A Primary Contributor to the U.S. Dioxin Burden
Pat Costner with
Charlie Cray, Gail Martin, Bonnie Rice, David Santillo and Ruth Stringer
also with contributions by Paul Johnston and by Allan Vincent, whose
technical and practical contributions were particularly crucial. published in
conjunction with the Greenpeace International Science Unit
Greenpeace launched a broad investigation of the PVC industry in 1994, specifically targeting U.S. chemical companies that manufacture the key chemicals for PVC production: ethylene dichloride (EDC) is converted into vinyl chloride monomer (VCM) which is then polymerized to form polyvinyl chloride, also known PVC or vinyl. Collecting and analyzing samples of process wastes and other materials, Greenpeace confirmed the following facts:
U.S. EDC/VCM facilities are creating large quantities of dioxin and PCBs. Waste samples from these facilities contained some of the highest dioxin concentrations ever reported in chemical processing wastes.
U.S. EDC/VCM facilities are releasing dioxin into the surrounding environment. A sediment sample taken downstream from one facility's wastewater discharge carried an extraordinarily high dioxin concentration; a sediment sample taken downstream from another facility, although not analyzed for dioxin, contained a chemical that has been reported to be an indicator for dioxin. No air, effluent or sludge samples were analyzed; however, dioxin has been identified in all of these media in industry and government studies of European EDC/VCM facilities, as well as in PVC itself.
Industry documents obtained by Greenpeace confirm that PVC production is inextricably linked to the formation and release of dioxin and PCBs, as follows.
- Dioxin is the unavoidable by-product of PVC manufacture, specifically including the oxychlorination process, which is crucial to EDC production; and
- PCBs are also unavoidable by-products of PVC production processes.
U.S. government documents reveal USEPA's responses to the dioxin/PVC connection as follows:
- USEPA first learned that dioxin is generated in PVC manufacture during the Reagan/Bush administrations. Then, as now, the Agency had the power to require PVC manufacturers to prevent the generation and release of dioxin but did not do so.
- After proposing to regulate dioxin in one PVC-related waste in 1988, USEPA conceded to industry pressure and, in 1990, deleted dioxin from the list of chemicals to be regulated; and
USEPA's documentation of the dioxin/PVC connection from the 1980s has not been acknowledged during the Agency's dioxin reassessment.
PVC plastic is the largest single use of chlorine in the U.S., accounting for about 34 percent of all chlorine production. PVC is produced by combining chlorine and ethylene gases (or ethylene, oxygen, and hydrochloric acid in a process called oxychlorination) to produce the intermediate EDC, which is then converted to VCM. VCM is polymerized to form PVC.
Eleven U.S. companies with fifteen facilities produce EDC. The combined VCM capacity of U.S. companies is almost fourteen billion pounds per year. In 1994, total PVC production in the U.S. was estimated at 10.88 billion pounds. A large body of evidence suggests that the greatest share of the nation's dioxin burden stems from the manufacture, use, recycling, and disposal of this enormous quantity of PVC plastic.
GREENPEACE'S SAMPLING AND ANALYSIS
After making several requests to USEPA to investigate the generation and release of dioxin during PVC manufacture, Greenpeace launched its own independent study in the summer of 1994. Greenpeace representatives gained access to areas for production, storage, treatment and disposal of wastes at twelve facilities in Louisiana and Texas where EDC and/or VCM are produced.
Fifty-one grab samples were retrieved from these facilities, 25 of which are addressed in this report. Samples were taken from containers bearing labels with USEPA waste codes and other content descriptions, from sediments in receiving streams for wastewater discharges, etc. All samples were shipped to an analytical laboratory, where chemical analyses were conducted for organic and metallic contaminants. Due to the extraordinary costs entailed, only four of these samples were selected for dioxin analysis and two for PCB analysis.
Concentrations of dioxin in the three process waste samples were extraordinarily high:
Vulcan Chemicals, Geismar, Louisiana: A sample of heavy ends1 from the distillation of EDC contained dioxin at a total concentration of 200,750 parts per billion (ppb);
- Formosa Plastics, Point Comfort, Texas: A sample of heavy ends from the distillation of VCM contained 761 ppb total dioxin;
- Georgia Gulf, Plaquemine, Louisiana: A waste sample, collected from a tank containing F024 waste, had a total dioxin content of 1,248 ppb.
In comparison, wastes from the manufacture of Agent Orange, which are regulated by USEPA as "7023 and F020 dioxin-listed wastes," contain estimated total dioxin concentrations ranging from 33 to 238 ppb and 24,000 to 50,000,000 ppb, respectively. Dioxin concentrations found in the PVC-related wastes sampled by Greenpeace exceed those of Agent Orange waste F023 and, in one case, fall in the mid-range of Agent Orange waste F020.
1 Heavy ends from distillation are the higher boiling residues remaining after the chemical mixture generated by a specific process or sequence of processes is heated to evaporate the desired products, such as EDC or VCM, which are then captured by condensation.
Twenty-five samples from nine EDC/VCM facilities contained one or more of the following chemicals that have been reported to signal the presence of dioxin: hexachlorobenzene, 1,1,2,3,4,4- hexachloro-1,3- butadiene, tetrachlorobenzene, pentachlorobenzene, and 1,1,3,4-tetrachloro-1,3-butadiene. These nine facilities include the three listed above, plus the following:
Geon Vinyl (formerly BFGoodrich), LaPorte, Texas; Borden Chemical, Geismar, Louisiana; Dow Chemical, Freeport/Oyster Creek, Texas; Occidental Chemicals, Ingleside, Texas; * PPG Industries, Lake Charles, Louisiana; and Vista Chemical, Lake Charles, Louisiana.
These findings suggest that the 21 samples - 20 waste samples and one sediment sample - that were not directly analyzed for dioxin content also contained significant quantities of dioxin.
In the fourth dioxin analysis, a sediment sample taken slightly downstream from the discharge point of the Geon Corporation (formerly BFGoodrich) in LaPorte, Texas, was found to cant' a total dioxin concentration of >2,911 parts per trillion (ppt). This dioxin concentration is approximately five times higher than the average concentration reported for North American sediments in USEPA's draft dioxin reassessment. Also, dioxin-indicator compounds were found in the sediment downstream from Dow Chemical's EDC/VCM facility in Oyster Creek, Texas.
Extrapolations from industry reports and academic studies of European EDC/VCM production suggest that the quantity of dioxin discharged into U.S. waterways from EDC/VCM facilities may rival that discharged from all 104 U.S. pulp and paper mills.
USEPA BOWS TO INDUSTRY PRESSURE
USEPA's failure, during the Reagan and Bush administrations, to investigate and curtail the generation and release of dioxin and PCBs during PVC production was due neither to ignorance nor to negligence.
The Agency has known since at least 1979 that very large quantities of PCB-contaminated wastes are produced in the production of EDC and VCM. Rather than using its power under the Toxic Substances Control Act (TSCA) to prohibit such PCB manufacture, USEPA actively solicited EDC/VCM producers to apply for special exemptions.
USEPA knew prior to 1988 that dioxin is produced during EDC/VCM production. However, after initially proposing in 1988 to regulate dioxin in one EDC/VCM waste, the Agency came under pressure from the Vinyl Institute and major producers. In 1990, USEPA deleted those portions of the regulations addressing dioxin, citing the potential costs to the industry. None of this information was included in USEPA's draft dioxin reassessment but was not.
Since 1989, the published scientific and governmental literature has contained numerous reports of dioxin in the wastes, discharges and surroundings of EDC/VCM facilities in Europe. Recent Swedish studies have detected dioxin and PCBs in the PVC product itself.
The inventory of dioxin sources in USEPA's draft dioxin reassessment serves as the basis for the Agency's dioxin elimination policies. However, the Agency has not included EDC/VCM facilities among its major dioxin sources. USEPA does not provide its own estimates of dioxin releases from U.S. EDC/VCM facilities but asserts that "monitoring efforts to collect these data are highly recommended." This situation suggests that USEPA does not yet acknowledge the inseparability of PVC, dioxin and PCBs, and the enormity of PVC's contribution to the national dioxin burden both during its manufacture as well as in the burning of associated wastes and discarded PVC products.
PVC: THE COMMON CONTRIBUTING FACTOR AMONG LARGEST DIOXIN SOURCES
Without chlorine, dioxin cannot be created. PVC is the primary donor of chlorine for most of the dominant dioxin sources identified by USEPA - various kinds of incinerators and other thermal processes. Considering what is already known about the dioxin output from various portions of the PVC lifecycle, this plastic must be regarded as the largest contributor to the nation's dioxin burden.
- PVC accounts for the majority of dioxins emitted by incinerators for medical and municipal wastes, the two largest dioxin sources identified by EPA. Disposable PVC products are responsible for the vast majority of the chlorine fed to these facilities and are thus the primary source of dioxin emissions.
- Dioxin output from the burning of large quantities of chlorine-rich EDC/VCM process wastes in on-site and commercial hazardous waste incinerators is almost entirely unquantified. It is, however, considerably greater than the estimate given in USEPA's dioxin reassessment. Existing dioxin emission data were obtained almost exclusively during trial bums using the USEPA-approved stack testing method, which achieved only 26 percent recovery of dioxin spiked into the sample collection system during its evaluation with a full-size incinerator. Moreover, these data do not reflect the higher dioxin emissions associated with incinerator upsets and by-passing of pollution control systems, both of which have been reported to occur with considerable frequency. In addition, Greenpeace analyses of EDC/VCM wastes found high concentrations of metals, including copper, which catalyzes the formation of dioxin during burning in incinerators and other combustion systems.
- PVC is the primary chlorine donor in recycling facilities for copper, steel, lead and other metals. These secondary smelting facilities - which receive PVC as residues on automobiles, cables, electronic equipment, and batteries - have been identified by USEPA as major dioxin sources.
- PVC residues on scrap wood are the primary cause of dioxin emissions from domestic and industrial wood burning, according to European studies. EPA has identified these sectors as major dioxin sources.
- PVC burning in accidental home and building fires appears to be one of the largest single sources of dioxin. PVC is now ubiquitous in modem buildings, and high concentrations of dioxin have been found in the residues from accidental fires in which uncontrolled PVC-burning occurred in homes, schools, office buildings, and industrial facilities. The total contribution of PVC fires to the national dioxin burden has not been estimated, but given the hundreds of thousands of fires that occur each year, the dioxin loadings are certain to be significant.
TIME FOR ACTION ON PVC AND DIOXIN
According to USEPA's dioxin reassessment, dioxin is extraordinarily toxic, persistent, and bioaccumulative. Dioxin is now distributed globally in the environment, food chain, and human tissues, and EPA has found that current "background" levels of dioxin are already at or near the range at which health effects are known to occur. These findings provide added weight to recommendations such as the International Joint Commission call for a phaseout of chlorine use in products such as PVC in order to eliminate on a rapid timetable the generation and release of dioxin into the environment.
PVC is the largest single contributor to the national dioxin burden. For the protection of public health and the environment, a PVC phase-out must be a priority in the national dioxin prevention program. Important first steps include:
- Prohibition of the oxychlorination process for the production of EDC and other chemicals;
- Prohibition of new facilities or capacity expansions for the production of EDC/VCM and PVC;
- Modification of existing permits for EDC/VCM plants to bring generation and releases of dioxin to zero;
- Classification of relevant wastes from EDC/VCM production as dioxin-listed and PCB-containing wastes, subject to all appropriate regulatory requirements that have been revised to reflect the inadequacies and limitations of incineration as a treatment technology;
- Prohibition of incineration of chlorine-rich wastes and/or wastes from EDC/VCM production that contain dioxin and PCBs;
- Rapid phase-out of PVC uses associated with the largest dioxin releases, including short-life PVC uses (packaging and disposable products sent to incinerators or otherwise burned), uses in areas susceptible to fire (construction, appliances and automobiles), and products recycled in smelters (cables and cars); and
- A longer-term phase-out of other uses of PVC, with priorities established according to environmental impact and the availability of alternatives.
Because the ultimate phase-out of PVC will have economic impacts in the communities where manufacturing facilities are located, transition planning processes must be an integral component of any phase-out plan. This process must be guided by participation from labor, community and other stakeholders and should seek to minimize the economic effects of the transition and insure that costs and benefits are equitably
distributed. For instance, the Oil, Chemical, and Atomic Workers Union has proposed a tax on chlorine and related chemicals; the revenue would be used to encourage reinvestment in affected communities and to provide income protection, continued health care, and meaningful opportunities for higher education and reemployment for workers and their families.
TABLE OF CONTENTS
1.1 EDC/VCM Production Processes
1.1.1 EDC Via Direct Chlorination: Opportunities for Dioxin Formation
1.1.2 EDC Via Oxychlorination: The Unavoidability of Dioxin Formation
1.1.3 EDC Pyrolysis to VCM: Opportunities for Dioxin Formation
1.2 EDC/VCM Wastes
1.2.1 Vent Gases
1.2.2 Organic Liquid Wastes
1.2.3 Aqueous Wastes
1.2.4 Solid Wastes
1.3 Dioxin and PCBs in PVC Plastic
2.1 USEPA's BDAT Standards
2.1.1 USEPA's Characterization of K019 and K020 Wastes
2.1.2 Greenpeace Characterization of K019 Wastes
2.1.3 Greenpeace Characterization of Probable K020 Wastes
2.1.4 USEPA's Regulation of K019 and K020 Wastes
2.1.5 F024 Wastes: USEPA Bows to Industry Pleas to Ignore Dioxin
2.1.6 Greenpeace Characterization of F024-Containing Wastes
2.1.7 Greenpeace Characterization of K016 Waste from EDC/VCM Facilities
2.1.8 Greenpeace Characterization of Other Wastes From EDC/VCM Facilities
2.1.9 Greenpeace Characterization of Sediment Samples Collected Downstream from EDC/VCM Facilities
2.2 Dioxin and PCBs in Other USEPA-Coded Hazardous Wastes From Organochlorine Production
3.1 Dioxin Releases from Metallurgical Industries
3.2 Dioxin Releases from Municipal and Medical Waste Combustors
3.3 Dioxin Releases from Miscellaneous Combustion Facilities and Devices
3.4 Dioxin Releases from Uncontrolled Fires
3.4.1 Accidental Fires at Facilities Manufacturing or Processing PVC and/or PVC Products
3.4.2 Accidental Fires in Commercial Buildings, Homes and Stockpiled Materials
APPENDICES (not included)
Appendix 1 Vista Chemical Waste characteristics data sheet
Appendix 2 Greenpeace International Submission to the Paris Convention for the Prevention of Marine Pollution
Appendix 3 List of EDC/VCM producers, with capacities and processes
Appendix 4 Race and siting of EDC/VCM facilities
Appendix 5 PVC suspension analyses
Appendix 6 EDC/VCM Heavy and Light Ends Reported by Generators
Appendix 7 Waste analysis data sheets for Dow, Plaquemine, LA showing 302 ppm PCBs.
Appendix 8 1994 notice of registration for Texas National Resource Conservation Commission, Dow EDC/VCM facility, two categories of K019 wastes: PCB contaminated solids and PCB contaminated chlorinated organics.
Appendix 9 Occidental's record of management describing generation of 10,000 pounds per year of "PCB-Contaminated Liquids" during "VCM Production."
Appendix 10 Greenpeace Analyses of K019 Wastes PU4015 PU4017
Appendix 11 Greenpeace Analyses of Probable K020 Wastes PU4009 PU4011 P U4043
Appendix 12 Greenpeace Analyses of F024-Containing Wastes PU4016 PU4027 PU4029 PU4034
Appendix 13 Greenpeace Analyses of K016 Wastes PU4014 PU4018 PU4020
Appendix 14 Greenpeace Analyses of Other EDC/VCM Wastes PU4013 PU4019 PU4021 PU4024 PU4025 PU4026 PU4028 PU4031 PU4037 PU4039 PU4041
Appendix 15 Greenpeace Analyses of EDC/VCM-Related Sediments PU4033 PU4036
Appendix 16 PVC Bans, Phase-outs and Resolutions
More than 99 percent of the dioxin1 now ubiquitous in the U.S. population and environment stems from human activities. Most of the dioxin released to the atmosphere from identified sources is produced during the combustion of artificial, chlorine-containing materials. (USEPA 1994a)
To create dioxin, four elements are required: hydrogen, oxygen, carbon and chlorine. Of these, chlorine is by far the least abundant in the general mass of artificial materials which give rise to the major portion of the national dioxin burden. As a result, chlorine is the limiting element for dioxin formation: if there is no chlorine, no dioxin can be created.
The chlorine in these materials can be traced back to the producers of chlorine-containing materials. As the destination for 34 percent of elemental? chlorine demand, polyvinyl chloride (PVC) manufacture stands as the largest single end use of elemental2 chlorine in the U.S. (Hileman et al. 1994). A compelling body of evidence suggests that, throughout its entire lifecycle3, PVC is responsible for a greater share of the nation's annual dioxin burden than any other industrial product.
1 In this paper, the tern "dioxin" includes all polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs), except where otherwise clearly indicated.
2 Elemental chlorine consists of two atoms of chlorine, an extremely reactive compound that is almost exclusively the product of modem technology, specifically the chloralkali industry.
3 The PVC lifecycle includes the following: (1) the production of chlorine and natural gas; (2) the production of EDC, followed by its conversion to VCM; (3) the polymerization of VCM into PVC; (4) the formulation and manufacture of products made of or containing PVC; (5) all gaseous, liquid and solid wastes generated in this sequence of events; (6) the gaseous, liquid and solid wastes generated during the treatment for disposal of the initial PVC-related industrial wastes as well as discarded PVC products; and (7) the gaseous, liquid and solid residues generated during the uncontrolled burning of PVC-containing materials. The pigments, stabilizers, flame retardants, and similar substances which are added to PVC during the formulation and production of PVC products are not addressed in this paper.
As documented in this report, PVC's contribution to the national dioxin burden begins during the manufacture of its precursors, ethylene dichloride (EDC) and vinyl chloride monomer (VCM), and continues through its use and disposal:
- Both dioxin and dioxin-like chemicals, including polychlorinated biphenyls (PCBs), are byproducts of EDC/VCM production processes. Both dioxin and PCBs are found in the PVC itself. Dioxin is released in air emissions, wastewater effluents and sludges from production facilities, while process wastes carry both dioxin and PCBs;
- Disposal by combustion of chlorine-rich process wastes from EDC/VCM facilities releases dioxin and dioxinlike compounds in the stack gases, pollution control residues and ashes of the on-site and off-site combustion facilities which bum such wastes. These releases consist of some portion of the innate dioxin and dioxinlike chemicals that escape destruction during combustion and, in greater part, the dioxin and dioxin-like compounds that are newly created within the combustion systems;
- Discarded PVC products contribute to the formation and release of dioxin and dioxinlike compounds in the stack gases, pollution control residues, and ashes of combustion systems (municipal waste incinerators, hospital waste incinerators, secondary metal smelters, etc.) as well as in their release in smoke and residues from the burning, both deliberate and accidental, of automobile fluff, construction debris, packaging, and other PVC-containing materials. Dioxin and dioxinlike compounds can also be expected in emissions and residues of household and building furnaces, on-site burners at grocery stores, etc. in which packaging, containers, and other PVC-based materials are burned; and
- PVC products in use (e.g., siding, flooring, pipes and other construction materials) are major contributors to the formation and release of dioxin in smoke, soot and other residues of building fires.
In 1994, U.S. PVC manufacturers used 9.37 billion pounds of chlorine to produce 10.88 billion pounds of PVC,4 (Lucas 1994; Coeyman, 1994) Based on an average chlorine content in PVC of 57.6 percent by weight, these figures suggest that some 67 percent of the chlorine used in PVC production is actually incorporated into the final PVC product. The remaining one-third of this chlorine apparently emerges as chlorine-containing chemicals other than PVC. Some of these chemicals may be reclaimed as marketable or usable by-products; the remainder become chlorine-rich wastes.
4 As shown in Appendix 3, eleven U.S. companies have fifteen facilities for manufacturing EDC, which is the precursor for VCM as well as other chlorinated hydrocarbons. Their combined EDC manufacturing capacity is over 22 billion pounds per year. (SRI 1992) Ten U.S. companies have twelve facilities with a combined VCM-manufacturing capacity of over 13 billion pounds per year. (Westervelt 1995) Virtually all VCM is polymerized to form PVC.
At some point in the lifecycle of PVC, an unknown but undoubtedly substantial portion of the 9.37 billion pounds of chlorine that goes into PVC manufacturing is subjected to conditions conducive to the generation of dioxin and PCBs. Such conditions exist at many points in the PVC lifecycle, including EDC and VCM production processes.
There is broad agreement that the manufacture of PVC adds to the national dioxin burden: (USEPA 1994a)
'There is no apparent dispute between the industry and Greenpeace regarding the formation of CDDs/CDFs (chlorinated dibenzo-p-dioxins/chlorinated dibenzofurans) during the [PVC] production process, nor that some CDDs/CDFs are released to various environmental media."
In fact, one EDC/VCM manufacturer, ICI Polymers & Chemicals Ltd., contends that dioxin is apparently an unavoidable by-product of oxychlorination, a key process in PVC manufacture, describing the situation as follows: (ICI 1994)
"It is difficult to see how any of these conditions could be modified so as to prevent PCDD/PCDF formation without seriously impairing the reaction for which the process is designed."
PCBs are also apparently unavoidable by-products of PVC manufacture, contrary to USEPA's assertion that PCBs "are no longer produced in industrialized countries" (USEPA 1994a). In formal comments submitted to the Agency on proposed regulations affecting EDC/VCM production, Dow Chemical Company offered the following statements: (Dow 1981)
'The technology to prevent the formation of trace quantities of PCB's does not exist in most instances. PCB's can be formed in trace quantities any time that chlorine in any reactive form is contacted with carbon or compounds of carbon at elevated temperatures and/or in the presence of a catalyst."
USEPA recently acknowledged that "[d]ue to the chemicals involved in the manufacture of PVC, some incidental generation of PCBs is not unexpected." (Goldman 1994) In fact, USEPA has been categorizing some, if not all, EDC/VCM producers as PCB manufacturers for many years. For example, a waste characteristics data sheet from the Vista Chemical EDC/VCM facility in Lake Charles, Louisiana, offers the following description: (Appendix 1)
'The VCM Plant is an inadvertent manufacturer of PCB's but is certified as an excluded manufacturing process."
One waste disposal firm reported to USEPA that their client, PPG Industries in Lake Charles, Louisiana, generates "250 tons per year of material ... [having] elevated PCB levels between 50 ppm and 1000 ppm." (Cooper 1993) Those wastes identified as containing the highest levels of PCBs were VCM distillation bottoms followed by liquid wastes from EDC production and still bottoms from perchloroethylene and trichloroethylene production. (Cooper 1994)
As discussed later in this document, dioxin and PCBs are co-products of EDC/VCM production and related processes. In other words, the detection of PCBs in chemical products and wastes suggests that dioxin is also present, and vice versa.
This dioxin/PCB relationship can be particularly useful in identifying additional dioxin sources. For example, in 1982, USEPA assembled a list of suspected inadvertent manufacturers of PCBs. Identified on this list were 467 chemical processes, including EDC and VCM production, as well as the 169 chemical manufacturing facilities using these processes. (Versar 1983) All of these processes can now be regarded as potential sources of both dioxin and PCBs.
European studies have confirmed the generation of dioxin during EDC/VCM production and its subsequent dispersal into the environment surrounding EDC/VCM facilities. As reviewed in a submission by Greenpeace International to the Paris Convention for the Prevention of Marine Pollution, these studies document the dispersal of dioxin in the air emissions, wastewater discharges and various wastes from EDC/VCM facilities. (GPI 1994; see Appendix 2)
Also included in the submission to the Paris Convention is an analysis commissioned by Greenpeace of wastewater effluent from a Dutch EDC/VCM facility, revealing a dioxin concentration of 15.7 TEQ5 ppt6 (GPI 1994; see Appendix 2). In comparison, ICI reported a dioxin concentration of 37.22 ppt TEQ in the wastewater discharge from its VCM unit. (ICI 1994)
These wastewater dioxin concentrations are of particular interest in relationship to dioxin in sediments. For example, in one study, up to 80 percent of the dioxin in sediment samples from the Rhine River (Netherlands) were attributed to an upstream
5 Individual PCDDs and PCDFs with chlorine atoms at the 2,3,7, and 8 positions have been assigned Toxicity Equivalence Factors (TEFs) that reflect each chemical's potency relative to that of 2,3,7,8-tetrachlorodibenzo-pdioxin (2,3,7,8-TCDD), which has a TEF of 1. (NATO-CCMS, 1988a, 1988b) After multiplying the concentrations of individual PCDDs and PCDFs by the assigned TEFs and summing the results, a PCDD/PCDF mixture can be described as containing the equivalent of a quantity of 2,3,7,8TCDD that is equal to the summed results, commonly expressed as Toxic Equivalents (TEQs).
6 "Ppt" denotes parts per trillion, e.g., picograms per gram, nanograms per kilogram, etc.
EDC/VCM facility. (Evers et al. 1988; Evers 1989) In another, dioxin concentrations ranging from 433 to 922 ppt TEQ were detected in sediments from a harbor on which an EDC/VCM facility is sited. (Wenning et al. 1992; Evers et al. 1989b)
In its estimate of dioxin emissions from U.S. EDC/VCM manufacturers, USEPA used emission factors provided by European PVC manufacturers (Miller 1993) to calculate total dioxin emissions ranging between 0.45 and 23 grams TEQ per year (USEPA 1994a). This estimate may be much lower than actual dioxin emissions. For instance, U.S. EDC/VCM production is estimated to be fifteen times greater than that in Norway (SRI 1992), where the government has estimated total dioxin releases from EDC/VCM production to be 10 grams TEQ per year (SFT 1993).
Both of these national estimates contrast sharply with the dioxin release rates reported from ICI's single EDC/VCM facility. In an inspection report to British authorities, ICI estimated dioxin releases from its "VC3" unit, which had an annual VCM production rate of 200,000 metric tons, as follows: (ICI 1994)
- Air emissions: No data;
- Wastewater discharges: 0.54-10 grams TEQ per year,
- Sludge: No data;
- "Heavies": 12-30 grams TEQ per year.
If U.S. VCM producers, with their total production capacity of 6.89 million tons VCM per year (see Appendix 3), are releasing dioxin in their wastewater at rates similar to those at the ICI facility, they are discharging dioxin into U.S. waterways at rates ranging from 16 to 313 grams TEQ per year. In comparison, USEPA has estimated that all 104 U.S. pulp and paper mills that use chlorine for bleaching discharged dioxin in their wastewaters at the rate of 105 grams TEQ per year in 1993. (USEPA 1994a) Moreover, with dioxin concentrations in ICI's VCM-related wastewater discharge as high as 37.22 ppt TEQ (ICI 1994), related wastewater treatment sludges can be expected to carry substantial quantities of dioxin.
The ICI report also indicates that, at their EDC/VCM facility, the "Heavies" from VCM production are used as feedstock for the production of 100,000 metric tons per year of perchloroethylene and trichloroethylene. Dioxin generated from this unit was estimated as follows: (ICI 1994)
- Air emissions: No data;
- Wastewater discharges: 0.6-1.5 grams TEQ per year;
- "DOPP" waste: 237-495 grams TEQ per year,
- "LEWA" waste: 108-131 grams TEQ per year; and
- Discarded catalyst: 0.3 grams TEQ per year.
PPG Industries (Brown 1989) and other U.S. EDC/VCM facilities, including Dow in Plaquemine, Louisiana and in Freeport, Texas, have integrated processes for the production of VCM, perchloroethylene and/or trichloroethylene. Consequently, it appears likely that such integrated facilities are adding a very substantial quantity of dioxin to the U.S. environment.
Based on the dioxin concentrations in ICI's "Heavies", their VCM production and the U.S. VCM production rate, the quantity of dioxin in the heavy ends from U.S. VCM production, which are coded by USEPA as K020 waste, may range between 376 and 940 grams TEQ per year. Whether these and other chlorine-rich wastes from EDC/VCM production are used as feedstock to produce other organochlorines or are treated by the most common method, incineration, they add to the annual dioxin burden.
ICI's data indicate that using dioxin-containing "Heavies" from EDC/VCM production as feedstocks to produce perchloroethylene and trichloroethylene leads to the generation of wastes with even higher dioxin levels. While dioxin concentrations in ICI's EDC/VCM "Heavies" ranged from 3,100 to 7,500 ppt TEQ, concentrations in the perchloroethylene/trichloroethylene wastes ranged from 30,000 to 90,000 ppt TEQ. (ICI 1994)
These dioxin concentrations can be compared to 6,365,700 ppt TEQ measured in a sample collected by Greenpeace from a container, located at a U.S. EDC/VCM facility, which was identified as heavy ends from the distillation of EDC. In another sample collected by Greenpeace, which was taken from a container labeled to indicate it contained heavy ends from the distillation of VCM, a dioxin concentration of 3,191.2 ppt TEQ was found. A third waste sample from another EDC/VCM facility contained dioxin at a concentration of 19,978 ppt TEQ. (See Section 2.1.2, 2.1.3, 2.1.6 and subsequent sections of this report for a more detailed description of the analyses commissioned by Greenpeace of these and other waste samples from U.S. EDC/VCM facilities.)
When dioxin-containing wastes such as those from EDC/VCM production and related processes are combusted, some portion of the dioxin present in these wastes will escape destruction. USEPA has reported that incinerators do not achieve the destruction and removal efficiency (DRE) of 99.9999 percent, which is the legal requirement for dioxin-listed wastes, on chemicals that are present at less than 10,000 ppm. Indeed, even the 99.99 percent DIRE required for ordinary hazardous waste is not achieved with waste constituents that are present at concentrations of less than 1,000 ppm. (Kramlich et al. 1989; Lowrance 1992) In other words, incinerators burning these wastes will release unburned a share of the dioxin indigenous to these wastes that exceeds legal standards.
In addition to the dioxin that escapes destruction, a far greater quantity of dioxin is created within the combustion systems themselves when the chlorine-rich wastes from EDC/VCM production are combusted. In their review of the relationship between chlorine content and dioxin emissions, researchers at Princeton University documented a proportional relationship: the greater the chlorine content of combusted materials, the greater the dioxin emissions, subject to other factors, such as operating conditions and pollution control efficiencies. (Thomas and Spiro 1994)
Average chlorine concentrations of 64 to 76 weight percent have been reported for various EDC/VCM wastes - VCM distillation bottoms and liquid EDC wastes - as well as still bottoms from the manufacture of perchloroethylene and trichloroethylene. (Cooper 1994) For dioxin-contaminated wastes with a far lower chlorine content than that of EDC/VCM wastes, USEPA has estimated that stack emissions of dioxin created during combustion may be 10 to 100 times greater than emissions of the dioxin that escapes combustion. (RTI 1994)
As described in the following sections, various EDC/VCM production processes offer many opportunities for the generation of dioxin and PCBs. To further substantiate the dioxin and PCB contamination of EDC/VCM production wastes, Greenpeace selected several U.S. EDC/VCM producers where representatives collected and had analyzed a total of 51 samples which included wastes, effluents and sediments. The samples were shipped to a certified analytical laboratory where all 51 samples underwent a general screening analyses by gas chromatography/mass spectrometry. The 25 samples which contained dioxin indicators are discussed in detail later in this report. Of these 25 samples, four were selected for dioxin testing and two were selected for PCB testing. Dioxin and PCBs were detected in all samples subjected to such analyses.
There can be no argument that throughout its lifecycle PVC manufacture is making a substantial contribution to the national dioxin burden. However, there also can be no argument that the communities in which EDC/VCM facilities are sited bear the brunt of the dioxin and dioxinlike chemicals released from these facilities. Airborne dioxin is emitted in process vent gases as well as in the stack gases of the on-site combustion units in which much of these facilities' chlorine-rich wastes are burned. The EDC/VCM facilities also discharge into waterways dioxin where it accumulates in sediments and in the fish, shrimp and other aquatic organisms that serve as the basic food supply as well as an economic resource for many of these communities.
In general, communities with EDC/VCM facilities in their midst have a higher percentage of people of color than the national average. (See Appendix 4) Emission problems and the risk of catastrophic accident from EDC/VCM facilities, as well as corporate fear of liability, has resulted in entire communities being literally wiped off the map. In 1987, 106 residents of Reveilletown, Louisiana, a small African-American community about ten miles south of Baton Rouge, filed a lawsuit against Georgia-Pacific and Georgia-Guff arguing that they had suffered health problems and property damage. After settling out-of-court for an undisclosed amount, Georgia-Gulf relocated the remaining families and then tore down every structure in town including the church. Management at Dow Chemical's neighboring EDC/VCM facility in Plaquemine followed suit soon afterwards, buying out all of the residents of the small town of Morrisonville. (Bowermaster 1993)
Subsistence and commercial shrimpers and fishermen, including a large contingent of people of Vietnamese origin have organized protests to stop present and future pollution of shrimping and fishing areas from EDC/VCM production near Point Comfort, Texas, where a Formosa Plastics facility is located.
In summary, EDC/VCM production, including releases of dioxin and dioxin-like chemicals from production processes and related waste disposal, contribute to the national dioxin burden. However, the impacts of EDC/VCM production weigh most heavily on nearby communities in which an above average percentage of residents are people of color.
1.1 EDC/VCM Production Processes
The generation and release of dioxin during at least one step in EDC/VCM production -oxychlorination - is apparently unavoidable. (ICI 1994) However, conditions favorable to the formation of dioxin also exist in other EDC/VCM processes.7
The first effective industrial route to VCM was the hydrochlorination of acetylene, catalyzed by mercuric chloride. One facility, Borden Chemicals and Plastics, still has the capacity to produce 300 million pounds of EDC by this method. However, most U.S. EDC/VCM facilities use the balanced process, in which EDC is produced by a combination of two processes: direct chlorination (the reaction of ethylene with elemental chlorine) and oxychlorination (the reaction of ethylene with hydrogen chloride (HCI) and oxygen). EDC is then pyrolyzed to yield VCM and HCI. The HCI is fed back into the oxychlorination process and the VCM is ready to polymerize into PVC. (Cowfer and Magistro 1983) (See Appendix 3 for a list of U.S. EDC/VCM facilities, their locations, processes and capacities.)
7 Given the evidence for the concurrent generation of PCBs and dioxin, except where otherwise stated, circumstances described in this report as conducive to dioxin formation can also be assumed to support the formation of PCBs. for PCB formation.
1.1.1 EDC Via Direct Chlorination: Opportunities for Dioxin Formation
In general, ferric chloride is used to catalyze the direct chlorination of ethylene to EDC. Although no oxygen is fed into the reaction, it is commonly present as an impurity in the chlorine. (Cowfer and Magistro 1983)
Like other inorganic chlorides (Vogg et al. 1987), ferric chloride can catalyze dioxin formation, as demonstrated with benzene by Nestrick et al. (1987). In direct chlorination, oxygen would appear to be the limiting element for dioxin formation but not EDC from direct chlorination is usually 99.5 percent pure, so little further purification is necessary other than removal of the ferric chloride catalyst. This is accomplished by adsorption of the ferric chloride on a solid, by distilling the EDC from the catalyst in a boiling reactor, or by removing the ferric chloride by washing with water. (Cowfer and Magistro 1983)
Besides the possibility of dioxin formation during direct chlorination, purification of the ferric chloride catalyst may also provide opportunity for dioxin formation. For example, the increased heat and oxygen availability during catalyst removal via a boiling reactor offers conditions favorable to dioxin formation. Depending on temperature and pH, catalyst removal by water washing may also provide an opportunity for dioxin formation.
Industry sources have reported the generation of PCBs during the direct chlorination of ethylene for the purpose of producing perchloroethylene (Hill 1979) and for the production of VCM, EDC, perchloroethylene, trichloroethylene, methyl chloroform, ethyl chloride, vinylidene chloride monomer, and muriatic acid (Duncan 1982). The concurrent generation of dioxin and PCBs in other chemical processes implies that dioxin is also formed during direct chlorination.
1.1.2 EDC Via Oxychlorination: The Unavoidability of Dioxin Formation
The formation of dioxin during oxychlorination is evidently inescapable. The British firm, ICI Chemicals & Polymers Ltd. (1994) describes this circumstance as follows:
"The processes used by ICI to produce vinyl chloride monomer, trichloroethylene and perchloroethylene involve oxychlorination stages. The oxychlorination stage in the vinyl chloride process makes 1,2-dichloroethane which is converted into vinyl chloride by pyrolysis. In the process for making trichloroethylene and perchloroethylene an oxychlorination reactor is used to convert raw materials directly into solvent products. It has been known since the publication of a paper in 1989 [Evers et al. 1989a] that these oxychlorination reactions generate polychlorinated dibenzodioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). The reactions include all of the ingredients and conditions necessary to form PCDD/PCDFs, ie air or oxygen, a hydrocarbon (ethylene etc), chlorine or hydrogen chloride, a copper catalyst, an ideal temperature and an adequate residence time. It is difficult to see how any of these conditions could be modified so as to prevent PCDD/PCDF formation without seriously impairing the reaction for which the process is designed." (ICI 1994)
In the oxychlorination process, dry HCI reacts with ethylene and oxygen, over copper chloride catalyst in either a fluidized bed reactor at 220-235 C or a fixed bed reactor at 230-300 C. (Cowfer and Magistro 1983) As described above by ICI (1994), these are excellent conditions for dioxin formation: available oxygen and heat energy as well as a copper catalyst that also catalyzes the formation of dioxin (Vogg et al. 1987; Gullett et al. 1990).
In comparison to direct chlorination, oxychlorination yields less EDC and more byproducts - carbon tetrachloride, chloroform and other "higher boiling compounds". Using air, instead of oxygen, gives lower yields of EDC and higher by-product yields in addition to vent gas volumes 20-100 times those of oxygen-based operations. The type of oxychlorination process used determines the levels of undesirable impurities, ethylene, and chlorinated hydrocarbons in the vent gas. (Cowfer and Magistro 1983)
Increased formation of highly chlorinated by-products and tars has been attributed to low levels of acetylene present in the HCI derived from EDC pyrolysis. (Cowfer and Magistro 1983) This latter circumstance suggests that the one U.S. facility that still produces EDC by the hydrochlorination of acetylene, Borden Chemicals and Plastics, may be generating more dioxin per unit of product than other EDC/VCM facilities.
Well prior to the ICI (1994) report, laboratory simulations at the University of Amsterdam demonstrated dioxin formation during oxychlorination at a rate equivalent to 419 grams TEQ per 100,000 tons of EDC produced. (Evens 1989)
1.1.3 EDC Pyrolysis to VCM: Opportunities for Dioxin Formation
The thermal cracking/pyrolysis of EDC to VCM and HCI is expedited by initiators or accelerators, which include carbon tetrachloride, chlorine, bromine, iodine, or oxygen. (Cowfer and Magistro 1983) The use of oxygen can be expected to facilitate dioxin formation in this process, as supported by observations that exclusion of oxygen can decrease the formation of nonvolatile byproducts, which is evidenced by less "fouling" of pyrolysis tube walls (Cowfer and Magistro 1983).
The cracking of EDC is generally conducted at temperatures of 500-550 C (Cowfer and Magistro 1983), which are well within ranges in which dioxin formation has been reported in combustion systems, but is slightly higher than the range of maximum dioxin formation (Vogg et al. 1992). The increased production of heavy ends and tars that accompanies slower cooling (Cowfer and Magistro 1983) can also be expected to be accompanied by increased generation of dioxin.
Both light and heavy ends are produced during EDC pyrolysis. These residues are either further processed or disposed of by incineration or other means. (Cowfer and Magistro 1983) Numerous by-products are formed during pyrolysis, some of which are generated concurrently with dioxin in combustion processes. Such products include "acetylene, ethylene, methyl chloride, butadiene, vinylacetylene, benzene, chloroprene, vinylidene chloride, 1,1dichloroethane, chloroform, carbon tetrachloride, 1,1,1trichloroethane, and other chlorinated hydrocarbons" (Cowfer and Magistro 1983).
1.2 EDC/VCM Wastes
Residues from EDC/VCM manufacture occur as gaseous, organic liquid, aqueous, and solid wastes. As discussed earlier, dioxin has been detected in some, if not all, of these wastestreams.
The combustion of these wastes creates a second generation of gaseous, liquid and solid residues. Chlorine, some portion of which is already in the form of dioxin and PCBs, is ubiquitous in virtually all of these wastes. Consequently, some or all of these combustion emissions, effluents and residues can be expected to contain dioxin and dioxinlike chemicals - both those indigenous to the waste and those formed as products of incomplete combustion.
1.2.1 Vent Gases
As discussed earlier, dioxin has been detected in vent gases from EDC/VCM facilities. (SFT 1993) Moreover, vent gases are commonly treated by incineration or catalytic combustion. (Cowfer and Magistro 1983) Since vent gases often contain HCI, vinyl chloride, chlorine and various hydrocarbons (Cowfer and Magistro 1983), such treatment is obviously amenable to the formation of dioxin and dioxinlike chemicals with their subsequent release in combustor air emissions, solid residues and, if present, scrubber liquids.
There is insufficient information to determine vent gas treatment practices among U.S. facilities or USEPA regulation of such treatment. However, vent gases are treated on-site by virtually all EDC/VCM producers. Some portion of vent gases may be treated in RCRA-permitted boilers, incinerators, thermal oxidizers, and halogen furnaces. Combustion of vent gases also takes place in non-RCRA devices designated as recycling facilities and/or in-process flares.
There is no dioxin monitoring data required of or available from these combustion systems, other than that occasionally gathered during trial bums for the RCRA-permitted combustion systems. However, trial bum data substantiate the emissions of dioxin from boilers and incinerators, as briefly reviewed in the Agency's draft dioxin reassessment. (USEPA 1994a) Also, studies have documented the emission of dioxin from flares
burning dilute mixtures of gases, such as those from landfills, which include vinyl chloride and other chlorinated species that are present in vent gases from EDC/VCM production. (USEPA 1991a; USEPA 1992)
As PCB manufacturers, EDC/VCM producers are required to restrict air emissions of PCBs to 10 ppm or less, but there are no requirements to monitor or report PCB emissions unless estimated PCB releases to air exceed 10 pounds per year. (40 CFR 761)
1.2.2 Organic Liquid Wastes
Organic liquid wastes include light and heavy ends from EDC purification. Light ends contain ethyl chloride, cis- and trans-1,2-dichloroethylene, chloroform, and carbon tetrachloride. The concentrations of dioxin in heavy ends have been reported by ICI (1994) and by Greenpeace (See Sections 2.1.2 and 2.1.3). Likewise, PCBs are found in heavy ends, as discussed in several other sections of this report. These circumstances suggest that dioxin may also be present in light ends. (Cowfer and Magistro 1983)
In addition to dioxin and PCBs, heavy ends contain 1,1,2-trichloroethane, tetrachloroethanes, chlorinated butanes, chlorinated aromatics, and many other compounds. Sometimes useful components are recovered by fractional distillation, with remaining materials incinerated. (Cowfer and Magistro 1983) As discussed earlier, heavy ends are also used as feedstock for the production via oxychlorination of perchloroethylene and trichloroethylene. ICI (1994)
In another approach, all organic liquid wastes are combusted in a fluidized-bed, catalytic oxidizer and resulting gaseous combustion products are fed into the oxychlorination process for HCI recovery as EDC. (Cowfer and Magistro 1983) This "recycling" can be expected to contribute further to the quantities of dioxin and PCBs formed during oxychlorination.
1.2.3 Aqueous Wastes
As discussed earlier, dioxin has been detected in the wastewater effluents of EDC/VCM facilities (SFT 1993; ICI 1994) as well as in the sludges (SFT 1993). Moreover, a study funded by PPG Industries, a U.S. EDC/VCM producer, ascertained the discharge in that facility's wastewater of hexachlorobenzene (HCB) and hexachlorobutadiene (HCBD) in concentrations as high as 2,000 ppb, with a total discharge of 0.28 and 0.39 pounds per day respectively. (Brown 1989) As discussed by Evers (1989), these two chemicals, HCB and HCBD, as well as pentachlorobenzene, tetrachlorobenzene, and 1,1,3,4-tetrachloro-1,3-butadiene "can be regarded as important indicators for the thermochemical formation of PCDDs and PCDFs."
USEPA (1994a) suggests that water effluent limitations for these facilities "are expected to control releases of CODs and CDFs to minimal levels," even though such limitations do not address dioxin. The Agency offers no data or documentation to support this claim and offers no definition of the tern "minimal." As presented in Section 2.1.9, the total PCDD/PCDFs, concentration of >2,911 ppt in a sediment sample taken by Greenpeace from a point immediately downstream from one U.S. EDC/VCM facility suggests that water effluent limitations are not controlling releases of dioxin to "minimal levels." USEPA also fails to acknowledge that any effort to control dioxin concentrations in wastewater discharges will necessarily increase the dioxin content of wastewater treatment sludges.
As PCB manufacturers, EDC/VCM facilities are required under the Toxic Substances Control Act (TSCA)8 to report PCB releases greater than 10 pounds per year in their effluents. However there are no monitoring requirements for effluents and no monitoring or reporting requirements for wastewater sludges. (40 CFR 761)
Aqueous waste streams are generally steam-stripped to remove volatile organics then neutralized, and the remaining nonvolatile organics are treated in an activated sludge system. Depending on its stringency, steam stripping can potentially carry some portion of any dioxin and PCBs in the sludge into the air. Dioxin and PCBs remaining in the sludge should be largely unaffected by biological treatment resulting in their release in both treated effluents and sludges.
8 Under the Toxic Substances Control Act (TSCA), USEPA is charged with enforcing certain prohibitions and restrictions on the manufacture and use of PCBs as well as certain disposal requirements for PCB-containing materials and equipment.
1.2.4 Solid Wastes
Solid wastes include "furnace coke, reboiler drillings, and inorganic sludge from waste water treatment." (Lewis 1989) Substantial concentrations of dioxin have been reported in both tars and sludges from Norwegian EDC/VCM facilities. (SFT 1993)
Greenpeace's analysis of a waste sample containing cracking furnace coke from an EDC/VCM facility isolated 198 compounds, including polychlorinated benzenes, at least one PCB and hexachlorobutadiene (HCBD). (See Section 2.1.8) The presence of the latter chemical, HCBD, suggests that dioxin is also present in this sample. (Evers 1989)
In summary, dioxin and PCBs permeate the production of PVC. Dioxin has been identified in the air emissions, wastewater effluents and other wastes generated during PVC manufacture. In many of the wastes for which no dioxin and PCB data are available, the presence of these contaminants is strongly suggested by analyses showing the presence of one or more of the chemicals identified by Evers (1989) as dioxin indicators. Sections 2.1.2, 2.1.3, 2.1.6, 2.1.7 and 2.1.8 of this report describe the presence of dioxin and/or PCBs and/or dioxin indicator chemicals in 23 waste samples collected by Greenpeace from the following EDC/VCM facilities: Borden Chemical, Geismar, Louisiana; Vulcan Chemical, Geismar, Louisiana; Georgia Gulf, Plaquemine, Louisiana; Vista Chemical, Lake Charles, Louisiana; PPG Industries, Lake Charles, Louisiana; BFGoodrich - Geon, LaPorte, Texas; Occidental Chemicals, Ingleside, Texas; and Formosa Plastics, Point Comfort, Texas.
1.3 Dioxin and PCBs in PVC Plastic
When PVC emerges from the final production step, the PVC itself carries dioxin and PCBs. The Swedish Environmental Protection Agency recently reported detectable concentrations of both dioxin and PCBs in PVC itself. (SEPA 1994) Analyses of pure PVC suspension from two Swedish PVC facilities revealed total concentrations, including PCBs, ranging from 0.86 to 8.69 ppt TEQ. (Appendix 5) Based on these figures and an annual U.S. production of 10.88 billion pounds of PVC (Coeyman 1994), a quantity of dioxin ranging between 4.2 and 42.9 grams TEQ per year is entrained in the PVC products manufactured in the U.S.
To date, neither USEPA nor any other U.S. governmental agency has released data describing the occurrence of dioxin and PCBs in U.S.-produced PVC or PVC products.
USEPA has been aware of the generation of dioxin and PCBs during PVC manufacture for more than seven years in the case of dioxin and more than fifteen years in the case of PCBs. In 1988, USEPA proposed RCRA regulations in which dioxin was listed among the constituents of concern in one PVC-related waste. (Berlow and Vorbach 1988)
In 1990, at the request of the Vinyl Institute and various PVC producers, USEPA deleted dioxin from this list. (Rosengrant and Craig 1990; Kinch and Vorbach 1990) Besides making the decision not to regulate dioxin in this case, the Agency has also failed to list dioxin as a constituent of concern among other regulated wastes in which dioxin can be expected to be present.
It is important to note that USEPA does not address all forms of dioxin when promulgating RCRA regulations governing the monitoring and treatment of hazardous waste. USEPA relies on a list of constituents to be considered, the so-called "BDAT list," which includes only one of the seventeen congeners of greatest toxicological concern - 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD). Moreover, the list includes only six of the ten groups of homologues, groups in which the different PCDD/PCDFs, are aggregated according to the number of chlorine atoms in their molecules. (Berlow and Jones 1988)
This regulatory limitation makes it impossible to calculate the total dioxin-related toxicity, in TEQ, of any RCRA-regulated waste. Further, exclusion of the hepta- and octachlorodibenzofurans is particularly significant in the context of wastes from the PVC industry since large quantities of the higher chlorinated dibenzofurans are produced. (Wenning et al. 1992)
When USEPA promulgated RCRA regulations for two of the PVC wastes known or suspected to contain PCBs, the Agency omitted PCBs from the list of constituents of concern for the wastes. Le., USEPA made a decision not to regulate PCBs under RCRA. Instead, USEPA opted to regulate these wastes under two separate laws: the listed constituents of concern under RCRA, and the PCBs under TSCA:
Under RCRA, PVC producers must report and treat to set standards the constituents of concern in certain wastes. USEPA's exclusion of dioxin and PCBs from listed constituents of concern in any PVC-related wastes relieved PVC manufacturers of requirements that they reveal the extent of their dioxin and PCB generation in RCRA biennial reports and that they treat such wastes so that dioxin and PCB concentrations in the treated wastes do not exceed specific limits.
Under TSCA, the Agency designated some, if not all, PVC producers as "excluded" PCB manufacturers. With this status, PVC producers do not have to ensure that PCB concentrations in treated wastes fall below certain limits, as can be required under RCRA. Instead, as long as PCB concentrations in the untreated wastes do not exceed 500 ppm, PVC producers can send their PCB containing wastes to chemical waste landfills or bum them in their own or others' boilers. Under TSCA, there are no limits placed on the PCB concentrations that can remain in the treated wastes. (40 CFR 761)
In short, USEPA has made regulatory decisions affecting PVC manufacturers that allow the full extent of this industry's contributions to the nation's dioxin burden, including its output of PCBs, to be hidden from the public as well as the Agency itself.
USEPA has assigned source-specific hazardous waste codes to only two EDC/VCM wastes. Other EDC/VCM wastes such as light ends, aqueous wastes and solids (furnace coke, reboiler drillings, and wastewater sludges) have no such designation. The two EDC/VCM wastes that are included in the Agency's listing of hazardous wastes from specific sources are as follows: (Below and Jones 1988)
K019: Heavy ends from the distillation of ethylene dichloride in ethylene dichloride production; and
K020: Heavy ends from the distillation of vinyl chloride in vinyl chloride monomer production.
In the absence of clear guidelines from USEPA and state agencies, EDC/VCM facilities often identify and report wastes, including heavy ends from EDC and VCM production, under several other waste codes, as indicated in Appendix 6. For example, wastes reported by one facility as K019 may be reported by another as F024. This and other waste codes sometimes used by EDC/VCM producers to identify their wastes are as follow:
F024: Process wastes, including, but not limited to, distillation residues, heavy ends, tars, and reactor clean-out wastes, from the production of certain chlorinated aliphatic hydrocarbons by free radical catalyzed processes. These chlorinated aliphatic hydrocarbons are those having carbon chain lengths ranging from one to and including five, with varying amounts and positions of chlorine substitution.
F025: Condensed light ends, spent filters and filter aids, and spent desiccant wastes from the production of certain chlorinated aliphatic hydrocarbons, by free radical catalyzed processes. These chlorinated aliphatic hydrocarbons are those having carbon chain lengths ranging from one to and including five, with varying amounts and positions of chlorine substitution.
D043: Vinyl chloride
D028: Ethylene dichloride
Waste-related data are not reported to USEPA in a useful or verifiable fashion, as Appendix 6 illustrates. In a forthcoming report, the Environmental Action Foundation concludes as follows:
"Lack of government agency program integration has made it difficult if not impossible to review and compare data reported by the 14 EDC/VCM producers in the U.S. in order to determine exactly how much waste they are generating, incinerating or managing in other manners." (Monsma and Kaerwer 1995)
Wastes from EDC/VCM facilities are managed, to a considerable extent, by onsite treatment, disposal, recycling and energy recovery systems. Inconsistencies in waste data submitted by these facilities - that required by the Emergency Planning and Community Right-to-Know Act for inclusion in the USEPA's Toxics Release Inventory (TRI) as well as in that required in the RCRA Biennial Reports - indicate that USEPA has not adequately characterized the wastes from these facilities. As a result, the Agency is unable to determine the quantities, characteristics and fates of many EDC/VCM wastes and cannot identify irregularities in the limited waste data that are reported. The Environmental Action Foundation has documented some of the inconsistencies in TRI data, using the manufacture of PVC as a case study. (Monsma and Kaerwer 1995)
2.1 USEPA's BDAT Standards
In regulating waste management via the best demonstrated available technology (BDAT) under RCRA, USEPA selects those waste chemicals to be regulated - the constituents of concern - from a list, which, as of 1988, contained some 249 substances. Commonly referred to as the BDAT constituent list, the 1988 list included 187 volatile and semivolatile organic chemicals; seventeen metals; three inorganics; 28 pesticides; seven PCBs; 2,3,7,8-tetrachlorodibenzo-p-dioxin; and the tetra-, penta-, and hexachlorinated dibenzo-p-dioxins and dibenzofurans. (Berlow and Jones 1988) As mentioned earlier, the other sixteen dioxin isomers of toxicological concern are not included on this list nor are the hepta- and octachlorodibenzo-p-dioxins and dibenzofurans.
Further, USEPA does not consider a constituent from the BDAT list for regulation if: "(1) the constituent was not detected in the untreated waste; (2) the constituent was not analyzed in the untreated waste; or (3) detection limits or analytical results were not obtained for the constituent due to analytical or accuracy problems." (Berlow and Jones 1988) The Agency also declines to regulate waste constituents in VCM waste (F024) to reduce costs: (Berlow and Jones 1988)
"... the Agency is not regulating all of the constituents considered for regulation in order to reduce the analytical cost burdens on the treater and to facilitate implementation of the compliance and enforcement program."
2.1.1 USEPA's Characterization of K019 and K020 Wastes
In USEPA's BDAT regulations for K019 and K020 wastes - the heavy ends from EDC distillation and VCM distillation, respectively - the Agency makes no mention of the known occurrence of PCBs or the suspected occurrence of dioxin in these wastes. (Berlow and Jones 1988) As noted earlier, USEPA obviously has long-standing knowledge of the presence of PCBs in wastes from EDC/VCM production, having certified some VCM facilities as "excluded [PCB] manufacturing process[es]" under TSCA. (Appendix 1) The Agency also had access to waste analysis data sheets from EDC/VCM facilities, such as those from Dow's EDC/VCM facility in Plaquemine, Louisiana, which describe a concentration of 302 ppm9 PCBs in "EDC I Heavies" identified as hazardous waste number "KO19". (Appendix 7)
9 "ppm" denotes parts per million, e.g., milligrams per kilogram, micrograms per gram, etc.
According to 1994 notices of registration for industrial and hazardous waste from the Texas Natural Resource Conservation Commission, the Dow EDC/VCM facility in Freeport, Texas, generates two categories of K019 wastes: "PCB contaminated solids" and "PCB contaminated chlorinated organics." (Appendix 8) Similarly, a record of hazardous waste management activities for Occidental's EDC/VCM facility in Deer Park, Texas, describes the generation of 10,000 pounds per year of "PCB-Contaminated Liquids" during "VCM Production." (Appendix 9)
In drawing up BOAT regulations for K019 and K020 wastes, USEPA evaluated data describing a portion of the chemicals identified in K019 waste. Based on the highest concentrations reported for all identified K019 components, at least 72 percent of the waste remained uncharacterized. (Berlow and Jones 1988)
In one set of K019 samples, USEPA reported detectable quantities of 21 of 155 organic constituents for which analyses were conducted and 8 of 16 metals. In analyses of K019 wastes burned in rotary kiln incinerator performance tests, USEPA identified 40 organic constituents in the waste (16 volatiles and 22 semivolatiles) and 11 metals. (Berlow and Jones 1988)
2.1.2 Greenpeace Characterization of K019 Wastes
At Vulcan Chemical's EDC/VCM facility in Louisiana, Greenpeace collected two waste samples from containers clearly labeled as K019 wastes. (For more detailed descriptions and analytical data, see Appendix 10.)
In the one K019 sample analyzed for dioxin, the total PCDD/PCDF concentration was 200,750,000 ppt; the concentration, expressed in dioxin equivalents, was 6,365,700 ppt TEQ. These extraordinarily high dioxin concentrations are comparable to those in the Agent Orange wastes at the Vertac Superfund site in Jacksonville, Arkansas. The two categories of Agent Orange waste, "2,4-D waste"10 and "2,4,5-T waste"11 can be estimated to contain a total PCDD/PCDF concentration range of 33,000 to 238,000 ppt and 24,000,000 to 52,847,000,000 ppt, respectively. (Wicklund 1992; USEPA 1981; CAE 1991)
No dioxin analysis was performed on the second sample. However, during the general screening analysis, the sample was found to contain four of the five chemicals recognized as indicators for dioxin (Evers 1989): hexachlorobenzene, pentachlorobenzene, tetrachlorobenzene, and hexachlorobutadiene. In other words, this sample almost certainly contains dioxin as well.
Both samples contained detectable levels of seven of the ten metals for which analyses were conducted. The metals present at highest concentrations and their ranges are as follows: copper, 161-181 ppm; nickel, 71-83 ppm; and aluminum, 26 ppm.
10 The 2,4-D waste from the Vertac site is dioxin-listed waste, assigned USEPA's hazardous waste code F023.
11 The 2,4,5-T waste is a dioxin-listed waste, assigned USEPA hazardous waste code F020.
These wastes are incinerated. Consequently, these relatively high copper concentrations, which stand in stark contrast to USEPA's maximum reported concentration of 3.6 ppm (Berlow and Jones 1988), are particularly important given the recognized role of this metal in catalyzing dioxin formation at elevated temperatures. (Vogg et al. 1987; Gullett et al. 1990)
2.1.3 Greenpeace Characterization of Probable K020 Wastes
At Borden Chemical's EDC/VCM facility in Louisiana and Formosa Plastic's Texas facility, Greenpeace collected three samples from containers which, based on their labels and locations, are most probably K020 wastes. (For analytical data sheets and more detailed sample descriptions, see Appendix 11.)
The one sample analyzed for dioxin, the Formosa waste sample, carried a total PCDD/PCDF concentration of 761,623 ppt, or 3,191.2 ppt TEQ. As with the earlier K019 samples, this dioxin concentration falls within the range of those estimated for the dioxin-listed waste remaining after Agent Orange manufacture.
Although the other two samples were not analyzed for dioxin, both contained indicator chemicals. One sample from the Borden facility contained two dioxin indicator chemicals, tetra- and pentachlorobenzene, while the other contained one, tetrachlorobenzene. As reported by Evers (1989) the presence of these chemicals suggests that dioxin is also present.
The one sample analyzed for PCBs contained eight of the ten PCB congeners analyzed at a total concentration of 214,700 ppt.
Due to interferences, metals results were obtained for only one of these samples. Eight of the ten metals analyzed were present above limits of detection. Those at highest concentrations were as follows: aluminum, 9,951 ppm; manganese, 113 ppm; and chromium, 52 ppm.
2.1.4 USEPA's Regulation of K019 and K020 Wastes
As discussed earlier, USEPA set no treatment standards for dioxin or PCBs in K019 (heavy ends from EDC distillation) and K020 wastes (heavy ends from VCM distillation). Greenpeace found the number of isolatable chemicals in these wastes to range from 50 to more than 200. (Appendices 10 and 11) However, USEPA set treatment standards for only 11 chemicals in the K019 waste and three chemicals in the K020 waste, as listed below:
Regulated Constituents of K019 and K020 Wastes from EDC/VCM Production
NA: Not applicable.
USEPA selected these chemicals and set their respective treatability standards based on these crucial assumptions: (Berlow and Jones 1988)
- If, following treatment by the best demonstrated available technology, the concentrations of these specific chemicals in the treated waste do not exceed these standards, all other chemicals in the treated waste will also be below levels of concern; and
- If all chemicals in the treated waste are below levels of concern, the treatment of K019 and K020 wastes by the selected technology and the subsequent deposition of the treated waste in a hazardous waste landfill affords adequate protection of public health and the environment.
The first assumption may be true for some of the chemicals with some waste treatment technologies; it is not true for incineration, the technology designated by USEPA as BDAT for the K019 and K020 wastes.
In promulgating the BDAT regulations for K019 and K020, USEPA erroneously assumed that all chemicals in these wastes having boiling points below those of the regulated constituents will be destroyed more thoroughly during incineration than the regulated constituents. This incinerability ranking system based on boiling points is not supported by scientific fact and is, in fact, contrary to USEPA's regulations and guidelines for hazardous waste incinerators. Historically, USEPA has recommended an incinerability ranking system based on heats of combustion. (Mitre 1983) Currently, the Agency also recommends an incinerability ranking system based on thermal stability under oxygen-starved conditions. (USEPA 1989)
Like USEPA's first assumption, the second assumption is not true for incineration In evaluating incineration, USEPA failed to assess stack emissions of constituents of concern that escaped destruction; other unburned waste components; and products of incomplete combustion (PICs), such as dioxin and dioxinlike chemicals. The Agency also failed to consider the deposition of other unburned waste components and PICs, such as in dioxin and dioxin-like chemicals, in scrubber water and ashes. (Berlow and Jones 1988)
Without this information, USEPA's evaluation of incineration is so incomplete as to be meaningless. Moreover, in the absence of these data, it was impossible for the Agency to assess the potential impacts on public health and the environment of designating incineration as BDAT for K019 and K020 wastes.
In summary, USEPA designated rotary kiln incineration as BDAT for K019 and K020 wastes without considering those factors that are most crucial to an adequate evaluation of this technology and to the protection of public health and the environment. In selecting constituents of concern in K019 and K020 wastes, USEPA omitted those constituents of greatest concern - PCBs and dioxin - despite having evidence that PCBs were present and dioxin was extremely likely to be present, as discussed in other sections of this report.
2.1.5 F024 Wastes: USEPA Bows to Industry Pleas to Ignore Dioxin
In 1988, USEPA proposed regulating dioxin - polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) - in F02412 wastes, which are process wastes, including, but not limited to, distillation residues, heavy ends, tars, and reactor clean-out wastes, from the production by free radical catalyzed processes of chlorinated aliphatic hydrocarbons having carbon chain lengths varying from one to five. Shown below, along with their respective concentration ranges in F024 wastes, are the PCDD/PCDFs listed by the Agency as constituents of concern: (Below and Vorbach 1988)
12 As described above, USEPA's description of F024 waste is sufficiently broad to be applied to the same wastes that are classified as K019 and K020. Consequently. EDC/VCM producers sometimes assign waste code F024 to the same wastes that are, at other times and other facilities, classified as K019 or K020. (See Appendix 6.)
PCDD/PCDFs13 Proposed for Regulation in F024 Waste
In comparison to the maximum dioxin concentration shown above, an F024 sample collected by Greenpeace was found to have a total PCDD/PCDF concentration more than ten times greater1,248,000 ppt, as discussed in Section 2.1.6.
In 1990, USEPA revised the F024 regulations in response to complaints that "treatment facilities that had previously treated F024 were refusing to do so because the treatment standards for F024 included standards for chlorinated dibenzodioxins and dibenzofurans." USEPA offered the following rationale for its actions: (Kinch and Vorbach 1990)
"The concentration-based treatment standards that were promulgated for the chlorinated dibenzodioxins and dibenzofurans in F024 (54 FR 26615) may hinder effective treatment because of the refusal of treatment facilities to accept these wastes due to the analytical costs to determine compliance with the treatment standards for these constituents and the perceived stigma of managing wastes containing chlorinated dibenzodioxins and dibenzofurans."
USEPA deleted the chlorinated dibenzo-p-dioxins and dibenzofurans from the F024 list of constituents of concern despite being "unable to select an appropriate surrogate which would ensure adequate treatment of these constituents." The Agency justified the deletions as follows: (Kinch and Vorbach 1990)
"The Agency believes these revisions will allay treatment firms' concerns with accepting F024;" and
"Additionally, the revised standards will reduce the analytical costs associated with compliance with this rule."
13 It is important to note that no data were presented describing the concentrations in F024 wastes of the other PCDD/PCDFs on USEPA's BDAT Constituent List (2,3,7,8-tetrachlorodibenzo-p-dioxin and tetrachlorodibenzo-p-dioxins) or of the following congeners excluded from that list: heptachlorodibenzo-p-dioxins, heptarhlorodibenzofurans, octachlorodibenzo-p-dioxin and octachlorodibenzofuran
Vista Chemical Company officials complained to USEPA that, given the company's limited ability to store its F024 waste and an alleged lack of available treatment capacity, "... Vista would eventually be forced to shut the Vinyl Chloride Monomer unit down ..." (Vista 1990) Obviously, this company categorizes certain waste from VCM production as F024.
Other companies and organizations that successfully argued against the regulation of dioxin in F024 waste include Vulcan Chemicals, the Hazardous Waste Treatment Council, BFGoodrich Company (Geon Vinyl Division), the Vinyl Institute, Dow Chemical Company, PPG Industries, Rollins Environmental Services, and Georgia Gulf Corporation. (Rosengrant and Craig 1990)
In the 1990 revised regulations, USEPA identified the regulated constituents of concern for F024 non-wastewater and their treatability standards as shown below. (Kinch and Vorbach 1990)
Regulated Constituents of Concern for F024 Non-wastewaters
2.1.6 Greenpeace Characterization of F024-Containing Wastes
Greenpeace obtained four samples identified as F024 wastes from the Georgia-Gulf and PPG Industries EDC/VCM facilities in Louisiana, and from the Dow Chemical EDC/VCM facility in Texas. (For more complete descriptions of the samples and their analytical data sheets, see Appendix 12.)
One of these samples came from a tank containing F024 waste located at the Georgia Gulf EDC/VCM facility in Plaquemine, Louisiana. This sample had a total PCDD/PCDF content of 1,248,000 ppt. Expressed as TEQ, the dioxin content was 19,978 ppt TEQ, more than 55 percent of which can be attributed to 1,2,3,4,7,8-hexachlorodibenzofuran. Seven of the ten PCB isomers for which analyses were performed were detected in this sample with a total concentration of 118,610 ppt.
Two of the other three samples contained all five of the chemicals identified as dioxin indicators (Evers 1989). The third sample contained four of the five. The presence of these indicator chemicals suggests quite strongly that dioxin is present as well.
In contrast to USEPA's decision to regulate nine chemicals in F024 waste, Greenpeace identified from 97 to 172 organic compounds in these waste samples.
Two samples contained detectable levels of all ten metals for which analyses were conducted; the third sample contained detectable levels of eight of these metals. In the two samples, aluminum was the most predominant metal, with a concentration range of 618-862 ppm; in the third sample, manganese was the most predominant at 4,362 ppm.
Copper, the metal known to catalyze dioxin formation at elevated temperatures (Vogg et al. 1987; Gullett et al. 1990), was present in two samples in the relatively high range of 235-545 ppm. It is also noteworthy that nickel concentrations in two samples ranged from 59 to 470 ppm.
2.1.7 Greenpeace Characterizations of K016 Waste from EDC/VCM Facilities
While gathering samples of various materials from U.S. EDC/VCM facilities, Greenpeace collected and analyzed three samples taken from containers labeled as K016 waste. The containers were located at Vulcan Chemical's Louisiana EDC/VCM facility. In USEPA's hazardous waste codes, K016 waste is defined as follows: (USEPA 1993)
"K016: Heavy ends or distillation residues from the production of carbon tetrachloride."
However, as noted below, the labels on all three containers from which these samples were taken indicate that this K016 waste was related to the production of perchloroethylene, not carbon tetrachloride. As described in Section 1.0, some facilities, such as ICI (1994), use VCM waste as the feedstock for perchloroethylene production.
One of the three samples contained three dioxin indicator chemicals (Evers 1989), while the other two samples contained two dioxin indicators. The presence of these chemicals suggests that dioxin is present as well. The likelihood that these K016 samples contain dioxin is also supported by the detection of dioxin at concentrations ranging from 30,000 to 90,000 ppt TEQ in "Heavies" from ICI's production of perchloroethylene and trichloroethylene by oxychlorination of VCM waste. (ICI 1994)
Among the ten metals for which analyses were conducted, these samples contained detectable levels of eight to nine metals. In two samples, the most
predominant metal was aluminum at concentrations ranging from 11,162 to 25,844 ppm. In one sample, nickel was the most predominant with a concentration of 1,324 ppm. All three samples contained chromium in concentrations ranging from 41 to 97 ppm. Copper was also present at the relatively high concentration of 178 ppm in one sample.
For a more detailed description of these samples as well as analytical data sheets, see Appendix 13.
2.1.8 Greenpeace Characterization of Other Wastes From EDC/VCM Facilities
Following are the descriptions and general analytical results for samples of other wastes of interest that were collected from various EDC/VCM facilities in the U.S. (For more complete sample descriptions and analytical data, see Appendix 14.)
PU4013: This sample was absorbent material from a barrel in a hazardous waste storage area at the Vulcan Chemical EDC/VCM facility in Louisiana.
Two of the five dioxin indicator chemicals were found in the sample, suggesting that dioxin is present as well. Seven of the ten metals analyzed were present above limits of detection. The two present at highest concentrations were aluminum, 183 ppm, and zinc, 125 ppm.
PU4019: This sample was taken from a container labeled "U077 General Plant" at Vulcan Chemical's EDC/VCM facility in Louisiana.
Four of the five identified dioxin indicator chemicals (Evers 1989) were found in this sample. This suggests strongly that dioxin is also present.
This sample contained detectable levels of nine of the ten metals analyzed. The metals present at highest concentrations were aluminum, 18,455 ppm; manganese, 462 ppm; zinc, 378 ppm; nickel, 60 ppm; copper, 37 ppm; chromium, 30 ppm, and lead, 24 ppm.
PU4021: This sample was taken from a container labeled "U077-D028" at Vulcan Chemical's EDC/VCM facility in Louisiana.
This sample contained all five of the identified dioxin indicator chemicals. Consequently, dioxin is, in all probability, present as well.
Nine of the ten metals analyzed were detected in this sample. Those present at highest concentrations included nickel, 1,872 ppm; aluminum, 1,609 ppm; copper, 782 ppm; manganese, 376 ppm; zinc, 260 ppm; and chromium, 54 ppm.
PU4024: This sample was collected at Vista Chemicals in Louisiana, from a container labeled to indicate that it contained EDC waste.
Two dioxin indicator chemicals were detected in this sample, suggesting that dioxin is present as well.
Nine of the ten metals analyzed were present above limits of detection. Those at the highest concentrations were aluminum, 4,085 ppm; zinc, 861 ppm, nickel, 96 ppm; chromium, 76 ppm.
PU4025: This sample was collected at the Vista Chemicals facility in Louisiana, from waste that included cracking furnace coke, according to the label.
The dioxin indicator chemical, hexachlorobutadiene, was detected in this sample, suggesting that dioxin is present.
All ten of the metals analyzed were detected. Those present at higher concentrations included nickel, 3,149 ppm; zinc, 2,553 ppm; aluminum, 1,868 ppm, chromium, 556 ppm, copper, 412 ppm-, manganese, 212 ppm; and titanium, 88 ppm.
PU4026: This sample, which was taken from a container labeled "U211, D043, D022, D019, Hazardous Waste Solid N.O.S., carbon tetrachloride, NA 3077 III, RQ U211, D043, D022, D0191, ERG 31." The container was at the Vista Chemical EDC/VCM facility in Louisiana.
The detection during the general screening analysis of the dioxin indicator, HCB, suggests that dioxin is present in this sample.
Detectable levels of nine of the ten metals analyzed were present in this sample. The metals present at highest concentrations were zinc, 970 ppm; manganese, 838 ppm; aluminum, 172 ppm; lead, 119 ppm; copper, 66 ppm, and nickel, 79 ppm.
PU4028: This sample was a hazardous waste from PPG Industries in Louisiana.
Two of the five dioxin indicator chemicals were identified in this sample, suggesting that dioxin is present as well.
All ten of the metals analyzed were present above detection limits. Those at highest concentrations were copper, 271 ppm; manganese, 34 ppm; and nickel, 25 ppm.
PU4031: This sample from the BFGoodrich - Geon facility in Texas contained the following wastes: "D032, D034, K019, K020, D039, D078."
The presence in this sample of four of the five dioxin indicator chemicals (Evers 1989) indicates that dioxin is, in all probability, also present.
Nine of the ten metals analyzed were detected in this sample. Those present at highest concentrations were aluminum, 5,834 ppm; zinc, 397 ppm; manganese, 118 ppm; and copper, 23 ppm.
PU4037: This sample was filter cake from Occidental Chemical's EDC/VCM facility in Texas.
This sample contained one dioxin indicator chemical, suggesting that dioxin is present as well.
Nine of the ten metals analyzed were present above limits of detection. Those at highest concentrations were aluminum, 456 ppm, zinc, 274 ppm-, manganese, 152 ppm; copper, 117 ppm; and nickel, 51 ppm,
PU4039: This sample was a solid hazardous waste from Occidental Chemical in Texas.
The presence of one of the dioxin indicator chemicals in this sample suggests that dioxin is present as well.
Nine of the ten metals analyzed were present at concentrations above limits of detection. Those at highest concentrations were zinc, 179 ppm and nickel, 43 PPM.
PU4041: This sample was collected at Occidental's EDC/VCM facility in Ingleside, Texas, from a drum labeled "Hazardous Waste Code # CC-047."
The identification during the screening analysis of two of the five dioxin indicator chemicals suggests that dioxin is present in this sample.
Eight of the ten metals analyzed were present in concentrations above the detection limit. The metals present at highest concentrations were aluminum, 1,790 ppm, zinc, 73 ppm; copper, 52 ppm, titanium, 29 ppm, and chromium, 22 ppm.
2.1.9 Greenpeace Characterization of Sediment Samples Collected Downstream from EDC/VCM Facilities
Greenpeace collected and analyzed one sediment sample taken downstream from the outfall of BFGoodrich - Geon facility in Texas and one sample taken downstream from Dow Chemical's facility in Texas. (For more complete sample descriptions and analytical data, see Appendix 15.)
The sample taken two meters downstream from outfall 001-002 of the BFGoodrich - Geon facility had a total PCDD/PCDF content of >2,911 ppt, or 15.4 ppt TEQ. By comparison, USEPA (1994a) reported North American sediments to have an average total dioxin content of 560 ppt, or 3.91 ppt TEQ.
Nine of the ten metals analyzed were present in concentrations above the limits of detection. Those present at highest concentrations were aluminum, 12,474 ppm; manganese, 92 ppm; zinc, 49 ppm; chromium, 22 ppm; lead, 13 ppm; and nickel, 11 PPM.
The sample taken from the center of the effluent canal near wastewater discharge outfall 001 of Dow Chemical's Oyster Creek Facility in Texas, was found to contain hexachlorobenzene, which suggests that dioxin is also present (Evers 1989).
Nine of the ten metals analyzed were present in measurable concentrations. Those metals present at highest concentrations were aluminum, 14,131 ppm, manganese, 204 ppm; zinc, 132 ppm; copper, 48 ppm; nickel, 45 ppm; and chromium, 25 ppm.
2.2 Dioxin and PCBs in Other USEPA-Coded Hazardous Wastes From Organochlorine Production
The presence of dioxin and PCBs in K019, F024 and, in all probability, K020 wastes suggests strongly that these chemicals also occur in wastes from the manufacture of other organochlorines, particularly those that are by-products of EDC/VCM production or are produced by similar processes. For example, as discussed in Section 2.1.7, the presence of dioxin indicator chemicals in K016 waste samples collected by Greenpeace suggests strongly that dioxin is also present in this waste.
These findings suggest that dioxin and PCBs also occur in the following wastes,which USEPA has aggregated into a single treatability group, based on "their similar physical and chemical characteristics," after examining, among other things, "the sources of the wastes, [and] the specific similarities in waste composition:" (Berlow and Jones 1988)
K016: Heavy ends or distillation residues from the production of carbon tetrachloride
K018: Heavy ends from the fractionation column in ethyl chloride production
K019: Heavy ends from the distillation of ethylene dichloride in ethylene dichloride production
K020: Heavy ends from the distillation of vinyl chloride in vinyl chloride monomer production
K030: Column bottoms or heavy ends from the combined production of trichloroethylene and perchloroethylene
According to USEPA, K016, K018, K019, K020, and K030 wastes are "all still bottoms generated by similar processes: the chlorination or oxychlorination of hydrocarbon feedstocks often at high temperatures and pressures." (Berlow and Jones 1988) K030 waste is especially likely to contain dioxin and PCBs, since trichloroethylene and perchloroethylene are, like EDC/VCM, produced by the oxychlorination and direct chlorination of ethylene dichloride as well as the chlorination of acetylene. (Berlow and Jones 1988) As discussed earlier, ICI (1994) contends that oxychlorination cannot be carried out without the formation of dioxin. Moreover, the listing by Occidental Chemical of its K030 waste as "PCB Contaminated Hydrocarbons, Chlorinated" (USEPA 1991b) attests to the concurrent generation of PCDD/PCDFs during oxychlorination.
As discussed earlier, the report from ICI (1994) attests to the fact that dioxin generation during the production of trichloroethylene and perchloroethylene via oxychlorination is considerably greater than that during VCM production.
In their review of the relationship of chlorine input to dioxin output from various combustion systems, researchers at Princeton University have documented a clear trend: the greater the chlorine content of combusted materials, the greater the dioxin emissions.14 (Thomas and Spiro 1994) Although PCBs are not addressed in this study, this same trend should be evident for these chemicals, which are, like dioxin, thermodynamically stable products of the combustion of chlorine-containing materials in hazardous waste incinerators, municipal waste incinerators, metal sintering facilities, etc. (Fiedler 1993)
14 This correlation can, of course, be obscured or overwhelmed by other factors, such as variations in operating conditions, efficiencies or pollution control systems, etc.
PVC is a primary chlorine donor, both by its content (at least 56 percent by weight chlorine) and by its quantity (some 10.88 billion pounds produced in 1994). (Coeyman 1994) As such, PVC and PVC products are the principal contributors to the formation of dioxin and PCBs in many combustion systems as well as in certain uncontrolled fires.
In USEPA's draft dioxin reassessment, the Agency neither acknowledges PVC as the predominant chlorine donor in the materials burned in various dioxin-emitting combustors and fires nor identifies all of the combustion systems and fires in which PVC and PVC products are burned. However, when these crucial factors are addressed; a simple and effective dioxin-prevention strategy emerges: eliminate chlorine donors, including PVC and other chlorine-containing materials, from combustor feedstocks and prohibit their use in circumstances vulnerable to uncontrolled fires. This strategy has already been recommended by the German Federal Environmental Authority, which has urged that no chlorine- containing fuels be burned in any furnace, from private fireplaces to industrial plants. (Wilken 1994) (For further discussion, see Thornton and Weinberg 1994).
3.1 Dioxin and PCB Releases from Metallurgical Industries
In the dioxin reassessment, USEPA identifies secondary smelters for copper and lead as major sources of dioxin. The Agency does not acknowledge that PVC is the major chlorine donor and, consequently, the primary contributor to dioxin formation in these processes:
- Secondary copper smelters: Large quantities of PVC-coated copper cables, as well as PVC telephone cases and other PVC-laden products are processed at high temperatures.. (Christmann et al. 1989)
- Secondary lead smelters: PVC plastic separators in lead-acid batteries are the major source of the chlorine requisite for dioxin formation. (USEPA 1994a)
- Secondary steel smelters: PVC residues in steel scrap, primarily from automobiles, are sources of the chlorine necessary for dioxin formation. (Aittola 1993)
Steel sintering plants have also been identified as major dioxin sources, with two possible chlorine donors: dusts and slag from the recycling of PVC-containing scrap steel, which occurs even in so-called "primary" steel production, and the use of chlorinated solvents and cutting oils in the production process. (Lahl 1993) It seems reasonable to assume that these same chlorine sources contribute to the emission from sintering facilities of PCBs in the range of nanograms per cubic meter as reported by Fiedler (1994).
PVC has also been identified as contributing to dioxin formation during the recycling of aluminum and zinc. (Aittola 1993)
3.2 Dioxin and PCB Releases from Municipal and Medical Waste Combustors
In their draft dioxin reassessment, USEPA identifies municipal and medical waste incineration as the largest sources of dioxin without identifying or otherwise alluding to the constituents in the wastes that are the primary chlorine donors for dioxin formation. Also, the Agency suggests that PCB emissions from municipal waste incinerators may be attributable to PCB contamination in the refuse, effectively dismissing the probability that, like dioxin, PCBs can be created as products of incomplete combustion. Following this pattern, USEPA does not acknowledge that PCBs may also be emitted from medical waste incinerators. (USEPA 1994a)
Danish authorities report that PVC contributes the vast majority of the available chlorine input to municipal waste incinerators. (Danish EPA 1993) A study commissioned by the German Federal Ministry for Research and Technology calculated that while PVC accounts for only 0.5% of municipal waste by weight, PVC is responsible for about 50% of the total chlorine -organically-bound or ionic - in the trash stream. (Brahms et al. 1989)
For medical waste incinerators, the predominant chlorine donor is the PVC in disposable plastic items (packaging, gloves, blood bags, infusion tubing, etc.). An estimated 9.4 percent of all "red-bag" waste is PVC, while virtually all available chlorine fed to medical waste incinerators comes from PVC. (Marrack 1988)
Numerous reports have found that burning PVC results in dioxin formation. (Thiessen 1989; Christmann et al. 1989; Kopponen et al. 1992) PCBs have also been identified among the combustion by-products of polyvinylidene dichloride (PVDC) (Blankenship et al. 1994), which is an organochlorine plastic similar to PVC. PVDC is most widely used as a copolymer with PVC in plastic films, filaments, etc. (Parker 1989)
Various studies have identified a relationship between PVC and dioxin formation from municipal and medical waste incinerators:
- Ozvacic et al. (1990) have reported a direct relationship between the quantity of PVC burned and dioxin emissions.
- The Danish EPA found that doubling the PVC content of an incinerator's wastefeed increases dioxin emissions by 34 percent. (Danish EPA 1993)
"These new experiments demonstrate clearly a relationship between the content of PVC in household waste and dioxin formation in waste incinerators. On the basis of these experiments, there is no reason to reconsider present policies regarding PVC applications: the main feature of these policies is that for PVC applications for which no feasible system of recycling and reuse can be established, the use of more environmentally sound alternative materials is to be preferred." (NME 1994)
In a 1993 study by the U.S. Department of Energy, the addition of PVC at 10 percent by weight to a waste mixture, which increased waste chlorine content from 1.5 percent to 6 percent by weight, had the following results: (Bums 1993)
- Increases in the relative proportion of submicron particulates emitted (at 1832 F, 26 weight, percent with no PVC versus 69 weight percent with PVC; at 1600 F, 26 weight percent with no PVC versus 60 weight percent with PVC);
- Increased emissions of volatile organic compounds (VOCs); and
- Increased emissions of dioxin - PCDD/F emissions of 35,900 nanograms per kilogram (ng/kg) of waste feed including PVC at 10 percent by weight versus 3760 ng/kg with no PVC (or, expressed as TEQ, 430 ng/kg with PVC versus 46.5 ng/kg with no PVC).
As mentioned earlier, the association between chlorine content of combusted materials and dioxin emissions was confirmed by Princeton University researchers. (Thomas and Spiro 1994) Similarly, the 1993 Dutch study mentioned above linked increased HCI emissions with increased dioxin emissions. (Boerekamps-Kanters and Louw 1993) In its draft dioxin reassessment, USEPA suggests that some portion of the dioxin formed in incinerators may occur after HCI is formed and released as an intermediate compound. (USEPA 1994a)
This theory is further corroborated by a 1993 study conducted by the Clean Combustion Technology Laboratory at the University of Florida, where researchers found "statistically significant relationships between HCI emissions (a surrogate for PVC in the waste) and the emissions of a number of chlorinated organic compounds." The researchers found: (Wagner and Green 1993)
- 'There is a direct dependence of HCI emissions on the level of PVC in the waste;
- 'There is a strong dependence of chlorinated benzene emissions on HCI emissions;; and
- 'There is almost certain evidence of nonlinear relations existing between chlorinated hydrocarbon emissions and CO and temperature."
The researchers went on to conclude as follows:
'These results, contrary to the prevailing opinion, lead to the physically reasonable conclusion that decreases in the levels of organically bound chlorine in the input leads to decreases in chlorinated organic emissions ....Thus we are convinced that, when all other factors are held constant, there is a direct correlation between input PVC and output PCDD/PCDF and that it is purposeful to reduce chlorinated plastics inputs to incinerators." (Wagner and Green 1993)
3.3 Dioxin Releases from Miscellaneous Combustion Facilities and Device
EPA identifies the industrial and municipal burning of wood as a large dioxin source. (EPA 1994a) Since much of the wood incinerated is construction debris, and markets for PVC siding, window frames, wallpaper and flooring are increasing rapidly due to the cost advantages and universal applicability of PVC, this source is likely to increase in the future.
According to recent studies conducted for the German EPA, the primary causes of dioxin emissions from wood combustion are PVC residues on scrap wood. (Wilken 1994) The German Federal Environmental Authority subsequently urged that no chlorine-containing fuels be burned in any furnace, from private fireplaces to industrial plants. (Wilken 1994)
PVC scrap is also combusted in household and commercial furnaces. Grocery stores and other businesses incinerate vinyl pallet wraps and discarded PVC and PVDC/PVC packaging, sometimes in dedicated burners and sometimes in open pits.
3.4 Dioxin Releases from Uncontrolled Fires
Dioxin emissions from the burning of PVC in uncontrolled fires have long been a matter of record, as documented below. However, USEPA's draft dioxin reassessment document fails to consider such fires, although they are one of the largest potential sources of dioxin releases. Scientists have made the following observations:
"Obviously, in many incineration and pyrolysis processes as well as fires, PVC can be considered a main source for the formation of dioxins and furans." (Christmann et al. 1989) and:
"...in fire accidents involving chlorinated organic plastics considerable concentrations of PCDF/Ds could be detected in now more than 200 samples." (Theisen et al. 1989)
Tests performed by other German scientists have also documented significant levels of PCDD/Fs as a result of accidental fires: (Theisen et al. 1989)
"Samples from real fire accidents as well as from laboratory combustion tests were analyzed for polychlorinated dibenzofurans (PCDFs) and dibenzo-p-dioxins (PCDDs). The results demonstrate that in case PVC-containing materials take part in combustion processes PCDF/Ds can be found in the decomposition products in considerable concentrations."
These laboratory combustion tests showed that "even at temperatures around 350 degrees Celsius dioxins and furans are formed by the charring of (PVC) cable coatings in amounts that exceed the German statutory regulation for dangerous substances considerably." (Theisen et al. 1989)
Accidental fires involving PVC can be split into two categories: fires which occur at sites where PVC, or PVC products, are manufactured or processed and, potentially, all other fires.
3.4.1 Accidental Fires at Facilities Manufacturing or Processing PVC and/or PVC Products
Significant levels of dioxin have been documented in samples taken in Europe and Canada following accidental fires at plants involved in the manufacture or processing of PVC and/or PVC products:
- Dioxin concentrations of 13.7 ppb TEQ were found by the German Environmental Protection Agency (UBA) in residues after a fire occurred at the Microplast PVC recycling company in Lengerich, Germany. Agriculture in the area was significantly affected: cabbages grown 600 meters away from the plant site showed an 88-fold increase in dioxin concentration after the fire. (UBA 1992)
- Dioxin concentrations as high as 87 ppt were measured in fire residues after a fire destroyed the Belgian Euromat plant, a producer of PVC granules for cars, cable, shoes, and medical products. Approximately 100 tons of PVC burned.
- Dioxin concentrations of 18 ppb TEQ were documented in fire residues of the Plastibec plastics plant in Canada; approximately 15,000 kg of PVC was in the factory at the time of the fire. Soil near the plant contained dioxin and furan concentrations of 0.55 ppb. Plastibec is the largest producer of extruded vinyl window frames. (QUO 1993)
- TCDD equivalents (using the Nordic model) ranging from 1.3 to 305 nanograms per square meter (ng/m2) were measured in snow samples after a fire at a plastic carpet company near Umea, Sweden, in 1987. Approximately 200 tons of pure PVC plastic and 500 tons of plastic carpets were destroyed by the fire. Samples were taken at various distances from 10 meters to 1500 meters downwind from the plant. Wipe samples were also taken two days after the fire, in parts of the plant which had been filled with smoke and also from the outside of a home downwind from the plant. Measured TCDD equivalents ranged from <4 ng/m2 to 180 ng/m2_15 The wipe sample from the firewall which separated the warehouse (where the fire took place) from the factory yielded dioxin equivalents higher than 50 ng/m2, the level which requires action by Swedish authorities. (Marklund et al. 1989)
3.4.2 Accidental Fires in Commercial Buildings, Homes and Stockpiled Material
The accidental burning of commercial buildings, homes and stockpiled materials may constitute a far larger source of dioxin than fires at PVC-related manufacturing sites. For example, German scientists have offered the following observation:
"...also in fires in public and industrial buildings, where PVC materials had been involved, significantly high levels of dioxins and furans were found." (Christmann et al. 1989)
- German scientists have identified concentrations of up to 10,000 ng/m2 on surfaces from accidental fires with large quantities of PVC present, and concentrations of up to 200 ng/m2 after "normal" fires in homes and offices. Soot samples from a fire at a German kindergarten, where substantial quantities of PVC construction material burned, revealed dioxin concentrations as high as 45 ppb TEQ. (Fiedler et al. 1993)
- PCDD/F concentrations in soot samples taken after a fire occurred at the Heinrich Heine Allee underground station in Dusseldorf, Germany, were as much as five times greater than the German level requiring cleanup action of 10 ng/m2 TEQ. Concentrations were attributed to the burning of PVC cable casings. (Lindert 1994)
- Dioxin and furan concentrations of 3.4 ng TEQ/gram were measured in soot samples after a fire in a New York University lecture hall building. The material that burned was determined to be floor scrubbing pads, trash bags, PVC floor tiles, quaternary ammonium chloride cleaning solution packets, chlorinated cleaner, paper products, and various types of furniture. In particular, intense heat melted several plastic chairs, a plastic waste container, plastic coverings, and electrical wiring. Authors of the study assert that no PCB or chlorobenzene pyrolysis occurred, ruling out PCDF/D formation due to these substances. (Deutsch and Goldfarb 1988) Given that PVC is the second most common type of plastic used today, and the most common of chlorinated plastic, it can be inferred that much, if not all, of the plastic involved in this fire was PVC-based. Comparing these results to control buildings analyzed in previous tests, the authors found:
15 These figures represent the sum of 2,3,7,8-tetrachlorodibenzofuran and 2,3,4,7,8-pentachlorodibenzofuran.
"The concentrations of PCDDs and PCDFs in a few of the samples produced in the fire were significantly higher than those reported in the literature from uncontaminated buildings. It can therefore be concluded that PCDDs and PCDFs were formed during the fire." (Deutsch and Goldfarb 1988)
According to one industry spokesman roughly three quarters of all PVC manufactured goes into the building and construction market for uses such as piping, siding, window frames, wallpaper, cabling, flooring, and other uses. (Reisch 1994) As a result, any fine in a modem building is likely to be a source of dioxin.
The contents of many buildings are also potential sources of dioxin. Furniture, appliance and computer housings, children's toys, packaging, and hundreds of other consumer products present in a typical home or place of business are PVC-based.
In 1993, there were approximately 470,000 residential fires across the United States, and over 151,000 fires in nonresidential buildings, totaling over 620,000 accidental fires in 1993. The National Fire Protection Association has determined that PVC was the material first ignited in over 15,000 of the 637,000 fires occurring in 1992. (NFPA undated) This number, of course, is only a small subset of the total number of fires (potentially all 620,000 in 1993) that involved PVC and therefore resulted in the formation and subsequent release of dioxin to the environment.
The extent of the problem may be quite significant and should be further defined, as recommended by German scientists:
"Given the fact that fires involving a source of chlorine and hydrocarbons can result in PCDD and PCDF formation, it may be prudent to test for these in a few samples of soot deposited in the immediate vicinity of any fire involving synthetic materials." (Deutsch and Goldfarb 1988)
It is impossible to control or capture dioxin emissions from accidental fires involving PVC. The only possible policy to eliminate dioxin releases from uncontrolled fires is to phase out the dioxin precursor, PVC.
As listed in Appendix 16, well-over 100 communities in European countries now have PVC bans or phaseout policies in place for PVC in public buildings due to concern over toxic releases resulting from accidental fires involving PVC, as well as persistent toxic pollution resulting its production, use, and ultimate disposal.
Substantial dioxin releases from the accidental burning of stockpiled waste containing PVC have also been documented. For example, several fires involving automobile fluff stockpiled at automobile reclamation facilities have occurred. In one fire, between 39,000 and 48,000 pounds of the fluff were burned. (Ryan and Lutes 1993)
Laboratory studies by USEPA indicate that burning one kilogram of fluff generates air emissions of approximately 0.0072 grams of PCDD/PCDFs16. The Agency estimates that approximately 2 billion pounds of automobile fluff are generated annually. (Ryan and Lutes 1993) Obviously, if all this fluff were burned in open fires, more than 31 kilograms of these PCDD/PCDFs would be released into the air. This large quantity of automobile fluff could be expected to contribute significantly to the national dioxin burden even if burned in controlled combustors.
16 Analysis of these air emissions did not include octachlorodibenzo-p-dioxins and octachlorodibenzofuran, so emissions of total PCDD/PCDFs can be expected to be considerably higher.
All major stages of the PVC life-cycle are associated with the generation of significant quantities of dioxin. As confirmed by an industry source, one particularly important dioxin-generating process is the production of EDC by oxychlorination.
PCBs are also described by industry sources as inescapable by-products of PVC manufacturing processes. However, the extent of their dispersal into the environment via vent gases, wastewater effluents and sludges, process wastes, and combustor emissions is even less well documented than that of dioxin.
USEPA has followed a strategy that minimizes and obscures to the point of uselessness its regulation of the PVC industry's generation and release of PCBs. Moreover, USEPA made a fully deliberated decision to forego regulating the industry's generation and release of dioxin. After proposing regulations that would have partially addressed the industry's dioxin output, the Agency turned, under industry pressure, to delete dioxin from the list of chemicals to be regulated.
In the dioxin reassessment draft, USEPA makes no mention of its own documentation of dioxin generation during PVC manufacture. Indeed, the Agency offers little acknowledgment in that document of the magnitude of this source of dioxins and PCBs. (USEPA 1994a) Similarly, the Agency has failed to address adequately the other constituents of PVC process wastes described in this document.
Greenpeace Response to USEPA Inaction
After receiving no response from USEPA to repeated requests for an investigation of dioxin releases from PVC manufacture, Greenpeace collected 51 samples of wastes and other materials associated with EDC/VCM production. Twenty-five of these samples are discussed in this report.
With the analysis of four of these samples for dioxin content, Greenpeace has documented the presence of dioxin concentrations in some of these materials that rival those in dioxin-listed wastes from the manufacture of Agent Orange. Although no direct dioxin determinations were made for the remaining 21 samples, Greenpeace analyses revealed the presence of certain chemicals which, by their presence, suggest that dioxin is also present in all of these samples.
One sediment sample collected and analyzed by Greenpeace contained a substantial concentration of dioxin. This finding supports, in part, extrapolations from data from industry and other sources that suggest that EDC/VCM facilities may be discharging dioxin into U.S. waterways at a rate that rivals that from all pulp and paper mills.
Greenpeace analyses also found these samples to contain copper, nickel, chromium and other metals at unexpectedly high concentrations. Copper is known to catalyze the generation of dioxin during incineration, which has been designated by USEPA as the disposal method of choice for many EDC/VCM wastes.
PVC: The Chlorine Donor to Dioxin Formation
Besides PVC manufacture's direct contributions to the national dioxin burden, PVC itself is the primary chlorine donor for dioxin generation and release from most of the dominant dioxin sources identified by USEPA (1994a). The sum of dioxin formation throughout the life-cycle of PVC clearly makes this one product the largest single contributor to the nation's dioxin burden.
The two largest dioxin sources identified by USEPA are medical and municipal waste incinerators. Among the materials burned in these incinerators, PVC is the major chlorine donor and, consequently, the primary contributor to dioxin formation.
Enormous quantities of chlorine-rich process wastes from the synthesis of EDC/VCM are incinerated in on-site or commercial hazardous waste incinerators, producing and releasing significant additional quantities of dioxin. Moreover, Greenpeace's sampling program found significant concentrations of copper in EDC/VCM wastes; copper is known to catalyze the formation of dioxin in incinerators.
PVC is the primary chlorine donor for dioxin formation in recycling facilities for copper, steel, lead and other metals. These secondary smelting facilities -- which receive PVC in residues from automobiles, cables, electronic equipment, and batteries have been identified by USEPA as major dioxin sources.
PVC is the primary chlorine donor for dioxin formation and emission from the burning of scrap wood in domestic woodstoves and industrial wood burning furnaces, according to European studies. USEPA has identified these sectors as major dioxin sources.
Accidental home and building fires may well be one of the largest single sources of dioxin. PVC products used in building construction and those in use within such buildings are the primary chlorine donors for the large quantities of dioxin created during the accidental burning of homes, schools, office buildings, and industrial facilities. The contribution of such fires to the national dioxin burden has not been estimated. However, given the hundreds of thousands of fires that occur each year, the dioxin loadings are certain to be significant.
Essential Steps in Reducing the National Dioxin Burden
In USEPA's dioxin reassessment, the Agency establishes Dearly that dioxin is extraordinarily toxic, persistent, and bioaccumulative and that dioxin is now distributed globally in the environment, food chain, and human tissues. USEPA has found that current "background" levels of dioxin are already within the range at which health damage is known to occur in the laboratory. These findings support the call for a national dioxin prevention program to eliminate on a rapid timetable the generation and release of dioxin into the environment.
USEPA's draft dioxin reassessment offered little insight into the formation and release of dioxin and PCBs during PVC manufacture. However, among those dioxin sources that were identified by the Agency, PVC provides the majority of the chlorine that is available for dioxin formation. Consequently, a PVC phaseout must be given priority for a national dioxin prevention program. Some important first steps are as follows:
- Prohibition of the oxychlorination process for the production of EDC and other chemicals.
- Prohibition of new facilities or capacity expansions for the production of EDC/VCM and PVC.
- Modification of existing permits for EDC/VCM plants to bring generation and releases of dioxin to zero.
- Classification of relevant wastes from EDC/VCM production as dioxin-listed and PCB-containing wastes, subject to all regulatory requirements that have been revised to reflect the inadequacies and limitations of incineration as a treatment technology.
- Prohibition of incineration of chlorine-rich wastes and/or wastes from EDC/VCM production that contain dioxin and PCBs.
- Rapid phaseout of PVC uses associated with the most copious dioxin generation, including short-life PVC uses (packaging and disposable products), uses in areas susceptible to fire (construction, appliances and automobiles), and products recycled in smelters (cables and cars).
- A longer-term phaseout of other uses of PVC, with priorities established according to environmental impact and the availability of alternatives.
Because the ultimate phaseout of PVC will have economic impacts in the communities where manufacturing facilities are located, transition planning must be an integral component of any phaseout plan. This process must be guided by participation
from labor, community and other stakeholders and should seek to minimize the economic effects of the transition and insure that costs and benefits are equitably distributed.
For instance, the Oil, Chemical, and Atomic Workers Union has proposed a tax on chlorine and related chemicals; the revenue would be used to encourage reinvestment in affected communities and to provide income protection, continued health care, and meaningful opportunities for higher education and reemployment for workers and their families.
(Aittola et al. 1993) Aittola, J. Paasivirta, J. and Vattulainen. Measurements of Organochloro Compounds at a Metal Reclamation Plant. Chemosphere, 27:65-72.
(Berlow and Jones 1988) Berlow, J.R., and Jones, L. Final Best Demonstrated and Available Technology (BDAT) Background Document for K016, K018, K019, K020, K030. Washington, D.C.: U.S. Environmental Protection Agency. EPA/530-SW-88-031 B. August 1988.
(Berlow and Vorbach 1988) Berlow, J.R., and Vorbach, J. Proposed Best Demonstrated and Available Technology (BDAT) Background Document for F024. Washington, D.C.: U.S. Environmental Protection Agency. EPA/530-SW-89-0171, December 1988.
(Blankenship et al. 1994) Blankenship, A., Chang, D.P.Y., Jones, A.D., Kelly, P.B., Kennedy, I.M., Matsumura, F., Pasek, R., and Yang, G. Toxic combustion by-products from the incineration of chlorinated hydrocarbons and plastics. Chemosphere 28(1): 183196. 1994.
(Boerekamps-Kanters and Louw 1993) Boerekamps-Kanters, J., and R. Louw. Final report of the RUL-VROM project: GFT, PVC, Afvalverbranding en 'Dioxine' (Green Waste Fraction, PVC, Waste Incineration and 'Dioxins'). Leiden: Centre for Chemistry and the Environment. Department of Chemistry, University of Leiden, report number CCESRS 9309, 1993.
(Bowermaster 1993) Bowermaster, J. "A town called Morrisonville," Audubon, July/August 1993, pp. 42-51.
(Brahms et al. 1988) Brahms, E. et al. "Papier Kunststoff Verpackungen. Ene Mengen and Schadstoffbetrachtung." Berlin: UBA Berichte, 1988.
(Brown 1989) Letter from T.G. Brown, PPG Industries, Inc., to Marion Fannaly, Louisiana Department of Environmental Quality. Re: PPG's Water and Sediment Reports. August 15, 1989
(Bums 1993) Bums, D.B. Final Report Consolidated Incineration Facility Metals Partitioning Test (U). Aiken,S.C.: Savannah River Technology Center. WSRC-TR-93 623. August 31, 1993
(CAE 1991) Clean Air Engineering. Report on Trial Bum Testing, Vertac Chemical Corporation Superfund Site, Jacksonville, Arkansas. CAE Project No: 5815-1. Palatine, Illinois. December 5, 1991.
(Christmann et al. 1989) Christmann, W., D. Kasiske, K.D. Kloppel, H. Partscht, W. Rotard. Combustion of polyvinylchloride - an important source for the formation of PCDD/PCDF. Chemosphere 19(1-6):387-392.
(Coeyman 1994) Coeyman, M. Banner year for plastics, with HDPE as volume leader. Chemicalweek 155 (24): 10. December 21/28, 1994.
(Cooper 1993) Letter from Eliot D. Cooper, Waste-Tech Services, Inc., to Kai D. Midboe, Louisiana Department of Environmental Quality. Re: Waste Tech Services, Inc., RCRA Permit No. LAD 981 514 441; Lake Charles Facility. January 8, 1992.
(Cooper 1994) Letter from Eliot D. Cooper, Ecova Corporation, to Glenn A. Miller, Louisiana Department of Environmental Quality. Re: Notice of Deficiency: Review of Class 3 RCRA Permit Modifications, Ecova Corporation - Lake Charles RCRA Incineration Facility Hazardous Waste Permit Number: LAD981514441. June 23, 1994.
(Costner 1992) Costner, P. The incineration of dioxin in Jacksonville, Arkansas: A review of trial bums and related air monitoring at Vertac Site Contractors mobile incinerator, Jacksonville, Arkansas. Washington, D.C.: Greenpeace USA. January 29, 1992.
(Cowfer and Magistro 1983) Cowfer, J.A. and Magistro, A.J. Vinyl Polymers, Vinyl Chloride. Volume 23 of (Eds: Martin Grayson and David Eckroth) Kirk-Othmer Encyclopedia of Chemical Technology. 3rd Edition. New York: Wiley Interscience. 1983.
(Danish EPA 1993) Miljoministeriet Miljostyrelsen (Danish EPA). PVC and Alternative Materials. (English translation). Copenhagen, 1993.
(Deutsch and Goldfarb 1988) Deutsch, D. G. and Goldfarb, T.D. PCDD/PCDF contamination following a plastics fire in a university lecture hall building. Chemosphere 17(12):2423-2431.
(Dow 1981) The Dow Chemical Company. Comments of the Dow Chemical Company Before the United States Environmental Protection Agency In Re: The Office of Toxic Substances, Advanced Notices of Proposed Rulemaking Concerning Polychlorinated Biphenyls. Under cover letter from Michael E. Nevill, Dow Chemical U.S.A., to Document Control Officer, Office of Toxic Substances, U.S. Environmental Protection Agency. November 13, 1981
(Duncan 1982) Letter from R.D. Duncan, PPG Industries, to Joni T. Repasch, Office of Toxic Substances, U.S. Environmental Protection Agency. Re: PPG Industries, Inc., Request for Renewal of PCB Exemption Petition. With Attachment: Petition for Renewal of PCB Manufacturing Exemption. July 21, 1982
(EAF 1989) Environmental Action Foundation. The Real Wrap on Polyvinyl Chloride Packaging. Solid Waste Alternatives Project. Solid Waste Action Paper #e. Available from EAF at (301) 891-1100.
(Evers et al. 1988) Evers, E.H.G., Ree, K.C.M. and Olie, K. Spatial variations and correlations in the distribution of PCDDs, PCDFs and related compounds in sediments from the river Rhine - Western Europe. Chemosphere 17(12):2271-2288. 1988.
(Evers 1989) Evers, E.H.G. De Vorming van PCDFs, PCDDs en gerelateerde verbindungen bij de oxychlorering van etheen. Publ. Vakgroep Milieu - en Toxicologische Chemie. University of Amsterdam 62 pp. 1989.
(Evers et al. 1989a) Evers, E., Suring, K., Olie, K., and Govers, E. Catalytic oxychlorination processes of aliphatic hydrocarbons as new industrial sources of PCDDs and PCDFs. Paper SOU 14, "Dioxin 89", Ninth International Symposium of Chlorinated Dioxins and Related Compounds. Toronto, Ontario, Canada. September 17-22, 1989.
(Evers et al. 1989b) Evers, E.H.G., van Berghem, J.W. and Olie, K. Exploratory data analysis of PCDD and PCDF measurements in sediments from industrialized areas. Chemosphere 19(106):459466. 1989.
(Fiedler 1993) Fiedler, H. Sources of dioxins and furans. In Conference Reports, Dioxin '93, Austria, pp. 60-61. 1993.
(Fiedler et al. 1993) Fiedler, H., O. Hutzinger, and J. Hosseinpour (1993). Analysis and remedial actions following an accidental fire in a kindergarten. Organohalogens 14:19-23.
(Goldman 1994) Memorandum: Lynn Goldman, Assistant Administrator, U.S. Environmental Protection Agency, to Rick Hind, Greenpeace. December 19, 1994.
(GPI 1994) Greenpeace International. PVC: The need for an industrial sector approach to environmental regulation. INDSEC 6/15/NGO.1-Add.1-E. Paris Convention for the Prevention of Marine Pollution. Sixth Meeting of the Working Group on Industrial Sectors. Oslo: 17021 January 1994.
(Gullett et al. 1990) Gullet, B.K., Bruce, K.R., and Beach, L.D. The effect of metal catalysts on the formation of polychlorinated dibenzo-p-dioxin and polychlorinated dibenzofuran precursors. Chemosphere 20:1945-1952.
(Hileman et al. 1994) Hileman, B., Long, J.R., and Kirschner, E.M. Chlorine Industry Running Flat Out Despite Persistent Health Fears. C&EN, 72 (47): 12-26. November 21, 1994
(Hill 1979) R.W. Hill, Senior Counsel, Diamond Shamrock. Letter to Document Control Officer, Office of Toxic Substances, U.S. Environmental Protection Agency. Re: Manufacturing Ban Exemption from PCB. December 17, 1979.
(ICI 1994) ICI Chemicals & Polymers Ltd, "Report to the Chief Inspector HMIP Authorisation AK6039," ICI Chemicals & Polymers Ltd., Safety & Environment Department, The Heath, Runcom, April 27, 1994
(Kinch and Vorbach 1990) Kinch, R. and Vorbach, J. Final Amendment to the Final Best Demonstrated Available Technology (BDAT) Background Document for Wastes from the Production of Chlorinated Aliphatic Hydrocarbons. Washington, D.C.: U.S. Environmental Protection Agency. EPA/530-SW-90-0600. May, 1990
(Kopponen et al. 1992) Kopponen at al., Comparison of cytochrome P4501A1 induction with the chemical composition of fly ash from combustion of chlorine containing material. Chemosphere 24(1992): 391-401.
(Kramlich et al. 1989) Kramlich, J.C., Poncelet, E.M., Charles, R.E., Seeker, W.R., Samuelsen, G.S., and Cole, J.A. Experimental Investigation of Critical Fundamental Issues in Hazardous Waste Incineration. Research Triangle Park, N.C.: U.S. Environmental Protection Agency. EPA/600/2-89/048. September 1989
(Lahl 1993) Lahl, U. Sintering plants of steel industry: the most important thermal PCDD/F source in industrial regions? Presented at Thirteenth International Symposium on Chlorinated Dioxins and Related Compounds. Organohalogens 11:311-314.
(Lewis 1989) Lewis, J. Chemistry Utilized in BFF's Unit Operations for Manufacture of EDC & VCM and Definitions of "Free Radical", "Initiator", and Catalyst". In (L. Rosengrant and R. Craig, authors) Final Response to BDAT Related Comments Document, Volume I-M. Washington, D.C.: USEPA Office of Solid Waste. EPA/530-SW-900610. May, 1990.
(Lindert 1994) Lindert, M. Dusseldorf City Environment Department. See Nach. Chem. Tech. Lab. 42, No. 718, 1994.
(Lowrance 1992) Memorandum: Sylvia K. Lowrance, Director, Office of Solid Waste, U.S. Environmental Protection Agency, to Waste Management Division Directors, Regions I-X, September 22, 1992
(Marklund et al. 1989) Marklund, S., R. Andersson, M. Tysklind, C. Rappe. Emissions of PCDDs and PCDFs from a PVC-fire in Holmsund, Sweden. Chemosphere 18(1-6):10311038.
(Marrack 1988) Marrack, D. Hospital Red Bag Waste: An Assessment and Management Recommendations. JAPCA 38:1305 1311.
(Miller 1993) Miller, A. Dioxin emissions from EDC/VCM plants. Environ. Sci. Technol. 27:1014-1015. 1993.
(Mitre 1983) Mitre Corporation, "Guidance Manual for Hazardous Waste Incinerator Permits," NTIS PB84-100577, July 1983.
(Monsma and Kaerwer 1995) Monsma. D., and Kaerwer, L. Data Quality in the 1991 and 1992 Toxic Chemical Release Inventory: A Case Study. Draft. Takoma Park, MD: Environmental Action Foundation. 1995.
(NATO/CCMS 1988a) North Atlantic Treaty Organization, Committee on the Challenges of Modem Society. International toxicity equivalency factor (ITEF) method of risk assessment for complex mixtures of dioxins and related compounds. Report No. 176.
(NATO/CCMS 1988b) North Atlantic Treaty Organization, Committee on the Challenges of Modem Society. Scientific basis for the development of international toxicity equivalency (I-TEF) factor method of risk assessment for complex mixtures of dioxins and related compounds. Report No. 178.
(Nestrick et al. 1987) Nestrick, T.J., Lamparski, L.L., Crummett, W.B. Thermolytic surface-reaction of benzene and iron (III) chloride to form chlorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 16(4):777790.
(NFPA undated) National Fire Protection Association, Fire Analysis and Research Division, Quincy, Massachusetts. 1980-1992 NFIRS, NFPA Survey.
(NME 1994) Netherlands Ministry of the Environment. Letter to Austrian Ecology Institute. Re: Advertising Code Commission (PVC). January 14, 1994.
(Oehme et al. 1985) Oehme, M., Mano, S., and Bjerke, B. Formation of polychlorinated dibenzofurans and dibenzo-p-dioxins by production processes for magnesium and refined nickel. Chemosphere 18 (7-8): 13791389.
(Ozvacic et al. 1990) Ozvacic, V., Wong, G., Marson, G., Clement, R., Rokosh, D., Suter, S., Horsnell, G., Hipfner, J.C., Bums, S., Corinthios, H., Young, M., Birmingham, B. Biomedical Waste Incinerator Testing Program. Chemosphere 20:1801-1808.
(Parker 1989) Parker, S.P., ed. McGraw-Hill Concise Encyclopedia of Science and Technology. Second Edition. New York: McGraw-Hill Publishing Co. 1989.
(OEM 1993). Quebec Environment Ministry. Resultats d'analyse des sois avoisinants I'usine Plastibec Ltee. 29 June 1993. See also Francoeur Le Devoir, L. "Contamination a la dioxine a Sainte-Therese." Le Devon. 2 July 1993. See also Picard, A. "Quebec fire site seriously contaminated." Toronto Globe and Mail, 3 July 1993, A4.
(Reisch 1994) Reisch, M. Plastics in building and construction. C&E News. May 30, 1994. p. 20.
(Research Triangle Institute 1994) Research Triangle Institute. Baird and McGuire Draft Risk Assessment. Research Triangle Park, North Carolina. August 4, 1994.
(Rosengrant and Craig 1990) Rosengrant, L. and Craig, R. Final Response to BDAT Related Comments Document, Volume I-M. Washington, D.C.: USEPA Office of Solid Waste. EPA/530-SW-90-0610. May, 1990.
(Ryan and Lutes 1993) Ryan, J.V., and Lutes, C.C. Characterization of Emissions from the Simulated Open-Burning of Non-Metallic Automobile Shredder Residue. Research Triangle Park, N.C.: U.S. Environmental Protection Agency. EPA/600/SR-93/044, May 1993.
(SEPA 1994) Swedish Environmental Protection Agency. Sample reports #610s09 and 0610s017: PVC suspension/ PVC plastic and PVC suspension used to make PVC. May 5, 1994.
(SFT 1993) Norwegian State Pollution Control Authority. Input of organohalogens to the convention area from the PVC industry. Submission to the Oslo and Paris Commissions. Oslo, November 1993.
(SRI 1992) Stanford Research Institute. Directory of U.S. Chemical Producers. Menlo Park, CA: Stanford Research Institute. 1992.
(Theisen et al. 1989) Theisen, J., W. Funcke, E. Balfanz, J. Konig. Determination of PCDFs and PCDDs in fire accidents and laboratory combustion tests involving PVC-containing materials. Chemosphere 19(16):423-428.
(Thomas and Spiro 1994) Thomas, V.M. and Spiro, T.G. An Estimation of Dioxin Emissions in the United States. Princeton, NJ: Princeton University, The Center for Energy and Environmental Studies. PU/CEES Report No. 185. August 1994 (Revised September 1994).
(Thornton 1994) Thornton, J. Achieving Zero Discharge: An Emergency Strategy for Dioxin Elimination. Washington, D.C.; Greenpeace USA. July 1994.
(Thornton and Weinberg 1994) Thornton, J., and Weinberg, J. "Dioxin Prevention: Focusing on Chlorine Chemistry." Adapted from testimony delivered at public hearings on U.S. EPA's draft Dioxin Reassessment. Washington, D.C.: Greenpeace USA. December 1994.
(TWC 1992) Texas Water Commission, Compliance Evaluation Inspection, January 15-28, 1992, by S.D. McCoy and R. Musick, Field Investigators, Texas Water Commission District 7 - Houston (with D.J. Hanson and K. Cameron of B.F. Goodrich). Transmitted by interoffice memorandum from S. D. McCoy to File through E. Heyer, Head, Program Services Unit, Field Operations Division. February 26, 1992.
(UBA 1992) German Federal Office of the Environment (UBA). Environmental Damage by PVC: An Overview. Berlin: Umweltbundesamt, June 1992. Reports. EPA Contract No. 68-01-6271. August 15, 1983.
(Vista 1990) Vista Chemical Company. Treatment standards for dioxins and furans. In (L. Rosengrant and R. Craig, authors) Final Response to BDAT Related Comments Document, Volume I-M. Washington, D.C.: USEPA Office of Solid Waste. EPA/530-SW-90-0610). May 1990.
(Vogg et al. 1987) Vogg, H., Metzger, M., Steiglitz, L. Recent findings on the formation and decomposition of PCDD/PCDF in municipal solid waste incineration. Waste Management and Research 5(3):285-294.
(Wagner and Green, 1993) Wagner, John C. and Alex E.S. Green Correlation of chlorinated organic compound emissions from incineration with chlorinated organic input. Chemosphere, Vol. 26, No. 11, pp 20392054, 1993.
(Wenning et al. 1992) Wenning, R.J., Harris, Ungs, M.J., Paustenbach, D.J. and Bedbury, H. Chemometric comparisons of polychlorinated dibenzo-p-dioxin and dibenzofuran residues in surficial sediments from Neward Bay, New Jersey and other industrialized waterways. Arch. Environ. Contam. Toxicol. 22_397-413. 1992.
(Westervelt 1995) Westervelt, R. "PPG and Vista plan VCM venture and capacity boost." Chemical Week, Feb 1, 1995, p 9.
(Wicklund 1992) Memorandum of Jan. 30, 1992: John Wicklund, USEPA Region 6, to Garret Bondy, USEPA, Region 6: "Historical Analytical Data on Drums of 2,4-D Wastes at Vertac."
(Wilken 1994) Wilken, M. Dioxin emissions from furnaces, particularly from wood furnaces. Berlin: Engineering Group for Technical Environmental Protection for the German Environmental Protection Agency, February 1994. See also German Environmental Protection Agency (UBA) "Dioxin emissions from wood furnaces." (Press release). Berlin, February 1994.
If you have come to this page from an outside location click here to get back to mindfully.org