Document yrMMZ9gkZLr9rRzqpMYeYL4E2

btP 8 1983 PREUMJMARY ASSESSMENT OF THE ENVIRONMENTAL PROBLEMS ASSOCIATED WITH VINYL CHLORIDE AND POLYVINYL CHLORIDE (Appendices) Report on the Activities and Findings of the Vinyl Chloride Task Force ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. SEPTEMBER 1974 GENC 000765 PRELIMINARY ASSESSMENT OF THE ENVIRONMENTAL PROBLEMS ASSOCIATED WITH VINYL CHLORIDE AND POLYVINYL CHLORIDE (Appendices) Report on the Activities and Findings of the Vinyl Chloride Task Force Environmental Protection Agency Washington, DC September 1974 GENC 000764 TABLE OF CONTENTS APPENDICES I. Selected Economic Considerations Production Levels Competitive Substitution International Aspects Control Technology II. Producers of Vinyl Chloride and Polyvinyl Chloride VC Producers PVC Producers PVC Copolymer Producers III. The Materials Balance at Vinyl Chloride and Polyvinyl Chloride Facilities Vinyl Chloride Production Facilities Polyvinyl Chloride Polymerization Facilities IV. Interim Method for Sampling and Analysis of Vinyl Chloride in Waste Water Effluents and Air Emissions Scope and Application Summary of Analytical Procedures Interferences Apparatus and Materials Reagents, Solvents, and Standards Sampling Calibration Procedure Quality Control V. Summary of Regional Activities Region I: Region II: Region III: Region IV: Region V: Region VI: Region IX: Leominster, Massachusetts Flemingtop,New Jersey Delaware City, Delaware S. Charleston, W. Virginia Louisville, Kentucky Paine sville,Ohio Plaquemine, Louisiana Long Beach, California 1 1 1 4 4 6 6 6 8 10 10 11 17 17 17 17 18 19 20 23 25 25 26 26 27 27 28 28 29 29 1 GEt-iC 000767 VI. VII. VIII. IX. Persistence of Vinyl Chloride Behavior of Vinyl Chloride in Air Behavior of Vinyl Chloride in Water Behavior of Vinyl Chloride in Closed Rooms Health Effects of Vinyl Chloride Occupational Cases of Liver Angiosarcoma Cases of Hepatic Angiosarcoma, Connecticut, 1935-1973 Observed Deaths/Expected Deaths in VC Workers Summary of Toxicological and Epidemiological Studies on Vinyl Chloride Disposal of Products Containing Polyvinyl Chloride Incineration Landfilling Resource Recovery Activities of Task Force 31 31 31 32 34 38 40 44 63 63 64 65 67 ii OEHC 00' 63 APPENDIX I SELECTED ECONOMIC CONSIDERATIONS Production Levels During 1973, VC production was at the 5.3 billion pound level with PVC and its copolymers at the 4.6 billion pound level. PVC has become a very important polymer as evidenced by the broad dependence of nearly every branch of industrial and commercial activity upon products and components fabricated from this plastic. In Table 1, major PVC products manufactured during 1973 are iden tified. The U. S. VC/PVC industry has been operating for more than forty years, and over the past five years has shown an average annual growth rate of 14 percent - - a rate of growth that had been expected to taper off only moderately in the next few years. The size of this industry can be appreciated by considering that the synthesis of the monomer is conducted in fifteenU. S. plants, and forty-three facilities are engaged in polymerization of PVC (including its use as a copolymer) with almost all of these plants currently operating at or near capacity. At least 7,500 plants are engaged in fabricating products from PVC. About 1,500 workers are employed in monomer synthesis and an additional 5,000 in polymerization operations. Estimates have suggested that up to 350, 000 workers may be associated with the fabrication plants. The wholesale value of the annual output of fabricated products based on PVC is at least several billion dollars. Coihpetitive Substitution Should requirements for worker safety or environmental controls drive the price of PVC resin upward, it seems likely that some PVC products would be displaced by products using other plastics or other materials. Other products dependent on PVC might disappear alto gether from the marketplace. Probably one-fourth to one-third of current PVC products by value are marginally competitive with other plastic products. At significantly higher prices a lesser number probably would find substitutes in other materials at higher costs. Identified in Table 2 are a few of the substitute materials that might be considered. For some uses, there are no apparent substitutes. 1 GENC 000769 Market Category I. Apparel II. Building and Construction III. Electrical IV. Home V. Packaging VI. Recreation VII. Transportation VIII. Miscellaneous Table 1 MAJOR PVC PRODUCTS Products Baby pants Footwear Outerwear Extruded foam moldings Flooring Lighting Panels and siding Pipe and conduit Pipe fittings Rainwater systems, soffits. facias Swimming pool liners Weatherstripping Windows Wire and cable Appliances Furniture Garden hose Housewares Wall coverings and wood surfacing films Blow molded bottles Closure liners and gaskets Coatings Film Sheet Phonograph records Sporting goods Toys Auto mats Auto tops Upholstery and seat covers Agriculture (incl. pipe) Credit cards Laminates Medical tubing Novelties Stationery supplies Tools and hardware Other 1973 1000 metric tons 12 66 31 26 211 5 39 525 44 16 18 16 26 194 20 145 18 51 54 36 9 9 59 35 66 25 88 18 15 83 66 8 23 23 7 18 8 45 Total 2158 2 GENC 000770 Table 2 SUBSTITUTE MATERIALS FOR PVC PRODUCTS PVC PRODUCT Pipe & Tubing Flooring Electrical Insulation Records Film & Sheet Products Coatings Household Goods Packaging SUBSTITUTES SAME PRICE RANGE Polyethylene Polypropylene Metals ABS resins X X Asphalt Wood ABS resins X Polyethylene Polypropy1ene EPDM rubbers SBR rubbers TFE plastics X X ABS resins Acrylics Polyvinylidene chloride Polyethylene Polypropylene Cellulosics X X Acrylics Polyurethanes Cellulosics Styrene Polyethylene Polyp ropylene Wood Metals Acrylics X X X Polyethylene Polypropylene Polyvinylidene chloride Cellulosics Acrylics Polyurethanes Glass X X rKJH KA rKSA KA^ Kn J kUS kMS M kS rK /S KJ KA kS kN rKSA knA Kr Ji Kr Ji rKSA KAn K /n Kr i/ KmA r S r kSJ HIGHER PRICE 3 6ENG 000771 International Aspects U. S. based manufacturers currently produce about one-third of th^k western world's supply of resins, with the U. S. market also consuming^ about one-third of the total. In 1973, 3.7 percent of PVC and 7.8 percent of VC manufactured in the United States were exported. Prior to the recent U. S. concern over worker and environmental controls at VC and PVC facilities, there was no reason to anticipate a major change in the U. S. share of production or market during the next few years. Recent increases in demand for PVC resins -- and concurrently for VC --at attractive prices have been of worldwide dimensions with expansion plans for PVC manufacturing being considered by a number of companies at home and abroad. There is presently an import duty on PVC resin from countries with status as Most Favored Nations of 1 1/4 cents per pound plus six percent ad valorem and from other nations of four cents per pound plus 30 per cent ad valorem. Given the current U. S. market price of 18 to 24 cents per pound for the general purpose uncompounded resin, there has been little incentive to import PVC resin. Also, there currently is little export incentive because of short U. S. supply and unattractive foreign prices. However, higher prices as a result of more stringent worker or environmental controls in PVC resin plants in the United States than abroad might well stimulate significantly increased imports. Control Technology While there appear to be a number of general approaches for reducing the discharge of VC into the environment at VC and PVC resin plants and the discharge of PVC at resin plants, in many respects the approaches^k must be tailored to the individual plants. All VC plants and some PVO^P resin plants are outdoors while other PVC plants are at least partially enclosed. A variety of production processes are used, and different kinds of technology are employed. However, there are some common measures that would reduce VC emissions. FOR VC PLANTS: 1. Reducing the escape into the atmosphere of VC when venting the tank car gauge tube, disconnecting the feeding line, and closing the valves during rail tank car loading. Mechanical disconnect de vices and double block and bleed piping are available to ease this problem. 2. Improving the quality of pumps to reduce the possibility of leakage due to failure of seals. Pumps are available today which could minimize this problem. 3. Venting unintentional leaks and spills into a system which is flared and, preferably, scrubbed. 4 GB-c oo772 FOR PVC RESIN PLANTS: 1. Collection and destruction of purge gases from the reaction kettles prior to opening for cleaning, sampling, or recharging. 2. Centralized collection and filtering of VC vapor discharges from dryers and centrifuges. With regard to PVC particulate in air and water discharges, improved housekeeping and relatively simple ventilation filtering systems are usu ally technically feasible and effective. Laboratory data have shown that VC can be adsorbed on activated carbon. Concentrated VC vapor streams have produced a recovery work ing capacity on carbon equivalent to about ten percent of the carbon weight. Ambient air contaminated with low levels of VC produces significantly lower adsorbent working capacities. Control of dilute VC is therefore possible but may not be practical using activated carbon. Carbon regene ration using steam or pressure swing appears possible, with recovery of desorbed VC for recycle. Clearly, these approaches will not eliminate losses but should mate rially reduce them. In the longer run, the development of continuous flow processes, the use of larger kettles, better housekeeping, and/or reductions in the number of feed lines might result in more dramatic reductions of VC leakage. REFERENCES 1. Modern Plastics, Jan 1974, p. 43 2. The 1972 Census of Manufacturers shows 7,574 plants manufacturing miscellaneous plastics products (SIC 3079), a substantial number of which use PVC. SIC 3079 probably covers most, but not all, PVC fabricators. 3. Discussions with representatives of the Department of Commerce, Manufacturing Chemists Association, and Society of Plastics Industry. V 5 GENC 00077 APPENDIX II PRODUCERS OF VINYL CHLORIDE AND POLYVINYL CHLORIDE The major producers of VC, PVC, and PVC copolymers are listed in this section with the plant location and available capacity data. VC Producers Location Annual Capacity (Millions of Pounds) Allied Chemical Corporation American Chemical Corporation Continental Oil Company Dow Chemical, U.S. A. Ethyl Corporation B.F. Goodrich Chemical Company Monochem, Inc. PPG Industries, Inc. Shell Chemical Company Tenneco, Inc. Baton Rouge, La. Long Beach, Calif. Westlake, La. Freeport, Tex. Oyster Creek, Tex. Plaquemine, La. Baton Rouge, La. Pasadena, Tex. Calvert City, Ky. Geismar, La. Lake Charles, La. Guayanilla, P.R. Deer Park, Tex. Norco, Tex. Houston, Tex. 300 175 650 200 700 390 300 150 1000 300 400 500 840 700 225 PVC Producers Air Products and Chemicals, Inc. American Chemical Corporation Borden, Inc. Continental Oil Company Calvert City, Ky. Pensacola, Fla. Long Beach, Calif. Rliopolis, HI. Leominster, Mass. Aberdeen, Miss. Oklahoma City, Okla. 150 50 150 140 180 285 240 6 GEMC ooq7 Company Locations Annual Capacity (Millions of Pounds) Diamond Shamrock Chemical Company Deer Park, Tex. Delaware City, Del, 270 100 Ethyl Corporation Baton Rouge, La. 180 The Firestone Tire &, Rubber Company Perryville, Md. Pottstown, Pa. 230 270 The General Tire & Rubber Company Ashtabula, Ohio Pleasants County, W. Va. 125 50 B. F. Goodrich Chemical Company Avon Lake, Ohio Henry, 111. Long Beach, Calif. Louisville, Ky. Pedricktown, N. J. 140 140 140 340 170 The Goodyear Tire & Rubber Company Niagara Falls, N.Y. Plaquemine, La. 100 100 Great American Chemical Corporation Fitchburg, Mass. 40 Hooker Chemical Corporation Burlington, N, J. Hicksville, N.Y. 180 15 Keyso r-Century Corporation Saugus, Calif. Delaware City, Del. 35 35 Monsanto Company Springfield, Mass. 70 National Starch & Chemical Corporation Meredosia, HI. 10 Olin Corporation Assonet,- Mass. 150 The Pantasote Co. of New York, Inc, Passiac, N.J. Point Pleasant, W. Va. 60 90 Robintech, Inc. Painesville, Ohio 250 Stauffer Chemical Company Delaware City, Del. 175 Tenneco Chemicals, Inc. Burlington, N.J. Flemington, N.J. 165 70 Union Carbide Corporation South Charleston, W, Va. Texas City, Tex. 160 240 Uniroyal, Inc. Painesville, Ohio 'a. 140 7 GENC 000775 PVC Copolymer Producers Company Locations A. Polyvinyl Chloride-Propylene Copolymer Resins Air Products and Chemicals, Inc. Calvert City, Ky. B. Polyvinyl Chloride-Vinyl Acetate Copolymer Resins Air Products and Chemicals, Inc. Calvert City, Ky. American Chemical Corporation Long Beach, Calif. Atlantic Tubing & Rubber Company Cranston, R. I. Borden, Inc. Bainbridge, N. Y. Compton, Calif. Demopolis, Ala. Illiopolis, 111. Leominster, Mass. The Firestone Tire & Rubber Comany Pottstown, Pa. B. F. Goodrich Chemical Company Avon Lake, Ohio Louisville, Ky. Hooker Chemical Corporation Hicksville, N. Y. Keysor-Century Corporation Saugus, Calif. National Starch and Chemical Corporation Meredosia, 111. Olin Corporation Assonet, Mass. The Pantasote Company of New York, Inc. Passaic, N. J. Point Pleasant, W. Va. C. Polyvinyl Chloride-Vinylidene Chloride Copolymer Resins BASF Wyandotte Corporation South Kearny, N. J. Borden, Inc. Bainbridge, N. Y. Compton, Calif. Demopolis, Ala. Illiopolis, 111. Leominster, Mass. 8 GENC 000776 Dow Chemical, U.S. A. B.F. Goodrich Chemical Company W. R. Grace & Company Morton-Norwich Products, Inc. National Starch and Chemical Corporation SCM Corporation Tenneco, Inc. Union Carbide Corporation Midland, Mich. Louisville, Ky. Owensboro, Ky. South Acton, Mass. Ringwood, 111. Meredosia, 111. Huron, Ohio Burlington, N. J. Flemington, N. J. Institute and South Charleston, W. Va. Texas City, Texas REFERENCES 1. 1974 Directory of Chemical Producers, USA, Chemical Information Services, Stanford Research Institute, Menlo Park, California, 1974. 2. Chemical Marketing Reporter, May 20, 1974. 9 6ENC 000777 APPENDIX III THE MATERIALS BALANCE AT VINYL CHLORIDE AND POLYVINYL CHLORIDE FACILITIES Vinyl Chloride Production Facilities Detailed, reliable data for estimating material losses at VC facilities with precision are not readily available. Therefore, only generalized estimates have been attempted. A simplified block diagram for production of VC from ethylene and chlorine is shown in Figure 1. Some VC complexes utilize oxychlorination units; others produce ethyl chloride from the by-product hydrogen chloride (HC1) and ethylene. However, the production of dichloroethane (EDC) allows for many approaches to recycling of light and heavy materials such that the losses of VC are reduced. Even vent streams of inerts can be scrubbed with EDC for maximum removal of VC before venting. Light ends such as methane are usually flared and VC is converted to water and small amounts of HC1. VC losses have come primarily from vent streams, the storage and transportation loading systems, and seepages from pumps. If vent streams are not scrubbed or flared, the amount of VC reaching the atmosphere increases considerably. This in turn is influenced by the purity of the ethylene and the chlorine being fed into the units. Usually, these inerts come out in the EDC unit but may be carried on depending upon the pro ducer' s philosophy regarding the purity of the EDC to be fed to the cracker. Experience has been that the higher the purity of EDC both with regard to light and heavy material, the greater the efficiency of the cracking. It is frequently difficult to pinpoint the areas and quantities of VC losses. However, some generalizations can be made for, as an example, a plant producing 500 million pounds per year of VC. (The industry is heading toward plants of this size and larger.) Tank car loading losses may be several hundred pounds per day. Vent stream losses could reach another 100 pounds per day while losses of VC entrapped in the water effluent might be a few pounds per day. In addition to these very small operating losses, there are undoubtedly unintentional losses from leaking pumps, flanges, and containment vessels, with total plant losses probably less than 0.1% or less than 500,000 pounds per year. From an environmental standpoint, the disposal of the heavy chlori nated hydrocarbons may also presents problem. Some are sold to solvent scrap dealers for salvage. In the past much of the material has been dumped at sea or put into landfills or deep wells. More recently, incin eration has been used, which is known to produce HC1 emissions. 10 Polyvinyl Chloride Polymerization Facilities Reasonably reliable data are available for estimating material losses at PVC facilities. However, generalizations applicable to the entire industry must be surrounded with many caveats. It must be emphasized that there are a number of PVC processes, and each plant has its own idiosyncrasies. VC losses will fluctuate depending on the care exercised in operating the PVC plant, types of products produced, frequency of product change, method of PVC shipment, and emergency situations. Estimates of losses have varied widely in the industry, indicating the complexity of establish ing precise losses for a given facility and overall losses on a nationwide basis. In general, older PVC plants are smaller than those being built today and are equipped with smaller sized reactors. With small reactors, the number of batches required to produce a given amount of PVC is greater, and thus the number of process steps are increased with a greater poten tial for loss of both VC and PVC. Further, a small plant has the disadvan tage of having to make frequent resin changes to meet customer demands. During these changeovers a certain amount of off-grade resin is produced. In addition, older plants have the added burden of higher maintenance than new plants, but this tends to stabilize after a few years. The handl ing of VC and the production of the high quality resins which are demanded by the marketplace require a reasonable maintenance program. Mainte nance consists primarily of the care of agitator seals, pump seals, and valves and the removal of polymer which slowly builds up in VC lines -primarily in the recovery system. Although many older facilities have been in operation for years, they are usually not the same as when first installed. Some of the operators have continually updated the plants for many reasons including labor savings systems, new product require ments, replacement of wornout equipment, addition of new product lines, and safety. When VC was cheap and there was little concern about its toxicity, the emphasis was almost exclusively on productivity. Often this resulted in high losses of VC to the environment as recovery cycles were reduced. Today, the picture is changing. Not only are the producers trying to reduce the direct VC losses, but they are also trying to minimize PVC losses by scheduling longer production runs between product changes. As an example, the newer large plants are setup with multiple production lines. This allows the dedication of one line to a given product which results in very low resin loss due to product change. It GEHC 00077? The traditional method of stating yield of VC in PVC plants has been based upon pounds of prime resin in the bag as compared to VC invoiced. This often has led to a misunderstanding about VC losses with the interpretation that a 94% yield means 6% VC loss to the envi ronment. In fact some VC may never actually be received because of the inability to measure the weight of tank cars accurately, some of the losses are in the form of PVC scrap, and some losses escape as PVC particles. A properly run and maintained suspension plant using technology that is ten years old should be capable of obtaining a 95% or higher yield unless some especially esoteric resin is being produced along with large amounts of scrap or off-grade resin. For the older plants, the losses will probably be significantly higher. Other than overall sloppy operation, the recovery system is the single most important part of the plant govern ing VC losses. If insufficient time is allowed or vacuum is not applied, then the VC content in the PVC/water slurry will be greater than neces sary. As a result, VC losses will occur in the centrifuge effluent water, drier/ product collector vent air, the venting of the reactor, and the slurry tank. The magnitude of VC and PVC losses in a typical PVC plant is described in Figure 2. These losses are expressed as a range of losses depending on the feed rate, reactor size, reactor cleaning procedures, batch sizes, level of technology, and general housekeeping and operating procedures. The following comments on manufacturing practices may help put these losses into perspective: 1. VC Feed - This is shipped as virtually 100% VC and does not normally contain an inhibitor. 2. VC Unloading - Considering normal losses in disconnecting the piping, sampling, tank gauging, pump and compressor seals to the tank cars, losses to the atmosphere should not be greater than 100 pounds per car. 3. VC Charging - A 0.05% loss between storage and polymerization should cover losses from flanges and seals throughout all VC handling equipment. 4. Polymerization - The loss from build-up of PVC on the walls of the reactor is split between reactor wash-out and the slurry strainer. 5. Reactor Venting - Before the reactor can be cleaned, residual VC is vented After recovery and emptying the PVC resin, the reactor is full of a mixture of air, moisture, and VC at ambient conditions. 12 6. Recovery - Processing schemes will vary, but one of the most widely used is the direct recovery of unreacted VC from the reactor. While the reaction can be carried out further, economically it is essen tially complete at 90% conversion or even less depending on the type of resin. At this point the residual VC is recovered by means of compres sors which evacuate VC from the reactor. The recovered VC is con densed and distilled before recycling to the reactor. 7. Drying - Unreacted VC is collected in the recovery system but there are losses of polymer in the drier due to coalescence of the resin and periodic clean-out. This is almost entirely scrap, 8. Product Collector - Most plants use bag collectors so that the loss of resin is less than one pound per hour, but there are losses due to product changes which raise the total. 9. Screening - Oversize resin is removed from the final product. This material consists of scrap and off-grade resin. With the current PVC shortage much of this off-grade resin is used as prime resin by special customers. 10. Miscellaneous - In addition to the above losses, others occur as scrap or off-grade polymer and as quality control samples. a. Bad Batches - Most plants experience batches which are off specification. These range from "just slightly off" to solid batches, with losses at 2 to 3 batches per month or about 0.4% or 40 pounds per hour average. Salvage value depends upon the degree of "off-grade" and market conditions. d. Samples - Probably about 0.05% or 5 pounds per hour and is usually destroyed in testing. c. Polymer Build-up - VC slowly polymerizes in the pipe lines, particularly the recovery system, and must be removed peri odically. No quantitative value is available for this loss. d. Spillage - Some of the product is shipped in bulk and some is bagged. While some spillage occurs in bulk handling, more occurs in bag filling and in bag breakage. e. Centrifuge Effluent - Some PVC enters the effluent water. 11. Product Change-Over - As indicated previously there are losses in the drier and collector due to cleaning for changes from one product to another. In addition one must segregate the first product that comes through this system. The amount can vary widely depending upon the number of changes and the sensitivity of the product to contamination from the previous product. 13 GENC 000731 The foregoing analysis, together with estimates provided by industry, suggests that the losses of VC at PVC polymerization facilities currently range from about 3.0 to 6.3% while PVC losses are on the order of 1.3%. 14 vjvJ^,7^ PRODUCTION OF VC FROM ETHYLENE AND CHLORINE SIMPLIFIED BLOCK DIAGRAM Figure 1 Water H .1 ic jOOf o--' PRELIMINARY ESTIMV-E OF LOSSES IN PVC SUSPENSION POLYMERIZATION (TYPICAL PROCESS! F igure GENC 000784 NJ APPENDIX IV INTERIM METHOD FOR SAMPLING AND ANALYSIS OF VINYL CHLORIDE IN WASTE WATER EFFLUENTS AND AIR EMISSIONS Scope and Application The initial basis for this method was developed during the moni toring program carried out by EPA Region IV in March and April. The techniques used by Region IV provided guidance for the monitoring activities of other Regions, and the experiences of all Regions were then incorporated into this refined version of the original Region IV approach. This method is applicable to VC determinations in water effluents, sludges and scums, and atmospheric emissions. The limit of detection is approximately 0.06 mg/1 in water and 0.06 ppm (v/v) in air samples. Summary of Analytical Procedures Water composite samples, air continuous composite bag samples, and air and water grab samples are analyzed without cleanup by gas chromatography (GC). Separations are effected by selection of one of two types of columns depending upon the nature of the sample. Detection is by means of the flame ionization detector (FID). Tetrahydrofuran extracts of sludges and scums are used for injection into the GC. Air continuous samples on activated carbon are extracted with carbon disulfide, and the extract is analyzed by direct injection into the GC. Calibration curves are developed using gravimetrically prepared calibration solutions, or by using known dilutions of VC in carrier gas. VC confirmation should be made by mass spectrometric analysis of the GC eluent if possible. Independent confirmation may also be made in the event of extraordinarily high VC concentration sam ples by using long path Fourier transform IR spectrophotometry. This IR technique requires special equipment and about 20 cubic feet of air samples. Interferences v Certain volatile hydrocarbons such as neopentane, butadiene, and freon 12 have elution characteristics similar to VC. However, on the GC column substrates specified in these procedures, these have not usually presented problems of resolution of the VC peak. When column substrates other than those specified have been used, impurities from solvents and carbon adsorbents have been 17 GENC 000785 f found to interfere with the VC elution peak. Under certain condi tions a peak is associated with the injection and subsequent with drawal of the microsyringe into and from the GC septum. These peaks can also give interferences with the VC peak. Withdrawal should be timed to avoid overlap of this peak with the VC peak. Apparatus and Materials Gas Chromatograph Flame Ionization Detector Recorder - any potentiometric strip chart recorder which is compatible with the detector system. An integrator is also desirable to estimate peak areas. Column Materials for Waste Water, Sludge, or Scum Samples Borosilicate glass tube or stainless steel tube - 6' x 2. 5 mm ID preferred. When GC configuration requires columns of other dimensions, these should be used. Solid support - 60. to 80 mesh Gas Chrom Q Liquid Phase - 4% FFAP on specified solid support (weight percent). Liquid phase on solid support can be purchased directly from commercial distributors. Column Materials for Air Samples Borosilicate glass tubing or stainless steel tubing - 8' x 2. 5 mm ID preferred. When GC configuration requires columns of other dimensions, these should be used. Solid support - Carbopak A Liquid phase - 0.4% Carbowax 1500 on solid support (weight percent). Liquid support on solid phase can be purchased directly from commercial distributors. Continuous Air Monitoring Materials - Carbon Adsorption Option Adsorption Tube - pyrex glass, 18" x 3/8" OD Activated coconut charcoal, 8-16 mesh. Any good commer cial grade, e.g. Fischer Scientific Company can be used. Becton-Dickson 27 gage 3/8" hypodermic needle flow control Vacuum pump Air flow meter 18 G8C- 0007S6 Continuous Air Monitoring Materials and Equipment - Bag Sampling Option Environmental Measurements, Inc., Programmable Bag Sampler Tedlar bags (or equivalent) Gas Pressure Regulator (0-5 PSIG) Microsyringes - 10, 25, 50, and 100 microliter (graduated) Gas-tight sample syringes - 1 and 50 ml (graduated) Vacuum Sampling Cans - 370 ml steel Vacu-Samplers, or glass sampling bottles. Cans and bottles should be flushed with clean air or nitrogen and evacuated prior to use. Evacuated containers should be protected from rough handling" to prevent implosion or collapse. Sampling Bags (Tedlar or equivalent) - 12" x 12", 36" x 36", equipped with sampling valves and speta for GC sample withdrawal Automatic water sampler - compositor (manual sampling is optional) equipped with sample refrigeration capabilities, and a a means to prevent loss of vinyl chloride from open bottles Glass sampling bottles with teflon lined screw type caps - 50 ml capacity or other sizes depending upon sampler requirements Septum-sealed vials - 1 to 10 ml capacity Volumetric Flask, Glass stoppered, 25 ml Medicine droppers Dedicated GC/M. S. for confirmatory tests (preferable) Barometer Thermometer Anemometer Reagents, Solvents, and Standards Carrier gases - zero nitrogen or helium FID gases - zero hydrogen, oxygen Tetrahydrofuran, reagent grade, peroxide-free 19 GENC 000787 Carbon tetrachloride (reagent grade) Carbon disulfide (reagent grade) Standards VC in zero air, 50 ppm (+ 2%) v/v VC, analyzed reagent grade (lecture bottle) Sampling A. Water Samples All waste water discharge points identified in NPDES permits should be sampled for VC. A minimum of three successive 24-hour composite samples of each site should betaken. Com positing interval should be one hour (manual or automatic sampling is optional). Compositing interval of 20 minutes may be used if the automatic sampler has this capability. Samples should be taken at waste treatment units such as clarifiers and scum and sludge separators. Two 8-hour composites should be taken from the effluents from each of these points, and one 8-hour composite should betaken of scum and sludge from each separator unit. Compositing interval should be one hour. Three grab samples of clean process water (city or private well) should be taken as blanks. Samples should betaken in50 ml bottles with gas-tight, teflonsealed, screw cap closures, or in equivalent containers re quired by the characteristics of automatic samplers. All water, sludge, and scum samples should be refrigerated dur ing collection and storage. Compositing volumes should be selected to assure head space above the sample is absent or minimized to avoid loss of VC by its partitioning into the gas phase when samples are sealed. Provisions should be made to avoid such losses during continuous monitoring operations. Estimates of discharge flows should be made using any appro priate measuring device (venturi, weir, magnetic meter, etc.). Samples should be preserved by refrigeration and protected from sunlight until they are ready for analysis. B. Air Samples Sampling sites should be selected which are downwind and in the plume of the atmospheric emissions from the plant. Samples should be collected only in areas where local residents or neighboring industries would be exposed. At a minimum. 20 6ENC 000788 sampling should be conducted over a period of five days. Sites should be selected in the following array: one site immediately upwind (A) and one immediately downwind (B) of the plant site; four sites about 0.4 miles from the plant site, one laterally left (C) and one laterally right (D) of the plant site on a line roughly perpendicular to the prevailing wind direction and two (E, F) downwind from the plant site; two sampling sites (G, H) approximately 0. Smiles downwind; single sampling sites, each at distances approximately 0.6 (I), 0.8 (J), 1.0 (K), and 3.0 (L) miles downwind from the plant site. If wind is fish-tailing severely, move sampling sites Gand H approximately 0.5 mile upwind of the fish-tailing wind direction from the plant. The sites specified are minimum. Additional sites may be selected contingent on overriding micrometeorological considerations. These should be determined in consultation with the Regional meteorologist. These may be at ground or some elevated level, as determined by the plume survey or as estimated by release of meteorological balloons, anemometer, and wind direction indicators, etc. SAMPLING SITES Prevailing Wind Direction __ Minimum Sampling Schedule Miles from plant site Site Symbol Time Mon Wed Fri 0.0 A 0.4 C Plant D 0800 A, A, B A, B, B A, A, B 0.0 B------ 1000 C,D,F C, D C, D, D 0.4 E F 1200 A, E A, G, G A, E 0.5 G H 1400 B, B, F B, H B, B, G 0.6 I 0.8 J 1.0 K 1600 1800 2000 C, G D.I - E, K - H,L,L I.J L,L - 3.0 L (Note: All times are + 30 minutes for manual grab samples, or + Tminutes for automatic, programmable bag samplers). Grab samples should be taken in 50 ml gas-tight syringes, 50 to 100 ml glass sampling bottles, 370 ml "Vacu-Sampler" metal cans, or 12" x 12" capacity Tedlar-type bags. Both the Vacu-Samplers and the glass sampling bottles should be evacuated prior to use. (Caution: These may implode or collapse when under vacuum. Use due care in their handling). The perfect gas laws should be assumed to estimate gas vol umes. Gas-tight syringes are flushed several times with am bient air before a sample is taken. After the sample is taken, the gas-tight syringe is locked and sealed until it is ready for analysis. 21 6EHC 000789 The Tedlar-type bag samplers may be filled by pulling the walls of the bag apart manually, or better, by placing the bag in an enclosure and pulling a vacuum on the outside sur faces of the bag. The bag is sealed until it is ready to be analyzed. Tedlar-type bags are preferred for grab sampling. All samples should be protected from sunlight. Continuous Sampling - Carbon Adsorption Option: Continuous samples are taken in pyrex tubes (approximately 3/8" O. D. x 18" long) packed with a good grade of activated coconut shell charcoal. The charcoal is added to the tube in three segments, each 3-inches long, and each separated by a glass wool plug. The two ends of the tube are also plugged with glass wool. Both ends of the pack adsorption tube are plugged with serum caps during transport and for storage pur poses. Flow rate through the tube is controlled by inserting a BectonDickson 27 gage, 3/8" hypodermic needle through one of the serum caps into the end glass wool plug. Air is sucked through the tube by connecting it to a conventional vacuum pump. The arrangement is similar to that used in the National Air Surveillance Network. Flow rate should be about 200 ml per minute. For each adsorption tube, the flow rate should be calibrated in the laboratory before the sample is taken and should be verified again in the laboratory after the sample is taken. Clean needles frequently to prevent plugging. The adsorption efficiency of the carbon in the adsorption tube should be verified in the laboratory by preparing a 5 ppm v/v VC mixture in the 36" x 36" Tedlar-type bag and drawing this through the adsorption tube. Flow rates should be verified before and after the experiment. It is important to note that all collections should be made with the adsorption tubes held in an upright position to minimize channeling. Adsorption tubes should be protected from sunlight either by wrapping with foil or by enclosing them in a box. Each segment of the adsorption tube is worked up separately by etching the tube in the middle^ of a 3" section with a file, successively breaking each segment and spilling its contents into measured volumes of carbon disulfide in glass stoppered test tubes. The additions should be effected cautiously and with cooling in an ice bath since the interaction of activated carbon with carbon disulfide is quite exothermic. A 2 microliter aliquot of the supernatant solution should be injected on the carbowax 1500 column for estimation of the adsorped VC. Suc cessive analysis of the three adsorption tube segments will indicate the amount of break-through of VC through the adsorb ent. 22 GENIC 000??o The same procedure should be used for taking samples in the field. Continuous Sampling - Programmable Bag Sampler Option: The sampler is programmed to take twenty-four consecutive one-hour composite samples. Each one-hour sample is analyzed separately for VC content. Sampling rate of the individual pumps should be verified before and after use of the sampling device. Record the temperature and atmospheric pressure at which the samples are taken. All gas volumes and concentrations should be corrected to 25C and one atmosphere (760 mm Hg). At a minimum, con tinuous samples should be taken at sites A, B, C, and D at ground level, unless otherwise indicated by micrometeorological conditions. Calibration A. Gas Analysis - Gas Dilution Option: Record ambient temperature and atmospheric pressure. Evaluate the 36" x 36" Tedlar-type bag. Add 1 liter of the standard VC gas mixture (50 ppm, v/v) to the bag. This addi tion maybe made with a flow meter or with a gas-tight syringe. Dilute with nine liters of zero nitrogen or helium carrier gas. This gives a concentration of 5.0ppm (v/v) of VC. (13 ng/ml at 25C and one atmosphere.) Evacuate a 12" x 12" Tedlar-type bag and add 0.5 1 of the 5.0 ppm (v/v) concentration mixture. Dilute with 2 liters of zero nitrogen or helium carrier gas. This gives a concentration of 1. 0 ppm (v/v) VC, (2.6 ng/ml at 25C and one atmosphere). Evaucate a 12" x 12" Tedlar-type bag and add 0.5 1 of the 1.0 ppm (v/v) VC calibration mixture. Dilute with 2 liters of zero nitrogen or helium carrier gas. This gives a concentration of 0.2 ppm (v/v) VC (about 0.52 ng/ml at 25C and one atmosphere). Evacuate a 12"x 12" Tedlar-type bag and add 0. 75 1 of the 0. 2 ppm (v/v) VC calibration mixture. Dilute with 1. 75 liters of zero nitrogen or helium carrier gas. This gives a concentra tion of 0.06 ppm (v/v) VC (about 0.16 ng/ml at 25C and one atmosphere). This is about the"limit of detection for direct injection into the GC. With a gas-tight syringe, inject 1 ml aliquots of the 5.0, 1.0, 0.20 and 0.06 ppm (v/v) VC calibration mixtures into a GC equipped with a Carbowax 1500 or Carbopak column and an FID detector. Use zero nitrogen or helium as carrier gas at a flow rate of 60 ml/min. Operate the inlet and the column isothermally at room temperature. 23 iiENC 0007yl Prepare a calibration curve. Repeat until the calibration curve is reproducible. B. Gas or Water Analysis - Gravimetric option: Stock solution of VC. Pipet 40.0 ml of carbon tetrachloride into a tared 50 ml glass stoppered volumetric flask and accurately weigh to 0.1 mg. Attach a tygon delivery tube to the VC lecture bottle valve. Attach the end of the delivery tube to a piece of glass tubing which has been constricted at one end. flush out the tube with VC, and slowly bubble VC into the CCI4 containing volumetric flask until about 5.0 mg of VC has been added. Precautions should be exercised to prevent loss of carbon tetrachloride during this operation. Reweigh the volumetric flask to determine the weight of added VC. Fill the volume tric flask to the 50 ml mark (approximately 100 ppm wt/vol). (These operations should be carried out in a hood). Transfer 1 ml of the stock solution of VC to a 25 ml volume tric flask and dilute to the 25 ml mark with carbon tetrachlo ride (approximately 4 ppm w/v). Transfer 5 ml of the 4 ppm VC solution to a 10 ml volume tric flask and dilute to the 10 ml mark (approximately 2 ppm, w/v). Repeat dilution for a solution approximately 1 ppm, and 0. 2 ppm. Transfer the stock solution to a teflon-lined screw capped bottle. This solution can be kept for extended periods of time Transfer the diluted solutions to serum vials and cap them with teflon-lined serum cap septa. Inject 1 ml aliquots of the calibration solutions in the GC equipped with Carbowax 1500 on Carbopak A packed columns and an FID detector. Use Zero nitrogen or helium carrier gas at a flow rate of 60 ml/min. Operate the inlet at 150<>C and the column at 60 C. After the VC peak has been eluted, program the column temperature, to 150 C to elute solvent. Cool column back to 60C for follow-on concentrations. Repeat procedure using a GC equipped with a 4% FFAP on Gas ChromQpacked column and FID detector. Operate under the same conditions. Prepare a calibration curve to be used be used with water samples. 24 GENC 000792 Procedure Water Sample Analysis Untreated water samples (1-5 microliter aliquots) are injected directly into the GC. A 4% FFAP on "Gas Chrom Q" packed column is used. Nitro gen zero gas or helium is used as the carrier gas at a flow rate of 60 ml/min. Inlet temperature is set at 1500C. The column is operated isothermally at 62C. Detection is by FID. Report concentration of VC in sample in mg/1. Sludge and Scum Samples Extract 5 grams of sludge or scum sample with 100 ml of tetrahydrofuran (THF). Analyze THF extract in the same manner used for water samples. If VC concentrations are too high, make appropriate dilutions of the THF extracts. Report concentration of VC in sample in mg /g of sample. Air Sample Analysis Grab samples. Use a 0.4% Carbowaxl500 on Carbopak A packed column. Use nitrogen zero gas or helium as the carrier gas with a flow rate of 60 ml/min. Operate the column and inlet at room temperature. Use a flame ionization detector. Untreated air samples (1 ml) are injected directly into the GC. VC contamination of syringes requires attention. Report concentration of VC in gas samples in ppm (v/v). Continuous Samples Use same procedure as previously discussed for calibration of adsorption tube efficiency. Quality Control , Duplicate sample analyses are recommended as a quality con trol check. 25 GEHC 000793 APPENDIX V SUMMARY OF REGIONAL ACTIVITIES This Appendix briefly summarizes the results of the preliminary VC monitoring activities conducted by EPA Regional Offices during the Spring of 1974 at the request of the Task Force. More detailed reports are available from the Regional Offices. The sampling and analyses were carried out in a very short period of time using new methods, based on the Agency's best scientific judge ment. They represent, in the Agency's opinion, the best methods then available. In large measure, the sampling and analysis methods were based on previous analytical studies in which similar chemicals were evaluated. However, they had not been thoroughly tested for accuracy and precision under field conditions. Prior to and during the sampling and measurement only limited quality control and standardization of procedures could be applied in the time available. The methods utilized were interim procedures which have already been subjected to further modification. The nature of the PVC manufacturing process results in the escape of VC pulses which could lead to widely fluctuating levels of VC in the ambient air. So, too, changes in air movement may influence concen trations at a given station at any one time. Therefore, the VC data reported are preliminary in nature and are subject to change as addi tional monitoring is performed. Individual measurements probably underestimate the VC levels due to the possibility of VC leakages and other inaccuracies in the monitoring system. Region I: Leominster, Massachusetts: Borden Chemical Company (PVC); May 9, 10, 13. 1. One hundred and fifty-seven discrete (grab) ambient air sam ples were collected on plant property and within a 3.0 mile radius of the plant. The VC concentrations ranged from less than the detectable limit of 0.06 ppm to 6.0 ppm. The samples exceeding 1 ppm were obtained on plant property near the fenceline. 2. Twelve 24-hour integrated ambient air samples were collected at the fenceline on plant property. The VC values ranged from less than the detectable limit of 0.06 ppm to 1 ppm. 3. VC concentrations in three 24-hour composite waste water samples taken from the lagoon effluent ranged from 0.15 to 0. 29 ppm. 4. VC concentrations in two sludge samples taken from the lagoon near the outlet measured at the 0.05 - 0.06 ppm level on a wet basis. 26 GENC 000?'?4 5. The plant is located in a residential/industrial area on the edge of Leominster with residential developments adjacent to plant pro perty. 6. Shifting meteorological conditions and rain hampered the sam pling program. Region II: Flemington, New Jersey: Tenneco Chemicals, Inc. (PVC); May 29-31. 1. Forty-three discrete ambient air samples were collected on plant property and within a 2. 0 mile radius of the plant. The VC con centrations outside the plant property ranged from less than detecta ble (0.01 ppm) to 0.05 ppm. On plant property a single sample collected on the dryer building roof contained 5.6 ppm. At ground elevation, the VC concentrations on plant property ranged up to 0.30 ppm. 2. Twenty-three integrated ambient air samples were collected for 24-hour periods on plant property and within 2. 0 miles of the plant. The VC values ranged from 0. 005 to 0. 038 ppm on plant property and from less than detectable to 0. 031 ppm outside the plant area. 3. Two integrated one-hour ambient air samples collected within 0.1 mile of the plant showed VC at levels of 0.32 ppm and 0.18 ppm. 4. A maximum level of 20 ppm was detected in three 24-hour composite samples taken from the water effluent discharge into the Bushkill Brook, which immediately flows into the Raritan River. This amounts to approximately 400 lbs/day. 5. VC concentrations in sludge samples taken from the lagoon areas on plant property ranged from less than detectable to 1,000 ppm in wet weight concentrations; however, the concentration at the sludge disposal area was 54 ppm. 6. The plant is located in an area in which manufacturing facili ties are interspersed with farmland and relatively large acreage residential properties. There are a number of small communities within a few miles of the plant. Region III: Delaware City, Delaware: Stauffer Chemical Company (PVC) and Diamond Shamrock Chemical Company (PVC); May 20-22. S. Charleston," West Virginia: Union Car bide Corporation (PVC); May 24. 1. The air sampling and analysis activity was organized around a mobile laboratory equipped with a gas chromatograph using a flame ionization detector. VC levels were later confirmed by mass spectro meter. 2. A single discrete ambient air sample at the fenceline of the Diamond Shamrock plant showed 0. 2 ppm VC. 27 GENC 000795 3. Four discrete ambient air samples taken near the Stauffer Chemical plant ranged from 0.3 to 0.7 ppm VC. The highest level was recorded 0. 5 miles from the plant and the lower levels at 0. 25 miles from the plant. 4. The area immediately adjacent to the Delaware City complex is light ly populated residential areas for several miles. 5. Water samples collected at the Union Carbide plant gave VC values of 1.1 and 0.8 ppm for grab samples at several outfalls and 0.35 for a 24-hour composite. Samples obtained from the Kanawha River did not have a detectable level of VC. 6. Sampling was attempted but was not feasible due to limited time and equipment difficulties at the PVC plants of the Firestone Plastics Company in Perryville, Maryland, and Pottstown, Pennsylvania. Region IV: Louisville, Kentucky: B.F. Goodrich Chemical Company (PVC); March 19-21 and May 8-16. 1. The initial air monitoring program conducted in March was pre liminary to the more extensive program in May which showed significant ly higher levels. 2. In May there were 39 discrete ambient air samples collected in the area designated industrial (within 0.8 miles from the plant center). The VC concentrations ranged from less than 0.05 to 5.6 ppm, with 10 samples exceeding 1 ppm. In the area designated residential/ industrial, 149 samples were collected within 0. 8 miles of the plant with VC concen trations ranging from less than 0.05 to 33 ppm. The average concentra tions at the site registering 33 ppm were between 0.5 and 1 ppm, but 18 samples had concentrations greater than 5.0 ppm. Four samples were obtained in strictly residential areas with VC values of 0.05 to 1.6 being observed. The 1.6 value was 0.8 miles from the plant. 3. Five sampling sites were established within 0.6 miles of the plant for integrated air sampling over 24 hours. VC values ranged from less than 0.001 to 0.53 ppm. The highest value was obtained from a sampling site 0. 2 miles from the plant center. 4. Wastewater from the clarifier discharge was measured in March at 2 to 3 mg/1 in 24-hour composite samples. 5. Dewatered clarifier sludge and clarifier scum contained 193 and 162 ppm of VC, respectively. Region V: Painesville, Ohio: Uniroyal, Inc. (PVC) and Robintech, Inc. Inc. (PVC); May 9-14. 1. Four of 137 ambient air samples taken at distances up to 3.0 miles from the plant showed levels exceeding 1 ppm of VC with the highest level being 2. 26 ppm. Many of the samples were less than 0.1 ppm. 28 6EWC 000? 2. Nine 24-hour integrated ambient air samples taken at various dis tances from the plant showed levels up to 0. 2 ppm of VC. 3. VC levels in 11 of 17 water effluent samples were less than 0. 2 ppm. with three samples exceeding 1 ppm, including a high of 3.7 ppm. 4. VC levels in nine sludge samples, as the sludge would leave the plant property, ranged from 9 to 3520 ppm. 5. The complex is surrounded by residential areas. Region VI: Plaquemine, Louisiana: The Goodyear Tire and Rubber Company (PVC) and Dow Chemical Company (VC); April 7-9. 1. There were 31 discrete ambient air samples collected within 3.0 miles of the complex with VC concentrations ranging from less than detec table (. 001 ppm) to 7.81 ppm. Most of the readings were less than 1 ppm, with the highest value at the .property line. 2. VC concentrations in wastewater effluent measured by 24-hour com posites were all below . 05 ppm. 3. VC concentrations in residual reactor scrapings at the Goodyear plant ranged from 23 to 31 ppm. 4. The small communities of Morrisonville and Eliza are located less than 1 mile north and northwest respectively of the Goodyear plant. A few homes from Morrisonville extend almost to the north property line of the Goodyear plant. 5. Very limited air sampling was conducted in the Houston area in the vicinity of the plants listed below. However, in view of the inadequacy of this activity, the sampling effort in this area is being continued. Deer Park, Tex., PVC Plant - Diamond Shamrock Corp., Diamond Sham rock Chemical Co. Deer Park, Tex., VC Plant - Shell Chemical Co., Industrial Chemicals Division Houston, Tex., VC Plant - Tenneco, Inc., Tenneco Chemicals, Inc. Pasadena, Tex., VC Plant - Ethyl Corporation Region IX: Long Beach, California: B.F. Goodrich Chemical Company (PVC); American Chemical Corporation (VC); American Chem ical Corporation (PVC); May 7-10. 1. One hundred and eighty 10-minute integrated ambient air samples were collected within 3.1 miles of the complex. About 11 percent of the 29 GENC 0007 readings exceeded 0. 5 ppm, while 5 percent exceeded 1.0 ppm. The maximum value measured was 3.4 ppm in a sample taken 3.1 miles from the plant; however, the average level measured at this point was about 0. 5 ppm. 2. Samples of wastewater effluents were composited for 8 to 24 hours and yielded values from 3. 5 to 8.9 ppm, with individual samples reading up to 22 ppm. 3. Sludge samples showed values ranging from 290 to 4200 micrograms of VC per gram of dry sludge. 4. The complex is surrounded by residential areas. Within the three mile radius of the plants there are eleven schools. 30 GENC 000798 APPENDIX VI PERSISTENCE OF VINYL CHLORIDE The available information on the stability and persistence of VC in the environment is currently very limited. Some literature and laboratory studies have recently been initiated by industry and by EPA. This discus sion summarizes the findings of EPA to date and particularly the results of research efforts at EPA research facilities undertaken in response to the needs of the Task Force for at least preliminary data on environmental fate. Results of related experiments reported by industry seem to be consistent with the discussion. Behavior of Vinyl Chloride in Air The peak absorption of VC in the ultraviolet region is very far below the solar cutoff of about 2900 A, indicating that VC would not undergo reaction in sunlight in the absence of other reactive chemicals. When irradiated with simulated solar radiation in the presence of nitrogen oxides (nitric oxide and nitrogen dioxide), VC reacts to form a variety of products. The available laboratory results indicate a rate of reaction of about 8 to 10% per hour for VC, recognizing that reaction rates may vary with concentrations. The direct and indirect reaction products identified included ozone, nitrogen dioxide, carbon monoxide, formalde hyde, formic acid, and formyl chloride. High eye irritation levels were found with human exposure panels which is consistent with the products identified. The low reaction rate of VC, including reactions in the presence of nitrogen oxides, indicates that within a few miles downwind of VC emission sources VC will persist and can be considered a stable pollutant. The usual meteorological dispersion equations for gases could be applied to approximate concentrations. Because of temperature inversions and the absence of sunlight at night during the fall and winter, buildup of VC might be of particular concern during such periods. Clearly at greater distances from emission sources, VC will have greater opportunity to disperse and degrade. The noxious gases which are products of VC reactions should not be ignored. In air quality regions with large industrial activities involving large volume production of these chemicals, suchproducts may contribute appreciably on particularly sunny days to eye, nose, throat, and lung irri tation. Behavior of Vinyl Chloride in Water The loss of VC from water at constant temperature and pressure de pends on the rate of agitation or aeration. Distilled water in a beaker spiked with 16 ppm VC, when rapidly stirred at 22C with a magnetic stirrer, lost 96% of VC in two hours, while quiescent water at the same concentration lost only 25% VC. There was no significant difference in the rate of VC losses from distilled water, river water, or effluent from a VC plant stirred at the same rate, indicating negligible adsorption effects with particulate matter. Plots of log water concentration versus time give straight lines, indicating volatility to be the only important loss mechanism. 31 GEMC 00079? Hydrolysis over a pH range of 4.3 to 9.4 does not appear to be an im portant pathway for loss of VC from water. Chemical reaction of VC in the clarifier effluent from a VC plant was followed at 50PC for 57 hours at pH 4.3, 8.0, and 9.4 in sealed septum vials. Concentrations indicated that VC at these three pH values decreased at the same rate. This lack of pH dependence suggests that the loss of VC occurred by volatilization rather than hydrolysis, or at least there is a very slow hydrolysis rate. This experiment should be repeated in leak-proof reaction vials. Very preliminary experiments do not show photolysis as an impor tant pathway for loss of VC in water. However, there are many uncertain ties in the experimental techniques, and additional studies are needed in this area. Earlier theoretical studies are consistent with these experimental re sults. One study on the transfer of small non-reactive molecules across the air-water interface (as in stream aeration) used a kinetic approach to predict that VC will be rapidly lost from an aqueous solution, with the rate of loss being a function of water turbulence, mixing efficiency, and molecular diameter. Another study, using a thermodynamic approach, predicted a rapid rate of evaporation of low solubility chlorinated hydro carbons, including compounds of low vapor pressure. Despite the foregoing efforts there is a general absence of data con cerning VC in aquatic systems. It is conceivable that as the result of poor or erratic mixing in lakes or ponds, together with slow but con tinuous release of VC from sediments and sludges, VC could persist long enough to accumulate biologically, via direct absorption or via the food chain, or to cause other ecological effects. Behavior of Vinyl Chloride in Closed Rooms Tables 1 and 2 present data concerning concentrations of VC in a typical room following release of a pesticidal spray containing VC. TABLE 1 One Hundred and Twenty Second Release of Insect Spray in 133, 000 Liter Room SAMPLE TIME No. 1 Collected at breathing zone during spray No. 2 15 minutes COLUMN I VC FREON-12 41.64 ppm 8.15 ppm 16.91 3.13 COLUMN II VC FREON-12 41. 9 ppm 7.94 ppm 17.1 3.30 No. 3 No. 4 30 minutes 60 minutes 1.38 0.08 0.27 0.018 1.32 0.061 0. 25 0.018 No. 5 120 minutes 0.012 - 0.010 - 32 QyO'oO'J TABLE II Thirty Second Release of Insect Spray in 21,400 Liter Room SAMPLE TIME COLUMN I VC FREON-12* COLUMN II VC FREON-12* No. 1 Collected one minute after spray 380. 1 ppm 84. 8 ppm 383. 6 ppm 83. 2 ppm No. 2 30 minutes later 52.1 9.9 48.7 10.3 No. 3 60 minutes 24.6 4.8 22.5 4.7 No. 4 150 minutes 10.3 2.1 9. 3 2.2 No. 5 Collected in adjacent hall 151 minutes 0.83 0.17 0.17 0.15 *Freon-12 concentrations were determined using hydrocarbon response factors to compare dilution effects; the actual concentration is higher by a factor of 5.3. REFERENCES 1. Unpublished results of experiments and analyses conducted at EPA laboratories in Research Triangle Park, N. C., and Athens, Georgia, during April and May 1974. 2. Unpublished results of experiments on persistence of VC in water conducted by Dow Chemical Company. 3. Tsiroglou, E. C. and J. R. Wallace, "Characterization of Stream Reaeration Capacity," EPA Ecological Research Series Report #EPAR3-72-012 (October, 1972). 4. MacKay, Donald and Aaron W. Wolkoff, "Rate of Evaporation of LowSolubility Contaminants from Water Bodies to Atmosphere," Environ mental Science & Technology, 7 (7):611-614 (July, 1973). 33 GENC 000SOI APPENDIX VII HEALTH EFFECTS OF VC This Appendix presents much of the epidemiological and toxi cological data available as of August 1974, on the health effects associated with exposure to VC, together with a few interpretive com ments supplementing information presented in the body of the report. However, the Appendix does not present an exhaustive review or evaluation of available information. Table 1 summarizes the data, collected by CDC/NIOSH, on the confirmed cases of angiosarcoma of the liver in VC/PVC workers in the United States and abroad. A total of 15 occupational cases have been discovered in the United States and confirmed as angiosarcoma of the liver. Of the 15 cases, 2 are still alive and undergoing treat ment. Fourteen of the 15 were employed in PVC production plants and the remaining one in a PVC fabrication plant. The average age at death for the U. S. PVC production workers was 48.5 years (with a range from 36 to 61 years) which is about seven years younger than the average age of death from liver cancer in the U. S. male population. Based on the data available for the workers, the latent period for this disease appears to be on the order of twenty years, a period consistent with latencies observed for other occupational, chemically induced cancers. In the U. S. PVC production worker cases, all of the men were at one time "pot cleaners", required to enter the reactors in order to chip the residue of the chemical reaction from the sides of the "pots." Since the residue often contained pockets of trapped gases that were literally released in the cleaner's face when they were ruptured by his chipping operation, the potential for exposure to high levels of VC while cleaning these tanks was particularly great during the early years of this operation. Ten cases of worker-related angiosarcoma of the liver have been reported from five foreign countries to date. Table 2 summarizes the epidemiological data, collected by CDC from the Connecticut Tumor Registry, on five confirmed cases of angiosarcoma of the liver, including one accountant in a PVC fabrication plant and two residents; near PVC fabrication plants. The case of occupational exposure occurred in a man who had been employed for 10 years as an accountant in a factory which pro duces vinyl sheets and processes PVC resins; it is reported that he frequently visited the production area of the plant. Of the two cases who had no occupational exposure to VC or PVC, one was a 73 year-old man who lived his entire life within two miles of a PVC wire insulation plant. The other was an 83 year-old woman, a housewife and retired cook, who had lived for 35 years "within one-half mile of the vinyl products plant at which the accountant had been employed. 34 6040 000S02 While these findings establish no causal connection between exposure to PVC and angiosarcoma of the liver," they do raise the possibility of such a relationship. Time will He needed to define the possible risk factors in persons who have worked with PVC since the latency period appears to be so long. Because of the rarity of this tumor, the additional finding in this study of angiosarcoma of the liver in persons who had no occupational exposure to VC, but who may have had community exposure, is also worrisome but again establishes no causal connection. Epidemiologic investigation of additional cases of hepatic angiosarcoma that may be found to have had possible community exposure to VC will be necessary to clarify the significance of these cases. Tables 3A - 3D present the findings of the MCA-funded mortality study of VC /PVC workers, conducted by Tabershaw/Cooper Asso ciates. In calculating the risk of death, the usual method is to express the number of deaths which actually occurred as a percentage of the number which would have been expected in a comparable population observed over the same age and time intervals. This statistic is called the Standardized Mortality Ratio (SMR). Using the U. S. male population as the standard population of comparison, the SMRs were calculated for each of the 35 cases of death for which detailed mortaility rates are published on a national basis. In the standard population each SMR would be equal to 100. The statistical signifi cance of the deviation of each SMR in the study population from the exn^cted value of 100 was tested. A single asterisk indicates those SMRs which differed significantly from 100 at the 5 percent level, that is, which had a probability of . 05 or less of occurring by chance. A double asterisk indicates those which were significant at the 1 percent level. SMRs based on fewer than 5 observed cases were not tested for significance. The overall mortality of the study population is statistically significantly lower than that of the U. S. male population. There were 352 observed deaths compared with 467 expected, for an SMR of 75. For each job, an exposure score was estimated by industrial hy giene and safety personnel in each plant. A score of 1 was given for low exposure, 2 for medium, and 3 for high. The number of months each worker spent on a given job was multiplied by the appro priate exposure score. The total for each worker was then divided by the total number of months of exposure to give an Exposure Index (El) for that worker. Table 3A shows the SMRs for workers with an El below 1.5 versus those at 1.5 or above. The dividing point of 1.5 represents a level halfway between low and medium exposure. Table 3B shows similar results for workers with less than 5 years exposure versus those with 5 years or more. In order to examine the possible interaction between duration and level of exposure, the study population was divided into 4 groups on the basis of both El (low vs. high) and duration of exposure (short vs. 35 GEN'" 0008U long) using the same dichotomization as Tables 3 A and 3B. Table 3C shows the results for short versus long exposure in the low El group, and Table 3A shows the same comparison in the high El group. When the study population is divided according to length and duration of exposure (Tables 3A and 3B) and combinations of these measurements (Tables 3C and 3D), three major patterns emerge. For malignant neoplasms as a whole, the SMR increases with increasing exposure, whether measured by level, duration, or both. In the high exposure group with 5 years or more exposure (Table 3D) there are 36 observed cases and 26.11 expected. For cardiovascular - renal diseases as a group, there are also increases in the SMR with increasing exposure, but the number of observed cases remain less than expected, the differences being statistically significant in all groups except the high exposure, long duration group. For all other causes, there are no con sistent relationships with exposure. Within the malignant neoplasms, the largest (although not statisti cally significant) SMR is in cancers of the buccal cavity and pharynx, with 5 observed, 2.84 expected, and an SMR of 189. However, Tables 3A and 3D show that all these cases have an El below 1.5, and 4 out of 5 have less than 5 years exposure. Cancer of the digestive system shows no excess in the study popu lation as a whole. However, in those workers with Els of 1.5 or higher, there are 12 observed cases where 9.14 are expected (Table 3A). In the subgroup of the above workers with 5 years or more exposure, there are 11 observed cases and 7.47 expected. Respiratory cancer shows a slight excess in the total group, and a similar pattern for different exposure categories, with 13 observed versus 10.28 expected when the El is 1.5 or higher, and 12 observed versus 8.50 expected when, in addition, the duration of exposure is 5 years or more. Malignant neoplasms of other and unspecified sites show an excess in the total group, and an increase with both level and duration of exposure (Tables 3A and 3B). The relationship with exposure is more pronounced, since those with exposures of less than 5 years have fewer cases than expected. The lymphosarcomas, although occurring at about the expected rate when the whole group is considered, are concentrated almost entirely in the high exposure long duration group. In that category there are 4 cases observed and 1.84 expected. The Tabershaw/Cooper Study is based on an examination of 328 death certificates. The authors acknowledge three areas where bias might have entered: (a) choice of the U. S. male population as the 36 G0-C standard, (b) absence of 15% of the study population (untraceable), and (c) discovery, as the study ended, of a group of 1500 workers whose exposures occurred up to 35 years ago and who are not included in the study group. Since the latency period fo r angiosarcoma of the liver is averaging 18 years at least, it would appear desirable to examine the data for these 1500 workers. In addition to the Tabershaw/Cooper study several other epidemio logical studies presented during the recent OSHA hearings suggest the possibility of a multiple cancer risk. Table 4 summarizes many of the published and unpublished toxi cological and epidemiological studies of human and animal exposures to VC. A list of the references cited in Table 4 completes this Appendix. 37 GENC 000905 Occupation 1. VC Monomer Production 2. PVC Polymerization 3. PVC Compounders, Fabricators, Etc. Country Sweden United States United States United States United States United States United States United States United States United States United States United States United States United States U. Germany W. 'Germany Great Britain Norway Sweden Czechoslovakia Czecho s1ovakla United States United States Great Britain Table 1 OCCUPATIONAL CASES OP LIVER ANGIOSARCOMA Case f BIRTH DATE 1st VC/PVC Diagnosis Work of Angiosar Age at coma Diagnosis 01 00-00-11 00-00-45 00-00-72 61 01 00-00-22 12-09-48 03-00-71 49 02 00-00-34 11-15-55 05-00-70 36 03 00-00-15 11-28-45 12-00-73 58 04 00-00-24 07-06-52 08-00-67 43 05 00-00-12 06-19-44 04-00-64 52 06 00-00-29 01-17-62 02-00-74 45 07 05-03-22 08-00-44 00-00-68 45 03 05-06-20 10-07-46 08-00-61 41 09 00-00-31 05-28-45 03-01-74' 43 10 08-16-13 06-00-51 05-00-68 55 11 05-27-09 10-14-46 03-00-70 61 12 11-17-18 09-13-49 05-00-69 50 13 12-01-21 08-19-44 05-00-74 53 01 07-26-31 10-14-57 00-00-71 40 02 06-04-30 10-01-57 00-00-69 39 01 00-00-01 00-00-46 12-00-72 71 0] 12-23-15 03-00-50 12-20-71 56 02 00-00-27 00-00-51 00-00-70 43 01 02 14 11-04-27 11-11-51 00-00--69 41 15 00-00-25 00-00-00 07-00-72 47 02 00-09-14 nruivju4 02-00-70 55 Yrs. 1st Total Yrs. VC/PVC Work VC/PVC To Diagnosis Exposure 27 23 22 16 14 13 28 28 15 15 20 IS 12 12 24 18 15 15 29 17 17 17 23 23 20 15 30 30 14 14 11 11 26 20 22 21 19 18 17 4 00 00 24 11 Date of Death 00-00-72 03-03-73 09-28-71 12-19-73 01-07-68 04-09-64 Alive 03-23-68 OB-29-61 Alive 05-10-68 03-16-70 05-02-69 07-04-74 12-14-71 01-25-69 12-00-72 01-04-72 00-00-70 03-27-65 02-15-73 12-00-70 4. Other VC Exposure W. Germany Note: '001 indicates unknown date UJ 43 SOURCE: NIOSH 14 Table 2 CASES OF HEPATIC ANGIOSARCOMA, CONNECTICUT, 1935-1973 Case NCI No. Age Sex Diagnosis Date of Original Date of Diagnosis Death Medical History 1 73 M Hepatic 11-25-67 12-3-67 2 months history of diarrhea, , anorexia, and Angiosarcoma 20 lb weight loss. Intermittent abdominal pain. Non-tender, firm epigastric mass. Died after 9 days with spontaneous ruptured liver leading to shock. Past history of alcohol In take. Occupation Place of Residence Fireman 1917-42 Aluminum worker 1942-44 Corset cutter 1945-61 Retired 1961-67 Bridgeport entire life 2 47 M Alcoholic 1-15-73 2-15-73 Initial symptoms RUQ abdominal pain with Accountant - Vinyl Co., Bridgeport - Cirrhosis vomiting. Cecal volvulus found, Rx cecopexy, 1963-73 1956-73 over next 6 weeks pain continued with weak Accountant - Plastic Previously many Portal ness. RUQ tenderness with 2 FB liver. Diagnosed Belt Co., 1956-63 locations Fibrosis by needle biopsy on 1-15-73. Deteriorated slowly Previously accountant- until death 31 days later. other states 3 83 F Hepatic 12-19-73 1-22-74 Admitted 12-2-73 with short history of RUQ Angiosarcoma abdominal pain radiating to R shoulder. Had Housewife Stratford Restaurant cook 35 yra 35 years RUQ tenderness. Open liver biopsy 12-19-73 showed large tumor. No resection. Deteriorated until death 34 days later. 4 76 F Hepatic 3-12-50 3-19-50 1 month history of anorexia with abdominal pain Housewife Angiosarcoma and back pain. Firm epigastric mass. Died 6 days after admission with carcinomatosis and pulmonary emboli. Windsor Locks 5 50 M Hepatic Angiosarcoma 5-4-73 Admitted for abdominal pain and jaundice 3-27-73. 4 FB liver. Discharged. Readmitted 4-29-73 with abd distension, general edema. Icterus, fever, shaking chills. Rapid down hill course with death due to renal and hepa tic failure, Past history of alcohol intake. Fisherman and carpen ter before 1959 Plas terer 1959-60. Unem ployed 1960-73. Puerto Rico 1923-59 New York City 1959-73 Bridgeport 1973 Table 3A OBSERVED DEATHS/EXPECTED DE.iTHS AND STANDARDIZED MORTALITY RATIOS IN VINYL CHLORIDE WORKERS, BY ESTIMATED LEVEL OF EXPOSURE Cause of death with I. C.D.+number El <1.5 obs/exp SMR1 121. 5 oba/exp S(*1 All causes 188/270.33 70** 157/195.68 60** Tuberculosis (001-019) Tuberculosis of respiratory system (001*008) Malignant neoplasms (140*205) Malignant neoplasma, buccal cavity and pharynx (140-146) Malignant neoplasms, digestive organa end peritoneum (150*159) Malignant neoplasms, respiratory system (160*164) MelIgnant neoplasms , genital organs (170*179) Malignant neoplasms, urinary organs (180-181) Malignant neoplasms, other and unspecified sites (190-199) Leukemia and aleukemia (204) Lymphosarcoma, lymphatic and hematopoietic tissues (200-203, 205) Diabetes mellitue (260) Major cardiovascular and renal diseases (330-334, 400*468, 592-596) Vascular lesions affecting CHS (330-334) Rheumatic fever & chronic rheumatic heart dls. (400-402, 410*416) Arteriosclerotic heart disease (420) Honrheumettc endocarditis (421, 422) hypertensive heart disease (440*443) Other hypertensive disease (444-447) Chronic & unspecified nephritis & renal sclerosis (592*594) Influenza end pneumonia (480*493) Ulcer of stomach and duodenum (540, 541) Appendicitis (550-553) Hernia and Intestinal obstruction (560, 561, 570) Gastritis, duodenitis, enteritis and colitis (543, 571, 572) Cirrhosis of liver (581) Hyperplasia of prostate (610) Symptosm, senility and ill-defined conditions (780-795) All other diseases (residual) M>tor vehicle accidents (810-83S) Other accidents (800-802, 840-962) Suicide (963, 970-979) Homicide (964, 980-965) 0/3.38 0/3.16 37/44.28 5/1.62 7/12.50 11/13.56 2/2.30 1/2.07 9/6.57 1/2.18 1/3.48 5/3.65 84/120.11 7/14.42 3/3.98 68/78.94 0/4.20 1/5.48 1/1.52 0/2.50 5/5.80 1/2.21 0/0.39 0/0.68 0/0.76 2/8.90 0/0.25 0/4.22 14/21.90 8/19.08 11/17.83 9/9.73 0/6.98 0 0 90 330 60* 86 93 51 146 49 31 146 75** 52** 80 92 0 19 70 0 92 48 0 0 0 23 0 0 68* 45** 66* 98 0 0/2.33 0/2.18 41/32.67 0/1.21 12/9.14 13/10.28 1/1.43 0/1.52 8/4.52 2/1.57 5/2.54 2/2.65 69/86.99 6/10.06 2/2.85 51/58.05 1/2.89 2/3.86 2/1.07 0/1.77 0/4.13 1/1.60 0/0.27 1/0.63 1/0.55 1/6.64 0/0.14 1/3.09 6/15.89 9/13.46 6/12.67 7/7.02 1/4.94 0 0 134 0 141 135 75 0 190 136 212 81 85* 64 75 95 38 56 201 0 0 68 0 171 196 16 0 34 41** 72 50** 107 21 Number of workers Person-years 4032 45354 3057 32108 ^SMt's adjusted for deaths with cause unknown. ^Significant at SI level. **SignlfIcant at 11 level. +International Classification of Diseases SOURCE: Tabershaw Cboper Associates, Inc., Epidemiological Study of Vinyl Chloride Workers. Final Report 6ENC 00080 Table 3B OBSERVED DEATHS/EXPECTED DEATHS AND STiiNB.'itDIZED HORTALITY RATIOS IH VINYL CHLORIDE WORKERS BY DURATION OP EXPOSED BHFLOYtCNT Cause of death with I.C.D. number All causes Tuberculoela (001-019) Tuberculosis of respiratory system (001-008) Malignant neoplasms (140-205) Malignant neoplasaa, buccal cavity and pharynx (140-148) Malignant aeopleans, digestive organa end perltonaus (150-159) Malignant neoplasms, respiratory system (160-164) Malignant neoplasms, genital organs (170-179) Malignant neoplasms, urinary organa (180-lfll) Malignant neoplasms, other and unspecified sites (190-199) Leukemia and aleukemia (204) Lymphosarcoma, lymphatic and hematopoietic tissues (200-203, 205) Diabetes mellltue (260) Major cardiovascular and renal diseases (330-334, 400-468, 592-594) Vsscular lesions effecting (MS (330-334) theumacic fever & chronic rheumatic heart dls. (400-402, 410-416) Arteriosclerotic heart dleease (420) Nonrheumatic endocarditis (421, 422) Hypertensive heart dlaaaae (440-443) Other hypertensive disease (444-447) Chronic & unspecified nephritis & renal sclerosis (592-594) Influence and pneumonia (480-493) Ulcer of stomach and duodenum (540,541) Appendicitis (550-553) Hernia and Intestinal obstruction (560, 561, 570) Gastritis, duodenitis, enteritis and colitis (543, 571, 572) Cirrhosis of liver (581) Hyperplasia of prostate (610) Symptoms, ssnility and ill-defined conditions (780-795) All other diseases (residual) Motor vehlcla accidents (BIO-835) Other accidents (800-802, 840-962) Suicide (963, 970-979) Homicide (964, 980-985) Number of workers Person-ysars ^SMt's adjusted for deaths with cause unknown. *SlgnIfleant at 5X level. **Slgnifleant at IX level. <60 months obs/exp SMI* 94/140.53 67** 0/2.23 0/2.07 13/19.96 4/0.70 2/5.26 3/5.52 0/0.99 0/0.83 2/3.40 1/1.25 1/2.01 2/1.80 .28/51.45 4/5.97 2/2.39 21/32.54 0/1.79 1/2.42 0/0.79 0/1.55 3/2.93 1/1.07 0/0.24 0/0.42 1/0.40 1/4.49 0/0.07 1/2.28 4/11.97 10/16.14 7/12.94 6/6.34 1/5.80 0 0 78 688 46 65 0 0 71 96 60 134 65** 81 101 78* 0 49 0 0 123 112 0 0 301 26 0 51 40 75 65* 114 20 2955 34201 ^60 months obs/exp SW1 251/329.30 76** 0/3.55 0/3.33 65/57.61 1/2.16 17/16.56 21/IB.51 3/2.76 1/2.79 15/8.23 2/2.53 5/4.07 5/4.54 125/157.39 9/18.71 3/4.53 98/105.39 1/5.35 2/7.01 3/1.82 0/2.76 2/7.09 1/2.78 0/0.43 1/1.10 0/0.92 2/11.18 0/0.33 0/5.09 16/26.34 7/16.65 10/17.82 10/10.28 0/6.20 0 0 116 47 106 116 112 37 187 81 126 113 81** 49** 68 96 20 30 170 0 29 37 0 93 0 18 0 0 63* 43** 58* 100 0 4134 43240 Table 3C OBSERVED DEATHS/EXPECTED DEATHS i.ND STANDARDIZED MORTALITY RATIOS IK VINYL CHLORIDE WORKERS WITH EXPOSURE INDICES BELOW 1.5, BY DURATION OP EXPOSED EMPLOYMENT Cause of death with I.C.D. number All cauaes Tuberculosis (001-019) Tuberculosis of respiratory system (001-008) Malignant neoplasms (140-205) Malignant neoplasms, buccal cavity and pharynx (140-148) Malignant neoplasias, digestive organs and peritoneum (150-159) Malignant neoplasms, respiratory system (160-164) Malignant neoplasms, genital organs (170-179) Malignant neoplasms, urinary organs (180-161) Malignant neoplasms, other and unspecified sites (190-199) Leukemia end aleukemia (204) Lymphosarcoma, lymphatic and hematopoietic tissues (200-203,205) Diabetes mellltua (260) Major cardiovascular and renal diseases (330-334, 400-468, 592-594) Vetculer lesions sffsctlng (US (330-334) Rheumatic fever & chronic rheumatic heart dla. (400-402, 410-416) Arteriosclerotic heart disease (420) Honrheumatl'c endocarditis (421, 422) Hypertensive heart disease (440-443) Other hypertensive disease (444-447) Chronic & unspecified nephritis & renal sclerosis (592-594) Influents and pneumonia (480-493) (fleer of stomach and duodenum (540, 541) Appsndlcltlt (550-553) Hernia and intestinal obstruction (560, 561, 570) Gastritis, duodenitis, enteritis and colitis (543, 571, 572) Cirrhosis of liver (58t) Hyperplasia of prostate (610) Symptoms, senility and ill-defined conditions (760-795) All other diseases (residual) Motor vehicle accidents (8LO-835) Other accidents (800-602, 840-962) Suicide (963, 970-979) Howl cldc (964 , 980-985) Number of workers Person-yer *SWs adjusted tor deaths with cause unknown. *Signific*nt at 5X level. **Slgnillcant at IS level. <60 months exposure *60 months exposure obs/exp SMRr obs/exp Stftl 56/89.23 63** 132/181.28 73** 0/1.41 0/1.31 8/12.86 4/0.45 1/3.43 2/3.58 0/0.68 0/0.55 1/0.97 0/0.79 0/1.26 2/1.15 21/33.38 2/3.93 2/1.50 16/21.14 0/1.18 1/1.58 0/0.50 0/0.97 3/1.86 0/0.68 0/0.15 0/0.27 0/0.26 1/2.80 0/0.05 0/1.43 4/7.45 3/9.87 3/7.98 3/4.08 0/3.55 0 0 73 1036 34 65 0 0 120 0 0 203 73** 59- 155 8B 0 74 0 0 IBS 0 0 0 0 42 0 0 63 35 44 86 0 0/1.97 0/1.85 29/31.46 1/1.17 6/9.08 9/10.00 2/1.62 1/1.53 8/4.43 1/1.40 1/2.23 3/2.50 63/86.82 5/10.51 1/2.49 52/57.87 0/3.02 0/3.90 1/1.02 0/1.53 2/3.95 1/1.53 0/0.24 0/0.61 0/0.51 1/6.09 0/0.20 0/2.79 10/12.92 5/9.22 8/9.85 6/5.66 0/3.43 0 0 95 88 68 93 127 67 187 73 46 124 75** 49* 41 93 0 0 101 0 53 67 0 0 0 16 0 0 79 56 83 109 0 1715 21418 2317 23920 GENC 000810 Table 3D OBSERVED DEATHS/EXPECTED DEATHS t HD STANDARDIZED MORTALITY RATIOS IH VINYL CHLORIDE WORKERS WITH EXPOSURE INDICES OF 1.5 OR GREATER, BY DURATION OP EXPOSED EMPLOYMENT Cause of death with I.C.D. number All causes Tuberculosis (001-019) Tuberculosis of retporstory system (001-008) Malignant neoplasms (140*205) Malignant neoplasias, buccal cavity and pharynx {140-148) Malignant neoplasms, digestive organs and peritoneum (150*159) Malignant neoplasms, respiratory system (180*144) Malignant neoplasms, genital organa (170*179) Malignant neopleama, urinary organs (180-181) Malignant neoplasms, other and unspecified sites (190-199) Leukemia and aleukemia (204) Lymphosarcoma, lymphatic and hematopoietic tissues (200*203, 205) Diabetes mellltus (260) Major cardiovascular and renal diseases (330*334, 400-468, 592*594) Vascular leslona affecting (MS (310-334) Rheumatic fever A chronic rheumatic heart dls. (400*402, 410*416) Arteriosclerotic heert disease (420) Bfonrheusatlc endocarditis (421, 422) Hypertensive heart disease (440*443) Other.hypertensive dlseeee (444-447) Chronic 8 unspecified nephritis 6 renal sclerosis (592-594) Influenza and pneumonia (480-493) Ulcer of stomach and duodenum (540, 541) Appendicitis (550-553) Hernia and Intestinal obstruction (580, 561, 570) Gastritis, duodenitis, enteritis and colitis (543, 571, 572) Cirrhosis of liver (581) Hyperplasia of prostate (610) Symptoms, senility and ill-defined conditions (780-795) All other diseases (residual) Motor vehicle accidents (810-835) Other accidents (800-80Z, 840-962) Suicide (963, 970-979) Homicide (964, 980-985) Humber of workers Person-years ^SMR's adjusted for deaths with cause unknown, `significant at 51 level. "Significant at U level. <60 months exposure 260 months exposure obs/exp SHR1 obs/exp s*1 38/47.93 79 119/147.81 81* 0/0.76 0/0.71 5/6.57 0/0.23 1/1.67 1/1.79 0/0.29 0/0.26 1/1.18 1/0.44 1/0.71 0/0.61 7/16.54 2/1.87 0/0.82 5/10,41 0/0.57 0/0.76 0/0.27 0/0.54 0/0.99 1/0.35 0/0.08 0/0.14 1/0.14 0/1.56 0/0.01 1/0.80 0/4.02 7/6.05 4/4.73 3/2.40 1/2.18 0 0 96 0 76 71 0 0 107 288 178 0 54" 135 0 61* 0 0 0 0 0 362 0 0 904 0 0 158 0 146 107 158 58 1240 12828 _____ 0/1.57 0/1.48 36/26.11 0/0.99 11/7.47 12/8.50 1/1.41 0/1.26 7/3.51 1/1.13 4/1.84 2/2.04 62/70.46 4/8.19 2/2.04 46/47.65 1/2.32 2/3.10 2/0.81 0/1.23 0/3.13 0/1.25 0/0.19 1/0.49 0/0.41 1/5.08 0/0.13 0/2.30 6/11.88 2/7.43 2/7.96 4/4.62 0/2.76 0 0 141 0 151 144 73 0 204 90 222 100 90 50 100 98 44 66 253 0 0 0 0 209 0 20 0 0 51* 28 26 88 0 1817 19305 GENr nr Table 4 SUMMARY OF TOXICOLOGICAL AND EPIDEMIOLOGICAL STUDIES ON VINYL CHLORIDE Authors____________ Species Von Oettlngen (1955) Human Gabor Kecca-Radu Manta (1962) Chem. Abstract Lester Greenberg Adams (1963) Human Human Gabor Radu Preda Abrudean Juanof Anca Valczkay (1964) Chem. Abstract Human Grlgorescu Toba (1966) Chem. Abstract Human Sex No. EXPOSURE Hrs. per Day___________ Days HUMAN DATA Cone. Total Dose ppm______ ppm-Days___________________Observations 12,000 10,000 25,000 Dangerous Narcosis Produced symptoms of dizziness, disorientation, headache and burning sensation on Boles of feet. S2 Workers exposed to DDT, Senzene, Hexachlorocyclo- hexane, VC, PVC. Blood: Decrease in catalase Increase In peroxidase, indoplienoloxldase and gluthathione Changes occurred during second year of work. M 3 Twice per day F 3 for 3 days. 0 4,000 0 83.3 1/5 slightly dizzy 0/6 had any effects 5 min. sessions 8,000 166.7 1/6 slightly dizzy at 6 houra in 12,000 250.0 2/6 definitely dizzy tervals 16,000 333.2 5/6 dizzy, nausea, blurred vision and heaving symptoms stopped after exposure 20,000 416.7 6/6 Intoxicated, one with persistent headaches 50X level of no effect is 1*31 No statement about repeated exposures 78 PVC Workers Decrease plasma albumin Increased B and 8 globulin Decrease In B/8 for serum lipoproteins Decrease In serum cholinesterase Decrease in pseudo cholinesterase Normal blood catalase Normal serum pyruvic acid Experimental: PVC Workers Control: Other clinically healthy people Hypothesis: VC+RjC^chioral + chloracetlc acid (1). Results: (1) was found in SOX of exptl. people, but Jn none of controls. Most of + findings were in people exposed 2-5 years., In these cases,cG-globulin is higher,^globulin Is lower than people with no (1) in urine> tenacity to metabolize (1) decreased after 2 years. Pathology None reported None reported None reported None reported Authors Species Sex No. EXPOSURE Rrs. Total per Cone. Dose Day Days ppm ppm-Days HUMAN DATA ObservatIons Pathology Karris Adams (1967) Human 2-- -- One worker had knee cap and toes involved in the acro-osteolysls. Other worker only hands* HiIson McCormick Tatum Creech (1967) Human M 31 - - -- No cases of acro-osteolysls diagnosed in 1000 individuals who handled finished resin or used for plastic product production Age range of affected workers 26-47. Incubation period greater than 12 months of polycleaning experience. 31/3000 (32) workmen associated VC polymerization founds to have aeroosteolysis. 22/31 Acro-osteolysls associated with Reynaud's sumptons* Bareeta Stewart tfutchler (1969) Human M 13 7.5 1 50 15.6 250 78.1 500 156.2 Breath decay curves, 0 to 20 hrs. after exposure were measured. Level In breath at 4 hrs. las' 12. No adverse effects noted. About the same set of breath decay curves following occupational exposure. None Reported Kudryautseva Human (1970) Abstract M 50 - - F 63 ' -- 1. Changes In ECG: rhythm, conductance, polarization. None Reported. 2. Increase In systolic index. Viola (1970) Unpublished Human -1 5 15 500* 8 8 1095 1825 - *ln several other factories Acro-osteolysls symptoms, Reynaud's syndrome, aversion to fats, enlarged liver Raynaud's syndrome Enlarged liver, minor liver Insufficiency. 13/500 had acro-osteolysls. Olefactory thres hold is 0.0 to 12. Acute nervous symptoms become evident when It Is easily perceptible. Aeronetry: (VC) on factory filters at air discharge tine: 2,000 ppm (VC) at point of -worker entry: 2)000 ppm in plants where acroosteoivais occurred-150 ppm (max) in plant with no disease [VC] on gangway and other parts of plant: 10 to 15 ppm. HUMAN DATA Authors______ Soeeles Sex Ho. Hrs. Total per Cone. Dose Day______Days ppm ppm-Days_____________ Observations Pathology_________________ Ditman Hunan 5011 21,510 Kan-yeara Experience Conditions associated with hand cleaning of 25 cases of acro-osteolysls, Cook Waterhouse Magnuson Dltcheck (1971) polymerlzers. There appeared to be correlation between reactor degassing time and acro-osteolysls. 16 other individuals questionable Acro-osteolysls appears to be systemic rather than local disease. Reynaud's phenomenon was statis tically related to acro-osteoly sis. Dodson Human M Dlnnan Whltehouse Nasr Magnuson (1971) On 4 1-23 Months All patients had worked as PVC reactor-vessel cleaners. Neg. Ca and P balance In one subject. PiethysinographIc abnormalities were present In 3 subjects. Esophageal motility within normal limits. Catecholamine, -HydroxyIndole Acetic Acid excretion normal. Reynaud's phenomenon anteceded osteolytic lesions in all four subjects. 1&P scintiscans correlated with radiographic lesions.1 No liver enlargement or hypothy roidism. Additional smaller abnormalities All other numerous clinical laboratory investigations found in ulnar styloid, oscalcls negative. and patella* Kramer Kutchler (1971) Human M 98 -Up to 25 years Experience Performed statistical correlation between Beveral clinical measurements and total dose and timeweighted average VC concentration. None reported. 2 liver function Indices show s positive correlation with total dose (abnormally high). a) Icterus Index b) Bromsulphaleln 3 other Indices are dose-related but are not outside normal limits: a) systolic and diastolic blood pressure. b) hemoglobin negative correlation. c) beta-protein. Authors Keyerson Meier (1972) Abstract Species Human Sex Ho. i Markowit2 McDonald Fethlere Kerzner (1972) Abstract Human Lange Juhe Stein Veltman (1973) Human 13 HUMAN DATA EXPOSURE Cone. Total Dose Hrs/Dav Days oom. oom-davs OBSERVATIONS PATHOLOGY Acroosteolysls Had unique papular skin lesions which have been described only in PVC workers Describes acroosteolysls symptoms. Incidence Is <3Z among workers Ages 29-32 Reactor cleaners 2-18 yrs. employment Latent period: lh-$i years in 11 patients; 7 and 11 years In other 2 patients. Acro-osteolyals symptoms. Peripheral vessel stenosis. Thrombopeny (low count) Is the first objective symptom described In all patients; lung fibrosis orglnating in portal system, large spleen, impaired lung function; 101 mortality. This is the first objective symptom described* HUMAN DATA Authors Marsteller Lelbach Miller Juhe lange Pohner Veltman (1973) Species Human Sex No. - 120 Hrs. perDay EXPOSURE Cone. Days ppm Total Dose ppm-Davs Observations l*i to 21 years. 20 PVC workers were studied out of 45 with suspected skin problems, 30 to 56 years old ________________ Pathology___________________ Liver enlarged In 13/20. Pain In Rt. upper abdomen in 2/20. Hyperlipidemia has been diagnosed In 1967 after 6 years In 1/20. Jaundice history in 1955 before exposure In 1968, liver dysfunc tion was diagnosed, no alcoholism. Spleenomegaly In 7/20. Total bilirubin was 1 mg/100 (which Is upper normal limit) In 3/20* Bromsulphaleln test was abnormal In 19/20 (>5Z retention after 45 minutes). SGOT was >12 mU/ml In 17/20. SGPT was elevated (15-30) in 14/20. Alkaline phosphatase was >48 uU/ml In 2/20 Hypothroabocytemla (<150x 103 /m3) vaa found 19/20100X103/n.3) inl2/ Acro-ostyeolysls was seen In 4/20. Varicose veins of esophagus In 3/20 Liver histology; Collagen transfor mation of walls of sinusoids in 5/20. Focal activation of Kupfet cells In 19/20. Focal fatty infiltration in 14/20. Fibrosis of septa and capsule intralobular and portal spaces in 17/20. 15 additional blood parameters were nor mal. 9 lmunologlcal tests were done once, not repeated}. Table 4 SUMMARY OF TOXICOLOGICAL AND EPIDEMIOLOGICAL STUDIES ON VINYL CHLORIDE ANIMAL DATA Authors ANIMALS , EXPOSURE Cone. Total Dose Species Sex No. Hrs/day Days___________ ppm_________ ppm-days_______________________________Observations___________________________Pathology Von Oettingen (1955) (Review Article) Cats Cats Cats ND ND ND ND ND <4 ND ND <4 ND 1 L Cats Cats ND ND <4 ND ND 4 1 1 Cats ND ND ND ND -sFXS> Rabbits ND ND 1 rain. Dogs 1 ND ND VC is promptly excreted by lungs; B2X Is eliminated after inhalation stops 100,000 to <1,200 130,000 180,000 <30,000 200,000 <33,000 250.000 to<40,000 to 300.000 50,000 ND ND 170,000 118 Blood VC concentration reaches 15*17 mgl This concentration causes same lntra-aurlcular pressure reduction as 13,000 ppm dichloroethylene, 30,000 ppm ether Cardiac insufficiency Even this does not produce complete cardiac failure Blood levels are 40 mgX at time of cardiac arrest, 27*30 mgZ at time of respiratory arrest This is the narcotic concentration Mice Dogs Dogs ND ND ND Dogs ND Mice ND G. Pigs ND ND 1 min. ND' <4 1 1 86,000 to 60 to 85 This is the narcotic concentration 123,000 100,000 <12,000 Cardiac irregularities, ECG abnormalities ND 3 7 10,000 475 No major change in liver or kidney for several wks ND 3 7 for 200,000 9,500 Marked salivation, vomiting, respiratory arrest several wks ND 10 min. 1 245.000 to 1,700 to This Is the lethal range for 10 minutes exposure 295.000 2,040 ND short 1 200,000 to ND All killed time 400,000 u<josi? ANIMAL DATA ANIMALS Authors Species Sex No. Hrs/dav Days Von Oettingen <1955) Review Article Continued G. Pigs ND ND 0.5-1 G. Pigs ND ND 0.5-1 G. Pigs ND ND 0.5-1 EXPOSURE Cone.' Total Dose ppmppg-daygObservations________________________________________________ Patho logy 1 100,000 2.000 to Dangerous to life 4.000 1 5,000 100 to Higher concentrations than this cause severe 200 lung edema and hyperemia of liver and kidney ND Order of toxicity is: carbon tetrachloride-^ chloroform < VC Methyl chloride Mastromatteo Fisher Christie Danzlger (1969) Fisher Christie Danxiger (I960) Mice ND 5 0.5 1 Rats G. Pigs ND 5 ND 5 0.5' 0.5 1 1 Humber of animals and duration of exposure same as above 100,000 2,080 200,000 4,160 Torkelson (1) Oyen Rove (19AL) Number of animals and duration of exposure same as above 300,000 6,250 Number of animals and duration of exposure same as above Rats M 10 7 (5d/wk) F 10 7 (4.5Mo) 400,000 8,330 500 14,000 Sequential effects were: 1. irritation, 2. increased motor activity, 3. twitching, A. tremor, incoordination, 5. unconscious, 6. deep narcosis. All animals recovered In 5 minutes. Mice: Light lung engorgment, kidney swelling; Rats and G. Pigs: same lung picture. 1/5 mice died after 30 minutes, same symptoms as above but appeared sooner. Guinea pigs unsteady for 20 min. after exposure. 5/5 mice and 5/5 rats died; 1/5 guinea pigs died; 4/5 guinea pigs recovered In 25 minutes. 2/5 guinea pigs died Growth and gross appearance were normal. Liver/body weight ratio and absolute liver weight larger than control in males. Liver/body weight and absolute liver weight not larger than control in females Blood SGOT, SGPT, SUN, alkaline phosphatase were normal. Lung engorgement, no edema in all species. One rat had fatty liver Liver and kidney was congested, tracheal epithelium damaged Same symptoms, more severe Central lobular liver degeneration. Kidney tubular damage. ANIMAL DATA Authors ANIMALS Suedes Sex No. Hrs/dav Torkelson Oyen Rove (1961) continued M5 7 F57 (2a) Rats H 12 7 F 12 7 G. Pigs M 10 7 F 87 Rabbits M 37 F 37 Dogs M 17 F 17 Matched controls, both exposed and onexposed groups Days (5d/wk) (4.5mo) EXPOSURE Cone. Total Dose ppp p pa-day a_______________________ Observations 00 Control animals 138 exposures In 204 days (6*4 months) 200 200 200 200 5d/vk 8,050 to 8,400 All groups ware normal in appearance, mortality and growth. Hematology (hemoglobin, hematocrit, cells) was normal. Liver function tests (SUN, SCOT, SGPT, alkaline phosphatase) were normal. All organ/body weight ratio normal except M and F rats, where liver/body weight ratio was increased. Pathology Gross pathology was normal. Microscopic pa thology was nor mal In all species except liver of M and P rabbits; Central lobular granular degenera tion and necrosis. (2b) Same protocols 100 4,000 to 4,200 All animals were normal In appearance, mortality and growth. Gross and micro Hematology (hemoglobin, hematocrit, cells) scopic appearance was normal. Liver function teats (SUN, SGOT, SGPT, alka line phosphatase) were normal. Liver/body Of tissues were normal. weight ratio of M and F rata were larger than controls. ANIUAL DATA ANIMALS EXPOSURE Authors Species Sex No. Hrs/day days Cone. Total Dose ppm oom-davs Observations Torkelson Oyen Roue (1961) (2c) Rats M54 MS 2 M5 1 M 5 0.5 M54 M5 2 M51 K 5 0.5 (3) Rats M 24 F 24 7 7 G. Pigs M 12 7 F 12 7 NLA> Rabbits M F 1 3 7 7 Dogs H1 7 F 1 '1 Matched controls, exposed and unexposed groups- 5d/wk for 6.5 months as above 200 4,600 200 2,300 200 1,150 200 575 100 2,300 100 1,150 100 575 100 265 no exposure in 189 days Liver/body/ weight ratio larger than controls, not statistically signifi- cant. M M" """ M " Liver/body weight ratio Bante as controls " M" " tr '* M Liver/body weight ratio higher than controls, not statistically slgnlfl- cant. " ,T " M 11 M 11 M " Normal in all respects II II tl II All parameters normal In all species* Lester Sherman ND 2 0-2 1 Greenberg rats ND 2 0-2 1 Adams Cl) nd 2 0-2 1 (1963) ND 2 0-2 1 ND 1 5 min. 1 42 mil. 1 ND 1 2 1 50,000 60,000 70,000 100,000 150,000 150,000 0-4,160 0-5,000 0-5,830 0-8,330 552 4,380 12,500 Moderate intoxication, righting reflex lost. More intense intoxication, righting reflex present More intense intoxication, righting reflex lost. Corneal reflex disappears* no gross pathology* Deep anesthesia. Respiratory failure of same animal. Deep anesthesia, complete recovery after exposure. No pathology observed. GEHC 000820 Authors EXPOSURE Hrs. per Cone. Suedes Sex. Ho. Dav Days ppm ANIMAL DATA Total Dose ppm-Days _______________________Observations_________________________________ Pathology Lester (2) Greenberg Adams (1963) Cont'd Sherman rats H F 9s 98 2 then 13 100,000 Variable 80,000 Same exposures <is above After animal deaths, replacements were made in chambers. Two malea survived all 15 expo sures. Remaining animals and replacements survived an average of eight exposures One died after two exposures at 100,000 and twelve at 80,000. One died after two exposures at 100,000 and twelve at 60,000. Six/nine survived all fifteen exposures. K 98 0 Control animals F 98 0 Control animals Growth stopped during exposures and resumed Lungs had focal pneumonia which at normal rate after exposures. healed after two weeks of recov oj External appearance normal* Liver color, ery from exposure. One-third of appearance, consistency, degree of congestion animals had parasitic cysts In was same as controls. liver. Liver pathology same as controls, but more variation In amount of fatty Infiltration. Spleen had advanced lymphocy tic hyperplasia. Kidney pathology same as con (3) Sherman M 15 8 5days/20,000 434,000 trols. rata F 15 8 week 20,000 for 3 months 434,000 1/30 died. External appearance of all animals normal. All organs had normal gross ap pearance. M 15 S Same 0 F 15 8 Same 0 0 0 Liver larger, spleen smaller than controls. White blood cells lower, lymphocytes higher, neutrophils lover than controls. Body weight and hemoglobin were same as controls. Control animals. 4/30 died. Liver parasitic cysts In all animals. Liver fat normal. No abnormal histology. Congestion and swelling greater in liver than controls. Congestion and swelling less in kidney than controls. Conges tion and swelling same In spleen as controls. AM I HAL DATA Authors EXPOSURE Hrs. pet Cone Species Sex Mo. Day Days ppm Total Dose ppm-DaVB Observation Pathology Lester Greenberg Adans (1963) Cont'd (4) Sherman M rats F 5 5 8 8 19 50,000 317,000 Mo mortality. On days 1-4, animals lost weight, showed 19 50,000 317,000 neuorologlcal symptoms. On days 4-19, weight gain was normal. Serum transaminase, hematocrit, and prothrombin times were normal. White and red cell counts were lower than Gross organ appearance was same as controls. Liver pathology showed congested cells. Liver parasitic cysts seen in all animals. controls. Hair: all 5 males had thin hair and scaly tails; females and controls were normal. Ltver/body weight ratio was higher than controls. Control animals. MSS 19 0 0 Control animals. aLn* F58 19 0 0 QEUr-.Eif 000S2i Authors Kuebler (1964) Abatract ANIMAL DATA Sued es Sex Ho. Hrs. per Dav EXPOSURE Cone. Davs onm Total Dose ootn-Dav Observations Rata Klee G. Pigs Klee ND HD HD ND ND 2 100 5,000 41,600 No effect at 5,000 and 150,000 ppm. ND 2 100 15,000 125,000 At 50,000, animals were hyperactive, but ND 2 100 50,000 416,000 returned to normal after exposure. ND 0,5 minutes Distance ND Animals sprayed with A shellac-based 20-25cm. hair spray Pathology No histological damage No change In lung hlstology Vazln Plokhova (1968a) Abstract UInl Vazln Plokhova (1968b) Abstract Vazln Plokhova (1969a) Abstract Rabbits ND ND "chronic" 3,500 3,900 ND Rabbits ND ND 4 Chin ND chilla rabbits 34 167 3,500 (5.5 to mas.) 3,900 ND 150 8 to 12 200 to 300 Brain electrical activity changes: Appearance of beta waves (60 Hertz) in anterior and pos terior hypothalamus along with circulatory changes. Altered f*wuves In EEG Decreased heart rate, arrhythmia from posterior hypothala Decreased ECO voltage mus* Potentials from Decreased duration of systole anterior and posterior Reduced blood flow. Increased arterial pressure. hypothalamus increased by 18-30X and 70-851 respectively. After 20 days, blood adrenaline rose from 3.5 AqmZ to 6.15 AqmZ; at 40 and more days. It was 6.6 xqmZ. Posterior hypothalamus electrical activity also changed. This is the direct cause of hypertension. Vazln Plokhova Rata (1969b) Abstract ND ND 150 {5 months) Disrupted cardiac work rhythm. Bradycardia and arrhythmia. Reduced relative duration of I-II and T-II Intervals. Relative duration of QRS complex did not change. After 15 days recovery: cardiac activity rhythm returned to normal, but the duration of the sound Interval remained below initial levels for another 15 days. Therefore, max. permissible VC concentration is significantly less than .03og/l (I2ppm). ANIMAL DATA AuthPUL. Clapp Rats Kaye Young (1969) Aba tract m Viola Ulster M (1970a) Rats M Viola' Rats - (1970b) Ulster 300 gm Viola Bigotti Caputo (1971) Rats Ulster H M per No. Dev ND Days i Cone* ppm Total Dose ppm-Davs _______________________ Observations_______________________________________________ Pathology SubCutaneous HD Urine contains allylmercapturic acid and 3-hvdroxypropylnercapturic acid. These compounds arise by the reactions of allyl compounds with glutathiones. 25 4 25 260 30,000 (5days 0 per vk for 12 months) 90 1 i 10,000 26 4 260 5day/wk 25 1 260 5day/wk 30,000 0 l,300xl03 417 1300xl03 0 Animals slightly sleepy during exposure. Gross behavior deteriorated after 10 months. 13/50 died of cardio-respiretory Most animals had pathological involve ment of brain, liver, kidney, thyroid. Severe proliferation of cartilege and complications. 2/50 died of bleeding In the peritoneal cavity. No mention of skin tumorB. bone abnormalities In small metatar sal bones. Severe tissue degenera tion In brain and liver and thyroid. Connective tissue Invaded small ar teries In feet. Enlarged, prolifera ting Kupfer cells in liver.' Distribution of VC in tissue; None observed. Red cells had much more VC then serum- high variation VC Is In urine, but major quantity Is lost via lungs. '(VC) falls rapidly In first hour in expired air, blood, urine, and brain, liver kidney. After 3 hre. no VC is measurable. Controls showed no tumors. Almost all exptl. animals developed skin and lung tumors. Very few bone tumors; when seen they were In all 4 extre mities. 651-701 of tumors were skin tumors near parotid and aubmaxlllary glands, frequencies: SKIN LUNGS BONE 26/26 16/26 16/26 Lung tumors were glandular. New cartilage and subsequent ossifica tion In 4 extremetles. Bard mass first seen after 10 months exposure. Authors Species Sex No. Basalaev Rabbits Vatin Rats Kochetkov (1972) Abstract ND ND ND ND EXPOSURE lira. per Day Days ND 6 mos. ND (130 to 180 days) Cone. ppm 12-16 ANIHAJL DATA Total Dose ppm-D. HD ...... ................................. Observations __________________ _________ Pathology Changes In electrical activity of hypothalamus Hyperadrenalineaia. Cardio-vascular function impaired. Bone resorption and osteoporosis. Theory: All symptoms are caused by hypothalamus disfunction and subsequent hormone Imbalance. vj Sftr OQs? ANIMAL DATA Authors AnInals Species Sex Maltoni 1974 Rats Sprague- Dawley M F Rats M Sprague- F fcg Dawley Rats SpragueDawley Rats breeders M F M F off spring No. 3rs/Day Exposures Days cone. ppm Total dose ppm-days Observations _ Survivors Total 309 4 268 5da/wk 265 4 280 5da/k 30 4 10 5da/wk 635 10,000 6,000 2,500 500 250 50 0 280 10,000 6,000 2,500 500 250 50 0 155 30,000 uo60xio3 <635X103 <265X103 952.9X1G3 (26.5X103 t 5.3X103 0 466X103 280X103 167X103 23X103 11.7X103 2.3X103 0 775X103 0769 0/72 0/7A 0/67 1/67 3/6A 1/68 36/60 43/60 54/60 56/60 44/60 50/60 183/190 60/60 27 21 21 16 11 0 0 3 1 0 0 0 0 0 2 liver Angiosarcomas Pathology Zymbal Sarcomas NephroBlastomaa 6 13 3 11 5 3 9 26 7 33 2 0 0 05 00 00 0 30 0 0 0 0 0 0 10 00 00 00 00 00 0 20 36 A 110 7 10,000 6,000 (day 12-1B of preg.) 10,000 6,000 11667 7000 11667 7000 28/30 28/30 0 0 30/34 O) 1 (subcutaneous angiosarcoma) 30/32 l (subcutaneous antlosarcoma) Authors Industrial Bio - test Laboratories UVOi Animals Species Sex mice rfiss CD-I nice H F rats Sprague Dawley outbred COB a M F hamster Golden Syrian M F ANIMAL DATA Kxposure Observations* No hra/day days cone. ppa Survivors 300 7 5da/wk 300 7 5da/wk 165 165 143/200 166/200 157/200 300 7 5da/wk 300 7 5da/wk 30Q 7 5da/wk 300 7 Sda/vk 165 2500 200 50 165 2500 200 50 165 2500 200 as of April 15, 1974 50 165 2500 200 50 Pathology liver ang losarcomas 17 4 2 GHf/C fry REFERENCES Baretta, E.D., R. D. Stewart, and J.E. Mutchler. Monitoring Expo sures to Vinyl Chloride Vapor: Breath Analysis and Continuous AL^ Sampling. American Industrial Hygiene Association Journal, Volume pp. 537-544. w Basalaev, A. V., A.N. Vazin and A.G. Kochetkov. Pathogenesis of Changes Developing Due to Long-term Exposure to the Effect of Vinyl Chloride. GIG TR Prof Zabol 16 (2) : 24-27. 1972. Clapp, J.J., C.M. Kaye, and L. Young. Metabolism of Alkyl Com pounds in the Rat. Biochem. Journal 114 (1), pp. 6-7. 1969. Dinman, B.D., W. A., Cook, W. M. Whitehouse, H. J. Magnuson, and T. Ditcheck. Occupational Acroosteolysis: I. An Epidemiological Study. Archives of Environmental Health, Volume 22, pp. 61-73, January, 1971. Dodson, V. N., B.D. Dinman, W.M. Whitehouse, A. N.M. Nasr, and H. J. Magnuson. Occupational Acroosteolysis: III. A Clinical Study. Archives of Environmental Health, Volume 22, pp. 83-91, January 1971. Gabor, S., M. Lecca-Radu, and I. Manta. Certain Biochemical Indexes of the Blood in Workers Exposed to Toxic Substances (Benzene, Chloroben zene, Vinyl Chloride). Prom. Toksikol. i Klinika Prof. Zabolevanii Khim. Etiol. Sb. 221-223. 1962. Gabor, S., M. Radu, N. Preda, S. Abrudean, L. Ivanof, Z. Anea, and C. Valaezkay. Inst. Hyg. Cluj., Romania. Bucharest 13 (5), 409-418. 1964. Grigorescu, I. and G. Tova. Vinyl Chloride; Industrial Toxicological A pects. Rev. Chim. 17(8): 499-501. 1966. Harris, D.K. and W.G.F. Adams. Acroosteolysis Occurring in Men En gaged in the Polymerization of Vinyl Chloride. Brit. Med. Journal, 5567, pp. 712-714. Ulus. 1967. Kramer, C.G., and J. E. Mutchler. The Correlation of Clinical and En vironmental Measurements for Workers Exposed to Vinyl Chloride. American Industrial Hygiene Association Journal, Volume 33(1): 19-30. 1971. Kudryavtseva, O.F. Characteristics df Electrocardiographic Changes in Patients with Vinyl Chloride Poisoning. GIG TR Prof Zabol 14(8):54-56. Kuebler, H. The Physiological Properties of Aerosol Propellants. Aero sol Age 9(4), 44,47-48, 50, 90-91. 1964. Lange, C.E., S. Juhe, G. Stein, and G. Veltman. Uber die Sogenannte Vinylchlorid-Krankheit. Dtsch. med. Wschr. 98, pp. 2034-2037. (Ger man) 1973. 60 Lester, D., L.A. Greenberg, and W. R. Adams. Effects of Single and Repeated Exposures of Humans and Rats to Vinyl Chloride. Amer ican Industrial Hygiene Association Journal, pp. 265-275, May-June, 1963. Maltoni, C. Preliminary Report on the Carcinogenicity Bio-assays of Vinyl Chloride. Presented at OSHA Vinyl Chloride Fact Finding Hearing, February 15, 1974. Markowitz, S. S., C.J. McDonald, W. Fethiere and M. S. Kerzner. Occupational Acroosteolysis. Arch Dermatol 106 (2):219-223. 1972. Marsteller, H. J. Chronic Toxic Liver Damage in Workers Engaged in PVC Production. Deutsche Medizinische Wochenschift 98 2311-2314. 1973. Mastromatteo, E.,' M.D., A.M. Fisher, H. Christie, and H. Danziger. Acute Inhalation Toxicity of Vinyl Chloride to Laboratory Ani mals. American Industrial Hygiene Association Journal, Volume 21, No. 5, October, 1960. Meyerson, L. B. and G.C. Meier. Cutaneous Lesions in Acroosteoly sis. Arch Dermatol 106(2):224-227. 1972. Torkelson, T.R., F. Oyen, and V.K. Rowe. The Toxicity of Vinyl Chloride as Determined by Repeated Exposure of Laboratory Animals. American Industrial Hygiene Association Journal, Volume 22, No. 5, : pp. 354-361. 1961. Vazin, A. N. and E. I. Plokhova. Creation of an Experimental Model of "toxic angioneurosis" Developing from the Chronic Action of Vinyl Chloride Vapors on an Organism. GIG TR Prof Zabol 12(7):47-49. 1968a. Vazin, A. N., E.I. Plokhova. Pathogenic Effect of Chronic Exposure to Vinyl Chloride on Rabbits. Farmakol Toksikol, 31(3):369-372. 1968b. Vazin, A.N., and E.I. Plokhova. Dynamic Changes in Epinephrine like Substances in Rabbit Blood Following Chronic Exposures to Vinyl Chloride fumes. GIG TR Prof Zabol 13(6):46-47. 1969a. Vazin, A. N., E. I. Plokhova. Changes in the Cardiac Activity of Rats Chronically Exposed to Vinyl Chloride Vapors. Farmakol Toksikol, 32(2): 220-222. 1969b. v Viola, P.L. Pathology of Vinyl Chloride. Medicina del Lavoro, Vol ume 61, No. 3 March, 1970. Translated from the Italian. 1970a. Viola, P.L. The Vinyl Chloride Disease, (unpublished translation) Sum mer, 1970. Viola, P. L., A. Bigotti, and A. Caputo. Oncogenic Response of Rat Skin, Lungs, and Bones to Vinyl Chloride. Cancer Research, Volume 31, pp. 516-522. 61 GENC 00082?' Von Oettingen, W. F., M.D. The Halogenated Aliphatic, Olefinic, Cyclic, Aromatic, and Aliphatic-aromatic Hydrocarbons including the Halogenated Insecticides .Their Toxicity and Potential Dangers. Public Health Service Publication No. 414, U. S. Department of Health, Edu cation, and Welfare, Washington, D. C. 1955. Wilson, R.H., W. E. McCormick, C.F. Tatum, andJ.L. Creech. Occupational Acroosteolysis, Report of 31 Cases. The Journal of the American Medical Association, Volume 201. No. 8, pp. 577-581. 1967. 62 30 APPENDIX VIII DISPOSAL OF PRODUCTS CONTAINING POLYVINYL CHLORIDE This discussion on disposal of PVC emphasizes incineration and landfilling, the only presently used large-scale methods for the disposal of solid wastes. There is also a limited discussion of resource recovery possibilities. Incineration The two areas of concern related to PVC incineration are incinerator air pollution and incinerator and gas scrubber corrosion. Hydrogen chloride is the major toxic material released when PVC is burned. It has been shown that virtually all of the chlorine is released from PVC on combustion, resulting in HC1. It is estimated that 0. 2 per cent of solid waste is PVC, and 16 x 10 tons per year of solid waste are incinerated in the United States. Thus, on the order of 32, 000 tons of PVC are burned annually, releasing approximately 18,500 tons per year of HC1 as air emissions. Other solid waste sources which can produce HC1 are chlorides in food waste, plants, grass clippings, and inorganic salts. The formation of compounds requires volatilization and reaction with incinerator flue gases. Achinger and Baker compiled data indicating an emission factor of six pounds of HC1 per ton of solid waste burned. Recent data on HC1 emissions obtained by Battelle show a factor of 5.1 pounds per ton. A value of five to six pounds per ton would be a reasonable emission factor to use for HC1 emissions from municipal incinerators. Using an emission factor of 5.5 pounds per ton gives 44,000 tons per year of HC1 produced by incineration of municipal solid waste. The amount of HC1 produced from PVC using the above calculation is 42 percent of the total. Much more HC1 is probably now emitted to the atmosphere from the nation's coal-burning power plants than from our municipal incinerators. However, there still could be a hazard in the immediate vicinity of an incinerator as a direct result of its HC1 emissions. Of particular concern is the possible dispersal of the stack gases to cause the ambient concen trations of HC1 at ground level to exceed harmful concentrations. How ever, HC1 is not at the present time regulated by EPA. Other air pollutants could be formed\from the additives in PVC dur ing incineration. Several additives are usually incorporated into the poly mer to emphasize particular properties not inherent in the base polymer. The types of additives are antioxidants, antistatics, colorants, fillers, plasticizers, and stabilizers. Some of the additive agents used are: anti oxidants--phenols, amines, phosphates, and sulfur compounds; antistatics --amine derivatives, quaternary ammonium salts, phosphate esters. 63 GENC 000831 polyethylene glycolesters; colorants--salts or oxides of metals, aluminum, copper and inorganic pigments; fillers--silica, glass, calcium carbonate, metallic oxides, carbon, cellulose fillers, asbestos; plasticizers--phthalates, organic phosphates; stabilizers--lead salts of acids, barium, cad mium, calcium, zinc, alkyl tin compounds. It is highly unlikely that large quantities of VC will be emitted during incineration of PVC. There is no evidence that PVC will chemically revert to VC. Some small amounts of entrapped monomer might conceivably survive incineration, but these quantities would be very low. The second area of concern with incineration of PVC is firebox corro sion and corrosion of pollution control equipment. HC1 can be a major factor related to corrosion of this equipment during incineration at certain temperatures. In the case of plastics, PVC is the major source of chlorine leading to HC1, but other plastics may also contain some chlorine. Incinerators with heat exchangers will have corrosion problems on the fire side of the exchange equipment when the combustion gases contact the outer metal surface. Other surfaces of concern are in the cooling area and in the gas scrubbers. Estimates indicate that in incinerators with heat-recovery systems PVC in the refuse will increase tube maintenance costs by 15 to 20 perpercent over that to be expected if PVC-free refuse was used as fuel. About 95 percent of the incinerators in this country have some type of air pollution control equipment that is exposed to the high chloride envi ronment resulting from refuse combustion. Because of the high chlorine content of the combustion products, the cooling and precipitating water from the scrubbers that contacts the flue gas contains large quantities of chloride and is extremely corrosive to the structure. In summary, technology exists for controlling the HC1 emissions that result from incineration of solid waste; however, the application of this technology will result in increased costs. If technology is not applied, then the contribution of PVC to the nation's air pollution problem will increase because of the projected increases in the usage and disposal. HC1 scrubbing technology is available, but its application results in corro sion problems. Depending on construction materials, design, and opera tion, these problems can be either large or small. Landfilling \ PVC does not decompose significantly within the normal time frame of most other municipal solid wastes. It comprises only about 0. 2 percent of the total municipal solid waste being landfilled today, and the effect of PVC on the reuse of the landfill site, at least in the short run, should be negligi ble. 64 Since PVC degrades very slowly, in the landfill environment it should not add significantly to the production of leachate or decomposition gases as do other parts of the refuse. The additives of greatest concern are probably the plasticizers. However, if a sanitary landfill is designed and operated with today's technology, disposal of PVC products in a sanitary landfill should pose no special problems to the operation or to the ultimate use of the site. Resource Recovery Recycling of solid waste is a growing industry. Technology has been developed to recover some resources from many of the items in the municipal waste stream. However, the technology to separate plastics or PVC from the waste stream has not yet been commercially demonstrated. The solution to the separation of plastic waste from other components of the municipal waste stream is one deterrent to direct recycling and reuse of plastics, including PVC. However, gathering and centralizing the waste products are also major problems. Some types of scrap PVC from the fabrication process are presently being recycled back into the manufacturing process. This reduces the solid waste from plastic fabrication plants and reduces the need for new raw materials. There is work underway to develop means for utilizing the benefits of recycling the total municipal waste stream. Examples of these recycling techniques are listed below: -- To recover heat given off during the incineration of solid waste containing PVC and other combustible materials as electricity or steam for heating. An example is EPA's research contract with the Combustion Power Company of Menlo Park, California, in which combustion gases are expanded through a turbine to produce power. -- To recover the products of a refuse pyrolysis operation either as a pipeline gas or as feed material for a nearby refinery. An example is EPA's research grant with West Virginia University in which refuse pyrolysis is being studied on a bench-scale. A second example is the Bureau of Mine's research effort to convert refuse to pipeline gas. Also, US and Japanese industrial firms are actively exploring this area. The recent change in the world's supply of crude oil should speed up research and development on new and existing ways to utilize more fully the resource of waste PVC. 65 6ENC 0 0 083d REFERENCES 1. E.A, Boettner, G.L. Bell, B. Weiss, "Combustion Products from the Incineration of Plastics, "Report No. EPA-670/2-73-049, July 1973. 2. "Compilation of Air Pollution Emission Factors," 2nd Edition, Publication No. AP-42, EPA, April 1973. 3. W.C. Achingerand R.L. Bakei', "Environmental Assessment of Muni cipal-Scale Incinerators," Report No. SW-111, EPA, 1973. 4. G.L. Huffman, "The Environmental Aspects of Plastics Waste Treat ment, " Symposium on the Disposal and Utilization of Plastics, New Paltz, New York, June 25, 1973. 5. "Threshold Limit Values," American Conference of Governmental and Industrial Hygienists, 1972. 6. Fessler, R., H. Leib, H. Spahn, "Corrosion in Refuse Incineration Plants," Mitt. Ver. Grosekesaelbets, 48 126 - 140, April 1973. 7. Vaughan, D.A., and P. D. Miller, "A Study of Corrosion in Municipal Incinerators," Cincinnati, Research Grant, April 1973. 8. Miller, P.D. et al, "Corrosion Studies in Municipal Incinerators," SHWRL - NERC, Report SW - 72-3-3. 9. Baum, B. and C. H. Parker, "Incinerator Corrosion in the Presence of Polyvinyl Chloride and Other Acid-Releasing Constituents," report by DeBell and Richardson, Inc. OJo date) 10. George L. Huffman and Daniel J. Keller, "The Plastics Issue, " SHWRL NERC, Cincinnati, Ohio* August 28, 1972. 11. "Incinerator Gas Sampling at Harrisburg, Pennsylvania," EPA Con tract No. 68-02-0230, Office of Air Programs, September 1973. 66 GEHC APPENDIX IX ACTIVITIES OF TASK FORCE The principal activities undertaken or stimulated by the Task Force are set forth below MARCH - Recognition of problem of pesticidal sprays containing VC-Responsibility assigned to Office of Pesticide Programs MARCH - Analysis of material losses during PVC polymerization pro cess MARCH 19-21 - Pilot monitoring effort at B. F. Goodrich Plant in Louisville MARCH - Preliminary evaluation of health effects data APRIL 2 - Meeting with representatives of PVC manufacturers organ ized by Manufacturing Chemists Association APRIL 4 - Meeting with representatives of interested environmental groups APRIL - Development of interim methodology for VC sampling and analysis APRIL/MAY - Visits to VC manufacturing facilities and to PVC polymeri zation, compounding, and fabrication facilities APRIL 12 - First of series of interagency meetings convened by EPA APRIL/MAY - Monitoring at seven complexes involving 10 PVC and 2 VC plants APRIL/MAY Review of health effects data APRIL 30 Review of Industrial Biotest toxicological experiments MAY 27-31 Preliminary VC water persistence studies MAY - Preliminary VC air persistence studies MAY/JUNE - Recognition of air emissions problem -- Responsibility assigned to Office of Air Quality Planning and Standards JUNE 3 - Technical review of monitoring activities JUNE 11 - Administrator's meeting with senior executives of 29 com panies producing PVC and VC JULY - Development of improved methodology for VC sampling and analysis 67 GENC 000835