.1
14.1
20
16.5
Open burn sites
Ash/soil (3 sites)
Geometric
mean
(Hg/kg)
a
0.24
0.13
0.33
0.39
1.2
3.4
1.7
0.58
0.66
2.7
0.76
0.66
0.49
4.3
0.71
6.6
a
2.8
0.98
2
3.4
5.6
7
7.6
7.4
6.6
1.3
48.4
Relative
percent of
total
CDD/CDF
0.5
0.3
0.7
0.8
2.5
7
3.5
.2
.4
5.6
.6
.4
8.9
1.5
13.6
a
5.8
2
4.1
7
11.5
14.4
15.7
15.3
13.6
aAnalytical method used had low sensitivity for TCDDs; results were not reported.
Source: Harnly etal. (1995).
7-44
-------
Van Wijnen et al. (1992) reported similar results for soil samples collected from
unpermitted incineration sites of former scrap wire and cars in the Netherlands. Total CDD/CDF
concentrations in the soil ranged from 60 to 98,000 ng/kg, with 9 of the 15 soil samples having
levels above 1,000 ng/kg. Chen et al. (1986) reported finding high levels of CDDs/CDFs in
residues from open air burning of wire in Taiwan, and Huang et al. (1992) reported elevated
levels in soil near wire scrap recovery operations in Japan. Bremmer et al. (1994) estimated an
emission rate to air of 500 ng I-TEQDF/kg of scrap for illegal, unregulated burning of cables in the
Netherlands.
7.7. DRUM AND BARREL RECLAMATION FURNACES
Hutzinger and Fiedler (1991a) reported detecting CDDs/CDFs in stack gas emissions
from drum and barrel reclamation facilities at levels ranging from 5 to 27 ng/m3. EPA measured
dioxin-like compounds in the stack gas emissions of a drum and barrel reclamation furnace as
part of the National Dioxin Study (U.S. EPA, 1987a).
Drum and barrel reclamation facilities operate a burning furnace to thermally clean used
55-gallon steel drums of residues and coatings. The drums processed at these facilities come
from a variety of sources in the petroleum and chemical industries. The thermally cleaned drums
are then repaired, repainted, relined, and sold for reuse. The drum-burning process subjects the
used drums to an elevated temperature in a tunnel furnace fired by auxiliary fuel for a sufficient
time so that the paint, interior linings, and previous contents are burned or disintegrated. Used
drums are loaded onto a conveyor that moves at a fixed speed. As the drums pass through the
preheat and ignition zone of the furnace, residual contents of the drums drain into the furnace ash
trough. A drag conveyor moves these sludges and ashes to a collection pit. The drums are air-
cooled as they exit the furnace. Exhaust gases from the burning furnace are typically drawn
through a breeching fan to a high-temperature afterburner.
The afterburner at the facility tested by EPA operated at an average of 827°C during
testing and achieved a 95% reduction in CDD/CDF emissions (U.S. EPA, 1987a). Emission
factors estimated for this facility are presented in Table 7-16. On the basis of the measured
congener and congener group emissions, the average TEQ emission factor is estimated to be 17.5
ng TEQDF-WHO98/drum (16.5 ng I-TEQDF/drum). The congener group profile is presented in
Figure 7-6.
Approximately 2.8 to 6.4 million 55-gallon drums are reclaimed by incineration annually
in the United States (telephone conversation between C. D. Ruiz, Versar, Inc., and P. Rankin,
Association of Container Reconditioners, December 21, 1992). This estimate is based on the
assumption that 23 to 26 incinerators are in operation; each incinerator, on average, handles 500
7-45
-------
Table 7-16. CDD/CDF emission factors for a drum and barrel reclamation
facility
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Mean facility emission factor" (ng/drum)
2.09
NR
NR
NR
NR
NR
37.5
36.5
NR
NR
NR
NR
NR
NR
NR
NR
22.4
NR
NR
16.5b
17.5
50.29
29.2
32.2
53.4
37.5
623
253
122
82.2
22.4
1,305.2
aNo nondetect values were reported for 2,3,7,8-TCDD, 2,3,7,8-TCDF, or any congener group in the three test runs.
'Estimated on the basis of the measured data for 2,3,7,8-TCDD, 2,3,7,8-TCDF, OCDD, and OCDF and congener
group emissions (i.e., for the penta-, hexa-, and hepta-CDDs and CDFs, it was assumed that the measured
emission factor within a congener group was the sum of equal emission factors for all congeners in that group,
including non-2,3,7,8-substituted congeners).
NR = Not reported
Source: U.S. EPA(1987a).
7-46
-------
Figure 7-6. Congener group profile for air emissions from a drum incinerator.
Source: U.S. EPA(1987a).
to 1,000 drums each day and, on average, each incinerator operates 5 days/wk, with 14 downtime
days annually for maintenance activities. The weights of 55-gallon drums vary considerably;
however, on average, a drum weighs 38 Ib (or 17 kg); therefore, an estimated 48 to 109 million
kg of drums are incinerated annually. For 1987 and 1995, EPA assumed that 4.6 million drums
were burned each year (i.e., the midpoint of the range); applying the emission factors developed
above, the estimated annual emission is 0.08 g TEQDF-WHO98 (0.08 g I-TEQDF).
The Reusable Industrial Packaging Association estimated that approximately 35 million
55-gallon barrels were reclaimed in 1997 (RIPA, 1997). Assuming the number of drums treated
has remained constant through 2000, the estimate for 2000 would be 0.61 g TEQDF-WHO98 (0.58
gI-TEQDF).
A low confidence rating is assigned to the activity level estimates for all reference years
because they are based on expert judgment rather than a published reference. A low confidence
7-47
-------
rating is also assigned to the emission factor because it was developed from stack tests conducted
at just one U.S. drum and barrel furnace and thus may not represent average emissions from
current operations in the United States. Based on these ratings, the emission estimates are
assigned a low confidence rating.
7.8. SOLID WASTE FROM PRIMARY/SECONDARY IRON/STEEL
MILLS/FOUNDRIES
Table 17 in Quab and Fermann (1997) contains summary data on the typical annual
quantities and ranges of TEQ (Norwegian-TEQ [NTEQ] and I-TEQ) from various solid residuals
from the metallurgical industries in Europe, but support information and specific congeners were
not discussed. The summary data for annual TEQ generation are presented below (in grams) for
informational purposes only and are not included in the inventory of dioxin releases presented in
this report because they are disposed of in permanent landfills and are not considered an
environmental release.
Grey iron foundries, FF dust and scrubber sludge: 0.817 NTEQ
Steel mill coke oven door leakage dust: 0.31 NTEQ
Steel mill coke oven door leakage dust: 0.04 I-TEQ
Pig iron tapping slag: 0.041 NTEQ
Basic oxygen furnace scrubber sludge: 1.53 NTEQ (range, 0.3-7.81)
Electric furnace FF dust: 3.1 I-TEQ (range of 0.4-2.4)
Electric furnace slag or FF dust: 19.2 NTEQ
7-48
-------
8. CHEMICAL MANUFACTURING AND PROCESSING SOURCES
8.1. BLEACHED CHEMICAL WOOD PULP AND PAPER MILLS
In March 1988, EPA and the U.S. pulp and paper industry jointly released the results of a
screening study that provided the first comprehensive data on the formation and discharge of
CDDs/CDFs from pulp and paper mills (U.S. EPA, 1988d). This early screening study of five
bleached kraft mills (the Five Mill Study) confirmed that the pulp bleaching process was
primarily responsible for the formation of CDDs/CDFs. The study results showed that 2,3,7,8-
TCDD was present in seven of nine bleached pulps, five of five wastewater treatment sludges,
and three of five treated wastewater effluents. The study results also indicated that 2,3,7,8-
TCDD and 2,3,7,8-TCDF were the principal CDDs/CDFs formed.
To provide EPA with more complete data on the release of these compounds, EPA and
the pulp and paper industry jointly conducted a survey during 1988 of 104 pulp and paper mills
in the United States to measure levels of 2,3,7,8-TCDD and 2,3,7,8-TCDF in effluent, sludge,
and pulp. The survey study, commonly called the 104 Mill Study, was managed by the National
Council of the Paper Industry for Air and Stream Improvement, Inc. (NCASI), with oversight by
EPA, and included all mills where chemically produced wood pulps were bleached with chlorine
or chlorine derivatives. The final study report (U.S. EPA, 1990e) was released in July 1990.
An initial phase of the 104 Mill Study involved the analysis of bleached pulp (10
samples), wastewater sludge (9 samples), and wastewater effluent (9 samples) from eight kraft
mills and one sulfite mill for all 2,3,7,8-substituted CDDs/CDFs. These analyses were conducted
to test the conclusion drawn in the Five Mill Study that 2,3,7,8-TCDD and 2,3,7,8-TCDF were
the principal CDDs/CDFs found in pulp, wastewater sludge, and wastewater effluent on a TEQ
basis. Although at the time of the study there were no reference analytical methods for many of
the 2,3,7,8-substituted CDDs/CDFs, the data obtained were considered valid by EPA for the
purposes intended because of the identification and quantification criteria used, duplicate sample
results, and limited matrix spike experiments. Table 8-1 presents a summary of the results
obtained in terms of the median concentrations and the range of concentrations observed for each
matrix (pulp, sludge, and effluent). Figures 8-1, 8-2, and 8-3 present congener profiles for each
matrix (normalized to total CDD/CDF) using the median reported concentrations.
After examination of the raw, mill-specific data, EPA concluded that the congener
profiles were fairly consistent across matrices within mills and that 2,3,7,8-TCDD and 2,3,7,8-
TCDF accounted for the majority of TEQ in the samples (U.S. EPA, 1990e). Using the median
concentrations and treating nondetect values as either zero or one-half the detection limit (DL),
3-1
-------
Table 8-1. CDD/CDF concentrations in pulp and paper mill bleached pulp, wastewater sludge, and wastewater
effluent (circa 1988)
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDDa-b
Total 2,3,7,8-CDFa-b
Total I-TEQDF
(nondetect = 0)b
Total I-TEQDF
(nondetect = !/2 DL)b
Total TEQDF-WHO98
(nondetect = 0)b
Total TEQDF-WHO98
(nondetect = !/2 DL)b
Total CDD/CDFb
Median
(ng/kg)
6.4
ND(0.3)
ND(0.4)
ND(0.5)
ND(0.5)
3.3
46
18
ND(0.7)
ND(0.2)
ND(0.3)
ND(0.3)
ND(0.3)
ND(0.3)
ND(0.6)
ND(0.6)
2.2
55.7
20.2
8.28
8.56
8.24
8.59
75.9
Bleached pulp
Range
(ng/kg)
0.4 to 124
ND(0.1)tol.4
ND(0.2) to 0.4
ND(0.2)to 1.6
ND(0.2)to0.5
2.3 to 8.4
28 to 81
1.4 to 716
ND(0.1)to3.9
ND(0.1)to4.7
ND(0.2) to ND(0.6)
ND(0.1)toND(0.4)
ND(0.1)toND(0.4)
ND(0.2) to ND(0.4)
ND(0.1)to0.8
ND(0.1)toND(2.1)
ND(2.8)to4.3
No. of
detects
(10 samples)
10
2
1
2
1
10
10
10
4
3
0
0
0
0
3
0
8
89.47
91.7
Median
(ng/kg)
63
ND(2.5)
ND(3.1)
ND(3.2)
ND(3.9)
37
698
233
6.2
4.7
ND(2.5)
ND(1.4)
ND(1.7)
ND(1.7)
6.6
ND(1.6)
22
798
272.5
90.12
91.72
1,070.5
Wastewater sludge
Range
(ng/kg)
ND(6.3)to 180
ND(1.4)to28
ND(1.5)to40
ND(1.7)to95
ND(1.7)to80
18 to 490
263 to 1,780
13 to 1150
ND(1.2)to22
ND(0.9)to38
ND(0.9)to31
ND(0.9)to33
ND(0.9) to ND(4)
ND(0.9)to34
ND(3.6)to70
ND(1.2)tolO
ND(54)tol68
No. of
detects
(9 samples)
8
1
1
1
1
9
9
9
6
6
2
1
0
1
7
1
8
Median
(Pg/L)
42
ND(9.6)
ND(12)
ND(12)
ND(12)
170
3,000
120
ND(7.2)
ND(6.3)
ND(8.4)
ND(7.1)
ND(6.2)
ND(8.2)
ND(23)
ND(22)
190
3,212
310
58.89
66.57
56.02
66.09
3,522
Wastewater effluent
Range
(Pg/L)
ND(ll)to98
ND(2.8)toND(25)
ND(6.6)toND(12)
ND(6.6) to ND(24)
ND(6.6) to ND(23)
77 to 270
1,000 to 4,600
12 to 840
ND(2.2) to 36
ND(2.2) to 33
ND(4.8)toND(15)
ND(4.8)toND(15)
ND(2.5)toND(15)
ND(4.8)toND(15)
ND(13)to44
ND(6.4)toND(41)
ND(180)to230
No. of
detects
(9 samples)
8
0
0
0
0
9
9
9
2
2
0
0
0
0
3
0
8
oo
to
Calculated assuming nondetect values were zero.
bSum of median values.
DL = Detection limit
ND = Not detected (value in parenthesis is the detection limit)
Source: U.S. EPA (1990e).
-------
Congener
Figure 8-1. 104 Mill Study full congener analysis results for pulp (nondetects
equal to zero).
Source: Median concentrations from U.S. EPA (1990e).
Congener
Figure 8-2. 104 Mill Study full congener analysis results for sludge.
Source: Median concentrations from U.S. EPA (1990e).
8-3
-------
Congener
Figure 8-3. 104 Mill Study full congener analysis results for effluent.
Source: Median concentrations from U.S. EPA (1990e).
EPA concluded that 2,3,7,8-TCDF accounted for 77.8 to 99.5% of the total TEQDF-WHO98 (77.3
to 99% of the total I-TEQDF) in pulp, 70.4 to 96.5% of the TEQDF-WHO98 (69.9 to 95.8% of the
I-TEQDF) in sludge, and 75.0 to 96.4% of the TEQDF-WHO98 (71.3 to 91.7% of the I-TEQDF) in
effluent.
NCASI reported on a similar full-congener analysis study for samples collected from
eight mills during the mid-1990s (letter dated February 10, 1997, from W. Gillespie, National
Council of the Paper Industry for Air and Stream Improvement, Inc., to G. Schweer, Versar,
Inc.). The results of these analyses are presented in Table 8-2. The frequency of detection of
2,3,7,8-TCDD and 2,3,7,8-TCDF was significantly lower than in the 1988 study; therefore,
deriving meaningful summary statistics concerning the relative importance of 2,3,7,8-TCDD and
2,3,7,8-TCDF to the total TEQ is difficult. With all nondetect values assumed to be zero,
2,3,7,8-TCDD and 2,3,7,8-TCDF accounted for 97% of the total effluent TEQDF-WHO98 (91% of
the I-TEQDF), 53% of the total sludge TEQDF-WHO98 (46% of the I-TEQDF), and 87% of the total
pulp TEQDF-WHO98 (87% of the I-TEQDF). Because of the high frequency of nondetects when all
nondetect values are one-half the DL, 2,3,7,8-TCDD and 2,3,7,8-TCDF accounted for only 13%
of the total effluent I-TEQDF, 13% of the total sludge I-TEQDF, and 28% of the total pulp I-
TEQDF.
8-4
-------
Table 8-2. CDD/CDF concentrations in pulp and paper mill bleached pulp, wastewater sludge, and wastewater
effluent (mid-1990s)
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDDa
Total 2,3,7,8-CDFa
Total I-TEQDF
(nondetect = 0)a
Total I-TEQDF
(nondetect = !/2 DL)a
Total TEQDF-WHO98
(nondetect = 0)a
Total TEQDF-WH098
(nondetect = !/2 DL)a
Bleached pulp
Mean
nondetect
= 0
(ng/kg)
0.3
0
0
0
0
0
2.4
10.3
0
0.4
0
0
0
0
0
0
0
2.7
10.7
1.53
6.4
1.5
7.6
Median
(ng/kg)
ND(1)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(10)
ND(1)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(10)
Range
(ng/kg)
ND(l)to5
ND(3) to ND(7)
ND(3) to ND(7)
ND(3) to ND(7)
ND(3) to ND(7)
ND(3) to ND(7)
ND(10)tol5
ND(l)tol70
ND(3) to ND(7)
ND(3) to 7
ND(3) to ND(7)
ND(3) to ND(7)
ND(3) to ND(7)
ND(3) to ND(7)
ND(3) to ND(7)
ND(3) to ND(7)
ND(6)toND(14)
No. of
detects/
samples
1/18
0/18
0/18
0/18
0/18
0/18
3/16
7/18
0/18
1/18
0/18
0/18
0/18
0/18
0/18
0/18
0/18
Wastewater sludge
Mean
nondetect
= 0
(ng/kg)
0.8
0
0.5
2.3
1.6
41.4
445
6.2
0
0.5
0
0
0
0.5
1.2
0
0
491.6
8.4
3
12.9
2.6
15.2
Median
(ng/kg)
ND(1)
ND(5)
ND(5)
ND(5)
ND(5)
7
150
3
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(5)
ND(10)
Range
(ng/kg)
ND(l)to4
ND(4) to ND(52)
ND(4) to 7
ND(4)tol8
ND(4) to 14
ND(4) to 330
21 to 2,900
ND(l)to31
ND(4) to ND(52)
ND(4) to 7
ND(4) to ND(52)
ND(4) to ND(52)
ND(4) to ND(52)
ND(4) to 6
ND(4)tolO
ND(4) to ND(52)
ND(9)toND(100)
No. of
detects/
samples
4/12
0/12
1/13
2/13
2/13
9/13
10/10
9/12
0/13
1/13
0/13
0/13
0/13
1/13
2/13
0/13
0/13
Wastewater effluent
Mean
nondetect
= 0
(Pg/L)
1.2
0
0
0
0
3.2
99
2.3
103
2.3
1.5
53.6
1.4
66.5
Median
(ng/kg)
ND(ll)
ND(53)
ND(53)
ND(53)
ND(53)
ND(53)
ND(llO)
ND(ll)
ND(53)
ND(53)
ND(53)
ND(53)
ND(53)
ND(53)
ND(53)
ND(53)
ND(106)
Range
(Pg/L)
ND(10)to21
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to 58
ND(100)to370
ND(10)to23
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(50) to ND(55)
ND(104)toND(HO)
No. of
detects/
samples
1/18
0/18
0/18
0/18
0/18
1/18
6/14
2/18
0/18
0/18
0/18
0/18
0/18
0/18
0/18
0/18
0/18
oo
a Sum of mean values.
DL = Detection limit
ND = Not detected (value in parenthesis is the detection limit)
Source: Letter dated February 10,1997, from W. Gillespie, National Council of the Paper Industry for Air and Stream Improvement, Inc., to G. Schweer, Versar, Inc.
-------
In 1992, the pulp and paper industry conducted its own NC AST-coordinated survey of
2,3,7,8-TCDD and 2,3,7,8-TCDF emissions (NCASI, 1993). Ninety-four mills participated in
the study, and NCASI assumed that the remaining 10 (of 104) operated at the same levels as
measured in the 1988 104 Mill Study. All nondetect values were counted as one-half the DL. If
a DL was not reported, it was assumed to be 10 pg/L for effluent and 1 ng/kg for sludge or
bleached pulp. The data used in the report were provided by individual pulp and paper
companies that had been requested by NCASI to generate the data using the same protocols used
in the 104 Mill Study.
In 1993, as part of its efforts to develop revised effluent guidelines and standards for the
pulp, paper, and paperboard industry, EPA published the development document for the
guidelines and standards being proposed for this industry (U.S. EPA, 1993b). The development
document presented estimates of the 2,3,7,8-TCDD and 2,3,7,8-TCDF annual discharges in
wastewater from the mills in this industry as of January 1, 1993. To estimate these discharges,
EPA used the most recent information about each mill from four databases (104 Mill Study, EPA
short-term monitoring studies at 13 mills, EPA long-term monitoring studies at eight mills, and
industry self-monitoring data submitted to EPA). The 104 Mill Study data were used for only
those mills that did not report making any process changes subsequent to the 104 Mill Study and
did not submit any more recent effluent monitoring data.
Gillespie (1994) and Gillespie (1995) reported the results of 1993 and 1994 updates,
respectively, to the 1992 NCASI survey. As in the 1992 survey, companies were requested to
follow the same protocols for generating data that were used in the 104 Mill Study. Gillespie
(1994, 1995) reported that fewer than 10% of mills had 2,3,7,8-TCDD and 2,3,7,8-TCDF
concentrations in effluent above the nominal DLs of 10 pg/L and 100 pg/L, respectively. EPA
obtained similar results in its short- and long-term sampling for 18 mills; 2,3,7,8-TCDD was
detected at four mills, and 2,3,7,8-TCDF was detected at nine mills (U.S. EPA, 1993b).
Gillespie (1994) reported that wastewater sludges at most mills (90%) contained less than
31 ng/kg of 2,3,7,8-TCDD and less than 100 ng/kg of 2,3,7,8-TCDF. Gillespie (1995) reported
that 90% of the mills had 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations in sludge of less than
17 ng/kg and 76 ng/kg, respectively, in 1994. U.S. EPA (1993b) reported similar results but
found detectable levels of 2,3,7,8-TCDD and 2,3,7,8-TCDF in sludges from 64% and 85%,
respectively, of the facilities sampled.
Gillespie (1994) reported that nearly 90% of the bleached pulps contained less than 2
ng/kg of 2,3,7,8-TCDD and less than 160 ng/kg of 2,3,7,8-TCDF. Gillespie (1995) reported that
90% of the bleached pulps contained 1.5 ng/ng or less of 2,3,7,8-TCDD and 5.9 ng/kg or less of
2,3,7,8-TCDF. The final levels in white paper products would correspond to levels in bleached
pulp, so bleached paper products would also be expected to contain less than 2 ng/kg of 2,3,7,8-
TCDD.
-------
On April 15, 1998, EPA promulgated effluent limitation guidelines and standards for
certain segments of the pulp, paper, and paperboard industry (Federal Register, 1998c). The
industry segments covered by this rulemaking (i.e., the bleached paper-grade kraft and soda
subcategory and the paper-grade sulfite subcategory) are those segments responsible for more
than 90% of the bleached chemical pulp production in the United States. For this rule, EPA
updated the estimates of baseline loadings made in 1993 for the proposed rule by using more
recent data collected by EPA, NCASI (including the 1994 NCASI survey), and individual
facilities (U.S. EPA, 1997d). These revised estimates are presented in the last column in Table
8-3. EPA projects that, after full compliance with these rules, annual TEQ discharges will be
reduced to 5 g in effluent and 7 g in sludge.
Table 8-3. Summary of bleached chemical pulp and paper mill discharges
(g/yr) of 2,3,7,8-TCDD and 2,3,7,8-TCDF
Matrix
Effluent
Sludge6
Pulp
Congener
2,3,7,8-TCDD
2,3,7,8-TCDF
TEQ
2,3,7,8-TCDD
2,3,7,8-TCDF
TEQ
2,3,7,8-TCDD
2,3,7,8-TCDF
TEQ
U.S. EPA
1988a
201
1,550
356
210
1,320
343
262
2,430
505
NCASI
1992"
22
99
32
33
118
45
24
124
36
U.S. EPA
1993C
71
341
105
~
~
177
~
~
149
NCASI
1993"
19
76
27
24
114
35
22
106
33
NCASI
1994"
14.6
49.0
19.5
18.9
95.2
28.4
16.2
78.8
24.1
U.S. EPA
1995"
16
120
28
~
~
50
~
~
40
aData from the 104 Mill Study (U.S. EPA, 1990e). The total discharge rate of congener or TEQ (based only on
2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations) was summed across all 104 mills.
bThe total discharge rate of congener or TEQ (based only on 2,3,7,8-TCDD and 2,3,7,8-TCDF concentrations) was
summed across all 104 mills. The daily discharge rates reported in NCASI (1993) and Gillespie (1994, 1995) were
multiplied by a factor of 350 days/yr to obtain estimates of annual discharge rates. NCASI 1992 survey (NCASI,
1993), 1993 update (Gillespie, 1994), and 1994 update (Gillespie, 1995).
The discharges in effluent and sludge were estimated in U.S. EPA (1993b, 1997d) for January 1, 1993. The TEQ
discharge in pulp was estimated by multiplying the 1988 discharge estimate by the ratio of the 1988 and!993
effluent discharge estimates (i.e., the estimate of the reduction in 1988 discharges achieved by pollution prevention
measures taken by the industry between 1988 and 1993).
dThe discharges in effluent and sludge were estimated in U.S. EPA (1997d) for mid-1995. The TEQ discharge in
pulp was estimated by multiplying the 1988 discharge estimate by the ratio of the 1988 and!995 effluent discharge
estimates (i.e., the estimate of the reduction in 1988 discharges achieved by pollution prevention measures taken by
industry between 1988 and 1995).
eApproximately 20.5% of the sludge generated in 1990 was incinerated. The remaining 79.5% was predominantly
landfilled (56.5%) or placed in surface impoundments (18.1%); 4.1% was land-applied directly or as compost, and
0.3% was distributed or marketed (U.S. EPA, 1993e).
NCASI = National Council of the Paper Industry for Air and Stream Improvement, Inc.
~ = No information given
3-7
-------
8.1.1. Estimates of National Emissions in 1987 and 1995
The U.S. annual discharges of 2,3,7,8-TCDD and 2,3,7,8-TCDF are summarized in Table
8-3 for each of the six surveys discussed above. EPA release estimates for 1988 (U.S. EPA,
1990e) and 1995 (U.S. EPA, 1997d) are believed to best represent emissions in reference years
1987 and 1995, respectively. During the period between EPA's 104 Mill Study and issuance of
the development document (U.S. EPA, 1993b), the U.S. pulp and paper industry reduced releases
of CDDs/CDFs, primarily by instituting numerous process changes to reduce the formation of
CDDs/CDFs during the production of chemically bleached wood pulp. Details on the process
changes implemented are provided in U.S. EPA (1993b) and Gillespie (1995). Much of the
reduction between 1988 and 1995 can be attributed to process changes for pollution prevention.
The confidence ratings for these release estimates are judged to be high because direct
measurements were made at virtually all facilities, providing a high level of confidence in both
the production and the emission factor estimates. The best estimates of annual emissions in 1987
(i.e., the 1988 estimates presented in Table 8-3) are 356 g TEQ/yr for effluent and 343 g TEQ/yr
for sludge. The best estimates of annual emissions in 1995 (i.e., the 1995 estimates presented in
Table 8-3) are 28 g TEQ/yr for effluent and 50 g TEQ/yr for sludge. The CDD/CDF content in
bleached chemical wood pulp as a product is estimated to be approximately 505 g TEQ and 40 g
TEQ in 1987 and 1995, respectively. Although EPA provided an estimate of contaminant levels
of CDDs/CDFs in wood pulp, it is currently not known whether the dioxin contamination in the
product actually resulted in a release to the open and circulating environment.
In 1990, the majority (75.5%) of the wastewater sludge generated by these facilities was
placed in landfills or in surface impoundments, with the remainder incinerated (20.5%), applied
to land directly or as compost (4.1%), or distributed as a commercial product (less than 1%)
(U.S. EPA, 1993e). Data on the disposition of wastewater sludges are available only for years
1988 through 1995. On the basis of these data, the best estimate of TEQ applied to land (i.e., not
incinerated or landfilled) is 14.1 g TEQ (4.1% of 343 g) for 1987 and 2 g (4.1% of 50 g) for
1995. These emission estimates are assigned a high level of confidence on the basis of the high
confidence ratings given to both the activity level and the emission factor estimates.
8.1.2. Estimates of National Emissions in 2000
In 2000, NCASI provided estimates of congener-specific CDD/CDF releases from the
pulp and paper industry in effluent, wastewater residuals, and pulp (Gillespie, 2002). Emission
factors were taken from "NCASI Handbook of Chemical Specific Information for SARA
(Superfund Amendments and Reauthorization Act) Section 313 Form R Reporting." Emission
factors were compiled from valid test data supplied to NCASI by a variety of sources, including
member companies that had performed the tests in response to a regulatory program. The mass
-------
throughput parameter of total pulp production (31.9 million metric tons/yr) was provided by the
American Forest and Paper Association and included data from 12 elemental chlorine-free mills.
The effluent flow from chemical pulp mills with aerated stabilization basins (1,509 million
gal/day) and with activated sludge treatment (660 million gal/day) was taken from the NCASI
database and included data from five aerated stabilization basin mills and three activated sludge
treatment mills. The primary waste treatment residuals from pulp mills (0.974 million dry metric
tons/yr) and the combined, secondary, and dredged waste treatment residuals from pulp mills
(1.37 million dry metric tons/yr) were also taken from the NCASI database and included data
from five mills for the primary residuals and data from three mills for the secondary residuals
(Gillespie, 2002).
Table 8-4 provides a breakdown of TEQDF-WHO98 concentrations and emissions by
congener. Total TEQDF.WHO98 concentrations were reported to be 0.49 pg/L, 1.76 ng/kg, and
0.02 pg/g for effluent, sludge, and pulp, respectively. CDD/CDF emission estimates were
reported as 1.02 g TEQDFWHO98/yr, 1.93 g TEQDFWHO98/yr, and 0.582 g TEQDF WHO98/yr for
effluent, sludge, and pulp, respectively.
Fifty-one percent of the sludge generated in 2000 was sent to landfills or lagoons. It is
uncertain how much of the remaining 49% of the sludge was applied to land. However, a
conservative estimate can be developed by applying the 4.1% used to develop the 1987 and 1995
estimates. In this case, 0.08 g TEQDF.WHO98/yr of sludge is estimated to have been applied to
land in 2000. These estimates are assigned a high confidence rating because they are based on
recent industry survey data; however, EPA is working with NCASI to develop a QA/QC protocol
to monitor the data being collected.
8.2. MANUFACTURE OF CHLORINE, CHLORINE DERIVATIVES, AND METAL
CHLORIDES
Testing of CDD/CDF emissions to air, land, or water from U.S. manufacturers of
chlorine, chlorine derivatives, and metal chlorides on which to base estimates of national
emissions has not been reported. Sampling of graphite electrode sludges from European chlorine
manufacturers indicates high levels of CDFs. Limited sampling of chlorine derivatives and metal
chlorides in Europe indicates low-level contamination in some products.
8.2.1. Manufacture of Chlorine
Chlorine gas is produced by electrolysis of brine electrolytic cells. Until the late 1970s,
the primary type of electrolytic process used in the chlor-alkali industry to produce chlorine
consisted of the use of mercury cells containing graphite electrodes. As shown in Table 8-5, high
levels of CDFs have been found in several samples of graphite electrode sludge from facilities in
8-9
-------
Table 8-4. CDD/CDF TEQ concentrations and emissions for the paper and
pulp industry by source
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
1,2,3,4,6,7,8,9-OCDF
TEQDF-WH098
Effluent
TEQ
cone."
(pg/L)
0
O.OOe+00
O.OOe+00
1.30e-01
9.00e-02
7.00e-02
7.37e-02
l.OOe-01
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
1.90e-02
5.00e-03
2.00e-03
0.49
TEQ
emissions
(ng/yr)
O.OOe+00
O.OOe+00
O.OOe+00
2.71e+08
1.88e+08
1.46e+08
1.54e+08
2.08e+08
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
3.96e+07
1.04e+07
4.17e+06
1.02e+09
Waste treatment
residuals not lagooned
or landfilled (sludge)
(49% not landfilled)
TEQ
cone."
(ng/kg)
4.00e-01
5.00e-02
l.OOe-02
8.00e-02
9.00e-02
1.82e-01
2.80e-01
4.00e-01
l.OOe-02
l.OOe-01
4.00e-02
l.OOe-02
5.00e-02
O.OOe+00
1.70e-02
O.OOe+00
3.70e-03
1.76
Residuals total
Residuals not landfilled
TEQ
emissions
(ng/yr)
4.63e+08
6.24e+07
1.25e+07
8.53e+07
9.05e+07
1.97e+08
2.81e+08
4.66e+08
1.25e+07
1.25e+08
4.63e+07
1.25e+07
5.15e+07
O.OOe+00
1.83e+07
O.OOe+00
3.93e+06
1.93e+09
9.44e+08
Pulp
TEQ
cone."
(pg/g pulp)
l.OOe-02
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
3.00e-03
3.04e-03
l.OOe-03
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
2.00e-03
O.OOe+00
l.OOe-03
O.OOe+00
6.00e-05
0.02
TEQ
emissions
(ng/yr)
2.90e+08
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
8.69e+07
8.80e+07
2.90e+07
O.OOe+00
O.OOe+00
O.OOe+00
O.OOe+00
5.79e+07
O.OOe+00
2.90e+07
O.OOe+00
1.74e+06
5.83e+08
aTEQ concentrations are in TEQDF-WHO98.
Source: Gillespie (2002).
Europe. The CDFs predominate in these sludges, and the 2,3,7,8-substituted congeners account
for a large fraction of the respective congener totals (Rappe et al., 1990a, 1991; Rappe, 1993;
Strandell et al., 1994). During the 1980s, titanium metal anodes were developed to replace
graphite electrodes (U.S. EPA, 1982b; Curlin and Bommaraju, 1991). Currently, no U.S. facility
is believed to use graphite electrodes in the production of chlorine gas (telephone conversation
between L. Phillips, Versar, Inc., and T. Fielding, U.S. EPA, Office of Water, February 1993).
8-10
-------
Table 8-5. CDD/CDF concentrations (jig/kg) in graphite electrode sludge
from chlorine production
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDDa
Total 2,3,7,8-CDFa
Total I-TEQDFa
Total TEQDF-WHO98a
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDFa
Sludge 1
ND (0.006)
ND (0.007)
ND (0.018)
ND (0.012)
ND (0.016)
0.095
0.92
26
25
12
32
7
1.3
0.87
9.1
8.1
31
1.02
152.37
14.2
14.1
ND (0.006)
ND (0.070)
ND (0.046)
0.22
0.92
64
75
68
24
31
263.14
Sludge 2
ND (0.009)
ND (0.009)
ND (0.026)
ND (0.016)
ND (0.022)
0.21
2
56
55
25
71
16
2.8
1.9
19
19
76
2.21
341.7
30.5
30.4
ND (0.009)
ND (0.009)
ND (0.064)
0.48
2
150
240
140
53
76
661.48
Sludge 3
ND (0.009)
ND (0.009)
ND (0.029)
ND (0.019)
ND (0.025)
0.25
2.2
57
56
24
73
15
2.6
2
19
20
71
2.45
339.6
30.2
30.2
ND (0.009)
ND (0.009)
ND (0.074)
0.56
2.2
140
240
140
54
71
647.76
Sludge 4
ND
ND (0.033)
ND (0.49)
ND (0.053)
ND(1.2)
0.055
0.65
52
55
27
44
12
1.7
1.3
15
14
81
0.7
303
27.7
27.6
~
~
~
~
0.65
—
—
—
—
81
-
Calculated assuming nondetect values were zero.
ND = Not detected (value in parenthesis is the reported detection limit)
~ = No information given
Sources: Rappe et al. (1991); Rappe (1993).
Although the origin of the CDFs in graphite electrode sludge is uncertain, chlorination of
the cyclic aromatic hydrocarbons (such as dibenzofuran) present in the coal tar used as a binding
agent in the graphite electrodes has been proposed as the primary source (Strandell et al., 1994).
8-11
-------
For this reason, sludges produced using metal electrodes were not expected to contain CDFs.
However, results of an analysis of metal electrode sludge from a facility in Sweden, analyzed as
part of the Swedish Dioxin Survey, showed that the sludge contained high levels of CDFs
(similar to those of the graphite sludge) and primarily nondetectable levels of CDDs (Strandell et
al., 1994). The sludge showed the same type of CDF congener pattern reported by Rappe et al.
(1991) and Rappe (1993). Strandell et al. suggested that chlorination of polyaromatic
hydrocarbons present in the rubber linings of the electrolytic cell may have produced the CDFs
found in the one sample analyzed.
Although EPA does not regulate CDDs/CDFs specifically, it issued restrictions under
RCRA on the land disposal of wastewater and sludges generated by chlorine manufacturers that
use the mercury cell process and the diaphragm process (with graphite electrodes) (waste codes
K071, K073, and K106) (40 CFR 268).
The Chlorine Chemistry Council (CCC), a trade association representing manufacturers
that produce and/or use chlorine, sampled the treated wastewater discharges to surface waters at
seven chlor-alkali facilities in the U.S. (CCC, 2004). The results of this study are presented in
Table 8-6. The measurements are intended to represent wastewater discharges in both 2000 and
1995. As indicated in Table 8-6, 1.79 g TEQDF-WHO98 (1.81 g I-TEQ) were released to water
from chlor-alkali facilities in 2000 and 1995. The emission estimates for 2000 and 1995 are
assigned a high confidence rating, because they were derived from the testing of chlor-alkali
facility treated wastewater.
The CCC reported on air emissions from two chlor-alkali production facilities in 2000
(CCC, 2004). Table 8-7 summarizes the CDD/CDF congener-specific and TEQ annual releases
to air from the Dow Chemical chlor-alkali facility in Midland, MI, and the PPG Industries facility
in Natrium, WV. Total releases to air in 2000 were 0.08 g TEQDF-WHO98 (0.08 g I-TEQ). There
are no data showing air releases from chlor-alkali facilities in 1995 and 1987. A high
confidence rating is to assigned the estimate for 2000 because it is based on actual measurements
of chlor-alkali facilities.
8.2.2. Manufacture of Chlorine Derivatives and Metal Chlorides
The limited sampling of chlorine-derivative products indicates that they contain very low,
if any, concentrations of CDDs/CDFs. Rappe et al. (1990b) analyzed a sample of chlorine bleach
consisting of 4.4% sodium hypochlorite. Most of the 2,3,7,8-substituted CDD/CDF congeners
were below the limits of detection (0.3 to 7 pg/L for all congeners except OCDD and OCDF,
which were 12 and 20 pg/L, respectively). No 2,3,7,8-substituted CDDs were detected. Tetra-,
penta-, and hexa-CDFs were detected at levels of 13 pg/L or lower. The TEQ content of the
sample was 4.9 pg I-TEQDF/L. Hutzinger and Fiedler (1991b) reported finding no CDDs/CDFs at
8-12
-------
Table 8-6. Releases of dioxin-like compounds in wastewater discharges from
chlor-alkali and mixed chemical manufacturing facilities to surface water in
reference years 2000 and 1995
Congener
2,3,7,8 TCDD
1, 2,3,7,8 PeCDD
1, 2,3,4,7,8 HxCDD
1, 2,3,6,7,8 HxCDD
1,2,3,7,8,9 HxCDD
1,2,3,4,6,7,8 HpCDD
OCDD
2,3,7,8 TCDF
1, 2,3,7,8 PeCDF
2,3,4,7,8 PeCDF
1,2,3,4,7,8 HxCDF
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 HpCDF
OCDF
Total I-TEQ
Total TEQDF-WHO98
Occidental Chemical Corporation
Battleground,
TX
0.00
0.00
0.00
0.00
0.00
0.00
0.48
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
4.83e-04
4.83e-05
Deer
Park, TX
0.02
0.00
0.00
0.00
0.00
0.08
21.50
0.63
1.20
0.21
2.11
0.38
2.03e-03
0.00
0.59
5.66e-03
4.88
0.53
0.51
Delaware
City, DE
0.00
0.00
0.00
0.00
0.00
0.00
4.09e-03
1.02e-03
0.00
0.00
0.00
0.00
0.00
0.00
2.31e-03
0.00
0.00
1.29e-04
1.26e-04
Hahnville,
LA
0.00
0.00
0.00
0.00
0.00
0.01
0.10
0.19
0.90
0.85
2.96
1.18
0.63
0.60
4.47
0.69
1.75
1.08
1.08
Mobile,
AL
0.00
0.00
0.00
0.00
0.00
0.00
1.15e-03
2.88e-04
0.00
0.00
0.00
0.00
0.00
0.00
6.49e-04
0.00
0.00
3.64e-05
3.54e-05
Muscle
Shoals,
AL
0.00
0.00
0.00
0.00
0.00
0.00
1.13e-09
3.94e-08
1.33e-07
7.99e-08
1.85e-07
9.76e-08
2.29e-08
3.28e-08
1.32e-07
6.30e-08
1.34e-07
8.65e-08
8.63e-08
PPG
Industries
Natrium,
WV
0.00
0.00
0.00
0.00
0.00
0.22
3.13
0.06
0.06
0.33
0.11
0.00
0.00
0.00
0.15
0.00
0.66
0.19
0.19
Total
0.02
0.00
0.00
0.00
0.00
0.31
25.22
0.89
2.16
1.39
5.18
1.56
0.63
0.60
5.22
0.69
7.29
1.80
1.59
Source: CCC (2004).
a DL of 4 |ig/kg in chlorine gas or in samples of 10% sodium hypochlorite, 13% sodium
hypochlorite, and 31 to 33% hydrochloric acid at a DL of 1 |ig/kg.
Hutzinger and Fiedler (1991b) reported the results of analyses of samples of ferric
trichloride (FeCl3), aluminum trichloride (A1C13), copper chloride (CuCl2), cuprous chloride
(CuCl), silicon tetrachloride (SiCl4), and titanium tetrachloride (TiCl4) for their HpCDF, OCDF,
HpCDD, and OCDD content. The sample of FeCl3 contained HpCDF and OCDF in the low
micrograms per kilogram range, but no HpCDD or OCDD was detected at a DL of 0.02 |ig/kg.
One of the two samples of A1C13 analyzed also contained a low (jig/kg) concentration of OCDF.
The samples of CuCl2 and CuCl contained concentrations of HpCDF, OCDF, and OCDD of less
than 1 |ig/kg. The results are presented in Table 8-8.
8-13
-------
Table 8-7. Congener-specific and TEQ annual releases to air (g/yr) from
chlor-alkali and mixed chemical production facilities in 2000
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDD
1,2,3,4,7,8,9-HpCDD
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Dow Chemical Co.
Midland, MI
0.027
0.004
0.008
0.000
0.003
0.020
0.086
0.023
0.009
0.008
0.066
0.000
0.002
0.009
0.148
0.028
0.225
0.047
0.049
PPG Industries
Natrium, WV
0.003
0.000
0.000
0.002
0.002
0.087
0.208
0.044
0.003
0.030
0.044
0.006
0.006
0.022
0.142
0.039
0.064
0.034
0.033
Source: Chlorine Chemistry Council (2004).
8.3. MANUFACTURE OF HALOGENATED ORGANIC CHEMICALS
Several chemical production processes generate CDDs/CDFs (Versar, Inc., 1985;
Hutzinger and Fiedler, 1991b). CDDs/CDFs can be formed during the manufacture of
chlorophenols, chlorobenzenes, and chlorobiphenyls (Versar, 1985; Ree et al., 1988).
Consequently, disposal of industrial wastes from manufacturing facilities producing these
compounds may result in the release of CDDs/CDFs to the environment. Also, the products
themselves may contain these compounds, and their use or consumption may result in additional
releases to the environment.
CDD/CDF congener distribution patterns indicative of noncombustion sources have been
observed in sediments in southwest Germany and the Netherlands. According to Ree et al.
(1988), the congener patterns found suggest that wastes from the production of chlorinated
organic compounds may be important historical sources of CDD/CDF contamination in these
8-14
-------
Table 8-8. CDD/CDF concentrations (jig/kg) in metal chlorides
Congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
FeCl3
—
—
—
ND
ND
—
—
—
12
42
AlCl3a
—
~
~
ND
ND
—
~
~
ND
ND
AlCl3a
—
—
—
ND
0.1
—
—
—
ND
34.0
CuCl2
—
~
~
0.03
0.60
—
~
~
0.10
0.50
CuCl
—
—
—
ND
0.03
—
—
—
0.08
0.20
TiCl4
—
~
~
ND
ND
—
~
~
ND
ND
SiCl4
—
—
—
ND
ND
—
—
—
ND
ND
aAlC!3 was tested twice.
ND = Not detected; detection limit of 0.02 ug/kg
~ = No information given
Source: Hutzinger and Fiedler (1991b).
regions. The production and use of many of the chlorophenols, chlorophenoxy herbicides, and
PCB products are now banned or strictly regulated in most countries. However, these products
may have been a source of the environmental contamination that occurred prior to the 1970s and
may continue to be a source of environmental releases under certain limited use and disposal
conditions (Rappe, 1992b).
8.3.1. Chlorophenols
Chlorophenols have been widely used for a variety of pesticidal applications. The higher-
chlorinated phenols (tetrachlorophenol and pentachlorophenol [PCP]) and their sodium salts
have been used primarily for wood preservation. The lower-chlorinated phenols have been used
primarily as chemical intermediates in the manufacture of other pesticides. For example, 2,4-
dichlorophenol is used to produce the herbicides 2,4-dichlorophenoxyacetic acid (2,4-D), 4-(2,4-
dichlorophenoxy)butanoic acid (2,4-DB), 2-(2,4-dichlorophenoxy)-propanoic acid (2,4-DP),
Nitrophen, Genite, and Zytron and 2,4,5-trichlorophenol was used to produce hexachlorophene,
2,4,5-T, Silvex, Erbon, Ronnel, and Gardona (Gilman et al., 1988; Hutzinger and Fiedler,
1991b). (Sections 8.3.7 and 8.3.8 contain information on EPA actions to control CDD/CDF
contamination of pesticides, including PCP and its salts, and to obtain additional data on
CDD/CDF contamination of pesticides.)
8-15
-------
The two major commercial methods used to produce chlorophenols are (1) electrophilic
chlorination of molten phenol by chlorine gas in the presence of catalytic amounts of a metal
chloride and organic chlorination promoters and stabilizers, and (2) alkaline hydrolysis of
chlorobenzenes under heat and pressure using aqueous methanolic sodium hydroxide. Other
manufacturing methods include conversion of diazonium salts of various chlorinated anilines and
chlorination of phenolsulfonic acids and benzenesulfonic acids, followed by the removal of the
sulfonic acid group (Oilman et al., 1988; Hutzinger and Fiedler, 1991b).
Because of the manufacturing processes employed, commercial chlorophenol products
can contain appreciable amounts of impurities (Oilman et al., 1988). During the direct
chlorination of phenol, CDDs/CDFs can form either by the condensation of tri-, tetra-, and
pentachlorophenols or by the condensation of chlorophenols with hexachlorocyclohexadienone
(which forms from excessive chlorination of phenol). During alkaline hydrolysis of
chlorobenzenes, CDDs/CDFs can form through chlorophenate condensation (Ree et al., 1988;
Oilman et al., 1988; Hutzinger and Fiedler, 1991b).
The limited information on CDD/CDF concentrations in chlorophenols published in the
1970s and early 1980s was compiled by Versar, Inc. (1985) and Hutzinger and Fiedler (1991b).
The results of several major studies cited by these reviewers (Firestone et al., 1972; Rappe et al.,
1978a, b) are presented in Table 8-9. Typically, CDDs/CDFs were not detected in mono- and
dichlorophenols but were reported in tri- and tetrachlorophenols. More recent results of testing
of 2,4-dichlorophenol, performed in response to the Toxic Substances Control Act (TSCA)
dioxin/furan test rule, showed no detectable concentrations of 2,3,7,8-substituted tetra- through
hepta-CDD/CDFs.
Other than a study by Hagenmaier (1986), which reported finding 2,3,7,8-TCDD at a
concentration of 0.3 |ig/kg in a sample of 2,3,4,5-tetrachlorophenol, no more recent data on
concentrations of CDDs and CDFs could be found in the literature for the mono- through
tetrachlorophenols. Tables 8-10 and 8-11 present summaries of several studies that reported
CDD/CDF concentrations in PCP and in PCP-Na products, respectively. Many of these studies
do not report congener-specific concentrations, and many are based on products obtained from
non-U.S. sources.
8.3.1.1. Regulatory Actions for Chlorophenols
Section 8.3.8 of this report describes regulatory actions taken by EPA to control the
manufacture and use of chlorophenol-based pesticides. In the mid-1980s, EPA's Office of Solid
Waste (OSW) promulgated, under RCRA, land disposal restrictions on wastes (wastewaters and
nonwastewaters) resulting from the manufacture of chlorophenols (40 CFR 268). Table 8-12
lists all wastes in which CDDs/CDFs are specifically regulated by EPA as hazardous
8-16
-------
Table 8-9. CDD/CDF concentrations (mg/kg) in mono- through tetrachlorophenols
Congener/
congener
group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
TOTAL
2-CPa
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
+
ND
ND
ND
ND
-
2,4-DCPa
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND
ND
ND
ND
ND
-
2,6-DCPa
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND
ND
ND
ND
ND
-
2,4,5-TrCP
(Na salt)3
ND (0.02) to 14
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND
ND
ND
ND
ND
-
2,4,5-TrCPa
ND (0.02) to 6.5
ND (0.02) to 1.5
ND (0.02)
ND (0.02)
ND (0.02)
ND
ND
ND
ND
ND
-
2,4,6-TrCPa
ND (0.02) to 49
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
+
+
+
ND
ND
-
2,4,6-TrCP
(Na salt)b
O.02
0.03
0.03
0.1
0.1
1.5
17.5
36
4.8
~
-
2,3,4,6-TeCPa
ND (0.02)
ND (0.02)
ND (0.02) to 15
ND (0.02) to 5.1
ND (0.02) to 0.17
+
+
+
+
+
-
2,3,4,6-TeCP
(Na salt)b
0.7
5.2
9.5
5.6
0.7
0.5
10
70
70
10
-
oo
aSource: Firestone et al. (1972); because of poor recoveries, the authors stated that actual CDD/CDF levels may have been considerably higher than those
reported.
bSources: Rappe et al. (1978a, b); common Scandinavian commercial chlorophenols.
ND = Not detected (value in parenthesis is the detection limit, if reported)
+ = Detected but not quantified
~ = No information given
-------
Table 8-10. CDD/CDF concentrations (historical and current) (ug/kg) in technical-grade pentachlorophenol
(PCP) products
Congener/
congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD1
Total 2,3,7,8-CDF1
Total I-TEQo,,1
Total TEQoF-WHCV
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF1
1973"
__
-
-
-
-
-
-
__
-
-
-
-
-
-
-
-
-
..
-
-
—
ND (20)
ND (30)
5,500
98,000
220,000
40
250
22,000
150,000
160,000
655,790
1978"
__
-
-
-
-
-
-
__
-
-
-
-
-
-
-
-
130,000
..
-
-
—
-
-
-
-
-
900
4,000
32,000
120,000
130,000
286,900
1979C
__
-
-
-
-
-
-
__
-
-
-
-
-
-
-
-
-
..
-
-
—
-
-
10,100
296,000
1,386,000
-
1,400
9,900
88,000
43,000
1,834,400
1984"
ND(10)
ND(10)
-
2,200
100
100,000
610,000
ND(10)
-
-
-
-
-
-
-
-
130,000
71l%a,°J?)0
1,970
1,304
ND(10)
ND(10)
4,500
135,000
610,000
ND(10)
-
-
62,000
130,000
941,500
1985"
ND (0.05)
ND(1)
6
2,565
44
210,000
1,475,000
ND (0.5)
ND(1)
ND(1)
49
5
5
ND(1)
34,000
4,100
222,000
1,687,615
260,159
4,445
2,918
ND
ND
4,694
283,000
1,475,000
6
10
1,982
125,000
222,000
2,111,692
1986"
ND (0.05)
ND(1)
8
1,532
28
106,000
930,000
ND (0.5)
ND(1)
ND(1)
34
4
ND(1)
ND(1)
29,000
6,200
233,000
1,037,568
268,238
O -7Q C
2., 1 JJ
ND
ND
2,925
134,000
930,000
ND
3
1,407
146,000
233,000
1,447,335
1987'
ND (0.03)
1
ND(1)
831
28
78,000
733,000
ND(O.l)
0.5
1.5
125
ND(1)
32
ND(1)
11,280
637
118,000
811,860
130,076
1,853
1,088
1.9
6.5
1,700
154,000
733,000
0.8
141
4,300
74,000
118,000
1,085,150
1987s
ND (0.05)
9
ND(1)
1,480
53
99,900
790,000
ND(O.l)
0.2
0.9
163
ND(1)
146
ND(1)
19,940
980
137,000
891,435
158,230
2,321
1,488
0.4
15.2
3,300
198,000
790,000
0.4
343
13,900
127,000
137,000
1,269,559
1985-88"
ND (0.05)
ND(1)
8
600
13
89,000
2,723,000
ND (0.5)
ND(1)
ND(1)
67
2
ND(1)
ND(1)
22,000
3,400
237,000
2,812,621
262,469
4,173
ND
ND
912
117,000
2,723,000
ND
200
1,486
99,000
237,000
3,178,598
1991'
ND
ND
-
-
-
-
1,100,000
ND
ND
ND
-
-
-
-
-
-
170,000
1,100,000
170,000
> 1,270
>127
ND(10)
ND(10)
8,900
130,000
1,100,000
ND(10)
ND(10)
14,000
36,000
170,000
1,458,900
1988-99"
__
-
-
-
-
-
-
__
-
-
-
-
-
-
-
-
-
..
-
-
—
ND(1)
ND(10)
1,440
55,560
-
ND(10)
ND(10)
3,070
36,530
-
960,000
1988-99'
ND(0.5)
-
-
-
-
-
-
__
-
-
-
-
-
-
-
-
-
..
-
-
—
ND
3
1,490
48,430
191,700
48
520
13,650
76,090
136,310
468,241
Un-
known11
ND(10)
ND(10)
ND(10)
860
20
36,400
296,810
ND(10)
ND(10)
ND(10)
200
ND (20)
ND (20)
ND (20)
2,000
140
19,940
334,090
22,280
810
525
-
-
-
-
-
-
-
-
-
-
--
oo
oo
aSource: Buser and Bosshardt (1976); mean of 10 samples of "high" CDD/CDF-content PCP received from Swiss commercial sources in 1973.
bSource: Rappe et al. (1978b); sample of U.S. origin, "presumably prepared by alkaline hydrolysis of hexachlorobenzene."
°Source: U.S. Department of Health and Human Services (1989); composite of technical-grade materials produced in 1979 by Monsanto Industrial Chemical Co.
(St. Louis, MO), Reichhold Chemicals, Inc. (White Plains, NY), and Vulcan Materials Co. (Birmingham, AL).
-------
Table 8-10. CDD/CDF concentrations (historical and current) in technical-grade pentachlorophenol (PCP)
products (ug/kg) (continued)
d Source: Cull et al. (1984); mean of four "recent" production batches from each of two manufacturers of technical PCP using three different analytical methods;
ANOVA showed no statistically significant difference in CDD/CDF concentrations between the eight samples (samples obtained in the United Kingdom).
e Source: Letter dated February 7, 1997, from John Wilkinson, Pentachlorophenol Task Force, to Matthew Lorber, U.S. EPA; average of monthly batch samples
for the period January 1987 to August 1996.
f Source: Hagenmaier and Brunner (1987); sample of Witophen P (Dynamit Nobel - Lot no. 7777) (obtained in Germany).
g Source: Hagenmaier and Brunner (1987); sample of PCP produced by Rhone Poulenc (obtained in Germany).
h Source: Letter dated February 7, 1997, from John Wilkinson, Pentachlorophenol Task Force, to Matthew Lorber, U.S. EPA; samples of "penta" manufactured in
1985, 1986, and 1988.
1 Source: Harrad et al. (1991); PCP-based herbicide formulation from the New York State Department of Environmental Conservation.
J Source: Letter dated March 5, 1997, from Thomas Mitchell, KMG-Bernuth, to Matthew Lorber, U.S. EPA; average of monthly batch samples for the period
February 1987 to December 1996 (excluding the following months, for which data were not available: February 1993, January 1992, December 1991,
September 1991, December 1988, and September 1988).
k Source: Schecter et al. (1997); sample found stored in a barn in Vermont.
1 Calculated assuming nondetects were zero.
ND = Not detected (value in parenthesis is the detection limit)
- = No information given
-------
Table 8-11. Historical CDD/CDF concentrations (ug/kg) in pentachlorophenol-Na
(PCP-Na)
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1, 2,3,4 ,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1, 2,3,4 ,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDDh
Total 2,3,7,8-CDFh
Total I-TEQDFh
Total TEQDF-WHO98h
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDFh
1969a
—
—
__
3,600
—
—
—
—
—
—
—
-
-
--
—
—
—
--
—
—
17,000
9,600
3,600
-
-
—
—
--
30,200
1973b
—
—
__
--
—
—
—
—
—
—
—
-
-
--
—
—
—
--
140
40
140
1,600
4,000
ND(20)
60
1,400
4,300
4,300
15,980
1973C
—
—
__
--
—
-
-
-
-
-
-
-
-
--
—
—
—
--
50
ND(30)
3,400
38,000
110,000
ND(20)
40
11,000
47,000
26,500
235,990
1987d
0.23
18.2
28.3
2,034
282
9,100
41,600
1.8
8.2
6.6
48
69
ND(1)
87
699
675
37,200
53,062.7
38,794.6
452
390
27
213
3,900
18,500
41,600
82
137
3,000
13,200
37,200
117,859
1987e
0.51
3.2
13.3
53
19
3,800
32,400
0.79
1.9
1.1
4.6
1.3
1.3
4.6
197
36
4,250
36,289
4,498.6
89.5
58.1
52
31
230
5,800
32,400
12
27
90
860
4,250
43,752
1992f
0.076
18.7
96
4,410
328
175,400
879,000
ND(1)
ND(4)
ND(4)
27.6
21.9
9.8
103
9,650
2,080
114,600
1,059,252.8
126,492.3
3,374
2,489
3.6
142.7
9,694
260,200
879,000
10.1
88.4
9,082.3
75,930
114,600
1,348,751
1980s8
ND(1.4)
28.3
ND(6.1)
4,050
ND(1.4)
33,800
81,000
149
319
324
ND(2.8)
225
480
ND(385)
6,190
154
36,000
118,878.3
43,841
1,201
1,110
1.9
140
14,000
100,000
81,000
1,200
6,400
49,000
91,000
36,000
378,742
aSource: Firestone et al. (1972); mean of two samples of PCP-Na obtained in the United States between 1967 and
1969.
bSource: Buser and Bosshardt (1976); mean of five samples of "low" CDD/CDF-content PCP-Na received from
Swiss commercial sources.
°Source: Buser and Bosshardt (1976); sample of "high" CDD/CDF-content PCP-Na received from a Swiss
commercial source.
dSource: Hagenmaier and Brunner (1987); sample of Dowicide-G purchased from Fluka; sample obtained in
Germany.
eSource: Hagenmaier and Brunner (1987); sample of Preventol PN (Bayer AG); sample obtained in Germany.
fSource: Santl et al. (1994); 1992 sample of PCP-Na from Prolabo, France.
gSource: Palmer et al. (1988); sample of a PCP-Na formulation collected from a closed sawmill in California in the
late 1980s.
hCalculated assuming nondetect values were zero.
ND = Not detected (value in parenthesis is the detection limit)
~ = No information given
8-20
-------
Table 8-12. Summary of specific dioxin-containing wastes that must comply
with land disposal restrictions"
EPA
hazardous
waste
number
Waste description
Land disposal
restriction
effective date
Regulated
waste
constituent
F020
Wastes (except wastewater and spent carbon
from HC1 purification) from the production or
manufacturing use (as a reactant, chemical
intermediate, or component in a formulating
process) of tri- or tetrachlorophenol or of
intermediates used to produce their pesticide
derivatives. (This listing does not include
wastes from the production of hexachlorophene
from highly purified 2,4,5-trichlorophenol.)
November 8, 1988
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
F021
Wastes (except wastewater and spent carbon
from HC1 purification) from the production or
manufacturing use (as a reactant, chemical
intermediate, or component in a formulating
process) of pentachlorophenol or of
intermediates used to produce its derivatives.
November 8, 1988
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
F022
Wastes (except wastewater and spent carbon
from HC1 purification) from the manufacturing
use (as a reactant, chemical intermediate, or
component in a formulating process) of tetra-,
penta-, or hexachlorobenzenes under alkaline
conditions.
November 8, 1988
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
F023
Wastes (except wastewater and spent carbon
from HC1 purification) from the production of
materials on equipment previously used for the
production or manufacturing use (as a reactant,
chemical intermediate, or component in a
formulating process) of tri- and
tetrachlorophenols. (This listing does not
include wastes from equipment used only for the
production or use of hexachlorophene from
highly purified 2,4,5-trichlorophenol.)
November 8, 1988
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
F026
Wastes (except wastewater and spent carbon
from HC1 purification) from the production of
materials on equipment previously used for the
manufacturing use (as a reactant, chemical
intermediate, or component in a formulating
process) of tetra-, penta-, or hexachlorobenzene
under alkaline conditions.
November 8, 1988
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
8-21
-------
Table 8-12. Summary of specific dioxin-containing wastes that must comply
with land disposal restrictions" (continued)
EPA
hazardous
waste
number
F027
F028
F039
K043
K099
Waste description
Discarded unused formulations containing tri-,
tetra-, or pentachlorophenol or discarded unused
formulations containing compounds derived
from these chlorophenols. (This listing does not
include formulations containing
hexachlorophene synthesized from prepurified
2,4,5 -trichlorophenol as the sole component.)
Residues resulting from the incineration or
thermal treatment of soil contaminated with
EPA Hazardous Wastes No. F020-F023, F026,
and F027
Leachate (liquids that have percolated through
land-disposed wastes) resulting from the
disposal of more than one restricted waste
classified as hazardous under Subpart D of 40
CFR 268. (Leachate resulting from the disposal
of one or more of the following EPA hazardous
wastes and no other hazardous wastes retains its
EPA hazardous waste number(s): F020, F021,
F022, F026, F027, and/or F028.)
2,6-Dichlorophenol waste from the production
of2,4-D
Untreated wastewater from the production of
2,4-D
Land disposal
restriction
effective date
November 8, 1988
November 8, 1988
August 8, 1990
(wastewater)
May 8, 1992
(nonwastewater)
June 8, 1989
August 8, 1988
Regulated
waste
constituent
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
TCDDs
PeCDDs
HxCDDs
TCDFs
PeCDFs
HxCDFs
Tor wastewater, the treatment standard for all regulated waste constituents except PeCDFs is 0.063 ng/L; the
standard for PeCDFs is 0.035 ng/L. For nonwastewater, the treatment standard for all regulated waste
constituents is 1 ng/kg. Treatment standards are based on incineration to 99.9999% destruction and removal
efficiency.
Source: 40 CFR 268.
8-22
-------
constituents, including chlorophenol wastes (waste codes F020 and F021). The regulations
prohibit the land disposal of these wastes until they are treated to a level below the routinely
achievable DLs for the EPA hazardous waste numbers listed in Table 8-12 for each of the
following congener groups: TCDDs, PeCDDs, HxCDDs, TCDFs, PeCDFs, and HxCDFs.
Wastes from PCP-based wood-preserving operations (waste codes K001 and F032) are also
regulated as hazardous wastes under RCRA (40 CFR 261).
EPA's Office of Water promulgated effluent limitations for facilities that manufacture
chlorinated phenols and discharge treated wastewater (40 CFR 414.70). These effluent
limitations do not specifically regulate CDDs or CDFs. The effluent limitations for the
individually regulated chlorinated phenols are less than or equal to 39 jig/L for facilities that use
biological end-of-pipe treatment.
Di- and trichlorophenols are subject to reporting under the dioxin/furan test rule, which is
discussed in Section 8.3.7 of this report. Since the effective date of that rule (June 5, 1987), only
the 2,4-dichlorophenol isomer has been commercially produced in (or imported to) the United
States, and as noted in Table 8-9, no CDDs/CDFs were detected in the product. Testing is
required for the other di- and trichlorophenols if manufacture or importation resumes. Similarly,
tetrachlorophenols were subject to reporting under the Dioxin/Furan Pesticide Data Call-In (DCI)
(discussed in Section 8.3.8 of this report). Since issuance of the DCI, the registrants of
tetrachlorophenol-containing pesticide products have elected to no longer support the registration
of their products in the United States.
In January 1987, EPA entered into a settlement agreement with PCP manufacturers that
set limits, effective in February 1989, on the allowed uses of PCP and its salts and the maximum
allowable concentrations of 2,3,7,8-TCDD and HxCDDs. Section 8.3.8 discusses the 1987 PCP
settlement agreement and includes estimates of current releases of CDDs/CDFs associated with
the use of PCP in the United States. Section 11.3.1 (Chapter 11) provides an estimate of the
amount of CDDs/CDFs that may have entered the environment or that are contained in treated
wood products as a result of prior use of PCP and PCP-Na.
Since the late 1980s, U.S. commercial production of chlorophenols has been limited to
2,4-dichlorophenol and PCP. As noted above, disposal of wastes generated during the
manufacture of chlorophenol s is strictly regulated, and thus releases to the environment are
expected to be negligible. With regard to releases associated with the use of 2,4-dichlorophenol,
no CDDs/CDFs have been detected in 2,4-dichlorophenol. Releases associated with the use of
PCP are presented in Sections 8.3.8 and 11.3.1.
-------
8.3.2. Chlorobenzenes
Chlorobenzenes have been produced in the United States since 1909. U.S. production
operations were developed primarily to provide chemical raw materials for the production of
phenol, aniline, and various pesticides based on the higher-chlorinated benzenes. Because of
(incremental) changes in the processes used to manufacture phenol and aniline and the phaseout
of highly chlorinated pesticides such as DDT and hexachlorobenzene, U.S. production of
Chlorobenzenes in 1988 had decreased to 50% of the peak production level, in 1969.
Chlorobenzenes can be produced via three methods: (1) electrophilic substitution of
benzene (in liquid or vapor phase) with chlorine gas in the presence of a metal salt catalyst, (2)
oxidative chlorination of benzene with HC1 at 150 to 300°C in the presence of a metal salt
catalyst, and (3) dehydrohalogenation of hexachlorocyclohexane wastes at 200 to 240°C with a
carbon catalyst to produce trichlorobenzene, which can be further chlorinated to produce higher-
chlorinated benzenes (Ree et al., 1988; Hutzinger and Fiedler, 1991b; Bryant, 1993).
All Chlorobenzenes currently manufactured in the United States are produced by the
electrophilic substitution process using liquid-phase benzene (i.e., temperature is at or below
80°C). FeCl3 is the most common catalyst employed. Although this method can be used to
produce mono- through hexachlorobenzene, the extent of chlorination is controlled to yield
primarily monochlorobenzene and dichlorobenzene. The finished product is a mixture of
Chlorobenzenes, and refined products must be obtained by distillation and crystallization (Bryant,
1993).
CDDs/CDFs can be produced inadvertently during the manufacture of Chlorobenzenes by
nucleophilic substitution and pyrolysis mechanisms (Ree et al., 1988). The criteria required for
production of CDDs/CDFs via nucleophilic substitution are oxygen as a nuclear substituent (i.e.,
presence of chlorophenols) and production or purification of the substance under alkaline
conditions. Formation via pyrolysis requires reaction temperatures above 150°C (Ree et al.,
1988; Hutzinger and Fiedler, 1991b). The liquid-phase electrophilic substitution process
currently used in the United States does not meet either of these criteria. Although Ree et al. and
Hutzinger and Fiedler state that the criteria for formation of CDDs/CDFs via nucleophilic
substitution may be present in the catalyst neutralization and purification/distillation steps of the
manufacturing process, Opatick (1995) states that the chlorobenzene reaction product in U.S.
processes remains mildly acidic throughout these steps.
Table 8-13 summarizes the very limited published information on CDD/CDF
contamination of chlorobenzene products. The presence of CDDs/CDFs has been reported in
tri-, penta-, and hexachlorobenzene. No CDDs/CDFs have been reported in mono- or
dichlorobenzene. Conflicting data exist concerning the presence of CDDs/CDFs in
8-24
-------
Table 8-13. CDD/CDF concentrations (ug/kg) in chlorobenzenes
Congener/
congener group
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
MCBza
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND
1,2-DCBz
(for
synthesis)3
ND (0.02)
ND, (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
ND (0.02)
NB. (0.02)
ND (0.02)
0.8
1,2,4-TrCBz
("pure")"
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND
Mixed
TrCBz
(47%)a
0.027
0.14
0.259
0.253
0.081
0.736
0.272
0.091
0.03
0.016
1.9
1,2,4,5-TCBz
(99%)a
N ( .Oj2j
0.5
0.8
0.4
0.03
0.2
0.8
1.5
2.1
6 ^
PeCBz
(98%)a
ND (0.02)
ND (0 02)
^ 0.02
0.02
0.05
0.02
ND (0.02)
ND (0 02)
v O.r
0.1
HCBz
(97%)a
ND (20)
ND (20)
ND (20)
470
6,700
ND (20)
ND (20)
ND (20)
455
2,830
10,455
HCBzb
~
~
~
~
50-212,000
~
~
~
~
350-58,300
400-270,300
oo
aSource: Hutzinger and Fiedler (1991b); unpublished results of tests performed at the University of Bayreuth, GeiQnany, and by Dr. H. Hagenmaier.
bSource: Villanueva et al. (1974); range of three samples of commercially available HCBz.
ND = Not detected (value in parenthesis is the detection limit, if reported)
~ = No information given
-------
trichlorobenzene. One study (Villanueva et al., 1974) detected no CDDs/CDFs in one sample of
1,2,4-TCBz at a DL of 0.1 ng/kg. Hutzinger and Fiedler (1991b) reported unpublished results of
a study by Dr. Hans Hagenmaier showing CDD/CDF congener group concentrations ranging
from 0.02 to 0.074 |ig/kg in a sample of mixed trichlorobenzene. Because the TCBz examined
by Hagenmaier contained about 2% hexachlorocyclohexane, it is reasonable to assume that it was
produced by dehydrohalogenation of hexachlorocyclohexane (a manufacturing process not
currently used in the United States).
8.3.2.1. Regulatory Actions for Chlorobenzenes
EPA determined, as part of the Federal Insecticide, Fungicide, and Rodenticide Act
(FIFRA) DCI (discussed in Section 8.3.8), that the 1,4-dichlorobenzene manufacturing processes
used in the United States are not likely to form CDDs/CDFs. Mono-, di-, and trichlorobenzene
are listed as potential precursor chemicals under the TSCA dioxin/furan test rule and are subject
to reporting (see Section 8.3.7). In addition, EPA issued a Significant New Use Rule (SNUR)
under Section 5(a)(2) of TSCA on December 1, 1993 (effective January 14, 1994) for
pentachlorobenzene and 1,2,4,5-tetrachlorobenzene (Federal Register, 1993c). This rule requires
that EPA be notified at least 90 days before the manufacture, import, or processing of either of
these compounds in amounts of 10,000 pounds or greater per year per facility for any use. All
registrations of pesticide products containing hexachlorobenzene were cancelled in the mid-
1980s (Carpenter et al., 1986).
OSW promulgated land disposal restrictions on wastes (i.e., wastewaters and
nonwastewaters) resulting from the manufacture of chlorobenzenes (40 CFR 268). Table 8-12
lists all solid wastes for which EPA specifically regulates CDDs and CDFs, including
chlorobenzene wastes, as hazardous constituents. The regulations prohibit the land disposal of
these wastes until they are treated to a level below the routinely achievable DLs in the waste
extract listed in Table 8-12 for each of the following congener groups: TCDDs, PeCDDs,
HxCDDs, TCDFs, PeCDFs, and HxCDFs.
EPA's Office of Water promulgated effluent limitations for facilities that manufacture
chlorinated benzenes and discharge treated wastewater (40 CFR 414.70). These effluent
limitations do not specifically address CDDs and CDFs. The following chlorinated benzenes are
regulated: chlorobenzene; 1,2-dichlorobenzene; 1,3-dichlorobenzene; 1,4-dichlorobenzene;
1,2,4-trichlorobenzene; and hexachlorobenzene. The effluent limitations for the individual
regulated chlorinated benzenes are less than or equal to 77 jig/L for facilities that use biological
end-of-pipe treatment and less than or equal to 196 jig/L for facilities that do not use biological
end-of-pipe treatment.
8-26
-------
Since at least 1993, U.S. commercial production of chlorobenzenes has been limited to
monochlorobenzene, 1,2-dichlorobenzene, 1,4-dichlorobenzene, and, to a much lesser extent,
1,2,4-trichlorobenzene. As noted above, CDD/CDF formation is not expected under the normal
operating conditions of the processes currently used in the United States to produce these four
chemicals. No tetra-, penta-, or hexachlorinated benzenes are now intentionally produced or used
in the United States (Bryant, 1993). Thus, releases of CDDs/CDFs from the manufacture of
chlorobenzenes in 1995 were estimated to be negligible. Because the information available on
CDD/CDF content of mono- through pentachlorobenzene is very limited and is based primarily
on unpublished European data, and because information on the chlorobenzene manufacturing
processes in place during 1987 is not readily available, no emission estimates can be made for
1987.
8.3.3. Chlorobiphenyls
PCBs are manufactured by the direct batch chlorination of molten biphenyl in the
presence of a catalyst, followed by separation and purification of the desired chlorinated biphenyl
fractions. During the manufacture of PCBs, the inadvertent production of CDFs also occurs.
This section focuses on levels of CDD/CDF contamination that may have been present in PCB
products.
CDFs have been shown to form when PCB-containing transformers and capacitors
undergo malfunctions or are subjected to fires that result in accidental combustion of the
dielectric fluid. The direct releases of dioxin-like PCBs from sources is reviewed in Chapter 10.
During the commercial production of PCBs, thermal oxidative cyclization under alkaline
conditions resulted in the inadvertent production of CDFs in most of the commercial PCB
mixtures (Brown et al., 1988; ATSDR, 1993). Bowes et al. (1975a) first reported detection of
CDFs in Aroclor products; samples of unused Aroclors manufactured in 1969 and 1970 were
found to have CDF (TCDF through HxCDF) concentrations ranging from 0.8 to 2 mg/kg. Bowes
et al. used congener-specific analytical methodology and detected 2,3,7,8-TCDF and 2,3,4,7,8-
PeCDF at concentrations ranging from 0.11 to 0.33 mg/kg and 0.12 to 0.83 mg/kg, respectively,
in unused samples of Aroclor 1254 and Aroclor 1260. The presence of CDDs in commercial
PCB mixtures, although at much lower concentrations than those of the CDFs, was reported by
Hagenmaier (1987) and Malisch (1994). Table 8-14 presents the CDF and CDD congener group
concentrations reported by Bowes et al. (1975a) and those reported in subsequent years for
unused PCBs by Erickson (1986), ATSDR (1993), Hagenmaier (1987), and Malisch (1994).
Several researchers have reported concentrations of specific CDD/CDF congeners in
commercial PCB mixtures (Bowes et al., 1975b; Brown et al., 1988; Hagenmaier, 1987; Malisch,
1994). Table 8-15 presents the results of these four studies. Only the Hagenmaier and Malisch
8-27
-------
Table 8-14. Concentrations of CDD/CDF congener groups in unused commercial polychlorinated biphenyl (PCB)
mixtures (mg/kg)
PCB mixture
Aroclor 1016
Aroclor 1242
Aroclor 1242
Aroclor 1242
ClophenA-30
ClophenA-30
Aroclor 1248
ClophenA-40
Kanechlor 400
Aroclor 1254
Aroclor 1254
Aroclor 1254
Aroclor 1254
Clophen A-50
Aroclor 1260
Aroclor 1260
Aroclor 1260
Aroclor 1260
Clophen A-60
Clophen A-60
Clophen A-60
Phenoclor DP-6
Clophen T-64
Prodelec 3010
Year of
manufacture
1972
—
—
—
1969
—
1969
1970
—
-
1969
—
—
-
—
-
CDD congener group concentrations
TCDD
—
—
—
0.0007
ND
ND
—
—
ND
-
—
—
0.0004
ND
—
-
PeCDD
—
—
—
ND
ND
ND
—
—
ND
-
—
—
0.002
ND
—
-
HxCDD
—
—
—
0.001
ND
__
ND
—
—
ND
-
—
—
0.002
ND
_
HpCDD
—
—
—
0.006
0.005
__
0.012
—
—
0.011
-
—
—
0.003
0.014
—
-
OCDD
—
—
—
0.031
0.025
0.030
—
—
0.027
-
—
—
0.015
0.032
—
-
Total
CDD
—
—
—
0.039
0.030
0.042
—
—
0.038
-
—
—
0.022
0.046
—
-
CDF congener group concentrations
TCDF
ND
0.07
2.3
0.25
6.377
0.713
0.5
1.289
0.1
0.2
0.02
0.05
5.402
0.3
0.1
0.8
0.2
15.786
16.34
1.4
0.7
0.3
1.08
PeCDF
ND
0.03
2.2
0.7
2.402
0.137
1.2
0.771
0.2
0.4
0.2
0.1
2.154
1
0.4
0.9
0.3
11.655
21.164
5
10
1.73
0.35
HxCDF
ND
0.003
ND
0.81
0.805
0.005
0.3
0.144
1.4
0.9
0.6
0.02
2.214
1.1
0.5
0.5
0.3
4.456
7.63
2.2
2.9
2.45
0.07
HpCDF
—
0.108
0.001
0.02
-
0.479
1.35
—
—
1.517
2.522
0.82
-
OCDF
—
—
—
0.016
ND
__
0.011
—
—
0.069
-
—
—
0.639
1.024
—
-
Total
CDF
ND
0.15
4.5
1.9
9.708
0.855
22.2352
1.7
1.5
0.8
0.2
10.318
3.8
1
2.2
0.8
34.052
48.681
8.6
13.6
5.4
2
Source
a
b,c
b,c
b
d
e
b
e
b,c
a
a
b, c
b
e
b, c
a
b,c
a
d
e
a
a
b
b
oo
to
oo
aSource: Bowes etal. (1975a).
bSource: Erickson (1986).
cSource: ATSDR(1993).
dSource: Malisch (1994).
eSource: Hagenmaier (1987).
ND = Not detected
- = No information given
-------
Table 8-15. 2,3,7,8-Substituted congener concentrations in unused polychlorinated biphenyl (PCB) mixtures (ug/kg)
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDFe
Total I-TECV
Total TEQop-WFKV
Congener concentrations in Clophens
A-30"
ND
ND
ND
0.8
ND
5.6
31.1
1,032.6
135.8
509.2
301.4
65.3
ND
50.6
43.7
22.5
15.7
0.7
ND
1.2
5.6
31.1
6,376.6
2,402.4
804.8
108.3
15.7
9,746.4
407.2
407.2
A-30"
ND
ND
ND
ND
ND
2.4
24.7
36.9
14.9
13.1
1.9
0.8
ND
0.1
0.6
ND
ND
ND
ND
ND
5.4
24.7
713
136.5
5.1
0.8
ND
885.5
11.3
11.3
A-40"
ND
ND
ND
ND
ND
4.4
30.3
250.2
52.7
171.3
48.4
19.6
0.7
6.8
7
2.8
11.4
ND
ND
ND
11.6
30.3
1,289.4
770.8
143.6
19.5
11.4
2,276.6
A-50"
ND
ND
ND
ND
ND
5.3
26.9
1,005.7
155.2
407.5
647.5
227.5
8.3
62.5
205.5
72 2
69.2
ND
ND
ND
11
26.9
5,402.3
2,153.7
2,213.8
478.8
69.2
10,355.7
409.6
409.5
A-60"
ND
0.1
0.2
ND
ND
2.5
14.9
2,287.7
465.2
1,921.9
1,604.2
157.6
42.8
369.5
480.6
321.7
639.2
0.4
2
1.8
3
14.9
15,785.7
11,654.6
4,455.8
1,517
639.2
34,074.4
1,439.2
1,439
A-60"
ND
ND
ND
ND
ND
6.8
32.3
3,077.2
1,750.8
2,917.0
2,324.1
351.3
19
4,08.3
1,126.1
304
1,024.3
ND
ND
ND
13.5
32.3
16,340
21,164
7,630.2
2,522.3
1,024.3
48,726.6
2,179
2,178
Congener concentrations in Aroclors
1016C
__
—
—
—
—
-
-
0.1
—
1.75
-
—
0.08
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1242C
__
—
—
—
—
-
-
40.1
—
40.8
-
—
0.26
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1248"
__
—
—
—
—
-
-
330
—
830
—
—
-
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1254C
__
—
—
—
—
-
-
28
—
110
—
—
28.8
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1254C
__
—
—
—
—
-
-
20.9
—
179
—
—
28.7
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1254C
__
-
-
—
—
-
-
55.8
—
105
—
—
19.4
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1254"
__
—
—
—
—
-
-
110
—
120
—
—
-
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1260C
__
—
—
—
—
-
-
63.5
—
135
—
—
5.1
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1260C
__
—
—
—
—
-
-
6.88
—
58.2
—
—
9.7
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
1260C
__
—
—
—
—
-
-
29
—
112
—
—
10.7
-
-
-
-
—
—
-
-
—
—
—
-
-
—
-
oo
to
VO
"Source: Malisch(1994).
bSource: Hagenmaier (1987).
"Source: Brown etal. (1988).
dSource: Bowes (1975b).
Calculated assuming nondetect values were zero.
ND = Not detected
~ = No information given
-------
studies, however, reported the concentrations of all 2,3,7,8-substituted CDDs and CDFs. It is
evident from the table that major variations are found in the levels of 2,3,7,8-TCDF and
2,3,4,7,8-PeCDF in the Clophen mixtures reported by Hagenmaier and Malisch and the
corresponding levels in the Aroclor mixtures reported by Bowes et al. and Brown et al.
Brown et al. (1988) compared the levels of 2,3,7,8-TCDF; 2,3,4,7,8-PeCDF; and
1,2,3,7,8,9-HxCDF in used samples (from previously used capacitors and transformers) and
unused samples of Aroclor 1016, 1242, 1254, and 1260. The concentration ranges reported for
the used and unused Aroclors were similar, leading Brown et al. (1988) to conclude that CDFs
are not formed during the normal use of PCBs in electrical equipment.
8.3.4. Ethylene Dichloride/Vinyl Chloride Monomer/Polyvinyl Chloride Manufacturing
In the United States the manufacture of PVC is an integrated manufacturing process.
This means that most manufacturing facilities produce all the precursors and chemical
intermediates necessary to manufacture PVC as well as PVC resins and products. For example,
ethylene dichloride (EDC) is used to produce vinyl chloride monomer (VCM). VCM is then
converted to PVC resins. The resins are used to manufacture various PVC products.
PVC resins are produced from the polymerization of VCM. VCM is typically produced
by the thermal dehydrochlorination (commonly known as cracking) of EDC. The cracking of
EDC requires elevated pressure (20 to 30 atm) and temperature (450 to 650°C) and yields VCM
and HC1 at about a 1:1 molar ratio. EDC is produced by two different methods: (1) direct
chlorination of ethylene with chlorine in the presence of a catalyst at a temperature of 50 to 60°C
and pressure of 4 to 5 atm, and (2) oxychlorination, which involves reaction of ethylene with HC1
and oxygen in the presence of a catalyst at temperatures generally less than 325°C. The primary
source of HC1 for the oxychlorination process is the HC1 produced from the cracking of EDC to
form VCM. Most VCM manufacturing facilities are integrated with EDC production facilities
(The Vinyl Institute, 1998).
Although it has been generally recognized that CDDs/CDFs can be formed during the
manufacture of EDC, VCM, and PVC, manufacturers and environmental public interest groups
have disagreed as to the quantity of CDDs/CDFs that are formed and released to the environment
in wastes and possibly in PVC products. Although EPA regulates emissions from EDC/VCM
production facilities under the Clean Water Act (40 CFR 61), the Clean Air Act (40 CFR 414),
and RCRA (40 CFR 268, waste codes F024, KOI9, and K020), CDDs/CDFs are not specifically
regulated pollutants; as a consequence, monitoring data for CDDs/CDFs in emissions were
generally lacking until the early 1990s.
Greenpeace International initially determined that CDDs and CDFs can be formed during
the manufacture of PVC. In 1993, it issued a report on CDD/CDF emissions associated with the
8-30
-------
production of EDC/VCM (Greenpeace, 1993). Greenpeace estimated that 5 to 10 g I-TEQDF
were released to the environment (air, water, and ground combined) annually for every 100,000
metric tons of VCM produced. This emission factor was based on data gathered by Greenpeace
on four European plants. The Vinyl Institute responded with a critique of the Greenpeace report
(ChemRisk, 1993). Miller (1993) summarized the differing views of the two parties. According
to Miller, European PVC manufacturers claimed the emission factor was 0.01 to 0.5 g I-
TEQDF/100,000 metric tons of VCM, but although Greenpeace and ChemRisk used basically the
same monitoring information to develop their emission factors, Greenpeace adjusted the
emission factor to account for unquantified fugitive emissions and waste products that contain
unspecified amounts of CDDs/CDFs.
In 1995, Greenpeace issued a second report (Stringer et al., 1995) reiterating the
organization's concern that the generation and emission of CDDs/CDFs may be significant and
urging that further work be initiated to quantify and prevent emissions. Stringer et al. presented
the results of analyses of three samples of chlorinated wastes obtained from U.S. EDC/VCM
manufacturing facilities. The three wastes were characterized according to EPA hazardous waste
classification numbers as an F024 waste (waste from the production of short-chain aliphatics by
free radical-catalyzed processes), a KOI9 waste (heavy ends from the distillation of ethylene from
EDC production), and a probable K020 waste (heavy ends from distillation of vinyl chloride in
VCM manufacture). Table 8-16 presents the analytical results reported by Stringer et al. This
study acknowledged that because EDC/VCM production technologies and waste treatment and
disposal practices are very site-specific, the limited information available on CDD/CDF
generation and emissions made it difficult to quantify amounts of CDDs/CDFs generated and
emitted.
In response to the lack of definitive studies, and at the recommendation of EPA, U.S.
PVC manufacturers began an extensive monitoring program, the Dioxin Characterization
Program (DCP). The objective of the DCP was to evaluate the extent and magnitude of
potential CDD/CDF releases to air, water, and land, as well as the potential for PVC product
contamination. Manufacturers performed emissions and product testing at several facilities that
were representative of various manufacturing and process control technologies. In 1998, The
Vinyl Institute completed studies of CDD/CDF releases in wastewater, wastewater treatment
plant solids, and stack gases, as well as studies of the CDD/CDF content of products (PVC resins
and EDC sold as products) (The Vinyl Institute, 1998).
In September 2002, the CCC met to review dioxin release estimates for 2000 submitted
by various EDC/VCM manufacturing facilities. Several companies provided stack gas emissions
and wastewater release data as well as a discussion of how they generated the release and transfer
estimates reported in the Toxics Release Inventory for 2000. In March 2004, the CCC met again
8-31
-------
Table 8-16. Reported CDD/CDF concentrations (jig/kg) in wastes from
polyvinyl chloride (PVC) manufacture
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQnF-WHO9S
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
F024 waste
0.37
0.14
0.3
0.14
0.11
4.2
15
0.91
9.5
1.6
110
24
9.5
3.1
250
51
390
20.3
849.6
20
19.7
3.1
3.6
1.3
5
15
15
65
300
450
390
1,248
KOI 9 waste
260
890
260
330
620
920
1,060
680
975
1,050
10,100
9,760
21,800
930
13,400
1,340
43,500
4,340
103,535
5,928
6,333
1,230
3,540
3,950
1,270
1,060
20,600
45,300
63,700
16,600
43,500
200,750
K020 waste
0.06
0.05
0.08
0.06
0.07
0.89
3
0.44
1.8
0.58
11
2.4
1.3
0.89
38
6
650
4.21
712.4
3.2
2.6
1.9
1.7
a
1.7
O
6
11
27
58
650
760.3
aCongener group concentration reported in source is not consistent with reported congener concentrations.
Source: Stringer etal. (1995).
8-32
-------
to discuss the results, to date, of its CDD/CDF data validation study for PVC/EDC/VCM and
chlor-alkali facilities. The study's goal was to provide facility-specific water, air, and land
release estimates for 2000 and 2002. As of the date of this report, data validation studies were
provided for 17 of 20 facilities in the CCC that were considered chlor-alkali production facilities
and PVC/EDC/VCM manufacturing plants.
8.3.4.1. Water Releases
This section presents estimates of releases of dioxin-like compounds in wastewater
discharges to surface waters for 2000 and 1995 from the integrated EDC, VCM, and PVC
manufacturing facilities in the United States. Site-specific testing of the wastewater from
facilities operated by Dow Chemical Co., Occidental Chemical Corp., Georgia Gulf Corp., PPG
Industries, and DuPont provide the basis for estimating annual releases of dioxin-like compounds
in 2000 (CCC, 2005, 2003a, b, c, d). In total, these tests represent site-specific analysis of 17
manufacturing facilities throughout the United States. The site-specific releases of dioxin-like
compounds for these facilities are shown in Table 8-17. In 2000, approximately 23.08 g TEQDF-
WHO98 (23.94 g I-TEQ) were released from 17 EDC/VCM/PVC manufacturing facilities in the
United States. These estimates are assigned a high confidence rating because they were derived
from the testing of individual manufacturing facilities and the activity level is known with a high
degree of certainty.
Estimating TEQ wastewater discharges for 1995 from EDC/VCM/PVC integrated
chemical manufacturing facilities is problematic. A report by The Vinyl Institute (1998) did not
represent a comprehensive testing of the wastewaters from all existing EDC/VCM/PVC
manufacturing facilities in 1995. The report presented results for treated wastewater samples
collected during April and May of 1995 at only 10 manufacturing sites (6 that manufactured only
PVC, 3 that manufactured EDC and VCM, and 1 that manufactured EDC, VCM, and PVC). In
terms of production, the 10 sites represented only about 27% of the total estimated 1995
EDC/VCM/PVC production.
The representativeness of these sites to total EDC/VCM/PVC manufacturing is
questionable because the testing program did not include some of the higher-emitting facilities
discovered in 2000 (CCC, 2003a). For example, the CCC reports that provided the basis for the
2000 release estimates (CCC, 2005, 2003a, b, c, d) indicated that the Dow facility in Freemont,
TX, and the Occidental facility in Ingleside, TX, released approximately 3 g and 1.6 g TEQDF-
WHO98, respectively, as wastewater discharges into surface waters. These two facilities are
about an order of magnitude higher in wastewater releases than are the 10 facilities tested in 1995
(The Vinyl Institute, 1998). In an attempt to overcome this deficiency, it is assumed that the
wastewater releases from EDC/VCM/PVC integrated chemical production facilities in 1995 were
8-33
-------
Table 8-17. Releases of dioxin-like compounds (g/yr) in wastewater discharges from EDC/VCM/PVC and
integrated chlorine chemical manufacturing facilities to surface water in reference years 2000 and 1995
Company
Dow3
Dow
Dow
DuPontb
DuPont
DuPont
Georgia Gulf c
Occidental4
Occidental
Occidental
Occidental
Occidental
Occidental
Occidental
Occidental
Occidental
PPG Industries
Facility
Freeport, TX
Midland, MI
Plaquemine, LA
DeLisle, MS
Edge Moor, DE
Johnsonville, TN
Plaquemine, LA
Convent, LA
Deer Park, TX
Ingleside, TX
LaPorte, TX
Mobile, AL
Battleground, TX
Delaware City, DE
Hahnville, LA
Muscle Shoals, AL
Lake Charles, LA
TOTAL
Q
Q
U
H
90^
r-~
rf
0.00
0.00
0.06
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.08
[,2,3,7,8 PeCDD
0.00
0.00
0.12
0.00
0.00
0.00
0.00
0.00
0.00
0.09
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.22
«
«
U
&
90
r-~
TT'
ff
rf
0.57
0.02
0.87
0.00
0.00
0.00
0.00
0.00
0.00
0.08
0.00
0.00
0.00
0.00
0.00
0.00
0.01
1.55
«
«
U
&
90
r-~
so'
ff
rf
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.09
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.10
«
«
u
&
ON
90~
r-~
ff
rf
0.35
0.00
0.33
0.00
0.00
0.00
0.00
0.00
0.00
0.08
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.77
§
u
&
90
1--'
so'
TT~
ff
rf
15.61
0.54
18.72
0.00
0.00
0.00
0.03
0.00
0.08
0.09
0.00
0.00
0.00
0.00
0.01
0.00
0.40
35.49
«
«
K
74.61
3.63
285.79
0.00
0.10
1.41
0.12
0.15
1.19
0.18
0.05
0.00
0.48
0.00
0.10
le-09
2.97
370.78
tu
«
U
H
90
1--'
ff
2.63
0.02
4.78
0.00
0.00
0.00
0.01
0.00
0.00
0.02
0.00
3e-04
0.00
le-03
0.19
4e-08
6.47
14.13
[,2,3,7,8 PeCDF
3.57
0.00
3.77
0.00
0.00
0.00
0.01
0.00
0.00
0.09
0.00
0.00
0.00
0.00
0.90
le-07
14.30
22.64
!,3,4,7,8 PeCDF
1.68
0.01
1.97
0.00
0.00
0.00
0.02
0.00
0.00
0.09
0.00
0.00
0.00
0.00
0.85
8e-08
10.83
15.46
tu
«
U
&
90
r-~
TT'
ff
rf
26.28
0.06
21.42
0.00
0.08
0.44
0.02
0.02
0.04
0.09
0.04
0.00
0.00
0.00
2.96
2e-07
12.29
63.73
tu
«
U
&
90
1--'
so'
ff
-------
Table 8-17. Releases of dioxin-like compounds (g/yr) in wastewater discharges from EDC/VCM/PVC and
integrated chlorine chemical manufacturing facilities to surface water in reference years 2000 and 1995
(continued)
aDow Chemical Company.
bDuPont.
"Georgia Gulf Corporation.
Accidental Chemical Corporation.
Source: Chlorine Chemistry Council (2004).
EDC = Ethylene dichloride
PVC = Polyvinyl chloride
VCM = Vinyl chloride monomer
oo
oo
-------
equal to wastewater discharges in 2000. Based on this assumption, approximately 23.08 g
TEQDF-WHO98 (23.94 g I-TEQ) were released from 17 EDC/VCM/PVC manufacturing facilities
in the United States in reference year 1995. These estimates are assigned a low confidence rating
because they may not be representative of wastewater discharges from facilities operating in
1995, although the activity level is known with a high degree of certainty and is assigned a high
confidence rating. The site-specific releases of dioxin-like compounds for facilities operating in
1995 are shown in Table 8-17.
A number of EDC/VCM/PVC manufacturing facilities have reported significant
reductions in CDD/CDF wastewater discharges to surface water in 2002 from 2000 levels. For
example, the Dow facility in Freemont, TX, achieved a 47% reduction (from 6.4 g TEQDF-
WHO98 in 2000 to 3.43 g TEQDF-WHO98 in 2002) (CCC, 2003a), the Dow facility in
Plaquemine, LA, achieved a 69% reduction (from 6.86 g TEQDF-WHO98 in 2000 to 2.16 g
TEQDF-WHO98 in 2002) (CCC, 2003a), and the PPG Industries facility in Lake Charles, LA,
achieved a 28% reduction (from 8.97 g TEQDF-WHO98 in 2000 to 6.47 g TEQDF-WHO98 in 2002)
(CCC, 2003b).
8.3.4.2. Land Releases
Only one EDC/VCM/PVC manufacturing facility (the Georgia Gulf facility in
Plaquemine, LA) reported CDD/CDF releases to land from the land application of wastewater
sludge in 2000 (CCC, 2003c). The congener-specific and TEQ releases are presented in Table
8-18. Releases to land from EDC/VCM/PVC manufacturing in 2000 and 1995 were 1.36 g
TEQDF-WHO98 (1.45 g I-TEQDF), assuming that the land application rates were the same in both
years.
These emission estimates for 2000 are assigned a high confidence rating because the
releases to land were determined from the actual measurements taken from the single facility that
applies the wastewater sludge to land. The emission estimates for 1995 are assigned a medium
confidence rating because the estimates were based on assuming that the CDD/CDF levels of
contamination in the wastewater solids and the land application rates were the same in both
reference years.
8.3.4.3. Air Releases
The Vinyl Institute conducted a study of releases of EDC/VCM/PVC to air (The Vinyl
Institute, 1998). Based on similarities in design and service, thermal destruction units at
EDC/VCM and/or PVC manufacturing units were subcategorized into three types: type A, vent
gas incinerators at PVC-only resin plants; type B, vent gas thermal oxidizers at EDC/VCM
plants; and type C, liquid-only and liquid/vent gas thermal oxidizers at EDC/VCM plants. Using
8-36
-------
Table 8-18. Congener-specific releases to land from an EDC/VCM/PVC
integrated chemical manufacturing facility" in reference years 2000 and 1995
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDD
1,2,3,4,7,8,9-HpCDD
OCDF
Total I-TEQDF
Total TEQDF-WHO98
Annual release to land (g/yr)
0.004
0.039
0.123
0.122
0.078
1.710
8.640
0.077
0.354
0.370
2.690
2.110
1.540
0.595
28.100
6.540
118.000
1.450
1.360
aGeorgia Gulf, Plaquemine, LA.
Source: CCC (2004).
EDC = Ethylene dichloride
PVC = Polyvinyl chloride
VCM = Vinyl chloride monomer
an industry-wide survey, The Vinyl Institute identified 22 type A units at 11 facilities, 23 type B
units at 10 facilities, and 17 type C units at 10 facilities. Test data were gathered from 5 of the 22
type A units (3 facilities representing 7% of total U.S. and Canadian EDC/VCM/PVC production
in 1995), 14 of the 23 type B units (8 facilities), and 13 of the 17 type C units (7 facilities). The
sampled type B and C units represented 70% of total U.S. and Canadian EDC/VCM/PVC
production in 1995.
Annual I-TEQDF emission estimates were generated by combining estimated emissions
from tested units (based on measured stack gas results and plant-specific activity data) with an
estimate of emissions from untested units. The emissions from the untested units were estimated
by multiplying the average emission factor for the tested units in the category (the most likely
8-37
-------
estimate) or the average emission factor of the tested units with the highest emissions in each
class (the upper-bound estimate) by the activity level for the untested units. It is not possible to
calculate emission factors for TEQDF-WHO98 using the data presented in The Vinyl Institute
report.
The Vinyl Institute estimates of most likely and upper-bound emissions during 1995 for
these three categories are as follows:
Most likely emissions Upper-bound emissions
Category estimate (g I-TEQDF/yr) estimate (g I-TEQDF/yr)
PVC-only incinerators 0.0014 0.0019
EDC/VCM liquid and liquid/vents 3.7 7.2
EDC/VCM vents for VCM only 6.9 21.6
The study also estimated emissions that may have resulted from incineration of
EDC/VCM/PVC wastes processed by off-site, third-party processing. Using the emission factors
for liquid and liquid/vents developed in its study, it was estimated that potential emissions to air
from this source category would be 0.65 g I-TEQDF/yr (most-likely estimate) and 2.3 g I-
TEQDF/yr (upper-bound estimate). Combining these third-party release estimates with those
developed above yields a 1995 estimate of 11.2 g I-TEQDF/yr.
Data validation studies by the CCC indicate that eight EDC/VCM/PVC manufacturing
facilities released 5.51 g TEQDF-WHO98 (5.56 g I-TEQDF) to air (CCC, 2004); more than 85% of
the releases occurred at two facilities. Congener-specific and TEQ release estimates to air from
EDC/VCM/PVC manufacturing facilities are shown in Table 8-19. The emission estimates for
1995 and 2000 are assigned a high confidence rating because they were based on emissions
testing of on-site incinerator and vent releases.
8.3.4.4. Transfers to Secure Landfills
The CCC reported on the amount of CDDs/CDFs contained in wastewater treatment plant
sludges used to secure landfills in 2000 (CCC, 2004). These data were determined on the basis
of sampling wastewater sludges at 16 integrated EDC/VCM/PVC manufacturing facilities. Table
8-20 summarizes the estimated CDD/CDF congener-specific and TEQ amounts (based on actual
test data) transferred from specific facilities in 2000. It should be noted that, because the wastes
were transferred to secure landfills, this is not considered to have been an environmental release;
therefore, these transfers were not incorporated into the inventory.
8-38
-------
Table 8-19. Congener-specific and TEQ releases to air (g/yr) from EDC/VCM/PVC integrated chemical
manufacturing facilities in reference years 2000 and 1995
Company
Dow"
Dow
Dow
DuPontb
DuPont
DuPont
Georgia Gulf c
Occidental4
Occidental
Occidental
Occidental
Facility
Freeport, TX
Midland, MI
Plaquemine, LA
DeLisle, MS
Edge Moor, DE
Johnsonville, TN
Plaquemine, LA
Convent, LA
Deer Park, TX
Ingleside, TX
LaPorte, TX
TOTAL
§
U
H
ao_
ff
0.01
0.026
0.00
0.000
0.000
0.000
0.00
0.00
0.00
0.01
0.00
0.05
Q
Q
PM
90^
ff
T— T
0.20
0.004
0.00
0.001
0.001
0.001
0.00
0.01
0.01
0.03
0.01
0.27
§
u
90^
ff
1.01
0.008
0.03
0.000
0.000
0.000
0.00
0.00
0.05
0.11
0.01
1.21
§
u
90^
so"
ff
T— r
0.00
0.000
0.00
0.000
0.000
0.000
0.00
0.01
0.03
0.12
0.02
0.18
§
u
o\
°°I
rf
T— T
0.39
0.003
0.01
0.000
0.000
0.000
0.00
0.00
0.06
0.09
0.01
0.56
u
o.
ao_
so"
ff
T— r
3.73
0.02
0.10
0.004
0.003
0.003
0.02
0.01
1.37
1.58
0.15
6.99
§
9.46
0.086
0.27
0.011
0.016
0.007
0.12
0.00
13.54
9.51
0.48
33.49
tu
U
H
ao_
ff
1.26
0.023
0.03
0.004
0.001
0.005
0.09
0.02
0.06
0.04
0.01
1.54
tu
«
PM
ao_
1.80
0.009
0.04
0.004
0.002
0.006
0.07
0.02
0.12
0.35
0.02
2.45
2,3,4,7,8 PeCDF
1.28
0.008
0.03
0.006
0.002
0.008
0.03
0.01
0.14
0.37
0.03
1.92
tu
Q
U
90^
ff
11.98
0.066
0.40
0.007
0.003
0.010
0.17
0.01
1.38
3.58
0.04
17.65
tu
Q
U
°°,
so^
0.00
0.000
0.00
0.005
0.002
0.008
0.07
0.01
0.58
3.38
0.04
4.09
1,2,3,7,8,9 HxCDF
0.34
0.002
0.00
0.002
0.005
0.004
0.04
0.00
0.55
0.77
0.01
1.74
tu
Q
U
°°,
so"
2.50
0.009
0.05
0.013
0.003
0.023
0.01
0.00
0.06
1.94
0.04
4.64
tu
Q
U
c.
ao_
scT
ff
32.35
0.148
1.47
0.028
0.011
0.043
0.32
0.02
11.88
25.71
0.25
72.22
tu
Q
U
c.
o\
°€
5.34
0.028
0.10
0.003
0.018
0.003
0.03
0.00
2.32
5.13
0.02
13.01
tu
Q
66.92
0.225
3.21
0.042
0.204
0.043
1.01
0.02
52.83
46.90
0.23
171.64
w
hH
3.08
0.05
0.09
0.007
0.004
0.011
0.07
0.02
0.58
1.61
0.04
5.56
CO
o"
w
H
3.11
0.05
0.09
0.007
0.004
0.011
0.07
0.03
0.53
1.58
0.04
5.51
oo
aDow Chemical Company.
bDuPont.
"Georgia Gulf Corporation.
Accidental Chemical Corporation.
Source: Chlorine Chemistry Council (2004).
EDC = Ethylene dichloride
PVC = Polyvinyl chloride
VCM = Vinyl chloride monomer
-------
Table 8-20. Congener-specific and TEQ transfers to secure landfills (g/yr)a from EDC/VCM/PVC integrated
chemical production facilities in 2000
Company
Dowb
Dow
Dow
DuPonf
DuPont
DuPont
Georgia Gulf c
Occidental6
Occidental
Occidental
Occidental
Occidental
Occidental
Occidental
Occidental
Occidental
Facility
Freeport, TX
Midland, MI
Plaquemine, LA
DeLisle, MS
Edge Moor, DE
Johnsonville, TN
Plaquemine, LA
Convent, LA
Deer Park, TX
Ingleside, TX
LaPorte, TX
Mobile, AL
Battleground, TX
Delaware City, DE
Hahnville, LA
Muscle Shoals, AL
TOTAL
§
U
90
1--"
ff
"
ff
"
TT"
r
-------
8.3.4.5 Products
The Vinyl Institute (1998) presented results for 22 samples from 14 of the 24 U.S. and
Canadian facilities manufacturing suspension and mass PVC resins (13 pipe resins, 3 bottle
resins, and 6 packaging resins). The results for U.S. manufacturers are summarized in Table
8-21. The 14 sampled sites represented approximately 74% of estimated 1995 U.S. and
Canadian suspension and mass PVC resin production. CDDs/CDFs were detected in only one
sample (0.043 ng I-TEQDF/kg, assuming nondetects equal to zero). The overall mean TEQ
concentrations were 0.002 ng I-TEQDF/kg (assuming nondetects equal to zero) and 0.7 ng
I-TEQDF/kg (assuming nondetects equal to one-half the DL). The DLs were 2 ng/kg or less for
all congeners in all samples except for OCDD and OCDF, which had DLs of 6 ng/kg or less.
The same study also presented results for six samples from four of the seven U.S.
facilities manufacturing dispersion PVC resins. CDDs/CDFs were detected in five of the
samples. The results are summarized in Table 8-21. In terms of production, the four sampled
sites represent approximately 61% of estimated 1995 U.S. dispersion PVC resin production. The
results ranged from not detected to 0.008 ng I-TEQDF/kg (overall mean = 0.001 ng I-TEQDF/kg,
assuming nondetects equal to zero, and 0.4 ng I-TEQDF/kg, assuming nondetects equal to one-
half the DL). The DLs were 2 ng/kg or less for all congeners in all samples except for OCDD
and OCDF, which had DLs of 4 ng/kg or less.
Results were also presented for five samples from 5 of the 15 U.S. facilities
manufacturing EDC. The results are summarized in Table 8-21. In terms of production, the five
sampled sites represented approximately 71% of the estimated EDC produced in the United
States in 1995. CDDs/CDFs were detected in only one sample (0.03 ng I-TEQDF/kg). The
overall mean TEQ concentrations were 0.006 ng I-TEQDF/kg (nondetects equal to zero) and 0.21
ng I-TEQDF/kg (nondetects equal to one-half the DL). The DLs for all congeners were 1 ng/kg or
less.
Using 1995 U.S. production data, 4.846 million metric tons of suspension and mass PVC,
0.367 million metric tons of dispersion PVC resins, and 1.362 million metric tons of EDC were
produced. Based on the average TEQ concentration observed, The Vinyl Institute estimated that
the total I-TEQDF contents of suspension/mass PVC resins, dispersion PVC resins, and EDC were
0.01 g, 0.004 g, and 0.008 g, respectively (nondetects equal to zero), and 3.39 g, 0.15 g, and 0.29
g, respectively (nondetects equal to one-half the DL). Therefore, total I-TEQDF present in PVC in
1995 was estimated to be between 0.02 g (nondetects equal to zero) and 3.83 g (nondetects equal
to one-half the DL). It is not possible using the data presented in The Vinyl Institute report to
calculate emission factors for TEQDF-WHO98. However, because neither 1,2,3,7,8-PeCDD nor
OCDD was detected in any sample, the TEQDF-WHO98 emission factors would be very similar to
the I-TEQDF emission factors.
8-41
-------
Table 8-21. CDD/CDF concentrations in products from U.S. EDC/VCM/PVC manufacturers
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Mean I-TEQDF (nondetect = 0)
Mean I-TEQDF (nondetect = !/2 DL)
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Suspension and mass PVC resins
No. detects/
samples3
0/22
0/22
0/22
0/22
0/22
1/22
0/22
0/22
0/22
0/22
0/22
0/22
0/22
1/22
0/22
0/22
0/22
0/22
0/22
0/22
1/22
0/22
0/22
0/22
1/22
0/22
0/22
Range" (ng/kg)
Min.
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.002
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Max.
ND
ND
ND
ND
ND
0.64
ND
ND
ND
ND
ND
ND
ND
0.37
ND
ND
ND
0.7
ND
ND
ND
0.64
ND
ND
ND
0.37
ND
ND
Dispersion PVC resins
No. of
detects/
samples
0/6
0/6
0/6
0/6
0/6
1/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
2/6
1/6
1/6
5/6
1/6
0/6
0/6
1/6
0/6
0/6
2/6
Range0 (ng/kg)
Min.
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.001
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Max.
ND
ND
ND
ND
ND
0.8
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.38
0.4
0.24
0.32
0.97
1.3
ND
ND
0.3
ND
ND
0.38
EDC sold as product11
No. detects/
samples
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
1/5
1/5
1/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
0/5
1/5
1/5
Range6 (ng/kg)
Min.
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.001
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Max.
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.1
0.4
11
0.21
ND
ND
ND
ND
ND
ND
ND
ND
2.02
11
oo
-k
to
-------
Table 8-21. CDD/CDF concentrations in products from U.S. EDC/VCM/PVC manufacturers (continued)
aTwo of these 22 samples were duplicate samples from two sites. The results were averaged and treated as one sample for each site.
bDetection limits (DLs) for individual samples were less than 2 ng/kg for all congeners and congener groups except OCDD and OCDF, which had
DLs less than 6 ng/kg.
°DLs for individual samples were less than 2 ng/kg for all congeners and congener groups except OCDD and OCDF, which had DLs less than 4 ng/kg.
d"Sales" EDC is defined as EDC sold commercially for non-VCM uses or exported from the United States.
eDLs were less than 1 ng/kg for all congeners in all samples.
DL = Detection limit
EDC = Ethylene dichloride
ND = Not detected
PVC = Polyvinyl chloride
VCM = Vinyl chloride monomer
Source: The Vinyl Institute (1998).
oo
-------
Approximately 6.55 million metric tons of PVC and 9.91 million metric tons of EDC
were produced in North America in 2000 and approximately 5.58 million metric tons of PVC and
7.83 million metric tons of EDC were produced in 1995 (C&EN, 2002). Approximately 94% of
PVC production and approximately 17% of EDC production in 1995 occurred in the United
States. Of the PVC produced, 87% was for suspension and mass PVC products and 7% was for
dispersion PVC resins. Assuming these product percentages remained the same for 2000, it is
estimated that approximately 5.69 million metric tons of suspension and mass PVC and 0.46
million metric tons of dispersion PVC resins were produced and 1.69 million metric tons of EDC
product were produced. Applying the same average TEQ observed in The Vinyl Institute
samples from 1998, EPA estimated the total I-TEQDF contents of suspension/mass PVC resins,
dispersion PVC resins, and EDC produced in 2000 to be 0.01 g, 0.0004 g, and 0.01 g,
respectively (nondetects equal to zero) and 3.99 g, 0.17 g, and 0.36 g, respectively (nondetects
equal to one-half the DL). Therefore, total I-TEQDF present in PVC in 2000 was estimated to be
between 0.02 g (nondetects equal to zero) and 4.52 g (nondetects equal to one-half the DL).
8.3.5. Other Aliphatic Chlorine Compounds
Aliphatic chlorine compounds are used as monomers in the production of plastics, as
solvents and cleaning agents, and as precursors for chemical synthesis (Hutzinger and
Fiedler, 1991b). These compounds are produced in large quantities. In 1992, 14.6 million metric
tons of halogenated hydrocarbons were produced in the United States (U.S. ITC, 1946-1994),
with 1,2-dichloroethane and vinyl chloride accounting for 82% of total production. Highly
chlorinated CDDs/CDFs (hexa- to octa-chlorinated congeners) have been found in nanograde-
quality samples of 1,2-dichloroethane (55 ng/kg of OCDF in one of five samples),
tetrachloroethene (47 ng/kg of OCDD in one of four samples), epichlorohydrin (88 ng/kg of
CDDs and 33 ng/kg of CDFs in one of three samples), and hexachlorobutadiene (360 to 425
ng/kg of OCDF in two samples) obtained in Germany from Promochem (Hutzinger and Fiedler,
1991b; Heindl and Hutzinger, 1987). No CDDs/CDFs were detected in two samples of allyl
chloride, three samples of 1,1,1-trichloroethane, and four samples of trichloroethylene (DL
ranged from 5 to 20 ng/kg) (Heindl and Hutzinger, 1987). Because no more recent or additional
data could be found in the literature to confirm these values for products manufactured or used in
the United States, no national estimates of CDD/CDF emissions were made for the inventory.
EPA's Office of Water promulgated effluent limitations for facilities that manufacture
chlorinated aliphatic chlorine compounds and discharge treated wastewater (40 CFR 414.70).
These effluent limitations do not specifically address CDDs/CDFs. Regulated limits for
chlorinated aliphatic compounds are 68 |ig/L for 1,2-dichloroethane and 22 |ig/L for
tetrachloroethylene. Similarly, OSW promulgated restrictions on land disposal of wastes
8-44
-------
generated during the manufacture of many chlorinated aliphatics (40 CFR 268); however, these
restrictions do not specifically regulate CDDs/CDFs.
8.3.6. Dyes, Pigments, and Printing Inks
Several researchers have analyzed various dyes, pigments, and printing inks obtained in
Canada and Germany for the presence of CDDs/CDFs (Williams et al., 1992; Hutzinger and
Fiedler, 1991b; Santl et al., 1994). The following subsections discuss the findings of those
studies.
8.3.6.1. Dioxazine Dyes and Pigments
Williams et al. (1992) analyzed the CDD/CDF content in dioxazine dyes and pigments
available in Canada. As shown in Table 8-22, OCDD and OCDF concentrations in the
nanogram-per-kilogram range and HpCDD, HxCDD, and PeCDD concentrations in the
microgram-per-kilogram range were found in Direct Blue 106 dye (three samples), Direct Blue
108 dye (one sample), and Violet 23 pigments (six samples). These dioxazine pigments are
derived from chloranil, which has been found to contain high levels of CDDs/CDFs and has been
suggested as the source of contamination among these dyes (Christmann et al., 1989b; Williams
et al., 1992; U.S. EPA, 1992d). In May 1990, EPA received test results showing that chloranil
was heavily contaminated with dioxins; levels as high as 2,903 jig TEQDF-WHO98/kg (3,065 jig I-
TEQDF/kg) were measured in samples from four importers (mean value of 1,388 jig TEQDF-
WHO98/kg [1,754 |ig I-TEQDF/kg]) (U.S. EPA, 1992d; Remmers et al., 1992). (See Section 8.3.6
for analytical results.)
In the early 1990s, EPA learned that I-TEQDF levels in chloranil could be reduced by
more than two orders of magnitude (to less than 20 |ig/kg) through manufacturing feedstock and
process changes. EPA's Office of Pollution Prevention and Toxics subsequently began efforts to
complete an industry-wide switch from the use of contaminated chloranil to low-dioxin chloranil.
Although chloranil is not manufactured in the United States, significant quantities are imported.
As of May 1992, EPA had negotiated agreements with all chloranil importers and domestic
dye/pigment manufacturers known to EPA that used chloranil in their products to switch to low-
dioxin chloranil. In May 1993, when U.S. stocks of chloranil with high levels of CDDs/CDFs
had been depleted, EPA proposed a SNUR under Section 5 of TSCA that would require industry
to notify EPA at least 90 days prior to the manufacture, import, or processing, for any use, of
chloranil containing CDDs/CDFs at a concentration greater than 20 jig I-TEQDF/kg (Federal
Register, 1993a; U.S. EPA, 1993d).
8-45
-------
Table 8-22. CDD/CDF concentrations (ug/kg) in samples of dioxazine dyes and pigments (Canada)
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
l,2,3,6,7,8-HxCDFa
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDFb
Total TEQDF-WHO98b
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF b
Blue 106
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
31
41,953
ND (0.3)
ND (0.3)
ND (0.3)
12
~
ND (0.3)
50
~
12,463
41,984
12,525
56.4
7.45
ND (0.3)
ND (0.3)
ND (0.3)
34
41,953
ND (0.3)
ND (0.3)
12
71
12,463
54,533
ND (0.3)
ND (0.3)
ND (0.3)
ND(03)
_L ~M^r \\s .>/ J
28,523
ND (0.3)
ND (0.3)
ND (0.3)
2
~
ND (0.3)
10
-
1,447
28,529
1,459
30.3
3.4
ND (0.3)
ND (0.3)
ND (0.3)
8
28,523
0.3
ND (0.3)
2
32
1,447
30,012.3
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
9
18,066
ND (0.3)
ND (0.3)
ND (0.3)
2
~
ND (0.3)
14
-
1,006
18,075
1,022
19.5
2.3
ND (0.3)
ND (0.3)
ND (0.3)
12
18,066
ND (0.3)
ND (0.3)
2
26
1,006
19,112
Blue 108
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
ND (0.3)
23
ND (0.3)
ND (0.3)
ND (0.3)
ND (0.3)
-
ND (0.3)
9
~
11
23
20
0.1
0.1
ND (0.3)
ND (0.3)
1
ND (0.3)
23
ND (0.3)
ND (0.3)
ND (0.3)
12
11
47
Violet 23
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
9
7,180
ND (0.3)
0.5
ND (0.3)
76
-
ND (0.3)
13
~
941
7,189
1,030.5
16
8.7
ND (0.3)
ND (0.3)
21
30
7,180
ND (0.3)
0.5
76
26
941
8,274.5
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
1
806
ND (0.3)
ND (0.3)
ND (0.3)
4
-
ND (0.3)
10
-
125
807
139
1.4
0.6
ND (0.3)
ND (0.3)
2
5
806
ND (0.3)
ND (0.3)
5
14
125
957
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
16
11,022
ND (0.3)
ND (0.3)
ND (0.3)
39
~
ND (0.3)
11
~
3,749
11,038
3,799
18.9
5.6
ND (0.3)
ND (0.3)
7
36
11,022
ND (0.3)
ND (0.3)
39
29
3,749
14,882
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
10
7,929
ND (0.3)
ND (0.3)
ND (0.3)
31
-
ND (0.3)
4
-
1,556
7,939
1,591
12.7
4.2
ND (0.3)
ND (0.3)
ND (0.3)
11
7,929
ND (0.3)
ND (0.3)
31
13
1,556
9,540
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
2
1,627
ND (0.3)
ND (0.3)
ND (0.3)
9
-
ND (0.3)
1
-
147
1,629
157
2.7
1.1
ND (0.3)
ND (0.3)
ND (0.3)
2
1,627
0.4
ND (0.3)
9
2
147
1,787.4
ND (0.3)
ND (0.3)
-
ND (0.3)
ND (0.3)
4
1,420
ND (0.3)
ND (0.3)
ND (0.3)
7
~
ND (0.3)
12
~
425
1,424
444
2.7
1
ND (0.3)
ND (0.3)
1
6
1,420
ND (0.3)
ND (0.3)
7
21
425
1,880
oo
-k
a\
-------
Table 8-22. CDD/CDF concentrations in samples of dioxazine dyes and pigments (ug/kg) (Canada) (continued)
aResults listed for 1,2,3,4,7,8-HxCDF include concentrations for 1,2,3,6,7,8-HxCDF.
bCalculations assume nondetected values were equal to zero.
ND = Not detected (value in parenthesis is the detection limit)
~ = Not reported
Source: Williams et al. (1992).
oo
-------
In 1983, approximately 36,500 kg of chloranil were imported (U.S. ITC, 1984). The U.S.
International Trade Commission has not published quantitative import data for chloranil since
1984. If it is assumed that this import volume reflects actual usage of chloranil in the United
States during 1987 and that the CDD/CDF contamination level was 1,388 |ig TEQDF-WHO98/kg
(1,754 jig I-TEQDF/kg), then the maximum release to the environment via processing wastes and
finished products was 50.6 g TEQDF-WHO98 (64 g I-TEQDF). If it is assumed that the import
volume in 1995 was also 36,500 kg but that the imported chloranil contained 10 jig I-TEQDF/kg
on average, then the total potential annual TEQ release associated with chloranil in 1995 was
50.6 g TEQDF-WH098 (64 g I-TEQDF).
In 1986, EPA promulgated the Inventory Update Rule (IUR) that requires the partial
updating of the TSCA Chemical Inventory database. Every four years, chemical manufacturers
and importers of chemicals listed in the TSCA inventory that produce at one plant site or import
at production volume levels of 10,000 Ib or more must report the range of chemical production or
import. According to information entered in the TSCA database, 10,000 to 500,000 Ib (4,540 to
227,000 kg) of chloranil were imported in 1994 and 2000 (http://www.epa.gov/opptintr/iur
/iur02/search03.htm). Assuming the imported chloranil contained the same concentration of
dioxin as the 1995 estimate (10 jig I-TEQDF/kg), the total potential annual TEQ release associated
with chloranil in 2000 was 0.05 to 2.27 g I-TEQDF-WHO98 (mean of 1.16 g I-TEQDF).
8.3.6.2. Phthalocyanine Dyes and Printing Inks
Hutzinger and Fiedler (1991b) found CDDs/CDFs (tetra-, penta-, and hexa-chlorinated
congeners) in the microgram-per-kilogram range in a sample of a Ni-phthalocyanine dye. No
CDDs/CDFs were detected (DL of 0.1 to 0.5 |ig/kg) in two samples of Cu-phthalocyanine dyes
and in one Co-phthalocyanine dye (Hutzinger and Fiedler, 1991b).
Santl et al. (1994) reported the results of analyses of four printing inks obtained from a
supplier in Germany. Two of the inks are used for rotogravure printing and two are used for
offset printing. The results of the analyses are presented in Table 8-23. The TEQDF-WHO98
content of the inks ranged from 17.7 to 87.2 ng/kg (15 to 88.6 ng/kg on an I-TEQDF basis).
Primarily non-2,3,7,8-substituted congeners were found. The identities of the dyes and pigments
in these inks were not reported.
Although EPA provided an estimate of potential environmental releases based on limited
information of contaminant levels of CDDs/CDFs in the product, the estimate is still too
uncertain to include in the quantitative inventory of sources. It is currently not known whether
the dioxin contamination in the product actually results in a release to the open and circulating
environment.
8-48
-------
Table 8-23. CDD/CDF concentrations (ng/kg) in printing inks (Germany)
Congener/congener group
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF a
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
Rotogravure
(2-color)
ND(1)
8
19
325
155
2,770
5,810
2.5
ND(2)
ND(2)
4
ND(3)
ND(3)
ND(3)
40
ND(4)
129
9,087
175.5
88.6
87.2
4
58
2,679
5,630
5,810
5.5
13
29
64
129
14,421.5
Rotogravure
(4-color)
ND(1.5)
ND(4)
ND(5)
310
105
1,630
2,350
14
ND(4)
ND(4)
7
ND(5)
ND(5)
ND(5)
14
ND(7)
ND (10)
4,395
35
62.4
60.3
ND(2)
145
2,485
3,460
2,350
28
ND(4)
45
14
ND(10)
8,527
Offset
(4-color)
ND(2)
15
16
82
42
540
890
7
ND(4)
ND(4)
27
ND(5)
ND(5)
ND(5)
315
11
960
1,585
1,320
35.4
41.2
77
35
660
1,100
890
90
340
95
566
960
4,813
Offset
(4-color)
ND(2)
6
11
21
14
240
230
7
ND(3)
ND(3)
35
ND(5)
ND(5)
ND(5)
42
ND(6)
165
522
249
15
18
38
25
246
445
230
35
110
94
63
165
1,451
""Calculations assume nondetect values were zero.
ND = Not detected (value in parenthesis is the detection limit)
Source: Santletal. (1994).
8-49
-------
8.3.7. TSCA Dioxin/Furan Test Rule
Citing evidence that halogenated dioxins and furans may be formed as by-products during
chemical manufacturing processes (Versar, Inc., 1985), EPA issued a rule under Section 4 of
TSCA that requires chemical manufacturers and importers to test for the presence of
CDDs/CDFs and brominated dibenzo-p-dioxins (BDDs) and brominated dibenzofurans (BDFs)
in certain commercial organic chemicals (Federal Register, 1987a). The rule listed 12
manufactured or imported chemicals that required testing and 20 chemicals not currently
manufactured or imported that would require testing if manufacture or importation resumed.
These chemicals are listed in Table 8-24. The specific dioxin and furan congeners that require
quantitation and the target limits of quantitation (LOQs) that are specified in the rule are listed in
Table 8-25. Under Section 8(a) of TSCA, the final rule also required that chemical
manufacturers submit data on manufacturing processes and reaction conditions for chemicals
produced using any of the 28 precursor chemicals listed in Table 8-26. The rule stated that
subsequent to this data-gathering effort, testing may be proposed for additional chemicals if any
of the manufacturing conditions used favored the production of dioxins and furans.
Twenty-three sampling and analytical protocols and test data for 10 of the 12 chemicals
that required testing were submitted to EPA (U.S. EPA, 2003a, c). Manufacture or import of two
substances (tetrabromobisphenol-A-bis-2,3-dibromopropylether and tetrabromobisphenol-A-
diacrylate) have stopped since the test rule was promulgated. (All data and reports in the EPA
TSCA docket are available for public review and inspection at EPA Headquarters in Washington,
DC.)
Table 8-27 presents the results of analytical testing for CDDs/CDFs for the chemicals that
have data available in the TSCA docket. Five of these 10 chemicals contained CDDs/CDFs.
Positive results were obtained for 2,3,5,6-tetrachloro-2,5-cyclohexadiene-l,4-dione (chloranil),
pentabromodiphenyloxide, octabromodiphenyloxide, decabromodiphenyloxide, and 1,2-
bis(tribromophenoxy)-ethane. Table 8-28 presents the quantitative analytical results for four
submitted chloranil samples, as well as the results of an EPA analysis of a sample of carbazole
violet, which is manufactured from chloranil.
Although testing conducted under this test rule for 2,4,6-tribromophenol indicated no
halogenated dioxins or furans above the LOQs, Thoma and Hutzinger (1989) reported detecting
BDDs and BDFs in a technical-grade sample of this substance. Total TBDD, TBDF, and PeBDF
were found at 84 |ig/kg, 12 |ig/kg, and 1 |ig/kg, respectively. No hexa-, hepta-, or octa-BDFs
were detected. The investigators also analyzed analytical-grade samples of two other brominated
flame retardants, pentabromophenol and tetrabromophthalic anhydride; no BDDs or BDFs were
detected (DLs not reported).
8-50
-------
Table 8-24. Chemicals requiring Toxic Substances Control Act Section 4
testing under the dioxin/furan rule
CAS No.
Chemical name
Currently manufactured or imported as of June 5,1987
79-94-7
118-75-2
118-79-6
120-83-2
1163-19-5
4162-45-2
21850-44-2
25327-89-3
32534-81-9
32536-52-0
37853-59-1
55205-38-4
Tetrabromobisphenol-A
2,3,5,6-Tetrachloro-2,5-cyclohexadiene-l,4-dione
2,4,6-Tribromophenol
2,4-Dichlorophenol
Decabromodiphenyloxide
Tetrabromobisphenol-A-bisethoxylate
Tetrabromobisphenol-A-bis-2,3-dibromopropylethera
Allyl ether of tetrabromobisphenol-A
Pentabromodiphenyloxide
Octabromodiphenyloxide
1,2-Bis(tribromophenoxy)-ethane
Tetrabromobisphenol-A-diacrylatea
Not manufactured or imported as of June 5,1987b
79-95-8
87-10-5
87-65-0
95-77-2
95-95-4
99-28-5
120-36-5
320-72-9
488-47-1
576-24-9
583-78-8
608-71-9
615-58-7
933-75-5
1940-42-7
2577-72-2
3772-94-9
37853-61-5
Tetrachlorobisphenol-A
3,4',5-Tribromosalicylanide
2,6-Dichlorophenol
3,4-Dichlorophenol
2,4,5-Trichlorophenol
2,6-Dibromo-4-nitrophenol
2[2,4-(Dichlorophenoxy)]-propanoic acid
3,5-Dichlorosalicyclic acid
Tetrabromocatechol
2,3 -Dichlorophenol
2,5 -Dichlorophenol
Pentabromophenol
2,4-Dibromophenol
2,3,6-Trichlorophenol
4-Bromo-2,5-dichlorophenol
3,5 -Dibromosalicylanide
Pentachlorophenyl laurate
Bismethylether of tetrabromobisphenol-A
Alkylamine tetrachlorophenate
Tetrabromobisphenol-B
aNo longer manufactured in or imported into the United States (memorandum dated May 4, 1993, from Gordon
Cash, U.S. EPA/OPPTS, to John Schaum, U.S. EPA/ORD).
bAs of August 5, 1995, neither manufacture nor importation of any of these chemicals had resumed in the United
States (memorandum dated August 2, 1995, from T.S. Holderman, U.S. EPA, to A. Adenuga, Versar, Inc.).
8-51
-------
Table 8-25. Congeners and limits of quantitation (LOQs) for which
quantitation is required under the dioxin/furan test rule and pesticide Data
Call-In
Chlorinated dioxins
and furans
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-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
1,2,3,4,7,8-HxCDF
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-HpCDF
Brominated dioxins
and furans
2,3,7,8-TBDD
1,2,3,7,8-PeBDD
1,2,3,4,7,8-HxBDD
1,2,3,6,7,8-HxBDD
1,2,3,7,8,9-HxBDD
1,2,3,4,6,7,8-HpBDD
2,3,7,8-TBDF
1,2,3,7,8-PeBDF
2,3,4,7,8-PeBDF
1,2,3,4,7,8-HxBDF
1,2,3,6,7,8-HxBDF
1,2,3,7,8,9-HxBDF
2,3,4,6,7,8-HxBDF
1,2,3,4,6,7,8-HpBDF
1,2,3,4,7,8,9-HpBDF
LOQ
(^g/kg)
0.1
0.5
2.5
2.5
2.5
100
1
5
5
25
25
25
25
1,000
1,000
8.3.8. Halogenated Pesticides and FIFRA Pesticides Data Call-In
In the late 1970s and early 1980s, attention began to focus on pesticides as potential
sources of CDDs/CDFs in the environment. Up to that time, CDD/CDF levels were not
regulated in end-use pesticide products. However, some of the active ingredients in pesticides,
particularly chlorinated phenols and their derivatives, were known or suspected to be
contaminated with CDDs/CDFs. During the 1980s and 1990s, EPA took several actions to
investigate and control CDD/CDF contamination of pesticides.
In 1983, EPA cancelled the sale of Silvex and 2,4,5-T for all uses (Federal Register,
1987a). Earlier, in 1979, EPA had ordered emergency suspension of the forestry, rights-of-way,
and pasture uses of 2,4,5-T. Emergency suspensions of the forestry, rights-of-way, pasture, home
and garden, commercial/ornamental turf, and aquatic weed control/ditch bank uses of Silvex
were also ordered (Federal Register, 1979; Plimmer, 1980). The home and garden,
commercial/ornamental turf, and aquatic weed control/ditch bank uses of 2,4,5-T had been
suspended in 1970.
8-52
-------
Table 8-26. Precursor chemicals subject to reporting requirements under
Toxic Substances Control Act Section 8(a)a
CAS No.
Chemical name
85-22-3
87-61-6
87-84-3
89-61-2
89-64-5
89-69-0
92-04-6
97-74-6
94-81-5
95-50-1
95-56-7
95-57-8
95-88-5
95-94-3
95-50-7
99-30-9
99-54-7
106-46-7
108-70-3
108-86-1
108-90-7
117-18-0
120-82-1
348-51-6
350-30-1
615-67-8
626-39-1
827-94-1
Pentabromoethylbenzene
1,2,3-Trichlorobenzene
l,2,3,4,5-Pentabromo-6-chlorocyclohexane
1,4-Dichloro-2 -nitrobenzene
4-Chloro-2-nitrophenol
2,4,5 -Trichloronitrobenzene
2-Chloro-4-phenylphenol
4-Chloro-o-toloxy acetic acid
4-(2-Methyl-4-chlorophenoxy) butyric acid
o-Dichlorobenzene
o-Bromophenol
o-Chlorophenol
4 -Chlorore sorcinol
1,2,4,5 -Tetrachlorobenzene
5 -Chloro-2,4-dimethoxyaniline
2,6-Dichloro-4-nitroaniline
1,2-Dichloro-4-nitrobenzene
p-Dichlorobenzene
1,3,5 -Trichlorobenzene
Bromobenzene
Chlorobenzene
l,2,4,5-Tetrachloro-3-nitrobenzene
1,2,4-Trichlorobenzene
o -Chlorofluorobenzene
3 -Chloro-4-fluoronitrobenzene
Chlorohydroquinone
1,3,5 -Tribromobenzene
2,6-Dibromo-4-nitroaniline
T)ibromobenzene (CAS No. 106-37-6) was identified in the preamble to 52 FR 21412 as one of 29 precursor
chemicals; however, it was inadvertently omitted from the regulatory text. Because the regulatory text identified
only 28 chemicals, 28 chemicals appear in 40 CFR 766.38 and in this table.
In 1984, EPA issued a notice of intent to cancel registrations of pesticide products
containing PCP (including its salts) for all wood preservative uses (Federal Register, 1984). This
notice specified modifications to the terms and conditions of product registrations that were
required in order to avoid cancellation of the products. In response to this notice, several trade
associations and registrants requested administrative hearings to challenge EPA's determinations.
After carefully considering the comments and alternatives suggested during the preheating stage
of the administrative proceedings, EPA concluded that certain changes to the 1984 notice were
appropriate. These changes, finalized in 1986 (Federal Register, 1986), included the following:
8-53
-------
Table 8-27. Results of analytical testing for dioxins and furans in the
chemicals tested to date under Section 4 of the dioxin/furan test rule
CAS No.
79-94-7
118-75-2
118-79-6
120-83-2
1163-19-5
25327-89-3
32536-52-0
378-53-59-1
32534-81-9
4162-45-2
Chemical name
Tetrabromobisphenol-A
2,3,5,6-Tetrachloro-2,5-
cyclohexadiene-
1,4-dione (chloranil)
2,4,6-Tribromophenol
2,4-Dichlorophenol
Decabromodiphenyl
oxide
Allyl ether of
tetrabromobisphenol-A
Octabromodiphenyl
oxide
l,2-Bis(tribromo-
phenoxy)-ethane
Pentabromodiphenyl
oxide
Tetrabromobisphenol-A-
bisethoxylate
No. of
chemical
companies
that
submitted
data
3
6
1
1
3
1
3
1
3
1
No. of
positive
studies
0
5
0
0
o
5
0
o
3
i
3
0
Congeners detected
(detection range in fig/kg)
a
See Table 8-28
a
a
2,3,7,8-PeBDD (ND-0.1)
l,2,3,4,7,8/l,2,3,6,7,8-HxBDD(ND-0.5)
1,2,3,7,8,9-HxBDD (ND-0.76)
1,2,3,7,8-PeBDF (ND-0.7)
l,2,3,4,7,8/l,2,3,6,7,8-HxBDF(ND-0.8)
1,2,3,4,6,7,8-HpBDF (17-186)
a
2,3,7,8-TBDD (ND-0.7 1)
1,2,3,7,8-PeBDD (ND-0.1)
2,3,7,8-TBDF (ND-12.6)
1,2,3,7,8-PeBDF (ND-6.3)
2,3,4,7,8-PeBDF (ND-83.1)
l,2,3,4,7,8/l,2,3,6,7,8-HxBDF(ND-67.8)
1,2,3,7,8,9-HxBDF (ND-56.0)
1,2,3,4,6,7,8-HpBDF (ND-330)
2,3,7,8-TBDF (ND-0.04)
l,2,3,4,7,8/l,2,3,6,7,8-HxBDF(ND-0.03)
1,2,3,4,6,7,8-HpBDF (ND-0.33)
1,2,3,7,8-PeBDD (ND-5.9)
l,2,3,4,7,8/l,2,3,6,7,8-HxBDD(ND-6.8)
l,2,3,4,7,8/l,2,3,6,7,8-HxBDD(ND-6.8)
1,2,3,7,8,9-HxBDD (ND-0.02)
2,3,7,8-TBDF(ND-3.1)
1,2,3,7,8-PeBDF (0.7-10.2)
2,3,4,7,8-PeBDF (0.1-2.9)
1,2,3,4,7,8/1,2,3,6,7,8-HxBDF (15.6-61.2)
1,2,3,4,6,7,8-HpBDF (0.7-3.0)
a
aNo 2,3,7,8-substituted dioxins and furans detected above the test rule target limits of quantitation (see
Table 8-20).
ND = Not detected
Source: Holderman and Cramer (1995).
8-54
-------
Table 8-28. CDD/CDF concentrations (ug/kg) in chloranil and carbazole
violet samples analyzed pursuant to the EPA dioxin/furan test rule
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total I-TEQDFa
Total TEQDF-WHO98a
Chloranil
Importer
1
ND(1)
ND(2)
ND(3)
ND(3)
ND(1)
110
240,000
ND(1)
ND(1)
ND(1)
35
ND(5)
6
ND(5)
33
ND (15)
18,000
263
31
Importer
2
ND(1)
ND(2)
ND(10)
75
48
8,200
180,000
ND(2)
ND(1)
ND(1)
ND (860)
ND (860)
ND (680)
ND (680)
240,000
ND (100)
200,000
2,874
2,532
Importer
3
ND(2)
ND(5)
ND(5)
ND(5)
ND(5)
390
760,000
ND(1)
ND(3)
ND(3)
ND(4)
ND(4)
ND(4)
ND(4)
36
ND(15)
50,000
814
85
Importer
4
ND(2)
ND(6)
ND(3)
6
9
2,300
71,000
ND(2)
ND(5)
ND(5)
5,600
ND (600)
ND (600)
ND (600)
230,000
ND (400)
110,000
3,065
2,903
Carbazole
violet
ND (0.8)
ND (0.5)
ND(1.2)
ND(1.2)
ND(1.2)
28
1,600
ND(1.6)
ND (0.9)
ND (0.9)
ND (20)
ND (20)
ND (20)
ND (20)
15,000
ND (20)
59,000
211
156
Calculated assuming nondetect values are zero.
ND = Not detected (value in parenthesis is the minimum detection limit)
Source: Remmersetal. (1992).
(a) all wood preservative uses of PCP and its salts were classified as "restricted use" only by
certified applicators, (b) specific worker protection measures were required, (c) limits were
placed on the HxCDD content of PCP, and (d) label restrictions for home and farm uses of PCP
prohibited its application indoors and to wood intended for interior use (with a few exceptions)
as well as its application in a manner that might result in direct exposure of domestic animals or
livestock or in the contamination of food, feed, or drinking and irrigation water.
EPA subsequently amended its Notice on the wood preservative uses to establish reliable
and enforceable methods for implementing certified limits for HxCDD and 2,3,7,8-TCDD in
registered wood preservative pesticide products (Federal Register, 1987b). Levels of 2,3,7,8-
TCDD were not allowed to exceed 1 ppb in any product, and after February 2, 1989, any
manufacturing-use PCP released for shipment could not contain HxCDD levels that exceeded an
8-55
-------
average of 2 ppm over a monthly release or a batch level of 4 ppm (a gradually phased-in
requirement). On January 21, 1987, EPA prohibited the registration of PCP and its salts for most
nonwood uses (Federal Register, 1987c). EPA deferred action on several uses (uses in
pulp/paper mills, oil wells, and cooling towers) pending receipt of additional exposure, use, and
ecological effects data. On January 8, 1993, EPA issued a press advisory stating that its special
review of these deferred nonwood uses was being terminated because all of these uses had been
either voluntarily cancelled by the registrants or cancelled by EPA for failure of the registrants to
pay the required annual maintenance fees (U.S. EPA, 1993e).
PCP was one of the most widely used biocides in the United States prior to the regulatory
actions to cancel and restrict certain of its wood and nonwood preservative uses. PCP was
registered for use as a herbicide, defoliant, mossicide, and mushroom house biocide. It also
found use as a biocide in pulp-paper mills, oil wells, and cooling towers. These latter three uses
were terminated on or before 1993 (U.S. EPA, 1993e). However, the major use (greater than
80% of consumption) of PCP was and continues to be wood preservation.
The production of PCP for wood preserving began on an experimental basis in the 1930s.
In 1947, nearly 3,200 metric tons of PCP were reported to have been used in the United States by
the commercial wood preserving industry. Use in this industry steadily increased through the
mid-1970s (American Wood Preservers Institute, 1977). Although domestic consumption
volumes are not available for all years, it is estimated, on the basis of historical production/export
data for PCP reported in Mannsville (1983), that 90 to 95% of production volume has typically
been consumed domestically rather than exported. A reasonable estimate of average annual
domestic PCP consumption during the period of 1970 to 1995 is about 400,000 metric tons. This
estimate assumes an average annual consumption rate of 20,000 metric tons/yr during the 1970s,
15,000 metric tons/yr during the 1980s, and 10,000 metric tons/yr during the 1990s.
Table 8-10 presents a compilation of published data on the CDD/CDF content of
technical-grade PCP. The only samples that have been analyzed for all dioxin-like CDDs/CDFs
were manufactured in the mid to late 1980s. Figure 8-4 presents these data in graphical form. It
is evident from the figures that the predominant congener groups are OCDD, OCDF, HpCDF,
and HpCDD and the dominant 2,3,7,8-substituted congeners are OCDD, 1,2,3,4,6,7,8-HpCDD,
and OCDF. Waddell et al. (1995) tested analytical-grade PCP (from Aldrich Chemical Co.) for
CDD/CDF content and found the same congener profile; however, the CDD/CDF levels were
three to four orders of magnitude lower. Table 8-11 presents a similar compilation of published
data on the CDD/CDF content of PCP-Na. The table shows the same patterns of dominant
congeners and congener groups reported for PCP.
8-56
-------
Congener
Figure 8-4. Congener profile for technical-grade PCP (developed from data
in last column in Table 8-10).
Samples of technical-grade PCP manufactured during the mid to late 1980s contained
about 1.7 mg TEQDF-WHO98/kg (3 mg I-TEQ/kg), based on the data presented in Table 8-10. No
published reports could be located that present the results of any congener-specific analyses of
PCP manufactured since the late 1980s. However, monthly measurements of CDD/CDF
congener group concentrations in technical PCP manufactured for use in the United States have
been reported to EPA from 1987 to the present (letter dated March 5, 1997, from Thomas
Mitchell, KMG-Bernuth, to Matthew Lorber, U.S. EPA; letter dated February 7, 1997, from John
Wilkinson, Pentachlorophenol Task Force, to Matthew Lorber, U.S. EPA; U.S. EPA, 1999a).
The average congener group concentrations reported to EPA for the years 1988 (i.e., one year
after EPA regulations were imposed limiting HxCDD and 2,3,7,8-TCDD concentrations in PCP)
to 1999 are presented in Table 8-10. In general, the average congener group concentrations
during the period of 1988 to 1999 are lower by factors of 2 to 4 than those observed in the mid to
late 1980s' full congener analysis samples. If it is assumed that the toxic CDD/CDF congeners
have also been reduced by similar factors, then the TEQ content of PCP manufactured since 1988
is about 0.6 mg TEQDF-WHO98/kg (1 mg I-TEQ/kg).
8-57
-------
An estimated 12,000 metric tons of PCP were used for wood preservation in the United
States in 1987 (WHO, 1991). An estimated 8,400 metric tons were used in 1994 (American
Wood Preservers Institute, 1995); for purposes of this report, it is assumed that an identical
amount was used in 1995. In 1999, approximately 7,710 metric tons of PCP were produced
annually in the United States (Council of Great Lakes Industries, 1999); for purposes of this
report, it is assumed that an identical amount was produced in 2000. Assuming that 95% of the
production volume was consumed domestically (Mannsville, 1983), and that all of the PCP
produced in 2000 was used for wood preservation, approximately 7,325 metric tons of PCP was
used in the United States for wood preservation. Combining these activity level estimates with
the TEQ concentration estimates presented above indicates that 20,000 jig TEQDF-WHO98
(36,000 |ig I-TEQDF), 4,800 |ig TEQDF-WHO98 (8,400 |ig I-TEQDF), and 4,175 |ig TEQDF-WHO98
(7,325 |ig I-TEQDF) were incorporated into PCP-treated wood products in 1987, 1995, and 2000,
respectively. These amounts in PCP products are not considered an environmental release and
therefore are not included in the inventory. As discussed below, there is some evidence that
releases could occur, but no consistent estimation approach could be found.
Although the estimates of the mass of TEQ in treated wood are fairly certain, no studies
are available that provide measured CDD/CDF release rate data from which a reliable estimate
can be made of the amount of CDDs/CDFs that have or will volatilize or leach from treated
wood. Several recent field studies, discussed in the following paragraphs, demonstrate that
CDDs/CDFs do apparently leach into soil from PCP-treated wood, but the studies do not provide
release rate data. No studies were located that provide any measured CDD/CDF volatilization
rates from PCP-treated wood. Although CDDs/CDFs have very low vapor pressures, they are
not bound to, nor do they react with, the wood in any way that would preclude volatilization.
Several studies, discussed below, have attempted to estimate potential CDD/CDF volatilization
releases using conservative assumptions or modeling approaches, but these estimates span many
orders of magnitude.
Gurprasad et al. (1995) analyzed three PCP-treated utility poles and their surrounding
surface soils for penta- through octa-CDD content. All three poles showed significant levels of
HxCDD (0.29 to 0.47 mg/kg), HpCDD (4.69 to 6.63 mg/kg), and OCDD (27.9 to 42.1 mg/kg),
but no PeCDD. Surface soils collected 2 cm from the poles also had detectable levels of
HxCDD, HpCDD, and OCDD; however, no consistent pattern was found between the CDD
concentrations in the poles and those in the adjacent soils. The soil concentrations did, however,
show the same relative congener group pattern observed in the wood. CDD concentrations in
soils obtained 20 cm from the poles were an order of magnitude less than those measured at 2
cm. Soils 26 m from the poles showed nondetect values or values close to the DL of 0.01 to 0.02
mg/kg.
8-58
-------
In a study of the leaching of PCP from 31 utility poles, the Electric Power Research
Institute (EPRI, 1995) found patterns of PCP distribution in soils surrounding poles similar to
those found by Gurprasad et al. (1995) for CDDs. PCP concentrations decreased by as much as
two orders of magnitude between 7.5 cm from the poles and 20 cm from the poles, with an
average decrease of slightly more than one order of magnitude over this distance. The EPRI
study also found no obvious trend between PCP concentration in the wood (eight poles analyzed)
and the age of the poles (4 to 11 years) or the PCP concentration in the surface soil. On the basis
of their results and those of the EPRI study, Gurprasad et al. concluded that CDDs probably leach
from PCP-treated utility poles with the PCP/oil carrier and travel in the soil in a similar manner.
Wan (1995) and Wan and Van Oostdam (1995) measured CDD/CDF concentrations in
waters and sediments from ditches surrounding utility poles and railroad ties and demonstrated
that chlorophenol-treated wood could serve as a source of CDD/CDFs to the aquatic
environment. Ten samples were collected at each of six utility pole sites and five railroad tie
sites 1 to 2 days after major rainfall events and then were composited into one sample per site
prior to analyses. Total CDDs (mean value of 76.7 mg/kg) and total CDFs (mean value of 18.7
mg/kg) detected in chlorophenol/creosote-treated utility poles were about six to eight times
greater, respectively, than the CDD and CDF concentrations detected in chlorophenol/creosote-
treated railroad ties.
Total CDDs found in water from railway ditches without utility poles (i.e., only treated
railroad ties were present) were approximately 20 times higher than the background level found
in farm ditch water. Total CDDs in railway ditches with utility poles were 4,300 times higher
than the background levels. Water from railway ditches without utility poles contained total
CDF levels 13 times higher than background levels, whereas water in ditches adjacent to poles
had levels 8,500 times higher than background levels. Total CDDs in ditch sediments adjacent to
and 4 m downstream of utility poles were about 5,900 and 2,200 times higher, respectively, than
background levels; total CDFs for the same sites were about 8,100 and 1,700 times higher,
respectively, than background levels. Total CDDs found in ditch sediments of railway and ditch
sediments adjacent to utility poles were about 5 and 700 times higher, respectively, than
background levels, and total CDFs were about 9 and 1,800 times higher, respectively, than
background levels. Both CDDs and CDFs were found in utility ditch sediments 4 m downstream
of treated power poles, but at levels 200 and 400 times lower, respectively, than those found
adjacent to poles, indicating that they were transported from point sources of contamination. The
corresponding values for CDFs were 5,400 and 8,000 times higher, respectively, in
concentration.
Bremmer et al. (1994) estimated an annual release of 15 to 125 g I-TEQDF from PCP-
treated wood in the Netherlands. The lower estimate was based on three basic assumptions:
8-59
-------
(1) the half-life of PCP in treated wood is 15 years (according to industry sources), (2) the half-
life of CDDs/CDFs in treated wood is 10 times that of PCP (i.e., 150 years) because of the lower
vapor pressures of CDDs/CDFs relative to PCP, and (3) the typical CDD/CDF concentration in
PCP has been 3,000 |ig/kg. The higher estimate was based on an assumed half-life of PCP in
wood of 15 years and the results of an indoor air study by Papke et al. (1989) conducted at
several kindergartens where PCP-treated wood had been used. Although Papke et al. found no
clear correlation between indoor air concentrations of CDDs/CDFs and PCP across the range of
CDD/CDF concentrations observed in the 20-plus samples (2.6 to 427 pg CDD/CDF/m3), there
did appear to be a positive correlation at the sites with more elevated CDD/CDF concentrations.
Bremmer et al. (1994) reported the average ratio of PCP to I-TEQ DF air concentrations at these
elevated sites to be 1:5 x 10"6 (or about the same ratio as the concentration of I-TEQ DF in
technical PCP). The results of the Papke et al. (1989) study imply that CDDs/CDFs may be
released from PCP-treated wood at the same rate as is PCP rather than at a rate 10 times slower.
Rappe (1995) used the emission factor approach developed by Bremmer et al. (1994) and
an assumed U.S. usage volume of PCP over the past 50 years (0.5 million metric tons) to
estimate that as much as 10.5 kg I-TEQDF could volatilize from PCP-treated wood in the United
States annually. Eitzer and Kites (1987) derived a dramatically different estimate: 3 kg/yr of
total CDDs/CDFs (or 66 g I-TEQDF per year, assuming an I-TEQDF content in PCP of 3 mg/kg).
Eitzer and Kites based their estimate on an assumption that 0.1% of the PCP produced annually
enters the atmosphere and that the CDD/CDF contaminants present in the PCP (assumed to be
130 mg/kg) are released to the atmosphere at the same rate as the PCP (i.e., 0.1%). The basis for
the first assumption by Eitzer and Kites is not clear because U.S. EPA (1980), which was cited as
the source of the 0.1% emission factor, does not appear to address volatilization of PCP from in-
service treated wood. The report does, however, estimate that most PCP in treated wood leaches
relatively rapidly from the wood, presumably to land, within a period of 12 years.
Eduljee and Dyke (1996) and Douben et al. (1995) estimated that 0.8 g I-TEQDF is
released to the air annually from PCP-treated wood in the United Kingdom. This estimate was
based on the assumed emission of 0.1% of the CDDs/CDFs present in PCP-treated wood during
the first year of the service life of the wood that was assumed by Eitzer and Kites (1987). No
emissions were assumed for subsequent years of use of the treated wood.
The California Air Resources Board (Chinkin et al., 1987) generated estimates of
CDD/CDF volatilization releases at wood treatment facilities from bundles of treated wood that
remain on site for 1 month prior to shipment. An "adapted" version of a model developed by
McCord (1981) was used for estimating volatile releases from a constantly filling lagoon. The
model is primarily driven by chemical-specific vapor pressures and air diffusivity coefficients.
Chinkin et al. did not provide all model input parameter values used to generate the emission
estimates. However, running the model with typical dimensions for treated poles yields an I-
8-60
-------
TEQDF emission rate on the order of 6E-12 g/yr-pole, an extremely low number (170 billion poles
would together emit 1 g TEQ/yr).
In addition to cancelling some pesticide registrations and establishing product standards,
EPA's Office of Pesticide Programs (OPP) issued two DCIs in 1987. Pesticide manufacturers
are required to register their products with EPA in order to market them commercially in the
United States. Through the registration process, mandated by FIFRA, EPA can require that the
manufacturer of each active ingredient generate a wide variety of scientific data through several
mechanisms. The most common process is the five-phase reregi strati on process, with which the
manufacturers (i.e., registrants) of older pesticide products must comply. In most registration
activities, registrants must generate data under a series of strict testing guidelines, 40 CFR
158—Pesticide Assessment Guidelines (U.S. EPA, 1988c). EPA can also require additional data
from registrants, when necessary, through various mechanisms, including the DCI process.
The purpose of the first DCI, dated June and October 1987, "Data Call-In Notice for
Product Chemistry Relating to Potential Formation of Halogenated Dibenzo-^-dioxin or
Dibenzofuran Contaminants in Certain Active Ingredients," was to identify, using an analysis of
raw materials and process chemistry, those pesticides that might contain halogenated dibenzo-p-
dioxin (HDD) and halogenated dibenzofuran (HDF) contaminants. The 93 pesticides (76
pesticide active ingredients) to which the DCI applied, along with their corresponding
Shaughnessey and Chemical Abstract code numbers, are presented in Table 8-29. (The
Shaughnessey code is an internal EPA tracking system. It is of interest because chemicals with
similar code numbers are similar in chemical nature [e.g., salts, esters, and acid forms of 2,4-D].)
All registrants supporting registrations for these chemicals were subject to the
requirements of the DCI unless their product qualified for a Generic Data Exemption (i.e., a
registrant exclusively used a FIFRA-registered pesticide product as a source of an active
ingredient identified in Table 8-29 in formulating a product). Registrants whose products did not
meet the Generic Data Exemption were required to submit the types of data listed below to
enable EPA to assess the potential for formation of tetra- through hepta-HDD or -HDF
contaminants during manufacture.
• Product identity and disclosure of ingredients. EPA required submittal of a
Confidential Statement of Formula (CSF), based on the requirements specified in 40
CFR 158.108 and 40 CFR 158.120, Subdivision D: Product Chemistry. Registrants
who had previously submitted still-current CSFs were not required to resubmit this
information.
• Description of beginning materials and manufacturing process. Under the
requirements mandated by 40 CFR 158.120, Subdivision D, EPA required submittal
of a manufacturing process description for each step of the manufacturing process,
including specification of the range of acceptable conditions of temperature, pressure,
or pH at each step.
8-61
-------
Table 8-29. Status of first pesticide Data Call-In: pesticides suspected of having the potential to become
contaminated with dioxins if synthesized under conditions favoring dioxin formation
Shaughnessey
code
000014
008706
009105
012001
012101
019201
019202
019401
025501
027401
28201
028601
029201
29601
029902
029906
030602
031301
031503
031516
031563
034502
035502
35505
35506
35901
Pesticide [active ingredient]
Dichlorodifluoromethane
O-(4-Bromo-2,5-dichlorophenyl) O,O-dimethyl phosphorothioate
Dimethylamine 2,3,5-triiodobenzoate
Neburon
Crufomate
MCPB, 4-butyric acid [4-(2-Methyl-4-chlorophenoxy)butyric acid]
MCPB, Na salt [Sodium 4-(2-methyl-4-chlorophenoxy)butyrate]
4-Chlorophenoxyacetic acid
Chloroxuron
Dichlobenil
Propanil [3 ',4'-Dichloropropionanilide]
Dichlofenthion [O-(2,4-Dichlorophenyl) O,O-diethyl phosphorothioate)]
DDT [Dichloro diphenyl trichloroethane]
Dichlone [2,3-dichloro-l,4-naphthoquinone]
Ammonium chloramben [3-amino-2,5-dichlorobenzoic acid]
Sodium chloramben [3-amino-2,5-dichlorobenzoic acid]
Sodium 2-(2,4-dichlorophenoxy)ethyl sulfate
DCNA [2,6-Dichloro-4-nitroaniline]
Potassium 2-(2-methyl-4-chlorophenoxy)propionate
MCCP, DBA Salt [Diethanolamine 2-(2-methyl-4-chlorophenoxy)propionate]
MCPP, IOE [Isooctyl 2-(2-methyl-4-chlorophenoxy)propionate]
Dicapthon [O-(2-chloro-4-nitrophenyl) O,O-dimethyl phosphorothioate]
Monuron trichloroacetate [3 -(4-chlorophenyl)- 1 , 1 -dimethylurea trichloroacetate]
Diuron [3-(3,4-dichlorophenyl)-l, 1 -dimethylurea]
Linuron [3 -(3 ,4-dichlorophenyl)-l -methoxy- 1 -methylurea]
Metobromuron [3 -(p-bromophenyl)- 1 -methoxy- 1 -methylurea]
CAS No.
75-71-8
2104-96-3
17601-49-9
555-37-3
299-86-5
94-81-5
6062-26-6
122-88-3
1982-47-4
1194-65-6
709-98-8
97-17-6
50-29-3
117-80-6
1076-46-6
1954-81-0
136-78-7
99-30-9
1929-86-8
1432-14-0
28473-03-2
2463-84-5
140-41-0
330-54-1
330-55-2
3060-89-7
Support
withdrawn
Yes
Yes
Yes
Yes
Yes
No
No
No
Yes
No
No
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
No
Yes
Yes
No
No
Yes
Testing
required
~
~
~
-
-
Yes
No
Yes
-
Yes
No
-
~
-
~
~
~
Yes
~
~
No
~
-
No
No
~
oo
ON
to
-------
Table 8-29. Status of first pesticide data call-in: pesticides suspected of having the potential to become
contaminated with dioxins if synthesized under conditions favoring dioxin formation (continued)
Shaughnessey
code
53501
55001
55005
55201
57501
58102
58301
58802
59401
61501
62201
62202
62203
62204
62206
62207
62208
62209
62210
62211
62212
62213
62214
62215
64202
Pesticide [active ingredient]
Methyl parathion [O,O-Dimethyl O-p-nitrophenyl phosphorothioate]
Dichlorophene [Sodium 2,2'-methylenebis(4-chlorophenate)]
Dichlorophene, sodium salt [Sodium 2,2'-methylenebis(4-chlorophenate)]
1 ,2,4,5-Tetrachloro-3 -nitrobenzene
Ethyl parathion [O,O-diethyl O-p-nitrophenyl phosphorothioate]
Carbophenothion [S-(((p-chlorophenyl)thio)methyl) O,O-diethyl phosphorodithioate]
Ronnel [O,O-dimethyl O-(2,4,5-trichlorophenyl) phosphorothioate]
Mitin FF [Sodium 5-chloro-2-(4-chloro-2-(3-(3,4-dichlorophenyl)ureido)phenoxy)
benzenesulfonate]
Orthodichlorobenzene
Paradichlorobenzene
Chlorophene [2-Benzyl-4-chlorophenol]
Potassium 2-benzyl-4-chlorophenate
Sodium 2-benzyl-4-chlorophenate
2-Chlorophenol
2-Chloro-4-phenylphenol
Potassium 2-chloro-4-phenylphenate
4-Chloro-2-phenylphenol
4-Chloro-2-phenylphenol, potassium salt
6-Chloro-2-phenylphenol
6-Chloro-2-phenylphenol, potassium salt
4-Chloro-2-phenylphenol, sodium salt
6-Chloro-2-phenylphenol, sodium salt
4 and 6-Chloro-2-phenylphenol, diethanolamine salt
2-Chloro-4-phenylphenol, sodium salt
4-Chloro-2-cyclopentylphenol
CAS No.
298-00-0
97-23-4
10254-48-5
117-18-0
56-38-2
786-19-6
229-84-3
3567-25-7
95-50-1
106-46-7
120-32-1
35471-49-9
3184-65-4
95-57-8
92-04-6
18128-16-0
NA
53404-21-0
85-97-2
18128-17-1
10605-10-4
10605-11-5
53537-63-6
31366-97-9
13347-42-7
Support
withdrawn
No
Yes
Yes
Yes
No
Yes
Yes
No
Yes
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Testing
required
No
~
~
-
No
~
~
No
-
No
No
In review
In review
-
-
-
-
~
-
-
-
-
-
~
-
oo
-------
Table 8-29. Status of first pesticide data call-in: pesticides suspected of having the potential to become
contaminated with dioxins if synthesized under conditions favoring dioxin formation (continued)
Shaughnessey
code
64208
64209
64214
64218
67707
69105
69144
77401
77406
78780
79202
79301
80403
80811
81901
82602
84101
84901
86801
97003
100601
101001
101101
104301
106001
Pesticide [active ingredient]
Fentichlor [2,2'-Thiobis(4-chloro-6-methylphenol)]
Fentichlor [2,2'-Thiobis(4-chlorophenol)]
4-Chloro-2-cyclopentylphenol, potassium salt of
4-Chloro-2-cyclopentylphenol, sodium salt
Chlorophacinone
ADBAC [Alkyl* dimethyl benzyl ammonium chloride *(50% C14, 40% C12, 10%
C16)]
ADBAC [Alkyl* dimethyl 3,4-dichlorobenzyl ammonium chloride *(61% C12, 23%
C14, 11%C16, 5% CIS)]
Niclosamide [2-Aminoethanol salt of 2',5-dichloro-4'-nitrosalicylanilide]
5-Chlorosalicylanilide
2-Methyl-4-isothiazolin-3 -one
Tetradifon [4-chlorophenyl 2,4,5-trichlorophenyl sulfone]
Chloranil [tetrachloro-p-benzoquinone]
6-Chlorothymol
Anilazine [2,4-Dichloro-6-(o-chloroanilino)-s-triazine]
Chlorothalonil [tetrachloroisophthalonitrile]
Sodium 2,3 ,6-Trichlorophenylacetate
Chlorfenvinphos
O-(2-Chloro-l-(2,5-dichlorophenyl)vinyl) O,O-diethyl phosphorothioate
PCMX [4-Chloro-3,5-xylenol]
Piperalin [3-(2-Methylpiperidino)propyl 3,4-dichlorobenzoate]
Fenamiphos
p-Chlorophenyl diiodomethyl sulfone
Metribuzin
Bifenox [methyl 5-(2,4-dichlorophenoxy)-2-nitrobenzoate]
Methazole [2-(3,4-dichlorophenyl)-4-methyl-l,2,4-oxadiazolidine-3,5-dione]
CAS No.
4418-66-0
97-24-5
35471-38-6
53404-20-9
3691-35-8
68424-85-1
NA
1420-04-8
4638-48-6
NA
116-29-0
118-75-2
89-68-9
101-05-3
1897-45-6
2439-00-1
470-90-6
1757-18-2
88-04-0
3478-94-2
NA
20018-12-6
21087-64-9
42576-02-3
20354-26-1
Support
withdrawn
Yes
Yes
Yes
Yes
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
No
No
No
Yes
No
Yes
Yes
Testing
required
~
~
~
-
No
No
No
No
-
~
~
~
~
~
Yes
-
~
~
No
No
No
~
No
~
~
oo
-------
Table 8-29. Status of first pesticide data call-in: pesticides suspected of having the potential to become
contaminated with dioxins if synthesized under conditions favoring dioxin formation (continued)
Shaughnessey
code
108201
109001
109301
109302
109801
109901
110902
111401
111601
111901
112802
113201
119001
123901
125601
128838
206600
Pesticide [active ingredient]
Diflubenzuron [N-(((4-chlorophenyl)amino)carbonyl)-2,6-difluorobenzamide]
Oxadiazon [2-tert-butyl-4-(2,4-dichloro-5-isopropoxyphenyl)- delta 2 -1,3,4-
oxadiazoline-5-one]
Fenvalerate
Fluvalinate [N-2-Chloro-4-trifluoromethyl)phenyl-DL-valine (+-)-cyano(3 -phenoxy-
phenyl)methyl ester]
Iprodione [3 -(3 ,5-Dichlorophenyl)-N-( 1 -methylethyl)-2,4-dioxo-l -
imidazolidinecarboxamide (9CA)]
Triadimefon
[ 1 -(4-Chlorophenoxy)-3 ,3 -dimethyl- 1 -( 1H-1 ,2,4-triazol-l -yl)-2-butanone]
Diclofop - methyl [methyl 2-(4-(2,4-dichlorophenoxy)phenoxy)propanoate]
Profenofos [O-(4-Bromo-2-chlorophenyl)-O-ethyl S-propyl phosphorothioate]
Oxyfluorfen [2-chloro-l-(3-ethoxy-4-nitrophenoxy)-4-(trifluoromethyl)benzene]
Imazalil [l-(2-(2,4-Dichlorophenyl)-2-(2-propenyloxy)ethyl)-lH-imidazole]
Bromothalin [N-Methyl-2,4-dinitro-n-(2,4,6-tribromophenyl)-6-
(trifuloromethyl)benzenamine]
Vinclozolin [3-(3,5-Dichlorophenyl)-5-ethenyl-5-methyl-2,4-oxazolidinedione (9CA)]
Fenridazon [Potassium l-(p-chlorophenyl)-l,4-dihydro-6-methyl-4-oxo- pyridazine-
3-carboxylate]
Tridiphane [2-(3,5-Dichlorophenyl)-2-(2,2,2-trichloroethyl) oxirane]
Paclobutrazol
Linalool
Fenarimol [a-(2-chlorophenyl)-a-(4-chlorophenyl)-5-pyrimidinemethanol]
CAS No.
35367-38-5
19666-30-9
51630-58-1
69409-94-5
36734-19-7
43121-43-3
51338-27-3
41198-08-7
42874-03-3
35554-44-0
63333-35-7
50471-44-8
83588-43-6
58138-08-2
76738-62-0
78-70-6
60168-88-9
Support
withdrawn
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Testing
required
Yes
Yes
In review
No
No
No
Yes
In review
In review
No
No
No
In review
No
No
In review
No
oo
NA = Not available
~ = No information given
-------
• Discussion of the formation of impurities. Under the requirements mandated by 40
CFR 158.120, Subdivision D, EPA required submittal of a detailed discussion and
assessment of the possible formation of HDDs and HDFs.
Registrants had the option of voluntarily canceling their product or "reformulate to remove an
active ingredient" to avoid having to comply with the DCI.
The second DCI, dated June and October 1987, "Data Call-In for Analytical Chemistry
Data on Polyhalogenated Dibenzo-/?-Dioxins/Dibenzofurans (HDDs and HDFs)," was issued for
68 pesticides (16 pesticide active ingredients) suspected to be contaminated with CDDs/CDFs
(see Table 8-30). All registrants supporting registrations for these pesticides were subject to the
requirements of this DCI unless the product qualified for various exemptions or waivers.
Pesticides covered by the second DCI were strongly suspected by EPA of containing detectable
levels of CDDs/CDFs. Under the second DCI, registrants whose products did not qualify for an
exemption or waiver were required to generate and submit the following types of data in addition
to the data requirements of the first DCI:
• Quantitative method for measuring CDDs or CDFs. Registrants were required to
develop an analytical method for measuring the HDD/HDF content of their products.
The DCI established a regimen for defining the precision of the analytical method.
Target LOQs were established in the DCI for specific CDD/CDF congeners (see
Table 8-25).
• Certification of limits of CDDs or CDFs. Registrants were required to submit a
"Certification of Limits" in accordance with 40 CFR 158.110 and 40 CFR 158.120,
Subdivision D. Analytical results were required that met the guidelines described
above.
Registrants could select one of two options to comply with the second DCI. The first
option was to submit relevant existing data, develop new data, or share the cost of developing
new data with other registrants. The second option was to alleviate the DCI requirements
through several exemption processes, including a Generic Data Exemption, voluntary
cancellation, reformulation to remove the active ingredient of concern, an assertion that the data
requirements did not apply, or the application or award of a low-volume, minor-use waiver.
The data contained in CSFs, as well as any other data generated under 40 CFR 158.120,
Subdivision D, are typically considered confidential business information (CBI) under the
guidelines prescribed in FIFRA because they usually contain information regarding proprietary
manufacturing processes. In general, all analytical results submitted to EPA in response to both
DCIs are considered CBI and cannot be released by EPA into the public domain. Summaries
8-66
-------
Table 8-30. Status of second pesticide Data Call-In: pesticides suspected of being contaminated with dioxins
Shaughnessey
code
29801
29802
29803
30001
30002
30003
30004
30005
30010
30011
30013
30014
30016
30017
30019
30020
30021
30023
30024
30025
30028
30029
30030
30033
30034
30035
Pesticide [active ingredient]
Dicamba [3,6-dichloro-o-anisic acid]
Dicamba dimethylamine [3,6-dichloro-o-anisic acid]
Diethanolamine dicamba [3, 6-dichloro-2 -anisic acid]
2,4-Dichlorophenoxyacetic acid
Lithium 2,4-dichlorophenoxyacetate
Potassium 2,4-dichlorophenoxyacetate
Sodium 2,4-dichlorophenoxyacetate
Ammonium 2,4-dichlorophenoxyacetate
Alkanol* amine 2,4-dichlorophenoxyacetate *(salts of the ethanol and
ispropanol series)
Alkyl* amine 2,4-dichlorophenoxyacetate *(100% C12)
Alkyl* amine 2,4-dichlorophenoxyacetate *(100% C14)
Alkyl* amine 2,4-dichlorophenoxyacetate *(as in fatty acids of tall oil)
Diethanolamine 2,4-dichlorophenoxyacetate
Diethylamine 2,4-dichlorophenoxyacetate
Dimethylamine 2,4-dichlorophenoxyacetate
N,N-Dimethyloleylamine 2,4-dichlorophenoxyacetate
Ethanolamine 2,4-dichlorophenoxyacetate
Heptylamine 2,4-dichlorophenoxyacetate
Isopropanolamine 2,4-dichlorophenoxyacetate
Isopropylamine 2,4-dichlorophenoxyacetate
Morpholine 2,4-dichlorophenoxyacetate
N-Oleyl-l,3-propylenediamine 2,4-dichlorophenoxyacetate
Octylamine 2,4-dichlorophenoxyacetate
Triethanolamine 2,4-dichlorophenoxyacetate
Triethylamine 2,4-dichlorophenoxyacetate
Triisopropanolamine 2,4-dichlorophenoxyacetate
CAS No.
1918-00-9
2300-66-5
25059-78-3
94-75-7
3766-27-6
14214-89-2
2702-72-9
2307-55-3
NA
2212-54-6
28685-18-9
NA
5742-19-8
20940-37-8
2008-39-1
53535-36-7
3599-58-4
37102-63-9
6365-72-6
5742-17-6
6365-73-7
2212-59-1
2212-53-5
2569-01-9
2646-78-8
32341-80-3
Support
withdrawn
No
No
Yes
No
No
Yes
No
Yes
Yes
Yes
Yes
Yes
No
Yes
No
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
No
No
Testing
required
Yes
Yes
~
Yes
No
-
No
-
~
~
~
-
No
-
No
-
-
-
-
No
-
-
-
~
No
No
oo
-------
Table 8-30. Status of second pesticide Data Call-In: pesticides suspected of being contaminated with dioxins
(continued)
Shaughnessey
code
30039
30052
30053
30055
30056
30062
30063
30064
30065
30066
30072
30801
30804
30819
30853
30856
30863
31401
31419
31453
31463
31501
31519
35301
44901
Pesticide [active ingredient]
N,N-Dimethyl oleyl-linoleyl amine 2,4-dichlorophenoxyacetate
Butoxyethoxypropyl2,4-dichlorophenoxyacetate
Butoxyethyl 2,4-dichlorophenoxyacetate
Butoxypropyl 2,4-dichlorophenoxyacetate
Butyl 2,4-dichlorophenoxyacetate
Isobutyl 2,4-dichlorophenoxyacetate
Isooctyl(2-ethylhexyl) 2,4-dichlorophenoxyacetate
Isooctyl(2-ethyl-4-methylpentyl) 2,4-dichlorophenoxyacetate
Isooctyl(2-octyl) 2,4-dichlorophenoxyacetate
Isopropyl 2,4-dichlorophenoxyacetate
Propylene glycol butyl ether 2,4-dichlorophenoxyacetate
4-(2,4-Dichlorophenoxy)butyric acid
Sodium 4-(2,4-dichlorophenoxy)butyrate
Dimethylamine 4-(2,4-dichlorophenoxy)butyrate
Butoxyethanol4-(2,4-dichlorophenoxy)butyrate
Butyl 4-(2,4-dichlorophenoxy)butyrate
Isooctyl4-(2,4-dichlorophenoxy)butyrate
2-(2,4-Dichlorophenoxy)propionic acid
Dimethylamine 2-(2,4-dichlorophenoxy)propionate
Butoxyethyl 2-(2,4-dichlorophenoxy)propionate
Isooctyl2-(2,4-dichlorophenoxy)propionate
MCPP acid [2-(2-Methyl-4-chlorophenoxy)propionic acid]
MCPP, DMA [Dimethylamine 2-(2-methyl-4-chlorophenoxy)propionate]
Bromoxynil [3,5-dibromo-4-hydroxybenzonitrile]
Hexachlorophene [2,2'-Methylenebis(3 ,4,6-trichlorophenol)]
CAS No.
55256-32-1
1928-57-0
1929-73-3
1928-45-6
94-80-4
1713-15-1
1928-43-4
25168-26-7
1917-97-1
94-11-1
1320-18-9
94-82-6
10433-59-7
2758-42-1
32357-46-3
6753-24-8
1320-15-6
120-36-5
53404-32-3
53404-31-2
28631-35-8
7085-19-0
32351-70-5
1689-84-5
70-30-4
Support
withdrawn
Yes
Yes
No
Yes
Yes
Yes
No
Yes
Yes
No
Yes
No
No
No
Yes
Yes
Yes
No
No
No
No
No
No
No
Yes
Testing
required
~
-
No
-
-
~
Yes
-
-
No
-
Yes
No
No
~
-
~
Yes
No
No
No
Yes
No
Yes
~
oo
ON
oo
-------
Table 8-30. Status of second pesticide Data Call-In: pesticides suspected of being contaminated with dioxins
(continued)
Shaughnessey
code
44902
44904
54901
63004
63005
63006
63007
64203
64212
64219
64220
64501
78701
79401
82501
83701
104101
Pesticide [active ingredient]
Hexachlorophene, Na salt [Monosodium
2,2'-methylenebis(3,4,6-trichlorophenate)]
Hexachlorophene, K salt [Potassium
2,2'-methylenebis(3,4,6-trichlorophenate)]
Irgasan [5-Chloro-2-(2,4-dichlorophenoxy)phenol]
Tetrachlorophenols
Tetrachlorophenols, sodium salt
Tetrachlorophenols, alkyl* amine salt*(as in fatty acids of coconut oil)
Tetrachlorophenols, potassium salt
Bithionolate sodium [Disodium 2,2'-thiobis(4,6-dichlorophenate)]
Phenachlor [2,4,6-Trichlorophenol]
Potassium 2,4,6-trichlorophenate
2,4,6-Trichlorophenol, sodium salt
Phenothiazine
Dacthal-DCPA [Dimethyl tetrachloroterephthalate]
Endosulfan [hexachlorohexahydromethano-2,4,3-benzodioxathiepin-3-oxide]
Silvex [2-(2,4,5-trichlorophenoxy)propionic acid]
Tetrachlorvinphos [2-Chloro-l-(2,4,5-trichlorophenyl)vinyl dimethyl
phosphate]
Edolan [Sodium l,4',5'-trichloro-2'-(2,4,5-trichlorophenoxy)
methanesulfonaniride]
CAS No.
5736-15-2
67923-62-0
3380-34-5
25167-83-3
25567-55-9
NA
53535-27-6
6385-58-6
88-06-2
2591-21-1
3784-03-0
92-84-2
1861-32-1
115-29-7
93-72-1
961-11-5
69462-14-2
Support
withdrawn
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
Yes
No
Yes
Testing
required
~
~
Yes
~
~
~
~
~
-
~
~
-
Yes
No
~
Yes
~
oo
vo
NA = Not available
~ = No information given
-------
based on the trends identified in those data as well as in data made public by EPA are
summarized below.
The two DCIs included 161 pesticides. Of these, 92 are no longer supported by
registrants. Following evaluation of the process chemistry submissions required under the DCIs,
OPP determined that formation of CDDs/CDFs was not likely during the manufacture of 43 of
the remaining 69 pesticides; thus, analysis of samples of these 43 pesticides was not required by
OPP. Evaluation of process chemistry data is ongoing at OPP for an additional 7 pesticides.
Table 8-30 indicates which pesticides are no longer supported and those for which OPP has
required further testing to determine whether CDDs/CDFs are present as contaminants (U.S.
EPA, 1995g). OPP required that analysis of production samples be performed on the remaining
19 pesticides (see Table 8-31). The status of the analytical data generation/evaluation to date is
summarized as follows: (a) no detection of CDDs/CDFs above the LOQs in registrant
submissions for 13 active ingredients, (b) detection of CDDs/CDFs above the LOQs for 2,4-D
acid (two submissions) and 2,4-D 2-ethyl hexyl acetate (one submission), and (c) ongoing data
generation or evaluation for four pesticides.
Table 8-32 presents a summary of results obtained by EPA for CDDs/CDFs in eight
technical 2,4-D herbicides; these data were extracted from program files in OPP. Because some
of these files contained CBI, the data in this table were reviewed by OPP staff to ensure that no
CBI was being disclosed (memorandum dated May 28, 1996, from S. Funk, U.S. EPA, to W.
Hazel, U.S. EPA). Figure 8-5 presents a congener profile for 2,4-D based on the average
congener concentrations reported in Table 8-32.
Schecter et al. (1997) reported the results of analyses of samples of 2,4-D manufactured
in Europe, Russia, and the United States (see Table 8-33). The total TEQ concentrations
measured in the European and Russian samples were similar to those measured in the EPA DCI
samples; however, the levels reported by Schecter et al. for U.S. samples were significantly
lower. Similarly, Masunaga et al. (2001) reported the analyses of two agrochemical formulations
containing 2,4-D manufactured in Japan (Table 8-33). The total TEQ concentration measured in
one of the samples was similar to what Schecter et al. (1997) reported for the U.S. samples; no
TEQ was detected in the other sample.
As discussed in Section 11.2.1 (Chapter 11), an estimated 28,100 metric tons of 2,4-D
were used in the United States in 2000, making it one of the top 10 pesticides in terms of quantity
used (EPA proprietary data). The pesticide 2,4-D is the only product judged to have the potential
for environmental release through its agricultural use. However, no estimate of environmental
release can be made for 2000. Since 1995, the chemical manufacturers of 2,4-D have been
undertaking voluntary actions to significantly reduce the dioxin content of the product. No
information is available on the level of dioxin contamination, if any, that may have been present
8-70
-------
Table 8-31. Summary of analytical data submitted to EPA in response to pesticide Data Call-In(s)
Shaughnessey
code
019201
019401
027401
029801
029802
030001
030063
030801
031301
031401
031501
035301
054901
078701
081901
083701
108201
109001
110902
Pesticide
Common name
MCPB, 4-butyric acid
4-CPA
Dichlobenil
Dicamba
Dicamba, dimethylamine
2,4-D
2,4-D, 2EH
2,4-DB
DCNA
2,4-DP
Mecoprop (MCPP)
Bromoxynil
Irgasan
Dacthal (DCPA)
Chlorothalonil
Tetrachlorvinphos
Diflubenzuron
Oxadiazon
Dichlofop-methyl
Chemical name
4-(2-methyl-4-chlorophenoxy)butyric acid
4-Chlorophenoxyacetic acid
2,6-Dichlorobenzonitrile
3,6-Dichloro-o-anisic acid
3,6-Dichloro-o-anisic acid, dimethylamine salt
2,4-Dichlorophenoxy acetic acid
Isooctyl(2-ethylhexyl)2,4-dichlorophenoxyacetate
4-(2,4-Dichlorophenoxy)butyric acid
2,6-Dichloro-4-nitroaniline
2-(2,4-Dichlorophenoxy)propionic acid
2-(2-methyl-4-chlorophenoxy)propionic acid
3,5-Dibromo-4-hydroxybenzonitrile
5-Chloro-2-(2,4-dichlorophenoxy)phenol
Dimethyl tetrachloroterephthalate
Tetrachloroisophthalonitrile
2-Chloro-l-(2,4,5-trichlorophenyl)vinyl dimethyl phosphate
N-(((4-chlorophenyl)amino)carbonyl)-2,6-difluorobenzamide
2-Tert-butyl-4(2,4-dichloro-5-isopropoxyphenyl)-delta2-l,3,4-oxadiazoline-5-one
Methyl-2-(4-(2,4-dichlorophenoxy)phenoxy)propanoate
Number of
positive"
submissions to
date
0
0
0
0
0
2
1
0
Pending
0
0
0
0
Pending
Pending
0
0
Pending
0
oo
Tositive is defined as the detection of any congener at a concentration equal to or exceeding the limits of quantitation listed in Table 8-27.
Sources: U.S. EPA (19951); facsimile dated March 27, 1996 from S. Funk, U.S. EPA, to D. Cleverly, U.S. EPA.
-------
Table 8-32. Summary of results for CDDs/CDFs in technical 2,4-D and 2,4-D
ester herbicides
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
A
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
TOTAL0
I-TEQDF
TEQDF-WH098
EPA
LOQa
(ug/kg)
0.1
0.5
2.5
2.5
2.5
100.0
-
1
5
5
25
25
25
25
1,000
1,000
-
Total
no. of
technicals
8
8
8
8
8
8
~
8
8
7
8
8
8
8
8
8
~
Number of
technicals
greater than
LOQ
2
3
0
0
0
0
-
0
0
0
0
0
0
0
0
0
-
Observed
maximum
cone.
(US/kg)
0.13
2.6
0.81
0.77
0.68
1.5
~
0.27
0.62
0.73
1.6
1.2
1.4
1.1
8.3
1.2
~
Average
conc.b
(u£/kg)
0.06
0.78
0.31
0.39
0.24
0.21
~
0.07
0.38
0.07
0.36
0.11
0.16
0.14
2.17
0.18
~
5.6
0.7
1.1
aLOQ required by EPA in the data call-in.
bAverage of the mean results for multiple analyses of four technical 2,4-D and/or 2,4-D ester products for which
detectable CDD/CDF congener concentrations less than the LOQs were quantified; nondetect values were
assumed to be zero.
Total equals the sum of the individual congener averages.
LOQ = Limit of quantitation
~ = Analyses not performed
Source: U.S. EPA Office of Pesticide Programs file.
in 2,4-D in 2000. An estimated 26,300 and 30,400 metric tons were used during 1995 and 1987,
respectively (U.S. EPA, 1997e, 1988d). On the basis of the average CDD/CDF congener
concentrations in 2,4-D presented in Table 8-33 (not including OCDD and OCDF), the
corresponding TEQDF-WHO98 concentration is 1.1 |ig/kg (0.7 jig I-TEQDF/kg). Combining this
8-72
-------
Congener
Figure 8-5. Congener profile for 2,4-D (salts and esters) (based on mean
concentrations reported in Table 8-34).
TEQ concentration with the activity level estimates for 1995 and 1987 indicates that 28.9 g
TEQDF-WHO98 (18.4 g I-TEQDF) were released in 1995 and 33.4 g TEQDF-WHO98 (21.3 g I-
TEQDF) in 1987. The release estimates for 1987 and 1995 are assigned a high confidence rating,
indicating high confidence in both the production and the emission factor estimates. Because no
estimate can be made for 2000, it is rated as Category E (no estimate of environmental release of
CDDs/CDFs can be made because of the poor quality of existing information).
8.4. OTHER CHEMICAL MANUFACTURING AND PROCESSING SOURCES
8.4.1. Municipal Wastewater Treatment Plants
8.4.1.1. Sources
CDDs/CDFs have been measured in nearly all sewage sludges tested, although the
concentrations and, to some extent, the congener profiles and patterns differ widely. Potential
sources of the CDDs/CDFs include microbial formation (discussed in Chapter 9), runoff to
8-73
-------
Table 8-33. CDD/CDF concentrations (ug/kg) in samples of 2,4-D and pesticide formulations containing 2,4-D
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-71,2,3,4,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-71,2,3,4,7,9
-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
(nondetect = 0)
Total 2,3,7,8-CDF
(nondetect = 0)
Total I-TEQDF
(nondetect = 0)b
(nondetect = 0)b
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD7CDF
Acbar
Super
(Gaza
City")
ND(O.l)
0.1
ND(O.l)
ND(O.l)
ND(O.l)
0.1
0.1
0.3
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
ND(O.l)
0.1
ND(O.l)
0.2
0.3
0.6
0.082
0.134
—
-
-
-
-
—
—
-
-
-
-
Amco
Super
(Gaza
City")
ND(O.l)
ND(O.l)
ND(O.l)
0.2
ND(O.l)
1.2
2.6
ND(O.l)
0.2
ND(O.l)
0.1
ND(O.l)
ND(O.l)
ND(O.l)
0.8
ND(O.l)
3.8
4
4.9
0.066
0.061
—
-
-
-
-
—
—
-
-
-
-
(Bethlehem)"
ND(O.l)
1.2
ND(O.l)
0.6
0.4
0.3
0.1
ND(O.l)
0.7
0.1
0.4
0.1
ND(O.l)
0.1
0.1
ND(O.l)
0.4
2.6
1.9
0.85
1.449
—
-
-
-
-
—
—
-
-
-
-
Chimprom
(Russia)
ND (0.02)
0.03
0.02
0.05
ND (0.02)
0.23
0.85
ND(O.l)
1.2
0.06
0.08
0.11
ND (0.02)
0.05
0.24
0.02
0.46
1.18
2.22
0.142
0.156
—
-
-
-
-
-
-
-
-
-
-
Dragon
Lawn Weed
Killer
ND (0.001)
0.0014
ND (0.001)
0.0024
0.001
0.0017
0.0063
0.0036
0.001
0.0011
0.0013
ND (0.001)
ND (0.001)
0.0011
0.0016
ND (0.001)
0.0039
0.0128
0.0136
0.0023
0.003
—
-
-
-
-
-
-
-
-
-
-
KGRO
(U.S.)
—
-
-
-
-
-
-
—
-
-
-
-
-
-
-
-
-
0.0144
0.1628
0.0009
—
-
-
-
-
-
-
-
-
-
-
Pro Care
Premium
(U.S.)
—
-
-
-
-
-
-
—
-
-
-
-
-
-
-
-
-
0.0143
0.4253
0.0012
—
-
-
-
-
-
-
-
-
-
-
Ortho
Weed-B-
Gone
(U.S.)
—
-
-
-
-
-
-
—
-
-
-
-
-
-
-
-
-
0.0091
0.1095
0.0014
—
-
-
-
-
-
-
-
-
-
-
Sigma
Co.
(U.S.)
—
-
-
-
-
-
-
—
-
-
-
-
-
-
-
-
-
0.127
3.0507
0.0013
—
-
-
-
-
-
-
-
-
-
-
American
Brand
Chemical
Co. (U.S.)
—
-
-
-
-
-
-
—
-
-
-
-
-
-
-
-
-
0.0278
0.0822
0.0019
—
-
-
-
-
-
-
-
-
-
-
Ishihara
Sangyo
Kaisha, Ltd.
(Japan)
0.0021
0.011
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND(O.Ol)
ND (0.002)
0.0038
ND (0.002)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.010)
0.013
0.004
0.0078
0.0130.041
0.018
0.008
ND (0.005)
ND(O.Ol)
2.7
0.89
0.019
0.006
ND(O.Ol)
3.7
Nissan
Chemical
Industries,
Ltd.
(Japan)
ND (0.002)
ND (0.002)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND(O.Ol)
ND (0.002)
ND (0.002)
ND (0.002)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND (0.005)
ND(O.Ol)
-
-
ND
ND
ND (0.002)
ND (0.002)
ND (0.005)
ND (0.005)
ND(O.Ol)
0.0093
ND (0.002)
ND (0.005)
ND (0.005)
ND(O.Ol)
0.0093
oo
-------
Table 8-33. CDD/CDF concentrations (ug/kg) in samples of 2,4-D and pesticide formulations containing 2,4-D
(continued)
a2,4-D manufactured in Europe and packaged in Palestine.
bCalculated assuming nondetect values are zero.
ND = not detected (value in parenthesis is the detection limit)
— = No information given
Sources: Schecter et al. (1997); Masunaga et al. (2001).
oo
-------
sewers from lands or urban surfaces contaminated by product uses or deposition of previous
emissions to air (discussed in Chapter 11, Section 11.2.1), household wastewater, industrial
wastewater, and chlorination operations within the wastewater treatment facility (Rappe, 1992b;
Rappe et al., 1994; Horstmann et al., 1992; Sewart et al., 1995; Cramer et al., 1995; Horstmann
and McLachlan, 1995).
The major source(s) for a given publicly owned treatment works (POTW) is likely to be
site specific, particularly in industrialized areas. For example, Rieger and Ballschmiter (1992)
traced the origin of CDDs/CDFs found in municipal sewage sludge in Ulm, Germany, to metal
manufacturing and urban sources. The characteristics of both sources were similar and suggested
generation via thermal processing. However, a series of studies by Horstmann et al. (1992,
1993a, b) and Horstmann and McLachlan (1994a, b, 1995) demonstrated that wastewater
generated by laundering and bathing could be the major source at many, if not all, POTWs that
serve primarily residential populations. Although runoff from streets during precipitation events,
particularly from streets with high traffic density, was reported by these researchers as
contributing measurably, the total contribution of TEQ from household wastewater was eight
times greater than that from surface runoff at the study city.
Horstmann et al. (1992) provided initial evidence that household wastewater could be a
significant source. Horstmann et al. (1993a) measured CDD/CDF levels in the effluent from four
different loads of laundry from two different domestic washing machines. The concentrations of
total CDDs/CDFs in the four samples ranged from 3,900 to 7,100 pg/L and were very similar in
congener profile, with OCDD being the dominant congener, followed by the hepta- and hexa-
CDDs. Because of the similar concentrations and congener profiles found, the authors concluded
that the presence of CDDs/CDFs in washing machine wastewater is widespread. A simple mass
balance performed using the results (Horstmann and McLachlan, 1994a) showed that the
CDDs/CDFs found in the four washing machine wastewater samples could account for 27 to
94% of the total CDDs/CDFs measured in the sludge of the local wastewater treatment plant.
Horstmann et al. (1993a) performed additional experiments that showed that detergents,
commonly used bleaching agents, and the washing cycle process itself were not responsible for
the observed CDDs/CDFs. To determine whether the textile fabric or fabric finishing processes
could account for the observed CDDs/CDFs, Horstmann et al. (1993b), Horstmann and
McLachlan (1994a, b), and Klasmeier and McLachlan (1995) analyzed the CDD/CDF content of
raw cotton cloth, white synthetic materials, and more than 100 new textile finished products.
Low concentrations were found in most products (less than 50 ng/kg of total CDDs/CDFs), but a
small percentage contained high concentrations, up to 290 |ig/kg of total CDDs/CDFs. On the
basis of the concentrations and patterns found, the authors concluded that neither unfinished new
fabrics nor common cotton finishing processes could explain the CDD/CDF levels found in
8-76
-------
wastewater; rather, the use of CDDVCDF-containing textile dyes and pigments and the use in
some developing countries of PCP to treat unfinished cotton appeared to be the sources of the
detected CDDs/CDFs.
Horstmann and McLachlan (1994a, b, 1995) reported the results of additional
experiments showing that the small percentage of clothing items with high CDD/CDF levels
could be responsible for the quantity of CDDs/CDFs observed in household wastewater and
sewage sludge. They demonstrated that the CDDs/CDFs can be gradually removed from the
fabric during washing; they can be transferred to the skin, subsequently transferred back to other
textiles, and then washed out, or they can be transferred to other textiles during washing and then
removed during subsequent washing.
8.4.1.2. Releases to Water
8.4.1.2.1. Emissions data. The presence of CDDs/CDFs in sewage sludge suggests that
CDDs/CDFs may also be present in the wastewater effluent discharges of POTWs; however, few
studies reporting the results of effluent analyses for CDDs/CDFs have been published.
Rappe et al. (1989a) tested the effluent from two Swedish POTWs for all 2,3,7,8-
substituted CDD/CDF congeners. OCDD was detected in the effluents from both facilities at
concentrations ranging from 14 to 39 pg/L. The investigators detected 1,2,3,4,6,7,8-HpCDD and
1,2,3,4,6,7,8-HpCDF in the effluent of one facility at concentrations of 2.8 and 2 pg/L,
respectively. No 2,3,7,8-substituted tetra-, penta-, and hexa-CDDs or CDFs were detected (DLs
of 0.2 to 20 pg/L).
Ho and Clement (1990) reported the results of sampling during the late 1980s of 37
POTWs in Ontario, Canada, for each of the five CDD/CDF congener groups with four to eight
chlorines. The sampled facilities included 27 secondary treatment facilities, seven primary
treatment facilities, one tertiary plant, and two lagoons. The facilities accounted for about 73%
of the sewage discharged by POTWs in Ontario. No CDDs/CDFs were detected (DL in the low
ng/L range) in the effluents from the lagoons and the tertiary treatment facility. Only OCDD and
TCDF were detected in the effluents from the primary treatment facilities (two and one effluent
samples, respectively). HpCDD, OCDD, TCDF, and OCDF were detected in the effluents from
the secondary treatment facilities (detected in four or fewer samples at levels ranging from 0.1 to
11 ng/L).
Gobran et al. (1995) analyzed the raw sewage and final effluent of an Ontario, Canada,
wastewater treatment plant for CDD/CDF congeners over a 5-day period. Although HpCDD,
OCDD, HpCDF, and OCDF were detected in the raw sewage (12 to 2,300 pg/L), no CDDs/CDFs
were detected in the final effluent at congener-specific DLs ranging from 3 to 20 pg/L.
8-77
-------
The California Regional Water Quality Control Board reported the results of effluent
testing at nine POTWs in the San Francisco area (memorandum dated March 21, 1996, from Lila
Tang, California Regional Water Quality Control Board, to David Cleverly, U.S. EPA). A total
of 30 samples were collected between 1992 and 1995, and 1 to 6 samples were analyzed for each
POTW. The mean concentrations of CDDs/CDFs are shown in Table 8-34. The overall mean
TEQ concentration was 0.27 pg TEQDF-WHO98/L (0.29 pg I-TEQDF/L). With the exception of
OCDD, most 2,3,7,8-substituted CDD/CDF congeners were seldom detected.
Rappe et al. (1998) analyzed effluent samples from 17 POTWs in Mississippi, 10 of
which receive input from industrial facilities. Treatment processes at the facilities include the
use of one or more of the following: lagoons, activated sludge, aerated digestion, wetlands,
oxidative ditch, and trickling filter. Additionally, 12 of the facilities use chlorine gas in the
treatment process. The wastewater flows at the facilities range from 0.11 to 39.75 million L/day;
however, wastewater flow rates were not known for two facilities. Table 8-35 presents the
concentrations of dioxins measured in the effluent samples for each facility and total TEQ
emission factors. Concentrations were congener-specific for only 2,3,7,8-TCDD; 2,3,7,8-TCDF;
1,2,3,7,8-PeCDD; 2,3,4,7,8-PeCDF; OCDD; and OCDF. Also provided were concentrations for
total HxCDD and total HpCDD. The total I-TEQ concentrations reported by Rappe et al.
(assuming nondetect values equal to one-half the DL) ranged from 0.274 to 3.84 pg I-TEQDF/L
(average of 0.81 pg/I-TEQDF/L). Because concentrations for all congeners were not provided,
emission factors could not be calculated in TEQDF-WHO98.
The CRWQCB data (memorandum dated March 21, 1996, from Lila Tang, California
Regional Water Quality Control Board, to David Cleverly, U.S. EPA) were collected to provide
representative effluent concentrations for the San Francisco area. These data cannot be
considered to be representative of CDD/CDF effluent concentrations at the 16,000-plus POTWs
nationwide. Therefore, the data can be used only to generate a preliminary estimate of the
potential mass of CDD/CDF TEQ that may be released annually by U.S. POTWs.
8.4.1.2.2. Activity level information. Based on the results of the 1996 and 2000 Clean Water
Needs Surveys, an estimated 122 billion L and 148 billion L of wastewater were treated daily by
POTWs in the United States in 1996 and 2000, respectively (U.S. EPA, 1997c, 2004).
Wastewater treatment data were not available for 1987; however, an estimate was
developed using the population of the United States as a surrogate. In 2000, the population of the
United States was approximately 281 million people. Using the estimate of water treated daily
by POTWs in 2000, approximately 527 L/person of wastewater were treated daily by POTWs. In
1990, the population of the United States was approximately 249 million people. Assuming the
8-78
-------
Table 8-34. Mean CDD/CDF concentrations and range of detection limits
(DLs) (pg/L) in effluents from nine U.S. publicly owned treatment works
(POTWs)
Congener/congener
group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDD
Total 2,3,7,8-CDF
Total I-TEQDF
Total TEQDF-WHO98
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
No. of
detections/
samples
0/30
0/30
0/30
0/30
0/30
3/30
13/30
1/27
1/30
1/30
1/30
1/30
1/30
1/30
2/30
0/30
1/30
—
4/27
0/27
1/30
3/30
13/30
2/30
1/30
1/30
2/30
1/30
Range of
DLs
0.31-8.8
0.45-15
0.43-9.8
0.81-10
0.42-9.7
0.75-18
6.2-57
0.74-4.4
0.64-9.4
0.61-14
0.25-6.8
0.23-6.8
0.57-10
0.25-7.9
0.36-6.9
0.19-11
0.86-28
—
1.2-8.8
0.62-200
0.84-11
0.75-18
6.2-57
0.39-6.8
0.64-25
0.93-17
0.36-19
0.86-28
Total CDD/CDF
Range of detected
concentrations
(POTW mean basis)
Minimum
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
ND
ND
ND
ND
ND
ND
Maximum
ND
ND
ND
ND
ND
5
99.75
1.3
2
2.8
2.4
1.5
2
ND
4.6
ND
3.2
104.75
19.8
2.42
2.33
9.7
ND
1.7
8.4
99.75
25
20
13
4.6
3.2
185.35
Overall mean
concentrations3
Nondetect
set to 0
0
0
0
0
0
1.06
29.51
0.14
0.22
0.31
0.27
0.17
0.22
0
0.68
0
0.36
30.57
2.37
0.29
0.27
1.23
0
0.19
1.83
29.51
6.61
2.22
1.44
0.68
0.36
44.07
Nondetect
set to l/2 DL
0.98
1.32
1.38
1.42
1.31
3.61
37.95
0.98
1.58
1.68
1.22
0.97
1.72
0.93
1.83
1.18
3.4
47.97
15.49
3.66
4.28
2.61
6.27
1.93
4.77
37.95
7.7
4.72
3.43
2.41
3.4
75.19
aThe overall means are the means of the individual POTW mean concentrations rather than the means of the
individual sample concentrations.
DL = Detection limit
ND = Not detected
~ = No information given
Source: Memorandum dated March 21, 1996, from Lila Tang, California Regional Water Quality Control Board, to
David Cleverly, U.S. EPA.
8-79
-------
Table 8-35. Effluent concentrations (pg/L) of CDDs/CDFs from publicly owned treatment works in Mississippi
Facility
Waynesboro
Meridian
Pascagoula
W. Biloxi
Gulrport
Laurel
Brookhaven
Natchez
Picayune
Picayune"
Waveland
Corinth
New Augusta
Beaumont
Leaksville
McLain
Hattiesburg South
Hattiesburg North
AVERAGE
2,3,7,8-
TCDD
ND(0.17)
0.18
ND(0.13)
0.18
0.16
ND(0.18)
ND(0.18)
ND(0.16)
ND (0.22)
ND(0.13)
ND(0.18)
ND(0.15)
ND(O.l)
ND(O.l)
ND(0.12)
ND (0.06)
ND(0.16)
ND(0.19)
0.17
2,3,7,8-
TCDF
0.18
0.12
0.15
0.24
0.24
0.15
0.54
0.41
0.56
0.54
17
0.17
1.3
0.14
0.72
ND (0.05)
ND (0.24)
0.18
1.42
1,2,3,7,8-
PeCDD
ND (0.2)
ND(0.16)
ND(0.15)
ND(0.15)
ND(0.15)
ND (0.23)
0.45
0.6
ND (0.27)
ND(0.12)
0.22
ND(0.16)
0.28
ND(0.13)
0.25
ND(O.IO)
ND (0.24)
ND (0.26)
0.36
2,3,4,7,8-
PeCDF
ND(O.l)
ND (0.09)
0.11
0.082
0.094
ND(0.12)
0.16
0.34
ND(0.14)
ND (0.07)
0.66
ND (0.09)
0.085
0.088
0.15
ND (0.06)
ND(O.ll)
ND(0.13)
0.2
Total
HxCDD
ND
1.3
ND
ND
ND
ND
0.85
2.5
6.5
6
ND
0.77
21
0.64
8.9
2.5
1.2
0.96
4.43
Total
HpCDD
3.5
7.6
0.82
0.9
2.3
2.9
3.2
2.4
38
30
3
2.7
120
2.4
46
14
4.5
9.1
16.3
OCDD
13
58
3.6
4
9.9
38
28
9.1
120
53
14
18
2,500
11
780
200
59
73
221.76
OCDF
1.8
1.8
0.46
ND (0.34)
0.78
ND (0.48)
1.7
1.8
2
106
0.9
1.1
0.66
3.2
0.77
2.9
8.99
Total
I-TEQ
0.316
0.445
0.264
0.378
0.371
0.334
0.796
1.03
0.715
0.397
2.4
0.276
3.84
0.274
1.6
0.377
0.32
0.457
0.81
oo
oo
o
"Blind double.
ND = Not detected (value in parenthesis is the detection limit). When calculating the average, NDs were not included.
Source: Rappe et al. (1998).
-------
population did not change drastically between 1987 and 1990, and assuming that the daily
domestic wastewater treatment per person remained constant between 1987 and 2000, EPA
estimates that approximately 131 billion L of wastewater were treated daily at POTWs in 1987.
8.4.1.2.3. Emission Estimates. Multiplying the amount of wastewater treated by 365 days/yrby
the overall mean TEQ concentrations reported by CRWQCB (memorandum dated March 21,
1996, from Lila Tang, California Regional Water Quality Control Board, to David Cleverly, U.S.
EPA) (0.27 pg TEQDF-WHO98/L [0.29 pg I-TEQDF/L]) yields annual TEQ release estimates of
12.9 g TEQDF-WH098 (13.9 g I-TEQDF), 12 g TEQDF-WHO98 (13 g I-TEQDF), and 14.6 g TEQDF-
WHO98 (15.7 g I-TEQDF) for 1987, 1995, and 2000, respectively. These estimates should be
regarded as preliminary indications of possible emissions from this source.
8.4.1.3. Sewage Sludge Land Disposal
Sewage sludge is the solid, semi-solid, or liquid residue generated during the treatment of
wastewater. During wastewater treatment, nutrients, pathogens, inorganic compounds (metals
and trace elements), and organic compounds (CDDs/CDFs, PCBs, and surfactants) from the
incoming wastewater are partitioned to the resulting sewage sludge (NRC, 2002). The sludge is
either disposed of through methods such as incineration or landfill/surface disposal or
beneficially used through methods such as land application.
Sewage sludge that is applied to land is referred to as biosolids. In order to be applied to
land, the biosolids must be treated to meet land application regulatory requirements (Federal
Register, 1993a). Biosolids are often used for crop production, gardening, forestry, turf growth,
and landscaping. Some other uses include strip mine and gravel pit reclamation and wetland
restoration. Land application of biosolids is beneficial because it improves the physical and
chemical properties of the soil needed for plant growth, it reduces the need for other disposal
methods, and it reduces or eliminates the need for commercial fertilizers. Commercial fertilizers
often have higher nutrient contents than do biosolids; therefore, the application of biosolids to
land in lieu of commercial fertilizers may reduce the impacts of high levels of excess nutrients
entering the environment (U.S. EPA, 1999b).
8.4.1.3.1. Emissions data. EPA conducted the National Sewage Sludge Survey in 1988 and
1989 to obtain national data on sewage sludge quality and management. As part of this survey,
EPA analyzed sludges from 174 POTWs that employed at least secondary wastewater treatment
for more than 400 analytes, including CDDs/CDFs. Although sludges from only 16% of the
POTWs had detectable levels of 2,3,7,8-TCDD, all sludges had detectable levels of at least one
CDD/CDF congener (U.S. EPA, 1996e). I-TEQDF concentrations as high as 1,820 ng/kg dry
8-81
-------
weight were measured. The congener-specific results of the survey are presented in Table 8-36.
If all nondetect values found in the study are assumed to be zero, then the mean and median I-
TEQDF concentrations of the sludges from the 174 POTWs are 50 and 11.2 ng/kg (dry-weight
basis), respectively. If the nondetect values are set equal to the DL, then the mean and median
I-TEQDF concentrations are 86 and 50.4 ng/kg, respectively (U.S. EPA, 1996e; Rubin and White,
1992).
Table 8-36. CDD/CDF concentrations (ng/kg) measured in the 1998/1999
National Sewage Sludge Surveya
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Percent
detected
16
18
25
49
39
98
100
65
22
26
43
35
16
27
71
26
80
Total I-TEQDF
Total 2,3,7,8-CDD/CDF
Maximum
concentration
detected
116
736
737
737
737
52,500
905,000
337
736
736
1,500
737
1,260
737
7,100
842
69,500
1,820
-
Median concentration
Nondetect
set to
detection
limit
6.86
9.84
22.5
27.3
28
335
3,320
17
9.6
10.4
28
18
18
18
57
23
110
50.4
-
Nondetect
set to
zero
0
0
0
0
0
335
3,320
3.9
0
0
0
0
0
0
36
0
80
11.2
-
Mean concentration
Nondetect
set to
detection
limit
~
—
~
—
~
-
~
..
~
—
~
—
~
—
~
-
~
86b
-
Nondetect
set to
zero
~
—
~
—
~
-
~
..
~
—
~
—
~
—
~
-
~
50b
-
Tor publicly owned treatment works with multiple samples, the pollutant concentrations were averaged before the
summary statistics presented in the table were calculated.
Values presented by Rubin and White (1992) for 175 rather than 174 publicly owned treatment works.
~ = No information given
Source: U.S. EPA(1996e).
8-82
-------
Green et al. (1995) and Cramer et al. (1995) reported the results of analyses of 99 samples
of sewage sludge collected from wastewater treatment plants across the United States during the
summer of 1994 as part of the 1994/1995 Association of Metropolitan Sewerage Agencies
(AMSA) survey. These data are summarized in Table 8-37. To calculate average results in units
of TEQ, Green et al. averaged the results from all samples collected from the same facility to
ensure that the results were not biased toward the concentrations found at facilities from which
more than one sample was collected. Also, eight samples were excluded from the calculation of
the overall TEQ averages because it was unclear whether they were duplicate samples from other
POTWs. POTW average TEQ concentrations were calculated for 74 POTWs. If all nondetect
values are assumed to be zero, then the overall study mean and median I-TEQDF concentrations
were 47.7 and 33.4 ng I-TEQDF/kg (dry weight basis), respectively (standard deviation of 44.7 ng
I-TEQDF/kg). The corresponding mean and median TEQDF-WHO98 concentrations were 36.3 and
25.5 ng/kg, respectively (standard deviation, 38.6).
The mean and median results reported by Green et al. (1995) and Cramer et al. (1995) are
very similar in terms of total TEQ to those reported by EPA for samples collected 5 years earlier
(U.S. EPA, 1996e; Rubin and White, 1992). The predominant congeners in both data sets are the
octa- and hepta-CDDs and CDFs. Although not present at high concentrations, 2,3,7,8-TCDF
was commonly detected.
In addition to effluents, Rappe et al. (1998) analyzed the levels of CDDs/CDFs in
municipal sewage sludge from the 17 POTWs in Mississippi. Table 8-38 presents the
concentrations of dioxins measured in the sewage sludge samples and total TEQ emission factors
reported by Rappe et al. Concentrations were only congener specific for 2,3,7,8-TCDD; 2,3,7,8-
TCDF; 1,2,3,7,8-PeCDD; 2,3,4,7,8-PeCDF; OCDD; and OCDF. Also provided were
concentrations for total HxCDD and total HpCDD. The TEQ emission factors (assuming
nondetects equal to one-half the DL) ranged from 2.26 to 1,270 ng I-TEQDF/kg. The
predominant congeners in all samples were the octa- and hepta-CDDs. The sludge with the
highest concentrations of octa- and hepta-CDDs was from the Picayune POTW, which receives
industrial inputs, including effluents from wood treatment facilities that likely contain PCP. In
general, the sludges with the lowest TEQ values were from the facilities that do not receive
effluent from industrial facilities. Additionally, the samples with the two lowest TEQ values
were from facilities that do not use free chlorine as a disinfectant.
In 1999, sewage sludge samples from a POTW in Ohio were collected and analyzed for
CDDs/CDFs (U.S. EPA, 2000b). The facility, which accepts both domestic and industrial
wastewater, employs secondary wastewater technology. Assuming nondetects are zero, the mean
TEQ emission factor was 21.9 ng TEQDF-WHO98/kg (dry-weight basis). These results are
presented in Table 8-39.
8-83
-------
Table 8-37. CDD/CDF concentrations (ng/kg) measured in 99 sludges collected from U.S. publicly owned
treatment works (POTWs) during 1994
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Percent
detected
40
23
34
87
64
98
99
76
21
42
48
17
4
35
64
31
93
Average I-TEQDF (facility basis)b
Total 2,3,7,8-CDD/CDF
Maximum
concentration
detected
12.3
37.5
45.6
130
88.8
5,380
65,500
156
60.3
155
170
200
115
356
1,460
213
11,200
246
85,279.5
Average TEQDF-WHO98 (facility basis)b
Median concentration
Nondetect
set to
detection
limit
1.95
8.23
5.25
25.6
12.3
642
6,630
7.53
7.91
9.7
11.5
14
7.53
9.85
91.7
11.7
286
49.6
7,782.8
44.6
Nondetect
set to
zero
0
0
0
24.7
9.48
642
6,630
6.28
0
0
0
0
0
0
31.8
0
281
33.4
7,625.3
25.5
Mean concentration
Nondetect
set to
detection limit3
2.72 (2.4)
10.9 (7.8)
11.1 (8.13)
33.8 (27.6)
20.2 (17.7)
981 (977)
11,890 (12,540)
12.8 (19.6)
10.7 (11.3)
15.7 (19.8)
20.4 (25.3)
30.4 (53.6)
11.1 (13.6)
21.8 (40.4)
223 (271)
27.1 (34.8)
786 (1,503)
64.5 (50.1)
14,108.7 (15,573)
57.2 (44.4)
Nondetect
set to
zero3
1.71 (2.86)
3.34 (7.43)
6.03 (10.2)
32.2 (28.8)
17 (19.8)
981 (977)
11,890 (12,540)
11.1 (20.2)
3.53 (9.36)
10.5 (21.6)
14 (25.9)
5.13 (21.9)
1.56 (11.7)
13.6 (41)
97.5 (207)
15 (33.4)
775 (1,506)
47.7 (44.7)
13,878.2 (15,484.2)
36.3 (38.6)
oo
oo
aValue in parenthesis is the standard deviation.
bFor POTWs with multiple samples, the sample TEQ concentrations were averaged to POTW averages before calculation of the total TEQ mean and median
values presented in the table. A total of 74 POTW average concentrations were used in the calculations. In addition, the following sample ID numbers were
not included in the averaging because, according to Green et al. (1995), it was not possible to determine whether they were duplicate or multiple samples from
other POTWs: 87, 88, 89, 90, 91, 97, 98, and 106.
Sources: Green et al. (1995); Cramer et al. (1995).
-------
Table 8-38. Concentrations of CDDs/CDFs (ng/kg dry matter) in sewage sludge from publicly owned treatment
works in Mississippi
Facility
Waynesboro
Meridian
Pascagoula
W. Biloxi
Gulrport
Laurel
Brookhaven
Natchez
Picayune
Picayune"
Waveland
Corinth
New Augusta
Beaumont
Leaksville
McLain
Hattiesburg S
Hattiesburg N
Average
2,3,7,8-
TCDD
2.1
ND (0.06)
2
0.84
1.9
0.17
2
ND (0.58)
5.3
4.1
1.6
0.3
ND(0.13)
0.17
ND (0.051)
0.076
1
ND (0.035)
1.7
2,3,7,8-
TCDF
2.9
2.1
3.6
2.4
9.1
0.3
2.5
8.3
69
66
2.6
1.8
0.17
0.67
0.14
0.17
1.1
1.7
9.7
1,2,3,7,8-
PeCDD
3.5
6.4
5.3
3.2
9.5
0.37
11
8.4
74
60
5.1
0.97
0.15
0.78
0.32
0.11
9.1
4
11.2
2,3,4,7,8-
PeCDF
1.4
2.8
3.5
1.3
3.4
0.25
2.5
ND(1.5)
24
17
1.9
0.93
0.094
0.37
0.11
0.031
2.2
2
3.8
Total
HxCDD
85
10
170
78
200
22
130
270
17,000
16,000
130
42
21
59
16
39
170
310
1,930.7
Total
HpCDD
920
100
970
280
1,100
160
1,400
1,100
250,000
210,000
580
230
140
470
92
140
1,3000
3,600
26,904.6
OCDD
7,400
7,400
4,300
1,800
7,700
2,700
9,300
6,800
480,000
420,000
3,500
3,300
1,400
1,900
560
2,600
4,400
27,000
55,114.4
OCDF
410
410
170
70
310
91
230
270
16,000
17,000
150
36
8.8
42
26
0.74
180
980
2,017.5
Total
I-TEQ
23.7
27.6
26.4
13.7
30.9
4.83
36.7
37.7
1,270
1,240
31.7
7.4
2.67
6.18
2.26
3.55
33
70.4
159.4 ±399
oo
oo
"Blind double.
ND = Not detected (value in parenthesis is the detection limit). When calculating the average, NDs were not included.
Source: Rappe et al. (1998).
-------
Table 8-39. CDD/CDF median concentrations (ng/kg) measured in 1999
from a publicly owned treatment works facility in Ohio
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
Average total TEQDF-WHO98
Nondetect set to zero
ND
ND
2.67
21.33
30.33
298
2,963
26.67
4.33
10
21
5.33
ND
9
171
ND
364.67
21.87
Nondetect set to
1A detection limit
0.0018
0.0082
2.67
21.33
30.33
298
2,963
26.67
4.34
10
21
5.33
0.0033
9
171
0.01
364.67
21.88
ND = Not detected
Source: U.S. EPA(2000b).
In 2000 and 2001, AMSA conducted another survey of dioxin-like compounds in sewage
sludge (Alvarado et al., 2001). A total of 200 sewage sludge samples were collected from 171
POTWs located in 31 states. Assuming nondetects are zero, TEQ emission factors range from
0.08 to 3,578.61 ng TEQDF-WHO98/kg. The mean and median TEQ emission factors were 34.5
and 11.79 ng TEQDF-WHO98/kg, respectively.
EPA conducted another National Sewage Sludge Survey to characterize the dioxin and
dioxin-like equivalence levels in biosolids produced by 6,857 POTWs operating in the United
States in 2001 (U.S. EPA, 2002d). Samples were collected from 94 POTWs using secondary or
higher treatment practices. All facilities had been sampled previously as part of the 1988/1989
National Sewage Sludge Survey. The overall mean and median TEQDF-WHO98 concentrations
were 75 and 15 ng/kg, respectively. However, when the data were weighted using the daily
influent wastewater flow rates (i.e., the number of facilities with wastewater flow rate
8-86
-------
>100 Mg/day, >10 but < 100 Mg/day, >1 but < 10 Mg/day, and < 1 Mg/day), the overall mean and
median TEQDF-WHO98 concentrations were 21.7 and 15.5 ng/kg, respectively. These data are
summarized in Table 8-40.
Table 8-40. CDD/CDF mean concentrations (ng/kg) measured in the 2001
National Sewage Sludge Survey
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Average total TEQDF-WHO98
Nondetect set to zero
1.41
5.76
11.8
21.3
3.6
492
6,780
3.11
2.61
6.03
1.37
0.27
5.21
5.5
9.13
167
802
21.7
Nondetect set to
1A detection limit
1.1
4.57
7.49
15.1
2.22
273
2,730
2.3
1.5
2.8
1
0
2.6
3.36
2.8
88.2
279
15.5
Source: U.S. EPA(2002d).
The CDD/CDF concentrations and congener group patterns observed in the U.S. surveys
are similar to those reported for sewage sludges in several other Western countries. Stuart et al.
(1993) reported mean CDD/CDF concentrations of 23.3 ng I-TEQDF/kg (dry weight) for three
sludges from rural areas, 42.3 ng I-TEQDF/kg for six sludges from light industry/domestic areas,
and 52.8 ng I-TEQDF/kg for six sludges from industrial/domestic areas collected during
1991-1992 in England and Wales. Naf et al. (1990) reported concentrations ranging from 31 to
40 ng I-TEQDF/kg (dry weight) in primary and digested sludges collected from the POTW in
Stockholm, Sweden, during 1989. Gobran et al. (1995) reported an average concentration of
15.7 ng I-TEQDF/kg in anaerobically digested sludges from an industrial/domestic POTW in
Ontario, Canada. In all three studies, the congener group concentrations increased with
8-87
-------
0 50%
1
•6 40%
<& <$t <&<&<&<&<*{ <& <& <$t
-y A) ,-y f-y AJ AJ .-y ry ry ,-y
-------
sludge were generated in 1989 (Federal Register, 1993a). EPA also used the results of the 1984
to 1996 Clean Water Needs Surveys to estimate that 6.3 million dry metric tons of sewage sludge
were generated in 1998. Because estimates for reference years 1987 and 1995 are not available,
the 1989 and 1998 activity level estimates were used for 1987 and 1995, respectively. Tables
8-41 and 8-42 list the volumes, by use and disposal practices, of sludge disposed of annually for
reference years 1987 and 1995, respectively.
U.S. EPA (1999b) estimated that 6.6 million dry metric tons of sewage sludge would be
generated in 2000. Table 8-43 lists the volumes, by use and disposal practices, of sludge
disposed of annually for reference year 2000. Similarly, the National Research Council (NRC)
analyzed the amount of biosolids being applied to land in 2002 (NRC, 2002). Citing 2001 data
(unpublished) from the Wisconsin Department of National Resources, the NRC estimated that
approximately 8,650 of the 16,000 POTWs operating in the United States generated sewage
sludge requiring use or disposal. Using data from 37 states, the investigators estimated that
5,900 of these sewage sludge generators either land applied or publicly distributed more than 3.4
million dry tons of biosolids annually. The volume of biosolids, by use and disposal practices, is
presented in Table 8-44. The volume of biosolids and their distribution among the various
categories estimated by the NRC are very close to those estimated by EPA.
8.4.1.3.3. Emission estimates. Annual potential releases of CDDs/CDFs were determined by
multiplying the mean total TEQ concentrations by the sludge volumes generated. The mean total
TEQ concentrations in sewage sludges for reference years 1987, 1995, and 2000 are reported in
Tables 8-41, 8-42, and 8-43, respectively. For 1987, the total annual potential release from
nonincinerated sludges was 151 g TEQDF-WHO98. Of this amount, 2.6 g TEQDF-WHO98 (3.5 g I-
TEQDF) entered commerce as a product for distribution and marketing and 76.6 g TEQDF-WHO98
(103 g I-TEQDF) was applied to land. The remaining 71.8 g TEQDF-WHO98 did not result in an
environmental release because it was sent to RCRA Subtitle D landfills or disposal sites. For
1995, the total annual potential release from nonincinerated sludges was 178 g TEQDF-WHO98.
Of this amount, 3 g TEQDF-WHO98 (4 g I-TEQDF) entered commerce as a product for distribution
and marketing and 116.1 g TEQDF-WHO98(156.5 g I-TEQDF) was applied to land. The remaining
58.9 g TEQDF-WHO98 did not result in an environmental release because it was sent to RCRA
Subtitle D landfills or disposal sites. For 2000, the total annual release of nonincinerated sludges
was 111 g TEQDF-WHO98. Of this amount, 1.9 g TEQDF-WHO98 (1.9 g I-TEQDF) entered
commerce as a product for distribution and marketing and 78.2 g TEQDF-WHO98 (78.2 g I-
TEQDF) was applied to land. The remaining 30.9 g TEQDF-WHO98 did not result in an
environmental release because it was sent to RCRA Subtitle D landfills or disposal sites.
8-89
-------
Table 8-41. Quantity of sewage sludge disposed of and potential dioxin TEQ
releases from primary, secondary, and advanced treatment publicly owned
treatment works for reference year 1987
Use/disposal practice
Land application
Distribution and marketing
Surface disposal site/other
Sewage sludge landfill
Co-disposal landfills0
Sludge incinerators and
co-incinerators'1
Ocean disposal
TOTAL
Volume
disposed of
(1,000 dry
metric
tons/yr)
1,714
71
396
157
1,819
865
(336)f
5,358
Percent of
total volume
32b
1.3
7.4
2.9
33.9
16.1
(6.3)f
100
Potential dioxin releasea
(gTEQ/yr)
I-TEQDF
84
3.5
19.4
7.7
89.1
e
f
203.7
TEQDF-
WHO98
62.2
2.6
14.4
5.7
66
e
f
150.9
Totential dioxin TEQ release for nonincinerated sludges was estimated by multiplying the sludge volume generated
(second column) by the average of the mean I-TEQDF concentrations in sludge reported by Rubin and White (1992)
(i.e., 50 ng/kg dry weight) and Green et al. (1995) and Cramer et al. (1995) (47.7 ng/kg). The calculations of
TEQDF-WHO98 used the mean concentration of 36.3 ng TEQDF-WHO98/kg for the results reported by Green et al.
(1995) and Cramer et al. (1995).
Includes 21.9% applied to agricultural land, 2.8% applied as compost, 0.6% applied to forestry land, 3.1% applied
to "public contact" land, 1.2% applied to reclamation sites, and 2.4% applied in undefined settings.
°Landfills used for disposal of sewage sludge and solid waste residuals.
dCo-incinerators treat sewage sludge in combination with other combustible waste materials.
eSee Section 3.5 for estimates of CDD/CDF releases to air from sewage sludge incinerators.
fThe Ocean Dumping Ban Act of 1988 generally prohibited the dumping of sewage sludge into the ocean after
December 31, 1991. Ocean dumping of sewage sludge ended in June 1992 (Federal Register, 1993a). The current
method of disposal of the 336,000 metric tons of sewage sludge that were disposed of in the oceans in 1988 has not
been determined.
These release estimates are assigned a high confidence rating for both the production and
the emission factor estimates. The high rating is based on the judgment that the 174 facilities
tested as part of the 1988/1989 National Sewage Sludge Survey (U.S. EPA, 1996e), the 74
facilities tested as part of the 1994/1995 AMSA Survey (Green et al., 1995 and Cramer et al.,
1995), and the 94 facilities tested as part of the 2001 National Sewage Sludge Survey (U.S. EPA,
2002d) were reasonably representative of the variability in POTW technologies and sewage
characteristics nationwide.
8-90
-------
Table 8-42. Quantity of sewage sludge disposed of and potential dioxin TEQ
releases from primary, secondary, and advanced treatment publicly owned
treatment works for reference year 1995
Use/disposal practice
Land application13
Advanced treatment0
Other beneficial used
Surface disposal/landfill
Incineration
Other disposal method
TOTAL
Volume disposed
of (1,000 dry
metric tons/yr)
2,500
700
500
1,100
1,400
100
6,300
Percent of
total volume
40
11
8
17
22
2
100
Potential dioxin release"
(g TEQ/yr)
I-TEQDF
122.3
34.2
24.5
53.8
e
4.9
239.7
TEQDF-
WHO98
90.7
25.4
18.2
39.9
e
3.6
177.8
Totential dioxin TEQ release for nonincinerated sludges was estimated by multiplying the sludge volume generated
(second column) by the average of the mean I-TEQDF concentrations in sludge reported by Rubin and White (1992)
(50 ng/kg dry weight) and Green et al. (1995) and Cramer et al. (1995) (47.7 ng/kg). The calculations of TEQDF-
WHO98 used the mean concentration of 36.3 ng TEQDF-WHO98/kg for the results reported by Green et al. (1995)
and Cramer et al. (1995).
bWithout further processing or stabilization, such as composting.
°Such as composting.
dEPA assumes that this category includes distribution and marketing (i.e., sale or give-away of sludge for use in
home gardens). Based on the 1988 National Sewage Sludge Survey and 1988 Needs Survey, approximately 1.3%
of the total volume of sewage disposed is distributed and marketed (Federal Register, 1993a). Therefore, it is
estimated that 3 g TEQDF-WHO98 (4 g I-TEQDF) were released through distribution and marketing in 1995.
eSee Section 3.5 for estimates of CDD/CDF releases to air from sewage sludge incinerators.
Sources: Federal Register (1990, 1993a).
8.4.2. Drinking Water Treatment Plants
There is no strong evidence that chlorination of water for drinking purposes results in the
formation of CDDs/CDFs. Few surveys of CDD/CDF content in finished drinking water have
been conducted. Those that have been published have only rarely reported the presence of any
CDDs/CDFs, even at low picogram per liter DLs, and in those cases, CDDs/CDFs were also
present in the untreated water.
Rappe et al. (1989b) reported the formation of tetra- through octa-CDFs when tap water
and double-distilled water were chlorinated using chlorine gas. The CDF levels found in the
single samples of tap water and double-distilled water were 35 and 7 pg I-TEQDF/L, respectively.
No CDDs were detected at DLs ranging from 1 to 5 pg/L. However, the water samples were
8-91
-------
Table 8-43. Quantity of sewage sludge disposed of and potential dioxin TEQ
releases from primary, secondary, and advanced treatment publicly owned
treatment works for reference year 2000
Use/disposal practice
Land application13
Advanced treatment0
Other beneficial used
Surface disposal/landfill
Incineration
Other disposal method
TOTAL
Volume disposed of
(1,000 dry metric
tons/yr)
2,800
800
500
900
1,500
100
6,600
Percent of
total volume
42
12
8
14
23
2
100
Potential dioxin release"
(g TEQ/yr)
TEQDF-WH098
60.80
17.40
10.90
19.50
e
2.17
111.00
Totential dioxin TEQ release for nonincinerated sludges was estimated by multiplying the sludge volume generated
(second column) by the average of the mean TEQDF-WHO98 concentrations in sludge reported in U.S. EPA (2002d).
bWithout further processing or stabilization, such as composting.
°Such as composting.
dEPA assumes that this category includes distribution and marketing (sale or give-away of sludge for use in home
gardens). Based on the 1988 National Sewage Sludge Survey and 1988 Needs Survey, approximately 1.3% of the
total volume of sewage disposed is distributed and marketed (Federal Register, 1993a). Therefore, it is estimated
that 1.9 g TEQDF-WHO98 were released through distribution and marketing in 2000.
eSee Section 3.5 for estimates of CDD/CDF releases to air from sewage sludge incinerators.
Table 8-44. Biosolids disposal practices for reference year 2000
Use/disposal practice
Land application
Surface disposal/landfill
Incineration
Other
TOTAL
Volume disposed of (1,000 dry
metric tons/yr)
3,100
940
1,000
64
5,104
Percent of
total volume
61
18
20
1
100
Source: NRC (2002).
chlorinated at a dosage rate of 300 mg/L, which is considerably higher (by one to two orders of
magnitude) than the range of dosage rates typically used to disinfect drinking water. The authors
hypothesized that the CDFs or their precursors were present in chlorine gas.
8-92
-------
Rappe et al. (1990c) analyzed a 1,500-L sample of drinking water from a municipal
drinking water treatment plant in Sweden. Although the untreated water was not analyzed, a
sludge sample from the same facility was analyzed. The large sample volume enabled DLs on
the order of 0.001 pg/L. The TEQ contents of the water and the sludge were 0.0029 pg I-
TEQDF/L and 1.4 ng/kg, respectively. The congener patterns of the drinking water and sludge
sample were very similar, suggesting that the CDDs/CDFs detected in the finished water were
present in the untreated water.
8.4.3. Soaps and Detergents
As discussed in Section 8.4.1, CDDs/CDFs were detected in nearly all sewage sludges
tested, whether the sludges were obtained from industrialized areas or from rural areas. Because
of the ubiquitous presence of CDDs/CDFs in sewage sludge, several studies have been conducted
to determine their source(s). A logical category of products to test, because of their widespread
use, is detergents, particularly those that contain or release chlorine during use (i.e., hypochlorite-
containing and dichloroisocyanuric acid-containing detergents). The results of studies conducted
to date, summarized below, indicate that CDDs/CDFs are not formed during use of chlorine-free
detergents, chlorine-containing or chlorine-releasing detergents, or chlorine bleach during
household bleaching operations.
Sweden's Office of Nature Conservancy (1991) reported that the results of a preliminary
study conducted at one household indicated that CDDs/CDFs may be formed during use of
dichloroisocyanurate-containing dishwasher detergents. A more extensive main study with
multiple runs was then conducted using standardized food, dishes, cutlery, and other household
items. Testing of laundry washing and fabric bleaching and actual testing of the CDD/CDF
content of detergents was also performed. The study examined (1) hypochlorite- and
dichloroisocyanurate-containing dishwasher detergents, (2) sodium hypochlorite-based bleach
(4.4% NaOCl) in various combinations with and without laundry detergent, and (3) sodium
hypochlorite-based bleach, used at a high enough concentration to effect bleaching of a pair of
imported blue jeans.
CDDs/CDFs were not detected in either the chlorine-free detergent or the detergent with
hypochlorite; 0.6 pg TEQ/g were detected in the detergent containing dichloroisocyanurate. The
results of all dishwasher and laundry washing machine tests showed very low levels of
CDDs/CDFs, often nondetected values. There was no significant difference between the controls
and the test samples; in fact, the control samples had a higher TEQ content than did some of the
test samples. The drainwater from the dishwasher tests contained <1 to <3 pg I-TEQDF/L (the
water-only control sample contained <2.8 pg I-TEQDF/L). The CDD/CDF content of the laundry
8-93
-------
drainwater samples ranged from <1.1 to <4.6 pg I-TEQDF/L (the water-only control sample
contained <4.4 pg I-TEQDF/L).
Thus, under the study's test conditions, CDDs/CDFs were not formed during dishwashing
or laundry washing or during bleaching with hypochlorite-containing bleach. No definitive
reason could be found for the difference in results between the preliminary study and the main
study for dishwashing with dichloroisocyanurate-containing detergents. The authors of the study
suggested that differences in the foods used and the prewashing procedures employed in the two
studies were the likely causes of the variation in the results.
Rappe et al. (1990b) analyzed a sample of a Swedish commercial soft soap, a sample of
tall oil, and a sample of tall resin for CDD/CDF content. Tall oil and tall resin, by-products of
the pulping industry, are the starting materials for the production of soft liquid soap. Crude tall
oil, collected after the Kraft pulping process, is distilled under reduced pressure at temperatures
of up to 280 to 290°C, yielding tall oil and tall resin. The measured TEQ content of the liquid
soap was found to be 0.647 ng TEQDF-WHO98/L (0.447 ng I-TEQDF/L). PeCDDs were the
dominant congener group, followed by HpCDDs, HxCDDs, PeCDFs, and OCDD, with some
tetra-CDFs and CDDs also present. The TEQ contents of the tall oil (12 ng TEQDF-WHO98/kg
[9.4 ng I-TEQDF/kg]) and tall resin (196 ng TEQDF-WHO98/kg [200 ng I-TEFDF/kg]) were
significantly higher than the level found in the liquid soap. The tall oil contained primarily tetra-
and penta-CDDs and CDFs, whereas the tall resin contained primarily HpCDDs, HxCDDs, and
OCDD. The investigators compared the congener patterns of the three samples and noted that
although the absolute values for the tetra- and penta-CDDs and CDFs differed among the tall oil,
tall resin, and liquid soap samples, the same congeners were present. The congener patterns for
the higher-chlorinated congeners were very similar. Table 8-45 presents the results of the study.
In 1987, 118 million L of liquid household soaps were shipped in the United States (U.S.
DOC, 1990b); shipment quantity data are not available in the 1992 U.S. Economic Census (U.S.
DOC, 1996). Because only one sample of liquid soap has been analyzed for CDD/CDF content
(Rappe et al., 1990b), no estimate of environmental release can be made.
8.4.4. Textile Manufacturing and Dry Cleaning
As discussed in Section 8.4.1, CDDs/CDFs have been detected in almost all sewage
sludges tested, whether they were obtained from industrialized areas or rural areas. To determine
whether textile fabric or fabric finishing processes could account for the observed CDDs/CDFs,
several studies were conducted in Germany. These studies, summarized in the following
paragraphs, indicate that some finished textile products do contain detectable levels of
CDDs/CDFs and that they can be released from the textile during laundering or dry cleaning;
8-94
-------
Table 8-45. CDD/CDF concentrations in Swedish liquid soap, tall oil, and
tall resin
Congener/congener group
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,4,8-71,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3, 4,7, 8/9-HxCDF
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-HpCDF
OCDF
Total 2,3,7,8-CDDa
Total 2,3,7,8-CDFa
Total I-TEQDFa
Total TEQDF-WHO98a
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDFa
Liquid soap
(ng/L)
ND (0.009)
0.4
ND (0.02)
0.32
0.18
1.9
1
0.62
0.29
0.2
0.013
ND (0.004)
ND (0.004)
ND (0.004)
ND (0.005)
ND (0.01)
NA
3.8
1.123
0.447
0.647
0.12
15
3.4
3.6
1
1
1.3
0.15
ND (0.01)
NA
25.57
Tall oil
(ng/kg)
3.6
5.3
ND(2)
ND(2)
ND(2)
ND(1)
5.3
17
4.2
1.9
1.4
0.7
ND (0.7)
ND (0.5)
ND (0.8)
ND(2)
NA
14.2
25.2
9.4
12
31
380
3.3
ND(1)
5.3
26
41
4.9
ND(2)
NA
491.5
Tall resin
(ng/kg)
ND(1)
3.1
ND(4)
810
500
5,900
6,000
ND(2)
ND (0.4)
ND (0.5)
24
ND(1)
ND (0.7)
10
9.0
NA
13,213.1
43
200
196
ND(1)
25
6,800
11,000
6,000
ND(2)
ND (0.5)
56
19
NA
23,900
""Calculations assume nondetect values were zero.
NA = Not analyzed
ND = Not detected (value in parenthesis is the detection limit)
~ = No information given
Source: Rappe et al. (1990b).
8-95
-------
however, textile finishing processes are typically not sources of CDD/CDF formation. Rather,
the use of CDD/CDF-containing dyes and pigments and the use in some countries of PCP to treat
unfinished cotton appear to be the sources of the detected CDDs/CDFs.
Horstmann et al. (1993b) analyzed the CDD/CDF content of eight different raw
(unfinished) cotton cloths containing fiber from different countries and five different white
synthetic materials (acetate, viscose, bleached polyester, polyamide, and polyacrylic). The
maximum concentrations found in the textile fabrics were 30 ng/kg in the cotton products and 45
ng/kg in the synthetic materials. Also, a cotton finishing scheme was developed that subjected
one of the cotton materials to a series of 16 typical cotton finishing processes; one sample was
analyzed following each step. The fabric finishing processes showing the greatest effect on
CDD/CDF concentration were the application of an indanthrene dye and the "wash and wear"
finishing process, which together resulted in a CDD/CDF concentration of about 100 ng/kg. On
the basis of the concentrations found, the authors concluded that neither unfinished new fabrics
nor common cotton finishing processes could explain the CDD/CDF levels found in laundry
wastewater.
Fuchs et al. (1990) reported that the dry-cleaning solvent redistillation residues collected
from 12 commercial and industrial dry-cleaning operations contained considerable amounts of
CDDs/CDFs. The reported I-TEQDF content ranged from 131 to 2,834 ng/kg, with the dominant
congeners being OCDD and HpCDD. Towara et al. (1992) demonstrated that neither the use of
chlorine-free solvents nor variation of the dry-cleaning process parameters lowered the
CDD/CDF content of the residues.
Umlauf et al. (1993) conducted a study to characterize the mass balance of CDDs/CDFs
in the dry-cleaning process. The soiled clothes (containing 16 pg total CDDs/CDFs per kg)
accounted for 99.996% of the CDD/CDF input. Input of CDDs/CDFs from indoor air containing
0.194 pg/m3 accounted for the remainder (0.004%). The dry-cleaning process removed 82.435%
of the CDDs/CDFs in the soiled clothing. Most of the input CDDs/CDFs (82.264%) were found
in the solvent distillation residues. Air emissions (at 0.041 pg/m3) accounted for 0.0008% of the
total input, which was less than the input from indoor air. The fluff (at a concentration of
36 ng/kg) accounted for 0.1697%, and water effluent (at a concentration of 0.07 pg/L) accounted
for 0.0000054%.
Horstmann and McLachlan (1994a, b, 1995) analyzed 35 new textile samples (primarily
cotton products) obtained in Germany for CDDs/CDFs. Low levels were found in most cases
(total CDD/CDF <50 ng/kg). The dominant congeners were OCDD and HpCDD. However,
several colored T-shirts from a number of clothing producers had extremely high levels, with
concentrations up to 290,000 ng/kg. Because the concentrations in identical T-shirts purchased
at the same store varied by up to a factor of 20, the authors concluded that the source of
8-96
-------
CDDs/CDFs was not a textile finishing process, because a process source would have resulted in
a more consistent level of contamination. Klasmeier and McLachlan (1995) subsequently
analyzed 68 new textile products obtained in Germany for OCDD and OCDF. Most samples had
nondetectable levels (42 samples <60 ng/kg). Only four samples had levels exceeding 500 ng/kg.
Horstmann and McLachlan (1994a, b) reported finding two different congener group
patterns in the more contaminated of the 35 textile products. One pattern agreed with the
congener pattern for PCP reported by Hagenmaier and Brunner (1987), whereas the other pattern
was similar to that reported by Remmers et al. (1992) for chloranil-based dyes. The authors
hypothesized that the use of PCP to preserve cotton, particularly when it is randomly strewn on
bales of cotton as a preservative during sea transport, was the likely source of the high levels
occasionally observed. Although the use of PCP for nonwood uses was prohibited in the United
States in 1987 (see Section 8.3.8), PCP is still used in developing countries, especially to
preserve cotton during sea transport (Horstmann and McLachlan, 1994a).
Horstmann and McLachlan (1994a, b) conducted additional experiments that
demonstrated that the small percentage of clothing items with high CDD/CDF levels could be
responsible for the quantity of CDDs/CDFs observed in household wastewater (see Section
8.4.1.1).
8-97
-------
9. INDICATIONS OF POSSIBLE NATURAL SOURCES OF CDDs/CDFs
Numerous laboratory and field research studies have demonstrated that biochemical and
photolytic formation of CDDs/CDFs from chlorophenol precursors is possible. In addition,
under certain conditions, some CDDs/CDFs can be biodegraded to form less chlorinated (and
possibly more toxic) CDDs/CDFs. Both of these mechanisms are discussed in this chapter;
however, the extent to which CDDs/CDFs are formed by either mechanism in the environment is
not known at present. The potential for releases of CDDs/CDFs from the application of animal
manure to farmland and the mining and use of ball clay is also discussed.
9.1. BIOLOGICAL SOURCES OF CDDs/CDFs
9.1.1. Biotransformation of Chlorophenols
Biochemical formation of CDDs/CDFs—particularly the higher-chlorinated congeners—
from chlorophenol precursors is possible, as indicated in laboratory studies with solutions of
trichlorophenols and pentachlorophenol (PCP) in the presence of peroxidase enzymes and
hydrogen peroxide (Svenson et al., 1989; Oberg et al., 1990; Wagner et al., 1990; Oberg and
Rappe, 1992; Morimoto and Kenji, 1995) and with sewage sludge spiked with PCP (Oberg et al.,
1992). However, the extent to which CDDs/CDFs are formed in the environment via this
mechanism cannot be estimated at this time.
In 1991, Lahl et al. (1991) reported finding CDDs/CDFs in all 22 samples of the various
types of composts analyzed. The hepta- and octa-substituted CDDs/CDFs were typically the
dominant congener groups found. The I-TEQDF content of the composts ranged from 0.8 to 35.7
ng I-TEQDF/kg. The CDDs/CDFs found in compost may primarily result from atmospheric
deposition onto plants that are subsequently composted, but they may also be caused by uptake of
CDDs/CDFs from air by the active compost (Krauss et al., 1994). CDDs/CDFs are also
frequently detected in sewage sludges, and they may come primarily from the sources identified
in Section 8.4.1.
Peroxidases are common enzymes in nature. For example, the initial degradation of the
lignin polymer by white- and brown-rot fungi is peroxidase catalyzed (Wagner et al., 1990). The
conversion efficiency of chlorinated phenols to CDDs/CDFs that has been observed is low. In
the solution studies, Oberg and Rappe (1992) and Morimoto and Kenji (1995) reported
conversion efficiencies of PCP to OCDD of about 0.01% and 0.8%, respectively, and Wagner et
al. (1990) reported a conversion efficiency of trichlorophenol to HpCDD of about 0.001%.
Oberg et al. (1990) reported a conversion efficiency of trichlorophenols to CDDs/CDFs of about
0.001%. In their sewage sludge study, Oberg et al. (1992) reported a conversion efficiency of
PCP to total CDDs of 0.0002 to 0.0004%.
9-1
-------
Several researchers have conducted both laboratory and field studies in an attempt to
better understand the extent of, and factors affecting, the fate or formation of CDDs/CDFs in
composts and sewage sludges. The findings of several of these studies are discussed in the
following paragraphs. These findings are not always consistent because the congener profiles
and patterns detected—and the extent of CDD/CDF formation, if any—may vary with compost
materials studied, differences in experimental or field composting design, and duration of the
studies.
Harrad et al. (1991) analyzed finished composts and active compost windrows from a
municipally operated yard waste composting facility in Long Island, NY. Concentrations
measured in 12 finished composts ranged from 14 to 41 ng I-TEQDF/kg (mean, 3 ng I-TEQDF/kg).
The concentrations in the five active compost samples (1 to 30 days in age) ranged from 7.7 to 54
ng I-TEQDF/kg (mean, 21 ng I-TEQDF/kg). The authors observed that CDD/CDF concentrations
measured in two soil samples from the immediate vicinity of the composting facility were
significantly lower (1 and 1.3 ng I-TEQDF/kg) than the levels found in the composts, suggesting
that the source(s) of CDDs/CDFs in the composts was different from the source(s) affecting local
soils.
Harrad et al. (1991) also noted a strong similarity between the congener profiles observed
in the composts and the congener profile of a PCP formulation (i.e., predominance of 1,2,4,6,8,9-
HxCDF and 1,2,3,4,6,8,9-HpCDF in their respective congener groups), which seemed to indicate
that leaching of CDDs/CDFs from PCP-treated wood in the compost piles was the likely source
of the observed CDDs/CDFs. The levels of PCP in the 12 finished composts ranged from 7 to
190 |ig/kg (mean, 33 |ig/kg), and the PCP levels in the active compost samples ranged from 17 to
210 |ig/kg (mean, 68 |ig/kg). The PCP level in both soil samples was 1.5 |ig/kg.
Goldfarb et al. (1992) and Malloy et al. (1993) reported the results of testing composts at
three municipal yard waste composting facilities (5 to 91 ng I-TEQDF/kg; mean, 30 ng I-
TEQDF/kg), two municipal solid waste composting facilities (19 to 96 ng I-TEQDF/kg; mean, 48
ng I-TEQDF/kg), and one municipal facility composting solid waste and dewatered sewage sludge
(37 to 87 ng I-TEQDF/kg; mean, 56 ng I-TEQDF/kg). All facilities were located in the United
States. Two general trends were observed for the three types of composts: a progressive increase
in analyte levels, with an increasing degree of chlorination for each compound type (CDDs,
CDFs, chlorophenols, and chlorobenzenes), and a progressive increase in concentration of each
congener or homologue group from yard waste to solid waste to solid waste/sewage sludge
composts. As noted above, the mean TEQ concentrations showed this same trend, which was
primarily due to increasing levels of 1,2,3,4,6,7,8-HpCDD and OCDD. The mean PCP
concentrations in the three compost types were 20 |ig/kg (yard waste), 215 |ig/kg (solid waste),
and 615 |ig/kg (solid waste/sewage sludge).
9-2
-------
Comparison of congener profiles by the authors indicated that the CDD/CDF residue in
PCP-treated wood in the compost feedstock was a major but not exclusive contributor of the
observed CDDs/CDFs. The authors postulated that biological formation of HxCDDs, HpCDDs,
and OCDD from chlorophenols (tri, tetra, and penta) in the compost could be responsible for the
elevated levels of these congener groups relative to their presence in PCP.
Oberg et al. (1993) measured the extent of CDD/CDF formation in three conventional
garden composts; two were spiked with PCP and one was spiked with hexachlorobenzene. One
PCP-spiked compost was monitored for 55 days and the other for 286 days. A significant
increase in the concentrations of the more highly chlorinated congeners, particularly the
HpCDDs, OCDD, and, to a lesser extent, OCDF, were observed. Similar results were reported
for the hexachlorobenzene-spiked compost, which was monitored for 49 days. Oberg et al. stated
that for a "typical" composting event, a two- to threefold increase in TEQ content corresponded
with an elevation of 0.2 to 0.5 ng I-TEQDF/kg dry weight.
Weber et al. (1995) subjected sewage sludges from two German communities to
anaerobic digestion in laboratory reactors for 60 days. The two sludges were spiked with 2,3,5-
trichlorophenol (10 to 25 mg/kg), a mixture of 2,3,5-trichlorophenol and dichlorophenols (2.5 to
25 mg/kg), or a mixture of di-, tri-, and tetrachlorobenzenes (4 to 40 mg/kg). The initial
CDD/CDF concentrations in the two sludges were 9 and 20 ng I-TEQDF/kg. In nearly all of the
digestion experiments, the addition of the precursors did not lead to any significant changes in
concentrations. The only exceptions were increased 2,3,7,8-TCDF concentrations in the mixed
chlorophenol experiments and decreased 2,3,7,8-TCDF concentrations in the mixed
chlorobenzene experiments. However, the same increases or decreases for this congener were
also observed in the controls (i.e., no precursors added).
Researchers at the U.S. Department of Agriculture (USD A) (Fries et al., 1997) reported
that dairy cows that were fed PCP-treated wood excreted amounts of OCDD almost four times
greater than the amounts ingested. Feil and Tiernan (1997) reported that rats fed technical PCP
had liver concentrations of HxCDD, HpCDD, HpCDF, OCDD, and OCDF two to three orders of
magnitude higher than those of rats fed purified PCP. These results suggest the in vivo
formation of CDDs/CDFs from pre-dioxins (i.e., chlorinated phenoxy phenols present as
contaminants in the PCP). A follow-up USDA study (Huwe et al., 1998) investigated the
metabolic conversion of a pre-dioxin (monochloro-2-phenoxyphenol) to OCDD in a feeding
study with rats. The results of the study demonstrated the formation of OCDD from the pre-
dioxin, although the conversion was estimated to be less than 2%. Interestingly, the study noted
that the presence of added OCDD in the feed material increased the percentage of pre-dioxin
conversion.
9-3
-------
Wittsiepe et al. (1998) demonstrated that CDDs/CDFs can be formed through reaction of
chlorophenols with myeloperoxidase (a component of neutrophile granulocytes, a subgroup of
human leucocytes). The CDDs/CDFs formed showed different homologue patterns and
formation rates depending on the degree of chlorination of the chlorophenol substrate. The
formation rates ranged from 1 to 16 jimol of CDD/CDF per mol of chlorophenol substrate.
9.1.2. Biotransformation of Higher CDDs/CDFs
Results of several studies that examined the fate of a range of CDD/CDF congeners in
pure cultures, sediments, and sludges indicate that under certain conditions some CDD/CDF
congeners will undergo biodegradation to form lower-chlorinated (and possibly more toxic)
CDDs/CDFs. However, the extent to which more toxic CDDs/CDFs are formed in the
environment via this mechanism cannot be estimated at this time.
Several reports indicate that CDDs and CDFs may undergo microbial dechlorination in
anaerobic sediments. Adriaens and Grbic-Galic (1992, 1993) and Adriaens et al. (1995) reported
the results of a series of microcosm studies using Hudson River sediment (contaminated with
Aroclor 1242) and aquifer material (contaminated with CDDs) from Pensacola, FL. Both types
of substrates were spiked with several CDDs (1,2,3,4,6,7,8-HpCDD; 1,2,3,4,7,8-HxCDD; and
l,2,4,6,8,9-/l,2,4,6,7,9-HxCDD) and CDFs (1,2,3,4,6,7,8-HpCDF and 1,2,4,6,8-PeCDF) and
monitored over a 16-month period at an incubation temperature of 30°C. The Hudson River
sediment was spiked with 144 |ig/kg of each congener, and the Pensacola aquifer material was
spiked with 63 |ig/kg of each congener.
All of the congeners, with the exception of 1,2,3,4,6,7,8-HpCDF, showed a slow decrease
in concentration over time, which was attributed to biologically mediated reductive
dechlorination, with net disappearance rates ranging from 0.0031 wk"1 to 0.0175 wk"1 (i.e., half-
lives of approximately 1 to 4 yr). However, Adriaens et al. concluded that actual half-lives may
be orders of magnitude higher. The experiment with 1,2,3,4,6,7,8-HpCDD yielded formation of
1,2,3,4,7,8-HxCDD and 1,2,3,6,7,8-HxCDD. Thus, removal of the peri-substituted (1,4,6,9)
chlorines was favored, with enrichment of 2,3,7,8-substituted congeners. No lower-chlorinated
congeners were identified from incubation with the other tested congeners. 1,2,4,6,8-PeCDF was
also examined in dichlorophenol-enriched cultures. After 6 months of incubation, several
TCDFs were identified, which also indicated that peri-dechlorination was the preferred route of
reduction.
Barkovskii and Adriaens (1995, 1996) reported that 2,3,7,8-TCDD extracted from Passaic
River sediments was susceptible to reductive dechlorination when incubated at 30°C under
methanogenic conditions in a mixture of aliphatic and organic acids inoculated with
microorganisms obtained from the sediments. The initial concentration of 2,3,7,8-TCDD (20 ± 4
9-4
-------
|ig/L) decreased by 30% to 14 ± 2 jig/L over a period of 7 months, with the consecutive
appearance and disappearance of tri-, di-, and mono-CDDs. Experiments were also conducted by
spiking the sediment with HxCDDs, HpCDDs, and OCDD. Up to 10% of the spiked OCDD was
converted to hepta-, hexa-, penta-, tetra-, tri-, di-, and mono-chlorinated isomers, but the reaction
stoichiometry was not determined. Two distinct pathways of dechlorination were observed: the
peri-dechlorination pathway of 2,3,7,8-substituted hepta- to penta-CDDs, resulting in the
production of 2,3,7,8-TCDD, and the peri-lateral dechlorination pathway of non-2,3,7,8-
substituted congeners.
Several studies have reported that CDDs/CDFs can be formed during composting
operations through biological action on chlorophenols present in the compost feed material. The
results of studies that specify likely involvement of chlorophenols are described in Section 9.1.
Another possible formation mechanism was suggested by Vikelsoe et al. (1994), who reported
that higher-chlorinated CDD/CDF congeners are formed when humic acid is reacted with a
peroxidase enzyme, hydrogen peroxide, and sodium chloride. It is expected that some organic
material in compost and sewage sludge has a humic-like structure. Several additional studies are
described below in which the potential involvement of chlorophenols could not be assessed
because chlorophenol concentrations in the composts were not reported.
Schafer et al. (1993) monitored the seasonal changes in the CDD/CDF content, as well as
the extent of CDD/CDF formation, in composts from a vegetable and garden waste composting
operation in Germany. Finished compost samples were collected and analyzed every 2 months
for 1 yr. An annual cycle was observed in TEQ concentrations, with peak concentrations in the
summer (approximately 8.5 ng I-TEQDF/kg) being 2.5 times higher than the lowest concentrations
observed in the winter (approximately 3.5 ng I-TEQDF/kg). No seasonal source was apparent that
could explain the observed differences in seasonal levels.
The CDD/CDF content of the starting waste materials for two compost cycles (March and
September) was measured to monitor the extent of CDD/CDF formation during composting. For
the March cycle sample, most 2,3,7,8-substituted CDD/CDF congeners decreased in
concentration during composting. Four CDF congeners showed a slight increase in concentration
(less than 10%). For the September cycle sample, OCDD and HpCDD concentrations increased
300% during composting. Increases of less than 10% were observed for HxCDDs and OCDF; all
other 2,3,7,8-substituted CDD/CDF congeners showed decreases in concentrations during
composting.
Krauss et al. (1994) measured the extent of CDD/CDF formation during the composting
of household waste using a laboratory compost reactor. After 11 wk, the TEQ content of the
compost increased from 3 to 4.5 ng. The largest increases in mass content were observed for
HpCDD (primarily 1,2,3,4,6,7,8-HpCDD) and OCDD. TCDD, PeCDD, and HxCDD showed no
9-5
-------
change in mass content. All CDF congener groups showed decreases in mass content; however,
the concentrations in both the starting and the finished compost were close to the analytical
detection limits.
Oberg et al. (1994) reported the results of monitoring two household waste composts and
two garden composts. The total CDD/CDF content of both household waste composts decreased
over the 12-wk test period. Total CDD content and PCB content decreased, but total CDF
content increased, in contrast to the findings of Krauss et al. (1994). However, a small increase
in OCDD content in both composts was observed. The two garden composts were monitored for
a 60-wk period. Total CDD/CDF concentration increased, with the largest increases observed for
OCDD and HpCDDs. The lower-chlorinated CDFs decreased in concentration.
As a follow-up to a preliminary study (Hengstmann et al., 1990) that indicated CDD/CDF
concentrations may increase and congener profiles may change during anaerobic digestion of
sewage sludge, Weber et al. (1995) subjected sewage sludges from two German communities to
anaerobic digestion and aerobic digestion in laboratory reactors for 60 days and 20 days,
respectively. The initial average I-TEQDF concentrations in the raw sludges were 20 and 200 ng
I-TEQDF/kg. No significant increase or decrease in total CDD/CDF content or congener group
content was observed with either sludge. In contrast, a significant decrease in CDD/CDF content
was observed in the aerobic digestion experiments with both sludges. The greatest percentage
decreases in congener group concentrations (greater than 40%) were observed for TCDF,
PeCDF, HxCDF, TCDD, and PeCDD in the sludge initially containing 20 ng I-TEQDF/kg and for
TCDF, TCDD, HpCDD, and OCDD in the initially high-content sludge. The greatest percentage
decreases in congener concentrations (greater than 40%) were observed for non-2,3,7,8-
substituted congeners.
The data presented in this section and in Section 9.1.1 do not provide a basis for making a
release estimate via biotransformation; therefore, biotransformation releases are classified as
Category E (not quantifiable).
9.1.3. Dioxin-Like Compounds in Animal Manure
In 2000, approximately 9 billion individual livestock and poultry animals were raised on
commercial farms in the United States (U.S. Census Bureau, 2001). It is estimated that beef
animals, dairy cows, chickens, turkeys, and pigs, combined, produced in excess of 190 billion kg
(dry weight) of manure in 2000 (Table 9-1). Because livestock and poultry manure can provide
valuable organic material and nutrients for crop and pasture growth, most of the animal manure
generated at commercial farms and animal feed lots is applied to farmland as fertilizer. To the
extent dioxin-like compounds may contaminate animal manures, the practice of land-spreading
animal waste may result in releases of CDDs/CDFs to the open and circulating environment.
9-6
-------
Table 9-1. Estimated quantity of animal manure produced in the United
States in 2000
Species
Swine
Layer
Broiler
Turkey
Beef
Dairy cow
Numbers of
individuals
on farms in
2000a
6.73e+07
4.35e+08
8.26e+09
2.7e+08
9.73e+07
9.21e+06
Average
weight of
animal
(lbs)b
135
4
2
15
800
1,400
Total live
weight on
farms (Ibs)
9.09e+09
1.74e+09
1.65e+10
4.05e+09
7.78e+10
1.29e+10
Manure
generation rate
factor
(dry weight Ib/lb
live unit
weight/day)0
8.2e-03
1.6e-02
2.1e-02
1.2e-02
6.9e-03
le-02
Total
Manure
generated
(Ib/yr dry
weight)
2.72e+10
1.02e+10
1.27e+ll
1.77e+10
1.96e+ll
4.71e+10
4.25e+ll
Manure
produced
(kg/yr dry
weight)
1.23e+10
4.61e+09
5.74e+10
8.04e+09
8.89e+10
2.13e+10
1.93e+ll
"Source: U.S. Census Bureau (2001).
bSource: U.S. EPA (200Id).
°Source: Stevens and Jones (2003).
Stevens and Jones (2003) published results of CDD and CDF detection in animal manure applied
to farmland in the United Kingdom. Manure from six milking dairy cows was sampled at six
farms in the northern United Kingdom. In addition, single samples of sheep, chicken, and pig
manure were collected from other farms in the region. The samples were shipped to a laboratory
for trace chemical analysis. Samples were analyzed using high-resolution gas chromatography
coupled with high-resolution mass spectrometry and a capillary column for the identification of
CDD/CDF congeners. Recoveries of the internal standard ranged from 51 to 94%, with a mean
of 74% for CDD/CDF congeners. Table 9-2 summarizes the results of the study. The pig and
chicken manure contained approximately 0.2 ng WHO-TEQ/kg, and the cow manure averaged
3.6 ng WHO-TEQ/kg in concentration.
This study provides extremely limited data on the possible levels and occurrences of
dioxin-like compounds in farm animal manure, and, therefore, these data are clearly not
representative of national releases of dioxin-like compounds from the land application of all farm
animal manure in the United States. Accordingly, EPA currently considers this source to be
unquantifiable (Category E) in terms of dioxin emissions.
9-7
-------
Table 9-2. CDD and CDF concentrations (ng/kg dry weight) in samples of
animal manure in the United Kingdom
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
Total 2,3,7,8-CDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
,2,3,4,7,8-HxCDF
,2,3,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,4,6,7,8-HpCDF
,2,3,4,7,8,9-HpCDF
OCDF
Total 2,3,7,8-CDF
Total CDD/CDF
WHO-TEQ
Cows (n = 6)
(mean)
0.17
0.46
2.4
4.5
2.6
120
460
590.1
0.3
0.3
0.28
0.6
0.51
1.9
0.4
7.6
12
35
58.9
649
3.6
Sheep (n = 1)
0.11
0.41
0.9
0.86
0.56
9.4
53
65.2
.2
.1
.2
.4
.1
0.15
1.4
5.2
0.56
5
18.3
83.5
2.1
Pig (n = 1)
0.01
0.07
0.26
0.1
0.07
0.8
11
12.3
0.03
0.04
0.06
0.05
0.06
0.04
0.06
0.48
0.04
0.73
1.6
13.9
0.2
Chicken
(n = l)
0.01
0.04
0.03
0.09
0.12
1.4
14
15.7
0.03
0.09
0.12
0.15
0.07
0.05
0.14
0.37
0.09
0.8
1.9
17.6
0.2
Source: Stevens and Jones (2003).
9.2. PHOTOTRANSFORMATION OF CHLOROPHENOLS
Several researchers have demonstrated that CDDs/CDFs can be formed via photolysis of
PCP under laboratory conditions. However, the extent to which CDDs/CDFs are formed in the
environment via this mechanism cannot be estimated at this time.
Lamparski et al. (1980) conducted laboratory studies to determine the effect of simulated
summer sunlight on the formation of OCDD, HpCDDs, and HxCDDs in wood pressure-treated
in the laboratory with PCP. In the first set of experiments, wood veneers (southern pine) treated
with purified PCP or Dowicide EC-7, using methylene chloride as the PCP carrier, were exposed
to light for 70 days. The PCP concentration in the treated wood was 5% by weight, which
approximates the concentration in the outer layer of PCP-treated wood utility poles. Photolytic
condensation of PCP to form OCDD was observed, with the OCDD concentration increasing by
a maximum factor of 3,000 for the purified PCP and by a factor of 20 for EC-7 at about day 20
9-8
-------
before leveling off. HpCDD and HxCDD were also formed, apparently by photolytic
degradation of OCDD rather than by condensation of PCP and tetrachlorophenols. The HxCDD
concentration increased by a factor of 760 for the purified PCP and by a factor of 50 for EC-7
over the 70-day exposure period. The predominant HpCDD congener formed was 1,2,3,4,6,7,8-
HpCDD as a result of an apparently preferential loss of chlorine at the peri position (positions 1,
4, 6, and 9).
In a second set of experiments conducted by Lamparski et al. (1980), a hydrocarbon oil
(P-9 oil) was used as the carrier to treat the wood. The increases observed in the OCDD,
HpCDD, and HxCDD were reported to be much lower relative to the increases observed in the
first set of experiments, which used methylene chloride as the carrier. Results were reported only
for OCDD. The OCDD concentration increased by a maximum factor of 1.5 for both EC-7 and
technical PCP and by a factor of 88 for purified PCP. The authors concluded that the oil either
reduced condensation of PCP to OCDD or accelerated degradation to other species by providing
a hydrocarbon trap for free-radical species.
Vollmuth et al. (1994) studied the effect of irradiating laboratory water and landfill
seepage water that contained PCP under conditions simulating those used to purify water with
ultraviolet (UV) radiation (5-hr exposure to 254 nm radiation from low-pressure mercury lamps).
Before irradiation, the three solutions tested contained approximately 1 mg/L of PCP or PCP-Na,
but the CDD/CDF content of one solution varied dramatically from those of the other two (1.5
vs. 2,066 and 2,071 pg I-TEQDF/L). Irradiation resulted in nearly total destruction of PCP
(greater than 99% loss) in all three experiments. An overall net increase in I-TEQDF-content was
observed in the initially low I-TEQDF-content water, but a net decrease was observed for the two
initially high I-TEQDF-content waters.
Irradiation of laboratory water containing purified PCP showed an increase in I-TEQDF
concentration from 1.5 pg/L to 214.5 pg/L. The increase was due entirely to the formation of
1,2,3,4,6,7,8-HpCDD, OCDD, and 1,2,3,4,6,7,8-HpCDF. Formation of non-2,3,7,8-substituted
HpCDDs and HpCDFs was also observed. The ratios of the concentrations of these non-2,3,7,8-
congeners to the concentrations of the 2,3,7,8-congeners were 0.6 for HpCDDs and 5 for
HpCDFs. The HpCDD and HpCDF congeners formed indicate that the operative mechanism
was photoinduced dechlorination of OCDD at a peri position and dechlorination of OCDF at
only the 1 and 9 peri positions.
Irradiation of water containing technical PCP-Na (Dowicide-G) resulted in a net loss in I-
TEQDF content, from 2,065.5 pg/L to 112.7 pg/L. The only 2,3,7,8-substituted congener showing
an increased concentration was 1,2,3,6,7,8-HxCDD. The other congeners originally present in
the technical PCP-Na showed reductions of 80.6 to 100%.
9-9
-------
The I-TEQDF content of seepage water from a landfill (2,071 pg I-TEQDF/L) was reduced
by a factor of 2, to 1,088 pg I-TEQDF/L. However, several 2,3,7,8-substituted congeners did
increase in concentration (1,2,3,6,7,8-HxCDD; 1,2,3,4,6,7,8-HpCDD; 1,2,3,4,6,7,8-HpCDF; and
OCDF).
Waddell et al. (1995) also studied the effect of irradiating distilled laboratory water
containing PCP under conditions simulating those used to purify water with UV radiation. The
results obtained were similar to those of Vollmuth et al. (1994). Analytical-grade PCP at a
concentration of 10 mg/L was exposed for 12 min to 200 to 300 nm radiation from a medium-
pressure mercury lamp. All CDD/CDF congener groups increased in concentration over the 12-
min exposure period, with the greatest increases observed for OCDD (75-fold increase) and
HpCDDs (34-fold increase). The I-TEQDF content of the solution increased from 4.2 pg I-
TEQDF/L to 137 pg I-TEQDF/L over the 12-min period. The dominant congeners formed, in terms
of both concentration and contribution to I-TEQDF, were 1,2,3,4,6,7,8-HpCDD, OCDD, and
1,2,3,7,8,9-HxCDD.
9.2.1. Photolysis of Higher CDDs/CDFs
Photolysis appears to be one of the few environmentally significant degradation
mechanisms for CDDs/CDFs in water, air, and soil. Although good mass balances were not
obtained and the photolytic pathways for CDDs/CDFs were not fully identified in most studies, a
major photolysis pathway appears to be photodechlorination, resulting in formation of lower-
chlorinated CDDs/CDFs. A preferential loss of chlorines from the peri positions (1, 4, 6, and 9)
rather than from the lateral positions (2, 3, 7, and 8) was reported for some congener groups
when irradiated as dry films and sorbed to soil and in gas-phase CDDs/CDFs (Choudhry and
Webster, 1989; Kieatiwong et al., 1990; Sivils et al., 1994, 1995; Tysklind et al., 1992). Several
researchers reported that carbon-oxygen cleavage and other mechanisms may be similarly or
more important pathways for CDDs/CDFs containing four or fewer chlorines.
Because of the difficulties inherent in controlling experimental variables for nonvolatile
and highly lipophilic compounds such as CDDs/CDFs, few photolysis studies have been
performed on natural waters, soils, atmospheric particulates, and atmospheric gases to examine
the rates and products of photolysis under environmentally relevant conditions. Thus, it is not
possible at this time to quantitatively estimate the mass of various CDD/CDF congeners formed
in the environment annually via photolytic mechanisms. The following sections summarize the
key findings of environmentally significant studies for water, soil, vegetation, and air.
9-10
-------
9.2.2. Photolysis in Water
Numerous studies have demonstrated that CDDs/CDFs will undergo photodechlorination
following first-order kinetics in organic solution, with preferential loss of chlorine from the
lateral positions. Photolysis is slow in pure water, but it increases dramatically when solvents
serving as hydrogen donors such as hexane, benzene, methanol, acetonitrile, hexadecane, ethyl
oleate, dioxane, and isooctane are present. However, only a few studies have examined the
photolysis of CDDs/CDFs using natural waters and sunlight.
Choudhry and Webster (1989) experimentally determined the sunlight photolysis half-life
of 1,3,6,8-TCDD in pond water to be 3.5 days (more than 10 times greater than the half-life
predicted by laboratory experiments using a water/acetonitrile solution). The authors attributed
this significant difference in photolysis rates to the light screening/quenching effects of dissolved
organic matter.
Friesen et al. (1990) examined the photolytic behavior of 1,2,3,4,7-PeCDD and
1,2,3,4,6,7,8-HpCDD in water:acetonitrile (2:3, v/v) and in pond water under sunlight at 50
degrees North latitude. The observed half-lives of these two compounds in the water: acetonitrile
solution were 12 and 37 days, respectively, but were much shorter in pond water, 0.94 and 2.5
days, respectively. Similarly, Friesen et al. (1993) studied the photodegradation of 2,3,7,8-TCDF
and 2,3,4,7,8-PeCDF by sunlight using water:acetonitrile (2:3, v/v) and lake water. The observed
half-lives were 6.5 and 46 days, respectively, in the water:acetonitrile solution and 1.2 and 0.19
days, respectively, in lake water. The significant differences between the natural water and the
water:acetonitrile solution results were attributed to indirect or sensitized photolysis due to the
presence of naturally occurring components in the lake and pond water.
Dung and O'Keefe (1992), in an investigation of aqueous photolysis of 2,3,7,8-TCDF and
1,2,7,8-TCDF, reported findings similar to those of Friesen et al. (1993). The photolysis rates of
the two TCDF congeners observed in river and lake water (half-lives of about 4 to 6 hr) were
double those observed in pure water (half-lives of about 8 to 11 hr). The authors attributed the
difference in rates to the presence of natural organics in the river and lake water that may act as
sensitizers.
9.2.3. Photolysis on Soil
Photolysis of CDDs/CDFs on soil has not been well characterized. According to the data
generated to date, however, photolysis is an operative degradation process only in the near-
surface soil where UV light penetrates (the top few millimeters or less of soil), and
dechlorination of peri-substituted chlorines appears to occur preferentially.
Miller et al. (1989) studied the CDD degradation products resulting from irradiation of
13C-labeled OCDD on two soil types using sunlamps. Approximately 38 to 42% of the OCDD
9-11
-------
was degraded by day 5 of the experiment; no significant further loss of OCDD was observed over
the following 10 days. Although the authors determined that photodechlorination was not the
dominant photolysis pathway, it was observed in both soils; approximately 10 to 30% of the
lower-chlorinated congeners were produced from the immediate higher-chlorinated congeners.
The HpCDD and HxCDD congeners observed as degradation products were present in
proportions similar to the number of congeners in each congener group. However, the
investigators observed greater yields of 2,3,7,8-TCDD and 1,2,3,7,8-PeCDD than would be
expected on the basis of the number of potential TCDD and PeCDD congeners. One-fifth to
one-third of the total yield of PeCDDs was 1,2,3,7,8-PeCDD, and one-half of the total yield of
TCDDs was 2,3,7,8-TCDD.
Kieatiwong et al. (1990) performed experiments similar to those of Miller et al. (1989)
using natural sunlight rather than sunlamps for irradiation of 13C-labeled OCDD on soils.
Photodechlorination was estimated to account for approximately 10% of the loss of OCDD.
One-third to one-half of the total yield of PeCDDs was 1,2,3,7,8-PeCDD, and one-half of the
total yield of TCDDs was 2,3,7,8-TCDD. These findings, along with those of Miller et al.,
indicate that the 2,3,7,8-substituted TCDD and PeCDD congeners were either preferentially
formed or were photochemically less reactive than the other congeners that were formed.
Tysklind et al. (1992) studied the sunlight photolysis of OCDD on soil and reported
results similar to those of Miller et al. (1989) and Kieatiwong et al. (1990). Photodechlorination
was observed with production of HpCDDs, HxCDDs, PeCDDs, and TCDDs over the 16-day
irradiation period. Photodechlorination at the peri-substituted positions was the preferred
photodechlorination mechanism; the proportions of 2,3,7,8-substituted congeners present in the
soils after 16 days for each congener group were as follows: HxCDD, 65%; PeCDD, 40%; and
TCDD, 75%. Tysklind et al. (1992) also studied the sunlight photolysis of OCDF on soil.
Photodechlorination was observed; however, unlike the case with OCDD, photodechlorination of
the lateral-substituted positions was found to be the dominant photodechlorination mechanism,
resulting in a relative decreasing proportion of 2,3,7,8-substituted congeners during the
irradiation period. 2,3,7,8-TCDF was not observed in any of the irradiated samples.
9.2.4. Photolysis on Vegetation
Photolysis of CDDs/CDFs sorbed on the surface of vegetation has not been well
characterized, and the findings to date are somewhat contradictory. McCrady and Maggard
(1993) reported that 2,3,7,8-TCDD sorbed on the surface of reed canary grass (Phalaris
arundinacea L.) underwent photolytic degradation, with a half-life of 44 hr in natural sunlight.
In contrast, Welsch-Pausch et al. (1995) found little difference in the CDD/CDF congener
9-12
-------
patterns between grass (Lolium multiflorum) grown on an outdoor plot and grass grown in a
greenhouse (i.e., UV light transmission blocked).
In an attempt to clarify this contradiction, Welsch-Pausch and McLachlan (1995) studied
the photodegradation of CDDs/CDFs on pasture grass (Arrhenatherion elatioris) during two
growing cycles (summer and autumn) using two greenhouses. One greenhouse was constructed
of glass that blocks UV transmission and the other was constructed of plexiglass (4 mm) with a
UV light transmission of greater than 50% in the 280 to 320 nm range. In both the summer and
the autumn exposure periods, the concentrations of CDDs/CDFs (on a congener-group basis)
were similar in the grass exposed to UV light and the grass that was not exposed. The authors
concluded that if photodegradation was occurring, it was a relatively insignificant factor in the
accumulation of CDDs/CDFs in pasture grass.
9.2.5. Photolysis in Air
Photolysis of CDDs/CDFs in the atmosphere has not been well characterized. On the
basis of data generated to date, however, photolysis appears to be a significant mechanism for
degradation (principally, dechlorination of the peri-substituted chlorines) of those CDDs/CDFs
present in the atmosphere in the gas phase. For airborne CDDs/CDFs sorbed to particulates,
photolysis appears to proceed very slowly, if at all. Because of the low volatility of CDDs/CDFs,
few studies have been attempted to measure actual rates of photodegradation of gas-phase
CDD/CDF, and only recently have studies examined the relative importance of photolysis to
particulate-bound CDDs/CDFs.
Sivils et al. (1994, 1995) studied the gas-phase photolysis of several CDDs (2,3,7-
TrCDD; 2,3,7,8-TCDD; 1,2,3,4-TCDD; 1,2,3,7,8-PeCDD, and 1,2,4,7,8-PeCDD) by irradiating
the effluent from a gas chromatograph with broadband radiation in the UV/visible region for
periods of up to 20 min. The irradiated sample was then introduced into a second gas
chromatograph to measure the extent of dechlorination. The results showed that degradation
followed first-order kinetics and that an inverse relationship existed between the degree of
chlorination and the rate of disappearance. Although the lack of photoproducts prevented an
independent confirmation of the preferential loss mechanism, the results indicate that laterally
substituted congeners (chlorines at the 2, 3, 7, and 8 positions) degrade at a slower rate than do
the peri-substituted congeners (chlorines at the 1, 4, 6, and 9 positions). Although Sivils et al.
(1994) did not present the rate constants, the degradation rate for 2,3,7,8-TCDD (30% loss in
20 min) was reported to be slower than the rates for all other tested CDDs. Also, 1,2,4,7,8-
PeCDD (with two perichlorines) degraded significantly faster than did 1,2,3,7,8-PeCDD (with
only one perichlorine).
9-13
-------
Mill et al. (1987) studied the photolysis of 2,3,7,8-TCDD sorbed onto small-diameter fly
ash particulates suspended in air. The results indicated that fly ash confers photostability on
2,3,7,8-TCDD. Little (8%) to no loss was observed on the two fly ash samples after 40 hr of
illumination. Tysklind and Rappe (1991) and Koester and Kites (1992) reported similar results
of photolysis studies with fly ash. Tysklind and Rappe subjected fly ash from two German
incinerators to various simulated environmental conditions. The fraction of photolytically
degradable CDD/CDF after 288 hr of exposure was in the range of 20 to 40% of the extractable
CDD/CDF. However, a 10 to 20% reduction was also observed in the darkened control samples.
With the exception of HpCDD and HpCDF, the concentration of all other congener groups either
increased or stayed the same during the exposure period from hour 144 to hour 288.
Koester and Kites (1992) studied the photodegradation of CDDs/CDFs naturally adsorbed
to fly ash collected from five electrostatic precipitators. They observed no significant
degradation in 11 photodegradation experiments performed on the ash for periods ranging from 2
to 6 days. The authors concluded that (a) the absence of photodegradation was not due to the
absence of a hydrogen-donor organic substance; (b) other molecules on the ash, as determined by
a photolysis experiment with an ash extract, inhibited photodegradation, either by absorbing light
and dissipating energy or by quenching the excited states of the CDDs/CDFs; and (c) the surface
of the ash itself may have hindered photolysis by shielding the CDDs/CDFs from light.
9.3. CDDs/CDFs IN BALL CLAY
9.3.1. Initial Discovery of CDD/CDF Contamination of Ball Clay
The presence of dioxin-like compounds in ball clay was discovered in 1996 as a result of
an investigation to determine the sources of relatively high levels of dioxin found in two chicken
fat samples during a national survey of poultry. The survey was conducted jointly by USD A, the
U.S. Food and Drug Administration (FDA), and EPA to assess the national prevalence and
concentrations of CDDs, CDFs, and coplanar PCBs in poultry (Ferrario et al., 1997).
The results of the investigation indicated that soybean meal added to chicken feed was the
source of dioxin contamination (Ferrario et al., 2000). Further investigation showed that the
CDD contamination came from the ball clay added to the soymeal as an anticaking agent. The
ball clay was added at approximately 0.3 to 0.5% of the soybean meal. Samples of raw ball clay
were subsequently taken at the mine of origin in Mississippi. Analysis of the samples showed
elevated levels of CDDs with a congener profile similar to the CDD profiles found in the
soymeal, chicken feed, and immature chickens.
9-14
-------
9.3.2. Characteristics of Mississippi Embayment Ball Clays
The ball clays from the mine discussed above are part of a larger ball clay resource that
spans portions of western Kentucky, Tennessee, and Mississippi. These clays were deposited
along the shores of the Mississippi embayment during the early to middle Eocene epoch, which
occurred approximately 40 to 45 million years ago. The Mississippi embayment ball clays are
secondary clays composed mainly of poorly defined crystalline kaolinite. Other minerals present
include illite, smectite, and chlorite. Quartz sand is the major nonclay mineral. These deposits
of ball clay occur in lenses surrounded by layers of sand, silt, and lignite. The clays can have a
gray appearance caused by the presence of finely divided carbonaceous particles. It is not
uncommon to find black carbonized imprints of fossil leaves and other plant debris in the clay
(Patterson and Murray, 1984).
The plasticity of ball clay makes it an important natural resource for the ceramic industry.
The breakdown of the ceramic uses of ball clay is 33% for floor and wall tile, 24% for sanitary
ware, 11% for pottery, and 32% for other industrial and commercial uses (Virta, 2000). A minor
use of ball clay was as an anticaking agent in animal feeds, but this use has been banned by the
FDA (Headrick et al., 1999). Total mining of ball clay in 1999 was 1.14 million metric tons
(Virta, 2000).
9.3.3. Levels of Dioxin-Like Compounds in Ball Clay
The joint EPA/FDA and USD A investigation of ball clay as a source of dioxin
contamination in animal feeds resulted in sampling the clay at an operational mine in western
Mississippi. Eight samples of raw (unprocessed) ball clay were collected from an open mining
pit at a depth of about 10 to 15m. Samples were prepared and analyzed by EPA using EPA
Method 1613 (Ferrario et al., 2000). The concentrations of the CDDs/CDFs present in the raw
ball clay samples from the one mine are shown in Table 9-3. The ratio of the limits of detection
to the limits of quantification for the CDDs/CDFs in the clay samples were 0.5:1 pg/g (ppt, dry
weight) for the tetras; 1:2 pg/g for the pentas, hexas, and heptas; and 5:10 pg/g for the octas. The
mean concentrations of all of the CDDs exceeded 100 ppt (dry weight).
OCDD was found at the highest concentration in all of the samples, followed by either
1,2,3,4,6,7,8-HpCDD or 1,2,3,7,8,9-HxCDD. The maximum OCDD concentration in the eight
samples was approximately 59,000 pg/g. The most toxic tetra and penta congeners were present
at unusually high concentrations in all of the samples, with average concentrations of 711 pg/g
and 508 pg/g for 2,3,7,8-TCDD and 1,2,3,7,8-PeCDD, respectively. Although the ball clays
showed elevated levels of 2,3,7,8-substituted CDDs, they showed very low levels of 2,3,7,8-
substituted CDFs. In addition, there was a consistent ratio within the HxCDD congener
9-15
-------
Table 9-3. Concentrations of CDDs (pg/g, dry weight) in eight ball clay
samples in the United States
Congener
2,3,7,8-TCDD
,2,3,7,8-PeCDD
,2,3,4,7,8-HxCDD
,2,3,6,7,8-HxCDD
,2,3,7,8,9-HxCDD
,2,3,4,6,7,8-HpCDD
OCDD
Total TEQ
Mean
711
508
131
456
2,093
2,383
20,640
Median
617
492
134
421
1,880
2,073
4,099
Minimum
253
254
62
254
1,252
1,493
8,076
Maximum
1,259
924
193
752
3,683
3,346
58,766
TEQDF-WH098
711
508
13
46
209
24
2
1,513
Source: Ferrario et al. (2000).
distribution across all samples (1,2,3,7,8,9-HxCDD was present at higher concentrations than the
other 2,3,7,8-substituted HxCDD congeners). The average percent distribution of the three
individual 2,3,7,8-hexa congeners was 5, 17, and 78%, respectively. This congener pattern was
observed in all the raw ball clay samples analyzed.
The mean total TEQDF-WHO98 for the raw ball clay was determined to be 1,513 pg/g dry
weight; 2,3,7,8-TCDD accounted for 47% of the TEQDF-WHO98, followed by 1,2,3,7,8-PeCDD
at 34%. As expected, even though present at the highest concentration, OCDD contributed less
than 1% percent of the total TEQDF-WHO98 due to its relatively small WHO-TEF. In
comparison, the typical ranges of background TEQDF-WHO98 concentrations in North American
urban and rural surface soil samples were found to be 2 to 21 pg/g and 0.1 to 6 pg/g, respectively
(U.S. EPA, 2000c). In soil samples, all 2,3,7,8-CDD/CDF congeners were detected, and 2,3,7,8-
TCDD represented less than 1% of total CDD/CDF present. The most prevalent congeners in
soils were OCDD, followed by OCDF. Table 9-4 compares the mean CDD/CDF congener group
concentrations in ball clay with those in rural and urban background soils. This comparison
indicates there are few similarities between the ball clay and soils in the congener group
distributions.
9.3.4. Evidence for Ball Clay as a Natural Source
Several lines of evidence suggest that dioxin-like compounds in ball clay are of natural
origin. The clay samples were obtained from undisturbed deposits. It is unknown how human
activity could have contaminated these deposits without disturbing them. The EPA laboratory in
Athens, GA, analyzed the Mississippi mine clays using a broad screen for anthropogenic
contaminants and no compounds were found outside of the normal range (Ferrario et al., 2000).
9-16
-------
Table 9-4. Comparison of the mean CDD/CDF congener group
concentrations in ball clay with those in urban and rural soils in North
America (pg/g, dry weight)
Congener group
TCDD
TCDF
PeCDD
PeCDF
HxCDD
HxCDF
HpCDD
HpCDF
OCDD
OCDF
Total CDD/CDF
Mean concentration
Raw ball clay
3,729
6
4,798
2
6,609
6
6,194
9
11,222
11
32,586
Urban background soil
36.1
23.5
18.1
40.8
31.7
23.5
194.4
46.4
2,596
40.2
3,050.7
Rural background soil
2.3
6.8
4.1
12.7
22.7
21.9
114.7
37.3
565.1
33.5
821.1
Sources: Adapted from U.S. EPA (2000c); Ferrario et al. (2000)
All known anthropogenic sources of dioxin have associated with them a wide variety of other
contaminants. The absence of elevated levels of other compounds is strong evidence that the
dioxins found in the clay were not the result of waste disposal.
The congener profiles of ball clay do not match those of known anthropogenic sources.
Cleverly et al. (1997) reported on the congener profiles that are typical of known anthropogenic
sources of dioxin-like compounds in the United States. The results of this study, presented
below, were used as a basis of comparison for the profiles of raw ball clay.
The congener pattern characteristic of waste combustion sources differs significantly
from the ball clay profile in several aspects. In combustion source emissions, all 2,3,7,8-
substituted CDD and CDF congeners are measured, and 2,3,7,8-TCDD is usually 0.1 to 1% of
total CDD/CDF mass emitted. In ball clay, 2,3,7,8-TCDD is approximately 5% of total mass of
dioxins present. As with the ball clay, the most prevalent 2,3,7,8-Cl-substituted CDD congeners
in most incinerator emissions are OCDD and 1,2,3,4,6,7,8-HpCDD; however, combustion
emissions contain appreciable amounts of CDFs, of which the 1,2,3,4,6,7,8-HpCDF, OCDF,
1,2,3,4,7,8-HxCDF, 2,3,7,8-TCDF, and 2,3,4,6,7,8-HxCDF congeners dominate.
The combustion of wood generates a congener profile not unlike that of waste
combustion (i.e., the ratio of CDD:CDF is <1), and all laterally substituted congeners can be
detected in emissions. The combustion of tree bark produces a congener profile in which the
9-17
-------
CDD:CDF ratio is >1, showing only minimal and barely detectable levels of CDFs in the smoke,
the exception being that 2,3,7,8-TCDF is present at approximately 2% of total mass. The
dominant congener in tree bark combustion emissions is OCDD (>30% total CDD/CDF mass),
followed by 1,2,3,4,6,7,8-HpCDD and 1,2,3,7,8,9-HxCDD.
The congener profile of 2,4-D salts and esters seems to mimic a combustion source
profile in the number of congeners represented and in the minimal amount of 2,3,7,8-TCDD
relative to all 2,3,7,8-Cl-substituted congeners. Nevertheless, unlike the combustion source
profile, the 1,2,3,7,8-PeCDD and the 1,2,3,4,6,7,8-HpCDF constitute major fractions of total
CDD/CDF contamination present in 2,4-D. The congener profile of technical-grade PCP is
clearly dominated by OCDD and 1,2,3,4,6,7,8-HpCDD; however, only trace amounts of 2,3,7,8-
TCDD are detected in PCP, and 1,2,3,4,6,7,8-HpCDF and OCDF constitute roughly 15% of
typical formulations.
Metal smelting and refining processes, such as secondary aluminum, copper, and lead
smelting, also have all the 2,3,7,8-Cl-substituted CDD/CDF congeners in stack emissions. In
secondary aluminum smelting, 2,3,7,8-TCDD is less than 0.1% of total CDDs/CDFs, whereas
PeCDF is nearly 25% of total emissions of dioxin-like compounds, and the CDD/CDF ratio is
<1. Secondary copper operations show a similar pattern of CDD/CDF emissions, but with six
compounds dominating emissions: 1,2,3,4,7,8-HxCDF; 1,2,3,6,7,8-HxCDF; 1,2,3,4,6,7,8-
HpCDF; OCDF; OCDD; and 1,2,3,4,6,7,8-HpCDD. In iron ore sintering, the dominant congener
in emissions of 2,3,7,8-Cl-substituted compounds is 2,3,7,8-TCDF.
A number of studies have shown that natural processes can produce chlorinated aromatic
compounds, including dioxin-like compounds. Gribble (1994) reviewed the biological
production of a wide variety of halogenated organic compounds in nature. The Mississippi salt
marsh grass "needlerush" (Juncus roemerianus) contains the aromatic compound 1,2,3,4-
tetrachlorobenzene, and the blue-green alga Anacystis marina naturally contains chlorophenol.
The soil fungus Penicillium sp. produces 2,4-dichlorophenol, and the common grasshopper is
known to secrete 2,5-dichlorophenol.
Urhahn and Ballschmiter (1998) also provide a good review of the chemistry of the
biosynthesis of chlorinated organic compounds under natural conditions. It has been
hypothesized that CDDs, CDFs, and other chlorinated aromatic compounds can be naturally
formed from halogenated humic substances, and halomethanes can be formed through
chloroperoxidase-mediated reactions in undisturbed peat bogs (Silk et al., 1997). A similar
chloroperoxidase-mediated biochemical formation of CDDs/CDFs from chlorophenols was
achieved under laboratory conditions by Oberg and Rappe (1992).
It has been observed that chlorophenols can be biosynthesized (Gribble, 1994; Silk et al.,
1997), and that chorophenols are readily adsorbed into peat-bentonite mixtures (Viraraghavan
9-18
-------
and Slough, 1999). Hoekstra et al. (1999) offers the hypothesis that 2,3,7,8-TCDD, 1,2,3,7,8-
PeCDD, and 1,2,3,7,8,9-HxCDD can be naturally formed in soils of coniferous forests from
chlorinated phenol. These same congeners are also the predominant congeners in the ball clay
from the Mississippi embayment. Although none of these natural processes can be directly
connected with the presence of dioxin in ball clay, the existence of such mechanisms lends
plausibility to a hypothesis that they are of natural origin.
CDDs/CDFs have been found in other clays quite distant from Mississippi embayment
ball clay deposits. No evidence of anthropogenic sources have been discovered in these areas
either. The presence of CDDs has been discovered in kaolinitic clay mined in Germany (Jobst
and Aldag, 2000). Because no anthropogenic source could be determined to explain the presence
and levels of CDDs in the ball clay, the authors speculated that they were the result of an
unknown geologic process. In addition, the German clay also has a congener profile similar to
that observed in the Mississippi ball clay, with an absence of CDFs at comparable concentrations
and the predominance of 1,2,3,7,8,9-HxCDD among the toxic hexa-CDDs. The similarity in the
congener profiles in ball clay mined in the United States and Germany suggests a common origin
to the CDDs present in these clays (Ferrario et al., 2000).
In summary, no anthropogenic sources have been identified that explain the levels and
profiles of CDDs/CDFs present in ball clay. On the other hand, no definitive scientific evidence
has been brought forward that identifies the principal chemical and physical mechanism involved
in the selective chemical synthesis of CDDs under the conditions inherent in the formation of
ball clays some 40 million years ago.
9.3.5. Environmental Releases of Dioxin-Like Compounds from the Mining and Processing
of Ball Clay
In 1995, approximately 993 million kg of ball clay was mined in the United States (Virta,
2000). Multiplication of the mean TEQDF-WHO98 concentration in mined ball clay by the total
amount of ball clay mined in 1995 gives an estimate of 1,502 g TEQDF-WHO98 contained in all
the ball clay mined in 1995. It is unknown whether any of these CDDs are released to the
environment during the mining, initial refining, and product handling. As discussed above, most
ball clay is used to produce ceramics through a process of high-temperature vitrification. The
temperatures found in ceramic kilns are well above the levels needed for both volatilization and
destruction of CDDs. Despite these high temperatures, it is unclear whether some release occurs,
and no stack measurements have yet been made. Therefore, insufficient evidence is available to
make even a preliminary estimate of releases, and this activity is classified as a Category E
source.
9-19
-------
10. SOURCES OF DIOXIN-LIKE POLYCHLORINATED BIPHENYLS (PCBs)
The purpose of this chapter is twofold: (1) to identify sources that release dioxin-like
PCB congeners into the environment and (2) to derive national estimates for releases from these
sources in the United States. PCBs have been found in all media and in all parts of the world.
PCBs were manufactured in relatively large quantities for use in commercial products such as
dielectrics, hydraulic fluids, plastics, coatings and paints, and although PCBs are no longer
commercially produced in the United States, they continue to be released to the environment
through the use and disposal of these products. PCBs may also be inadvertently produced as by-
products during the manufacture of certain organic chemicals and also as products of the
incomplete combustion of some waste materials.
10.1. GENERAL FINDINGS OF THE EMISSIONS INVENTORY
Table 10-1 provides a compilation of known or suspected dioxin-like PCB-emitting
source categories in the United States for which emission measurements of dioxin-like PCB
congeners, Aroclors, or PCB congener groups have been reported in government, industry, and
trade association reports; conference proceedings and journal articles; and comments submitted
to EPA on previous versions of this document. The intent of Table 10-1 is to clearly identify
those source categories and media (air, water, land, and products) for which the available data are
adequate for reliably quantifying emissions of dioxin-like PCBs and those for which the data are
inadequate.
Nationwide emission estimates for the United States inventory are presented in
Table 10-2 (emissions to air, water, land, and product) for those source categories for which
estimates can be reliably quantified (the category has been assigned a confidence rating of A, B,
or C) (see Section 1.2.3 for details on confidence ratings). Table 10-2 also lists preliminary
estimates of the potential magnitude of emissions from "unquantified" sources (i.e., sources
assigned a confidence rating of D) in reference year 2000. Because of large uncertainties for
these Category D estimates, they are not included in the quantitative inventory.
Currently, no significant releases of newly formed dioxin-like PCBs are occurring in the
United States. Unlike CDDs/CDFs, PCBs were intentionally manufactured in the United States
in large quantities from 1929 until production was banned in 1977. Releases to the environment
of "old" dioxin-like PCBs (dioxin-like PCBs manufactured prior to the production ban) can
occur from ongoing use and disposal practices. Prior to regulations enacted beginning in the late
1970s that limited the manufacture/use/disposal of PCBs, significant quantities were released to
the environment in association with (a) the manufacture of PCBs, (b) the manufacture of
products containing PCBs, and (c) the use and disposal of products containing PCBs as well as
10-1
-------
Table 10-1. Confidence rating classes for 2000 for releases from all known and
suspected source categories of dioxin-like PCBsa
Source category
Approved PCB disposal
Accidental PCB releasees
Municipal wastewater treatment sludge
Municipal waste combustion
Industrial wood combustion
Medical waste incineration
Tire combustion
Cigarette combustion
Sewage sludge incineration
Backyard barrel burning
Petroleum refining catalyst regeneration
Air
E
E
E
E
E
E
D
C
E
E
Land
E
A
Water
E
a Blank cells mean not applicable or no data.
A= Characterization of the source category judged to be adequate for quantitative estimation with high confidence in
the emission factor and high confidence in the activity level.
C= Characterization of the source category judged to be adequate for quantitative estimation with low confidence in
the emission factor and/or the activity level.
D= These are preliminary indications of the potential magnitude of emissions from "unqualified" sources in
Reference Year 1995. These estimates were assigned a "confidence category" rating of D and are not included in
the Inventory.
E= Not quantifiable.
materials that may have been contaminated with trace levels of PCBs from prior PCB use or
disposal. Following the ban on PCB production, releases from these first two categories ceased.
The third type of releases, those associated with product use and disposal, will continue in at
least four ways:
1. Disposal of products containing greater than 2 Ib of PCBs (e.g., dielectric fluids in
transformers and large capacitors), which is controlled by disposal regulations that
have minimized environmental releases;
2. Disposal of products containing small quantities of PCBs (e.g., small capacitors,
fluorescent lighting fixtures) or trace quantities of PCBs (e.g., wastepapers), which is
subject to disposal as municipal solid waste but which may result in some release to
the general environment;
10-2
-------
Table 10-2. Inventory of contemporary releases of dioxin-like PCBs in the United States for 1987,1995, and
2000 and preliminary release estimates of dioxin-like PCBs for 2000 (g TEQp-WHO98/yr)
Emissions source category
2000 Inventory
A
B
C
1995 Inventory
A
B
C
1987 Inventory
A
B
C
Preliminary
estimate for
2000
D
Releases to air
Combustion sources
Cigarettes
Sewage sludge incineration
Total quantified releases to air
0.7
0.7
1.1
1.1
0.4
0.4
0.01
0.01
Releases to land
Municipal sludge (land
application and farming)
Total quantified releases to land
18.8
18.8
77.4
77.4
51.1
51.1
Releases to products
Municipal sludge as soil
ammendment
Total quantified releases to
products
0.5
0.5
2.0
2.0
1.7
1.7
A= Characterization of the source category judged to be adequate for quantitative estimation with high confidence in the emission factor and high confidence in
the activity level.
C= Characterization of the source category judged to be adequate for quantitative estimation with low confidence in the emission factor and/or the activity level.
D= These are preliminary indications of the potential magnitude of emissions from "unquantified" sources in Reference Year 1995. These estimates were
assigned a "confidence category" rating of D and are not included in the Inventory.
E= Not quantifiable.
-------
3. Leaks and spills of still-in-service PCBs; and
4. Illegal disposal of PCBs.
Although it has been demonstrated that small quantities of dioxin-like PCBs can be
emitted into the air during waste combustion, no strong evidence exists that they are emitted in
significant quantities as by-products during combustion. The widespread occurrence of dioxin-
like PCBs in the U.S. environment most likely reflects past releases associated with PCB
production, use, and disposal. Further support for this finding is based on observations of
reductions since the 1980s in PCB concentrations in Great Lakes sediment and in other areas.
10.2. RELEASES OF COMMERCIAL PCBs
PCBs were commercially manufactured by the direct batch chlorination of molten
biphenyl with anhydrous chlorine in the presence of a catalyst, followed by separation and
purification of the desired chlorinated biphenyl fractions. The degree of chlorination was
controlled by the chlorine contact time in the reactor. Commercial PCB production is believed to
have been confined to 10 countries. Total PCBs produced worldwide since 1929 (the first year
of known production) has been estimated at 1.5 million metric tons.
Initially, PCBs were used primarily as dielectric fluids in transformers. After World War
n, PCBs found steadily increasing use as dielectric fluids in capacitors, as heat-conducting fluids
in heat exchangers, and as heat-resistant hydraulic fluids in mining equipment and vacuum
pumps. PCBs also were used in a variety of "open" applications (i.e, uses from which PCBs
cannot be recollected), including plasticizers, carbonless copy paper, lubricants, inks, laminating
agents, impregnating agents, paints, adhesives, waxes, additives in cement and plaster, casting
agents, dedusting agents, sealing liquids, fire retardants, immersion oils, and pesticides (DeVoogt
andBrinkman, 1989).
U.S. production peaked in 1970, with a volume of 39,000 metric tons. In 1971,
Monsanto Corporation, the major U.S. producer, voluntarily restricted the sale of PCBs for all
applications, with the exception of "closed electrical systems." Annual production fell to 18,000
metric tons in 1974. Monsanto ceased PCB manufacture in mid-1977 and shipped the last
inventory in October of that year. Regulations issued by EPA beginning in 1977, principally
under the Toxic Substances Control Act (TSCA) (40 CFR 761), have strictly limited the
production, import, use, and disposal of PCBs. The estimated cumulative production and
consumption volumes of PCBs in the United States from 1930 to 1975 were 635.03 million kg
produced, 1.36 million kg imported (primarily from Japan, Italy, and France), 568.35 million kg
10-4
-------
sold in the United States, and 68.04 million kg exported (ATSDR, 1993; DeVoogt and
Brinkman, 1989). The reliability of these values is +5% and -20% (Versar, Inc., 1976).
Monsanto Corporation marketed technical-grade mixtures of PCBs primarily under the
trade name Aroclor. The Aroclor mixtures are identified by a four-digit numbering code in
which the last two digits indicate the chlorine content by weight percent. The exception to this
coding scheme is Aroclor 1016, which contains only mono- through hexachlorinated congeners
with an average chlorine content of 41%. From 1957 until 1972, Monsanto also manufactured
several blends of PCBs and polychlorinated terphenyls (PCTs) under the trade names Aroclor
2565 and Aroclor 4465; manufacture and sales volumes are not available for these blends. Listed
below are the percentages of total Aroclor production during the years 1957 to 1977 by Aroclor
mixture, as reported by Brown (1994).
1957-1977
U.S. production
Aroclor (%)
1016 12.88
1221 0.96
1232 0.24
1242 51.76
1248 6.76
1254 15.73
1260 10.61
1262 0.83
1268 0.33
The trade names of the major commercial PCB technical-grade mixtures manufactured in
other countries included Clophen (Germany), Fenclor and Apirolio (Italy), Kanechlor (Japan),
Phenoclor and Pyralene (France), Sovtel (USSR), Delor and Delorene (Czechoslovakia), and
Orophene (German Democratic Republic) (DeVoogt and Brinkman, 1989). The mixtures
marketed under these trade names had similar chlorine content (by weight percent and average
number of chlorines per molecule) to those of various Aroclor mixtures. Listed below are
comparable mixtures in terms of chlorine content marketed under several trade names.
Aroclor Clophen Pyralene Phenoclor Fenclor Kanechlor
1232 2000 200
1242 A-30 3000 DP-3 42 300
1248 A-40 DP-4 400
1254 A-50 DP-5 54 500
1260 A-60 DP-6 64 600
10-5
-------
Major advances in analytical separation and resolution techniques beginning in the 1970s
enabled various researchers to identify and quantify PCB congeners present in Aroclors,
Clophens, and Kanechlors (Jensen et al., 1974; Albro and Parker, 1979; Huckins et al., 1980;
Albro et al., 1981; Duinker and Hillebrand, 1983; Kannan et al., 1987; Tanabe et al., 1987;
Duinker et al., 1988; Schulz et al., 1989; Himberg and Sippola, 1990; Larsen et al., 1992; deBoer
et al., 1993; Schwartz et al., 1993; Frame et al., 1996a, b; Frame, 1997). Schulz et al. (1989)
were the first to identify and quantify all PCB congeners present in a series of Aroclors and
Clophens. Frame (1995) reported preliminary results of a nearly completed round robin study,
one goal of which was to determine the distribution of all PCB congeners above 0.05 weight
percent in various Aroclors (1221, 1016, 1242, 1260, and 1262) using 18 state-of-the-art gas
chromatography/mass spectrometry (GC/MS) or electron capture detector (GC/ECD) systems.
Table 10-3 presents mean summary statistics on the concentrations of the dioxin-like
PCBs in each mixture group (e.g., Aroclor 1248, Clophen A-40, and Kanechlor 400 are in one
mixture group) reported by these researchers. Table 10-3 also presents the mean TEQ
concentration of each congener in each mixture group as well as the total mean TEQ
concentration in the mixture group. Because of the wide variability in the reported results, the
uncertainty associated with these mean concentrations is very large.
For each mixture group, the congeners detected were generally similar. There was,
however, wide variability in the concentrations reported by some researchers for some congeners.
Brown et al. (1995) compiled similar statistics using a somewhat different set of studies and
derived significantly lower mean concentrations of some congeners in several Aroclors. Frame
(1995) and Larsen (1995) attributed such differences to either potential limitations in the GC
columns used by various researchers to separate similar eluting congeners or actual differences in
the congener concentrations in the Aroclor, Clophen, and Kanechlor lots analyzed by various
research groups.
The congener distributions also vary among the different mixtures. Therefore, the
calculated TEQs also vary. The congener distributions for various lots of Aroclor 1254, and the
corresponding TEQs, are presented in another study (Frame, 1999) in which the relative TEQs
for late production lots were reported to be much higher than those for the earlier production lots;
however, the late production lots were estimated to account for only about 1% of the total
production volume of Aroclor 1254. Therefore, the data for the later production lots were not
included in the average TEQ calculation for Aroclor 1254 in Table 10-3.
In the environment, PCBs also occur as mixtures of congeners, but their composition
differs from those of the commercial mixtures because after release to the environment the
mixtures change over time through partitioning, chemical transformation, and preferential
bioaccumulation (U.S. EPA, 1996f). Dioxin-like PCB congeners differ by up to one to two
10-6
-------
Table 10-3. Weight percent concentrations of dioxin-like PCBs in Aroclors, Clophens, and Kanechlors
Dioxin-like PCB congener
Aroclor 1016
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
Total TEQP-WHO98
Total TEQP-WH094
Aroclor 1221
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
Total TEQP-WHO98
Total TEQp-WHO94
IUPAC
number
77
81
105
114
118
123
126
156
157
167
169
170
180
189
77
81
105
114
118
123
126
156
157
167
169
170
180
189
No. of
samples
analyzed
5
3
4
4
4
4
4
4
4
4
5
4
4
4
4
4
4
4
4
4
4
4
4
4
4
3
3
4
No. of
detections
0
0
1
0
1
0
0
0
0
0
0
0
0
0
4
1
3
0
4
0
0
0
0
0
0
0
0
0
Mean cone.
(nondetect set to
zero) (g/kg)
0
0
0.0375
0
0.0125
0
0
0
0
0
0
0
0
0
1.075
0.0875
0.3875
0
1.725
0
0
0
0
0
0
0
0
0
TEQP-WHO98 cone.
(nondetect set to
zero) (mg/kg)
0
0
0.00375
0
0.00125
0
0
0
0
0
0
0
0
0
0.005
0.005
0.1075
0.00875
0.03875
0
0.1725
0
0
0
0
0
0
0
0
0
0.328
0.749
Mean conc.a
(nondetect set to !4
detection limit)
(g/kg)
0
0
0.109
0
0.091
0
0
0
0
0
0
0
0
0
1.078
0.116
0.4
0
1.725
0
0
0
0
0
0
0
0
0
TEQP-WHO98
conc.a (nondetect
set to 1A detection
limit) (mg/kg)
0
0
0.011
0
0.009
0
0
0
0
0
0
0
0
0
0.0200
0.0200
0.108
0.012
0.04
0
0.173
0
0
0
0
0
0
0
0
0
0.333
0.752
-------
Table 10-3. Weight percent concentrations of dioxin-like PCBs in Aroclors, Clophens, and Kanechlors (continued)
Dioxin-like PCB congener
Aroclor 1242, Clophen
A-30, and Kanechlor 300
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
Total TEQP-WHO98
Total TEQp-WHO94
Aroclor 1248, Clophen
A-40, and Kanechlor 400
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
Total TEQP-WHO98
Total TEQP-WH094
IUPAC
number
77
8
105
114
118
123
126
156
157
167
169
170
180
189
77
81
105
114
118
123
126
156
157
167
169
170
180
189
No. of
samples
analyzed
15
7
11
8
9
9
14
9
8
8
14
6
5
7
13
6
9
7
8
7
11
8
7
7
12
5
4
6
No. of
detections
15
6
11
5
9
7
8
8
2
2
2
2
2
0
13
4
8
6
8
7
6
8
3
3
3
4
4
1
Mean cone.
(nondetect set to
zero) (g/kg)
3.3
1.09
4.02
1.13
8.04
1.12
0.049
0.39
0.021
0.021
0.000013
0.19
0.16
0
4.36
1.76
10.12
3.39
20.98
1.48
0.11
1.13
0.19
0.16
0.01
0.96
1.24
0.0018
TEQP-WHO98 cone.
(nondetect set to
zero) (mg/kg)
0.33
0.11
0.4
0.57
0.8
0.11
4.94
0.2
0.011
0.00021
0.00013
0
0
0
7.47
8.70
0.44
0.18
1.01
1.69
2.1
0.15
10.55
0.56
0.09
0.0016
0.1006
0
0
0.0001833
16.87
18.55
Mean conc.a
(nondetect set to !4
detection limit)
(g/kg)
3.301
1.089
4.024
1.201
8.044
1.157
0.094
0.424
0.096
0.096
0.048
0.244
0.218
0
4.36
1.77
10.12
3.4
20.98
1.48
0.14
1.13
0.2
0.16
0.041
0.97
1.24
0.06
TEQP-WH098
conc.a (nondetect
set to 1A detection
limit) (mg/kg)
0.33
0.109
0.402
0.601
0.804
0.116
9.404
0.212
0.048
0.001
0.476
0
0
0
12.5
13.74
0.44
0.18
1.01
1.7
2.1
0.15
13.51
0.56
0.1
0.0016
0.41
0
0
0.006
20.16
21.83
o
oo
-------
Table 10-3. Weight percent concentrations of dioxin-like PCBs in Aroclors, Clophens, and Kanechlors (continued)
Dioxin-like PCB congener
Aroclor 1254, Clophen
A-50, and Kanechlor 500
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
Total TEQP-WHO98
Total TEQp-WHO94
Aroclor 1260, Clophen
A-60, and Kanechlor 600
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
Total TEQP-WHO98
Total TEQP-WH094
IUPAC
number
77
81
105
114
118
123
126
156
157
167
169
170
180
189
77
81
105
114
118
123
126
156
157
167
169
170
180
189
No. of
samples
analyzed
15
6
12
9
11
8
14
10
9
10
14
8
7
7
15
6
11
9
11
8
14
11
8
10
14
8
7
8
No. of
detections
12
1
11
6
11
8
12
10
8
9
6
8
7
2
6
1
10
4
10
1
7
11
8
9
5
8
7
8
Mean cone.
(nondetect set to
zero) (g/kg)
0.8
7.85
35.83
12.17
81.65
4.59
0.99
11.08
1.91
2.74
0.08
5.06
5.79
0.045
125.94
126.04
0.13
0.08
1.59
0.71
9.51
0.0005
1.81
6.89
1.59
2.87
0.16
32.94
82.61
1.74
TEQP-WHO98 cone.
(nondetect set to
zero) (mg/kg)
0.0795
0.79
3.58
6.08
8.17
0.46
99.46
5.54
0.95
0.0274
0.8
0
0
0.0045429
0.01256
0.0075
0.16
0.35
0.95
0.00005
180.89
3.45
0.79
0.03
1.64
0
0
0.1739792
188.45
192.62
Mean conc.a
(nondetect set to !4
detection limit)
(g/kg)
0.83
7.86
35.83
12.23
81.65
4.59
1.02
11.08
1.93
2.74
0.12
5.06
5.79
0.13
0.17
0.1
1.59
0.77
9.51
0.08
1.84
6.89
1.59
2.87
0.19
32.94
82.61
1.74
TEQP-WH098
conc.a (nondetect
set to 1A detection
limit) (mg/kg)
0.08
0.79
3.58
6.11
8.17
0.46
101.7
5.54
0.97
0.03
1.23
0
0
0.013
128.67
128.78
0.017
0.01
0.16
0.39
0.95
0.008
183.82
3.45
0.79
0.03
1.92
0
0
0.17
191.71
195.89
-------
o
o
Table 10-3. Weight percent concentrations of dioxin-like PCBs in Aroclors, Clophens, and Kanechlors (continued)
Calculated for a congener only when at least one sample contained detectable levels of that congener.
Sources: Adapted from Schulz et al. (1989); Duinker and Hillebrand (1983; deBoer et al. (1993); Schwartz et al. (1993); Larsen, et al. (1992); Kannan et al.
(1987); Huckins et al. (1980); Albro and Parker (1979; Jensen et al. (1974); Albro et al. (1981); Duinker et al. (1988); Tanabe et al. (1987); Himberg and
Sippola (1990); Frame et al. (1996a, b); Frame (1997).
-------
orders of magnitude in their water solubility, vapor pressure, Kow value, and Henry's Law
constant. Thus, although all the dioxin-like PCB congeners are poorly soluble in water and have
very low vapor pressures, they will volatilize and leach at different rates. Similarly, because the
congeners differ somewhat in their rates of biodegradation, bioaccumulation, and
photodegradation, the congener patterns found in environmental media and biota will vary from
those found in commercial mixtures.
Although environmental mixtures are often characterized in terms of Aroclors, this
characterization can be both imprecise and inappropriate. Qualitative and quantitative errors can
arise from judgements in comparing GC/MS peaks for a sample with the characteristic peak
patterns for different Aroclors, particularly for environmentally altered patterns (U.S. EPA,
1996f). For the same reason, it can be both imprecise and inappropriate to infer concentrations
of dioxin-like PCB congeners in an environmental sample on the basis of characterization of the
sample's Aroclor content and knowledge of the dioxin-like congener content in the commercial
Aroclor. Safe (1994) wrote, "Regulatory agencies and environmental scientists have recognized
that the composition of PCBs in most environmental extracts does not resemble the compositions
of the commercial product." Similarly, ATSDR (1993) stated, "It is important to recognize that
the PCBs to which people may be exposed are likely to be different from the original PCB source
because of changes in congener and impurity composition resulting from differential partitioning
and transformation in the environment and differential metabolism and retention."
10.2.1. Approved PCB Disposal/Destruction Methods
In 1978, EPA began regulating the disposal of PCBs and PCB-contaminated waste under
TSCA, PL 94-469. The disposal regulations, published in the Code of Federal Regulations, 40
CFR, Part 761, state that the preferred disposal method is incineration at 1,200°C or higher. If
the waste contains material that cannot be destroyed by incineration, EPA clearance must be
obtained to dispose of the waste in a chemical waste landfill or by another approved manner.
The PCB disposal regulations describe disposal of three distinct types of PCB waste:
PCBs, PCB articles (items containing PCBs), and PCB containers. Within these categories,
further distinctions are made on the basis of the PCB concentration in the waste, with the
acceptable disposal methods being based on the concentrations in the specific waste to be
destroyed. The acceptable disposal methods are Annex I incinerators, high-efficiency boilers,
Annex II chemical waste landfills, and other approved methods. The following paragraphs and
Table 10-4 provide brief descriptions of these disposal methods. More complete descriptions of
the specific methodologies are provided in 40 CFR, Part 761.
10-11
-------
Table 10-4. Disposal requirements for PCBs and PCB items
PCBs/items
PCBs
PCB articles
Waste characterization
Mineral oil dielectric fluids from
PCB transformers
Mineral oil dielectric fluids from
PCB-contaminated transformers
PCB liquid wastes other than
mineral oil dielectric fluid
Nonliquid PCB wastes (e.g.,
contaminated materials from
spills)
Dredged materials and municipal
sewage treatment sludges
containing PCBs
Transformers
PCB capacitors'1
PCB hydraulic machines
Those analyzing >500 ppm
PCB
Those analyzing 50-500
ppm PCB
Those analyzing >500 ppm
PCB
Those analyzing 50-500
ppm PCB
PCB transformers
PCB contaminated
transformers
Those containing > 1,000
ppm PCB
Those containing < 1,000
ppm PCB
Disposal requirements
Annex I incinerator3
Annex I incinerator
High-efficiency boiler (40 CFR 761.10(a)(2)(iii))
Other approved incineratorb
Annex II chemical waste landfill0
Annex I incinerator
Annex I incinerator
High-efficiency boiler (40 CFR 761.10(a)(2)(iii))
Other approved incineratorb
Annex II chemical waste landfill0
Annex I incinerator
Annex II chemical waste landfill
Annex I incinerator
Annex II chemical waste landfill
Other approved disposal method, 40 CFR
761.10(a)(5)(iii)
Annex I incinerator
Drained and rinsed transformers may be disposed of
in Annex II chemical waste landfill
Disposal of drained transformers is not regulated
Annex I incinerator
Drained and rinsed machines may be disposed of as
municipal solid waste or salvaged
Drained machines may be disposed of as municipal
solid waste or salvaged
o
to
-------
Table 10-4. Disposal requirements for PCBs and PCB items (continued)
PCBs/items
PCB containers
Waste characterization
Other PCB articles
Those used to contain only PCBs
at a concentration <500 ppm
Other PCB containers
Those containing PCB fluids
Those not containing PCB
fluids
Disposal requirements
Drained machines may be disposed of by Annex I or
Annex II
Annex I incinerator or Annex II chemical waste
landfill
As municipal solid waste provided any liquid PCBs
are drained prior to disposal
Annex I incinerator
Annex II, provided any liquid PCBs are drained
prior to disposal
Decontaminate per Annex IV
o
oo
aAnnex I incinerator is defined in 40 CFR 761.40.
bRequirements for other approved incinerators are defined in 40 CFR 761.10(e).
"Annex II chemical waste landfills are described in 40 CFR 761.41. Annex II disposal is permitted if the PCB waste contains less than 500 ppm PCB and is
not ignitable as per 40 CFR Part 761.41(b)(8)(iii).
dDisposal of containerized capacitors in Annex II landfills was permitted until March 1, 1981; thereafter, only Annex I incineration has been permitted.
Source: U.S. EPA(1987d).
-------
10.2.1.1. Approved Incinerators/High-Efficiency Boilers
PCB Annex I incinerators must meet the specific technical standards and criteria listed in
Annex I of EPA's PCB regulations. The minimum operating requirements for disposal of liquid
wastes are 2 sec at 1,200°C with 3% excess oxygen (measured in the stack gas) or 1.5 sec at
1,600°C with 2% excess oxygen (measured in the stack gas). Monitoring requirements, approval
conditions, and trial burn requirements are prescribed in Annex I. Operators of commercial or
industrial incinerators who intend to destroy liquid PCB wastes must demonstrate the
incineration's compliance with the Annex I requirements through a comprehensive trial burn
program. Annex I incinerators operating at optimum performance level should destroy 99.997%
of liquid PCB waste, with a resulting maximum emission factor of 0.03 g/kg.
Criteria for Annex I incinerators were established for the destruction of liquid PCB
wastes; however, these incinerators also may be used for disposal of nonliquid PCB items (such
as capacitors), provided that a destruction and removal efficiency of 99.9999% and a maximum
emission factor of 0.001 g/kg are met.
High-efficiency boilers may be used to destroy PCBs and PCB-contaminated waste with
PCB concentrations not exceeding 500 ppm. Conventional industrial and utility boilers may be
designated as high-efficiency boilers if they are operated under the prescribed combustion
conditions defined in the PCB disposal regulations. The PCB regulations do not specify a
minimum destruction efficiency for high-efficiency boilers; however, EPA-approved boilers
operated according to the regulations have reported destruction efficiencies in excess of 99.99%,
with a corresponding maximum emission factor of 0.1 g/kg (U.S. EPA, 1987d).
10.2.1.2. Approved Chemical Waste Landfills
Approved chemical waste landfills can be used for the disposal of some but not all PCB
wastes. PCB-contaminated materials acceptable for land disposal in an approved landfill include
PCB mixtures (e.g., certain PCB-contaminated soil/solid debris, PCB-contaminated dredged
materials, and PCB-contaminated municipal sewage sludge), PCB articles that cannot feasibly be
incinerated (e.g., drained and flushed transformers), and drained PCB containers. Written
approval must be obtained from EPA in order to landfill PCB articles other than transformers.
PCB-contaminated materials not acceptable for land disposal in an approved landfill include
nonliquid PCB mixtures in the form of contaminated soil, rags, or other solid debris, and sealed
capacitors. Typically, PCBs disposed of in these landfills are placed in sealed containers, thereby
minimizing any PCB emissions.
10-14
-------
10.2.1.3. Other Approved Disposal Methods
Other thermal and nonthermal destruction techniques may be approved by EPA Regional
Administrators if these processes can effect a level of destruction of PCBs equivalent to that of
incinerators or boilers. After April 29, 1983, all other PCB disposal technologies (thermal and
nonthermal) used in more than one EPA Region had to be approved by EPA Headquarters.
Examples of thermal technologies approved for commercial-scale use or for research and
development projects include a pyrolysis process to treat contaminated soils, a fluid wall reactor,
a cement kiln, a diesel engine, a steam-stripping operation, an aluminum melting furnace, and a
molten salt process. Examples of approved nonthermal processes include chemical
dechlorination processes, physical/chemical extraction techniques, and biological reduction
methods. The physical/chemical techniques extract the PCBs from transformers or capacitors
and concentrate them for disposal; they do not destroy the PCBs.
10.2.2. Emission Estimates
Tables 10-5 and 10-6 list the amounts of PCBs reported in EPA's Toxics Release
Inventory (TRI) as transferred off site for treatment, energy recovery, or disposal and the amounts
released between 1988 and 2000, respectively. These quantities do not necessarily represent
entry of PCBs into the environment. If it is assumed that all transferred PCBs are incinerated in
high-efficiency boilers with a destruction and removal efficiency of 99.99%, then annual
emissions of PCBs to air during 1988, 1995, and 2000 could have been as high as 264 kg, 31 kg,
and 15 kg, respectively. Because no stack testing data are available for dioxin-like PCBs, it is
not possible to estimate what fraction of these potential PCB releases would have been dioxin-
like congeners.
10.2.3. Accidental Releases of Still-in-Service PCBs
After the 1977 ban on production of PCBs, releases of commercially produced PCBs to
the environment (aside from minimal releases occurring during approved disposal or destruction)
have been limited to accidental release of in-service PCBs (U.S. EPA, 1987d). Accidental
releases are the result of leaks or spills during failure/breakage of an existing piece of PCB-
containing equipment or of incomplete combustion during accidental fires involving PCB-
containing equipment. These two types of accidental releases are discussed in the following
sections.
10-15
-------
Table 10-5. Off-site transfers of PCBs reported in the Toxics Release
Inventory (TRI) (1988-2000)
Year
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
No. of TRI
forms filed
NA
NA
NA
NA
NA
NA
NA
16
20
26
NA
NA
122
Reported transfers (kg)
Transfers to
POTWs
102
0
0
a
0
0
0
120
0
0
0
0.5
113
Transfers for
treatment/disposal
150,888
434,666
386,903
471,319
160,802
308,347
466,948
463,385
766,638
402,535
1,181,961
2,002,237
2,642,133
Total transfers
150,990
434,666
386,903
471,319
160,802
308,347
466,948
463,505
766,638
402,535
1,181,961
2,002,237
2,642,246
Tacilities left that particular cell blank on the Form R submissions.
NA = Not available
POTWs = Publicly owned treatment works
Sources: U.S. EPA(1993f, 1995h, 1998b, 2003c).
10.2.3.1. Leaks and Spills
PCBs that remain in active service at this time are those contained in "closed systems"
(i.e., those pieces of electrical equipment that completely enclose the PCBs and do not provide
direct atmospheric access for the PCBs during normal use). This equipment includes PCB
transformers, capacitors, voltage regulators, circuit breakers, and reclosures. With the exception
of PCB transformers—and probably small PCB capacitors—the majority of the PCB-containing
electrical equipment in service during 1981 was owned by the electrical utility industry.
Approximately 70% of the estimated 140,000 PCB transformers in service in 1981 were owned
by nonutilities. No information was available on the relative distribution of small PCB
capacitors (Versar, Inc., 1988).
10-16
-------
Table 10-6. Releases of PCBs reported in the Toxics Release Inventory (TRI) (1988-2000)
Year
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
No. of TRI
forms filed
NA
NA
NA
NA
NA
NA
NA
16
20
26
NA
NA
122
Reported releases (kg)
Fugitive or
nonpoint air
emissions
158
0
0
0
2.3
0
0
0
0
0
2.3
0
2.7
Stack or point
air emissions
2,497
0
0
0
114
0
0
0
0
0
0
0
0
Surface water
discharges
13
a
0
0
0
0
0
0
0
0
0
120
4.5
Underground
injection
0.5
a
a
a
0
0
0
0
0
0
0
0
0
On-site
releases to
land
648,128
60,854
3,081
4,179
0
0
120
0.5
0
32,372
453
341
Total
on-site
Releases
650,796.5
0.0
60,854.0
3,081.0
4,295.3
0.0
0.0
120.0
0.5
0.0
32,374.3
573.0
348.2
Tacilities left that particular cell blank on the Form R submissions.
NA = Not available
Sources: U.S. EPA(1993f, 1995h, 1998b, 2003c).
-------
The number of each of these items owned by the utility industry, the quantity of PCBs
contained in each, and an estimate of the annual quantity of PCBs leaked and/or spilled were
investigated by the Edison Electric Institute and the Utility Solid Wastes Activity Group
(EEI/USWAG) for EPA in 1981. The findings of this investigation, which were reported in a
proposed modification to the PCB regulations (Federal Register, 1982a), indicated that more than
99% of the total quantity of PCBs contained in utility-owned electrical equipment in 1981
(73,700 metric tons) was in 40,000 PCB transformers (those containing >500 ppm of PCBs) and
large PCB capacitors (those containing >3 Ib of PCBs). An upper-bound estimate of the mass of
PCBs that leached or spilled from this equipment in 1981 was 177 metric tons. Approximately
95% of the estimated releases were the result of leaks from large PCB capacitors (Federal
Register, 1982a). Leaks/spills typically occur in transformers when the gasket joining the top to
the body corrodes, tears, or physically fails. PCBs can then leak past this failed section and
potentially spill onto the surrounding ground. PCB capacitors typically fail by rupturing,
exposing the contained PCBs to the environment. Failure is caused by environmental and
weathering effects (e.g., lightning) or material failures (e.g., metal fatigue).
As of mid-1988, the total population of in-service PCB transformers and large PCB
capacitors was estimated to have decreased from 140,000 to 110,000 and from 3.3 million to 1.9
million, respectively (Versar, Inc., 1988). PCB transformers have normal operating lifetimes of
30 years and 40 years, respectively. EPA's PCB Electrical Use Rule (Federal Register, 1982b)
required the removal of 950 food/feed industry transformers by 1985 and 1.1 million
unrestricted-access large PCB capacitors by October 1988. In addition, EPA's PCB Transformer
Fires Rule (Federal Register, 1985b) required the removal by 1990 of 7,600 480-volt network
transformers.
More recent inventories of PCB-containing electrical equipment are not available.
However, an Information Collection Request submitted by EPA to the Office of Management
and Budget for information on uses, locations, and conditions of PCB electrical equipment
estimated that there may be 150,000 owners of PCB-containing transformers used in industry,
utilities, government buildings, and private buildings (Federal Register, 1997b). It is expected,
and is demonstrated by the reported PCB transfers in the EPA's TRI (see Table 10-5), that many
owners of PCB electrical equipment have removed PCB-containing equipment to eliminate
potential liability.
10.2.3.2. Accidental Fires
The available information is not adequate to support an estimate of potential annual
releases of dioxin-like PCBs from accidental electrical equipment fires. For fires involving PCB
transformers or capacitors, the amount of PCBs released is dependent on the extensiveness of the
10-18
-------
fire and the speed at which it is extinguished. A number of these fires are documented. A New
York fire involving 200 gal of transformer fluid containing some 65% by weight PCBs resulted
in a release of up to 1,300 Ib of PCBs. A capacitor fire that burned uncontrolled for 2 hr in
Sweden resulted in the destruction of 12 large utility capacitors containing an estimated 25
pounds each of PCBs, for a total potential release of 300 Ib. However, data are incomplete on the
exact amount of PCBs released as a result of these two fires.
EPA has imposed reporting requirements to ensure that the National Response Center is
informed immediately of fires involving PCB transformers (40 CFR 761). The recordkeeping
requirements are used to document the use, location, and condition of PCB equipment.
Responses are mandatory, but the submitter may claim them to be confidential information. The
number of PCB transformer fires is estimated to be approximately 20 per year; the number of
PCB capacitor fires is unknown (U.S. EPA, 1987d). As these PCB-containing items reach the
end of their useful lives and are retired, their susceptibility to fires will be eliminated, and the
overall number of PCB transformer and capacitor fires will be reduced.
10.2.4. Municipal Wastewater Treatment
EPA conducted the National Sewage Sludge Survey in 1988 and 1989 to obtain national
data on sewage sludge quality and management. As part of this survey, EPA tested for more than
400 analytes, including seven of the Aroclors, in sludges from 175 publicly owned treatment
works (POTWs) that employed at least secondary wastewater treatment. Sludges from 19% of
the POTWs had detectable levels of at least one of the following Aroclors: 1248, 1254, or 1260;
none of the other Aroclors were detected in any sample (the detection limit [DL] was typically
about 200 |ig/kg dry weight) (U.S. EPA, 1996e). Analyses were not performed for dioxin-like
PCB congeners. The Aroclor-specific results of the survey are presented in Table 10-7.
Gutenmann et al. (1994) reported similar results in a survey of sludges from 16 large U.S.
cities for Aroclor 1260 content. At a DL of 250 |ig/kg (dry weight), the investigators detected
Aroclor 1260 (4,600 ng/kg) at only one facility. These results indicate that PCBs are not likely
to be formed at POTWs, but rather are present because of disposal of PCB products or
recirculation of previously disposed of PCBs.
Although PCBs, measured as Aroclors, were not commonly detected in sewage sludge at
microgram-per-kilogram levels in studies by EPA (U.S. EPA, 1996e) and Gutenmann et al.
(1994), the presence of dioxin-like PCB congeners at lower concentrations may be more
common. Green et al. (1995) and Cramer et al. (1995) reported the results of analyses of 99
samples of sewage sludge for PCB congener numbers 77, 81, 126, and 169. The sludge samples
were collected from 74 wastewater treatment plants across the United States during the summer
10-19
-------
Table 10-7. Aroclor concentrations (ng/kg) measured in EPA's National
Sewage Sludge Survey"
Aroclor
1016
1221
1232
1242
1248
1254
1260
Any Aroclor (total)
Percent
detected
0
0
0
0
9
8
10
19
Maximum
concentration
~
~
-
-
5.2
9.35
4.01
14.7
Median concentration
Nondetects set to
detection limit
-
-
~
~
0.209
0.209
0.209
1.49
Nondetects set
to zero
0
0
0
0
0
0
0
0
Tor publicly owned treatment works with multiple samples, the pollutant concentrations were averaged before the
summary statistics presented in the table were calculated.
~ = No information given
Source: U.S. EPA(1996e).
of 1994. These data are summarized in Table 10-8. Results from all samples collected from the
same facility were averaged by Green et al. and Cramer et al. to ensure that results were not
biased toward the concentrations found at facilities from which more than one sample was
collected. If all nondetect values were assumed to be zero, then the POTW mean TEQP-WHO94
and TEQP-WHO98 concentrations were 25.1 and 24.2 ng TEQ/kg (dry-weight basis), respectively.
If the nondetect values were set equal to the DLs, then the POTW mean TEQP-WHO94 and
TEQP-WHO98 concentrations were 25.2 and 24.3 ng TEQ/kg, respectively.
In 1999, sewage sludge samples from a POTW in Ohio were collected and analyzed for
PCBs (U.S. EPA, 2000b). The facility, which accepts both domestic and industrial wastewater,
employs secondary wastewater technology. Assuming nondects were zero, the mean TEQ
emission factor was 141 ng TEQP-WHO98/kg. These results are presented in Table 10-9.
In 2000 and 2001, the Association of Metropolitan Sewage Agencies conducted a survey
of dioxin-like PCB compounds in sewage sludge (Alvarado et al., 2001). A total of 200 sewage
sludge samples were collected from 171 POTWs located in 31 states. Assuming nondetects were
zero, the mean and median TEQ emission factors were reported as 8.3 and 3.37 ng TEQP-
WHO98/kg, respectively.
10-20
-------
Table 10-8. Dioxin-like PCB concentrations measured in sludges collected from 74 U.S. publicly owned
treatment works (POTWs) during 1994a b
Congener
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
IUPAC
number
77
81
105
114
118
123
126
156
157
167
169
170
180
189
Percent
detected
100
86
99
22
Maximum
concentration
(ng/kg)
22,900
1,250
3,020
1,470
Total TEQP-WHO98
Median concentration (ng/kg)
Nondetect set to
1A detection
limit
783
27.3
91.6
8.5
9.3
Nondetect
set to zero
783
27
91.6
0
9.2
Mean concentration (ng/kg)
Nondetect set
to 1A detection
limit
2,243
65.2
237
32.5
24.3
Nondetect set
to zero
2,243
63.5
237
26.2
24.2
o
to
Tor POTWs with multiple samples, the sample concentrations were averaged by Cramer et al. (1995) to POTW averages before calculation of the total TEQ
mean and median values presented in the table. The TEQP-WHO94 and TEQP-WHO98 values were calculated on a facility-level basis.
bBlank cells indicate that no measurements of these congeners were made.
Source: Green et al. (1995); Cramer et al. (1995).
-------
Table 10-9. Dioxin-like PCB concentrations in sewage sludge collected from
U.S. publicly owned treatment works during 1999
Congener
3,3',4,4'-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2',3,4,4',5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
IUPAC
number
77
105
114
123
118
126
156
157
167
169
170
180
189
Total TEQP-WHO98
Mean emission factor (ng/kg)
Nondetect set to Vz
detection limit
42,467
7,230
701
249
12,867
1,270
1,843
524
935
570
2,627
6,497
199
141
Nondetect set to zero
42,467
7,230
701
249
12,867
1,270
1,843
524
935
570
2,627
6,497
199
141
Source: U.S. EPA(2000b).
For 2001, EPA conducted another National Sewage Sludge Survey to characterize the
dioxin and dioxin-like equivalence levels in biosolids produced by the 6,857 POTWs operating
in the United States in 2001 (U.S. EPA, 2002d). Sewage sludge samples were collected from 94
POTWs that used secondary or higher treatment practices. All the facilities had been sampled as
part of the 1988/1989 National Sewage Sludge Survey. To determine the mean and median TEQ
emission estimates of the dioxin-like PCBs, EPA weighted the values on the basis of wastewater
flow rates of all POTWs in the United States (i.e., number of facilities with wastewater flow rate
>100 mg/day, >10 but < 100 mg/day, >1 but < 10 mg/day, and < 1 mg/day). The weighted mean
and median TEQP-WHO98 concentrations of the dioxin-like PCB congeners were 5.22 and 2.05
ng/kg, respectively.
According to the results of its 1988/1989 National Sewage Sludge Survey, EPA estimated
that approximately 5.4 million dry metric tons of sewage sludge were generated in 1989 (Federal
Register, 1993a). EPA also used the results of the 1984 to 1996 Clean Water Needs Surveys to
10-22
-------
estimate that 6.3 million dry metric tons of sewage sludge were generated in 1998 and 6.6
million dry metric tons were generated in 2000 (U.S. EPA, 1999b). Because estimates for 1987
and 1995 are not available, the 1989 and 1998 activity level estimates are used for reference
years 1987 and 1995, respectively. Tables 10-10, 10-11, and 10-12 list the volume, by use and
disposal practices, of sludge disposed of annually for reference years 1989, 1995, and 2000.
Table 10-10. Quantity of sewage sludge disposed of annually in 1989 by
primary, secondary, or advanced treatment publicly owned treatment works
(POTWs) and potential dioxin-like PCB TEQ releases
Use/disposal practice
Land application
Distribution and marketing
Surface disposal site/other
Sewage sludge landfill
Co-disposal landfills0
Sludge incinerators and co-
incinerators'1
Ocean disposal
TOTAL
Volume
disposed of
(1,000 dry
metric tons/yr)
1,714
71
396
157
1,819
865
336
5,358
Percent
of
total
volume
32b
1.3
7.4
2.9
33.9
16.1
6.3
100
Potential TEQP-
WHO98 release3
(gofTEQ/yr)
41.5
1.7
9.6
4.2
44
e
0
101
Potential TEQP-
WHO94 release3
(gofTEQ/yr)
43
1.8
9.9
3.9
45.6
e
0
104.2
Totential TEQ release for nonincinerated sludges was estimated by multiplying the sludge volume generated
(column 2) by the mean dioxin-like PCB TEQ concentration in 74 POTW sludges reported by Green et al. (1995)
and Cramer et al. (1995) (i.e., 24.2 ng TEQP-WHO98/kg and 25.1 ng TEQP-WHO94/kg).
Includes 21.9% applied to agricultural land, 2.8% applied as compost, 0.6% applied to forestry land, 3.1%
applied to "public contact" land, 1.2% applied to reclamation sites, and 2.4% applied in undefined settings.
°Landfills used for disposal of sewage sludge and solid waste residuals.
dCo-incinerators treat sewage sludge in combination with other combustible waste materials.
eSee Section 10.4.6 for a discussion of dioxin-like PCB releases to air from sewage sludge incinerators.
The Ocean Dumping Ban Act of 1988 generally prohibited the dumping of sewage sludge into the ocean after
December 31, 1991. Ocean dumping of sewage sludge ended in June 1992 (Federal Register, 1993a). The
current method of disposal of the 336,000 metric tons of sewage sludge that were disposed of in the oceans in
1988 has not been determined.
Sources: Federal Register (1990, 1993a); Green etal. (1995); Cramer etal. (1995).
10-23
-------
Table 10-11. Quantity of sewage sludge disposed of annually in 1995 by
primary, secondary, or advanced treatment publicly owned treatment works
(POTWs) and potential dioxin-like PCB TEQ releases
Use/disposal practice
Land application13
Advanced treatment0
Other beneficial used
Surface disposal/Landfill
Incineration
Other disposal method
TOTAL
Volume disposed of
(1,000 dry metric
tons/yr)
2,500
700
500
1,100
1,400
100
6,300
Percent of
total volume
39.7
11.1
7.9
17.5
22.2
1.6
100
Potential dioxin release"
(g TEQ/yr)
TEQP-WHO98
60.5
16.9
12.1
26.6
e
2.4
118.5
TEQP-WHO94
62.8
17.6
12.6
27.6
e
2.5
123.1
Totential TEQ release for nonincinerated sludges was estimated by multiplying the sludge volume generated
(column 2) by the mean dioxin-like PCB TEQ concentration in 74 POTW sludges reported by Green et al. (1995)
and Cramer et al. (1995) (i.e., 24.2 ng TEQP- WHO98/kg and 25.1 ng TEQP-WHO94/kg).
bWithout further processing or stabilization, such as composting.
°Such as composting.
dEPA assumed that this category includes distribution and marketing (i.e., sale or give-away of sludge for use in
home gardens). Based on the 1988 National Sewage Sludge Survey and 1988 Needs Survey, approximately 1.3%
of the total volume of sewage disposed was distributed and marketed (Federal Register, 1993a). Therefore, it is
estimated that 2 g (TEQP-WHO98 and TEQP-WHO94) were released through distribution and marketing in 1995.
eSee Section 3.5 for estimates of CDD/CDF releases to air from sewage sludge incinerators.
Sources: U.S. EPA (1999b); Green et al. (1995); Cramer et al. (1995).
These tables also list the estimated amount of dioxin-like PCB TEQs that may be present
in sewage sludge and potentially released to the environment. For reference years 1987 and
1995, these values were estimated using the POTW mean TEQP-WHO98 concentration calculated
from the results reported by Green et al. (1995) and Cramer et al. (1995). For reference year
2000, they were estimated using the POTW mean TEQP-WHO98 concentration reported by EPA
(U.S. EPA, 2002d) as part of the 2001 National Sewage Sludge Survey. Multiplying these TEQ
concentrations by the sludge volumes generated yields annual potential total releases of 101 g
TEQP-WHO98 (104.2 g TEQP-WHO94) in 1987, 118.5 g TEQP-WHO98 (123.1 g TEQP-WHO94) in
1995, and 26.6 g TEQP-WHO98 in 2000 for nonincinerated sludges.
10-24
-------
Table 10-12. Quantity of sewage sludge disposed of annually in 2000 by
primary, secondary, or advanced treatment publicly owned treatment works
(POTWs) and potential dioxin-like PCB TEQ releases
Use/disposal practice
Land application13
Advanced treatment0
Other beneficial used
Surface disposal/landfill
Incineration
Other disposal method
TOTAL
Volume disposed of
(1,000 dry metric
tons/yr)
2,800
800
500
900
1,500
100
6,600
Percent of
total volume
42.4
12.1
7.6
13.6
22.7
1.5
100
Potential
TEQDF-WHO98 release3
(g TEQ/yr)
14.6
4.2
2.6
4.7
e
0.5
26.6
Totential dioxin TEQ release for nonincinerated sludges was estimated by multiplying the sludge volume
generated (column 2) by the average of the mean TEQDF-WHO98 concentrations in sludge reported by U.S. EPA
(2002c) (i.e., 5.22 ng TEQDF-WHO98/kg).
bWithout further processing or stabilization, such as composting.
°Such as composting.
dEPA assumed that this category includes distribution and marketing (i.e., sale or give-away of sludge for use in
home gardens). Based on the 1988 National Sewage Sludge Survey and 1988 Needs Survey, approximately 1.3%
of the total volume of sewage disposed of was distributed and marketed (Federal Register, 1993a). Therefore, it
is estimated that 0.5 g TEQDF-WHO98 were released through distribution and marketing in 2000.
eSee Section 3.5 for estimates of CDD/CDF releases to air from sewage sludge incinerators.
Sources: U.S. EPA (1999b, 2002d).
Of the 101 g TEQP-WHO98 released in 1987, 1.7 g entered commerce as a product for
distribution and marketing and the remainder was applied to land (41.5 g to land application and
9.6 g to surface disposal sites) or landfilled (48.2 g). Of the 118.5 g TEQP-WHO98 released in
1995, 60.5 g were applied to land without further processing or stabilization, 16.9 g underwent
advanced treatment such as composting, 26.6 g were disposed of on the surface or landfilled, and
the remainder was either used or disposed of in other ways. Of the 26.6 g TEQP-WHO98 released
in 2000, 14.6 g were applied to land without further processing or stabilization, 4.2 g underwent
advanced treatment such as composting, 4.7 g were disposed of on the surface or landfilled, and
the remainder was either used or disposed of in other ways. The PCBs in landfilled sludge were
not considered releases to the environment under the definition established in this document.
The other disposal practices were considered releases and were summed to get total land releases,
as shown in Table 10-2 (above).
10-25
-------
The 1987 and 1995 release estimates are assigned a confidence rating of B, indicating
high confidence in the production estimate and medium confidence in the emission factor
estimates. The medium rating was based on the judgment that, although the 74 facilities tested
by Green et al. (1995) and Cramer et al. (1995) may be reasonably representative of the
variability in POTW technologies and sewage characteristics nationwide, the sample size was
still relatively small, and not all dioxin-like PCB congeners were monitored. The 2000 release
estimates are assigned a confidence rating of A, indicating high confidence in both the
production estimate and the emission factor estimates. High confidence was placed in the
emission factors estimated because they were weighted on the basis of wastewater flow rates of
all POTWs in the United States.
10.3. CHEMICAL MANUFACTURING AND PROCESSING SOURCES
In the early 1980s, EPA investigated the extent of inadvertent generation of PCBs during
the manufacture of synthetic organic chemicals (Hammerstrom et al., 1985). For example,
phthalocyanine dyes and diarylide pigments were reported to contain PCBs in the milligram-per-
kilogram range. EPA subsequently issued regulations under TSCA (40 CFR 761.3) that ban the
distribution in commerce of any products containing an annual average PCB concentration of 25
mg/kg (50 mg/kg maximum concentration at any time). In addition, EPA requires manufacturers
with processes that inadvertently generate PCBs and importers of products that contain
inadvertently generated PCBs to report to EPA any process or import for which the PCB
concentration is greater than 2 mg/kg for any resolvable PCB gas chromatographic peak.
10.4. COMBUSTION SOURCES
10.4.1. Municipal Waste Combustors
Municipal waste combustors (MWCs) have long been identified as potential PCB air
emission sources. Stack gas concentrations of PCBs for three MWCs were reported (U.S. EPA,
1987d); the average test results yielded an emission factor of 18 |ig/kg refuse. Stack gas
emissions of PCBs from the three MWCs were quantified without determining the MWCs' PCB
destruction efficiency.
EPA also analyzed the PCB content of various consumer paper products (U.S. EPA,
1987d). The results indicated that paper products such as magazine covers and paper towels
contained up to 139 |ig/kg paper. These levels, which were reported in 1981, were attributed to
the repeated recycling of waste paper containing PCBs. For example, carbonless copy paper
manufactured prior to 1971 contained PCB levels as high as 7%. This copy paper then became a
component of waste paper, which was recycled. The PCBs were inevitably introduced into other
paper products, resulting in continued measurable levels in municipal refuse some four years
10-26
-------
after the PCB manufacturing ban was imposed. Refuse-derived fuel manufactured from these
paper products had PCB levels of 8,500 |ig/kg, indicating that this fuel could be a source of
atmospheric PCBs. Therefore, it was assumed that municipal refuse does contain detectable
levels of PCBs and that some of these PCBs may enter the atmosphere when the refuse is
incinerated (U.S. EPA, 1987d).
Shane et al. (1990) analyzed fly ash from five MWCs for PCB congener group content.
Total PCB levels ranged from 99 to 322 |ig/kg in the ash, with the tri, tetra, and penta congener
groups occurring in the highest concentrations. The investigators also analyzed seven bottom ash
and eight bottom ash/fly ash mixtures for total PCB measured as Aroclor 1254. The DL for this
Aroclor analysis was 5 |ig/kg. Aroclor 1254 was detected in two of the seven bottom ash
samples (26 and 8 |ig/kg) and in five of the eight fly ash/bottom ash mixtures (range, 6 to 33
Sakai et al. (2001) analyzed the PCB levels in fly ash and bottom ash from a newly
constructed MWC in Japan. The I-TEQ values derived from the data give a total TEQ value of
31.6 ng/kg for fly ash and 0.85 ng/kg for bottom ash.
The development of more sensitive analytical methodologies has enabled researchers in
recent years to detect dioxin-like PCB congeners in the stack gases and fly ash from full-scale
and pilot-scale MWCs (Sakai et al., 1993a, b, 1994, 2001; Boers et al., 1993; Schoonenboom et
al., 1993). Similarly, the advances in analytical techniques have enabled researchers to determine
that dioxin-like PCBs can be formed during the oxidative solid combustion phase of incineration,
presumably due to dimerization of chlorobenzenes. Laboratory-scale studies have also
demonstrated that dioxin-like PCBs can be formed from heat treatment of fly ash in air
(Schoonenboom et al., 1993; Sakai et al., 1994); however, the available data are not adequate to
support development of a quantitative estimate of a dioxin-like PCB emission factor for this
source category. MWCs are designated as a Category E source.
10.4.2. Industrial Wood Combustion
Emissions of PCB congener groups (but not individual congeners) were measured during
stack testing at two industrial wood-burning facilities (CARB, 1990d, e). Table 10-13 presents
the average of the congener group (monochlorobiphenyl [MCB] through decachlorobiphenyl
[DCB]) emission factors for these two facilities. No tetra- or higher-chlorinated congeners (the
congener groups containing the dioxin-like PCBs) were detected at either facility at DLs
corresponding to emission factors in the low range of nanogram per kilogram of wood
combusted.
10-27
-------
Table 10-13. PCB congener group emission factors for industrial wood
combustors"
Congener group
Monochlorobiphenyls
Dichlorobiphenyls
Trichlorobiphenyls
Tetrachlorobiphenyls
Pentachlorobiphenyls
Hexachlorobiphenyls
Heptachlorobiphenyls
Octachlorobiphenyls
Nonachlorobiphenyls
Decachlorobiphenyls
Number of
detections
1
1
1
0
0
0
0
0
0
0
Maximum
concentration
detected
(ng/kg wood)
32.1
23
19.7
-
-
~
~
-
~
-
Mean concentration
(ng/kg)
Nondetect set
to detection
limit
39.4
50.9
42.3
22.7
17.6
17
17.9
15.8
25
36.3
Nondetect set
to zero
16
11.5
9.8
-
~
~
~
~
~
~
aTwo sites for each congener group.
~ = No information given
Source: CARB (1990d, e).
In CARB (1990d), PCBs were measured in the emissions from two spreader stoker wood-
fired boilers operated in parallel by an electric utility for generating electricity. The exhaust gas
stream from each boiler was passed through a dedicated electrostatic precipitator (ESP), after
which the gas streams were combined and emitted to the atmosphere through a common stack.
Stack tests were conducted both when the facility burned fuels allowed by existing permits and
when the facility burned a mixture of permitted fuel supplemented by urban wood waste at a ratio
of70:30.
In CARB (1990e), PCBs were measured in the emissions from twin fluidized-bed
combustors designed to burn wood chips to generate electricity. The air pollution control device
(APCD) system consisted of ammonia injection for controlling nitrogen oxides and a multiclone
and an ESP for controlling PM. During testing, the facility burned wood wastes and agricultural
wastes allowed by existing permits. The available data are not adequate to support development
of a quantitative estimate of dioxin-like PCB emissions from this source. Industrial wood
combustion is designated as a Category E source.
10-28
-------
10.4.3. Medical Waste Incineration
As discussed in Chapter 3, Section 3.3, EPA has issued nationally applicable emission
standards and guidelines that address CDD/CDF emissions from medical waste incinerators
(MWIs). Although PCBs are not addressed in these regulations, the database of stack test results
at MWIs compiled for this rulemaking does contain limited data on PCB congener group
emission factors. Data are available for two MWIs lacking add-on APCD equipment and for two
MWIs with add-on APCD equipment in place. The average congener group emission factors
derived from these test data are presented in Table 10-14. Because data are available for only 4
of the estimated 1,065 facilities that make up this industry, and because these data do not provide
congener-specific emission factors, no national estimates of total PCB or dioxin-like PCB
emissions are being made at this time. Medical waste incineration is designated as a Category E
source.
Table 10-14. PCB congener group emission factors for medical waste
incinerators (MWIs)a
Congener group
Monochlorobiphenyls
Dichlorobiphenyls
Trichlorobiphenyls
Tetrachlorobiphenyls
Pentachlorobiphenyls
Hexachlorobiphenyls
Heptachlorobiphenyls
Octachlorobiphenyls
Nonachlorobiphenyls
Decachlorobiphenyls
Mean emission factor (ng/kg)
(2 MWIs without APCD)
Nondetects set to
detection limit
0.059
0.083
0.155
4.377
2.938
0.238
0.155
0.238
0.155
0.155
Nondetects
set to zero
0.059
0.083
0.155
4.377
2.938
0.238
0.155
0.238
0.155
0.155
Mean emission factor (ng/kg)
(2 MWIs with APCD)
Nondetects set to
detection limit
0.311
0.34
0.348
1.171
17.096
1.286
0.902
0.205
~
0.117
Nondetects
set to zero
0
0
0
0
9.996
1.078
0
0
~
0
aSee Section 3.3 for details on tested facilities.
APCD = Air pollution control device
~ = No information given
10-29
-------
10.4.4. Tire Combustion
As discussed in Chapter 3, Section 3.6, tires are burned in a variety of facilities, including
dedicated tire burners, cement kilns, industrial boilers, and pulp and paper combustion facilities.
Emissions of PCB congener groups (but not individual congeners) were measured during stack
testing of a tire incinerator (CARB, 1991). The facility consisted of two excess air furnaces
equipped with steam boilers to recover the energy from the heat of combustion. Discarded whole
tires were fed to the incineration units at rates ranging from 2,800 to 5,700 kg/hr during the three
testing days. The furnaces were equipped to burn natural gas as auxiliary fuel. The steam
produced from the boilers drove electrical turbine generators that produced 14.4 megawatts of
electricity. The facility was equipped with a dry acid gas scrubber and a fabric filter for the
control of emissions prior to exiting the stack. Table 10-15 presents the congener group (MCB
through DCB) emission factors for this facility. The emission factor for the total of the tetra-
through heptachlorinated congener groups was about 1.2 |ig/kg tire processed. Because these
data do not provide PCB congener-specific emission factors, no estimates of emissions of dioxin-
like PCBs can be made. Tire combustion is designated as a Category E source.
Table 10-15. PCB congener group emission factors for a tire combustora
Congener group
Monochlorobiphenyls
Dichlorobiphenyls
Trichlorobiphenyls
Tetrachlorobiphenyls
Pentachlorobiphenyls
Hexachlorobiphenyls
Heptachlorobiphenyls
Octachlorobiphenyls
Nonachlorobiphenyls
Decachlorobiphenyls
Number of
detections
0
1
1
0
2
1
1
0
0
0
Maximum
emission factor
(ng/kg)
-
34.8
29.5
~
2,724
106.5
298.6
-
-
-
Mean emission factor
(ng/kg)
Nondetect set to
detection limit
0.04
11.7
11.8
10
1,092
55.9
107.7
20.9
17.7
41.9
Nondetect set
to zero
~
11.6
9.8
~
1,092
35.5
99.5
~
~
~
Three samples for each congener group.
~ = No information given
Source: CARB (1991).
10-30
-------
10.4.5. Cigarette Smoking
Using high-resolution mass spectrometry, Matsueda et al. (1994) analyzed tobacco from
20 brands of commercially available cigarettes collected in 1992 from Japan, the United States,
Taiwan, China, the United Kingdom, Germany, and Denmark for the PCB congeners 77, 126,
and 169. Table 10-16 presents the results of the study. However, no studies examining tobacco
smoke for the presence of these congeners have been reported. Thus, it is not known whether the
PCBs present in the tobacco are destroyed or volatilized during combustion or whether PCBs are
formed during combustion. At least 1,200 tobacco constituents (e.g., nicotine, n-paraffin, some
terpenes) are transferred intact from the tobacco into the smoke stream by distillation in this area,
and it is plausible that PCBs present in the unburned tobacco would be subject to similar
distillation.
Cigarette consumption and the combustion processes operating during cigarette smoking
are discussed in Chapter 5, Section 5.5.
A preliminary rough estimate of potential emissions of dioxin-like PCBs can be made
using the following assumptions: (a) the average TEQP-WHO98 content of seven brands of U.S.
cigarettes reported by Matsueda et al. (1994), 0.64 pg/pack (0.032 pg/cigarette), is representative
of cigarettes smoked in the United States; (b) dioxin-like PCBs are neither formed nor destroyed,
and the congener profile reported by Matsueda et al. (1994) is not altered during combustion of
cigarettes; and (c) all dioxin-like PCBs contributing to the TEQ are released from the tobacco
during smoking. On the basis of these assumptions, the calculated annual emissions would be
0.018 g TEQP-WHO98, 0.016 g TEQP-WHO98, and 0.014 g TEQP-WHO98 for reference years
1987, 1995, and 2000, respectively. These estimates are assigned a confidence rating of D
because the emission factor is clearly not representative of cigarette smoke.
10.4.6. Sewage Sludge Incineration
EPA (U.S. EPA, 1996g) derived an emission factor of 5.4 jig PCBs/kg dry sewage sludge
incinerated. This emission factor was based on measurements conducted at five multiple-hearth
incinerators controlled with wet scrubbers. However, it is not known what fraction of the
emissions was dioxin-like PCBs.
In 1999, stack tests were conducted at a multiple-hearth incinerator in Ohio equipped
with a venturi scrubber and a three-tray impingement conditioning tower (U.S. EPA, 2000b). Of
the four test runs conducted, the first test run was aborted and the results from the fourth test run
were determined to be statistical outliers (p>0.05). The back-half CDD/CDF concentrations for
test run 4 were 50 to 60% lower than back-half emission concentrations for test runs 2 and 3.
Because of the problems associated with test run 4, the results were not used to calculate an
emission factor for dioxin-like PCBs. The average TEQ emission factor (excluding test run 4)
10-31
-------
Table 10-16. Dioxin-like PCB concentrations in cigarette tobacco in brands from various countries (pg/pack)a
Congener
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
IUPAC
number
77
81
105
114
118
123
126
156
157
167
169
170
180
189
Total TEQP-WHO98
U.S.
(avg. of 7
brands)
105.7
6.2
0.9
0.64
Japan
(avg. of 6
brands)
70.2
7.8
0.9
0.8
United
Kingdom
(avg. of 3
brands)
53
6.1
0.9
0.62
Taiwan
(1 brand)
133.9
14.5
2.4
1.49
China
(1 brand)
12.6
2.4
0.4
0.24
Denmark
(1 brand)
21.7
2.2
0.5
0.23
Germany
(1 brand)
39.3
7.3
1.6
0.75
o
to
aBlank cells indicate that no measurements of these congeners were made.
Source: Matsuedaetal. (1994).
-------
was 0.51 ng TEQP-WHO98/kg (see Table 10-17). This emission factor was assigned a low
confidence rating because it is based on limited surveys that are judged to be possibly
nonrepresentative.
Table 10-17. Dioxin-like PCB concentrations in stack gas collected from a
U.S. sewage sludge incinerator
Congener
3,3',4,4'-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2',3,4,4',5-PeCB
2,3',4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2,3,3',4,4',5,5'-HpCB
IUPAC
number
77
105
114
123
118
126
156
157
167
169
170
180
189
Total TEQP-WHO98
Mean emission factor (ng/kg)
Nondetect set to Vz
detection limit
92.37
18
2.56
0.82
38.65
4.51
4.25
1.41
2.55
3.61
7.19
17.79
0.6
0.51
Nondetect set to zero
92.37
18
2.56
0.82
38.65
4.51
4.25
1.41
2.55
3.61
7.19
17.79
0.6
0.51
Source: U.S. EPA(2000b).
Approximately 0.865 million dry metric tons of sewage sludge were incinerated in 1988
(Federal Register, 1993a), approximately 2.11 million dry metric tons in 1995 (e-mail dated July
13, 1998, from K. Maw, Pacific Environmental Services, to G. Schweer, Versar, Inc.), and an
estimated 1.42 million dry metric tons in 2000 (U.S. EPA, 1999b). Using the above estimated
amounts of sewage sludge incinerated per year and the average TEQ emission factor of 0.51 ng
TEQP-WHO98/kg, the estimated annual releases of total PCBs to air were 0.44 g TEQP-WHO98 in
1987, 1.1 g TEQP-WHO98 in 1995, and 0.72 g TEQP-WHO98 in 2000. These emissions were
assigned a low confidence rating (Category C) because the emission factor was given a low
rating.
10-33
-------
10.4.7. Backyard Barrel Burning
The low combustion temperatures and oxygen-starved conditions associated with
backyard barrel burning may result in incomplete combustion and increased pollutant emissions
(Lemieux, 1997). EPA's Control Technology Center, in cooperation with New York State's
departments of health and environmental conservation, conducted a study to examine,
characterize, and quantify emissions from the simulated open burning of household waste
materials in barrels (Lemieux, 1997). A representative waste to be burned was prepared on the
basis of the typical percentages of various waste materials disposed of by New York State
residents (i.e., nonavid recyclers); hazardous wastes such as chemicals, paints, and oils were not
included in the test waste. A variety of compounds, including dioxin-like PCBs, were measured
in the emissions from the simulated open burning. The measured TEQ emission factors for
waste that had not been separated for recycling purposes were 1.02 x 10"2 |ig TEQP-WHO94/kg
and 5.26 x 10"3 jig TEQP-WHO98/kg waste burned (see Table 10-18). These limited emissions
data were judged to be inadequate for estimating national emissions (a Category E source). The
activity level for backyard barrel burning is discussed in Section 6.5.2.
10.4.8. Petroleum Refining Catalyst Regeneration
As discussed in Section 5.4, regeneration of spent catalyst used in catalytic reforming to
produce high-octane reformates is a potential source of CDD/CDF air emissions. In 1998,
emissions from the caustic scrubber used to treat gases from the external catalyst regeneration
unit of a refinery in California were tested for CDDs/CDFs as well as PCB congener groups
(CARS, 1999) (see Chapter 5, Section 5.4 for details).
All PCB congener groups were detected in each of the three samples collected. The
average congener group emission factors in units of nanograms per barrel of reformer feed are
presented in Table 10-19. The total PCB emission factor was 118 ng/barrel. This emission
factor assumes that emissions are proportional to reforming capacity; emission factors may be
more related to the amount of coke burned, APCD equipment present, and/or other process
parameters.
Because emissions data are available for only one U.S. petroleum refinery (which
represents less than 1% of the catalytic reforming capacity at U.S. refineries), and because these
data do not provide congener-specific emission factors, no national estimates of total PCB or
dioxin-like PCB emissions are being made at this time. This is a Category E source.
10-34
-------
Table 10-18. Dioxin-like PCB emission factors from backyard barrel
burninga
Congener
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2 3 3' 4 4' 5-HxCB
2 3 3' 4 4' 5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,4',5-HpCB
2,2',3,4,4',5,5'-HpCB
2 3 3' 4 4' 5 5'-HpCB
IUPAC
number
77
81
105
114
118
123
126
156
157
167
169
170
180
189
Total TEQP-WHO98
Emission factors (ng/kg)
Testl
9.3
5.9
8.3
18.6
4.21e-03
Test 2
15.2
4.9
14.3
28.7
6.31e-03
Average
12.3
5.4
11.3
23.7
5.26e-03
aBlank cells indicate that the congener was not detected in either of the two duplicate samples.
Source: Lemieux (1997).
Table 10-19. PCB congener group emission factors for a petroleum catalytic
reforming unita
Congener group
Monochlorobiphenyls
Dichlorobiphenyls
Trichlorobiphenyls
Tetrachlorobiphenyls
Pentachlorobiphenyls
Hexachlorobiphenyls
Heptachlorobiphenyls
Octachlorobiphenyls
Nonachlorobiphenyls
Decachlorobiphenyls
Total PCBs
Mean
concentration
(ng/dscm)
(at 12% O2)
166
355
743
849
914
780
1,430
698
179
41.3
6,155.3
Mean emission
rate
(Ib/hr)
5.51e-08
1.17e-07
2.45e-07
2.81e-07
3.02e-07
2.57e-07
4.73e-07
2.32e-07
5.99e-08
1.39e-08
2.04e-06
Mean emission
factor
(lb/1000 bbl)
7.11e-09
1.52e-08
3.17e-08
3.62e-08
3.88e-08
3.30e-08
6.01e-08
2.95e-08
7.59e-09
1.76e-09
2.61e-07
Mean emission
factor
(ng/barrel)
3.23e+00
6.89e+00
1.446+01
1.64e+01
1.766+01
1.50e+01
2.73e+01
1.34e+01
3.44e+00
7.98e-01
1.18e+02
"Three samples and three detections for each congener group.
Source: CARB (1999).
10-35
-------
10.5. NATURAL SOURCES
This section discusses biotransformation and photochemical transformation of other
PCBs. Although there is some evidence that these processes occur, the data were considered
insufficient for developing release estimates.
10.5.1. Biotransformation of Other PCBs
Studies show that under anaerobic conditions, biologically mediated reductive
dechlorination to lower-chlorinated congeners, followed by slow anaerobic and/or aerobic
biodegradation, is a major pathway for destruction of PCBs in the environment. Research
reported to date and summarized below indicates that biodegradation should result in a net
decrease rather than a net increase in the environmental load of dioxin-like PCBs.
Laboratory studies (e.g., Bedard et al., 1986; Pardue et al., 1988; Larsson and
Lemkemeier, 1989; Hickey, 1995; Schreiner et al., 1995) have revealed that more than two dozen
strains of aerobic bacteria and fungi that are capable of degrading most PCB congeners with five
or fewer chlorines are widely distributed in the environment. Many of these organisms are of the
genus Pseudomonas or Alcaligenes. The major metabolic pathway involves addition of oxygen
at the 2,3-position by a dioxygenase enzyme, with subsequent dehydrogenation to the catechol
followed by ring cleavage. Several bacterial strains have been shown to possess a dioxygenase
enzyme that attacks the 3,4-position.
Only a few strains have demonstrated the ability to degrade hexachlorobiphenyl (HxCB)
and the higher-chlorinated biphenyls. The rate of aerobic biodegradation decreases with
increasing chlorination. The half-lives for biodegradation of tetrachlorobiphenyls (TCBs) in
fresh surface water and soil are 7 to 60+ days and 12 to 30 days, respectively. For
pentachlorobiphenyls (PeCBs) and the higher-chlorinated PCBs, the half-lives in fresh surface
water and soil are likely to exceed 1 year. PCBs with all or most chlorines on one ring and PCBs
with fewer than two chlorines in the ortho position tend to degrade more rapidly. For example,
Gan and Berthouex (1994) monitored over a 5-yr period the disappearance of PCB congeners
applied to soil with sewage sludge. Three of the tetra- and pentachlorinated dioxin-like PCBs
(IUPAC Nos. 77, 105, and 118) followed a first-order disappearance model, with half-lives
ranging from 43 to 69 months. A hexa-substituted congener (IUPAC No. 167) and a hepta-
substituted congener (IUPAC No. 180) showed no significant loss over the 5-yr period.
Prior to the early 1990s, little investigation focused on anaerobic microbial dechlorination
or degradation of PCBs, even though most PCBs eventually accumulate in anaerobic sediments
(Abramowicz, 1990; Risatti, 1992). Environmental dechlorination of PCBs via losses of meta
and para chlorines has been reported in field studies for freshwater, estuarine, and marine
anaerobic sediments, including those from the Acushnet Estuary, the Hudson River, the
10-36
-------
Sheboygan River, New Bedford Harbor, Escambia Bay, Waukegan Harbor, the Housatonic
River, and Woods Pond (Brown et al., 1987; Rhee et al., 1989; Van Doit and Bedard, 1991;
Abramowicz, 1990; Bedard et al., 1995; Bedard and May, 1996). The altered PCB congener
distribution patterns found in these sediments (i.e., different patterns with increasing depth or
distance from known sources of PCBs) have been interpreted as evidence that bacteria may
dechlorinate PCBs in anaerobic sediment.
Reported results of laboratory studies confirm anaerobic degradation of PCBs. Chen et
al. (1988) found that "PCB-degrading" bacteria from the Hudson River could significantly
degrade the MCB, dichlorobipheyl (DiCB), and trichlorobiphenyl (TrCB) components of a 20
ppm Aroclor 1221 solution within 105 days. These congener groups make up 95% of Aroclor
1221. No degradation of higher-chlorinated congeners (present at 30 ppb or less) was observed,
and a separate 40-day experiment with TCB also showed no degradation.
Rhee et al. (1989) reported degradation of mono- to penta-substituted PCBs in
contaminated Hudson River sediments held under anaerobic conditions in the laboratory (N2
atmosphere) for 6 months at 25°C. Amendment of the test samples with biphenyl resulted in
greater loss of PCBs. No significant decreases in the concentrations of the higher-chlorinated
congeners (more than five chlorines) were observed. No evidence of degradation was observed
in samples incubated in CO2/H2 atmospheres. Abramowicz (1990) hypothesized that this result
could be an indication that, in the absence of CO2, a selection is imposed favoring organisms
capable of degrading PCBs to obtain CO2 and/or low-molecular-weight metabolites as electron
receptors.
Risatti (1992) examined the degradation of PCBs at varying concentrations (10,000 ppm,
1,500 ppm, and 500 ppm) in the laboratory with "PCB-degrading" bacteria from Waukegan
Harbor. After nine months of incubation at 22°C, the 500 ppm and 1,500 ppm samples showed
no change in PCB congener distributions or concentrations, thus indicating a lack of degradation.
Significant degradation was observed in the 10,000 ppm sediment, with at least 20 congeners
ranging from the TrCBs through the PeCBs showing decreases.
Quensen et al. (1988) also demonstrated that microorganisms from PCB-contaminated
sediments (Hudson River) dechlorinated most TrCBs through HxCBs in Aroclor 1242 under
anaerobic laboratory conditions. The Aroclor 1242 used to spike the sediment contained
predominantly TrCBs and TCBs (85 mol percent). Three concentrations of the Aroclor,
corresponding to 14, 140, and 700 ppm on a sediment dry-weight basis, were used.
Dechlorination was most extensive at the 700 ppm test concentration: 53% of the total chlorine
was removed in 16 weeks, and the proportion of TCBs through HxCBs decreased from 42 to 4%.
Much less degradation was observed in the 140 ppm sediment, and no observable degradation
was found in the 14ppm sediment. These results and those of Risatti (1992) suggest that the
10-37
-------
organism(s) responsible for this dechlorination may require relatively high levels of PCBs as a
terminal electron acceptor to maintain a growing population.
Quensen et al. (1990) reported that dechlorination of 500 ppm spike concentrations of
Aroclor 1242, 1248, 1254, and 1260 by microorganisms from PCB-contaminated sediments in
the Hudson River and Silver Lake occurred primarily at the meta and para positions; ortho-
substituted MCBs and DiCBs increased in concentration. Significant decreases over the
incubation period (up to 50 weeks) were reported for dioxin-like PCBs 156, 167, 170, 180, and
189. Of the four dioxin-like TCBs and PeCBs detected in the Aroclor spikes (IUPAC Nos. 77,
105, 114, and 118), all decreased significantly in concentration, with the possible exception of
PeCB 114 in the Aroclor 1260-spiked sediment.
Nies and Vogel (1990) reported similar results with Hudson River sediments incubated
anaerobically and enriched with acetone, methanol, or glucose. Approximately 300 ppm of
Aroclor 1242 (31 mol percent TCBs, 7 mol percent PeCBs, and 1 mol percent HxCBs) were
added to the sediments prior to incubation for 22 weeks under an N2 atmosphere. Significant
dechlorination was observed, primarily at the meta and para positions on the higher-chlorinated
congeners (TCBs, PeCBs, and HxCBs), resulting in the accumulation of lower-chlorinated,
primarily ortho-substituted mono- through tri-substituted congeners. No significant
dechlorination was observed in the control samples (samples containing no added organic
chemical substrate and samples that were autoclaved).
Bedard and May (1996) also reported similar findings in the sediments of Woods Pond,
which was believed to be contaminated with Aroclor 1260. Significant decreases in the sediment
concentrations of PCBs 118, 156, 170, and 180 (relative to their concentrations in Aroclor 1260)
were observed. No increases or decreases were reported for the other dioxin-like PCBs.
Bedard et al. (1995) demonstrated that it is possible to stimulate substantial microbial
dechlorination of the highly chlorinated PCB mixture Aroclor 1260 in situ with a single addition
of 2,6-dibromobiphenyl. The investigators added 365 g of 2,6-dibromobiphenyl to 6-ft-diameter
submerged caissons containing 400 kg sediment (dry weight) and monitored the change in PCB
congener concentrations for a period of one year. At the end of the observation period, the
MCBs through HxCBs decreased by 74% in the top of the sediment and by 69% in the bottom.
The average number of chlorines per molecule dropped 21%, from 5.83 to 4.61, with the largest
reduction observed in meta chlorines (54% reduction) followed by para chlorines (6%). The
dechlorination stimulated by 2,6-dibromobiphenyl selectively removed meta-chlorines positioned
next to other chlorines.
The findings of these latter studies are significant, because removal of meta and para
chlorines from the dioxin-like PCBs should reduce their toxicity and bioaccumulative potential
10-38
-------
and also lead to the formation of lower-chlorinated congeners that are more amenable to aerobic
biodegradation.
Van Doit and Bedard (1991) reported the first experimental demonstration of biologically
mediated ortho-dechlorination of aPCB and stoichiometric conversion of thatPCB congener
(2,3,5,6-TCB) to lower-chlorinated forms. In that study, 2,3,5,6-TCB was incubated under
anaerobic conditions with unacclimated methanogenic pond sediment for 37 weeks, with
reported dechlorination to 2,5-DiCB (21%); 2,6-DiCB (63%); and 2,3,6-TrCB (16%).
10.5.2. Photochemical Transformation of Other PCBs
Photolysis and photo-oxidation may be major pathways for destruction of PCBs in the
environment. Research reported to date and summarized below indicates that ortho-substituted
chlorines are more susceptible to photolysis than are meta- and para-substituted congeners; thus,
photolytic formation of more toxic dioxin-like PCBs may occur. Oxidation by hydroxyl radicals,
however, apparently occurs preferentially at the meta and para positions, resulting in a net
decrease rather than a net increase in the environmental load of dioxin-like PCBs.
On the basis of the data available in 1983, Leifer et al. (1983) concluded that all PCBs,
especially the higher-chlorinated congeners and those that contain two or more chlorines in the
ortho position, photodechlorinate. In general, as the chlorine content increases, the photolysis
rate increases. More recently, Lepine et al. (1992) exposed dilute solutions (4 ppm) of Aroclor
1254 in cyclohexane to sunlight for 55 days in December and January. Congener-specific
analysis indicated that the amounts of many higher-chlorinated congeners, particularly mono-
ortho-sub stituted congeners, decreased, whereas those of some lower-chlorinated congeners
increased. The results for the dioxin-like PCBs indicated a 43.5% decrease in the amount of
PeCB 114, a 73.5% decrease in the amount of HxCB 156, and a 24.4% decrease in the amount of
HxCB 157. However, TCB 77 and PeCB 126 (the most toxic of the dioxin-like PCB congeners),
which were not detected in unirradiated Aroclor 1254, represented 2.5% and 0.43%, respectively,
of the irradiated mixture.
With regard to photo-oxidation, Atkinson (1987) and Leifer et al. (1983), using assumed
steady-state atmospheric OH concentrations and measured oxidation rate constants for biphenyl
and MCB, estimated atmospheric decay rates and half-lives for gas-phase PCBs. Atmospheric
transformation was estimated to proceed most rapidly for those PCB congeners containing either
a small number of chlorines or those containing all or most of the chlorines on one ring. Kwok et
al. (1995) extended the work of Atkinson (1987) by measuring the OH radical reaction rate
constants for 2,2'-, 3,3'-, and 3,5-DiCB. These reaction rate constants, when taken together with
Atkinson's measurements for biphenyl and MCB and the estimation method described in
Atkinson (1991), were used to generate more reliable estimates of the gas-phase OH radical
10-39
-------
reaction rate constants for the dioxin-like PCBs. The persistence of the PCB congeners increased
with increasing degree of chlorination. Table 10-20 presents these estimated rate constants and
the corresponding tropospheric lifetimes and half-lives.
Table 10-20. Estimated tropospheric half-lives of dioxin-like PCBs with
respect to gas-phase reaction with the OH radical
Congener group
Tetrachlorobiphenyls
Pentachlorophenyls
Hexachlorobiphenyls
Heptachlorobiphenyls
Dioxin-like
congener
3,3',4,4'-TCB
3,4,4',5-TCB
2,3,3',4,4'-PeCB
2,3,4,4',5-PeCB
2,3',4,4',5-PeCB
2',3,4,4',5-PeCB
3,3',4,4',5-PeCB
2,3,3',4,4',5-HxCB
2,3,3',4,4',5'-HxCB
2,3',4,4',5,5'-HxCB
3,3',4,4',5,5'-HxCB
2,2',3,3',4,41,5-HPCB
2,2',3,4,4',5,51-HPCB
2,3,3',4,4',5,5'-HpCB
Estimated
OH reaction
rate constant
(10 12 cm3/
molecule-sec)
0.583
0.71
0.299
0.383
0.299
0.482
0.395
0.183
0.214
0.214
0.266
0.099
0.099
0.125
Estimated
tropospheric
lifetime
(days)3
20
17
40
31
40
25
30
65
56
56
45
121
121
95
Estimated
tropospheric
half-life
(days)3
14
12
28
22
28
17
21
45
39
39
31
84
84
66
Calculated using a 24-hr, seasonal, annual, and global tropospheric average OH radical concentration of
9.7 x 105 molecule/cm3 (Prinnetal., 1995).
Source: Telephone conversation on November 16, 1995, between Roger Atkinson, Air Pollution Research Center,
University of California, and Greg Schweer, Versar, Inc. (based on Atkinson, 1991, and Kwok et al., 1995).
Sedlak and Andren (1991) demonstrated in laboratory studies that OH radicals generated
with Fenton's reagent rapidly oxidized PCBs (2-MCB and the DiCBs through PeCBs present in
Aroclor 1242) in aqueous solutions. The results indicated that the reaction occurs via addition of
a hydroxyl group to one nonhalogenated site; reaction rates are inversely related to the degree of
chlorination of the biphenyl. The results also indicated that meta and para sites are more reactive
than ortho sites due to stearic hindrance effects. On the basis of their kinetic measurements and
reported steady-state aqueous system OH concentrations or estimates of OH radical production
rates, the authors estimated environmental half-lives for dissolved PCBs (MCB through
10-40
-------
octachlorobiphenyl) in fresh surface water and in cloud water to be 4 to 11 days and 0.1 to 10
days, respectively.
10.6. PAST USE OF COMMERCIAL PCBs
An estimated total of 1.5 million metric tons of PCBs were produced worldwide
(DeVoogt and Brinkman, 1989). Slightly more than one-third of these PCBs (568,000 metric
tons) were used in the United States (Versar, Inc., 1976). Although the focus of this section is on
past uses of PCBs within the United States, it is necessary to note that the use and disposal of
PCBs in many countries, coupled with the persistent nature of PCBs, have resulted in their
movement and presence throughout the global environment. The ultimate sink of most PCBs
released to the environment is aquatic sediments. Currently, however, large quantities of PCBs
are estimated to be circulating between the air and water environments or are present in landfills
and dumps, some of which may offer the potential for re-release of PCBs into the air. Tanabe
(1988) presented a global mass balance for PCBs that indicated that as of 1985, 20% of the total
PCBs produced were present in seawater, whereas only 11% were present in sediments (see
Table 10-21). Nearly two-thirds of total global PCB production was estimated by Tanabe to still
be in use in electrical equipment or to be present in landfills and dumps.
As discussed in Section 10.2, an estimated 568,000 metric tons of PCBs were sold in the
United States between 1930 and 1975 (Versar, Inc., 1976). Table 10-22 presents annual
estimates of domestic sales by year for each Aroclor from 1957 to 1974. Estimates of PCB usage
in the United States by usage category from 1930 to 1975 are presented in Table 10-23. Prior to
voluntary restrictions by Monsanto Corporation in 1972 on sales for uses other than "closed
electrical systems," approximately 13% of the PCBs were used in "semi-closed applications,"
and 26% were used in "open-end applications." Most of the usage for semi-closed and open-end
applications occurred between 1960 and 1972 (Versar, Inc., 1976).
Table 10-24 presents estimates of the amounts of individual Aroclors that were directly
released to the environment (water, air, or soil) between 1930 and 1974. Because detailed usage
data were not available for the period 1930 to 1957, Versar, Inc. (1976) assumed that the usage
pattern for this period followed the average pattern for the period of 1957 to 1959. The basic
assumption used by Versar in deriving these estimates was that PCBs were released on the order
of 5% of those used in closed electrical systems, 60% of those used in semi-closed applications,
and 25% of those used for plasticizers and that 90% of PCBs used for miscellaneous industrial
uses had escaped. The reliability of these release estimates was assumed to be ±30%.
Versar, Inc. (1976) also estimated that 132,000 metric tons of PCBs were landfilled. This
total comprised 50,000 metric tons from capacitor and transformer production wastes, 36,000
metric tons from disposal of obsolete electrical equipment, and 46,000 metric tons from disposal
10-41
-------
Table 10-21. Estimated PCB loads in the global environment as of 1985
Environment
Terrestrial and coastal
Air
River and lake water
Seawater
Soil
Sediment
Biota
Total
Open ocean
Air
Seawater
Sediment
Biota
Total
Total load in environment
Degraded and incinerated
Land-stocked3
World production
PCB load
(metric tons)
500
3,500
2,400
2,400
130,000
4,300
143,100
790
230,000
110
270
231,170
374,000
43,000
783,000
l,200,000b
Percentage of
PCB load
0.13
0.94
0.64
0.64
35
1.1
39
0.21
61
0.03
0.07
61
100
Percentage of
world production
31
4
65
100
aStill in use in electrical equipment and other products, and deposited in landfills and dumps.
bThis value is from Tanabe (1988). DeVoogt and Brinkman (1989) estimated worldwide production to have been
1,500,000 metric tons.
Source: Tanabe (1988).
of material from open-end applications. An estimated additional 14,000 metric tons of PCBs,
although still "in service" in various semi-closed and open-end applications in 1976, were
ultimately destined for disposal in landfills.
An estimated 3,702 kg of TEQP-WHO98 were released directly to the U.S. environment
between 1930 and 1977 (see Table 10-25). These estimates are based on the Aroclor release
estimates presented in Table 10-22 and the mean TEQP-WHO98 concentrations in Aroclors
presented in Table 10-3.
10-42
-------
Table 10-22. Estimated domestic sales of aroclors and releases of PCBs, 1957-1974 (metric tons)
Year
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
TOTAL
% of Total
Estimated domestic sales
Aroclor
1016
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1,512
9,481
10,673
9,959
31,625
8.8
Aroclor
1221
10
7
115
47
43
64
164
270
167
239
200
62
230
670
1,005
78
16
26
3,413
0.9
Aroclor
1232
89
51
109
70
109
102
6
6
3
7
11
41
124
118
78
0
0
0
924
0.3
Aroclor
1242
8,265
4,737
6,168
8,254
8,993
9,368
8,396
10,692
14,303
17,943
19,529
20,345
20,634
22,039
9,970
330
2,812
2,815
195,593
54.2
Aroclor
1248
807
1,161
1,535
1,282
1,825
1,571
2,274
2,376
2,524
2,275
2,134
2,220
2,563
1,847
97
366
0
0
26,857
7.4
Aroclor
1254
2,023
3,035
3,064
2,761
2,855
2,869
2,681
2,849
3,509
3,191
3,037
4,033
4,455
5,634
2,114
1,585
3,618
2,805
56,118
15.6
Aroclor
1260
3,441
2,713
3,002
3,325
2,966
2,991
3,459
3,871
2,645
2,665
2,911
2,382
2,013
2,218
782
138
0
0
41,522
11.5
Aroclor
1262
14
83
163
148
164
196
188
202
253
348
381
327
323
464
0
0
0
0
3,254
0.9
Aroclor
1268
0
33
46
86
72
95
129
86
89
129
130
127
136
150
0
0
0
0
1,308
0.4
Total
PCB
releases
14,649
11,820
14,202
15,973
17,027
17,256
17,297
20,352
23,493
26,797
28,333
29,537
30,478
33,140
15,558
11,978
17,119
15,605
360,614
100
o
oo
Source: Versar, Inc. (1976).
-------
Table 10-23. Estimated U.S. usage of PCBs by use category, 1930-1975
Use class
Closed electrical
systems
Semi -closed
applications
Open-end
applications
Use category
Capacitors
Transformers
Heat transfer
fluids
Hydraulics and
lubricants
Plasticizer uses
Carbonless copy
paper
Misc. industrial
Petroleum
additives
TOTAL
Amount used
(1,000 metric
tons)
286
152
9
36
52
20
12
1
568
Percent of total
usage
50.3
26.8
1.6
6.3
9.2
3.5
2.1
<1
100
Reliability of
estimate (%)
±20
±20
±10
±10
±15
±5
±15
±50
Source: Versar, Inc. (1976).
10-44
-------
Table 10-24. Estimated direct releases of Aroclors to the U.S. environment,
1930-19743 (metric tons)
Year
1930-56
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
TOTAL
% of Total
Estimated environmental releases
Aroclor
1016
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
76
474
534
498
1,582
2.0
Aroclor
1242
8,486
903
649
1,042
1,340
1,852
1,811
1,655
2,085
2,689
3,180
3,376
3,533
4,165
4,569
1,466
22
141
141
43,105
54.9
Aroclor
1248
2,447
319
483
724
556
792
659
935
980
1,025
876
814
853
993
697
51
0
0
0
13,204
16.8
Aroclor
1254
2,269
307
416
518
449
587
554
529
555
660
566
525
733
985
1,168
325
104
181
140
11,571
14.7
Aroclor
1260
1,614
423
355
507
540
611
571
682
755
497
472
504
433
452
474
121
9
0
0
9,020
11.5
Total PCB
releases
14,816
1,952
1,903
2,791
2,885
3,842
3,595
3,801
4,375
4,871
5,094
5,219
5,552
6,595
6,908
2,039
609
856
779
78,482
100
T)oes not include an additional 132,000 metric tons estimated to have been landfilled during this period.
Source: Versar, Inc. (1976).
10-45
-------
Table 10-25. Estimated releases of dioxin-like PCB TEQs to the U.S.
environment, 1930-1977
Aroclor
1016
1221
1232
1242
1248
1254
1260
1262
1268
Percent of
U.S. sales3
(1957-1974)
12.88
0.96
0.24
51.76
6.76
15.73
10.61
0.83
0.33
Estimated
PCB releases
(1930-1974)"
(metric tons)
1,582
—
—
43,103
13,205
11,572
9,019
—
—
Estimated
mean TEQP-
WHO98
concentration0
(mg/kg)
d
0.328
—
7.47
16.87
125.94
188.45
—
—
TOTAL
Estimated
total TEQp-
WHO98
released
(kg)
d
—
—
322
223
1,457
1,700
—
—
3,702
"Sales during the period 1957-1974 constituted 63% of all PCB sales during 1930-1977. Sales data for
individual Aroclors are not available for years prior to 1957; however, sales of Aroclors 1221, 1232, 1262, and
1268 were minor even prior to 1957.
bFrom Table 10-24.
Trom Table 10-3 (assumes nondetect values are zero).
dData are available for only a few samples of Aroclor 1016 where only two dioxin-like PCB congeners were
detected. The total TEQP-WHO98 released is less than 0.01 kg.
- = Indicates that release estimates were not made because of relatively low usage amounts
Source: Versar, Inc. (1976).
10-46
-------
11. RESERVOIR SOURCES OF CDDs/CDFs AND DIOXIN-LIKE PCBs
National CDD/CDF source inventories have been conducted in several nations, including
the United Kingdom (U.K.), the Netherlands, Germany, Austria, and Sweden, to characterize
emissions from various source categories and estimate annual CDD/CDF emissions to air (and
sometimes other media). These inventories focused mainly on emissions from primary sources
(i.e., emissions from the site or process where the CDDs/CDFs are formed).
The authors of these inventories (Rappe, 1991; Harrad and Jones, 1992; Bremmer et al.,
1994; Thomas and Spiro, 1995, 1996; Eduljee and Dyke, 1996; Jones and Alcock, 1996; Duarte-
Davidson et al., 1997) indicated that the annual estimates of releases to air provided in these
inventories may, for several reasons, be underestimates of actual emissions. First, on an
empirical basis, estimates of the amounts of CDDs/CDFs deposited annually from the
atmosphere were greater than the estimates of annual CDD/CDF emissions to the atmosphere.
Second, because the emission test data were limited, the inventories may have underestimated
releases from known sources or may not have identified all primary sources. Third, the
investigators were not able to reliably quantify emissions from potential reservoir (secondary)
sources, including volatilization of CDDs/CDFs from PCP-treated wood, volatilization from soil,
and resuspension of soil particles. Relatively little research of either a monitoring or a
theoretical nature has been performed to identify reservoir sources and to quantify the magnitude
of current or potential future releases from these sources.
This chapter presents background information on the major reservoir sources of
CDDs/CDFs and PCBs, including the potential magnitude (mass) of CDDs/CDFs and PCBs in
each reservoir, the chemical/physical mechanisms responsible for releases of these compounds,
and estimates of potential annual releases from each reservoir, if such estimates are feasible.
11.1. POTENTIAL RESERVOIRS
Chapters 2 through 10 of this document discuss both known and suspected sources of
releases of newly formed dioxin-like compounds to the environment in the United States. Once
released into the open environment, CDDs, CDFs, and PCBs partition to air, soils, water,
sediments, and biota according to both the nature of the release and the contaminant's chemical
and physical properties.
For this analysis, reservoirs are defined as materials or places that contain previously
formed CDDs/CDFs or dioxin-like PCBs and have the potential for redistributing and circulating
these compounds into the environment. Potential reservoirs include soils, sediments, biota,
water, and some anthropogenic materials. The atmosphere is not considered a reservoir, but
serves as a medium for transporting and distributing CDDs and CDFs over large geographical
11-1
-------
areas. Atmospheric deposition of CDDs/CDFs to the earth's surface results in an exchange of
CDDs/CDFs from the atmosphere to surface waters and soils.
Dioxin-like compounds are sequestered by a reservoir only until physical processes
cause these contaminants to become released into the open environment over a defined time and
space. When this occurs, reservoirs become sources of dioxin-like compounds in the circulating
environment. Reservoir sources are not included in the quantitative inventory of contemporary
sources because they do not involve original releases but rather the re-circulation of past releases
from anthropogenic sources. They can, however, contribute to human exposure and, therefore,
are important to consider.
The rate of movement from one environmental medium to another is termed "flux," and
it refers to the direction and magnitude of flow and exchange over a reference time period and
space. Figure 11-1 presents a conceptual diagram of flux and exchange of dioxin-like
compounds to multiple environmental compartments such as soils, water, air, sediments, and
biota. This dynamic system consists of fluxes in and out of the atmosphere as well as other
exchanges between reservoirs and the atmosphere (recall that the atmosphere is not formally
considered an environmental reservoir, rather it is a transport media for dioxin-like compounds).
Movement of dioxin-like compounds between media can be induced by the physical processes of
volatilization, wet and dry atmospheric particle and vapor deposition, adsorption, erosion and
runoff, resuspension of soils into air, and resuspension of sediments into water.
p. Deep Soil
Deep
Sediment'
Fluxes Among Dioxin Reservoirs
Figure 11-1. Fluxes among environmental reservoirs.
11-2
-------
11.2. CHARACTERIZATION OF RESERVOIR SOURCES
This section is organized according to each principle reservoir, i.e., soil, water, sediment,
and biota. The discussion of each environmental reservoir provides information in three parts:
(1) the potential amount or mass of dioxin-like compounds contained within the reservoir, (2) the
chemical/physical mechanisms responsible for releases of these compounds back into the open
and circulating environment, and (3) estimates of potential annual releases from the reservoir if
such estimates are feasible, given the available state of knowledge. Although anthropogenic
structures (e.g., PCP-treated fenceposts, telephone poles) are potential reservoir sources, they are
not discussed here because they are covered in Chapter 8 (the most detailed discussion is on
PCP, Section 8.3.8).
11.2.1. Soil
11.2.1.1. Potential Mass of Dioxin-Like Compounds Present
In estimating burdens for the U.K., Harrad and Jones (1992) and Duarte-Davidson et al.
(1997) assumed that the majority of CDDs/CDFs in soil is present in the top 5 cm (except
possibly in cropland, where they may be present at greater depths due to plowing) and that the
soil density is 1,000 kg/m3. Coupling these assumptions with the rural and urban U.S. surface
areas and TEQ concentrations yielded soil burden estimates of 1,350 kg TEQDF-WHO98 (1,530
kg I-TEQDF) in rural soils and 220 kg TEQDF-WHO98 (250 kg I-TEQDF) in urban soils in the
United States. These calculations are not definitive, and only serve to indicate approximate
amounts of CDDs/CDFs that may be contained in soils in rural and urban areas of the United
States.
Higher concentrations of CDDs/CDFs for background urban and rural soils may be
present in soils underlain by municipal and industrial waste and in soils at contaminated
industrial sites. Higher concentrations may also be present in the soils of areas that have been
treated with pesticides contaminated with CDDs/CDFs. The lack of comprehensive data on
CDD/CDF concentrations in these soils, as well as the lack of data on the mass of these soils
nationwide, precludes estimating total national soil burdens of CDDs/CDFs. Because of the lack
of data, it is not possible to estimate current soil burdens of CDDs/CDFs associated with past
pesticide use; however, estimates can be made of the total mass of CDD/CDF TEQs that have
been applied to soil from past use of the pesticides 2,4-dichlorophenoxyacetic acid (2,4-D) and
2,4,5-trichlorophenoxyacetic acid (2,4,5-T).
2,4-D (and its salts and esters) are widely used as post-emergence herbicides in
agricultural and nonagricultural settings in the United States for control of broadleaf weeds and
brush. Commercial production of 2,4-D in the United States started in 1944 (Esposito et al.,
1980), and it has been in large-scale, large-volume commercial use for many years (U.S. EPA,
1975). In terms of annual volume, 2,4-D ranks among the top 10 pesticides used in the United
11-3
-------
States (U.S. EPA, 1994d, 1997e). Table 11-1 presents a compilation of domestic production,
sales, and usage volumes for 2,4-D and its salts and esters.
Table 11-1. Historical production, sales, and usage of 2,4-dichlorophenoxy-
acetic acid (2,4-D) (metric tons)a
Year
2000
1998/99
1996/97
1994/95
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
1966
2,4-D, acid
Production
volume
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
5,859
6,164
5,763
—
—
—
—
—
—
24,948b
—
19,766
21,354
35,953
34,990
30,927
Sales
volume
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
3,275
3,137
6,187
—
—
—
—
—
—
—
5,619
7,159
8,521
10,352
15,432
12,710
Domestic usage/
disappearance
23,600-28,100°
23,600-28, 100d
23,600-27,200d
21,800-26,300e
16,800-20,400f
16,800-20,400f
18,100-29,500g
18,100-29,500g
18,100-29,500h
23,600-30,400'
23,600-30,400J
-
-
-
-
-
-
-
~
-
-
17,418k
-
-
-
21,772k
15,700k
-
-
-
-
28,985'
2,4-D, esters and salts
(as reported)b
Production
volume
—
—
—
—
—
—
—
—
—
—
—
8,618
—
—
7,702
8,762
8,987
11,313
11,874
8,958
12,552
10,913
16,134
6,558
13,400
10,192
—
—
25,854
42,690
37,988
32,895
Sales
volume
—
—
—
—
—
—
—
—
—
—
—
12,150
0
0
8,234
8,400
8,002
11,147
13,453
9,256
10,196
7,813
13,414
5,991
13,698
10,899
18,654
19,920
20,891
30,164
29,300
25,075
11-4
-------
Table 11-1. Historical production, sales, and usage of 2,4-dichlorophenoxy-
acetic acid (2,4-D) (metric tons)a (continued)
Year
1965
1964
1963
1962
1961
1960
1959
1958
1957
1956
1955
1954
1953
1952
1951
1950
1949
1948
1947
1946
1945
2,4-D, acid
Production
volume
28,721
24,364
21,007
19,503
19,682
16,413
13,282
14,036
15,536
13,079
15,656
—
11,761
13,933
—
6,421
6,852
9,929
2,553
2,479
416
Sales
volume
11,816
11,343
9,446
7,716
7,591
—
7,240
6,234
6,871
6,465
5,924
4,838
—
—
—
4,301
2,991
4,152
2,320
2,330
286
Domestic usage/
disappearance
22,906'
19,958'
15,059'
16,284'
14,107'
14,107'
15,468'
9,662'
-
-
-
~
-
-
~
~
-
-
-
~
-
2,4-D, esters and salts
(as reported)b
Production
volume
28,740
24,660
20,178
16,831
16,683
15,436
12,438
11,295
12,392
9,635
13,390
10,268
10,733
11,358
—
5,274
5,829
2,458
1,468
515
—
Sales
volume
21,454
18,263
16,333
13,075
12,533
13,661
7,070
5,649
7,125
7,294
8,121
6,886
8,855
9,637
—
3,219
3,211
1,598
1,108
81
—
"All values are from USITC (1946-1994) unless footnoted otherwise.
bNo data were reported for domestic usage/disappearance of 2,4-D esters and
cSource: U.S. EPA(1991c).
dSource: U.S. EPA (2000a).
eSource: U.S. EPA(1997e).
fSource: U.S. EPA (1994d).
gSource: U.S. EPA(1992g).
hSource: U.S. EPA(1991d).
'Source: U.S. EPA(1990c).
JSource: U.S. EPA(1988d).
kSource: U.S. EPA (1975).
'Source: USDA(1970).
— = Not reported to avoid disclosure of proprietary data
~ = No information given
salts.
11-5
-------
As described in Section 8.3.8, CDDs/CDFs were detected in several formulations of 2,4-
D and its derivatives during analyses performed to comply with EPA's 1987 Data Call-In (DCI)
for CDDs/CDFs. Although the analytical results of these tests indicated that CDDs/CDFs were
seldom above the regulatory limits of quantification (LOQs) established by EPA for the DCI,
several registrants detected and quantified CDDs/CDFs at lower LOQs. The results of these
tests are summarized in Table 8-28 (Chapter 8). The average TEQ in these tests was 1.1 |_ig
TEQDF-WHO98/kg (0.7 [ig I-TEQDF/kg). Schecter et al. (1997) reported similar concentrations in
2,4-D samples manufactured in Europe and Russia; lower levels were observed in U.S. products.
Some of the results from Schecter et al. are presented in Table 8-34 (Chapter 8).
If it is assumed that the EPA DCI results are typical of CDD/CDF levels in 2,4-D
pesticides over the past 20 yr and that the average annual use of these pesticides in the United
States has been approximately 25,000 metric tons, then the estimated CDD/CDF TEQ released
to the environment from 2,4-D use during the period of 1975 to 1995 was 550 g TEQDF-WHO98
(350gI-TEQDF).
2,4,5-T was used in the United States for a variety of herbicidal applications until the late
1970s to early 1980s. The major use of 2,4,5-T (about 41% of annual usage) was for control of
woody and herbaceous weed pests on rights-of-way. The other major herbicidal uses were
forestry (28% of usage), rangeland (20% of usage), and pasture (5% of usage). Uses of 2,4,5-T
for home or recreation areas and for lakes, ponds, and ditches were suspended by EPA in 1970;
rights-of-way, forestry, and pasture uses were suspended by EPA in 1979; and all uses were
canceled in 1983.
Table 11-2 presents a compilation of domestic production, sales, and usage volumes for
2,4,5-T and its salts and esters. As shown in Table 11-2, production and use of 2,4,5-T generally
increased each year following its introduction in the 1940s until the late 1960s. Production,
sales, and usage information for the 1970s are generally not available but are reported to have
steadily declined during that decade (Federal Register, 1979; Esposito et al., 1980).
Some information is available on the 2,3,7,8-TCDD content of 2,4,5-T, but little
information is available on the concentrations of the other 2,3,7,8-substituted CDD/CDFs that
may have been present. Plimmer (1980) reported that 2,3,7,8-TCDD concentrations as high as
70,000 |ag/kg were detected in 2,4,5-T during the late 1950s. In a study of 42 samples of 2,4,5-T
manufactured before 1970, Woolson et al. (1972) found 500 to 10,000 [ig/kg of TCDDs in 7
samples, and another 13 samples contained 10,000 to 100,000 |-ig/kg of TCDDs. HxCDDs were
found in 4 samples at levels between 500 and 10,000 |-ig/kg and in 1 sample at a concentration
exceeding 10,000 |-ig/kg but less than 100,000 |-ig/kg. The detection limit in the study was 500
11-6
-------
Table 11-2. Historical production, sales, and usage of 2,4,5-trichlorophen-
oxyacetic acid (2,4,5-T) (metric tons)a
Year
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
1972
1971
1970
1969
1968
1967
1966
1965
1964
1963
1962
2,4,5-T
Production
volume
—
—
—
—
—
—
—
—
—
—
—
—
3,200-4, 100d
—
—
—
—
—
—
—
—
—
2,268
7,951
6,601
7,026
5,262
5,186
4,123
3,796
Sales
volume
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
1,329
757
2,312
—
1,691
1,928
1,021
Domestic usage/
disappearance
-
-
-
-
-
-
-
-
-
-
-
900C
-
3,200e
4,100d
-
3,200e
900f
-
-
694g
3,200e
-
~7,000h'1
~7,000h'1
7,756h
3,266h
4,037h
3,266h
3,674h
2,4,5-T, esters and salts
(as reported)1"
Production
volume
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
5,595
5,273
19,297
12,333
8,191
6,131
5,880
4,543
4,765
Sales
volume
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
1,675
3,272
2,576
15,021
11,657
4,553
5,977
3,128
2,585
2,543
11-7
-------
Table 11-2. Historical production, sales, and usage of 2,4,5-trichlorophen-
oxyacetic acid (2,4,5-T) (metric tons)a (continued)
Year
1961
1960
1959
1958
1957
1956
1955
1954
1953
1952
1951
1950
1949
1948
1947
1946
1945
2,4,5-T
Production
volume
3,134
2,874
2,516
1,668
2,419
2,345
1,327
1,223
2,395
1,583
—
852
—
—
—
—
—
Sales
volume
1,196
—
1,039
692
—
816
662
639
—
—
—
297
—
—
—
—
—
Domestic usage/
disappearance
2,449h
2,676h
2,495h
l,724h
-
-
l,300e
-
-
-
l,100e
-
-
-
-
-
-
2,4,5-T, esters and salts
(as reported)b
Production
volume
3,536
3,594
3,644
2,372
3,098
3,196
1,720
1,761
2,443
1,423
—
—
—
—
—
—
—
Sales
volume
2,372
1,891
1,843
1,151
1,337
1,473
1,077
615
1,817
569
—
—
—
—
—
—
—
aAll values are from USITC (1946-1994) unless footnoted otherwise.
bNo data were reported for domestic usage/disappearance of 2,4,5-T esters and salts.
cSource: Esposito etal. (1980).
dSource: Federal Register (1979).
eSource: Thomas and Spiro (1995).
Source: U.S. EPA (1977).
BSource: USDA (1971); reflects farm usage only.
hSource: USDA (1970); values include military shipments abroad.
'Source: Kearney et al. (1973) reports slightly lower domestic consumption for the years 1967 and 1968 than for
1966.
— = Not reported to avoid disclosure of proprietary data
~ = No information given
The average 2,3,7,8-TCDD concentration in 200 samples of Agent Orange, a defoliant
containing about a 50/50 mixture of the butyl esters of 2,4,5-T and 2,4-D that was used by the
U.S. Air Force in Vietnam, was 1,910 |-ig/kg (Kearney et al., 1973). Of the 200 samples, 64
(32%) contained more than 500 |-ig/kg of 2,3,7,8-TCDD, with the highest concentration reported
to be 47,000 [ig/kg. Storherr et al. (1971) reported detecting 2,3,7,8-TCDD at concentrations
11-8
-------
ranging from 100 to 55,000 |-ig/kg in five samples of 2,4,5-T. Kearney et al. (1973) reported that
production samples of 2,4,5-T obtained from the three principal 2,4,5-T manufacturers in 1971
contained 2,3,7,8-TCDD at levels of <100 |-ig/kg, 100 ng/kg, and 2,300 [ig/kg.
A 1975 survey of 10 lots of a commercial formulation containing 2,4,5-T showed 2,3,7,8-
TCDD concentrations ranging from 10 to 40 |-ig/kg (Lewert, 1976). Analyses by EPA of 16
technical-grade 2,4,5-T samples from five different manufacturers revealed 2,3,7,8-TCDD
contents ranging from <10 to 25 |ig/kg (Federal Register, 1979). Schecter et al. (1997) reported
the analytical results of one sample of 2,4,5-T purchased from Sigma Chemical Co. (product
number T-5785, lot number 16H3625). The results, presented in Table 11-3, indicate a total
TEQDF-WHO98 concentration of 3.26 [ig/kg (2.88 [ig I-TEQDF/kg).
Because of the wide variability (three orders of magnitude) in the limited available
information on the 2,3,7,8-TCDD content of 2,4,5-T (particularly the 2,4,5-T used in the 1950s)
and incomplete information on domestic usage, it is difficult to reliably estimate the amount of
2,3,7,8-TCDD that was released to the U.S. environment as a result of 2,4,5-T use. A very
uncertain estimate can be made using the following assumptions: (1) average annual
consumption during the 1950s, 1960s, and 1970s was 2,000, 4,000, and 1,500 metric tons/yr,
respectively; and (2) the average 2,3,7,8-TCDD concentration in 2,4,5-T used over these three
decades was 10,000 [ig/kg in the 1950s, 4,000 [ig/kg in the 1960s, and 100 [ig/kg in the 1970s.
Based on these assumptions, the very uncertain estimate of 2,3,7,8-TCDD input from 2,4,5-T use
over the period of 1950 to 1979 was 36,000 g.
Another contributing source to the soil reservoir is CDD/CDF in sewage sludge applied
to land (i.e., surface disposal or land farming), estimated to have been 75 g TEQDF-WHO98 (103
g I-TEQDF) in 1995 (see Section 8.4.1 for details). If this same amount of TEQ had been applied
each year during the period of 1975 to 1995, the total amount applied would have been 1,500 g
TEQDF-WH098 (2,000 g I-TEQDF).
11.2.1.2. Mechanisms Responsible for Releases from Surface Soils
The atmospheric deposition of dioxin-like compounds is believed to be the primary
means whereby surface soils, in general, become contaminated with dioxin-like compounds.
CDDs/CDFs and PCBs are highly lipid soluble and have low volatility, and they tend to partition
to soil rather than into air or water. Once present in or on soils, physical/chemical and biological
mechanisms (photolysis and biodegradation) can slowly alter the composition and amount of
compound present. Studies indicate that the dioxin-like compounds (particularly the higher-
chlorinated CDDs/CDFs) exhibit little downward mobility to lower soil depths after they are
deposited in or on soil (Puri et al. 1989; Freeman and Schroy, 1985; Orazio et al., 1992;
Paustenbach et al., 1992). However, re-mobilization of the contaminants to the atmosphere is
possible through volatilization and resuspension of soil particles.
11-9
-------
Table 11-3. CDD/CDF concentrations (|ig/kg) in recent sample of
2,4,5-trichloro-phenoxyacetic acid (2,4,5-T)
Congener/congener group
2,4,5-T sample
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
1.69
0.412
0.465
2.28
1.35
18.1
33.9
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
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-HpCDF
OCDF
0.087
0.102
0.183
1.72
0.356
ND (0.012)
0.126
2.9
0.103
3.01
Total 2,3,7,8-CDDa
Total 2,3,7,8-CDFa
Total I-TEQDFa
Total TEQDF-WHO98a
58.2
8.59
2.88
3.05
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDD/CDF
a Calculated assuming nondetect values were zero.
ND = Not detected (value in parenthesis is the detection limit)
~ = No information given
Source: Schecteretal. (1997).
11-10
-------
Young (1983) conducted field studies on the persistence and movement of 2,3,7,8-TCDD
from 1973 to 1979 on a military test area that had been aerially sprayed with 73,000 kg of
2,4,5-T from 1962 to 1970. TCDD levels of 10 to 1,500 ng/kg were found in the top 15 cm of
soil 14 yr after the last application of herbicide at the site. Although actual data were not
available on the amount of 2,3,7,8-TCDD originally applied as a contaminant of the 2,4,5-T, best
estimates indicated that less than 1% of the applied 2,3,7,8-TCDD remained in the soil after
14 yr. Young suggested that photodegradation at the time of and immediately after aerial
application was responsible for most of the disappearance; however, once incorporated into the
soil, the data indicated a half-life of 10 to 12 yr. Similarly, Paustenbach et al. (1992) concluded
that the half-life of 2,3,7,8-TCDD in soils at the surface might be 9 to 15 yr and the half-life
below the surface could be 25 to 100 yr.
Ayris and Harrad (1997) studied the mechanisms affecting volatilization fluxes of several
PCB congeners (PCB 28, 52, 101, 138, and 180) from soil and found positive correlations
between flux and soil temperature, soil moisture content, and soil PCB concentration. For PCBs,
secondary releases from soils (primarily via volatilization) are believed to currently exceed
primary emissions in the U.K. (Harner et al., 1995; Jones and Alcock, 1996). Lee et al. (1998)
quantified PCBs in air samples taken every 6 hr over a 7-day period in the summer at a rural site
in England and found a strong correlation between air temperature and PCB congener
concentrations. The concentrations followed a clear diurnal cycle, thus providing some evidence
that rapid, temperature-controlled soil-to-air exchange of PCBs influences air concentrations and
enables regional/global scale cycling of these compounds.
CDDs/CDFs and PCBs sorbed to soil and urban dust particles can also be moved from
the terrestrial environment to the aquatic environment through contaminated storm water runoff
and surface erosion of contaminated soils. Results of recent research indicate that, for at least
some water bodies, soil erosion and storm water runoff can be an important way of transporting
CDDs/CDFs into the surface water. This is especially true for water bodies lacking nonpoint
sources of dioxin-like compounds. Smith et al. (1995) analyzed CDD/CDF concentrations in
sediment cores, air, precipitation, soil, and storm water runoff in an effort to determine the
contributing sources of these compounds to the lower Hudson River. The mass balance
estimates developed from these data for 1990 to 1993 are as follows: stormwater runoff entering
tributaries (76% of total CDD/CDF input), anthropogenic wastes (19%), atmospheric deposition
(4%), and shoreline erosion (less than 1%). The authors projected the percent contribution of
these same sources for 1970 as anthropogenic wastes (70%), stormwater runoff into tributaries
(15%), atmospheric deposition (15%), and shoreline erosion (0.1%).
Studies conducted by Paustenbach et al. (1996), Mathur et al. (1997) and Fisher et al.
(1999) discerned the relative concentrations of CDDs and CDFs in storm water. Paustenbach et
al. and Mathur et al. reported that stormwater runoff from 15 sites in the San Francisco area
11-11
-------
contained CDD/CDF TEQ at levels ranging from 0.01 to 65 pg I-TEQDF/L; most samples
contained less than 15 pg I-TEQDF/L. The sites differed widely in land use; the highest levels
measured were obtained from an urban but nonindustrialized area. A distinct variability was
noted in the results obtained at the same sampling location during different rain events. The
profiles of CDDs/CDFs in the urban stormwater samples were similar, particularly in samples
collected at the onset of rain events. Stowe (1996) reported similar findings from analyses of
sediments from three stormwater basins collecting runoff from a military base, a city street, and
parking lots.
Fisher et al. (1999) reported that urban runoff samples from eight sites (15 samples) in
the Santa Monica Bay watershed contained CDD/CDF TEQ at levels ranging from 0.7 to 53 pg
I-TEQDF/L (all but one sample were in the range of 0.7 to 10 pg I-TEQDF/L). The samples were
collected in 1988/1989 from continuously flowing storm drains during both dry and storm
periods. The mean concentration measured during storm events, 18 pg I-TEQDF/L, was higher
than concentration observed during dry periods, 1 pg I-TEQDF/L.
Atmospheric deposition of CDDs and CDFs to lakes and the watershed area of water
bodies is another important means of contaminating bottom sediments. For example, Lebeuf et
al. (1996) analyzed sediment cores from different locations in the lower St. Lawrence River
Estuary and the Gulf of St. Lawrence. The congener group profiles found in the samples
indicated that the input of CDDs/CDFs was primarily from the atmosphere. Comparison of the
CDD/CDF concentrations in sediments collected from areas where sediment accumulation is due
primarily to fluvial transport with sediments from areas where sediment accumulation is due
primarily to direct atmospheric deposition onto the water indicates that the contribution of
CDDs/CDFs from direct atmospheric deposition represents less than 35% of the sediment
burden. Thus, the primary source of CDDs/CDFs is emissions to the atmosphere upwind of the
estuary that are deposited within the watershed and subsequently transported downstream by
fluvial waters.
11.2.1.3. Estimated Annual Releases from Soil to Water
Nonpoint sources of CDDs/CDFs to waterways include stormwater runoff from urban
areas and soil erosion in rural areas during storms. Approaches for estimating national loadings
to water for both of these sources are described below. The estimate derived for the potential
annual national loading of CDDs/CDFs in urban runoff to waterways is uncertain, but it suggests
that the loading may be comparable to the contribution from known industrial point sources (at
least 20 g I-TEQDF in 1995). Similarly, the estimate derived for the potential annual national
loading of CDDs/CDFs in rural eroded soils to waterways is uncertain, but it has a stronger
analytical base than does the urban runoff estimate. This loading estimate, however, is
significantly higher than the contribution from known industrial point sources.
11-12
-------
Urban runoff. Few data on CDD/CDF concentrations in urban runoff have been
reported. The most recent and largest data sets were reported in studies conducted in the San
Francisco Bay and Santa Monica Bay regions (Mathur et al., 1997; Fisher et al., 1999). These
studies found a wide range of CDD/CDF levels in samples of stormwater runoff from 23 sites,
varying from 0.01 to 83 pg I-TEQDF/L. The wide variability and limited geographic coverage of
these data preclude derivation of a national emissions estimate at this time. However, by making
a number of assumptions, a preliminary estimate of the potential CDD/CDF magnitude from this
source can be made.
In order to estimate the amount of rainfall in urbanized areas of the conterminous United
States, a Geographic Information System (GIS) analysis was performed to determine the total
area of every U.S. Census urbanized area and the 30-yr annual average rainfall for each of those
areas and to calculate the product of the total areas of urbanized areas with the annual average
rainfall (Lockheed Martin Corp., 1998). This approach yields an estimate of 1.9 x 1014 L/yr. If
it is assumed that urban runoff in the United States averages 1 pg TEQDF-WHO98/L (1 pg I-
TEQDF/L) (i.e., approximately the midpoint of the range reported by Mathur et al., 1997, and
Fisher et al., 1999), this source could contribute a total of 190 g TEQDF-WHO98 or I-TEQDF/yr to
U.S. waterways. No data were available to make similar estimates for PCBs.
A similar analysis was conducted using historical precipitation data from the National
Oceanic and Atmospheric Administration (NOAA, 2004) and metropolitan/urban area statistics
from the 1990 and 2000 census. The 30-year annual average rainfall for each state was
calculated for 1987, 1995, and 2000. An approximation of the urban area for each state was
estimated by summing the acreage for each metropolitan area identified in the 1990 census.
Assuming that the amount of land classified as urban did not change significantly from 1987 to
1990, the urban areas for each state in 1990 were assumed to be equal to those in 1987.
Similarly, an approximation for urban area for each state was estimated by summing the urban
area acreage available from the 2000 census. An approximation of the 1995 urban area for each
State was estimated by taking the average of the 1990 and 2000 estimates. Multiplying the 30-
year average rainfall by the urban area for each state and summing the results provides an
estimated amount of urban runoff for the conterminous United States. The urban runoff was
1.24 x 1014, 1.33 x 1014, and 1.42 x 1014L/year for 1987, 1995, and 2000, respectively.
Applying the emission factors generated above, urban runoff contributed 124, 133, and 142 g I-
TEQDF or TEQDF-WHO98 to U.S. waterways in 1987, 1995, and 2000, respectively. These
numbers are in agreement with the estimate developed using Lockheed Martin Corp. (1998)
data.
Rural soil erosion. Using acreage and erosion factors for cropland provided in the 2001
Annual National Resources Inventory (USDA, 2003), 1.36, 1.07, 0.96, and 0.91 billion metric
tons of soil and rill erosion were generated in 1987, 1992, 1997, and 2001, respectively.
11-13
-------
Likewise, using acreage data for rangeland from USDA (2003) and a soil and rill erosion factor
of 4.2 tons/acre/yr (USDA, 1995), approximately 1.55 billion metric tons of soil and rill erosion
were generated in 1987, 1992, 1997, and 2001. For purposes of estimating values for reference
years 1995 and 2000, it was assumed that the 1995 erosion estimate was the average of soil and
rill erosion estimates developed for 1992 and 1997, and that the 2001 numbers approximated
those generated in 2000. The total amount of eroded soil entering waterways was greater than
this value, because this value did not include soil erosion from construction areas, forests, and
other non-croplands and non-rangelands.
The data summarized in U.S. EPA (2000c) suggest that typical concentrations of
CDDs/CDFs in soils in rural areas is about 2.8 ng TEQDF-WHO98/kg. It is not known how well
this estimate represents eroded soil from cropland and rangeland. If these soils contain an
average of 1 ng TEQDF/kg (i.e., a lower value than the background value for all types of rural
soil), they would have contributed 2,900, 2,600, and 2,500 g TEQDF-WHO98 to the nation's
waterways in 1987, 1995, and 2000, respectively. Given the uncertainties in both the amount of
eroded soil and dioxin levels, these estimates are considered preliminary (i.e., Category D). As
with urban runoff, no data were available to make similar estimates for PCBs.
11.2.1.4. Estimated Annual Releases from Soil to Air
No quantitative estimates of the mass of dioxin-like compounds that may be released to
the atmosphere annually from U.S. soils have been published in the literature and none were
developed for this report. As noted above, the vapor flux of these compounds from soil to air is
dependent on the soil and air concentrations of dioxin-like compounds and the temperature,
moisture content, and organic carbon content of the soil. Most of these parameters are not
characterized well enough for the United States as a whole to enable a reliable estimate to be
made at present. Particle flux is dependent on many factors, including wind speed, vegetative
cover, activity level, particle size, soil type/conditions, moisture content, and particle density.
Through use of models and various assumptions, Kao and Venkataraman (1995) estimated the
fraction of ambient air CDD/CDF concentrations in the upper midwestern section of the United
States that may be the result of atmospheric re-entrainment of soil particles. Similarly, through
use of models and various assumptions, Jones and Alcock (1996) and Harner et al. (1995)
reached tentative conclusions about the relative importance of volatilization of dioxin-like
compounds from soils in the U.K.
Modeling re-entrainment of soil to the atmosphere was conducted by Kao and
Venkataraman (1995). Their model incorporated information on particle sizes, deposition
velocities, and concentrations of CDDs/CDFs in soils. Smaller particulates, with median
diameters ranging from about 0.01 |_im to 0.3 |_im, are primarily formed from combustion sources
when hot vapors condense and through accumulation of secondary reaction products on smaller
11-14
-------
nuclei. Particles at the upper end of this size range will deposit to the ground in several days.
Large or coarse particles, having median diameters of about 8 |am, are generated from wind-
blown dust, sea spray, and mechanically generated particles. CDDs/CDFs absorbed onto
re-entrained soil would be included in this larger particle size. These larger particles have a
lifetime in the atmosphere from a few to many hours.
The fraction of ambient air concentration of CDDs/CDFs that results from soil
re-entrainment was established on the basis of the contribution of crustal sources to the ambient
aerosol. Data on typical crustal soil concentrations in air (15 to 50 |_ig/m3 for rural areas and 5 to
25 |-ig/m3 for urban areas) were combined with data on the average concentrations of
CDDs/CDFs in soils (73 ng/kg for rural, 2,075 ng/kg for urban, and 8,314 ng/kg for industrial
soils) published by Birmingham (1990) for Ontario, Canada, and several U.S. midwestern states.
This analysis estimated the concentrations of CDDs/CDFs in the ambient aerosol that originate
from soils to be 1 x 10"3 to 4 x 10"3 pg/m3 in rural areas and 0.01 to 0.05 pg/m3 in urban areas.
These particulate dioxin concentrations were compared with average total particulate dioxin
levels of 1.36 pg/m3 in Eitzer and Kites (1989) to arrive at the conclusion that soil re-entrainment
could account for only 1 to 4% of the particulate dioxins in the atmosphere in urban areas and
0.1 to 0.3% of those in rural regions (Kao and Venkataraman, 1995).
This information on the size distribution of ambient aerosols and relative CDD/CDF
concentrations in different particle size fractions was integrated with particle size deposition
velocities to estimate the relative contribution to the total mass deposition flux for small and
large particle sizes. Even though re-entrained soil may constitute only a small fraction of the
atmospheric levels of CDDs/CDFs, the contribution of dioxins in re-entrained surface soil to the
total deposition flux could be significant because coarse particles dominate in dry deposition.
Soil re-entrainment could possibly account for as much as 70 to 90% of the total dry deposition
of CDDs/CDFs in urban areas and 20 to 40% in rural regions (Kao and Venkataraman, 1995).
Two approaches were used by Jones and Alcock (1996) to assess the potential
significance of CDD/CDF volatilization from soils: the fugacity quotient concept and a simple
equilibrium partitioning model. The fugacity quotient model compares the fugacity (the
tendency of a chemical to escape from a phase) of individual CDD/CDF compounds in different
environmental media to determine the tendency for these compounds to accumulate in particular
environmental compartments (McLachlan, 1996). Fugacities for individual compounds, by
media, were estimated by Jones and Alcock on the basis of physical/chemical properties of the
compounds as well as the concentrations in the media. In this instance, fugacity quotients were
calculated for air and soil by dividing each compound's fugacity for air by that of soil.
Quotients near 1 indicate equilibrium conditions between media; values greater than 1 represent
a tendency for flux (volatilization) from soil to air, and values less than 1 indicate a net flux to
the soil from the air. The equilibrium partitioning model used by Jones and Alcock predicts the
11-15
-------
maximum (possible "worst case") flux of CDDs/CDFs from soil to the atmosphere. Air phase-
to-soil partition coefficients were calculated using the ratios of soil and air fugacity capacities.
Equilibrium air concentrations were then calculated using typical U.K. soil concentrations for
both urban and rural settings.
From the fugacity quotient model, Jones and Alcock (1996) concluded that the lower-
chlorinated CDDs/CDFs may be close to soil-air equilibrium in the U.K., whereas for other
congeners, soil is a sink rather than a source to the atmosphere. The authors reported that the
equilibrium partitioning model predicted that 0.15 kg I-TEQ volatilizes annually from soil in the
U.K. However, they discounted this estimate and concluded that soil volatilization is unlikely to
be a significant contributor to emissions. The likelihood that these estimates were high was
attributed to the fact that assumptions were made that the concentrations of CDDs/CDFs in air
were zero and the model does not consider the resistance of CDDs/CDFs to volatilize from soil.
Harner et al. (1995) developed a model to predict the long-term fate of PCBs in soils,
with emphasis on soil-to-air exchanges. Using data on levels of PCBs in air, soil, and vegetation
in the U.K., the investigators developed a mass balance model to simulate the fate of PCBs in
U.K. soils from 1935 to 1994. Specifically, monitoring data and physical/chemical property data
were compiled to calculate fugacities for PCB congeners 28, 52, 138, and 153. The model was
designed to provide an order-of-magnitude level of accuracy, due in part to the inherent
variability in the input data. The mass balance equations in the model included a bell-shaped
function for rates of emissions of PCBs, with the maximum emission rate occurring in 1967.
From these emissions rates, fluxes between air and soil over several decades were estimated.
Table 11-4 summarizes the calculated fluxes.
Table 11-4. PCB 138 fluxes predicted by Harner et al. (1995)
Year
1950
1965
1975
1980
1994
Concentration
in air
(pg/m3)
4
280
-
49
6
Fugacity
in air
(Pascals x 10 9)
0.24
1.5
~
~
-
Fugacity
in soil
(Pascals x 10 9)
1.1
12
16
~
8.3
Concentration
in soil
(ng/g)
-
-
-
-
~
Net flux/direction
air -> soil (444 kg/yr)
air- soil (1,000 kg/yr)
soil - air (820 kg/yr)
soil - air (700 kg/yr)
~ = No information given
During the 1960s and 1970s, levels of total PCBs in U.K. soils reached average levels of
approximately 300 |-ig/kg as a result of atmospheric deposition. Because of restrictions on PCB
11-16
-------
use during the last two decades, air concentrations have fallen, and the primary source to the
atmosphere is now believed to be volatilization from soils. The mass balance model estimated a
net flux of 700 kg/yr of total PCBs from soils to the atmosphere in 1994. However, this estimate
is presented with the caveat that the model tends to underestimate the rate of reduction of PCB
concentrations in recent years, which could be attributed to other mechanisms such as
biodegradation, photolysis, and other degradation processes.
11.2.2. Water
11.2.2.1. Potential Mass of Dioxin-Like Compounds Present
The surface area of inland waters (including the Great Lakes) in the United States is
about 359,000 km2 (U.S. DOC, 1995b). Assuming that the mean depth of inland water is 10 m
(Duarte-Davidson et al., 1997), the total inland water volume is approximately 3,600 billion m3.
No compilation of CDD/CDF measurements in inland surface waters was made for this report;
however, if it is assumed that the "typical" value used by Duarte-Davidson et al. for rivers in the
U.K., 38 pg I-TEQDF/m3, is representative of U.S. waters, then the burden is calculated to be 137
gI-TEQDF.
11.2.2.2. Mechanisms Responsible for Supply to and Releases from Water
As discussed in Section 11.2.1.2, dioxin-like compounds enter surface water from
atmospheric deposition, stormwater runoff erosion, and discharges of anthropogenic wastes.
Volatilization is the primary mechanism for release of dioxin-like compounds from the water
column to the atmosphere. Several studies have addressed the water-air exchange of dioxin-like
PCBs through volatilization in the Great Lakes (Achman et al., 1993; Hornbuckle et al., 1993;
Swackhamer and Armstrong, 1986; Baker and Eisenreich, 1990). No similar body of literature
has been developed to address volatilization of CDDs/CDFs from water.
Most studies that have addressed PCB water-air exchange have used the two-film model
developed by Whitman (1927) and made popular by Liss and Slater (1974). When assessing gas
exchange between air and water, the interface between the two phases can be considered as a
two-layer (film) system consisting of well-mixed gas and liquid films adjacent to the interface;
the rate of transfer is controlled by molecular diffusion through the stagnant boundary layer
(Achman et al., 1993). Liss and Slater (1974) applied the model to assess the flux of various
gases, specifically in the air-sea systems, and indicated the possibility of its use at any air-water
interface in the environment if the necessary data are available. Hornbuckle et al. (1993)
concluded that the two-film model is the best available tool for estimating regional and local flux
of PCBs from natural waters. The following paragraph, from Achman et al. (1993), succinctly
summarizes the model.
11-17
-------
The basic equation used to describe the rate of transfer across the interface is
F = Kol(Cw-C*) (11-1)
where F is the flux (mol/m2-day), Cw (mol/m3) is the dissolved PCB concentration in the bulk
water, and C* is the air concentration expressed as a water concentration in equilibrium with the
air (P/H, mol/m3). The variable P is the vapor-phase air concentration measured (mol/m3) and
converted to units of pressure using the ideal gas law; H is Henry's Law constant (atm-m3/mol).
The overall mass-transfer coefficient, Kol, has units of velocity (m/day). The concentration
gradient determines the direction of flux and drives the mass transfer, whereas Kol is a kinetic
parameter that quantifies the rate of transfer. The value of Kol is dependent on the physical and
chemical properties of the compound as well as environmental conditions. The reciprocal of Kol
is the total resistance to transfer expressed on a gas (RT/Hka)- and liquid (l/kw)-phase basis:
l/Kol = l/kw + RT/Hka (11-2)
where kw is the water-side mass transfer coefficient (m/day) and ka is the air-side mass transfer
coefficient (m/day). R is the universal gas constant (8.2057 x 10"5 atm-m3/mol K), T is the
absolute temperature, and H is Henry's Law constant.
Achman et al. (1993) and Hornbuckle et al. (1993) calculated the volatilization rates of
PCBs from Green Bay on Lake Michigan on the basis of air and water samples simultaneously
collected over a 14-day period above and below the air-water interphase and analyzed for 85
PCB congeners. Air samples collected over nearby land were also analyzed for the 85 PCB
congeners. The direction and magnitude of flux for each congener were then calculated using
Henry's Law and meteorological and hydrological parameters in the "two-film" model (see eq
11-1).
The net total PCB transfer rate (i.e., the sum of all congener transfer rates) was found to
be from water to air (i.e., volatilization). However, during cool water temperature periods
(October), the direction of transfer reversed for many congeners. Calculated transfer rates to air
ranged from 15 to 300 ng/m2/day at low wind speeds (1 to 3 m/sec) to 50 to 1,300 ng/m2/day at
higher wind speeds (4 to 6 m/sec). On a congener basis, the lower-chlorinated congeners
dominated total fluxes. The summary of flux calculations is presented in Table 11-5. The most
important factors influencing the magnitude of volatilization were the water concentration of
PCBs, wind speed, and water temperature. In addition, Achman et al. (1993) and Hornbuckle et
al. (1993) found that atmospheric PCB concentrations were higher over contaminated water than
over nearby land, atmospheric PCBs over water tended to increase with increasing dissolved
11-18
-------
Table 11-5. Summary of flux calculations for total PCBs in Green Bay
Date
6-4
6-5
6-6
6-7
6-10
6-11
7-28
7-29
7-30
7-31
8-1
10-21
10-22
10-23
Site
18
18
10
10
4
10
18
21
14
10
4
14
10
4
Flux3
(ng/m2-day)
40
40
95
155
325
13
330
70
225
90
800
555
1,300
30
aNumbers indicate water-to-air transfer of total PCBs. They represent the sum of individual PCB congener fluxes
and are described as "daily" fluxes because they correspond to air samples collected over 5-10 hr and water samples
collected over ~1 hr.
Source: Achmanetal. (1993).
PCB concentrations, and the congener distribution in the atmosphere correlated linearly with the
congener distributions in the adjacent water.
Achman et al. (1993) also summarized the PCB volatilization rates reported by other
researchers (Baker and Eisenreich, 1990; Swackhamer and Armstrong, 1986; Strachan and
Eisenreich, 1988; Swackhamer et al., 1988) for Great Lakes water bodies. The results of these
other studies, presented below, also show net flux of PCBs from water to air.
Water body
Lake Superior
Lake Michigan
Lake Superior
Si ski wit Lake
Lake Michigan
Total PCB
volatilization rate
(ng/m2/day)
141
240
63
23
15
Reference
Baker and Eisenreich (1990)
Strachan and Eisenreich (1988)
Strachan and Eisenreich (1988)
Swackhamer et al. (1988)
Swackhamer and Armstrong (1986)
11-19
-------
11.2.3. Sediment
11.2.3.1. Potential Mass of Dioxin-Like Compounds Present
EPA conducted congener-specific measurements of CDDs/CDFs in the sediments from
11 U.S. lakes located in areas relatively unimpacted by nearby industrial activity. The mean
TEQ concentration in the uppermost sediment layers from these 11 lakes was 5.3 ng TEQDF-
WHO98/kg (5.3 ng I-TEQDF/kg) dry weight. For most of the lakes, the uppermost layer
represents about 10 years worth of sedimentation. CDD/CDF concentrations in lakes impacted
by industrial activity may have higher concentrations. For example, Duarte-Davidson et al.
(1997) reported a TEQ concentration of 54 ng I-TEQDF/kg for urban sediments in the U.K.
As noted above, the surface area of inland waters in the United States is approximately
359,000 km2 (U.S. DOC, 1995b). In their calculations of sediment burdens in the U.K., Duarte-
Davidson et al. (1997) assumed that (1) the sediment surface area equals the water surface area,
(2) the majority of CDDs/CDFs were located in the top 5 cm of sediment, and (3) sediment
density was 0.13 g dry weight/cm3. Applying these assumptions to the water surface area and
background TEQ concentration for U.S. sediments yields a burden of at least 120 kg TEQDF-
WH098 (120 kg I-TEQDF).
11.2.3.2. Mechanisms Responsible for Supply to and Releases from Sediment
Because sediment is closely connected to the water column above it, evaluating the
potential for sediment to act as a reservoir of dioxin-like compounds is complex and likely to be
more difficult than studying dioxin-like compounds in a single medium, such as water or soil.
Volatilization and sedimentation are two mechanisms whereby persistent chemicals such as
CDDs/CDFs and PCBs are lost from water bodies/columns. Numerous authors (Swackhamer
and Armstrong, 1986; Muir et al., 1985; Ling et al., 1993) have noted that sediments are a likely
sink for persistent hydrophobic organic compounds because these compounds are likely to be
strongly bound to organic particles in the sediment.
For example, Muir et al. (1985) radiolabeled 2,3,7,8-TCDD and studied its dissipation
from sediments (collected from a farm pond and a lake) to the water column in laboratory
studies under static aerobic conditions at 10EC. After 675 days, more than 80% of the labeled
TCDD was still present in the pond sediment and 87% was still present in the lake sediment.
Aeration had little effect on the dissipation rates.
The concept of fugacity is a useful way to estimate the behavior of dioxin-like
compounds in sediments. Fugacity is expressed in units of pressure (pascals, or Pa) and is the
partial pressure exerted by the chemical in each medium. Fugacity models estimate equilibrium
concentrations in specific media at given chemical concentrations in the environment. Clark et
al. (1988) suggested evaluating contaminant concentrations in multiple environmental media by
comparing fugacity of adjoining media (e.g., comparing sediment fugacity with water column
11-20
-------
fugacity to determine a chemical's tendency to move from one to the other). The authors
evaluated fugacities of certain organochlorine compounds, including PCBs, in air, water,
sediment, fish, and fish-eating birds and their eggs. The authors presented PCBs fugacities
developed from data collected in a study of the Lake Ontario region. The fugacities of PCBs in
various media can be ranked as birds>fish>water>bottom sediment, indicating that PCBs and
other similar chemicals are likely to remain in bottom sediment and are less likely to re-enter the
water column.
11.2.3.3. Releases from Sediment to Water
Given the lack of data, no quantitative estimates of annual releases can be made. Ling et
al. (1993) evaluated the fate of various chemicals, including PCBs, in Hamilton Harbour, located
in Ontario, Canada, using a modified version of the Quantitative Water Air Sediment Interaction
(QWASI) fugacity model. Among the processes evaluated were diffusion between air and water
and sediment and water; sediment deposition, resuspension, and burial; and sediment
transformation. Three primary compartments were studied: air, water, and bottom sediments.
The sediment was treated as a simple, well-mixed surface layer of active sediment and the buried
sediment underneath. Chemicals in the active sediment were assumed to be able to exchange
with the overlying water; chemicals in the buried sediment were assumed to be isolated from the
sediment-water exchange. Sediment was assumed to be homogenous rather than heterogenous.
The epi- and hypolimnetic compartments of the water column were defined on the basis of a
thermocline, and the atmosphere was defined as a semi-infinite medium of constant, defined
composition.
Ling et al. estimated rates of PCB movement on the basis of 1987 loadings using two
models: one with a thermocline and one without a thermocline. The results for the water-
sediment transfer using the model with a thermocline were -32 kg/yr entering the hypolimnion
from the epilimnion, -27 kg/yr entering the surface sediment from the hypolimnion, and -18
kg/yr (>50%) going to burial. For sediment-to-water transfer, -7 kg/yr transferred to the
hypolimnion and then 12.5 kg/yr transferred to the epilimnion. Similar numbers were found in
the single water column model (the model without a thermocline).
Both the model with a thermocline and the model without a thermocline predicted
volatilization from the water to the atmosphere—1.6 kg/yr and 1.8 kg/yr, respectively.
However, the actual contribution of PCBs from sediment to air was not determined. A
comparison of estimated concentrations with observed values are presented in Table 11-6. For
PCBs, 68% were buried in the sediment, 20% were exported to Lake Ontario, 5.4% degraded in
the water and sediment, and 6% volatilized. The authors noted that these percentages were
uncertain. At the sediment-water exchange, more than 90% of each chemical was contained in
the sediment because of particle deposition and the high affinity of the chemical for sediment.
11-21
-------
Table 11-6. Comparison of model-estimated PCB concentrations with
observed values
Variable
PCBs
Observed concentration
Sediment
Water
0.23-1.04 jig/g
<20 |ig/m3
Estimated concentration from model without thermocline
Sediment
Water
Amount in sediment
Amount in water
Total mass
0.518ng/g
8.33 |ig/m3
74.9 kg
2.33 kg
77.2 kg
Estimated concentration from model with thermocline
Sediment
Hypolimnion
Epilimnion
Amount in sediment
Amount in hypolimnion
Amount in epilimnion
Total mass
0.527
8.48 |ig/m3
7.93 |ig/m3
76.3 kg
1.28kg
1.02kg
78.6kg
Source: Ling etal. (1993).
There was no indication that contaminants buried in the bottom sediments are transferred
through diffusion mechanisms back to the surface sediments; however, episodic release of these
chemicals from surface sediments can occur through mechanisms such as resuspension during
flooding or lake inversions and uptake/ingestion by benthic biota.
11.2.4. Biota
11.2.4.1. Potential Mass of Dioxin-Like Compounds Present
The mass of CDDs/CDFs in biota in the United States was not estimated as part of this
report. However, to place perspective on the potential magnitude of this reservoir, 82 g I-TEQDF
have been estimated to be present in biota in the U.K. (50 g in humans and 32 g in vegetation),
which is about three orders of magnitude less than the mass estimated to be present in U.K.
surface soils (Duarte-Davidson et al., 1997; Eduljee and Dyke, 1996). No data are available to
estimate the biota burden in the United States.
11.2.4.2. Mechanisms Responsible for Supply to and Releases from Biota
Apparently, very little of the dioxin-like compounds contained in contaminated soil is
ultimately taken up by the vegetation growing in the soil. Kjeller et al. (1991) analyzed
11-22
-------
concentrations of CDDs/CDFs in archived soil and grass samples collected from the mid-1840s
to 1989 at an English experimental station and found that only 0.006 to 0.02% of the soil burden
of CDDs/CDFs was taken up by the grass. In addition, scientists generally agree that, once taken
up by plant tissue, CDDs/CDFs are not translocated to other parts of the plant (e.g., fruits or
shoots) (Bacci and Gaggi, 1985; Hulster and Marschner, 1993, 1994; Nakamura et al., 1994).
Researchers have found that the concentration of dioxin-like compounds in a plant should
reach equilibrium with the vapor phase concentrations of dioxin-like compounds in the
surrounding air (Bacci et al., 1990a, b; Frank and Frank, 1989; Horstmann and McLachlan,
1992; McCrady and Maggard, 1993; McLachlan et al., 1995; Paterson et al., 1991; Simonich and
Kites, 1994; Tolls and McLachlan, 1994; Welsch-Pausch et al., 1995). Horstmann and
McLachlan (1992) stated that the leaf-air transfer of volatile compounds is a reversible process
governed by concentration gradients. If CDD/CDF concentrations are higher in the surrounding
air than they are in the air spaces within plant tissue, CDDs/CDFs should diffuse into the plant.
Once equilibrium is reached and CDD/CDF concentrations in the plant equal that of surrounding
air, no more CDDs/CDFs should be taken into the plant. When CDD/CDF concentrations in
surrounding air begin to decrease, CDDs/CDFs should diffuse (probably at a slow rate) out of
the plant tissue. Apparently, CDDs/CDFs are not bioconcentrated to a significant extent in the
lipid portion of the leaf cuticle (Gaggi et al., 1985). The CDDs/CDFs present in the leaf tissue
are predominantly released from the plant through leaf fall onto soil. Therefore, vegetation is
not likely to be a long-term reservoir of dioxin-like compounds.
Research suggests that dioxin-like compounds in animal tissue, unlike in vegetation,
seldom, if ever, reach equilibrium with vapor phase concentrations in the surrounding
atmosphere (or water column concentrations in the case of aquatic life). Rather, animals
exposed to dioxin-like compounds are known to bioaccumulate these compounds, primarily in
body fat (U.S. EPA, 1993g, h). Nonetheless, animals, unlike plants, can metabolize certain
chlorinated hydrocarbons after they enter the body (Carlberg et al., 1983). Dioxin-like
compounds can be released from an animal's body (at congener-specific rates) through
metabolic processes or through weight loss, breast-feeding, or sweating. McLachlan (1996)
reported the half-life for the clearance of 2,3,7,8-TCDD from humans to be 7 yr. As a result,
animal life has a greater potential than does vegetation for being a long-term reservoir source of
CDDs/CDFs. The majority of the dioxin-like compounds released by animals in the form of
waste materials will be released to water or soil. Similarly, upon death, the dioxin-like
compounds remaining in the body will be deposited onto soil or aquatic sediments or will be
ingested by other animals.
11-23
-------
11.2.4.3. Approaches for Measuring and Estimating Releases from Biota
Researchers have investigated the uptake and release of CDDs/CDFs by vegetation
through measurement of actual concentrations during uptake and release by vegetation grown in
closed systems (greenhouses). Bacci et al. (1992) conducted uptake and release studies of
1,2,3,4-TCDD by plant foliage in a closed system (a specially constructed greenhouse).
Concentrations of TCDD vapor in the greenhouse air were maintained during the 370-hr uptake
phase at a mean concentration of 0.0062 ng/L (air concentration varied slightly from 0.005 to
0.0075 ng/L). To begin the release phase, the TCDD vapor source (amended sand) as well as the
greenhouse walls were removed, and release of CDDs/CDFs from the leaves was measured for
500 hr. The authors concluded that, during uptake, TCDD concentration in the leaves varied as a
function of time and was dependent on the concentration of vapor-phase TCDD in the
surrounding air. They estimated the release of TCDD from the vegetation to be relatively slow,
with a half-life of TCDD of 3,300 hr.
McCrady and Maggard (1993) conducted a mass balance study of uptake and release of
dioxin in grass foliage. The results indicated a half-life of dioxin in grass of 128 hr. These
researchers also noted that, in addition to volatilization, photodegradation of dioxins on the
foliage appeared to be a significant removal mechanism. They calculated the photodegradation
half-life to be 44 hr.
Interpretation of uptake and release data over variable exposure times and contaminant
concentrations has led to the development of models describing air-to-vegetation equilibrium
and kinetics controlling the behavior of dioxin in vegetation. Some earlier fugacity modeling
attempts described the leaf of a plant as behaving as a single compartment. One-compartment
models were described by Bacci et al. (1990a, b), Trapp et al. (1990), Schramm et al. (1987), and
Tolls and McLachlan (1994). Researchers presenting most of the recently developed models
claim that the available data better support the concept of a leaf behaving as two compartments
(Riederer, 1990; Paterson et al., 1991; Horstmann and McLachlan, 1992; McCrady and
Maggard, 1993; Tolls and McLachlan, 1994; McLachlan et al., 1995). Input parameters
considered by most models include critical chemical characteristics of the contaminant,
characteristics of the plant, exposure times, and contaminant concentrations measured within the
plant. Riederer (1990) suggested treating a leaf as multiple compartments having different
accessibilities to the atmosphere and different diffusion resistances.
Input parameters for the two-compartment model are octanol-water coefficients, cuticle-
water partition coefficients, aqueous solubility, and saturation vapor pressure of the chemical of
concern. Outputs of the model are prediction of equilibrium concentration in different leaf
tissues, estimates of air-to-vegetation bioconcentration equilibria, and identification of leaf
compartments in which compounds are likely to accumulate. Riederer (1990) also presented an
11-24
-------
approach for using the model to semi quantitatively assess the potential for revolatilization of
dioxins from vegetation.
One advantage of the model presented by Riederer (1990) is that it considers critical
plant characteristics in the release of dioxins. A plant is an active organism, responding to
changes in its environment and acting accordingly to ensure its survival. Certain plant
characteristics, such as the action of stomata (specialized cells usually on the lower leaf surface
that open and close to control passage of vapors into and out of the leaf interior) and total leaf
volume, are important factors that affect the release rates of vapor phase contaminants from
vegetation.
Paterson et al. (1991) also presented a two-compartment model for release of dioxin-like
compounds from vegetation. This model describes a plant as being made up of compartments in
terms of volume fractions of air, water, and nonpolar (lipid-soluble, or octanol-equivalent)
organic matter. Paterson et al. attempted to show that leaf-air equilibrium and kinetics can be
correlated with chemical properties of the contaminant and properties of the leaf. The authors
suggested that the clearance rate constant (&2) can be correlated with the bioconcentration factor.
This model does not consider critical plant characteristics, such as action of the stomata, and for
this reason it may be less reliable than models that do consider plant characteristics, such as the
model presented by Riederer (1990).
Horstmann and McLachlan (1992) developed a fugacity model to describe release of
semivolatile organic compounds from the surface of a solid (spruce needles). Their approach
was slightly different in that their goal was instrument/method development, but their data
supported the behavior of a leaf as a two-compartment system.
McCrady and Maggard (1993) also collected data supporting the importance of viewing a
leaf as a two-compartment system. They used a two-compartment model similar to the one
described by Paterson et al. (1991) that also does not consider critical plant characteristics and
thus may be less reliable than models that do (e.g., Reiderer, 1990).
Tolls and McLachlan (1994) exposed grass cultures for up to 240 hr to several
semivolatile organic compounds and then measured the release of contaminants from the grass.
They developed a two-compartment partitioning model based on the data they collected. The
model consists of a small surface compartment (the leaf cuticle) and a large interior reservoir (air
spaces within the leaf). Their model assumes that the flux of a chemical is the product of the
fugacity difference (surface fugacity minus reservoir fugacity) and the conductance between the
leaf compartments.
In an attempt to validate this model, McLachlan et al. (1995) compared concentrations of
semivolatile organic compounds measured in grass grown under field conditions with
concentrations predicted by their previous laboratory work with a fugacity meter. The
11-25
-------
concentrations measured in the grass cultures agreed with results predicted by the mathematical
model described by Tolls and McLachlan (1994).
11.3. SUMMARY AND CONCLUSIONS
As depicted in Figure 11-1 a set of complex relationships exists among reservoirs and
between reservoirs and contemporary formation sources. The significance of reservoirs for
human exposure is more dependent on their ability to affect the concentration of dioxin-like
compounds in other media than on their size or net release rate. This section first summarizes
and draws conclusions from the limited information available regarding the character and
magnitude of reservoir sources. This information is then used to discuss the implications of
reservoir sources to human exposure.
11.3.1. Reservoir Sources
The following are summary statements about soil reservoir sources.
• Soil is likely to be the reservoir source with the greatest potential for release of
CDDs/CDFs to other environmental media, particularly to water. This is due in part
to its relatively large mass of stored CDDs/CDFs, but more importantly, it is due to
the existence of demonstrated transport mechanisms for intermedia exchange, e.g.,
soil erosion to surface waters and particle resuspension to air.
• The preliminary estimates of CDD/CDF runoff from urban areas to waterways is
comparable to known industrial point source releases, and runoff from agricultural
areas to surface waters is more than 100 times greater. It is unclear how much of the
soil erosion and runoff represents recently deposited CDDs/CDFs from primary
sources or longer-term accumulation. Much of the eroded soil comes from tilled
agricultural lands, which would include a mix of CDDs/CDFs from various
deposition times. The age of CDDs/CDFs in urban runoff is less clear.
• Based on the limited information currently available (i.e., primarily fugacity
modeling), volatilization of CDDs/CDFs from soils is not believed to significantly
alter ambient air concentrations. However, volatilization of PCBs from soil may be a
significant process.
• Based on the limited information currently available, resuspension of soil may
account for a small fraction (-4%) of CDD/CDF concentrations in air. This
resuspended soil may, however, constitute a more significant portion of dry
deposition.
The following are summary statements about water reservoir sources.
11-26
-------
• It is unclear whether volatilization of CDDs/CDFs from water can significantly alter
air concentrations. For PCBs, however, the water-air exchange appears to be
significant, and for some water bodies results in a net transfer from water to air.
• Water is the major media contributing CDDs/CDFs and PCBs to sediment. Note that
most of the CDDs/CDFs in sediments originally came from soils. For specific water
bodies, however, the CDDs/CDFs and PCBs in sediments may have been dominated
by local industrial discharges to water.
The following are summary statements about sediment reservoir sources.
• It is important to distinguish between surface and deep sediments. Surface sediments
are commonly resuspended and introduced back into the water; deep sediments
generally do not interact with the water column. Surface sediments can contribute
significantly to the CDD/CDF and PCB concentrations in water, whereas deep
sediments do not.
• There is little, if any, movement of dioxin-like compounds once they are buried in the
bottom sediments. Bottom sediments may be considered as sinks.
The following are summary statements about biota reservoir sources.
• The mass of CDDs/CDFs in vegetation at any given time is likely to be small when
compared with the mass in soil. Vegetation does play an important role in
transferring CDDs/CDFs from the air to the soil via the decay of plant biomass.
• Release by volatilization from vegetation has been studied and modeled using the
fugacity approach, and half-lives have been estimated. Based on these results,
volatilization is not believed to be a significant mechanism for release of
CDDs/CDFs and PCBs except possibly during forest/brush fires.
• The mass of CDDs/CDFs in animals at any given time is likely to be small when
compared with the mass in soil. Similarly, releases are small and occur primarily by
excretion and decomposition of dead biomass.
11.3.2. Implications for Human Exposure
Although, the ability to make quantitative estimates of releases from reservoir sources is
limited at present, it is reasonable to conclude that the contribution of reservoir sources to human
exposure may be significant. Diet accounts for more than 95% of human exposure. Although
the size of the biota reservoir is small when compared with the soil and sediment reservoirs, it is
clearly the key contributor to human exposure. The potential contribution of the other reservoirs
to human exposure is discussed below.
11-27
-------
PCB reservoir releases. Because current sources of newly formed PCBs are most likely
negligible, human exposure to the dioxin-like PCBs is thought to be derived almost completely
from current releases of old PCBs stored in reservoir sources. Key pathways involve releases
from both soils and sediments to both aquatic and terrestrial food chains. One-third of general
population TEQDFP exposure is due to PCBs. Thus, at least one-third of the overall risk to the
general population from dioxin-like compounds comes from reservoir sources.
CDD/CDF releases from soil and sediments to water and exposure via the aquatic
pathway. The earlier discussion has shown that soils can have significant inputs to waterways
via soil erosion and runoff. Similarly, the sediment reservoir contributes significantly to
CDD/CDF concentrations in water. These releases appear to be greater than those from the
primary sources included in the inventory. Dioxins in waterways bioaccumulate in fish, and fish
consumption causes human exposure. Fish consumption makes up about one-third of the total
general population CDD/CDF TEQ exposure. This suggests that a significant portion of the
CDD/CDF TEQ exposure could be due to releases from the soil and sediment reservoir.
CDD/CDF releases from soil to air and exposure via the terrestrial pathway.
Potentially, soil reservoirs could have vapor and particulate releases that deposit on plants and
enter the terrestrial food chain. The magnitude of this contribution, however, is unknown. EPA
plans future studies in agricultural areas that will compare modeled air concentrations from
primary sources with measured levels as a way to gain further insight into this issue.
11-28
-------
REFERENCES
Abad, E; Adrados, A; Caixach, J; et al. (2002) Dioxin abatement strategies and mass balance at a municipal waste
management plant. Environ Sci Technol 36:92-99.
Abramowicz, D. (1990) Aerobic and anaerobic biodegradation of PCBs: a review. Crit Rev Biotech 10(3):241-251.
Achman, DA; Hornbuckle, KC; Eisenreich, SJ. (1993) Volatilization of polychlorinated biphenyls from Green Bay,
Lake Michigan. Environ Sci Technol 27(l):75-87.
Addink R; Altwicker, ER. (1999) Formation experiments of polychlorinated dibenzo-p-dioxins/dibenzofurans with
Na 37C1. Organohalogen Compounds 4:69-72.
Addink, R; Olie, K. (1995) Mechanisms of formation and destruction of polychlorinated dibenzo-p-dioxins and
dibenzofurans in heterogeneous systems. Environ Sci Technol 29:1425-1435.
Addink, R; Van Bavel, B; Visser, R; et al. (1990) Surface catalyzed formation of polychlorinated dibenzo-p-
dioxins/dibenzofurans during municipal waste incineration. Chemosphere 20(10-12): 1929-1934.
Addink, R; Drijver, DJ; Olie, K. (1991) Formation of polychlorinated dibenzo-p-dioxins/dibenzofurans in the
carbon/fly ash system. Chemosphere 23(8-l):1205-1211.
Addink, R; Bakker, WCM; Olie, K. (1995) Influence of HC1 and C12 on the formation of polychlorinated dibenzo-p-
dioxins/dibenzofurans in a carbon/fly ash mixture. Environ Sci Technol 29:2055-2058.
Adriaens, P; Grbic-Galic, D. (1992) Effect of cocontaminants and concentration on the anaerobic biotransformation
of PCDD/F in methanogenic river sediments. Organohalogen Compounds 8:209-212.
Adriaens, P; Grbic-Galic, D. (1993) Reductive dechlorination of PCDD/F by anaerobic cultures and sediments.
Organohalogen Compounds 12:107-110.
Adriaens, P; Fu, Q; Grbic-Galic, D. (1995) Bioavailability and transformation of highly chlorinated dibenzo-p-
dioxins and dibenzofurans in anaerobic soils and sediments. Environ Sci Technol 29(9):2252-2260.
Advanced Technology Systems, Inc. (1995) Report on measurement of hazardous air pollutants from aluminum melt
furnace operations at Ravenswood Aluminum Corporation. Monroeville, PA.
AGES (Applied Geotechnical and Environmental Services Corp). (1992) Source sampling report: comprehensive
emissions testing; blast furnace. May/June 1992, Franklin Smelting and Refining Corp, Valley Forge, PA. Report
No. 42614.01-01.
AHA (American Hospital Association). (1995) Comments by the American Hospital Association in review of EPA's
external review draft Estimating Exposures to Dioxin-Like Compounds (EPA/600/688/005Ca-c). Public Comment
Number EXP-48-A1, received January 17, 1995.
Ahlborg, VG; Becking, GC; Birnbaum, LS; et al. (1994) Toxic equivalency factors for dioxin-like PCBs.
Chemosphere 28(6): 1049-1067.
Ahling, B; Lindskog, A. (1982) Emission of chlorinated organic substances from combustion. In: Hutzinger, O;
Safe, S, eds. Chlorinated dioxins and related compounds, impact on the environment. New York, NY: Pergamon
Press.
Ahling, B; Bjorseth, A; Lunde, G. (1978) Formation of chlorinated hydrocarbons during combustion of polyvinyl
chloride. Chemosphere 7:799-806.
R-l
-------
AISI (American Iron and Steel Institute). (1990) Annual statistical report, 1989. Washington, DC.
Aittola, JP. (1993) Environmental emissions from landfill fires, preliminary investigation. Viatek Topiola Og,
Espoo, Finland.
Aittola, J; Paasivirta, J; Vattulainen, A. (1992) Measurements of organochloro compounds at a metal reclamation
plant. Organohalogen Compounds 9:9-12.
Albro, PW; Parker, CE. (1979) Comparison of the compositions of Aroclor 1242 and Aroclor 1016. J Chromatog
169:161-166.
Albro, PW; Corbett, JT; Schroeder, JL. (1981) Quantitative characterization of polychlorinated biphenyl mixtures
(Aroclors 1248, 1254, and 1260) by gas chromatography using capillary columns. J Chromatog 205:103-111.
Altwicker, ER; Milligan, MS. (1993) Formation of dioxins: competing rates between chemically similar precursors
and de novo reactions. Chemosphere 27(1-3):301-307.
Altwicker, ER; Konduri, R; Lin C; et al. (1993) Formation of precursors to chlorinated dioxin/furans under
heterogeneous conditions. Combust Sci Tech 88:349-368.
Alvarado, MJ; Armstrong, BS; Crouch, E. (2001) The AMSA 2000/2001 survey of dioxin-like compounds in
biosolids: statistical analysis. Cambridge, MA: Cambridge Environmental Inc.
Amendola, GA; Barna, DR. (1989) U.S. EPA petroleum refinery dioxin screening survey: BP Oil-Lima, Ohio
refinery. U.S. Environmental Protection Agency, Region 5, Environmental Sciences Division, Westlake, OH.
American Forest and Paper Association. (1997) Paper, paperboard, and wood pulp statistics: data through 1996.
Washington, DC.
American Paper Institute. (1992) Statistics of paper, paperboard, and wood pulp: data through 1991. New York,
NY.
American Wood Preservers Institute. (1977) Pentachlorophenol: a wood preservative. McLean, VA.
American Wood Preservers Institute. (1995) 1994 Wood preserving industry production statistical report. McLean,
VA.
ASME (American Society of Mechanical Engineer). (1995) The relationship between chlorine in waste streams and
dioxin emissions from waste combustor stacks. Vol 36. New York, NY: Report number CRTD.
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.
Atkinson, R. (1991) Atmospheric lifetimes of dibenzo-p-dioxins and dibenzofurans. Sci Total Environ 104:17-33.
ATSDR (Agency for Toxic Substances and Disease Registry). (1993) Toxicological profile for selected PCBs
(Aroclor-1260, -1254, -1248, -1242, -1232, -1221, and -1016). Atlanta, GA. Available from the National Technical
Information Service, Springfield, VA; PB93-182517.
Ayris, S; Harrad, S. (1997) Factors affecting the volatilization of PCBs from soil. Organohalogen Compounds
33:250-254.
BAAQMD (Bay Area Air Quality Management District). (1996) Air emissions of dioxins in the Bay area. San
Francisco, CA.
R-2
-------
Bacci, E; Gaggi, C. (1985) Polychlorinated biphenyls in plant foliage: translocation or volatilization from
contaminated soils? Bull Environ Contam Toxicol 35:673-681.
Bacci, E; Calamari, D; Gaggi, C; et al. (1990a) Bioconcentration of organic chemical vapors in plant leaves:
experiments, measurements and correlation. Environ Sci Technol 24(6):885-889.
Bacci, E; Calamari, D; Gaggi, C; et al. (1990b) Bioconcentration of organic chemical vapors in plant leaves: the
azalea model. Chemosphere 21(4-5):525-535.
Bacci, E; Cerejeira, MJ; Gaggi, C; et al. (1992) Chlorinated dioxins: volatilization from soils and bioconcentration in
plant leaves. Bull Environ Contam Toxicol 48:401-408.
Bacher, R; Swerev, M; Ballschmiter, K. (1992) Profile and pattern of monochloro-through octachlorodibenzodioxins
and dibenzofurans in chimney deposits from wood burning. Environ Sci Technol 26(8): 1649-1655.
Baker, JE; Eisenreich, SJ. (1990) Concentrations and fluxes of polycyclic aromatic hydrocarbons and
polychlorinated biphenyls across the air-water interface of Lake Superior. Environ Sci Technol 24(3):342-352.
Ball, M; Papke, O; Lis, A. (1990) Polychloridibenzodioxine und polychlordibenzofurane in cigaretterauch. Beitrage
zur Tabakforschung International 14(6):393-402.
Ballschmiter, K; Buchert, H; Niemczyk, R; et al. (1986) Automobile exhausts versus municipal waste incineration as
sources of the polychloro-dibenzodioxins (PCDD) and -furans (PCDF) found in the environment. Chemosphere
15(7):901-915.
Barkovskii, AL; Adriaens, P. (1995) Reductive dechlorination of tetrachloro-dibenzo-p-dioxin partitioned from
Passaic River sediments in an autochthonous microbial community. Organohalogen Compounds 24:17-21.
Barkovskii, AL; Adriaens, P. (1996) Microbial dechlorination of historically present and freshly spiked chlorinated
dioxins and diversity of dioxin-dechlorinating populations. App Environ Microbiol 62(12):4556-4562.
Beard, A; Krishnat, PN; Karasek, FW. (1993) Formation of polychlorinated dibenzofurans by chlorination and de
novo reactions with FeCl3 in petroleum refining processes. Environ Sci Technol 27(8): 1505-1511.
Bedard, DL; May, PJ. (1996) Characterization of the polychlorinated biphenyls in the sediments of Woods Pond:
evidence for microbial dechlorination of Aroclor 1260 in situ. Environ Sci Technol 30(l):237-245.
Bedard, DL; Unterman, R; Bopp, LH; et al. (1986) Rapid assay for screening and characterizing microorganisms for
the ability to degrade polychlorinated biphenyls. Appl Environ Microbiol 51(4):761-768.
Bedard, DL; Smullen, LA; DeWeerd, KA; et al. (1995) Chemical activation of microbially-mediated PCB
dechlorination: a field study. Organohalogen Compounds 24:23-28.
Behrooz, GS; Altwicker, ER. (1995) Formation of polychlorinated dioxins, furans, benzenes and phenols in the post-
combustion region of a heterogeneous combustor: effect of bed material and post-combustion temperature. Environ
Sci Technol 29:1156-1162.
Behrooz, GS; Altwicker, ER. (1996) Rapid formation of polychlorinated dioxins/furans during the heterogeneous
combustion of 1,2-dichlorobenzene and 2,4-dichlorophenol. Chemosphere 32(1): 133-144.
Bell, A. (1999) AB2588 emissions testing at California Portland Cement Company's Colton Plant; coal firing and
coal with tires firing. Prepared by Delta Air Quality Services, Inc., California Portland Cement Company, Glendora,
CA.
R-3
-------
Bergstrom, J; Bjorner, B. (1992) Dioxiner ock brander vid ovfallsuplag. Mamo, Miljokonsulterna i Studsvik AB.
Stiftelsen reforsk. FoU nr 68. 94p. [Dioxins and fires at landfills. The Reforsk Foundation, Studsvik, Sweden.
Report No. 68, p. 94].
Bingham, AG; Edmunds, CJ; Graham, BW; et al. (1989) Determination of PCDDs and PCDFs in car exhaust.
Chemosphere 19(l-6):669-673.
Birmingham, B. (1990) Analysis of PCDD and PCDF patterns in soil samples: use in the estimation of the risk of
exposure. Chemosphere 20(7-9):807-814.
Blumenstock, M; Zimmermann, R; Schraumm, L-W; et al. (1998) Presence of polychlorinated dibenzo-p-dioxins
(PCDD), dibenzofurans (PCDFs), biphenyls (PCB), chlorinated benzenes (PCBz) and polycyclic aromatic
hydrocarbons (PAH) under various combustion conditions in a post combustion chamber. Organohalogen
Compounds 36:59-63.
Boers, JP; de Leer, EWB; Gramberg, L; et al. (1993) Levels of coplanar PCBs in flue gases of high temperature
processes and their occurrence in environmental samples. Organohalogen Compounds 11:233-236.
Born, JGP; Mulder, P; Louw, R. (1993) Fly ash mediated reactions of phenol and monochlorophenols:
oxychlorination, deep oxidation, and condensation. Environ Sci Technol 27:1849-1863.
Bowes, GW; Mulvihill, MJ; Simoneit, BRT; et al. (1975a) Identification of chlorinated dibenzofurans in American
polychlorinated biphenyls. Nature 256:305-307.
Bowes, GW; Mulvihill, MJ; DeCamp, MR; et al. (1975b) Gas chromatographic characteristics of authentic
chlorinated dibenzofurans; identification of two isomers in American and Japanese polychlorinated biphenyls. J
Agric Food Chem 23(6): 1222-1223.
Bramley, MJ. ( 1998) Dioxin and hexachlorobenzene releases from magnesium production in North America:
lessons from Noranda's project in Asbestos, Quebec. Greenpeace Canada, Toronto.
Bremmer, HJ; Troost, LM; Kuipers, G; et al. (1994) Emissions of dioxins in The Netherlands. National Institute of
Public Health and Environmental Protection (RIVM) and Netherlands Organization for Applied Scientific Research
(TNO). Report No. 770501018.
Broker, G; Geueke, KJ; Hiester, E; et al. (1992) Emission polychloierter dibenzo-p-dioxins und -furane aus
hausbrand-feuerungen. Essen, Germany: Landesanstalt Ftir Immissionsschutz Nordhein-Westfalen. ISSN 0720-
8499.
Brown, JF. (1994) Determination of PCB metabolic, excretion, and accumulation rates for use as indicators of
biological response and relative risk. Environ Sci Technol 28(13):2295-2305.
Brown, A. (2002) 2002 North Carolina flue-cured tobacco production guide. Environ Sci Technol 29:2090-2098.
Brown, JF; Bedard, DL; Brennan, MJ; et al. (1987) Polychlorinated biphenyl dechlorination in aquatic sediments.
Science 236:709-712.
Brown, JF; Carnahan, JC; Dorn, SB; et al. (1988) Levels of bioactive PCDF congeners in PCB dielectric fluids from
capacitors and transformers. Chemosphere 17(9): 1697-1702.
Brown, JF; Frame, GM; Olson, DR; et al. (1995) The sources of the coplanar PCBs. Organohalogen Compounds
26:427-430.
Bruce, KR; Beach, LO; Gullet, RK. (1991) The role of gas-phase C12 in the formation of PCDD/PCDF during waste
combustion. Waste Management 11:97-102.
R-4
-------
Brunner, CR. (1984) Incineration systems, selection and design. New York, NY: Van Nostrand Reinhold; pp. 56-99.
Brunner, CR. (1992) Sewage sludge. In: Buonicore, AJ; Davis, WT; eds. Air pollution engineering manual. Air
and Waste Management Association. New York, NY: Van Nostrand Reinhold; pp. 296-311.
Bryant, JG. (1993) Chlorinated benzenes. In: Kirk-Othmer encyclopedia of chemical technology. 4th ed. Vol.6.
pp. 87-100.
Buckland, SJ; Dye, EA; Leathern, SV; et al. (1994) The levels of PCDDs and PCDFs in soil samples collected from
conservation areas following brush fires. Organohalogen Compounds 20:85-89.
Buekens, A; Stieglitz, L; Huang, H; et al. (1997) Preliminary investigation of formation mechanism of chlorinated
dioxins and furans in industrial and metallurgical processes. Organohalogen Compounds 31:516-520.
Bumb, RR; Crummett, WB; Cutie, SS; et al. (1980) Trace chemistries of fire: a source of chlorinated dioxins.
Science 210(4468):385-390.
Buonicore, AJ. (1992a) Control of gaseous pollutants. In: Buonicore, AJ; Davis, WT; eds. Air pollution
engineering manual. New York, NY: Van Nostrand Reinhold. pp. 46-201.
Buonicore, AJ. (1992b) Medical waste incineration. In: Buonicore, AJ; Davis, WT; eds. Air pollution engineering
manual. New York, NY: Van Nostrand Reinhold; pp.27-49.
Buser, HR. (1979) Formation of polychlorinated dibenzofurans (PCDFs) and dibenzo-p-dioxins (PCDDs) from the
pyrolysis of chlorobenzenes. Chemosphere 8:415-424.
Buser, HR; Bosshardt, HP. (1976) Determination of polychlorinated dibenzo-p-dioxins and dibenzofurans in
commercial pentachlorophenols by combined gas chromatography-mass spectrometry. Journal of the AOAC
59(3):562-569.
Cains, PW; Dyke, P. (1994) Emissions of chlorinated dibenzodioxins and dibenzofurans from waste combustion
plants in the UK. Organohalogen Compounds 20:271-276.
Calcagni, J; Ciolek, M; Mulrine, P. (1998) Dioxin emissions from U.S. sinter plants. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC.
CANA (Cremation Association of North America). (2002) Confirmed 2000 statistics: 2000-2025. Available online
at http://www.cremationassociation.org/docsAVebOOConfirmed.pdf.
CANA (Cremation Association of North America). (2006) Historical statistics (provides data for 1876-2005, based
on the number of deaths, the number of crematories, and the number of cremations for both the United States and
Canada). Available online at http://www.cremationassociation.org/html/statistics.html.
Capes, CE. (1983) Size enlargement. In: Kirk-Othmer encyclopedia of chemical technology. 3rd ed. New York,
NY: John Wiley and Sons; pp. 21:77-105.
CARB (California Air Resources Board). (1987) Determination of PCDD andPCDF emissions from motor vehicles
[Draft]. State of California Air Resource Board, Sacramento, CA. Test Report No. C-86-029.
CARB (California Air Resources Board). (1990a) Evaluation test on two propane fired crematories at Camellia
Memorial Lawn Cemetery. Engineering Evaluation Branch, Monitoring and Laboratory Division, Sacramento, CA.
Test Report No. C-90-004.
R-5
-------
CARD (California Air Resources Board). (1990b) Evaluation on a woodwaste fired incinerator at Koppers
Company, Oroville, California. Engineering Evaluation Branch, Monitoring and Laboratory Division, Sacramento,
CA. Test Report No. C-88-065.
CARB (California Air Resources Board). (1990c) Compliance testing for non-criteria pollutants at a landfill flare.
Engineering Evaluation Branch, Monitoring and Laboratory Division, Sacramento, CA. Confidential Report No.
ERC-2.
CARB (California Air Resources Board). (1990d) Evaluation test on a wood waste fired incinerator at
Pacific Oroville Power Inc. Engineering Evaluation Branch, Monitoring and Laboratory Division, Sacramento, CA.
Test Report No. C-88-050.
CARB (California Air Resources Board). (1990e) Evaluation test on twin fluidized bed wood waste fueled
combustors located in Cental California. Engineering Evaluation Branch, Monitoring and Laboratory Division,
Sacramento, CA. Test Report No. C-87-042.
CARB (California Air Resources Board). (1990f) Evaluation test on a wood waste fired incinerator at Louisiana
Pacific Hardboard Plant, Oroville, CA. Engineering Evaluation Branch, Monitoring and Laboratory Division,
Sacramento, CA. Test Report No. C-88-066.
CARB (California Air Resources Board). (1991) Emissions test of the Modesto Energy Project tires-to-energy
facility. Engineering Evaluation Branch, Monitoring and Laboratory Division, Sacramento, CA. Project No. C-87-
072.
CARB (California Air Resources Board). (1992a) Dioxin/furan estimates from a secondary aluminum facility.
Engineering Evaluation Branch, Monitoring and Laboratory Division, Sacramento, CA. Confidential Report No.
ERC-9.
CARB (California Air Resources Board). (1992b) Emission measurements on delacquering system. Engineering
Evaluation Branch, Monitoring and Laboratory Division, Sacramento, CA. Confidential Report No. ERC-32.
CARB (California Air Resources Board). (1993) Emissions measurement of toxic compounds from a cupola
baghouse at a steel foundry. Engineering Evaluation Branch, Monitoring and Laboratory Division, Sacramento, CA.
Confidential Report No. ERC-61.
CARB (California Air Resources Board). (1999) Determination of emissions from the no. 3 reformer as Tosco
Refining Company San Francisco Area Refinery at Avon. Engineering and Laboratory Branch, Monitoring and
Laboratory Division, Sacramento, CA. Project No. C-96-0396.
CARB (California Air Resources Board). (2002) Staff report: initial statement of reasons for the proposed airborne
toxic control measure to reduce emissions of toxic air contaminants from outdoor residential waste burning. State of
California Air Resource Board, Sacramento, CA.
Carlberg, GE; Ofstad, EB; Drangsholt, H. (1983) Atmospheric deposition of organic micropollutants in Norway
studied by means of moss and lichen analysis. Chemosphere 12(3):341-356.
Carpenter, C; Schweer, G; Stinnett, G; et al. (1986) Exposure assessment for hexachlorobenzene. U.S.
Environmental Protection Agency, Office of Toxic Substances, Washington, DC; EPA 560/5-86-019.
Carroll, WF. (1996) Is PVC in house fires the great unknown source of dioxin? Fire and Materials 20:161-166.
CCC (Chlorine Chemistry Council). (2003a) CDD/CDF data validation project. CDD/CDF release estimates for
Dupont: Midland, MI; Plaquemine, LA, Texas Operations, Freeport, TX. Prepared by Amendola Engineering, Inc.,
Westlake, OH, and PD Consulting, Herefordshire, U.K.
R-6
-------
CCC (Chlorine Chemistry Council). (2003b) CDD/CDF data validation project. CDD/CDF release estimates for
PPG Industries, Inc.: Lake Charles, LA; Natrium, WV. Prepared by Amendola Engineering, Inc., Westlake, OH,
and PD Consulting, Herefordshire, U.K.
CCC (Chlorine Chemistry Council). (2003c) CDD/CDF data validation project. CDD/CDF release estimates for The
Georgia Gulf Corporation: Plaquemine, LA. Prepared by Amendola Engineering, Inc., Westlake, OH, and PD
Consulting, Herefordshire, U.K.
CCC (Chlorine Chemistry Council). (2003d) CDD/CDF data validation project. CDD/CDF release estimates for
Occidental Chemical Corporation: Convent, LA; Deer Park (VCM PLANT), TX; Ingleside, TX; Laporte, TX;
Mobile, AL. Prepared by Amendola Engineering, Inc., Westlake, OH, and PD Consulting, Herefordshire, U.K.
CCC (Chlorine Chemistry Council). (2004) CDD/CDF data validation project. CDD/CDF release estimates for
occidental chemical corporation: Convent, LA; Deer Park (VCM Plant), TX; Ingleside, TX; LaPorte, TX; Mobile,
AL.
CCC (Chlorine Chemistry Council). (2005) CDD/CDF data validation project. CDD/CDF release estimates for
Dupont: Delisle, MS; Edge Moor, DE; Johnsonville, TN. Prepared by Amendola Engineering, Inc., Westlake, OH,
and PD Consulting, Herefordshire, U.K.
C&EN (Chemical & Engineering News). (2002) Production: down but not out. June 24, 2002. Available online at
http://pubs.acs.org/cen.
CEQ (Council on Environmental Quality). (1997) Environmental quality 1990. 25th Annual Report. Council on
Environmental Quality, Executive Office of the President of the United States, Washington, DC.
ChemRisk. (1993) Critique of dioxin factories. Prepared for The Vinyl Institute.
Chen, YR; Chen, CJ; Lee, CC; et al. (1986) Reproductive effects of emissions from waste electrical products
combustion. Presented at Dioxin '86, 6th International Symposium on Chlorinated Dioxins and Related Compounds;
September, Fukuoka University, Fukuoka, Japan.
Chen, M; Hong, CS; Bush, B; et al. (1988) Anaerobic biodegradation of polychlorinated biphenyls by bacteria from
Hudson River sediments. Ecotoxicol and Environ Safety 16:95-105.
Chinkin, LR; Weir, BR; Latimer, DA. (1987) Inventory of chlorophenol use in the forest products industry and
investigation of related emissions of chlorinated dibenzodioxins and dibenzofurans. California Air Resources Board,
Sacramento, CA. Contract No. A5-125-32.
Choudhry, GG; Hutzinger, O. (1983) Mechanistic aspects of the thermal formation of halogenated organic
compounds including polychlorinated dibenzo-p-dioxins. New York, NY: Gordon and Breach.
Choudhry, GG; Webster, GRB. (1989) Environmental photochemistry of PCDDs. 2. Quantum yields of direct
phototransformation of 1,2,3,7-tetra-, 1,3,6,8-tetra-, 1,2,3,4,6,7,8-hepta-, and 1,2,3,4,6,7,8,9-octachlorodibenzo-p-
dioxin in aqueous acetonitrile and their sunlight half-lives. J Agric Food Chem 37:254-261.
Christmann, W; Kasiske, D; Kloppel, KD; et al. (1989a) Combustion of polyvinyl chloride: an important source for
the formation of PCDD/PCDF. Chemosphere 19(l-6):387-392.
Christmann, W; Kloppel, KD; Partscht, H; et al. (1989b) Tetrachlorobenzoquinones, a source of PCDD/PCDF.
Chemosphere 18(l-6):789-792.
Cianciarelli, D. (2000) Measurement of PCDD/PCDF and hexachlorobenzene emissions from an electric arc furnace
shop, Dofasco Inc., Hamilton, Ontario. Emissions Research Advancement Division, Environment Canada. Report
ERMD 99-08.
R-7
-------
Cianciarelli, D. (2001) Characterization of organic emissions from an electric arc furnace, Gerdau Courtice Steel
Inc., Cambridge, Ontario. Emissions Research and Measurement Division, Environmental Technology
Advancement Directorate, Environment Canada. Report ERMD 2000-05.
Charles E. Napier Company, Ltd. (2000) Draft: background technical discussion paper on the release and control of
dioxins/furans from the steel sector. Prepared for Environment Canada, Minerals and Metals Division.
Charles E. Napier Company, Ltd. (2002) Final: generic dioxin/furan emission testing protocol for the base metals
smelting sector. Prepared for Environment Canada.
Clark, T; Clark, K; Paterson, S; et al. (1988) Wildlife monitoring, modeling, and fugacity. Environ Sci Technol
22(2): 120-127.
Clement, RE; Tashiro, C. (1991) Forest fires as a source of PCDD and PCDF. Presented at Dioxin '91, 11th
International Symposium on Chlorinated Dioxins and Related Compounds; Research Triangle Park, NC; September
1991.
Clement, RE; Tosine, HM; Ali, B. (1985) Levels of polychlorinated dibenzo-p-dioxins and dibenzofurans in wood
burning stoves and fireplaces. Chemosphere 14:815-819.
Clement, RE; Tosine, HM; Osborne, J; et al. (1988) Gas chromatographic/mass spectrometric determination of
chlorinated dibenzo-p-dioxins and dibenzofurans in incinerator stack emissions and fly-ash: a 13-test study.
Biomedical and Environmental Mass Spectrometry 17:81-96.
Cleverly, DH; Morrison, RM; Riddle, BL; et al. (1991) Regulatory analysis of pollutant emissions, including
polychlorinated dibenzo-p-dioxins (CDDs) and dibenzofurans (CDFs), from municipal waste combustors. In:
Hattemer-Frey, HA; Travis, C; eds. Health effects of municipal waste incineration. Boca Raton, FL: CRC Press; pp.
47-65.
Cleverly, D; Schaum, J; Schweer, G; et al. (1997) The congener profiles of anthropogenic sources of chlorinated
dibenzo-p-dioxins and chlorinated dibenzofurans in the United states. Organohalogen Compounds 32:430-435.
Commoner, B; McNamara, M; Shapiro, K; et al. (1984) Environmental and economic analysis of alternative
municipal solid waste disposal technologies, part II: the origins of chlorinated dioxins and dibenzofurans emitted by
incinerators that burn unseparated municipal solid waste, and an assessment of methods of controlling them. Center
for the Biology of Natural Systems, Queens College, CUNY, Flushing, New York.
Commoner, B; Webster, T, Shapiro, K; et al. (1985) The origins and methods of controlling polychlorinated
dibenzo-p-dioxin and dibenzofuran emissions from MSW incinerators. Presented at the 78th Annual Meeting of the
Air Pollution Control Association; Detroit, MI; June 1985.
Commoner, B; Shapiro, K; Webster, T. (1987) The origin and health risks of PCDD and PCDF. Waste Management
and Research 5:327-346.
Commonwealth Aluminum Corp (1995). Item II-D-64 in USEPA secondary aluminum MACT air docket number
A-92-61; letter and attachment, S. Bruntz, Commonwealth Aluminum Corp., July 5, 1995, enclosing test report and
discussion of test report results, test report entitled MACT emission testing of the 'B' rotary kiln at
Commonwealth Aluminum Corporation, Lewisport, Kentucky (9 volumes total).
Costner, PK. (1998) Correlation of chlorine input and PCDD/PCDF emissions at a full-scale hazardous waste
incinerator. Organohalogen Compounds 36:147-151.
Council of Great Lakes Industries. (1999) Octachlorostyrene and suggested industrial sources. A report to the Great
Lakes Binational Toxics Strategy OCS Workgroup. Council of Great Lakes Industries, Ann Arbor, MI.
R-8
-------
Cramer, PH; Heiman, J; Horrigan, M; et al. (1995) Results of a national survey for polychlorinated dibenzo-p-
dioxins, dibenzofurans, and coplanar polychlorinated biphenyls in municipal biosolids. Organohalogen Compounds
24:305-308.
CRE (Coal Research Establishment). (1994) Emissions of environmental concern from coal utilization. Coal
Research Establishment, National Coal Board, Stoke Orchard, United Kingdom. Report number ECSC 7220-
EC/011.
Cull, MR; Dobbs, AJ; Goudot, M; et al. (1984) Polychlorodibenzo-p-dioxins and dibenzofurans in technical
pentachlorophenol-results of a collaborative analytical exercise. Chemosphere 13(10): 1157-1165.
Curlin, LC; Bommaraju, TV. (1991) Alkali and chlorine products. In: Kirk-Othmer encyclopedia of chemical
technology. 4th ed. pp. 938-1025.
deBoer, J.; Stronck, CJN; Traag, WA; et al. (1993) Non-ortho and mono-ortho substituted chlorobiphenyls and
chlorinated dibenzo-p-dioxins, and dibenzofurans in marine and freshwater fish and shellfish from The Netherlands.
Chemosphere 26(10): 1832-1842.
De Fre, R; Rymen, R. (1989) PCDD and PCDF formation from hydrocarbon combustion in the presence of hydrogen
chloride. Chemosphere 19(l-9):331-336.
Dempsey, CR; Oppelt, ET. (1993) Incineration of hazardous waste: a critical review update. Air and Waste 43:25-
73.
Deutsch, DG; Goldfarb, TD. (1988) PCDD/PCDF contamination following a plastic fire in a university lecture hall
building. Chemosphere 17(12):2423-2431.
DeVoogt, P; Brinkman, UA. (1989) Production, properties and usage of polychlorinated biphenyls. In: Kimbrough,
R; Jensen, AA, eds. Halogenated biphenyls, terphenyls, napthalenes, dibenzodioxins and related products. New
York, NY: Elsevier Science Publishers.
Dickson, LC; Karasek, FW. (1987) Mechanism of formation of polychlorinated dibenzo-p-dioxins produced on
municipal incinerator flyash from reactions of chlorinated phenols. J Chromatography 389:127-137.
Dickson, LC; Lenoir, D; Hutzinger, O. (1992) Quantitative comparison of de novo and precursor formation of
polychlorinated dibenzo-p-dioxins under simulated municipal solid waste incinerator post combustion conditions.
Environ Sci Technol 26:1822-1828.
Domalski, ES; Churney, KL; Ledford, AE; et al. (1986) Monitoring the fate of chlorine from MSW sampling
through combustion, part 1: analysis of the waste stream for chlorine. Chemosphere 15(9-12): 1339-1354.
Donnelly, JR. (1992) Waste incineration sources: refuse. In: Buonicore, AJ; Davis, WT; eds. Air pollution
engineering manual. Air and Waste Management Association. New York, NY: Van Nostrand Reinhold. pp. 263-
275.
Douben, PET; Eduljee, GH; Dyke, P. (1995) A review of potential PCDD and PCDF emission sources in the UK.
Organohalogen Compounds 24:131-136.
Duarte-Davidson, R; Sewart, A; Alcock, RE; et al. (1997) Exploring the balance between sources, deposition and the
environmental burden of PCDD/Fs in the U.K. terrestrial environment: an aid to identifying uncertainties and
research needs. Environ Sci Technol 31(1):1-11.
Duinker, JC; Hillebrand, MTJ. (1983) Characterization of PCB components in Clophen formulations by capillary
GC-MS and GC-ECD techniques. Environ Sci Technol 17(8):449-456.
R-9
-------
Duinker, JC; Schulz, DE; Patrick, G. (1988) Multidimensional gas chromatography with electron capture detection
for the determination of toxic congeners of polychlorinated biphenyl mixtures. Analytical Chemistry 60:478-482.
Dumler-Gradl, R; Thoma, H; Vierle, O. (1995) Research program on dioxin/furan concentration in chimney soot
from house heating systems in the Bavarian area. Organohalogen Compounds 24:115-118.
Dung, M; O'Keefe, PW. (1992) Comparative rates of photolysis of polychlorinated dibenzofurans in organic
solvents and in aqueous solutions. Organohalogen Compounds 8:233-236.
Dyke, P; Coleman, P. (1995) Dioxins in ambient air, Bonfire Night 1994. Organohalogen Compounds 24:213-216.
Edelstein, DL. (1995) Minerals yearbook: copper. U.S. Geological Survey, Geologic Division, Reston, VA.
Edelstein, DL. (2000) Minerals yearbook: copper. U.S. Geological Survey, Geologic Division, Reston, VA.
Edelstein, DL. (2001) Minerals yearbook: copper. U.S. Geological Survey, Geologic Division, Reston, VA.
Eduljee, GH; Dyke, P. (1996) An updated inventory of potential PCDD and PCDF emission sources in the UK. The
Science of the Total Environment 177:303-321.
EIA (Energy Information Administration). (1991) Estimates of U.S. biofuels consumption in 1990. U.S. Department
of Energy, Office of Coal, Nuclear, Electric and Alternate Fuels, Washington, DC; DOE/EIA-0548(90).
EIA (Energy Information Administration). (1992) Fuel oil and kerosene sales 1991. U.S. Department of Energy,
Office of Oil and Gas, Washington, DC; DOE/EIA-0535(91).
EIA (Energy Information Administration). (1993) Petroleum supply annual 1992. Vol. 1. U.S. Department of
Energy, Office of Oil and Gas, Washington, DC; DOE/EIA-0340(92)/1.
EIA (Energy Information Administration). (1994) Estimates of U.S. biomass energy consumption 1992. U.S.
Department of Energy, Office of Coal, Nuclear, Electric, and Alternative Fuels, Washington, DC; DOE/EIA-
0548(92).
EIA (Energy Information Administration). (1995) Annual energy review 1994. U.S. Department of Energy, Office of
Energy Markets and End Use, Washington, DC; DOE/EIA-0884(94).
EIA (Energy Information Administration). (1997a) Fuel oil and kerosene sales 1996. U.S. Department of Energy,
Office of Oil and Gas, Washington, DC; DOE/EIA-0535(96).
EIA (Energy Information Administration). (1997b) Annual energy review 1996. U.S. Department of Energy, Office
of Energy Markets and End Use, Washington, DC; DOE/EIA-0384(96).
EIA (Energy Information Administration). (1999) State energy data report 1996, consumption estimates. U.S.
Department of Energy, Office of Energy Markets and End Use, Washington, DC; DOE/EIA-0214(96).
EIA (Energy Information Administration). (2002) Annual energy review 2000. Available online at
www.eia.doe.gov.
EIA (Energy Information Administration). (2003a) 2000 State energy data. Available online at
http ://www. eia. doe. gov/emeu/states/sep_fuel/html/pdf/fuel_all.pdf).
EIA (Energy Information Administration). (2003b) Monthly energy review, energy consumption by sector for the
year 2000. Available online at www.eia.doe.gov/mer/consump.html.
R-10
-------
Eitzer, BD; Hites, RA. (1987) Reply to comment on "Airborne dioxins and dibenzofurans: sources and fate."
Environ Sci Technol 21:924.
Eitzer, BD; Hites, RA. (1989) Dioxins and furans in the ambient atmosphere: A baseline study. Chemosphere 18(1-
6):593-598.
Eklund, G; Pedersen, JR; Stromberg, B. (1986) Phenol and HC1 at 550°C yield a large variety of chlorinated toxic
compounds. Nature 320:155-156.
Eklund, G; Pedersen, JR; Stromberg, B. (1988) Methane, hydrogen chloride and oxygen form a wide range of
chlorinated organic species in the temperature range 400°C-959°C. Chemosphere 17(3):575-586.
Environics Research Group. (2001) Household garbage disposal and burning: Ontario survey. Prepared for
Environment Canada.
Environment Canada. (1985) The national incineration testing and evaluation program: two-stage combustion
(Prince Edward Island). Environment Canada, Ottawa, Ontario, Canada. Report EPS 3/UP/l.
Environment Canada. (2000) Characterization of organic compounds from selected residential wood stoves and
fuels. Environment Canada, Emissions Research and Measurement Division, Ottawa, Canada. Report ERMD 2000-
01.
Environmental Risk Sciences, Inc. (1995) An analysis of the potential for dioxin emissions in the primary copper
smelting industry. Prepared for the National Mining Association.
EPRI (Electric Power Research Institute). (1994) Electric utility trace substances synthesis report. Palo Alto, CA;
EPRITR-104614.
EPRI (Electric Power Research Institute). (1995) Pentachlorophenol (PCP) in soils adjacent to in-service utility
poles in New York State. Palo Alto, CA; EPRITR-104893.
ERG (Eastern Research Group, Inc.) (1998) Report of the meeting to peer review inventory of sources of dioxin in
the United States. Prepared for the U.S. Environmental Protection Agency, Office of Research and Development,
Washington, DC. EPA Contract No. 68-D5-0028, Work Assignment No. 98-05.
Erickson, M. (1986) Analytical chemistry of PCBs. Boston, MA: Butterworth Publishers.
Erickson, MD; Cole, CJ; Flora, JD; et al. (1984) Thermal degradation products from dielectric fluids. U.S.
Environmental Protection Agency, Office of Toxic Substances, Washington, DC; EPA-560/5-84-009.
Esposito, MP; Tiernan, TO; Drydent, FE. (1980) Dioxins. Prepared for the U.S. Environmental Protection Agency,
Cincinnati, OH; EPA-600/2-80-197.
Fangmark, I; Stomberg, G; Berge, N; et al. (1994) Influence of postcombustion temperature profiles on the
formation of PCDDs, PCDFs, PCBzs, and PCBs in a pilot incinerator. Environ Sci Technol 28:624-629.
Federal Register. (1979) Decision and emergency order suspending registrations for certain uses of 2,4,5-T and
Silvex. FR (March 15) 44:15874-15920.
Federal Register. (1982a) Polychlorinated biphenyls (PCBs) use in electrical equipment. FR (April 22) 47:17426.
Federal Register. (1982b) Polychlorinated biphenyls (PCBs) manufacturing, processing, distribution in commerce
and use prohibitions; use in electrical equipment. FR (August 25) 47:37342.
R-ll
-------
Federal Register. (1984) Pentachlorophenol, and inorganic arsenicals; intent to cancel registrations of pesticide
products containing creosote, pentachlorophenol (including its salts) and the inorganic arsenicals; determination
concluding the rebuttal presumption against registration of the wood preservative uses of pesticide products;
availability of position document. FR(July 13)49:28666-28689.
Federal Register. (1985a) Regulation of fuels and fuel additives; gasoline lead content: final rule and supplemental
notice of proposed rulemaking. FR (March 7) 50:9386-9408.
Federal Register. (1985b) Fob/chlorinated biphenyls in electrical transformers. FR (July 17) 50:29170.
Federal Register. (1986) Creosote, pentachlorophenol, and inorganic arsenicals; amendment of notice of intent to
cancel registrations. FR (January 10) 51:1334-1348.
Federal Register. (1987a) Polyhalogenated dibenzo-p-dioxins/dibenzofurans; testing and reporting requirements;
final rule. FR (June 5) 52:21412-21452.
Federal Register. (1987b) Pentachlorophenol; amendment of notice of intent to cancel registrations. FR (January 2)
52:140-148.
Federal Register. (1987c) Final determination and intent to cancel and deny applications for registrations of pesticide
products containing pentachlorophenol (including but not limited to its salts and esters) for non-wood uses. FR
(January 21) 52:2282-2293.
Federal Register. (1990) National sewage sludge survey: availability of information and data, and anticipated
impacts on proposed regulation. FR (November 9) 55:47210-47283.
Federal Register. (199 la) Standards of performance for new stationary sources and final guidelines for municipal
waste combustors. FR (February 11) 56:5488-5527.
Federal Register. (1991b) Burning of hazardous waste in boilers and industrial furnaces-final rule. FR (February 21)
56:7134.
Federal Register. (1991c) Standards of performance for new stationary sources and guidelines for control of existing
sources: municipal solid waste landfills: proposed rule and guideline. FR (May 30) 56:24468.
Federal Register. (1993a) Standards for the use or disposal of sewage sludge; final rules. FR (February 19) 58:9248-
9404.
Federal Register. (1993b) Standards of performance for new stationary sources and guidelines for control of existing
sources: municipal solid waste landfills: notice of availability. FR (June 21) 58:33790-33792.
Federal Register. (1993c) Chemicals; toxic chemical release reporting; community right-to-know; significant new
use rule. FR (December 1) 58:63500-63518.
Federal Register. (1993d) 2,3,5,6-Tetrachloro-2,5-cyclohexadiene-l,4-dione; proposed significant new use of a
chemical substance. FR (May 12) 58:27980-27986.
Federal Register. (1995a) Standards of performance for new stationary sources and emission guidelines for existing
sources: municipal waste combustors, final rule. FR (December 19) 60:65387-65436.
Federal Register. (1995b) Hazardous waste management system; identification and listing of hazardous waste:
petroleum refining process wastes; land disposal restrictions for newly identified wastes; and CERCLA hazardous
substance designation and reportable quantities. FR (November 20) 60:57747-57800.
R-12
-------
Federal Register. (1996a) Revised standards for hazardous waste combustors: proposed rule. FR (April 19)
61:17357-17536.
Federal Register. (1996b) Standards of performance for new stationary sources and guidelines for control of existing
sources: municipal solid waste landfills: final rule and guideline. FR (March 12) 61:9905-9944.
Federal Register. (1997a) Standards of performance for new stationary sources and emission guidelines for existing
sources: hospital/medical/infectious waste incinerators; final rule. FR (September 15) 62:48347-48391.
Federal Register. (1997b) Toxic chemicals; PCBs; submission of IRC No. 1000 to OMB; information collection
activities. FR (January 21) 62:3035-3036.
Federal Register. (1998a) Source category listing for Section 112(d)(2) rulemaking pursuant to Section 112(c)(6)
requirements. FR (April 10)63:17838-17855.
Federal Register. (1998b) Hazardous waste management system; identification and listing of hazardous waste;
petroleum refining process wastes; land disposal restrictions for newly identified wastes; and CERCLA hazardous
substance designation and reportable quantities. FR (August 8) 63:42110-42189.
Federal Register. (1998c) National emission standards for hazardous air pollutants for source category: pulp and
paper production; effluent limitations guidelines, pretreatment standards, and new source performance standards:
pulp, paper, and paperboard category. FR (April 15) 63:18504-18751.
Federal Register. (1999a) NESHAPS: final standards for hazardous air pollutants for hazardous waste combustors.
FR (September 30) 64:52828-53077.
Federal Register. (1999b) National emission standards for hazardous air pollutants for source categories; Portland
Cement manufacturing industry. FR (June 14) 64:31897-31962.
Federal Register. (1999c) Standards for the management of cement kiln dust; proposed rule. FR (August 20) 64:161.
Federal Register. (2003) Federal plan requirements for small municipal waste combustion units constructed on or
before August 30, 1999. F.R. (January 31) 68:5144-5187.
Federal Register. (2004) National emission standards for hazardous air pollutants: proposed standards for hazardous
air pollutants for hazardous waste combustors (phase 1 final replacement standards and phase II): proposed rule. FR
(April 20) 69:21347-21385.
Feil, VJ; Tiernan, T. (1997) Pentachlorophenol as a source of dioxins and furans. Organohalogen Compounds
33:353-354.
FEMA (Federal Emergency Management Agency). (1997) Fire in the United States. United States Fire
Administration, Washington, DC. FA-173.
FEMA (Federal Emergency Management Agency). (2001) The national fire problem. United States Fire
Administration, Washington, DC. Available online at http://www.usfa.fema.gov/nfdc/national.htm.
Fenton, M. (1996) Minerals yearbook: iron and steel. U.S. Department of the Interior, U.S. Geological Survey,
Washington, DC.
Fenton, M. (2001) Minerals yearbook: iron and steel. U.S. Department of the Interior, U.S. Geological Survey,
Washington, DC.
Ferrario, J; Byrne, C; Lorber, M; et al. (1997) A statistical survey of dioxin-like compounds in United States poultry
fat. Organohalogen Compounds 32:245-251.
R-13
-------
Ferrario, J; Byrne, C; Cleverly, D. (2000) 2,3,7,8-Dibenzo-p-dioxins in mined clay products from the United States:
evidence for possible natural formation. Environ Sci Technol 34:4524-4532.
Fiedler, H; Hutzinger, O. (1992) Sources and sinks of dioxins: Germany. Chemosphere 25:1487-1491.
Fiedler, H; Lindert, M. (1998) Dtisseldorf Airport fire 1996: results and comparison with accidental fires.
Organohalogen Compounds 36:69-72.
Fiedler, H; Hutzinger, O; Hosseinpour, J. (1993) Analyses and remedial actions following an accidental fire in a
kindergarten. Organohalogen Compounds 14:19-22.
Firestone, D; Ress, J; Brown, NL; et al. (1972) Determination of polychlorodibenzo-p-dioxins and related
compounds in commercial chlorophenols. J Assoc Off Anal Chem 55(l):85-92.
Fisher, TS; Hayward, DG; Stephens, RD; et al. (1999) Dioxins and furans in Los Angeles urban runoff. J Environ
Engineering 125:185-191.
Fleischer, O; Wichmann, H; Lorenz, W. (1999) Release of polychlorinated dibenzo-p-dioxins and dibenzofurans by
setting off fireworks. Chemosphere 39:925-932.
Frame, GM. (1995) A comprehensive, multicenter study of PCB elution on capillary GC columns. Pittcon '95.
Sponsored by the Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, Inc. New Orleans, LA;
March 8, 1995; Paper 717.
Frame, GM. (1997) A collaborative study of 209 PCB congeners and 6 aroclors on 20 different HRGC columns.
Fresenius Journal of Analytical Chemistry 357(6):714-722.
Frame, GM. (1999) Improved procedure for single DB-XLB column GC MS-SIM quantitation of PCB congener
distributions and characterization of two different preparations sold as "Aroclor 1254". J High Resol Chromatogr
22(10):533-540.
Frame, GM; Cochran, JW; Bowadt, SS. (1996a) Complete PCB congener distributions for 17 aroclor mixtures
determined by 3 HRGC systems optimized for comprehensive, quantitative, congener-specific analysis. J High
Resol Chromatogr 19(12):657-668.
Frame, GM; Wagner, RE; Carnahan, JC; et al. (1996b) Comprehensive, quantitative, congener-specific analyses of
eight aroclors and complete PCB congener assignments on DB-1 capillary GC columns. Chemosphere 33(4):603-
623.
Frank, H; Frank, W. (1989) Uptake of airborne tetrachloroethane by spruce needles. Environ Sci Technol
23(3):365-367.
Freeman, RA; Schroy, J.M. (1985) Environmental mobility of dioxins. Aquatic Toxicology and Hazard Assessment:
Eighth Symposium, ASTM STP 891, Banner, RC; Hansen, DJ; eds. American Society for Testing and Materials,
Philadelphia, PA; pp. 422-439.
Fries, GF; Dawson, TE; Paustenbach, DJ; et al. (1997) Biosynthesis of hepta- and octa-chlorodioxins in cattle and
evidence for lack of involvement by rumen microorganisms. Organohalogen Compounds 33:296-301.
Friesen, JK; Muir, DCG; Webster, GRB. (1990) Evidence of sensitized photolysis of polychlorinated dibenzo-p-
dioxins in natural waters under sunlight conditions. Environ Sci Technol 24(11): 1739-1744.
Friesen, KJ; Loewen, MD; Foga, MM. (1993) Environmental aquatic photodegradation of chlorinated dibenzofurans
and their photoproducts. Organohalogen Compounds 12:135-137.
R-14
-------
Froese, KL; Hutzinger, O. (1994) The formation of chlorinated benzenes and phenols in fly ash catalyzed reactions
of trichloroethylene. Chemosphere 28 (11): 1977-1987.
Froese, KL; Hutzinger, O. (1996a) Polychlorinated benzene and polychlorinated phenol in heterogeneous
combustion reactions of ethylene and ethane. Environ Sci Technol 30:1009-1013.
Froese, KL; Hutzinger, O. (1996b) Polychlorinated benzene, phenol, dibenzo-p-dioxin, and dibenzofuran in
heterogeneous combustion reactions of acetylene. Environ Sci Technol 30: 998-1008.
Froese, K; Hutzinger, O. (1997) Mechanisms of formation of polychlorinated benzenes and phenols from
heterogeneous reactions of Cs aliphatics. Environ Sci Technol 31:542-547.
Fuchs, R; Towara, J; Hutzinger, O. (1990)PCDD/Finthe dry-cleaning process. Organohalogen Compounds 3:441-
444.
Funcke, W; Theisen, J; Balfanz, E; Konig, J. (1988) Polychlorierte dibenzofurane (PCDF) und polychlorierte
dibenzo(p)dioxine (PCDD) in Ruckstanden und emissionen eines Brandes in Anwesenheit von PBC-haltigen
materialien. Staub-Reinhaltung der Luft 48:393-396.
Gaggi, C; Bacci, E; Calamari, D; et al. (1985) Chlorinated hydrocarbons in plant foliage: an indication of the
tropospheric contamination level. Chemosphere 14(11-12): 1673-1686.
Galson Corporation. (1995) Source test report: source emission testing at Roth Brothers Smelting Corporation.
Galson Corporation, East Syracuse, NY.
Gambogi, J. (1996) Minerals yearbook: titanium. U.S. Geological Survey, Geologic Division, Reston, VA.
Gan, DR; Berthouex, PM. (1994) Disappearance and crop uptake of PCBs from sludge-amended farmland. Water
Environment Research 66(l):54-69.
Gertler, AW; Sagebiel, JC; Dippel, WA; et al. (1996) A study to quantify on-road emissions of dioxins and furans
from mobile sources: phase 2. Desert Research Institute, Reno, NV.
Gertler, A; Sagebiel, J; Dippel, W; et al. (1998) Measurement of dioxin and furan emissions factors from heavy duty
diesel vehicles. J of the Air and Waste Mgmt Assoc 48:276-278.
Geueke, K; Gessner, A; Quass, U; et al. (1999) PCDD/F emissions from heavy duty vehicle diesel engines.
Chemosphere 38:2791-2806.
Gillespie, WJ. (1994) A summary of data reflective of U.S. pulp and paper industry progress in reducing the
TCDD/TCDF content of effluents, pulps and wastewater treatment sludges. Organohalogen Compounds 20:255-
260.
Gillespie, WJ. (1995) Progress in reducing the TCDD/TCDF content of effluents, pulps and wastewater treatment
sludges from the manufacturing of bleached chemical pulp. National Council of the Paper Industry for Air and
Stream Improvement, Inc., Research Triangle Part, NC.
Oilman, AP; Douglas, VM; Newhook, RC; et al. (1988) Chlorophenols and their impurities: a health hazard
evaluation. Health and Welfare Canada. Document No. H46-2/88-110E.
Gobran, T; Khurana, V; MacPherson, K; et al. (1995) PCDD/PCDF levels in raw sewage, final effluent, sludge, and
ash samples from an Ontario wastewater treatment plant. Organohalogen Compounds 24:75-80.
R-15
-------
Goldfarb, TD; Malloy, TA; Surico, M. (1992) PCDDs, PCDFs, PCBs, chlorophenols (CPs) and chlorobenzenes
(CBzs) in samples from various types of composting facilities in the United States. Organohalogen Compounds
8:253-256.
Green, LC; Lester, RL; Armstrong, SR. (1995) Comments on EPA's external review draft Estimating Exposures to
Dioxin-Like Compounds. Cambridge Environmental Inc., Cambridge, MA. Dated January 12, 1995, and May 11,
1995.
Greenpeace. (1993) Dioxin factories: a study of the creation and discharge of dioxins and other organochlorines
from the production of PVC. Greenpeace International, Amsterdam, The Netherlands.
Greer, WL; Johnson, MD; Raught, EC; et al. (1992) Portland cement. In: Buonicore, AJ; Davis, WT; eds. Air
pollution engineering manual. Air and Waste Management Association. New York, NY: Van Nostrand Reinhold.
Gribble, GW. (1994) The natural production of chlorinated compounds. Environ Sci Technol 28(7):310A-319A.
Guerin, MR; Jenkins, RA; Tomkins, BA. (1992) The chemistry of environmental tobacco smoke: composition and
measurement. Chelsea, MI: Lewis Publishers.
Gullett, BK; Lemieux, PM. (1994) Role of combustion and sorbent parameters in prevention of polychlorinated
dibenzo-p-dioxin and polychlorinated dibenzofuran formation during waste combustion. Environ Sci Technol
28(1):107-118.
Gullett, BK; Ryan, JV. (1997) On-road sampling of diesel engine emissions of polychlorinated dibenzo-p-dioxin and
polychlorinated dibenzofuran. Organohalogen Compounds 32:451-456.
Gullett, B; Touati, A. (2003) PCDD/F emissions from forest fire simulations. Atmospheric Environment 37:803-
813.
Gullett, BK; Bruce, KR; Beach, LO. (1990a) Formation of chlorinated organics during solid waste combustion.
Waste Management and Research 8:203-214.
Gullett, BK; Bruce, KR; Beach, LO. (1990b) The effect of metal catalysts on the formation of polychlorinated
dibenzo-p-dioxin and polychlorinated dibenzofuran precursors. Chemosphere 20:1945-1952.
Gullett, BK; Bruce, KR; Beach, LO. (1991a) The effect of sulfur compounds on the formation mechanism of PCDD
and PCDF in municipal waste combustors. U.S. EPA, Research Triangle Park,NC; Acurex Corp., Research Triangle
Park,NC. In: Conference papers from the second international conference on municipal waste combustion, April 15-
19, 1991, Tampa, FL. Air and Waste Management Association, Pittsburgh, PA; pp. 16-34.
Gullett, BK; Bruce, KR; Beach, LO; et al. (1991b) Mechanistic steps in the production of PCDD and PCDF during
waste combustion. Presented at Dioxin '91, 11th International Symposium on Chlorinated Dioxins and Related
Compounds; Research Triangle Park, NC; September 1991.
Gullett, B; Bruce, K; Beach, L. (1992) Mechanistic steps in the production of PCDD and PCDF during waste
combustion. Chemosphere 25(7-10): 1387-1392.
Gullett, BK; Lemieux, PM; Lutes, CC; et al. (1999) PCDD/F emissions from uncontrolled, domestic waste burning.
Organohalogen Compounds 41:27-30.
Gullett, BK; Lemieux, PM; Winterrowd, CK; et al. (2000) PCDD/F emissions from uncontrolled, domestic waste
burning. Organohalogen Compounds 46:193-196.
Gullett, BK; Lemieux, PM; Lutes, CC; et al. (2001) Emissions of PCDD/F from uncontrolled, domestic waste
burning. Chemosphere 43:721-725.
R-16
-------
Gurprasad, N; Constable, M; Haidar, N; et al. (1995) Polychlorinated dibenzo-p-dioxins (PCDDs) leaching from
pentachlorophenol-treated utility poles. Organohalogen Compounds 24:501-504.
Gutenmann, WH; Rutzke, M; Kuntz, HT; et al. (1994) Elements and polychlorinated biphenyls in sewage sludges of
large cities in the United States. Chemosphere 28(4):725-728.
Hagenmaier, H. (1986) Determination of 2,3,7,8-tetrachlorodibenzo-p-dioxin in commercial chlorophenols and
related products. Fresenius Z Anal Chem 325:603-606.
Hagenmaier, H. (1987) Belastung der unwell mit dioxinen [Environmental loading with dioxins]. Institut fur
Organische Chemie der Universitat Tubingen.
Hagenmaier, H. (1994) Contributions of diesel-powered vehicles and wood burning to overall PCDD/PCDF
emissions. Organohalogen Compounds 20:267-270.
Hagenmaier, H; Brunner, H. (1987) Isomer-specific analysis of pentachlorophenol and sodium pentachlorophenate
for 2,3,7,8-substituted PCDD and PCDF at sub-ppb levels. Chemosphere 16:1759-1764.
Hagenmaier, H; Dawidowsky, V; Weber, UB; et al. (1990) Emission of polyhalogenated dibenzodioxins and
dibenzofurans from combustion-engines. Organohalogen Compounds 2:329-334.
Hagenmaier, H; Lindig, C; She, J. (1994) Correlation of environmental occurrence of polychlorinated dibenzo-p-
dioxins and dibenzofurans with possible sources. Chemosphere 29(9-ll):2163-2174.
Haglund, P; Egeback, K-E; Jansson, B. (1988) Analysis of PBDD/F in vehicle exhaust. Presented at: Dioxin '88, 8th
International Symposium on Chlorinated Dioxins and Related Compounds; Umea, Sweden; August 1988.
Halonen, L; Tarhanen, J; Ollikainen, S; et al. (1994) The effect of inorganic and organic chlorine on formation of
highly chlorinated organic compounds during incineration: laboratory pilot study. Chemosphere 28(12):2129-2138.
Hammerstrom, K; Schweer, G; Adkins, L; et al. (1985) Exposure assessment for polychlorinated biphenyls (PCBs):
incidental production, recycling, and selected authorized uses. U.S. Environmental Protection Agency, Office of
Toxic Substances, Washington, DC. EPA Contract No. 68-01-6271, Task No. 21.
Harner, T; Mackay, D; Jones, KC. (1995) Model of the long-term exchange of PCBs between soil and the
atmosphere in the southern U.K. Environ Sci Technol 29(5): 1200-1209.
Harnly, M; Stephens, R; McLaughlin, C; et al. (1995) Polychlorinated dibenzo-p-dioxin and dibenzofuran
contamination at metal recovery facilities, open burn sites, and a railroad car incineration facility. Environ Sci
Technol 29(3):677-684.
Harrad, SJ; Jones, KC. (1992) A source inventory and budget for chlorinated dioxins and furans in the United
Kingdom environment. The Science of the Total Environment 126:89-107.
Harrad, SJ; Malloy, TA; Khan, MA; et al. (1991) Levels and sources of PCDDs, PCDFs, chlorophenols (CPs) and
chlorobenzenes (CBzs) in composts from a municipal yard waste composting facility. Chemosphere 23 (2): 181 -191.
Hay, DJ; Finkelstein, A; Klicius, R. (1986) The national incineration testing and evaluation program: an assessment
of two-stage incineration and pilot scale emission control. Paper presented to the U.S. EPA Science Advisory
Board, Washington, DC. Environment Canada, Ottawa, Ontario, Canada.
Headrick, M; Hollinger, K; Lovell, R; et al. (1999) PBBs, PCBs, and dioxins in food animals, their public health
implications. Vet Clin North Am Food Anim Pract 15(1):109-131.
R-17
-------
Heindl, A; Hutzinger, O. (1987) Search for industrial sources of PCDD/PCDF, III. short-chain chlorinated
hydrocarbons. Chemosphere 16(8/9): 1949-1957.
Helble, JJ. (1993) Analysis of dioxin emissions from the incineration of hazardous waste. U.S. Environmental
Protection Agency, Office of Solid Waste, Permits and State Programs Division, Washington, DC.
Hengstmann, R; Hamann, R; Weber, H; et al. (1990) Impact of sewage sludge and sewer slimes by polychlorinated
dibenzo-p-dioxins and polychlorinated dibenzofurans. Organohalogen Compounds 4:407-412.
Hickey, WJ. (1995) Enhancing PCB biodegradatin in soil with chlorobenzoate-degrading inoculants.
Organohalogen Compounds 24:29-32.
Himberg, K; Sippola, E. (1990) Multidimensional gas chromatography: mass spectrometry as a method for the
determination of coplanar PCB congeners. Organohalogen Compounds 4:183-186.
HMIP. (1995) A review of dioxin emissions in the UK. Her Majesty's Inspectorate of Pollution.
HMIP/CPR2/41/1/38.
Ho, AKW; Clement, RE. (1990) Chlorinated dioxins/furans in sewage and sludge of municipal water pollution
control plants. Chemosphere 20(10-12):1549-1552.
Hoekstra, EJ; De Leer, WB; Brinkman, UT. (1999) Natural formation of chlorinated phenols, dibenzo-p-dioxins, and
dibenzofurans in soil of a Douglas fir forest. Environ Sci and Technol 33:2543-2549.
Holderman, TS; Cramer, PH. (1995) Updated results of polyhalogenated dibenzo-p-dioxin/dibenzofuran testing and
reporting under the Toxic Substances Control Act (TSCA). Organohalogen Compounds 24:137-140.
Hornbuckle, KC; Achman, DR; Eisnreich, SJ. (1993) Over-water and over-land, polychlorinated biphenyls in Green
Bay, Lake Michigan. Environ Sci Technol 27(l):87-97.
Horstmann, M; McLachlan, MS. (1992) Initial development of a solid-phase fugacity meter for semivolatile organic
compounds. Environ Sci Technol 26(8): 1643-1649.
Horstmann, M; McLachlan, MS. (1994a) Textiles as a source of polychlorinated dibenzo-p-dioxins and
dibenzofurans (PCDD/CDF) in human skin and sewage sludge. Environ Sci and Pollut Res 1(1): 15-20.
Horstmann, M; McLachlan, MS. (1994b) Polychlorinated dibenzo-p-dioxins and dibenzofurans in textiles and their
transfer to human skin, sewage sludge and other matrices. Organohalogen Compounds 20:251-254.
Horstmann, M; McLachlan, MS. (1995) Concentrations of polychlorinated dibenzo-p-dioxins (PCDD) and
dibenzofurans (PCDF) in urban runoff and household wastewaters. Chemosphere 31(3):2887-2896.
Horstmann, M; McLachlan, M; Reissinger, M. (1992) Investigation of the origin of PCDD/CDF in municipal
sewage. Organohalogen Compounds 9:97-100.
Horstmann, M; McLachlan, M; Reissinger, M. (1993a) Further investigations of the sources of PCDD/CDF in
municipal sewage sludge. Organohalogen Compounds 11:293-296.
Horstmann, M; McLachlan, M; Reissinger, M; et al. (1993b) An investigation of PCDD/CDF formation during
textile production and finishing. Organohalogen Compounds 11:417-420.
Houck, G. (1991) Minerals yearbook: iron and steel. U.S. Geological Survey, Geologic Division, Reston, VA.
Huang, H; Buekens, A. (1995) On the mechanisms of dioxin formation in combustion processes. Chemosphere 31
(9):4099-4117.
R-18
-------
Huang, CW; Miyata, H; Lu, J; et al. (1992) Levels of PCBs, PCDDs and PCDFs in soil samples from incineration
sites for metal reclamation in Taiwan. Chemosphere 24(1 1): 1669-1676.
Huang, H; Biesemans, M, Buekens, A. (1999) Characterization of the nature of de novo synthesis by 13C NMR.
Organohalogen Compounds 41:105-108.
Huckins, JN; Stalling, DL; Petty, JD. (1980) Carbon-foam chromatographic separation of non-o-o '-chlorine
substituted PCBs from Aroclor mixtures. J Assoc Off Anal Chem 63(4):750-755.
Htilster, A; Marschner, H. (1993) Soil-plant transfer of CDD/PCDF to vegetables of the cucumber family
(Cucurbitaceae). Environ Sci Technol 28(6): 1 1 10-1 1 15.
Htilster, A; Marschner, H. (1994) The influence of root exudates on the uptake of CDD/PCDF by plants.
Organohalogen Compounds 20:3 1-34.
Hutzinger, O; Fiedler, H. (1991a) Formation of dioxins and related compounds from combustion and incineration
processes. In: Bretthauer, EW; Kraus, HW; di Domenico, A; eds. Dioxin perspectives. A pilot study on
international information exchange on dioxins and related compounds. New York, NY: Plenum Press.
Hutzinger, O; Fiedler, H. (1991b) Formation of dioxins and related compounds in industrial processes. In:
Bretthauer, EW; Kraus, HW; di Domenico, A; eds. Dioxin perspectives. A pilot study on international information
exchange on dioxins and related compounds. New York, NY: Plenum Press.
Hutzinger, O; Essers, U; Hagenmaier, H. (1992) Untersuchungen zur emission halogenierter dibenzodioxine und
dibenzofurane aus verbrennungsmortoren beim betrieb mit handelstiblichen betriebsstoffen. Universities of
Bayreuth, Stuttgart and Tubingen, Germany. GSF-Forschungszentrum, Munich, Germany, ISSN 0937-9932.
Huwe, JK; Feil, VJ; Tiernan, TO. (1998) In vivo formation of octachlorodibenzo-p-dioxin from a predioxin.
Organohalogen Compounds 36:93-95.
Imagawa, T; Lee, C. (2001) Correlation of polychlorinated naphthalenes with polychlorinated dibenzofurans formed
from waste incineration. Chemosphere 44:1151-1520.
Jager, J; Wilken, M; Zeschmar-Lahl, B. (1992) Dioxin und furan emissionen in Berlin eine hochrechnung [Dioxin
and furan emissions in Berlin, a computer forecast]. Staub-Reinhaltung der Luft 52:99-106.
Jansson, B; Sundstrom, G; Ahling, B. (1977) Formation of polychlorinated dibenzo-p-dioxins during combustion of
chlorophenol formulations. Science in the Total Environment 10:209-217.
Jarmohamed, W; Mulder, P. (1994) Oxychlorination and combustion of propene on fly-ash formation of chlorinated
benzenes, dibenzodioxins and mono- and dibenzofurans. Chemosphere 29 (9-11): 1911-1917.
Jensen, S; Sundstrom, G; Frescati, L. (1974) Structures and levels of most chlorobiphenyls in two technical PCB
products and in human adipose tissue. Ambio 3(2):70-76.
Jobst, H; Aldag, R. (2000) Dioxine In Lagersttatten-Tonen. UWSF-Zeitschrift furUmweltchem. Okotoxikologie
Jones, KC; Alcock, RE. (1996) Dioxin inputs to the environment: a review of temporal trend data and proposals for
a monitoring programme to detect past and future change in the U.K. Department of Environment report.
Kannan, N; Tanabe, S; Wakimoto, T; et al. (1987) Coplanar polychlorinated biphenyls in Aroclor and Kanechlor
mixtures. J Assoc Off Anal Chem 70(3):45 1-454.
R-19
-------
Kanters, J; Louw, R. (1994) Chlorine input and output in combustion of municipal solid waste in a lab-scale mini-
reactor system. Chemosphere 29 (9-11): 1919-1925.
Kao, AS; Venkataraman, C. (1995) Estimating the contribution of reentrainment to the atmospheric deposition of
dioxin. Chemosphere 31(10):4317-4331.
Katami, T; Yasuhara, A; Okuda, T; et al. (2002) Formation of PCDDs, PCDFs, and coplanar PCBs from
polyvinylchloride during combustion in an incinerator. Environ Sci Technol 36:1320-1324.
Kearney, PC; Woolson, EA; Isensee, AR; et al. (1973) Tetrachlorodibenzodioxin in the environment: sources, fate,
and decontamination. Environ Health Perspect 5:273-277.
Kieatiwong, S; Nguyen, LV; Hebert, VR; et al. (1990) Photolysis of chlorinated dioxins in organic solvents and on
soils. Environ Sci Technol 24(10): 1575-1580.
Kim, K-S; Hong, K-H; Ko, Y-H (2004) Emission characteristics of PCDD/Fs, PCBs, chlorobenzenes,
chlorophenols, and PAHs from polyvinylchloride bomustion at various temperatures. J Air Waste Manage Assoc
54:555-562.
Kirby, R. (1994) Petroleum refining industry: presence of dioxins and furans in wastewater generated by reforming
operations. U.S. Environmental Protection Agency, Office of Water, Engineering and Analysis Division,
Washington, DC.
Kjeller, L; Jones, KC; Johnston, AE; et al. (1991) Increases in the polychlorinated dibenzo-p-dioxin and -furan
content of soils and vegetation since the 1840s. Environ Sci Technol 25(9): 1619-1627.
Klasmeier, J; McLachlan, MS. (1995) A screening method for detection of OCDD/F in textiles. Organohalogen
Compounds 23:131-134.
Knepper, WA. (1981) Iron. In: Kirk-Othmer encyclopedia of chemical technology. 3rd ed. 13:735-753. New York,
NY: John Wiley and Sons.
Knittel, D. (1983) Titanium and titanium alloys. In: Kirk-Othmer encyclopedia of chemical technology. 3rd ed.
23:98-130. New York, NY: John Wiley and Sons.
Kobylecki, R; Ohira, K; Ito, I; et al. (2001) Dioxin and fly ash free incineration by ash pelletization and reburning.
Environ Sci Technol 35:4313-4319.
Koester, CJ; Kites, RA. (1992) Photodegradation of polychlorinated dioxins and dibenzofurans adsorbed to fly ash.
Environ Sci Technol 26(3):502-507.
Konduri, R; Altwicker, E. (1994) The effect of the electric field on the formation of PCDD in electrostatic
precipitators. Hazardous Waste and Hazardous Materials 11(1): 145-156.
Kramer, DA. (1995) Minerals yearbook : magnesium. U.S. Geological Survey, Geologic Division, Reston, VA.
Krauss, T; Krauss, P; Hagenmaier, H. (1994) Formation of PCDD/PCDF during composting? Chemosphere
28(1):155-158.
Kuck, PH. (1995) Minerals yearbook: nickel. U.S. Geological Survey, Geologic Division, Reston, VA.
Kwok, ESC; Atkinson, R; Arey, J. (1995) Rate constants for the gas-phase reactions of the OH radical with
dichlorobiphenyls, 1-chlorodibenzo-p-dioxin, 1,2-dimethoxybenzene, and diphenyl ether: estimation of OH radical
reaction rate constants for PCBs, PCDDs, and PCDFs. Environ Sci Technol 29(6): 1591-1598.
R-20
-------
Lahl, U. (1993) Sintering plants of steel industry: the most important thermical PCDD/CDF source in industrialized
regions. Organohalogen Compounds 11:311-314.
Lahl, U. (1994) Sintering plants of steel industry: PCDD/F emission status and perspectives. Chemosphere 29(9-
11):1939-1945.
Lahl, U; Wilken, M; Zeschmar-Lahl, B; Jager, J. (1991) PCDD/PCDF balance of different municipal waste
management methods. Chemosphere 23(8-10):1481-1489.
Lamparski, LL; Stehl, RH; Johnson, RL. (1980) Photolysis of pentachlorophenol-treated wood. Chlorinated
dibenzo-p-dioxin formation. Environ Sci Technol 14(2): 196-200.
Lamparski, LL; Nestrick, TJ; Cutie, SS. (1990) The impact on the environment of airborne paniculate matter from
the eruption of Mount Saint Helens in May 1980. In: Clement, R; Kogel, R; eds. Emissions from combustion
processes: origin, measurement, control. Chelsea, MI: Lewis Publishers.
Larsen, BR. (1995) HRGC separation of PCB congeners. JHighRes Chromatogr 18:141-151.
Larsen, B; Bewadt, S; Facchetti, S. (1992) Separation of toxic congeners from PCB mixtures on two series coupled
narrow-bore columns (50 m SIL-8 and 25 m HT-5). International Journal of Environmental Analytical Chemistry
(47): 147-166.
Larssen, S; Brevick, EM; Oehme, M. (1990) Emission factors of PCDD and PCDF for road vehicles obtained by a
tunnel experiment. Organohalogen Compounds 1:453-456.
Larsson, P; Lemkemeier, K. (1989) Microbial mineralization of chlorinated phenols and biphenyls in sediment-water
systems from humic and clear-water lake. Water Research 23(9):1081-1085.
Lasagni, M; Collina, E; Tettamanti, M; et al. (2000) Kinetics of MSWI fly ash thermal degradation, 1. Empirical rate
equation for native carbon gasification. Environ Sci Technol 34:130-136.
Launhardt, T; Thoma, H. (2000) Investigation on organic pollutants from a domestic heating system using various
solid biofuels. Chemosphere 40:1149-1157.
Lebeuf, M; Gobeil, C; Brochu, C; et al. (1996) Direct atmospheric deposition versus fluvial inputs of PCDD/Fs to
the sediments of the Lower St. Lawrence Estuary. Organohalogen Compounds 28:20-24.
Lee, RG; Hung, H; Mackay, D; et al. (1998) Measurement and modeling of the diurnal cycling of atmospheric PCBs
and PAHs. Environ Sci Technol 32(14):2172-2179.
Leifer, A; Brink, RH; Thorn, GC; et al. (1983) Environmental transport and transformation of polychlorinated
biphenyls. U.S. Environmental Protection Agency, Office of Toxic Substances, Washington, DC; EPA-56 0/5-83-
025.
Lemieux, PM. (1997) Evaluation of emissions from the open burning of household waste in barrels, U.S.
Environmental Protection Agency, National Risk Management Research Laboratory, Research Triangle Park, NC;
EPA-600/R-97-134a.
Lemieux, PM; Gullett, BK; Lutes, CC; et al. (2000) Parameters influencing emissions of PCDDs/Fs from open
burning of household waste in barrels. Presented at AWMA/Environment Canada Specialty Conference "Recent
Advances in the Science and Management of Air Toxics," April 10-12, 2000, Banff, Alberta, Canada.
Lepine, F; Milot, S; Vincent, N. (1992) Formation of toxic PCB congeners and PCB-solvent adducts in a sunlight
irradiated cyclohexane solution of Aroclor 1254. Bull Environ Contain Toxicol 48:152-156.
R-21
-------
Lewert, HV; ed. (1976) A closer look at the pesticide question for those who want the facts. Dow Chemical
Company, Midland, Michigan. Report No. 41.
Lexen, K; De Wit, C; Jansson, B; et al. (1993) Polychlorinated dibenzo-p-dioxin and dibenzofuran levels and
patterns in samples from different Swedish industries analyzed within the Swedish dioxin survey. Chemosphere
27(1-3):163-170.
Liberson, GL; Belanger, WD. (1995) An analysis of dioxin production as a function of chlorine feed. In: Kilgroe, J,
ed. Solid waste management: thermal treatment & waste-to-energy technologies. Proceedings of an interantional
specialty conference sponsored by the Air and Waste Management Association, Washington, DC, April 18-21, 1995;
pp. 425-432.
Liberti, A; Brocco, D. (1982) Formation of polychlorodibenzodioxins and polychlorodibenzofurans in urban
incinerator emissions. In: Hutzinger, O; Frei, RW; Merian, E; et al; eds. Chlorinated dioxins and related
compounds. New York, NY: Pergamon Press.
Ling, H; Diamond, M; Mackay, D. (1993) Application of the QWASI fugacity/aquivalence model to assessing
sources and fate of contaminants in Hamilton Harbour. Journal of Great Lakes Research 19(3):582-602.
Liss, PS; Slater, PG. (1974) Flux of gases across the air-sea interface. Nature 247:181-184.
Lockheed Martin Corp. (1998) Urbanized areas rainfall. EPA Contract No. 68-C5-0091.
Lockwood, LF; Ansel, G; Haddad, PO. (1981) Magnesium and magnesium alloys. In: Kirk-Othmer encyclopedia of
chemical technology. 3rded. New York, NY: John Wiley and Sons; pp. 570-615.
Lofroth, G; Zebtihr, Y. (1992) Polychlorinated dibenzo-p-dioxins (PCDD) and dibenzofurans (PCDFs) in
mainstream and sidestream cigarette smoke. Bull Environ Contam Toxicol 48:789-794.
Lorang, PA. (1996) Comments on the API document: a study to quantify on-road emissions of dioxins and furans
from mobile sources: phase 2. Memorandum dated December 1996. U.S. Environmental Protection Agency,
National Vehicle and Fuel Emissions Laboratory, Ann Arbor, MI.
Lorber, M; Braverman, C; Gehring, P; et al. (1996) Soil and air monitoring in the vicinity of a municipal solid waste
incinerator, part I: soil monitoring. Organohalogenhalogen Compounds 28:255-261.
Lorenz, W; Wichmann, H; Bahadir, M. (1996) Balancing of the release of polychlorinated dibenzo-p-dioxins and
dibenzofurans in fire accidents: a contribution to the discussion. Reinhaltung der Luft. 56:49-53.
LUA (Landesumweltamt Nordrhein-Westfalen). (1997) Identification of relevant industrial sources of dioxins and
furans in Europe. MaterialienNo. 43. North Rhine-Westphalia State Environment Agency, Essen, Germany. ISSN
0947-5206.
Luijk, R; Covers, HA; Neilssen, L. (1992) Formation of polybrominated dibenzofurans during extrusion of high
impact polystyrene/decabromodiphenyl ether/antimony (III) oxide. Environ Sci Technol 26(11):2191-2198.
Luijk, R; Akkerman, DM; Slot, P; et al. (1994) Mechanism of formation of polychlorinated dibenzo-p-dioxins and
dibenzofurans in the catalyzed combustion of carbon. Environ Sci Technol 28:312-321.
Lustenhouwer, JA; Olie, K; Hutzinger, O. (1980) Chlorinated dibenzo-p-dioxins and related compounds in
incinerator effluents: a review of measurements and mechanisms of formation. Chemosphere 9:501-522.
Luthe, C; Karidio, I; Uloth, V. (1996) Controlling dioxins emissions in coastal power boilers: a status report. Pulp
and Paper Canada 97(9):76-82.
R-22
-------
Luthe, C; Karidio, I; Uloth, V. (1998) Dioxins formation in salt-laden power boilers: a mass balance. Chemosphere
36(2):231-249.
Lykins, BW; Clark, R; Cleverly, DH. (1987) Polychlorinated dioxin and furan discharge during carbon reactivation.
Journal of Environmental Engineering 114(2): 330-316.
Malisch, R. (1994) Determination of PCDD/PCDF inPCB products and milk samples; correlation between PCB-
and PCDD/PCDF-contamination of milk samples. Organohalogen Compounds 20:209-214.
Malloy, TA; Goldfarb, TD; Surico, MTJ. (1993) PCDDs, PCDFs, PCBs, chlorophenols (CPs) and chlorobenzenes
(CBzs) in samples from various types of composting facilities in the United States. Chemosphere 27(l-3):325-334.
Maniff, K; Lewis, M. (1988) MIS A monitoring discovers dioxins and furans in wastewater stream of petroleum
refinery. Ontario Ministry of the Environment Communique, Ontario, Canada.
Mannsville. (1983) Mannsville chemical products synopsis. Cortland, NY: Mannsville Chemical Products Corp.
December 1983.
Marklund, S; Rappe, C; Tysklind, M; et al. (1987) Identification of polychlorinated dibenzofurans and dioxins in
exhausts from cars run on leaded gasoine. Chemosphere 16(l):29-36.
Marklund, S; Andersson, R; Tysklind, M; et al. (1989) Emissions of PCDDs and PCDFs from a PVC-fire in
Holmsund, Sweden. Chemosphere 18(1-6): 1031-1038.
Marklund, S; Andersson, R; Tysklind, M; et al. (1990) Emissions of PCDDs and PCDFs in gasoline and diesel
fueled cars. Chemosphere 20(5):553-561.
Masunaga, S; Takasuga, T; Nakanishi, J. (2001) Dioxin and dioxin-like PCB impurities in some Japanese
agrochemical formulations. Chemosphere 44:873-885.
Mathur, DB; Stephenson, MJ; Wenning, PJ; et al. (1997) PCDD/PCDFs in storm water outfalls adjacent to urban
areas and petroleum refineries in San Francisco Bay, California, USA. Organohalogen Compounds 32:1-5.
Matsueda, T; Kurokawa, Y; Nakamura, M; et al. (1994) Concentrations of PCDDs, PCDFs and coplanarPCBs in
cigarettes from various countries. Organohalogen Compounds 20:331-334.
McAloon, K. (2003) Memorandum: 2000 medical waste incinerator (MWI) update for sources and environmental
releases of dioxin-like compounds in the United States: year 2000 update. Versar, Inc., Springfield, VA.
McCord, AT. (1981) A study of the emission rate of volatile compounds from lagoons.
Presented at the 5th National Conference on Management of Uncontrolled Hazardous Waste Sites, Washington, DC,
October 28-30.
McCrady, JK; Maggard, SP. (1993) Uptake and photodegradation of 2,3,7,8-tetrachlorodibenzo-p-dioxin sorbed to
grass foliage. Environ Sci Technol 27:343-350.
McLachlan, M. (1996) Bioaccumulation of hydrophobic chemicals in agricultural food chains. Environ Sci Technol
30:252-259.
McLachlan, MS; Welsch-Pausch, K; Tolls, J. (1995) Field validation of a model of the uptake of gaseous SOC in
Lolium multiflorum (rye grass). Environ Sci Technol 29(8): 1998-2004.
Merk, M; Schramm, KW; Lenoir, D; et al. (1995) Determination of the PCDD/F concentration in the fumes from a
PVC fire. Organohalogen Compounds 23:491-494.
R-23
-------
Metal Producing. (1991) Sinter plant roundup. 33 Metal Producing 29(5):24.
Metal Producing. (1996) Census of the North American steel industry. 33 Metal Producing 34(3):40-52.
Mill, T; Rossi, M; McMillen, D; et al. (1987) Photolysis of tetrachlorodioxin and PCBs under atmospheric
conditions. Internal report prepared by SRI International for U.S. Environmental Protection Agency, Office of
Health and Environmental Assessment, Washington, DC.
Miller, A. (1993) Dioxin emissions from EDC/VCM plants. Environ Sci Technol 27:1014-1015.
Miller, GC; Hebert, VR; Miille, MJ; et al. (1989) Photolysis of octachlorodibenzo-p-dioxin on soils: production of
2,3,7,8-TCDD. Chemosphere 18:1265-1274.
Milligan, MS; Altwicker, ER. (1995) Mechanistic aspects of the de novo synthesis of polychlorinated dibenzo-p-
dioxins and furans in fly ash from experiments using isotopically labeled reagents. Environ Sci Technol 29:1353-
1358.
Milligan, MS; Altwicker, ER. (1996) Chlorophenol reactions on fly ash. Equilibrium surface coverage and global
kinetics. Environ Sci Technol 30:230-236.
Mitchell, D; Loader, A. (1993) Investigation of pollutant emissions from crematoria. United Kingdom. Warren
Spring Laboratory. Report No. LR908(PA).
Mitchell, W; Suggs, J. (1998) Emission factors for the disposal of energetic materials by open burning and open
detonation (OB/OD). National Exposure Research Laboratory, U.S. Environmental Protection Agency, Research
Triangle Park, NC; EPA/600/R-98/103.
Miyabara, Y; Hashimoto, S; Sagai, M; et al. (1999) PCDDs and PCDFs in vehicle exhaust particles in Japan.
Chemosphere 39:143-150.
Morimoto, K; Kenji, T. (1995) Effect of humic substances on the enzymatic formation of OCDD from PCP.
Organohalogen Compounds 23:387-392.
Muir, CG; Yarechewski, AL.; Corbet, RL; et al. (1985) Laboratory and field studies on the fate of 1,3,6,8 -
tetrachlorodibenzo-p-dioxin in soil and sediments. J Agric Food Chem 33:518-523.
Musdalslien, VI; Nokleby, PH; Wallevik, O. (1998) Formation of dioxins from carbonaceous materials in production
of anhydrous magnesium chloride. Organohalogen Compounds 36:81-84.
Muto, H; Takizawa, Y. (1989) Dioxins in cigarette smoke. Archives of Environmental Health 44(3): 171-174.
Muto, H; Takizawa, Y. (1992) Potential health risk via inhalation/ingestion exposure to polychlorinated dibenzo-p-
dioxins and dibenzofurans. Bull Environ Contam Toxicol 49:701-707.
Naf, C; Broman, D; Ishaq, R; et al. (1990) PCDDs and PCDFs in water, sludge and air samples from various levels
in a waste water treatment plant with respect to composition changes and total flux. Chemosphere 20(10-12): 1503-
1510.
Naikwadi, KP; Karasek, F W. (1989) Prevention of PCDD formation in MS W incinerators by inhibition of catalytic
activity of fly ash produced. Chemosphere 19(l-6):299-304.
Nakamura, M; Matsueda, T; Kurokawa, Y; et al. (1994) Levels and profiles of CDDs and PCDFs in soils and plants.
Organohalogen Compounds 20:103-106.
NATO (North Atlantic Treaty Organization). (1988) Pilot study on international information
R-24
-------
exchange on dioxins and related compounds: emissions of dioxins and related compounds from combustion and
incineration sources. North Atlantic Treaty Organization, Committee on the Challenges of Modern Society. Report
#172.
NCASI (National Council of the Paper Industry for Air and Stream Improvement). (1993) Summary of data
reflective of pulp and paper industry progress in reducing the TCDD/TCDF content of effluents, pulps and
wastewater treatment sludges. New York, NY.
NCASI (National Council of the Paper Industry for Air and Stream Improvement). (1995) NCASI summary of
PCDD/F emission from wood residue and black liquor combustion. Attachment 2 to comments on EPA's external
review draft Estimating Exposures to Dioxin-Like Compounds. Submitted January 13, 1995, New York, NY.
Nestrick, TJ; Lamparski, LL. (1982) Isomer-specific determination of chlorinated dioxins for assessment of
formation and potential environmental emission from wood combustion. Anal Chem 54(13):2292-2299.
Nestrick, TJ; Lamparski, LL. (1983) Assessment of chlorinated dibenzo-p-dioxin formation and potential emission to
the environment from wood combustion. Chemosphere 12(4/5):617-626.
Nestrick, TJ; Lamparski, LL, Crummett, WB. (1987) Thermolytic surface-reaction of benzene and iron (III) chloride
to form chlorinated dibenzo-p-dioxins and dibenzofurnas. Chemosphere 16(4):777-790.
Nies, L; Vogel, TM. (1990) Effects of organic substrates on dechlorination of Aroclor 1242 in anaerobic sediments.
Applied and Environmental Microbiology 56(9):2612-2617.
Nito, SN; Akimoto, Y; Imagawa,T; et al. (1997) Comparative study on formations of polychlorinated dibenzo-p-
dioxin, chlorinated dibenzofuran and related compounds by pyrolysis of some precursors on unused sand for
fluidized bed incinerator and long term used sand. Chemosphere 35:1717-1727.
NOAA (2004) Data retrieval from National Climatic Data Center. Available online at http//wwwl.ncdc.noaa.gov/
drd964x.pcp.txt.
NRC (National Research Council). (2000) Waste incineration and public health. Committee on Health Effects of
Waste Incineration, Board of Environmental Studies and Toxicology, Commission on Life Sciences. Washington
DC: National Academy Press.
NRC (National Research Council). (2002) Biosolids applied to land. Advancing standards and practices. Committee
on Toxicants and Pathogens in Biosolids Applied to Land, Board on Environmental Studies and Toxicology,
Division on Earth and Life Sciences. Washington, DC: National Academies Press.
Oberg, LG; Rappe, C. (1992) Biochemical formation of PCDD/Fs from chlorophenols. Chemosphere 25(l-2):49-
52.
Oberg, T; Warman, K; Berstrom, J. (1989) Production of chlorinated aromatics in the post-combustion zone and
boiler. Chemosphere 19(l-6):317-322.
Oberg, LG; Glas, B; Swanson, SE; et al. (1990) Peroxidase-catalyzed oxidation of chlorophenols to polychlorinated
dibenzo-p-dioxins and dibenzofurans. Arch Environ Contam Toxicol 19:930-938.
Oberg, LG; Andersson, R; Rappe, C. (1992) De novo formation of hepta- and octachlorodibenzo-p-dioxins from
pentachlorophenol in municipal sewage sludge. Organohalogen Compounds 9:351-354.
Oberg, LG; Wagman, N; Andersson, R; et al. (1993) De novo formation of PCDD/Fs in compost and sewage sludge:
a status report. Organohalogen Compounds 11:297-302.
R-25
-------
Oberg, LG, Wagman, N; Koch, M; et al. (1994) Polychlorinated dibenzo-p-dioxins, dibenzofurans and non-ortho
PCBs in household organic-waste compost and mature garden waste compost. Organohalogen Compounds 20:245-
250.
Oehme, M; Mano, S; Bjerke, B. (1989) Formation of polychlorinated dibenzofurans and dibenzo-p-dioxins by
production processes for magnesium and refined nickel. Chemosphere 18(7-8): 1379-1389.
Oehme, M; Larssen,S; Brevik, EM. (1991) Emission factors of PCDD/CDF for road vehicles obtained by a tunnel
experiment. Chemosphere 23:1699-1708.
O'Keefe, PW; Smith, RM. (1989) PCB capacitor/transformer accidents. In: Kimbrough, RD; Jensen, AA; eds.
Halogenated biphenyls, terphenyls, napthalenes, dibenzodioxins and related products. New York, NY: Elsevier
Science Publishers.
Opatick, RE. (1995) Chlorobenzene Producers Association comments on EPA's external review draft Estimating
Exposures to Dioxin-Like Compounds. Chlorobenzene Producers Association, Washington, DC. January 10.
Orazio, CE; Kapila, S; Puri, RK; et al. (1992) Persistence of chlorinated dioxins and furans in the soil environment.
Chemosphere 25(7-10): 1469-1474.
Ozvacic, V. (1985) A review of stack sampling methodology for PCDDs/PCDFs. Chemosphere 15(9-12): 1173-
1178.
Paciorek, KL; Kratzer, RH; Kaufman, J; et al. (1974) Oxidative thermal decomposition of poly (vinyl chloride)
compositions. J Appl Polym Sci 18:3723-3729.
Palmer, FH; Sapudar, RA; Heath, JA; et al. (1988) Chlorinated dibenzo-p-dioxin and dibenzofuran contamination in
California from chlorophenol wood preservative use. California State Water Resources Control Board. Report No.
88-5WQ.
Papke, O; Ball, M; Lis, ZA; et al. (1989) PCDD and PCDF in indoor air in kindergartens in Northern W. Germany.
Chemosphere 18:617-626.
Pardue, JH; Delaune, RD; Patrick, WH. (1988) Effect of sediment pH and oxidation-reduction potential on PCB
mineralization. Water, Air and Soil Pollution 37:439-447.
Paterson, S; Mackay, D; Bacci, E; et al. (1991) Correlation of the equilibrium and kinetics of leaf-air exchange of
hydrophobic organic chemicals. Environ Sci Technol 25(5):866-871.
Patterson, SH; Murray, HH. (1984) Kaolin, refractory clay, ball clay, and halloysite in North America, Hawaii, and
the Caribbean region. U.S. Geological Survey. Professional Paper No. 1306.
Paustenbach, D; Wenning, RJ; Lau, V; et al. (1992) Recent developments on the hazards posed by 2,3,7,8-TCDD in
soil: implications for setting risk-based cleanup levels at residential and industrial sites. Journal of Toxicology and
Environmental Health 36:103-149.
Paustenbach, DJ; Wenning, RJ; Mathur, D; et al. (1996) PCDD/PCDFs in urban stormwater discharged to San
Francisco Bay, California, USA. Organohalogen Compounds 28:111-116.
PCA (Portland Cement Association). (2001) U.S. and Canadian Portland Cement Industry: plant information
summary. Skokie, Illinois.
Pechan and Associates. (2002) Draft technical memorandum no. 1: Survey results for open burning activity and rule
effectiveness in the MANE-VU region. Prepared for Mid-Atlantic Regional Air Management Association.
R-26
-------
Pechan and Associates. (2004) Final technical memorandum numbers 1: A survey results for open burning activity
and rule effectiveness in the MANE-VU Region, and 2: Activity data and emission estimates for residential open
burning in the MANE-VU Region.
Persson, PE; Bergstrom, J. (1991) Emission of chlorinated dioxins from landfill fires. In: Proceedings of the
Sardinia 91: Third International Landfill Symposium. Vol. 1:1635-1645.
Peters, A. (1988) Minerals yearbook: iron and steel. U.S. Geological Survey, Geologic Division, Reston, VA.
Plimmer, JR. (1980) Herbicides. In: Kirk-Othmer encyclopedia of chemical technology. 3rd ed. Vol. 12. pp. 297-
351.
Pohlandt, K; Marutzky, R. (1994) Concentration and distribution of polychlorinated dibenzo-p-dioxins (PCDD) and
polychlorinated dibenzofurans (PCDF) in wood ash. Chemosphere 28(7): 1311-1314.
Prinn, RG; Weiss, RF; Miller, BR; et al. (1995) Atmospheric trends and lifetime of CH3CC13 and global OH
concentrations. Science 269:187-192.
Procaccini, C; Bozzelle, JW; Longwell, JP; et al. (2003) Formation of chlorinated aromatics by reactions of Cl*, C12
and Hcl with benzene in the cool-down zone of a combuster. Environ Sci Technol 37:1684-1689.
Puri, RK; Clevenger, TE; Kapila, S; et al. (1989) Studies of parameters affecting translocation of tetrachlorodibenzo-
p-dioxin in soil. Chemosphere 18(1-6):1291-1296.
Quab, U; Fermann, M. (1997) Identification of relevant industrial sources of dioxins and furans in Europe. North
Rhine-Westphalia State Environment Agency. Report No. 43.
Quensen, JF; Tiedje, JM; Boyd, SA. (1988) Reductive dechlorination of polychlorinated biphenyls by anaerobic
microorganisms from sediment. Science 242:752-754.
Quensen, JF; Boyd, SA; Tiedje, JM. (1990) Dechlorination of four commercial polychlorinated biphenyl mixtures
(Aroclors) by anaerobic microorganisms from sediments. Appl Environ Microbiol 56(l):2360-2369.
Radian Corporation. (1991) Results of dioxin testing on the catalytic reformer unit #1 exhaust Texas Refinery;
Bakersfield, California. Sacramento, CA.
Raghunathan, K; Gullett, BK (1996) Role of sulfur in reducing PCDD and PCDF formation. Environ Sci Technol
30(6): 1827-1834.
Randall, D. (1995) Memorandum dated April 8, 1995, from David Randall and Brian Hardee, Midwest Research
Institute, Gary, North Carolina, to Richard Copland, Combustion Group, Office of Air Quality Planning and
Standards, U.S. Environmental Protection Agency, Research Triangle Park, NC. Subject: Emission factors for
medical waste incinerators (MWIs), EPA Contract No. 68-01-0115. EPA Docket Number A-91-61; IV B 042.
Rappe, C. (1991) Sources of human exposure to CDDs and PCDFs. In: Gallo, M; Scheuplein, R; van der Heiden, K;
eds. Biological basis for risk assessment of dioxin and related compounds. Plainview, NY: Cold Spring Harbor
Laboratory Press. Banbury Report No. 35.
Rappe, C. (1992a) Sources of exposure, environmental levels and exposure assessment of PCDDs and PCDFs.
Organohalogen Compounds 9:5-8.
Rappe, C. (1992b) Sources of PCDDs and PCDFs. Introduction. Reactions, levels, patterns, profiles and trends.
Chemosphere 25(l-2):41-44.
R-27
-------
Rappe, C. (1993) Sources of exposure, environmental concentrations and exposure assessment of PCDDs and
PCDFs. Chemosphere 27(l-3):211-225.
Rappe, C. (1995) Comments on EPA's external review draft Estimating Exposures to Dioxin-Like Compounds.
Comments submitted to EPA by the Integrated Waste Services Association; Washington, DC.
Rappe, C; Marklund, S; Buser, HR; et al. (1978a) Formation of polychlorinated dibenzo-p-dioxins (PCDDs) and
dibenzofurans (PCDFs) by burning or heating chlorophenates. Chemosphere 3:269-281.
Rappe, C; Gara, A; Buser, HR. (1978b) Identification of polychlorinated dibenzofurans (PCDFs) in commercial
chlorophenol formulations. Chemosphere 12:981-991.
Rappe, C; Kjeller, LO; Bruckmann, P; et al. (1988) Identification and quantification of PCDD/CDFs in urban air.
Chemosphere 17(1):3-20.
Rappe, C; Kjeller, LO; Andersson, R. (1989a) Analyses of PCDDs and PCDFs in sludge and water samples.
Chemosphere 19(1-6): 13-20.
Rappe, C; Swanson, S; Gals, B; et al. (1989b) Formation of PCDDs and PCDFs by the chlorination of water.
Chemosphere 19(12): 1875-1880.
Rappe, C; Glas, B; Kjeller, LO; et al. (1990a) Levels of PCDDs and PCDFs in products and effluent from the
Swedish pulp and paper industry and chloralkali process. Chemosphere 20(10-12): 1701-1706.
Rappe, C; Andersson, R; Lundstrom, K; et al. (1990b) Levels of polychlorinated dioxins and dibenzofurans in
commercial detergents and related products. Chemosphere 21(l-2):43-50.
Rappe, C; Kjeller, LO; Kulp, SE. (1990c) Sampling and analysis of PCDDs and PCDFs in surface water and
drinking water at 0.001 ppq levels. Organohalogen Compounds 2:207-210.
Rappe, C; Kjeller, L; Kulp, S; et al. (1991) Levels, profile and pattern of PCDDs and PCDFs in samples related to
the production and use of chlorine. Chemosphere 23(11-12): 1629-1636.
Rappe, C; Anderson, R; Karlaganis, G; et al. (1994) PCDDs and PCDFs in samples of sewage sludge from various
areas in Switzerland. Organohalogen Compounds 20:79-84.
Rappe, C; Anderson, R; Bonner, M; et al. (1998) PCDDs and PCDFs in municipal sewage sludge and effluent from
POTW in the State of Mississippi, USA. Chemosphere 36:315-328.
Ree, KC; Evers, EHG; Van Der Berg, M. (1988) Mechanism of formation of polychlorinated dibenzo(p)dioxins
(PCDDs) and polychlorinated dibenzofurans (PCDFs) from potential industrial sources. Toxicological and
Environmental Chemistry 17:171-195.
Remmers, J; Dupuy, A; McDaniel, D; et al. (1992) Polychlorinated dibenzo-p-dioxin and dibenzofuran
contamination in chloranil and carbazole violet. Chemosphere 25(7-10): 1505-1508.
Rhee, GY; Bush, B; Brown, MP; et al. (1989) Anaerobic biodegradation of polychlorinated biphenyls in Hudson
River sediments and dredged sediments in clay encapsulation. Water Research 23(8):957-964.
Riederer, M. (1990) Estimating partitioning and transport of organic chemicals in the foliage/atmosphere system:
discussion of a fugacity-based model. Environ Sci Technol 24(6):829-837.
Rieger, R; Ballschmiter, K. (1992) Search for sources of CLxDD/CLxDF in sewage sludge of mixed
industrial/domestic origin. Organohalogen Compounds 9:203-206.
R-28
-------
Riggs, KB; Brown, TD; Schrock, ME. (1995) PCDD/PCDF emissions from coal-fired power plants. Organohalogen
Compounds 24:51-54.
Rigo, H.G. (1998) Dioxin and chlorine: new evidence of no material relationship in a commercial scale system.
Organohalogen Compounds 36:261-264.
Rigo, GH; Chandler, JA; Larder, SW. (1995) The relationship between chlorine and dioxin emissions from waste
combustors. In: Kilgroe, J; ed. Solid waste management: thermal treatment & waste-to-energy technologies.
Proceedings of an international specialty conference sponsored by the Air and Waste Management Association,
Washington, DC. April 18-21, 1995. ASME/CRTD. Vol. 36.
RIPA (Reusable Industrial Packaging Association). (1997) 1997 Statistics as found on the Reusable Industrial
Packaging Association (RIPA) webpage. Available online at http://www.reusablepackaging.org/Stats.html.
Risatti, JB. (1992) Rates of microbial dechlorination of polychlorinated biphenyls (PCBs) in anaerobic sediments
from Waukegan Harbor. The Illinois Hazardous Waste Research and Information Center, Champaign, IL. HWRIC
RR-061.
Retard, W. (1993) Fires. In: Current views on the impact of dioxins and furans on human health and the
environment. Toxicology Forum, Inc., Washington, DC.
Rubber Manufacturers Association. (2002) Statistical report on scrap tire markets. Rubber Manufacturers
Association, Washington, DC. Available online at http://www.rma.org/scrap_tires/.
Rubin, A; White, C. (1992) Computer printout from National Sewage Sludge Survey Data
Base. U.S. Environmental Protection Agency, Office of Science and Technology, Health and Ecological Criteria
Division, Sludge Risk Assessment Branch. December 21, 1992.
Ruokojarvi, P; Ettala, M; Rahkonen, P; et al. (1995) Polychlorinated dibenzo-p-dioxins and -furans (PCDDs and
PCDFs) in municipal waste landfill fires. Chemosphere 30(9): 1697-1708.
Safe, S. (1994) Polychlorinated biphenyls (PCBs): environmental impact, biochemical and toxic responses, and
implications for risk assessment. Crit Rev Toxicol 21(l):51-88.
SAIC (Science Applications International Corporation). (1994) Identification of sources of dioxin in catalytic
reformers and potential mitigation. U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory,
Cincinnati, OH. EPA Contract No. 68-CO-0044.
Sakai, S; Hiraoka, M; Takeda, N; et al. (1993a) Coplanar PCBs and PCDD/PCDFs in municipal waste incineration.
Chemosphere 27(l-3):233-240.
Sakai, S; Hiraoka, M; Takeda, N; et al. (1993b) On emission and formation of non-ortho CBs and mono-ortho CBs
in municipal waste incineration. Organohalogen Compounds 11:245-248.
Sakai, S; Hiraoka, M; Takeda, N; et al. (1994) Behavior of coplanar PCBs in oxidative condition of municipal waste
incineration. Organohalogen Compounds 20:309-313.
Sakai, S; Hayakawa, K; Takatsuki, H; et al. (2001) Dioxin-like PCBs released from waste incineration and their
deposition flux. Environ Sci Technol 35:3601-3607.
Santl, H; Gruber, L; Stohrer, E. (1994) Some new sources of polychlorinated dibenzodioxins (PCDDs) and
dibenzofurans (PCDFs) in waste papers and recycled pulps. Chemosphere 29(9-ll):1995-2003.
Schafer, K; McLachlan, MS; Reissinger, M; et al. (1993) An investigation of PCDD/F in a composting operation.
Organohalogen Compounds 11:425-428.
R-29
-------
Schatowitz, B; Brandt, G; Gafner, F; et al. (1993) Dioxin emissions from wood combustion. Organohalogen
Compounds 11:307-310.
Schaub, WM; Tsang, W. (1983) Physical and chemical properties of dioxins in relation to their disposal. In: Tucker,
RE; Young, AL, Gray, A.P; eds. Human and environmental risks of chlorinated dioxins and related compounds.
New York, NY: Plenum Press.
Schecter, A; Papke, O; Isaac, J; et al. (1997) 2,3,7,8 Chlorine substituted dioxin and dibenzofuran congeners in 2,4-
D, 2,4,5-T and pentachlorophenol. Organohalogen Compounds 32:51-55.
Schoonenboom, MH; Tromp, PC; Olie, K. (1993) The formation of coplanar PCBs, PCDDs and PCDFs in a fly-ash
model system. Organohalogen Compounds 11:277-280.
Schramm, K; Reischl, A; Hutzinger, O. (1987) UNITTree: A multimedia compartment model to estimate the fate of
lipophilic compounds in plants. Chemosphere 16:2653-2663.
Schreiner, M; Reifenhauser, W; Vierle, O. (1992) Measurements of the emission concentrations of PCDD/PCDF in a
digester-torch of a municipal sewage plant. Organohalogen Compunds 9:255-257.
Schreiner, G; Schimmel, H; Ballschmiter, K. (1995) Influence of the substitution pattern on the microbial
degradation of mono- to tetrachlorinated dibenzofurans and dibenzodioxins. Organohalogen Compounds 24:43-46.
Schulz, DE; Petrick, G; Duinker, JC. (1989) Complete characterization of polychlorinated biphenyl congeners in
commercial Aroclor and Clophen mixtures by multidimensional gas chromatography-electron capture detection.
Environ Sci Technol 23:852-859.
Schwartz, TR; Tillitt, DE; Feltz, KP; et al. (1993) Determination of mono- and non~o-o'-chlorine substituted
polychlorinated biphenyls in Aroclors and environmental samples. Chemosphere 26(8): 1443-1460.
Schwind, K-H; Thoma, H; Hutzinger, O; et al. (1991) Emission halogenierter dibenzodioxine (PXDD) und
dibenzofurane (PXDF) aus verbrennungsmotoren [Emission of halogenated dibenzo-p-dioxins (PCDD) and
dibenzofurans(PCDF) from combustion engines]. UWSF-Z. Umweltchem. Oekotox. 3, 291-298. [English
translation]
Schwind, KH; Hosseinpour, J; Fiedler, H; et al. (1995) Human exposure to consumer products: analysis of
chlorinated organic compounds in candles and their exhaust fumes. Organohalogen Compounds 26:113-116.
Secor International, Inc. (1995a) Emission measurement results for furans and dioxins from the main stack at the
Phelps Dodge-Hidalgo Smelter, Playos, New Mexico. Prepared for the Phelps Dodge Mining Company by Secor
International, Inc., Redmond, WA.
Secor International, Inc. (1995b) Furans and dioxins emission measurement results for Cyprus Miami Mining
Corporation. Prepared for the Cypress Amax Minerals Company by Secor International, Inc., Redmond, WA.
Sedlak, DL; Andren, AW. (1991) Aqueous-phase oxidation of polychlorinated biphenyls by hydroxyl radicals.
Environ Sci Technol 25(8):1419-1427.
Sewart, A; Harrad, SJ; McLachlan, MS; et al. (1995) PCDD/Fs and non-o-PCBs in digested U.K. sewage sludges.
Chemosphere 30(l):51-67.
Shane, BS; Henry, CB; Hotchkiss, JH; et al. (1990) Organic toxicants and mutagens in ashes from eighteen
municipal refuse incinerators. Arch Environ Contam Toxicol 19:665-673.
Shaub, WM; Tsang, W. (1983) Dioxin formation in incinerators. Envrion Sci Technol 17:721-730.
R-30
-------
Shibata, E; Yamamoto, S; Kasai, E; Nakamura, E. (2003) Formation behavior of PCDD/Fs in PVC pyrolysis with
copper oxide. Chemosphere 50:1235-1242.
Shin, K-J; Chang, Y-S. (1999) Characterization of polychlorinated dibenzo-p-dioxins, dibenzofurans, biphenyls, and
heavy metals in fly ash produced from Korean municipal solid waste incinerators. Chemosphere 38:2655-2666.
Sidhu, S. (1999) Role of aliphatic precursors in the formation mechanism of PCDD/F, PCB, and PCN.
Organohalogen Compounds 41:59-62.
Silk, PJ; Lonergan, GC; Arsenault, TL; et al. (1997) Evidence of natural organochlorine formation in peat bogs.
Chemosphere 35:2865-2880.
Simonich, SL; Hites, RA. (1994) Vegetation-atmosphere partitioning of polycyclic aromatic hydrocarbons. Environ
SciTechnol28(5):939-943.
Sivils, LD; Kapila, S; Yan, Q; et al. (1994) Studies on gas-phase phototransformation of polychlorinated dibenzo-p-
dioxins. Organohalogen Compounds 19:349-353.
Sivils, LD; Kapila, S; Yan, Q; et al. (1995) Studies on vapor phase phototransformation of polychlorinated dibenzo-
p-dioxins (PCDDs): effect of environmental parameters. Organohalogen Compounds 24:167-172.
Smith, RM; O'Keefe, PW; Hilker, D; et al. (1995) Direct and indirect contribution of atmospheric PCDDs and
PCDFs to Hudson River National Estuarine Research Reserve sediment cores. Organohalogen Compounds 24:141-
145.
Solid Waste Technologies. (1994) A comprehensive status report onMSW waste-to-energy facilities, incinerators
refuse-derived-fuel, and tire processing facilities throughout North America. Kansas City, MO: HCI Publications.
Someshwar, AV; Pinkerton, JE. (1992) Wood processing industry. In: Buonicore, AJ; Davis, WT; eds. Air
pollution engineering manual. Air and Waste Management Association. New York, NY: Van Nostrand Reinhold. pp.
835-847.
St. Lawrence County. (1993) Legal open burning and on-premises burial of solid waste in St. Lawrence County, NY:
The environmental impacts. St. Lawrence County Planning Office, Canton, NY.
Stanmore, BR. (2004) The formation of dioxins in combustion systems. Combustion and Flame 136:398-427.
Stehl, RH; Lamparski, LL. (1977) Combustion of several 2,4,5-trichlorophenoxy compounds: formation of 2,3,7,8-
tetrachlorodibenzo-p-dioxin. Science 197:1008-1009.
Stehl, RH; Papenfuss, RR; Bredeweg, RA; et al. (1973) The stability of pentachlorophenol and chlorinated dioxins
to sunlight, heat, and combustion. In: Blair, EH; ed. Chlorodioxins: origin and fate. American Chemical Society,
Washington, DC.
Stevens, JL; Jones, KC. (2003) Quantification of PCDD/F concentrations in animal manure and comparison of the
effects of the application of cattle manure and sewage sludge to agricultural land on human exposure to PCDD/F.
Chemosphere 50:1183-1191.
Stieglitz, L; Zwick, G; Beck, J; et al. (1989) Carbonaceous particles in fly ash: a source for the de novo synthesis of
organochlorocompounds. Chemosphere 19(l-6):283-290.
Stieglitz, L; Vogg, H; Zwick, G; etal. (1991) On formation conditions of Organohalogen compounds from paniculate
carbon of fly ash. Chemosphere 23(8-10): 1255-1264.
R-31
-------
Storherr, RW; Watts, RR; Gardner, AM; et al. (1971) Steam distillation technique for the analysis of 2,3,7,8-
tetrachlorodibenzo-p-dioxin in technical 2,4,5-T. Journal of the AOAC 54(1):218-219.
Stowe, RA. (1996) A needle in a haystack. Environmental Protection. August 1996. pp. 22-25.
Strachan, WMJ; Eisenreich, SJ. (1988) Mass balancing of toxic chemicals in the Great Lakes: the role of
atmospheric deposition. International Joint Commission, Windsor, Ontario.
Strandell, ME; Lexen, KM; deWit, CA; et al. (1994) The Swedish dioxin survey: summary of results from PCDD/F
and coplanar PCB analyses in source-related samples. Organohalogen Compounds 20:363-366.
Stringer, RL; Costner, P; Johnston, PA. (1995) PVC manufacture as a source of PCCD/Fs. Organohalogen
Compounds 24:119-123.
Strong, B; Hanks, K. (1999) Memorandum "Emissions inventory for hospital, medical, infectious waste incinerators
covered by the proposed Section 1 l(d)/129 Federal Plan." EPA Docket number A-98-24, item II-B-1.
Stuart, B; Carter, D; Hoodless, RA. (1993) Report on the examination of sewage sludges for polychlorinated
dibenzo-p-dioxins and polychlorinated dibenzofurans. Laboratory of the Government Chemist, London, England.
Svenson, A; Kjeller, LO; Rappe, C. (1989) Enzyme-mediated formation of 2,3,7,8-tetrasubstituted chlorinated
dibenzodioxins and dibenzofurans. Environ Sci Technol 23(7):900-902.
Sverdrup Corp. (1991) Chemetco, Inc., compliance testing, final report. Sverdrup Corp., St. Louis, MO.
Swackhamer, DL; Armstrong, DE. (1986) Estimation of the atmospheric and nonatmospheric contributions and
losses of polychlorinated biphenyls for Lake Michigan on the basis of sediment records of remote lakes. Environ Sci
Technol 20(9): 879-883.
Swackhamer, DL; McVeety, BD; Hites, RA. (1988) Deposition and evaporation of polychlorobiphenyl congeners to
and from Siskiwit Lake, Isle Royale, Lake Superior. Environ Sci Technol 22(6):664-672.
Sweden's Office of Nature Conservancy. (1991) Organochlorine compounds from dishwashing detergents and
bleach? Sweden Office of Nature Conservancy, Solna, Sweden. Report No. 4009.
Takeda, N; Takaoka, M; Fujiwara, T; et al. (1998) Research on PCDDs/DFs emissions from crematories in Japan.
Organohalogen Compounds 36:167-170.
Takeda, N; Takaoka, M; Fujiwara, T; et al. (2001) Measures to prevent emissions of PCDDs/DFs and coplanar
PCBs from crematories in Japan. Chemosphere 43:763-771.
Takizawa, Y; Asada, S; Muto, H. (1994) Dioxins in dust fall and volcanic ash from the active volcanos Fugendake
and Sakurajima. Organohalogen Compounds 20:359-362.
Tanabe, S. (1988) PCB problems in the future: foresight from current knowledge. Environ Pollut 50:5-28.
Tanabe, S; Kannan, N; Wakimoto, T; et al. (1987) Method for the determination of three toxic non-orthochlorine
substituted coplanar PCBs in environmental samples at part-per-trillion levels. Intern J Environ Anal Chem 29:199-
213.
Tashiro, C; Clement, RE; Stocks, BJ; et al. (1990) Preliminary report: dioxins and furans in prescribed burns.
Chemosphere 20:1533-1536.
Taylor, A; Zannes, M. (1996) The 1996 IWSA municipal waste combustion directory of United States facilities.
Integrated Waste Services Association, Washington, DC.
R-32
-------
The Aluminum Association, Inc. (1997) U.S. aluminum industry plant directory, November 1997. The Aluminum
Association, Washington, DC.
The Tobacco Institute. (1995) Tobacco industry profile 1995. The Tobacco Institute, Washington, DC.
The Vinyl Institute. (1998) Dioxin characterization program. Phase I Report. The Vinyl Institute, Morristown, NJ.
Thiesen, J; Funcke, W; Balfanz, E; et al. (1989) Determination of PCDFs and PCDDs in fire accidents and
laboratory combustion tests involving PVC-containing materials. Chemosphere 19(l-6):423-428.
Thoma, H; Hutzinger, O. (1989) Content and formation of toxic products in flame retardants. Presented at the
Workshop on Brominated Aromatic Flame Retardants. Slokloster, Sweden; October 24-26, 1989. Sponsored by the
Swedish Environmental Protection Agency.
Thomas, VM; Spiro, TG. (1995) An estimation of dioxin emissions in the United States. Toxicological and
Environmental Chemistry 50:1-37.
Thomas, VM; Spiro, TG. (1996) The U.S. dioxin inventory: are there missing sources? Environ Sci Technol
30(2):82A-85A.
Thompson, TS; Clement, RE; Thornton, N; et al. (1990) Formation and emission of PCDDs/PCDFs in the petroleum
refining industry. Chemosphere 20(10-12): 1525-1532.
Thub, U; Popp, P; Ehrlich, C; et al. (1995) Domestic lignite combustion as source of polychlorodibenzodioxins and
furans (PCDD/F). Chemosphere 31(2):2591-2604.
Tolls, J; McLachlan, MS. (1994) Partitioning of semivolatile organic compounds between air and Lolium
multiflorum (Welsh rye grass). Environ Sci Technol 28(1): 159-166.
Tosine, H; Clement, R; Osvacic, V; et al. (1983) Levels of chlorinated organics in a municipal incinerator. Presented
at the 186th National Meeting of the American Chemical Society; August 28-September 2, Washington, DC.
September, 1, 1983.
Towara, J; Hiller, B; Hutzinger, O; et al. (1992) PCDD/F in distillation residues from dry cleaning machines.
Chemosphere 25:1509.
Trapp, S; Matthies, M; Scheunert, I.; et al. (1990) Modeling the bioconcentration of organic chemicals in plants.
Environ Sci Technol 24:1246-1252.
TRRCPO (Two Rivers Region Council of Public Officials and Patrick Engineering, Inc.) (1994) Emission
characteristics of burn barrels. Prepared for U.S. Environmental Protection Agency, Region V, Chicago, IL.
Tysklind, M; Rappe, C. (1991) Photolytic transformation of polychlorinated dioxins and dibenzofurans in fly ash.
Chemosphere 23(8-10): 1365-1375.
Tysklind, M; Soderstrom; Rappe, C; et al. (1989) PCDD and PCDF emissions from scrap metal melting processes at
a steel mill. Chemosphere 19(1-6):705-710.
Tysklind, M; Carey, AE; Rappe, C; et al. (1992) Photolysis of OCDF and OCDD on soil. Organohalogen
Compounds 8:293-296.
U.K. Department of the Environment. (1995). Comments on EPA's external review draft Estimating Exposures to
Dioxin-Like Compounds. Submitted by the Department of the Environment, Food, and Rural Affaires, London,
England.
R-33
-------
Umlauf, G; Horstmann, M; Klein, P; et al. (1993) Mass balance of PCDD/F in a dry cleaning machine.
Organohalogen Compounds 11:413-416.
Umweltbundesamt. (1996) Determination of requirements to limit emissions of dioxins and furans. Report from the
Working Group of the Subcommittee Air/Technology of the Federal Government/Federal States Emission Control
Committee. Berlin, Germany.
Urhahn, T; Ballschmiter, K. (1998) Chemistry of the biosynthesis of halogenated methanes: Cl-organohalogens as
pre-industrial chemical stressors in the environment. Chemosphere 37:1017-1032.
U.S. Census Bureau. (2000) Census 2000. Washington, DC.
U.S. Census Bureau. (2001) Statistical abstract of the United States: 2001. Washington, DC.
USDA (U.S. Department of Agriculture). (1970) Restricting the use of phenoxy herbicides: costs to farmers.
Economic Research Service and Agricultural Research Service, Washington, DC. Agricultural Economic Report
No. 194.
USDA (U.S. Department of Agriculture). (1971) Farmer's use of pesticides in 1971 and quantities. Economics
Research Service, Washington, DC. Agricultural Economic Report No. 252.
USDA (U.S. Department of Agriculture). (1995) Graphic highlights of natural resource trends in the United States
between 1982 and 1992. U.S. Department of Agriculture, Natural Resources Conservation Service, National
Resources Inventory, Washington, DC.
USDA (U.S. Department of Agriculture). (1997) Tobacco leaf and products statistics and analysis. U.S. Department
of Agriculture, Economic Research Service, Washington, DC.
USDA (U.S. Department of Agriculture). (2003) National Resources Inventory 2001 Annual Report. Natural
Resources Conservation Service, U.S. Department of Agriculture, Washington, DC.
U.S. Department of Health and Human Services. (1989) Toxicology and carcinogenesis studies of two
pentachlorophenol technical-grade mixtures in B6C3F1 mice (feed studies). U.S. Department of Health and Human
Services, National Toxicology Program, Research Triangle Park, NC. NIH Publication No. 89-2804.
U.S. DOC (U.S. Department of Commerce). (1990a) 1987 Census of manufactures. Smelting and refining of
nonferrous metals and alloys. U.S. Bureau of the Census, Washington, DC. Report No. 33C-18.
U.S. DOC (U.S. Department of Commerce). (1990b) 1987 Census of manufacturers. Soap, cleaners, and toilet
goods. U.S. Bureau of the Census, Washington, DC. Report No. MC-87-I-28D.
U.S. DOC (U.S. Department of Commerce). (1990c) 1987 Census of manufactures. Industrial inorganic chemicals.
Bureau of the Census, Washington, DC. Report No. MC87-I-28A.
U.S. DOC (U.S. Department of Commerce). (1995a) 1992 Census of transportation. Truck inventory and use
survey. U.S. Bureau of the Census, Washington, DC. Report No. TC92-T-52.
U.S. DOC (U.S. Department of Commerce). (1995b) Statistical abstract of the United States, 1995. 118th ed.
Washington, DC.
U.S. DOC (U.S. Department of Commerce). (1995c) 1992 Census of manufactures. Smelting and refining of
nonferrous metals and alloys. U.S. Bureau of the Census, Washington, DC. Report No. MC92-I-33C.
U.S. DOC (U.S. Department of Commerce). (1996) 1992 Economic census report series, disc 1G. U.S. Bureau of
the Census, Washington, DC. Report No. CD-EC92-1G.
R-34
-------
U.S. DOC (U.S. Department of Commerce). (1997) Statistical abstract of the United States, 1997. 117th ed. U.S.
Bureau of the Census, Washington, DC.
U.S. DOC (U.S. Department of Commerce). (2002) International database. U.S. Bureau of the Census. Available
online at http://blue.census.gov/cgi-bin/ipc/idbagg.
U.S. DOC (U.S. Department of Commerce). (2003) Census 2000 Summary File 1 (SF 1) 100-Percent Data. U.S.
Bureau of the Census, Washington DC.. Available online at
http://factfinder.census.gov/servlet/DTTable?_ts=64261331968.
U.S. DOT (Department of Transportation) (2002) 2002 Status of the nation's highways, bridges, and transit:
conditions and performance report to Congress. Federal Highway Administration, Federal Transit Administration,
Washington, DC.
U.S. EPA (Environmental Protection Agency). (1975) Production, distribution, use and environmental impact
potential of selected pesticides. Office of Pesticide Programs and Office of Water and Hazardous Materials,
Washington, DC.
U.S. EPA (Environmental Protection Agency). (1977) Pesticide usage survey of agricultural, governmental, and
industrial sectors in the United States, 1974. Office of Pesticide Programs, Washington, DC; EPA-540/9-78-007.
U.S. EPA (Environmental Protection Agency). (1980) Materials balance for chlorophenols. Level I-preliminary.
Office of Toxic Substances, Washington, DC; EPA-560/13-80-004.
U.S. EPA (Environmental Protection Agency). (1982a) Development document for effluent limitations guidelines
and standards for the iron and steel manufacturing point source category. Vol. II. Office of Water, Washington, DC;
EPA-440/1-82/024.
U.S. EPA (Environmental Protection Agency). (1982b) Development document for effluent limitations guidelines
and standards for the inorganic chemicals manufacturing point source category. Office of Water Regulations and
Standards, Washington, DC; EPA 440/1-82/007.
U.S. EPA (Environmental Protection Agency). (1984) Thermal degradation products from dielectric fluids. Office
of Toxic Substances, Washington, DC; EPA-560/5-84-009.
U.S. EPA (Environmental Protection Agency). (1987a) National dioxin study Tier 4: combustion sources.
Engineering analysis report. Office of Air Quality Planning and Standards, Research Triangle Park, NC; EPA-
450/4-84-014h.
U.S. EPA (Environmental Protection Agency). (1987b) Municipal waste combustion study. Characterization of the
municipal waste combustion industry. Office of Air and Radiation, Washington, DC; EPA/530-SW-87-021b.
U.S. EPA (Environmental Protection Agency). (1987c) Hospital waste combustion study: Data gathering phase.
Office of Air Quality Planning and Standards, Research Triangle Park, NC; EPA-450/3-88-017.
U.S. EPA (Environmental Protection Agency). (1987d) Locating and estimating air emissions from sources of
polychlorinated biphenyls (PCB). Office of Air Quality Planning and Standards, Research Triangle Park, NC; EPA-
450/4-84-007n.
U.S. EPA (Environmental Protection Agency). (1988a) Estimating exposure to 2,3,7,8-TCDD. Office of Health and
Environmental Assessment, Washington, DC; EPA/600/6-88/005A.
U.S. EPA (Environmental Protection Agency). (1988b) U.S. EPA cooperative dioxin screening study. Office of
Water Regulations and Standards, Washington, DC; EPA-440/1-88-025.
R-35
-------
U.S. EPA (Environmental Protection Agency). (1988c) Pesticide assessment guidelines, subdivision D: product
chemistry. Series 61-4, addendum 1 on data reporting. Office of Pesticide Programs, Washington, DC. Available
from the National Technical Information Service, Springfield, VA; PB88-191705.
U.S. EPA (Environmental Protection Agency). (1988d) Pesticide industry sales and usage: 1987 market estimates.
Office of Pesticide Programs, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1989) Interim procedures for estimating risks associated with
exposures to mixtures of chlorinated dibenzo-p-dioxins and -dibenzofurans (CDDs and CDFs) and 1989 update.
Risk Assessment Forum, Washington, DC; EPA/625/3-89/016.
U.S. EPA (Environmental Protection Agency). (1990a) Characterization of municipal waste combustion ash, ash
extracts, and leachates. Office of Solid Waste and Emergency Response, Washington, DC; EPA 530-SW-90-029A.
U.S. EPA (Environmental Protection Agency). (1990b) Operation and maintenance of hospital medical waste
incinerators. Office of Air Quality Planning and Standards, Research Triangle Park, NC; EPA/625/6-89/024.
U.S. EPA (Environmental Protection Agency). (1990c) Pesticides industry sales and usage: 1988 market estimates.
Office of Pesticides and Toxic Substances, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1990d) Emissions of metals, chromium and nickel species, and
organics from municipal wastewater sludge incinerators. Office of Research and Development, Risk Reduction
Engineering Laboratory, Cincinnati, OH. EPA Contract No. 68-CO-0027, Work Assignment No. 0-5.
U.S. EPA (Environmental Protection Agency). (1990e) Summary report on U SEP A/industry cooperative dioxin
study. "The 104 Mill Study." Office of Water Regulations and Standards, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1991a) Methodology for assessing environmental releases of and
exposure to municipal solid waste combustor residuals. Office of Research and Development, Washington, DC;
EPA/600/8-91/031.
U.S. EPA (Environmental Protection Agency). (1991b) Medical and institutional waste incineration: regulations,
management, technology, emissions, and operations. Center for Environmental Research Information, Cincinnati,
OH; EPA/625/4-91/030.
U.S. EPA (Environmental Protection Agency). (1991c) Workshop report on toxicity equivalency factors for
polychlorinated biphenyls congeners. Risk Assessment Forum, Washington, DC; EPA/625/3-91/020.
U.S. EPA (Environmental Protection Agency). (1991d) Pesticides industry sales and usage: 1989 market estimates.
Office of Pesticides and Toxic Substances, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1992a) Economic impact and preliminary regulatory impact analysis
for proposed MACT-based emission standards and guidelines for municipal waste combustors. Office of Air Quality
Planning and Standards, Research Triangle Park, NC; EPA-450/3-91-029.
U.S. EPA (Environmental Protection Agency). (1992b) Markets for scrap tires. Office of Solid Waste and
Emergency Response, Washington, DC; EPA/530-SW-90-074A.
U.S. EPA (Environmental Protection Agency). (1992c) Secondary lead smelter emission test report, East Penn
Manufacturing, Lyon Station, Pennsylvania. Office of Air Quality Planning and Standards, Emission Measurement
Branch, Research Triangle Park, NC. EMB Report No. 92-SLS-3.
U.S. EPA (Environmental Protection Agency). (1992d) Industry agrees to switch to low-dioxin chloranil from
contaminated chloranil. Chemicals in Progress Bulletin 12(2):23.
R-36
-------
U.S. EPA (Environmental Protection Agency). (1992e) Pesticides industry sales and usage: 1990 and 1991 market
estimates. Office of Prevention, Pesticides and Toxic Substances, Washington, DC; 733-K-92-001.
U.S. EPA (Environmental Protection Agency). (1993a) Report to Congress on cement kiln dust. Office of Solid
Waste, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1993b) Development document for proposed effluent limitations
guidelines and standards for the pulp, paper and paperboard point source category. Office of Water, Washington,
DC;EPA-821-R-93-019.
U.S. EPA (Environmental Protection Agency). (1993c) Economic analysis of impacts of integrated air/water
regulations for the pulp and paper industry on disposal of wastewater sludge. Office of Pollution Prevention and
Toxics, Regulatory Impacts Branch, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1993d) EPA publishes proposed rule for chloranil. Chemicals in
Progress Bulletin 14(2):26.
U.S. EPA (Environmental Protection Agency). (1993e) Press advisory: EPA terminates special review of non-wood
pesticide uses of pentachlorophenol. Office of Pesticide Programs, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1993f) 1991 Toxics release inventory: public data release. Office
of Pollution Prevention and Toxics, Washington, DC. EPA-745-R-93-003.
U.S. EPA (Environmental Protection Agency). (1993g) Interim report on data and methods for assessment of
2,3,7,8-tetrachlorodibenzo-p-dioxin risks to aquatic life and associated wildlife. Office of Research and
Development, Washington, DC; EPA/600/R-93/055.
U.S. EPA (Environmental Protection Agency). (1993h) Wildlife exposure factors. Vol. I. Office of Health and
Environmental Assessment, Office of Research and Development, Washington, DC; EPA/600/R-93/187a.
U.S. EPA (Environmental Protection Agency). (1994a) Combustion emissions technical resource document
(CETRED). Draft Report. Office of Solid Waste and Emergency Response, Washington, DC; EPA/530-R-94-014.
U.S. EPA (Environmental Protection Agency). (1994b) Preliminary source assessment for the secondary copper
smelting industry. Draft report. Office of Air Quality Planning and Standards, Research Triangle Park, NC. EPA
Contract No. 68-D1-118, Work Assignment Nos. 25, 40, and 46.
U.S. EPA (Environmental Protection Agency). (1994c) Secondary lead smelting industry: background information
document for proposed standards. Final report. Office of Air Quality Planning and Standards, Research Triangle
Park, NC; EPA-450/R-94-024.
U.S. EPA (Environmental Protection Agency). (1994d) Pesticide industry sales and usage: 1992 and 1993 market
estimates. Office of Prevention, Pesticides and Toxic Substances, Washington, DC; EPA 733-K-94-001.
U.S. EPA (Environmental Protection Agency). (1995a) Compilation of air pollutant emission factors. 5th ed.
Office of Air Quality Planning and Standards, Research Triangle Park, NC.
U.S. EPA (Environmental Protection Agency). (1995b) Locating and estimating air emissions from sources of
dioxins and furans. Draft final report. Office of Air Quality Planning and Standards, Research Triangle Park, NC.
Contract No. 68-D2-0160.
U.S. EPA (Environmental Protection Agency). (1995c) Secondary aluminum plant emission test report: Rochester
Aluminum Smelting Corporation. Office of Air Quality Planning and Standards, Research Triangle Park, NC. EMC
Report 95-SAL-01.
R-37
-------
U.S. EPA (Environmental Protection Agency). (1995d) Final emission test report, HAP emission testing on selected
sources at a secondary lead smelter, Schuylkill Metals Corporation, Forest City, Missouri. Office of Air Quality
Planning and Standards, Emission Measurement Branch, Research Triangle Park, NC. EMB Report No. 93-SLS-2.
U.S. EPA (Environmental Protection Agency). (1995e) Final emission test report, HAP emission testing on selected
sources at a secondary lead smelter, Tejas Resources, Inc., Terrell, Texas. Office of Air Quality Planning and
Standards, Emission Measurement Branch, Research Triangle Park, NC. EMB Report No. 93-SLS-l.
U.S. EPA (Environmental Protection Agency). (1995f) SRRD's dioxin/furan status report. Office of Pesticide
Programs, Special Review and Reregistration Division, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1995g) An SAB report: A second look at dioxin. Review of the
Office of Research and Development's reassessment of dioxin and dioxin-like compounds by the Dioxin
Reassessment Review Committee. Science Advisory Board, Washington, DC; EPA-SAB-EC-95-021.
U.S. EPA (Environmental Protection Agency). (1995h) 1993 Toxics release inventory: public data release. Office of
Pollution Prevention and Toxics, Washington, DC; EPA-745-R-95-010.
U.S. EPA (Environmental Protection Agency). (1996a) Draft technical support document for hazardous waste
combustor MACT standards. Volume I: Description of source categories. Office of Solid Waste and Emergency
Response, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1996b) EPA OSW hazardous waste combustion data base. Office of
Solid Waste, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1996c) EIIP. Vol. II, Chapter 3, Preferred and alternative methods
for estimating air emissions from hot-mix asphalt plants. Office of Air Quality Planning and Standards, Research
Triangle Park, NC.
U.S. EPA (Environmental Protection Agency). (1996d) Municipal solid waste factbook, version 3.0. Office of Solid
Waste, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1996e) Percentile estimates used to develop the list of pollutants for
round two of the Part 503 Regulation. Office of Science and Technology, Washington, DC. EPA Contract No. 68-
C4-0046.
U.S. EPA (Environmental Protection Agency). (1996f) PCBs: Cancer dose-response assessment and application to
environmental mixtures. Office of Research and Development, Washington, DC; EPA/600/P-96/001.
U.S. EPA (Environmental Protection Agency). (1996g) Emissions inventory of Section 112(c)(6) pollutants:
polycyclic organic matter (POM), 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)/2,3,7,8-tetrachlorodibenzofuran
(TCDF), polychlorinated biphenyl compounds (PCBs), hexachlorobenzene, mercury, and alkylated lead. Office of
Air Quality Planning and Standards, Research Triangle Park, NC.
U.S. EPA (Environmental Protection Agency). (1997a) Locating and estimating air emissions from sources of
dioxins and furans. Office of Air Quality Planning and Standards, Research Triangle Park, NC. DCN No. 95-298-
130-54-01.
U.S. EPA (Environmental Protection Agency). (1997b) 1996 Clean Water Needs Survey Report to Congress. Office
of Water, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1997c) Draft technical support document for hazardous waste
combustors MACT standards (NODA). Vol. I. MACT evaluations based on revised database. Office of Solid
Waste and Emergency Response, Washington, DC.
R-38
-------
U.S. EPA (Environmental Protection Agency). (1997d) Supplemental technical development document for effluent
limitations guidelines and standards for the pulp, paper, and paperboard category. Office of Water, Washington,
DC;EPA-821-R-97-011.
U.S. EPA (Environmental Protection Agency). (1997e) Pesticide industry sales and usage: 1994 and 1995 market
estimates. Office of Prevention, Pesticides and Toxic Substances, Washington, DC; EPA-733-R-97-002.
U.S. EPA (Environmental Protection Agency). (1998a) Inventory of sources of dioxin in the United States.
Workshop Review Draft. National Center for Environmental Assessment, Office of Research and Development,
Washington, DC; EPA/600/P-98/002Aa.
U.S. EPA (Environmental Protection Agency). (1998b) 1996 Toxics release inventory-public data release. Office
of Pollution Prevention and Toxics, Washington, DC; EPA 745-R-98-005.
U.S. EPA (Environmental Protection Agency). (1998c) AP-42: compilation of air pollutant emissions factors.
Volume I: stationary point and area sources. Supplement D to the 5th edition. Office of Air Quality Planning and
Standards, Research Triangle Park, NC.
U.S. EPA (Environmental Protection Agency). (1999a) Emission test evaluation of a crematory at Woodlawn
Cemetery in the Bronx, NY. Office of Air Quality Planning and Standards, Washington, DC; EPA-454/R-99-049
U.S. EPA (Environmental Protection Agency). (1999b) Biosolids generation, use, and disposal in the United States.
Office of Solid Waste and Emergency Response, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1999c) Technical background document for the Report to Congress
on remaining wastes from fossil fuel combustion: waste characterization. Office of Solid Waste and Emergency
Response, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1999d) Technical background document for the Report to Congress
on remaining wastes from fuel combustion: industry statistics and waste management practices. Office of Solid
Waste and Emergency Response, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1999e) EPA OSWER hazardous waste combustor emissions data
base. Office of Solid Waste and Emergency Response, Washington, DC.
U.S. EPA (Environmental Protection Agency). (1999f) Monthly reporting of CDD/CDF content in technical
pentachlorophenol manufactured by KMG-Bermuth, Inc. and Vulcan Chemicals, 1987 to 1999. Office of Pesticide
Programs, Washington, DC.
U.S. EPA (Environmental Protection Agency). (2000a) Emission test evaluation of a pathological waste incinerator
at the University of Georgia. Office of Air Quality Planning and Standards, Research Triangle Park, NC; EPA-
456/R-00-004.
U.S. EPA (Environmental Protection Agency). (2000b) Source characterization for sewage sludge incinerators.
Office of Air Quality Planning and Standards, Washington, DC; EPA-454/R-00-038a.
U.S. EPA (Environmental Protection Agency). (2000c) Exposure and human health reassessment of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Part 1: Estimating exposure to dioxin-like
compounds. National Center for Environmental Assessment, Office of Research and Development, Washington,
DC; EPA/600/P-00/001B(b-d).
U.S. EPA (Environmental Protection Agency). (200la) Documentation report for national and county-level emission
estimates for residential wood combustion. Prepared by Eastern Research Group, Inc., for Emission Inventory and
Analysis Group, Office of Air Quality Planning and Standards, Research Triangle Park, NC.
R-39
-------
U.S. EPA (Environmental Protection Agency). (200Ib) Municipal solid waste in the United States: 1999 facts and
figures. Office of Solid Waste and Emergency Response, Washington, DC; EPA530-R-01-014.
U.S. EPA (Environmental Protection Agency). (200Ic) Final titanium dioxide listing background document for the
inorganic chemical listing determination. Office of Solid Waste, Washington, DC.
U.S. EPA (Environmental Protection Agency). (2001d) Environmental assessment of proposed revisions to the
national pollutant discharge elimination system regulation and the effluent guidelines for concentrated animal
feeding operations. Office of Water Regulations and Standards, Washington DC; EPA-821-B-01-001.
U.S. EPA (Environmental Protection Agency). (2002a) Hazardous waste MACT database. Office of Solid Waste.
Available at: http://www.epa.gov/epaoswer/hazwaste/combust/ph2nodal/page2-3.htnrfWords. Accessed February
25, 2003.
U.S. EPA (Environmental Protection Agency). (2002b) Documentation for version 2 of the 1999 NEI for criteria air
pollutants, area sources. Office of Air Quality Planning and Standards, Emission Factor and Inventory Group,
Research Triangle Park, NC.
U.S. EPA (Environmental Protection Agency). (2002c) Municipal solid waste in the United States: 2000 facts and
figures. Office of Solid Waste and Emergency Response, Washington, DC; EPA530-R-02-001.
U.S. EPA (Environmental Protection Agency). (2002d) Statistical support document for the development of round 2
biosolids use or disposal regulations. Office of Science and Technology, Washington, DC; EPA-822-R-02-034.
U.S. EPA (Environmental Protection Agency). (2003a) EPA TRI Explorer (Ver 4.01). Office of Environmental
Information. Available online at http://www.epa.gov/triexplorer.
U.S. EPA (Environmental Protection Agency). (2003b) Documentation of emission estimation methods for year
2000 and 2001 mobile and nonpoint source dioxin inventories. Prepared for the Office of Air Quality Planning and
Standards, Washington, DC, by E.H. Pechan & Associates, Inc.
U.S. EPA (Environmental Protection Agency). (2003c) Chemical information collection and data development
(testing) web site. Office of Prevention, Pesticides and Toxic Substances, Washington, DC. Available online at
http ://www. epa. gov/opptintr/chemtest/polyha.pdf.
U.S. EPA (Environmental Protection Agency). (2004) Clean water needs survey 2000, discharge report. Office of
Water, Office of Wastewater Management, Washington, DC. Available online at
http ://cfpub. epa.gov/cwns/rpt_discharge 1_00. cfm.
USGS (U.S. Geological Survey). (1997a) Statistical compendium: copper. U.S. Department of the Interior, U.S.
Geological Survey, Washington, DC.
USGS (U.S. Geological Survey). (1997b) Statistical compendium: aluminum. U.S. Geological Survey, Geologic
Division, Reston, VA.
USGS (U.S. Geological Survey). (1997c) Mineral commodities summary: aluminum. U.S. Geological Survey,
Geologic Division Reston, VA.
USGS (U.S. Geological Survey). (1997d) Statistical compendium: titanium. U.S. Geological Survey, Geologic
Division, Reston, VA.
USGS (U.S. Geological Survey). (1997e) Minerals yearbook: recycling: nonferrous metals. U.S Geological
Survey, Geologic Division, Reston, VA.
R-40
-------
USGS (U.S. Geological Survey). (1997f) Mineral commodity survey: iron and steel scrap. U.S. Department of the
Interior, U.S. Geological Survey, Washington, DC.
USGS (U.S. Geological Survey). (2000) Mineral commodity summaries for 2000: iron and steel. U.S. Geological
Survey, Reston, VA.
USGS (U.S. Geological Survey). (2001) Mineral commodity summaries for 2001: iron and steel. U.S. Geological
Survey, Reston, VA.
USGS (U.S. Geological Survey). (2002) Mineral commodity summaries for 2002. U.S. Geological Survey,
Geologic Division, Reston, VA.
U.S. ITC (International Trade Commission). (1946 through 1994) Synthetic organic chemicals: United States
production and sales. U.S. International Trade Commission, Washington, DC.
U.S. ITC (International Trade Commission). (1984) Imports of benzenoid chemicals and products 1983;
investigation no. 332-135. U.S. International Trade Commission, Washington, DC. USITC Publication No. 1548.
Van den Berg, M; Birnbaum, L; Bosveld, ATC; et al. (1998) Toxic equivalency factors (TEFs) for PCBs, PCDDs,
PCDFs for humans and wildlife. Environ Health Perspect 106:775-792.
VanDort, HM; Bedard, DL. (1991) Reductive ortho and meta dechlorination of a polychlorinated biphenyl congener
by anaerobic microorganisms. Appl Environ Microbiol 57:1576-1578.
Van Oostdam, JC; Ward, JEH. (1995) Dioxins and furans in the British Columbia environment. BC Environment,
Environmental Protection Department, Victoria, British Columbia.
Van Wijnen, JH; Liem, AKD; Olie, K; et al. (1992) Soil contamination with PCDDs and PCDFs of small (illegal)
scrap wire and scrap car incineration sites. Chemosphere 24(2): 127-134.
Versar, Inc. (1976) PCBs in the United States: industrial use and environmental distribution. Prepared for the U.S.
Environmental Protection Agency, Office of Toxic Substances, Washington, DC; EPA Contract No. 68-01-3259,
Task I.
Versar, Inc. (1985) List of chemicals contaminated or precursors to contamination with incidentally generated
polychlorinated and polybrominated dibenzodioxins and dibenzofurans. Prepared for the U.S. Environmental
Protection Agency; EPA Contract No. 68-02-3968, Task No. 48.
Versar, Inc. (1988) PCB exposure assessment of scrap/salvage facilities. Prepared for the U.S. Environmental
Protection Agency, Office of Toxic Substances, Washington, DC; EPA Contract No. 68-02-4254, Task No. 136.
Vikelsoe, J; Madsen, H; Hansen, K. (1993) Emission of dioxins from Danish wood-stoves. Organohalogen
Compounds 11:405-408.
Vickelsoe, J; Lassen, P; Johansen, E; et al. (1994) Formation of PCDD/PCDFs via an enzymatically mediated
chlorination of humic acids. Organohalogen Compounds 20:351-354.
Villanueva, EC; Jennings, RW; Burse, VW; et al. (1974) Evidence of chlorodibenzo-p-dioxin and
chlorodibenzofuran in hexachlorobenzene. J Agr Food Chem 22(5):916-917.
Viraraghavan, T; Slough, K. (1999) Sorptionof pentachlorophenol on peat-bentonite mixtures. Chemosphere
39:1487-1496.
Virta, RL. (2000) Clays. U.S. Geological Survey mineral commodity summaries. February 2000. Available online
at http://www.usgs.gov/.
R-41
-------
Vogg, H; Metzger, M; Steiglitz, L. (1987) Recent findings on the formation and decomposition of PCDD/PCDF in
municipal solid waste incineration. Waste Management and Research 5(3):285-294.
Vogg, H; Nunsinger, H, Merz, A; et al. (1992) Influencing the production of dioxin/ruran in solid waste incineration
plants by measures affecting the combustion as well as the flue gas cleaning systems. Chemosphere 25(1-2): 149-
152.
Vollmuth, S; Zajc, A; Niessner, R. (1994) Formation of polychlorinated dibenzo-p-dioxins and polychlorinated
dibenzorurans during the photolysis of pentachlorophenol-containing water. Environ Sci Technol 28(6):1145-1149.
Waddell, DS; Boneck-Ociesa, H; Gobran, T; et al. (1995) PCDD/PCDF formation by UV photolysis of
pentachlorophenol with and without the addition of hydrogen peroxide. Organohalogen Compounds 23:407-412.
Wagner, JC; Green, AES. (1993) Correlation of chlorinated organic compound emissions from incineration with
chlorinated organic input. Chemosphere 26(11):2039-2054.
Wagner, HC; Schramm, KW; Hutzinger, O. (1990) Biogenic polychlorinated dioxin and furan from trichlorophenol.
Organohalogen Compounds 3:453-456.
Walsh, PJJ; Brimblecombe, P; Creaser, CS; et al. (1994) Biomass burning and polychlorinated dibenzo-p-dioxins &
furans in soil. Organohalogen Compounds (20):283-287.
Wan, MT. (1995) Treated wood as a source of dioxin/furan releases. Organohalogen Compounds 24:109-114.
Wan, MT; Van Oostdam, J. (1995) Utility and railway rights-of-way contaminants: dioxins and furans. J Environ
Qual 24:257-265.
Wang, LC; Lee, WJ; Lee, WS; et al. (2003) Effect of chlorine content in feeding wastes of incineration on the
emission of polychlorinated dibenzo-p-dioxins/dibenzofurans. The Science of the Total Environment 302:185-198.
Ward, DE; McMahon, CK; Johansen, RW. (1976) An update on paniculate emissions from forest fires. Presented at
69th Annual Meeting of the Air Pollution Control Association. Portland, OR; June 27-July 1.
Ward, DE; Peterson, J; Hao, W.M. (1993) An inventory of paniculate matter and air toxic emissions from prescribed
fires in the USA for 1989. Presented at 86th Annual Meeting & Exhibition, Air & Waste Management Association;
Denver, Colorado; June.
Washington State Department of Ecology. (1998) Washington State dioxin source assessment. Washington State
Department of Ecology, Olympia, Washington. Publication No. 98-320.
Watanabe, I; Tatsukawa, R. (1987) Formation of brominated dibenzorurans from the photolysis
of flame retardant decabromobiphenyl ether in hexane solution by UV and sun light. Bull Environ Contam Toxicol
39:953-959.
Water Pollution Control Federation. (1990) Operation of municipal wastewater treatment plants. Manual of practice
No. 11. Water Pollution Control Federation, Alexandria, VA.
Weber, R; Hagenmaier, H. (1999) Mechanism of formation of polychlorinated dibenzo-p-dioxins and dibenzorurans
from chlorophenols in gas phase reactions. Chemosphere 38:529-549.
Weber, H; Disse, G; Hamann, R; et al. (1995) Influence of the aerobic and anaerobic digestion on the levels of
chlorinated dibenzofurans and dibenzodioxins in sewage sludge. Organohalogen Compounds 24:81-85.
Weber, P; Altwicker, E; Dinjus, E; et al. (1999) The role of aliphatic organic chlorine in the formation of
PCDD/PCDF on fly ash in comparison to inorganic chlorine. Organohalogen Compounds 41:11-14.
R-42
-------
Welsch-Pausch, K; McLachlan, MS. (1995) Photodegradation of PCDD/Fs on pasture grass. Organohalogen
Compounds 24:509-512.
Welsch-Pausch, K; McLachlan, MS; Umlauf, G. (1995) Determination of the principal pathways of polychlorinated
dibenzo-p-dioxins and dibenzofurans to Lolium multiflorum (Welsh Ray Grass). Environ Sci Technol 29:1090-
1098.
Western Environmental Services and Testing, Inc. (2000) Magnesium Corporation of America, Rowley, Utah.
Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans. Melt reactor and cathode stacks. File
Number 00-1043.
Weston, R. (1996). ItemII-D-82 in USEPA secondary aluminum MACT air docket number A-92-61: scrap dryer
and reverberatory furnace emission test report, Wabash, Indiana. Prepared for Wabash
Alloys, Work Order No. 11590-001-001.
Wevers, M; De Fre, R. (1995) Estimated evolution of dioxin emissions in Belgium from 1988 to 1995.
Organohalogen Compounds 24:105-108.
Wevers, M; De Fre, R; Rymen, T. (1992) Dioxins and dibenzofurans in tunnel air. Vol. 9 (sources of exposure) of
extended abstracts. Organohalogen Compounds 9:321-324.
Whitman, WG. (1927) The two-film theory of gas absorption. Chemical and Metallurgical Engineering 29(4): 146-
148.
WHO (World Health Organization). (1991) IARC monographs on the evaluation of carcinogenic risks to humans:
occupational exposures in insecticide application, and some pesticides. World Health Organization, International
Agency for Research on Cancer, Geneva, Switzerland.
Wichmann, H; Zelinski, V; Scholz-Bottcher, B; et al. (1993) Sampling strategy to investigate the distribution
behavior of low volatile pollutants like "dioxins" during vehicle fires in traffic tunnels. Chemosphere 26(6): 1159-
1166.
Wichmann, H; Lorenz, W; Bahadir, M. (1995) Release of PCDD/F and PAH during vehicle fires in traffic tunnels.
Chemosphere 31(2):2755-2766.
Wienecke, J; Kruse, H; Wassermann, O. (1992) Organic compounds in the flue gas from a power station with
circulatory fluid bed combustion of coal. Chemosphere 25(12): 1889-1895.
Wikstrom, E; Lofvenius, G; Rappe, C; et al. (1996) Influence of level and form of chlorine on the formation of
chlorinated dioxins, dibenzofurans, and benzenes during combustion of an artificial fuel in a laboratory reactor.
Environ Sci Technol 30:1637-1644.
Wikstrom, E; Ryan, S; Touati, A; et al. (2003) Importance of chlorine speciation on de Novo formation of
polychlorinated dibenzo-p-dioxins and polychlorinated dibanzofurans. Environ Sci Technol 37:1108-1113.
Wikstrom, E; Persson, A; Marklund, S. (1998) Secondary formation of PCDDs, PCDFs, PCBs, PCBzs, PHPhs and
PAHs during MSW combustion. Organohalogen Compounds 36:65-68.
Wikstrom, E; Ryan, S; Touati, A; Telfer; M; et al. (2003) Importance of chlorine speciation on de Novo formation of
polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans. Environ Sci Technol 37:1108-1113.
Williams, CH; Prescott, CL; Stewart, PB; et al. (1985) Formation of polychlorinated dibenzofurans and other
potentially toxic chlorinated pyrolysis products in PCB fires. In: Keith, LH; Rappe, C; Choudhary, G; eds.
Chlorinated dioxins and dibenzofurans in the total environment II. Boston, MA: Butterworth Publishers.
R-43
-------
Williams, DT; LeBel, GL; Benoit, FM. (1992) Polychlorodibenzodioxins and polychlorodibenzofurans in dioxazine
dyes and pigments. Chemosphere 24(2): 169-180.
Wittsiepe, J; Kullmann, Y; Schrey, P; et al. (1998) Myeloperoxidase: catalyzed formation of PCDD/F from
chlorophenols. Organohalogen Compounds 36:351-356.
Woolson, EA; Thomas, RF; Ensor, PDJ. (1972) Survey of polychlorodibenzo-p-dioxin content in selected pesticides.
J AgrFood Chem20(2):351-354.
Young, AL. (1983) Long-term studies on the persistence and movement of TCDD in a natural ecosystem. In:
Tucker, RE; Young, AL; Gray, AP; eds. Human and environmental risks of chlorinated dibenzodioxins and related
compounds. New York, NY: Plenum Press.
Zenith Research Group, Inc. (2000) Summary report on burn barrels. Prepared for the Western Lake Superior
Sanitary District, Duluth, MN.
Zenith Research Group, Inc. (2001) A survey of residents of Northeast Minnesota & Northwest Wisconsin.
Prepared for the Western Lake Superior Sanitation District, Duluth, MN.
Zheng, M-H; Liu, P-Y; Xu, X-B. (1999) Mechanism of formation of polychlorinated biphenyls from the pyrolysis of
chlorobenzenes. Organohalogen Compounds 41:15-18.
R-44
------- |