Document JovD08Ep5O7ZvZd193KvK5KO

Safet y and Environmental Concerns for Compounds and Products John T. Barr Air (froducts and Chemicals, Inc. Box 538 Allentown, PA 18105 Volume III, Chapter 26 Encyclopedia of PVC Second Edition Working Draft October 1985 0472C-FI Table of Contents (continued) IV. Glossary. V. References Ei31 92 94 Table 1 - Generally Used Stabilizers for PVC and CPVC Table 2 - Environmental Dc t a for Some Commonly Used Plasticizers Table 3 - 1982 Injury Stat isties for Plastic Producers and Processors Table 4 - Toxicity Data fer Several Common Pipe Cement Solvents Table 5 - Solvent Migrati cn Data From Cemented Pipe Tests Table 6 - Static Extract! on of Chloroorganics From New PVC or CPVC Pipe Table 7 - Summary of Seve r a 1 Prominent Combustion Toxicity Test Methods 7 22 44 53 60 64. 73 SPI-00484 - II. Compoundina Processes All PVC must be mixed with other materials (compounded) before it is suitable for use. The reasons for his, and the methods for accomplishing it, are covered in detail in Volum- s .1 and II and the earlier chapters of this volume of this Encyclopedia. Producers of end products \ hich handle more than a few million pounds per year of materials usually find t more economical to combine the compounding and fabrication steps in the s; me facility. Smaller fabricators, and those who produce small amounts of sp ecialized products, may purchase their supply in ready-to-use form, usually as processed pellets, but sometimes as a dry powder blend. This chapter will c iscuss the compounding and fabrication steos as separate operations because of the differences in hazards and risks which are seen in those areas. Produ ct application is treated as a third area. A. Raw Materials Fabrication operations range in size from small operations which receive the raw materials in b igs, drums, and other small packages, to operations large enough to use bu k storage for mmost or all ingredients. The opportunity for worker exposure or release to the environment therefore covers a wide spectrum with the manual handling at small operations usually providing a hi her probability of spillage, but also having more 1imi ted opportunity fo a significant release. In any case, the hazard of the substance is the s<ame; it is the risk which changes. The discussion which follows centers jn these hazards. 2 SPI-00486 (16CFR1500.45), and various architectural codes for the flammability of interior furrliishings. The toxic properjt ies of PVC were discussed in Chacter 5 of Volume I of this encyclo pledia and will not be repeated here. Chlorinated PVC (CPVC) has very similar gross properties and should be handled in the same manner. At one time chloroform was used as a swel1ing-agent to assist in the chI orination step, but that practice was stopped several years ago and thus it is no longer a source for even the minimal release of chloroform which may have been possible earlier, A further result of the chlorination step is that essentially all of any residual VC Ikas been removed. Concentrations of VC above the OSHA action level (0.5 pom) can develop in the free space after extended storage of PVC in closed containers. Provisions for ventilation or respiratory protection should be provided if workers enter silos or railcars which have been used for PVC storage or transport. To a lesser degree, seme buildup can occur in warehouses or transportation vehicles used for bagged product. Generally, there is enough natural ventilation to prevent exposures over the action level, but the possibility of such levels exists in tightly closed areas where a large volume of bagged material is prese ht. Similarly, some VC can be expected to escape during the ccmoounding step, especially at' the first heating or fluxing stage. The 4 SPI-00488 After liquid stc bilizer or plasticizer has been added to PVC, the mixture no longe r is dusty. Before that stage, adequate ventilation should be proviced to hold the amount of dust in the air to less than 15 mg/M^ of total dust for a time-weighted average (TWA) over 8 hours, or 5 mg/M^. of respirable dust (particle size below 10 mi crons). Ve ry little suspension PVC contains a significant respirable fract ion. This is the OSHA standard of 29CFR1910.1000(a[) The American Conference of Governmental and Industrial Hygieilists (ACGIH) recommends 10 mg/M,3 for total dust, rather than the |alue of 15 listed by OSHA. Neither organization lists PVC as a specific substance controlled by these limits, but prudence dictate that these values be observed. Any PVC that has been spilled on concrete or steel walkways can create a very un 4afe walking condition and should be removed at once. This is e specially true of stairs and catwalks around silos and railcar uni oalding stations. 2. Stabilizers Stabilizers may be divided conveniently into two major categories for the purpose of th is discussion: FDA-approved and non-FDA-approved, Table 1 contains i list of classes of substances which are in common use for this purpb se. 6 S PI-00490 The Food and Drug Administration is responsible *or the implementation of the Food Drue and Cosmetic Act wnicn includes :re setting of regulatli ons for food additives. Polymers and other substances which c>me into contact with food are classed as indirect food addifives, be: ause of the presumption that seme ccmDonent of the package can migrati into the food and thus become an additive under the terms of the regulations. Rules for controlling food additives are codified at 210FR170-189, with polymers being grouted under Part 177, and most stabilizers under Part 173. Antioxidants, plasticizers, and ucn can be found at Part 175 anc ctner locations. Packaging material: or containers intended for aseotic conditions such as in meat or poultry packaging plants, must meet several FDA standards for appre val, but the actual compliance is monitored by cne Department of Agricu 1ture under rules published at 9 CFR 300 et seq. It is necessary for a producer to obtain specific aoproval from the FDA to use any subs tance in a food-contact use that '5 not on the approved lists. Apjp roved substances are those specifically listed in the rules named abo|v e, those on the Generally Regarded as Safe (GRAS) list, and those wi t|h prior sanction because of broad general use before 1958. In adjd ition, any substance approved as a direct food additive is acceptad le also as an indirect additive. The metallic compon ent of FDA-approved stabilizers is limited to salts of calcium, z inc, or tin, all of whicn nave limited to<icity to 3 SPI-00492 substantial 1 eve 1s it has been subjected to more rigorous controls. The FDA has approv ad several dialkyl tin salts for use in food applications, afte reviewing test results (21CFR178). Nevertheless, the Interagency Te sting Committee (ITC) recommenaed a group of alkyl _tj_n stab ill z_er_s_to EPA for further testing (47FR54626, 3 December 1982). The EPA ha concluded that a proposed test urogram by a group of manufacturers w 11 be adequate to answer the questions raised by the ITC, and this i jrogram will explore the chemical and environmental fate of this class of substances. The EPA stated (481 R51361) that the tin stabilizers are not expected to be released rap dly into the environment, and that the material which i.s released ` s the reaction product formed during the stabi1ization step. such as dibutyl tin dichloride, rather than the intact alkyl tin (I ). Total tin release to the environment from either compounding operations or use in pipe was estimated to be very low by the EPA. Mi gration rates from new PVC pipe was projected to O* be about 10 mg/m /sec. EPA calculated that the added tin concentration for a n average receiving stream resulting from plastic pipe use would appr oximate 7 x 10~ ppb. This is well below background in most areas. Alkyl tins decompos e slowly in the environment by hydrolysis, photolysis, and bio degradation, with a halflife in the range of a few weeks. The bioconc entration factor (BCD for the organosalts are in 10 . ... .. ^ SPI-00494 The non-FDA-approv ed stabilizers in common use include salts of barium, cadmium and lead, all of which are toxic heavy metals. Leap stabi1izers are app roved for use in potable water pine throughout the world except in Nor th America. These materials are used widely for _n0/1-food _(or d_rj/iki ng water) applications because of their greater price efficiency as compared to the approved staoilizers. Each of these metal s is listed in Appendix VIII of the Resource Conservation and R ;covery Act (RCRA) rules (40CFR251) which makes them potential 1y subject to the hazardous waste rules of RCRA. In genera 1 , if the Ex raction Procedure Toxicity Test (the EP test) of Appendix II, 40CFRt 61, applied to a substance produces an extract containing more thjt n a specified concentration of the metal, any waste that is disc arded of that composition must follow the hazardous waste rules. The est consists of contacting approximately 1/3-inch particles with 16 imes their weight of dilute acetic acid solution, and measuring for vhe extracted substance after 24 hours. The specified maximum concentrations for these three metals are: Barium Cadmiurn Lead 100 mg/1 1 mg/1 5 mg/1 Thus waste stabi1i ier is a hazardous waste and scrap compound may be, depending on its e xtractabi1ity. However, very few compounds will fai1 the EP test. especially when formed into the finished article. SPI-00496 as hazardous waste when discarded but not when in use. California has such a rule and -is issuing broad variances to avoid the obvious impossible situation s which would result from its strict enforcement. There is no question as to the cumulative toxicity of many of these metals to humans, end workers should receive full training, and protective equipment as appropriate, to prevent overexposure. OSHA and the ACGIH have set (29CFR1910.1000 et seq.) the following exposure limits (me /M3) : OSHA 8-hr TWA Calcium Cadmium (dust) 0.2 Barium 0.5 Lead 0.05 Tin 0.1 Zinc chloride (fume) 1 Ceiling 0.6 ACGIH 5 0.05 0.5 0.15 0.1 1 As in many other cas es, the ACGIH recommends lower values for some of these metals than dpes OSHA. Current OSHA Hazard Communication rules (48FR53280, 25 Nov. 1983, new section 29CFR1910. lflOO), as well as many local community "right-to-know" law ;, require extensive disclosure by the supplier of the composition and hazards of all commercial substances and require that workers be in farmed of these hazards and be supplied with a 14 SPI-00498 Phthalates - The p incipal phthalate ester in use in PVC 's di-(2-ethyl hexyl) ihthalate (DEHP, sometimes mistakenly termed DOP). It amounts to abou a third-of the total plasticizer use. primarily because of its hig efficiency and good properties at normal use temperatures..___ It has very lew act te toxicity, with rodent LDj-q values (.cose necessary to Rill 1 alf the animals in 14 days) in the 30 g/kg range (about the same as sugar) and mild irritation properties (12). CSHA nas set an 3-hour LV of 5 mg/NT. The ACGIH recommends tne same value, with a Shorl -Term Exposure Limit (STEL) of 10 mg/fA. The ERA recommends a w; ter quality criteria of 15 mg/1 in human drinking' water, .based on ani mal feeding studies (13), but the soluoility in fresh water is onlj 0.34 mg/1, and it is about half that soluble in seawater (14). Esters with alcohol chain lengths of four carbons or greater do not have sufficient sol ubility in water to present significant acute toxicity symptoms i n a variety of aquatic species (15,15). DEHP does not hydro lyze as rapidly in water as do some of the shorter chain analogs (17). It has the potential to bioaccumulate in the food chain because of its low water/oil solubility ratio, but it also is absorbed strongl / by sediments, and biota have been reported to contain less DEHP t nan the sediments above which they live (.13,19). In addition, it is netabolized readily by the host. The uoiquitous 16 SPI-00500 metaoolites are pop itive in a variety of teratogenicity and jjn vi tro mutagenicity tests (29-31). That concept is supported by the fact that DEHP i s a pro:ifioter of mouse liver tumors, but not of skin cancer, and is not an initiator of either liver or skin cancer (32). One study reported inhibition of the development of neoplastic indications in rat treated with diethylnitrosoamine (33), while another reported p rcmotion of this effect (34). Scienti sts at the Ii'jTP have reviewed the evidence for the careinogenicity of DEHP and have concluded "that DEHP has been shown to ce carcinogenic to rodents in a valid chronic test. Further experimental inquirly will be required, however, to accurately assess the ootential health risks posed to humans by exposure to small amounts of this plas ticizer (35)." The NTP has included DEHP in the Third Annual Report on Carcinogens (36). The IARC has concluded that there is "sufficient evidence" for animal carcinogenicity, out that there is inadequate evidence to evaluate the carcinogenicity of DEHP in humans (37). Metaoolism occurs after absorption in the gastrointestinal tract following nydrolysi s to the monoester (38). The alcohol metabolites appear to be the pri mary toxic substances, as concluded from specific studies on members of the series and the fact that the esters of lower alcohols are ;ven less toxic and are not carcinogenic in bioassays. In add i |t ion, phthalic anhydride itself (39), and of course, its immedia :e hydrolysis product, phthalic acid, is not 13 SPI-00502 no-effect level ex sts which is below normally expected ambient exposures (61). - The monoethyl hexyl phthalate ester, the primary metabolite, has a no-effect level of 50 mg/kg as a teratogen in rats; above that dose maternal toxicity i nd weight loss of the embryos were seen (45-46). The parent ester ir creased fetus malformations when fed at the maternally-toxic dc ses of 17. or more in the diet (47). DEHP causes testici lar atrophy in rats and is positive in the dominant lethal ass ay at 1 ml/kg dosage (48-50). An insufficieny of toxicity data in 1980 (13,51) led to listing of the substance by th e Interagency Testing Commission (ITC) for consideration for f urther testing to determine if regulations should be imposed by the E PA. That agency established a multi-tier testing program to be perfo rmed by a group of manufacturers. The results were reported in 19 34 (15). The Agency had concluded earlier (5Z) that, prior to obta ining data indicative of more serious consequences, the e iter presents risks below a level at which regulation normally would be taken. The NTP is exploring the question of genotox icity and other related factors (53). The Consumer Product Sa rety Commission- (CPSC) established a Chronic Hazard Advisory Pan :1 to advise it on hazards for consumer products containing DEHP (48 :R56623, Dec. 22, 1983). The Panel released its report in September 1985, concluding that DEHP is a nongenotoxic animal carcinogen, tnd that more research is needed to determine if 20 SPI-00504 SPI-00506 <w> a> a> a> a> c > >> -- > oj_ 4-4 4-- 4-* 4-4 cn a3 03 03 03 o CP CP CP CP 4-4 CD a> a OJ wrti Z z <D' 1 H0--)1 CD > a> > 0) > <D > O) > O: |"C 4_, <-- 03 4-4 4-- 03 03 cn uO cn CP CP > a> o a. oj z a> z a> z c: OOO CNJ o-- C\J 03 .-73 ---CQL) Q03 -- Q_ *0o) 03 03 <-o G> >> 5Q >o O<-)j <o^i C U-, o 03 cu NJ 1-- o -Q 03 4~ O) s: r~ CP 03 0) IN i-- o .a 03 4-* at "O a> Ni 03 25 -- a> O .-- --4 _i O 0) _Q U 03 <-J 4--* X CD UJ S -O *rOQ i "O o o *oaco>n o a 3 oOo a CC oo oo -- o> -- \I 1 'T3 cn rn o oo 1 1 1 1 1 1 Od oOn --C oi -- oj *- -- -- cx O <-> C(rTC> I a CL. Oo) 03 Ol =i_ O -u G OUJ aO_ Sebacates - The e! ters of sebacic acid can be considered hcmolcgue of the adipates, DErS has an LD-q in rats of 1.3 g/kg, and like tne adipates, is negatjiive in the short-term mutagenicity tests (12). Citrates - Acetyl tributyl citrate is an FDA-approved comoonent in some food wrap and other food acplication compounds. It has lev/ acute toxicity and is not a mutagen in the short-term Ames tests (12,60). The Belgian Plastibs Association issued a summary recently (34) data relating to human exposure to plasticizers. It was estimated that the maximum e<posure was less than 200 mg/day under extreme conditions. Studi ?s were discussed which indicate that at coses higher than those necessary to induce liver damage in rodents primates detoxify DEHP metabolites by combination with glucuronic acid through a procedure not found in rodents. This detoxification avoids the induction of excess peroxisomes in liver cells which are thought to be the 1 ase cause of rodent tumor development. Fire-rgtardant Pla? ticizers - The addition of organic substances such as plasticizers to PVC reduces its inherent nonflammability. For example. DEHP has a flash point of 325F (open cup) which places it in the NFPA flammaqiility hazard group 1, or "must be heated to ignite". However it and compositions with PVC will burn if exposed to a sufficiently s trong ignition source. Appropriate fire 24 SPI-00508 tests at this time. Also proposed are tests to define further the neurotoxicity of th e ortho TCP isomer, and its chronic toxicity in lower aquatic 1 ife forms. In the proposal the E3A noted that because of the genera 11y 1: w toxicity of TCP and its strong tendency to remain in the finis hed article or absorbed on soil, it was not likely to present signifi ant environmental exposure as the result of waste disposal. Iris (2 etnylnexyl) pnospnate has been reported by the NTP to show "equivocal evidence of increasing adrenal tumors in male rats. liver cancer in female mice, and thyroid tumors in male and female mice (64). The reDorts noted that no mutagenic effects were seen in bacteria. The phosonates as a class are delayed neurotoxins. and ingestion shoul d be avoided. Chlorinated paraffi ns with chain length of 10-30 carbon atoms and 40-707. chlorine are useful in many higher temoerature applications. These are viscous c ils at room temperature with insignificant vapor pressure or water s olubility. They have very low acute toxicity, do not biodegrade, ano are not mutagenic or teratogenic. They also induce increased ac tivity by liver enzymes, as do the other plasticizers discus sed here (65). Because of their pe 'sistence in the environment and tendency to bioaccumulate in th ? aquatic environment, the ITC placed this class on the recommended resting list in 1977, and the EPA accepted the testing proposal of an industry group in 1932 (47FR1017). Partial 26 ^ - 'T- ~ ' SPI-00510 app11 cation, but 1 iter decided to refer this problem to QSHA, then reversed itself agfeiin. The oresent status is not clear. As will be discuss nd in more detail later in this chaoter, the inorganic fillers reduce the combustibility of the final product and often make it more sturdy. Therefore, the net environmental effect is positive, as we 1 as naving a desirable economic impact. Colorants - Several different terms have been used in the oast for materials used to impart color to or mask color development in plastics. Some of the most common are dyes, pigments, colors, and so on. A more general term is colorants (48FR46773, Oct. 14, 1933), which include both dyes (generally thought of as chromopnoric organic molecules) and pig rn^nrs (generally inorganic materials). Colorants may range from col o|rless materials designed to improve whiteness oy their reflectant c abacity, or by obscuring the yellowing which may occur during proces sing, to substances which possess strong color values. Here, again, if food , drug, or cosmetic contact is intended, and the material may be ex p^cted to become a component of the product, the FDA must approve thii suostance and this has led to a series of numbered "FD and C dyes. Recently several of these have been restricted in use or banned because of the carcinogenicity of the dye, or one of its precursors. Those which are approved are listed in 21CFR, Parts 74 31 , and 32. 23 SPI-00512 Substances which yield extractable concentrations greater than that allowed are classed as hazardous and must follow RCRA rules if discarded from a commercial ODeration. This rule does not apply to household or retail trash. Most commercial applications of heavy metal colorants cai pass this test, and therefore there is little concern for leachiig of the colorant from discarded products. There is the usual problem of controlling exposures from dusts of these heavy metal : alts. One method in common use is to prepare a master blend of the color or dye with a part of the resin ana some liquid component, ' ucn as a stabilizer, for cnarging to the blenaer or mixer. This oft en is done away from the general work area and bypersons of greater tecnnical training. This procedure helps avoid cross-contaminatior of colors and conserves expensive materials as wel1 as reducing th e dust exposure. The same stringent rules of personal hygiene should be applied here that were discussed under lead stabi1izers. Liqht Stabi1izers - These materials are added in small proportions to some formulations. The usual FDA regulations apply to food contact uses. Many of the 1ight s :abilizers belong to the class of substituted benzophenones. The :e have very low toxicity, are nonirritating, and 30 SPI-00514 Storage and Hand 1i ig Some of the hazard connected with the raw materials have been discussed earlier. These hazards usually become apparent only if the materials are spi1 ed or otherwise mishandled. Good housekeeping in a workp1 ace is a r flection of good safety practices also, and efforts spent in r $ducing spillage is rewarded in reducea accident rates. Good light ng, open aisles, proper storage, and segregation of material sail at e necessary for an efficient operation, and contribute signifi dant1j to safety, as well. Pallets of bags or drums should never be stacked more than three hign, and preferably only two high, to avoid tilting and slippage, Adequate maneuvering space for forklift trucks must be available, ana operation on slopes or ramps should be avoided. Glued or strapped pallets add to the protection from falling containers, but an excess of glue can sometime s cause the bags to rip and increases spillage, Care should be taken in cutting strapping to avoid a whipping effect, or bags falling when the tension is released, and all strapping materials should be removed to avoid a tripping hazard. Workers should enter silos, transport vessels, or process vessels only under the con t ro 1 of an effective confined space entry program and after al1 elect' ical equipment has been locked and tagged out of service. A rescue larness and an outside observer are required by many safety program t>. NIOSH recently announced plans for developing 32 SPI-00516 elevated temperatures, often with electrical heating units closely spaced to cooling water troughs. The operation is principally a materials handling unit process, and thus a variety of physical hazards is present Proper mechanical c uards and electrical safety devices have been the key to maintaining safe conditions. Equipment should have, as a minimum, the mechar ical protection required by the OSHA standards of 29CFR1910.216, and conform to the e1ectrica1 standards of Subpart S of that rule. Secs. 303-4. Equipment manufacturers often design to higher standards, i nd insurance or local codes may require these improved levels of protection. For example, ground fault protection'. especial 1y in area; subject to wet floors, is highly desirable and frequently mandator y- Walking areas shoul d be designed to prevent slips and falls in the event of spill age c f water or plasticizer/stabilizer oils (or both combined - particul arly) and also from resin or pellets on the dry floors. Sturdy out door grade carpets are useful in nonmanufacturing areas because of tt e better protection they provide against slips and falls as compared t o tile, concrete, or steel walking surfaces. Equipment and walkv< ays must be arranged so that workers cannot fall into or on heated c r moving equipment or product. In addition to the requisite guardrai1 s, triplines and similar protective devices are useful here. 34 SPI-00518 Wastes and Emissions Empty containers s tiould be collected and removed promptly, as should al1 scrap and waste from the ooeration. Whether this waste will be _regu 1 a_ted__under_RCl:A depends on the type and the results of the EP Extraction test di cussed earlier, and whether the total amount of "hazardous" waste Exceeds the RCRA small generator exemption. In any event, current soc ;al (and legal) policy places the ultimate responsibi1ity for the future on the generator of the waste, regardless of who dually hauled the waste away or disposed of it. It is far easier tc take the necessary steps to assure proper disposal at the tin e of generation than it is to try to convince local politicians <|or the EPA) or concerned neighbors years later that the red color or the oil seeping from a landfill is FDA-approved, and therefore is not a health hazard. The EPA promulgatec effluent limitation standards for the plastics molding and forminc industries on December 17, 1984. See 49FR49026, for the new rules tjo be codified at 4QCFR4S3. The molding and for ming industries are defined by EPA to include "processes that ble nd, mold, form, or otherwise process plastic materials into inte rmediate or final products. The regulations apply to both existing ar d new facilities, and are divided into three categories of effU ents. Only direct discharges are affected. Those sending their waste s to public treatment systems are not limited to 35 SPl-00520 for these operati 3ns and successfully removed the major portion of the incoming wastfe s. The EPA' has not s heduled the plastics forming industry for a specific air emis ions rule under a New Source Performance Standard (NSPS). Some ass<j> ciated operations which are controlled include storage of volati e liquids, and vinyl coating and printing' (mid-1984). Under this latter rule, printing inks must contain no more than 507. vol tile solvent, or abatement equipment must be installed that re (juce emissions by 857.. Final rules for pressure-sensitivf tapes and adhesives were promulgated at 48FR48353, Oct. 18, 1983. Tie se rules set allowable limits for quantities of volatile organic Solvents released in the air per unit of production. The ceneral trend is to phase out organic solvents and replace them at 1 east partially with water-based coatings. The EPA has commen ced study of the application of polymeric coatings to substrates (Doc|ket No. A-83-42) and has issued a source category survey for comment Existing plants in these process categories will not be covered by these new source p erformance rules. However, they, and all existing plants in nonspeci ried categories, are covered by other sections of the Clean Air Act vhich require the states to produce rules to reduce the same classes o: emissions. The states must develop a state SPI-00522 38 EPA has under cons deration revisions for the RQ's based on its reevaluation of tiv chronic health effects of these materials, and " many of the salts < >f the heavy metals will be assigned RQ's in the 10-100 lb. range. Users of these salts as stabilizers should be ale.rt for issuance of these revised values. See 50FR13514, 4 April 1985 for a discuss on of this reevaluation process. The discussion in his section applies equally to both the compounding and the processing industries. Fabrication Processes 1. General Fabrication of fina 1 or intermediate products may begin with the basic raw materials , in which case the discussions in Part II of this chapter apply, or i t may use preprocessed compounds and have to deal only with situation s specific to the final stages. In either case there are many simi larities in the hazards and potential risks which may be present. Th e larger the operation, the more likely it is that both processes wi11 be located in the same facility. One significant dif ference between the two processes is that the wastes from the fab rication step are likely to be innocuous as far as having significant mealth and environmental hazards. The absence of risk from exposure to fabricated products has been recognized by 40 SPI-00524 Guards and shield ng are required for all such machinery oy OSHA rule 29CFR1910.216, bui aggressive preventive maintenance programs are necessary to assur e proper operation and integrity of these devices. Many of these oper ations are on a large scale, using high-speed. automatic equipmer t which feed conveying lines and packaging machinery in compl ex arrangements. Operators and maintenance workers must act promptly in case of a malfunction to prevent expensive losses of material s and products. Adeauate care is necessary in both equipment design a nd supervisory instructions to avoid accidents in such situations. Accident report an a 1yses show that the great majority of disabling accidents in the p lastics industry, as in the other manufacturing groups, is caused 3y slips, falls, striking against or by, or the various other cate jories that cover hitting or being hit by solid objects. Some of :he causes which might be expected to be related to the plastic materi ils being processed, such as burns or toxic effects, represent a surprisingly small proportion of such causes. Table 3 contains a comparison of Bureau of Labor Statistics data for 1982 injuries with similar data collected by the SPI in a survey of 437 companies with 589 facilities employing 77,344 workers. Both the resin producers ant the processors groups exceeded the safety record of the total priva e sector in all categories. Within these groups the larger companit s and the larger facilities generally had better 42 . ... * "H ~ - -- ' SPI-00526 .... ~ .... Table 3 1982 Injury Statistics for Plastic Producers and Processors National Data Private Sector Manufactur i ng Plastics Industry Resin/Material Suppliers ( 113)* Processors (223) Machinery, Equipment (68) Moldmakers (40) Total (437) Incidence cer 100 emolovees A1 1 Lost Iniuries/111 nesses Work Oavs S e v e ritv 7.7 3.5 53.7 10.2 5.4 75.0 5.01 6.05 13.0 13.1 8.7 1 .97 2.16 3.92 2.51 3.23 49.5 35.7 50.5 40.4 * 50.5 * number of facilities repor:ing 44 SPI-00528 It is general 1y be 1 ieve that PVC has GRAS status at the FDA because of its widespread use for 'ood packaging before 1958. In 1975 the FDA proposed (40FR40529, Se it. 3, 1975), however, to withdraw this status for rigid formulations beca se of concern for migration of residual VC into the package contents. lo further action has been taken on that proposal, and major reductions ha' e been made in the residual VC concentrations, so that the FDA has stated on several occasions over the years that -it is considering a formal wi hdrawal of the 1975 proposal. It has not acted yet, however. The food additive petition number 4T2959 was assigned to this matter in November 1984. The miniature liquor bol tie market had been denied to PVC in 1973 when it' was found that the conts nts were "adulterated" with ppm levels of VC. The Bureau of Alcohol, Tax, and Firearms (BATF) has asked the FDA for advice as to the suitabi1ity of reinstituting this use, and the SPI survey referred to above is one step which the FDA required before it would respond to the BATF reqi est. However, plastic bottles as a class now are permitted for liquor use (47FR43944, codified at 27CFR 19.11). Most of the 7.87. of food s packaged in PVC are contacted by flexible materials. The survey s howed 5.357. packaged at least partially in flexible fi1ms, and 1.95 '/. in packages with PVC-containing coatings. The safety of these classes pf packages never was questioned by the FDA, but nevertheless, a "cloud" was placed over all forms of PVC for food contact use by the 1975 proposal 46 SSPPJI--0005300 ....... ~ r - ' ............... *.. ^ ...... ~ increasing sensitivity of analytical methods reveals a growing number of trace contaminants. both natural and man-made, in all of our foodstuff, It will be necessary for both Congress and the agencies to adopt a more realiStic and scientif i|cally based attitude toward minor components cefore this situation can be rlesolved in an acceptable manner. The issue in the case or PVC packaging or potable water pipe revolve; around the question of whether the components of the package, and in seme cases the impurities in these components, can migrate into the package contents in sufficient quantity so as to present a significant, or even a detectable risk to humaip health. Before the Monsanto decision, the FDA had assumed that all of the potential migrations would occur into tne package, ignoring the ld>ss from the wall in the opposite direction or the retention of a portion of the migrant within the polymer matrix. The diffusion of small molecules from and through plastic matrices follows a Fickian process involving a relaxation process of the localized swollen polymer (73-75). A concentration gradient is established in the wall of a pipe or food container vhich provides the driving force for solute movement, and at equilitrium the migration rate is a function of the square of time, the tempjerature, and the residual concentration. The principal limiting facte:r is the original concentration, but theoretical and practical considerations also set a lower limit of concentration below which little significant migration occurs (76-78). Experience has confirmed the theoretical prediction that at VC concentrations in rigid PVC of 1 ppm or less, the concentrations in water will be below the 43 SPI-00532 successor, the Nations' Toxicology Program (86,87). The standard practic has been to accept any positive study in animals as indicative of human hazard, and to equate t he human risk as being equal to that of the most sensit i ve test species Quantitative risk is estimated by a series of "prudent" conservative assumptions, ignoring all comparative pnarmacokinetics or ottje r relevant data, to give a result that may be several orders of magni tude too high (38), if indeed, it has any applicabi1ity to humans (89). The National Academy of Science has urged that a better scientific foundation be laid for this process (82), and this is one of the problems with which the FDA is struggling in its effort to develop a carcinogen pol icy (44FR1 7070, 47FRfl 972). The FDA has indicated, lowever, that it will apply its so-called constituent policy, whi :h it has been using on an informal basis for direct food additives s jch as dyes, to indirect additive also. See 49FR13018, April 2, 198{l Under this policy additives which themselves are not carcinogenic, b j t which contain impurities which are suspected of carcinogenicity, will bu evaluated by a conservative risk assessment procedure. If the resu ting risk is found to be sufficiently small, the additive may be deemed o be safe, this policy allows FDA to avoid application of the De1 an ey Clause to this particular application. A much wider variety of foods is packaged in PVC in Europe than here (61 ,90), including butt* r and butter substitutes, cooking oil, wine, and 50 SPI-00534 employees as that descri bed for meat wrappers. A more realistic, ana higher risk hazard does develco occasionally when an inadequately stabi1ized batch of combound is -processed. An explosive release of HC1 can occur under these c<t> nditions and there are reports of extruder heads being. bJowq. off .the. e.qu. pment with serious re.suIts. Simple "press heat stabi1ity" tests of each batch of compound can eliminate this hazard. Sangha, et al., have shown (94) that, in the 150--175C temperature range (300-35QF) normally us< d for PVC processing, the emitted vapors are less toxic than those from Dc uglas fir at the same temperature. The toxicity increases rapidly at hi cjher temperatures, but of course PVC never is exposed deliberately to such temperatures. Workers at Harvard report tha.t little HC1 is found in Ijot wire cutting of PVC film below 150C ( 300F) but reaches a maximum bj 200C (400F), and that the acid is associated with particulate plastici zer emissions (95). Studies by NIOSH have 1 ed to their conclusion that the solution to the "asthma" problem is to supply adequate point ventilation at the workplace and/or to use lower temp erature cutting devices (96). Potable Water Use One of the major growth areas of PVC has been in the extrusion of rigid shapes for water use. If rigation pipe, drains, conduits, and vents have made up a substantial por tion of this application, but potable water pipe 52 SPI-00536 A contractor for the EFA has estimated that the average home contains about 500" lb. of PVC in its building materials, exclusive of consumer goods and plumbing or siding (106). This compares to about 200 ft. of PVC pipe for a house fully plumbed with plastic and several tons of combusti bj e_ cl_oth_i ng_and _furn i ture . As is discussed in the next section, it is unlikely that the plumbing, specifically, or plastics in general, make any significant contribution to fire toxicity under real life conditions. The controversy over let ching has several facets. These include the extraction of substancefc from the pipe, the passage of external ground contaminants through thu pipe into the water, and the addition of solvents from joint cements to tip e water. The argument reached a <rescendo when a union coalition persuaded the California Housing and Community Development Commission to reject the recommendation of its sijaff that California adopt the Uniform Plumbing Code which would extend the use of CPVC and other plastic pipe to uses inside residential struc tures. These materials had been acceptable for distribution lines, for drain, waste, and vent lines inside homes, and for all uses in commercial tluildings for several years. One basis for this decision was a study (1C7) which reported the presence of several organic solvents and DEHP in water after laboratory tests of simulated use conditions. The DEHP la|ter was found to be the result of laboratory contamination of the samlples, and the solvents were those expected as 54 SPI-00538  In fire-rated cons truction, fire stops and other construction details permit the plastic drain, waste, and vent systems to pass the fire rating tests. - The report_su^ges.te.d_fu vther .studies of worker health effects from solvents, and pointed o t that clastic pipe reduced the frequency of traumatic injury to plui ibers as ccmoared to the use of metal fittings. Tni s latter cone 1 us ion as suooorted by an analysis of the OSHA data for lost work, days of Califc rnia plumbers in 1979 (111), which showed a significantly lower inc' Pence of lost-time accidents for those causes traceable to plastic pif e use than for accidents caused by metal pioe working. 'here had been suggestic ns that persons working with plastic pipe are exposea to harmful mater ials during the cutting of the pipe and from the cementing steps. Much p lastic pipe is cut with knives or tubing cutters which generate no partic ulate matter, but when cut with saws plastics yield "sawdust" that is coarser than that from either metal or wood. In one test (79) an 18-toot h handsaw gave no material below 150 micron size when cutting PVC or CPVC pipe, and the exposure of the worker to respirable dust was 0.4 ng/m^, or approximately background, during the test. Under the same co iditions copper and iron pipe gave material down to, out not below, 30 mi :rons, while maple wood produced 0.17. of particles be 1ow 30 mi crons. Thus, while the particles from plastic are coarser than 56 SPI-00540 Table 4 Tox i ci-ty Data for Several Common Pipe Cement Solvents CHO Cyclohex anone Oral Rat LD50, g/kg 1.6 Irritation Eye, ppm Skin 75 Mild OSHA Exposure Limit, ppm 50 Aquatic Toxicity, TL96, ppm 10-100 Biodegradabi1ity Good Bioassay -- ITC Test Program Yes MEK Methyl Ethyl Ketone 3.4 350 Moderate 200 >1000 Good -- Yes THF Tetrahydrofuran TOL Toluene 3.0 5.0 Moderate 300 Mild 200 200 -- Good -- No 10-100 Good Negative No CMF Dimethyl formamide 2.8 Mild 10 Good ` No Data primarily from Ref. 12. 58 SPI-00542 Table 5 Pipe Cenjent Solvents Extracted from PVC pe as a Function of Time Sol vent First Fi11inq MEK THF CHO OMF Second Fi11inq MEK THF CHO DMF 2 hrs. 0.2 1 .0 0.1 3.1 0.1 0.5 0.7 1 .9 Data from Reference 110. 8 hrs. Concentration (ppm) 24 48 hrs . hrs. 120 hrs. 0.53 0.7 1 .5 2.2 2.9 2.7 7.8 5.6 0.2 0.1 1 .2 1.9 4.9 16 1 1 31 -- 0.4 0.4 0.5 -- 1 .9 2.0 2.4 -- 1.9 0.3 0. 1 -- 8.3 7.1 5.6 240 hrs 2. 3 4.2 0.5 7. 1 -- -- -- ___ 60 SPI-00544 Federal regulation of potable water pipe has been assigned to the E?A ay a memorandum of understarding between it and the FDA (44FR42775). No direct rulemaking has been started by the EPA, but many potential contaminants are regulated by the Ir terim Primary Drinking Water Regulations at 4QCFR141.. Local use is controlled by a variety of building codes and local regulations, most of which are permissive for the use of plastic plumbing in the absence of specific regulations to the contrary.' Induced in those codes which al low unlimited application of plastic piping are those of the Southern B uilding Conference, the Building Officials and Code Administrator Internati anal, the National PlumDing Code, ana me Department of Housing a nd Urban Development of the Federal Housing Administration (29CFR20 3.929 et seq.). The aim of the industry has been to get direct approvals in place, rather than simply the absence of disapproval . These eff arts undoubtedly will be the center of continuing controversy by their op aonents. Self-policing of potabl ; water pipe is supplied through particiDation in the voluntary National science Foundation (NSF) program of inspection and certification. Many pi jmbing codes require the presence of the NSF seal on plumbing supplies, e ther directly, or indirectly by requiring adherence to the ASTM s randards which in turn specify compliance with NSF standard No. 14. Under this system the NSF performs on-site inspection of production facilities a id makes nonscheduled tests of pipe and resin formulations for physic il properties and extractabi1ity. Extraction levels for water contam nants must not exceed the maximum contaminant levels (MCL) set by the EPA under tne Safe Drinking Water Act. Typical results from the NSF pr< igram are shown in Table 6 (85). 62 SPI-00546 Table 6 SfatTc Extraction of Ch 1 oroorganics from New PVC or CPVC Pipe Substance PVC Pipe Total Haloforms^ 1,1,1-trichloroethane Trich1oroethy 1ene Tetrachlorethane CPVC Pipe Total Haloforms 1,1,1-trichioroethane Trichloroethylene Tetrachloroethylene Ex raction No. 1 1_ 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 No. of Samo1es /7 7 7 6 6 6 6 6 6 6 6 6 11 12 12 11 11 11 11 11 11 11 11 11 Results, ODD ND^ 1-3 3-5 5-10 - 70 43 52 60 51 60 42 42 33 51 60 51 o 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 34 27 17 33 10 1 11 0 92 11 0 92 47 10 0 91 4 3 2 0 0 0 0 0 0 0 1 1 0 0 2 0 0 0 0 0 0 0 0 0 Notes: 1 1st extraction, 24 hrs. at 37C 2nd extraction, 24 hrs . at 37C 3rd extraction, 72 hrs . at 37C 2 Limit of detection , 0.3 ppb 2 Primarily chlorofo m Adapted from ref. 79. 54 SPl-00548 T *' ...... 1 ahead with their own prjogram in an effort to get some credible data on the record. The EPA is expected to promulgate Maximum Concentration Limits (MCL's) for a variety of halogenateb materials, including vinyl chloride during 1986. Recommended MCL's (RMCL's, not enforceable as standards) are to be set at zero for all suspect carcinogens, while MCL's are to be set at 1-10 ppo. depending on the streng:h of the evidence of human carcinogenicity. This is not to be confu ed with the strength of the carcinogen, or potency. That criterioip is not considered in this rulemaking. Vinyl chloride is expected to have a 1 ppb MCL assigned, while the other halomethanes and ethylenes will be limited to the 5-10 ppb range. Noncarcinogens such as riethyl chloroform will have health-based MCL's in the several hundred ppb range. Meanwhile, EPA has rece ived mixed advice from its National Drinking Hater Advisory Council on its inquiry as to the desirability of banning lead water pipe, and/or lead- containing solders. Considerable data has been accumulated on public ex posure to heavy metals through these sources (121), and the EPA conti nues to consider the steps needed to reduce the health risks involved fr om these exposures-. There appears to be no health, safety, or environmental reason why PVC or CPVC pipe should not continue to participate fully in the residential and commercial plumbing markbt. The interrelationship between the self-policing by the indjstry and the regulatory and building code systems 66 SPI-00550 free chlorine and hydrp gen chloride from the chlorine components, and water. Even under str nuous incineration conditions, however, the free chlorine content is mi iior. The tendency to form free chlorine is reduced even further when PVC is burned because of the facile removal of HC1 from the polymer chain uoon heating (123,124). Ccnbustlon proceeds at the interface between the air (oxygen) supply and the fuel surface, or interface if a gas (125). Thus extensive preheating occurs before solid surfaces are not enough to ignite themselves or to supply a source of combustible gas because of pyrolosis, ordinarily, both solid surfaces and evolved gases are burning at the same time, especially at the flame front, as is apparent in watching the burning of a fresh piec ; of firewood burn in a fireplace. Articles of PVC thus are subject to con liderable preheating, and most of the cnlorine content is removed as H(fl, reducing the opportunity of the chlorine to participate in further Reactions. The normal combustion products from PVC are found to be, in decreasing amounts: carbon dioxide, hydrogen chloride, and carbon moroxide, with much smaller amounts of benzene, and a variety of C^-Cy hydroc;rbons. Benzene, and smaller amounts of toluene, are formed by cyclization of fragments of the dehydrochlorinated polymer chain. The shorter alkane and alkene products also are formed by fragmentation and re'comt i nations. The authors cited above show that the fraction of carbon evolving as carbon monoxide is a function of the temperature and the air supply. Reduced oxygen availability increases the carbon monoxide ratio, as expected. Increased temperature also increases that ratio, due to more rapid char formation, with resultant poorer penetration of the oxygen into the burning mass. 68 SPI-00552 The key question relevan t to the safety of the public is whetner these combustion products are more hazardous than those oroduce: by ootential suDStitutes, particu1I ar 1 y the "natural" materials used before me introduction of synthetic materials, such as wood, cotton, and wool. A companion issue is whetp er the use of syntheti.c materials has increased the comoustibi1ity of interior furnishings, that is, the ease of ignition, and the rate of flame s D read once a fire is initiated. Eacn of these points will be discusseb below. The unfortunate fact is that there are no reliable tests wnicn can measure these important properti es under real fire conditions, desDite the significant research ef ort that has been expended during the last 15 years (103,128-134). One major problem is that results vary with time as the condition of the sample changes during static tests (135). Laboratory simulation of the amount and toxicity of combustion croducts can be made under car ef di 11y measured conditions, but none of these can depict adequately an are a fire of mixed fuels under real life conditions. An advisory committee of NFPA recently reviewed the results of a workshop on toxic hazard assessment of combustion products that was held on February 23-24, 1984 (116). They concluded that lethality was the most appropriate endpoint for presently-available tests, ana that a ten-fold difference in test resuits is needed to constitute a significant difference in lethality between test substances. The participants were not optimiStic that a me re accurate test system would be available in the near future (137). An active test development program is underway at the National Bureau of Stanc ards (138) 70 SPI-00554 Nevertheless, test results do find their way into regulatory procedures, and they are useful for making relative evaluations of new products and new combinations of materials under fixed conditions. Some of the more prominent of the toxicity tests are summarized briefly in Table 7. It can be seen that the`e is considerable variation among these protocols as to whether the system is static or flowing, the sample is burning or smoldering, and the resu Its are reported as a measure of time exposed or size of the sample used Smoldering conditions are recognized as producing hazardous cont itions no matter what is the source of combustible material (131 ,134,141). Kaplan and coauthors (134), Clarke (131), ana Huidener (142), have sun marized the advantages and disadvantages of eacn ' method recently and the) and the NFPA Standards Council (143) have suggested directions for future research to develop more meaningful results. In general, these test it ethods show that PVC is about equal to or scmewnat more toxic than the conv entional materials such as redwood, oak, cotton, or wool, depending on th e test conditions used. In all test methods all test materials can impar t lethality, and the differences usually are seen in time to incapacitatio i, or degree of irritation (123,134). The primary cause of death routinely is from the carbon monoxide. The rate at wnicn this result occurs is a "unction of the interaction of the oxygen and carbon monoxide (CO) con :entrations, which are controlled by the amount and type of fuels and by the air supply and temperature (141,144-147). 72 SPI-00556 The acute toxicity of C 0 is due to its displacement of oxygen in me hemoglobin to form care oxy hemoglcoin (CHG), thus starving the organs of oxygen. Minor effects appear at acout 157. CHG, severe effects are seen at 307., and death can occu r above 607. reolacement. These effects are seen earlier .in persons with reduced circulatory ability, and blood concentration is increa sed more rapidly for persons who are undergoing heavy exertion, or are under stress. Persons in a fire environment may also have their breathi ng rates increased by high levels of carbon dioxide or other acids. Hydrog en cyanide will act in a very similar manner to CO in displacing oxygen (1 12,143). An equilibrium is estab ished between the ambient and blood levels of C0,` so long-term exposure d< >es not result in cumulative effects, and there is a short period at the bi ginning of high exposure before harmful oicod levels are established. A recent estimate suggests 2,000 ppm-hrs of constant exposure as a hreshold for fatalities (144), if the exposure time is less than 4 hour s. The results generally are reversible unless extremely high exposure; have been received (149). Kaplan and Hartzeil (150) have reviewed the incapacitation effects of CO on rodents and primates and have conck ded that rodents are reasonable surrogates for humans. Their analysis suggest that 750-1,000 ppm-hrs will 'result in incapacitation in humans undergoing light activity. Studies have found an ur usually high percentage of victims of fires in public places have high blood alcohol levels as well as high CHG levels. Impaired orientation anc escape ability may be contributing factors in these cases (130,145). 74 SPI-00558 Several variations of :his tunnel test are in use, such as the Union Carbide 4--ft. tunnel, "he Monsanto 2-ft. tunnel, and the Forest Produce: 3-f t. tunnel (ASTM E286 69 (1975-)). A-variation. on-ttie-ASTty-test-is descr i bed. under NFPA-253-79 and resorts the heat flux on the sa|mple from the radiant panel at the flame-cur coin* (if any). Results are given in Btu/ft 2 or Watts/cm-7 . Other ASTM tests are des igned for specific forms of products, rather than for finishes and surfac|e s in general. These include: ASTM D1929-77, Ignition Properties of Elastics, which measures the self-ignition temDerature of plastics in a heated chamber, and the flash-ignition temDerarure-of the pyroll ysis gases; 01230-61 (1972); Flammability of Clothing Textiles, whlcfi reports the flame-spread time for a standard sample after a 1-sec. e ( posure to a gas flame; and 02633-76, Thermoolasti Insulated and Jacketed Hire and Cable, which includes a flame resistance test, among others. ASTM reports a normal i rtralaboratory variation of 15-50% on these tests and inter1aboratory var ations of up to 1007.. Useful discussions of the principles and applicat ;ons of these tests can be found in references 133,134,146,147,151. Several other groups have published or recommended tests for plastic product f1ammabi1ity. llhese include NFPA, NEMA, ANSI, and UL, and the rests are used for apprc va 1 or certification of many types of products. 76 SPI-00560 The CPSC rule is the on y flammability standard in force on a national basis. The Federal Avi ition Authority regulates the flammability of passenger aircraft inte iors (14CFR25) and recently has proposed a more stringent test for seat (48FR46250. Oct. 11, 1983). Many localities have adopted building codes ieveloped by major organizations such as the Uniform Building Code ( IJBC) of the International Conference of Building Official s or the BOCA B,i sic Building Code by the Building Officials and Code Administrators Inti! rnational , The UBC contained for s fveral years a requirement that the combustion product of plastic builijli ng materials be no more toxic than wood, but withdrew that criterion in the mid-701 s because of the lack, of a suitable test (156). The current code makes no mention of smoke density or toxicity, and sets flame spread criteria, based on the tunnel test at a maximum of 25 for Class I occupancy, 26-75 for Class II, and 76-200 for Class III (general resi dency) uses. Many natural products cannot meet these requirements for public buildings, or Class I and II occupancy, which exclude common ma t|erials such as red oak, because it is used to establish the calibration point of 100 in the test. The UBC nd BOCA codes set similar standards fcr public buildings. Perhaps the most effecti ve force in controlling the flammability of construction materials i s the insurance ratings imposed by the risk carriers. Their experiejn ce with past losses guides their assignment of premium rates and their recommendations for changes to reduce the risks and therefore the premiu ns. Such "se 1 f-pol i cijji ng" by users through 78 SPI-00562 test within a laboratory varies from 10-507., depending on the material, while the inter 1 aborato|ry precision ranges from 30 to over 1007. for most combustible materials, Wood giv-es higher results under the smoldering conditions than when f1 iming, while most plastics give higher density Vjal ues_unde/_fJ amJ rig_ca lditions. The density and thickness of the sample has considerable effect on the results, with high density and thicker samples giving higher sitioke generation (159). The NFPA 253-1982 test uses the same procedure. AS M D2843-77 also is very similar. A widely used flow test method was developed by Ohio State University, and is under study by ASTM <(159) . It also uses either flaming or smoldering conditions, and measure' the optical density of the exit gases from the comoustion chamber under controlled flow rates. Here, too, flaming conditions produce relatji vely more smoke from plastics than from wood, Smoke production also i deceases with air flow, and ASTM E-34, with three times the flow rate of he OSU test, yields higher smoke values than the OSU test. The same is true if the NBS test is run under flow conditions. All of the ASTM, NBS, NF3A, and other tests include a disclaimer at the introduction which cauti ans that the tests "should not be used to describe or appraise the fire hazird or risk of materials, products, or assemblies under actual fire condit ons". Efforts to measure these hazards under conditions more relevant to actual use situations have led to two types of large-scale studies, coni) er tests and room tests. S9 SPJ-00564 Two large studies to measure the combustion products of actual fires have been reported recently (167,163). Both of these studies founa that, in the accessible sampling areas, c-arbon monoxide exceeded the ACGIH-recommended short-term levels and that oxygen levels were below s_hort-t.erm_suryiv.al_ lev els greater than 90% of the time.- Hyarogen chloride above acutely toxic levels (500 ppm) was found 36-53% of the time. Other toxic gases reported frequently were acrolein and hydrogen cyanide. One result of the Bostodi study (165) was a rule requiring all firefighters to wear breathing equipment when involved in structural fires. This action was reported to lave reduced the inhalation injuries to firemen there by 807.. A recent test of the effectiveness of smoke detectors and SDrinklers in typical hotel settings showed little change in either the smoke density or room temperature from the presence of typical plastic-upholstered furniture from the room contents (169). The Coast Guard has confirmed that furniture in general is a major contributor to full-scale room fires (170). As expected, det ;ctors and sprinklers were very effective in these settings. Ionization de*|tectors operated faster than photoelectric types except in smoldering sitnations. Sprinklers appeared to increase the smoke density in the period after activation. SPI-00566 32 Detailed presentation of the results obtained from these many tests is beyond the scope of this discussion. The reader is referred to the monograph (152) or a brief review paper (175) by Hi1 ado for summaries and leading references to nany combustibility tests, and to Kaplan and coworkers (134) for extensive analysis of toxicity tests. The 1981-32 edition of the Modern Plastics Encyclopedia is a good reference for flammability data on specific plastic compositions. Research continues on methods for more meaningful test methods. Some of the most active institu ions include the National Bureau of Standards, Southwest Research Inst tute, and the University of Utah. The Society of the Plastics Industries or its mempers are sponsoring or cosponsoring several studies at thesi and other institutions, and is engaged in a series of demonstration' with fire schools and fire departments illustrating the combusi ion properties of building and furnishing materials. The Vinyl Institute (VI) , a prancn of the SPI, has prepared a series of Technical Information Bt lletins presenting detailed information on many of the issues related to th e combustion toxicity of common building materials, and has devel oped a computer-based bibliography of references on combustion studies. It has also been very active in testing and in educational programs wit h fire fighter associations. Several effective and informative audio-vi sual aids are available from their headquarters in New York. One is a slid ;/taoe program, "Vinyl in Todays Fire Environment", which was produced by the International Society of Fire Service Instructors. A ri 1m, ./ n i c h also is available on video tape, 34 SPI-00568 Heating systems remain :he primary source of home fires, and their relative contribution has been, increasing rapidly since 1980 (176). looking, incendiary act , electr-ical systems, and smoking are, in descending order, also major causes. Smoking remains the major cause of fjre _deaths_._with_ over _ iVo'f the cases ascribed.to that act. It is encouraging that Congre s recently passed a bill establishing an interagency task force inner the C?SC to study a means to obtain a more nearly "fire-safe" cigarette. Heating systems, incendiary acts, children playing, electrical systems, cooking, and open flame are other major identified causes of fire deaths. It is apparent that the najor contribution to continuing the current downward trend in injury and deatn from fires can be made in the area of reducing their incidence It is more productive to prevent fires than to try to protect from thei comousticn products, no matter what the level of toxicity. Proper applic <ltion of synthetic building materials and furnishings, and especia Iv of PVC. with its inherent noncombustibi1itv, can make a major contribi tion to that effort. There is no evidence that the increased use of pla: tics has added to either the incidence of or fatalities from resident' al fires, or of firemen (143,177-131). Nevertheless, because of the economic pressure, and perhaps because of misunderstanding of the relative toxicity of substitutes, pressure continues at the local level to prescribe rules that would ban or restrict the use of plastics in construction (136). SPI-00570 36 Instances have arisen vhere PVC was suspected of environmental problems, but these appear to be attributable to other causes. The case of vinyl chloride contamination of the Dade Co., Florida water supply has been found due to biodegradation of higher chlorinated olefins, as was discussed in Chapter III of Volume I of this Encyclopedia. Vinyl chloride has been found in the e nissions of the West Covina, California landfill (134). There are no kn 3wn sources of VC in that area, but VC-contaminated PVC was suspected as a :ause. However, substantial quantities of the other chlorinated olefiins vwere found there also, and it is most probaole that they are the sourc s of the VC in this instance as well as in Florida. Concentration of VC in the neighborhood since mitigation measures were taken at the dump have not exceeded the California standard of 10 ppb, and the repor t states that the situation does "not constitute a public health emergency'1 The gases from several British landfills also contain VC, and in each of these the probable precursor chlorinated olefins were present (185). Earlier studies at other U.S. landfills also found these precursors, DUt did not use an analytical method suitable to detect VC (186), but more recent data show VC present at almost all such facilities in trace amouits. In a third instance the PA has measured vinyl chloride above a landfill known to have received hu avily contaminated PVC sludges before 1974 (187). None of these ca; es suggest any concern for low-VC resins of the type produced today, nor do they represent any reason why PVC itself is not suitable for landfi11 53 SPl-00572 components act both as sink and a source of chlorine. In any event the quantities of dioxins f 3und are quite low (190-193) and are not expected to add significantly to the natural levels of these substances. In addition, the newer and larger mass burn incinerators are required to nave particulate removal equ pment that lowers substantially the amount wnich may be expected to escai ie. The EPA has made several public statements that municipal refuse ir cinerators of the type now being installed in many cities wi11 have no dels terious effect on human health. There are many positive features to incineration of plastics. Their presence helps control t he water content of the garbage, ana thus the ease of handling and heat yie Id are improved. They contribute a significant heat output to those uni ts having heat recovery, and the heat value of PVC is about twice that of t he average garbage. Operators preparing pel 1etized fuel for snip ment to central incineration units rely on the thermoplastic components to help adhere the pellets for easier and cleaner shipment. Mass burn incinerators a_e in much more common use in Europe than here, and the level of per capita use of plastics, including PVC, also is several times higher the*e than here. Several thorougn studies of European experience are presented in reference (182), and that technology now is coming into much wider use in this country. In particular, hydrogen chloride emission has not been a problem even in the absence of specific air pollution control equipment for that component. T' ... ~r - ~ r '' SPI-00574 ............... .. ACGIH ANSI ASME ASTM BOD 5 CFR CHG CMA CPSC DEHP EP EPA FDA FR GRAS ITC LD50 NBS NFPA NSF Glossary of Acronyms - American Conference of Government and Industrial Hygienists, Cincinnati, 01 45211 American National Standards Institute American Socie ty of Mechanical Engineers, New York American Socie ty for Testing Materials Five-day biolcgicaJ oxyge-n demand Code of Federal Regulations, a compilation of promulgated Federal rules. 29CFR1910 refers to Chapter 29, Section 1910. Carboxyhemoglobin, formed by reaction of carbon monoxide with the blood. Chemical Manufacturers Association, Washington, DC 20037 Consumer Produ:t Safety Commission, Washington, DC di-2-EthylhexyI Phthalate - Extraction Pro edure, designated by EPA for use in testing hazardous wasta candidates Environmental Protection Agency - Food and Drug administration - Federal Regist< r. The official daily publication of the federal government. Tie number before the letters gives the volume, tne following numbfrs are the page. Volume 49 is being published in 1 984. Generally Regarded as Safe by the FDA because of broad use before 1958 Interagency Tes ting Committee, which recommends substances for consideration ty EPA for toxicity testing - Lethal dose for 507. death of the experimental animal within 14 days - National Bureau of Standards, Washington, DC National Fire Protection Association, Quincy, MA National Sanitation Foundation, Ann Arbor, MI 92 SPI-00576 References 1. Kimura M., Takashima, ., Nakatoni, M., Kikkawa, H., Nagao, S., and Itckowa, Y. (1934), ,h Chrcmotoq., 295 503. 2. Kander, 0. S. (1984), , . Tox. Environ. Health, 14 471. 3. National Toxicology Prog ram (1982), "Carcinogenic Bioassay Stannous Chloride in F344/N Rat! and B6C3F1/N Mice (Feed Study)," NIH/Pub-82-1737. NTP-31-33, July , PB 81 -242553. See also Schafer, S. G. and Femfert, U.. (1934), Reg. Toxicol. Flharmacol . , 4 57 for a review of the toxicology of tin. 4. Takanashi, M., Furukawa , F., kokubo, T., Kurata, Y., and Hayashi, Y. (1933), Cancer Letters, 20 271. 5. Cordarel1i, N. F., Corda re 11i, B. M., Libby, E. P., and Dobbins E. (1984), Aust. J. Exp. Si o 1 . Med. Sci . , 62 (2) 209. 6. Schafer S. G. and Femert U. (1934), Reg. Tox. Pharmacol., 4 57. See also "Tin as a Vital Nutrien : ," CRC Press, i n press, 1 985. 7. Boettner, E. A., Ball, L., Hollingsworth, Z., and Aquino, R. (1932) "Organic and Organotin Compounds leacned from PVC and PVC Pipe," EPA-600/1 -81 -062, Febru.i ry. SPI-00578 16. Hobson, J. F., Cor ter.. D. E.-, and Lightner, D. V. (1984), Envi ron . deal th , J_3 95 ?. Tox. 17. Keith, L. H. (1981), E iviron. Sci. Techno 1., 15 156. 18. Ray, L. E., Murphy, H. E.,-and Giam, C. S. (1983), Chemosohere. 12 1031 19. ron, Health Persgect., as, 20. Samfield, M. and Raw 1 i rjgs, G. D. ( 1 979), Environ. Sc i . Technol . , J_3 150. 21. Atlas, E. and Giam, C. S. (1981), Science. 211 163. 22. Sugar:, R. h., O'Grady, D. P., Banerjee, S., Howard, P. H., and Gleoni 11. ,-t. E. (1984), Appl, Environ. Microbiol., 47 601. 23. Hornwig, H. E., II, Robinson, E. L. , and Pet.rasek, A. C., Jr. (1934) Arch. Environ. Contam. oxicol.. 13 191. 24. Engineering-Science, Inc . (1982), "CMA Five-Plant Study," prepared for the Chemical Manufacture rs Assn., Hashing ton, DC, April. 25. Catalytic, Inc. (1979), "Laboratory Study of Biological System Kinetics unpuolished study for tl e EPA, Philadelphia, PA. SPI-00580 33. DeAngelo, A. B. and Garrett, C. T. (1983), Can. Letters, 20 199. 34. Ward, J. M., Ohshima, `4., Lyncrr, P., and Riggs, C. (1984), Cancer Let:., 24 49. 35. K1ume, W. M., Haseman, J. K. and Huff, J. E. (1983), J. Toxicol. Environ. Health, 12 153. 36. National Toxicology Program, Third Annual Report on Carcinogens (1983), PB33-135355, Washington, DC. 37. International Agency fq>r Research on Cancer (1982), "IARC Monograph No. 29 on the Evaluati (j>n of the Carcinogenic Risk of Chemicals to Humans," Lyons, France May. 38. Fatiadi, A. J. (1983), "Priority Toxic Pollutants in Human Urine: Occurrence and Analysi National Bureau of Standards Report NBSIR83-2690, June, PB 83-22588. he i r 39. National Cancer Institute (1979), "Sicassay of Phtnalic Anhydride for Possible Carcinogenicity," TR-159, Wasnington, DC. 40. Kornbrust, 0. J., Barfknecht, T. R. , Ingram, P., and Shalburne, J. 0. (1984), J. Toxicol. Environ. Health. 13 99; Zieger, E., Haworth, S., Mortelmans, K., and Speck, W (1985) , Environ. Mutagen,, 7 213. ga SPI-00582 50. Chemical Week (1984) February 15, p. 26. Probing Chemical Causes of Infertility," 51 . Environmental Protection Agency (1979), "Phthalate Esters Quality Criteria," PB2(6804, Washington, DC. Ambient Water 52. J. A. Todhunter (1982), "Review of Data Available to the Administrator Concerning Formaldehyde and Di(2-Ethyl hexyl) Phthalate (DEHP)," memorandum dated Februalry 2, U.S. EPA, Washington, DC. 53. Department of Health and Human Services, Public Health Service (1933), "Third Annual Report on Carcinogens," NTP-32-330, Washington, DC, September, PB83-135855. 54. Turnbull, D. and Rodricss J. V. (1985), J. Am. Coll. Toxicol., 4111. 55. National Toxicology Program (1982), "Carcinogenic Bioassay of Butyl Benzyl Phthalate," Technical Report Series 213, NIH Pub. No. 32-1769, Bethesda, MD, August. 56. National Toxicology Program, "Carcinogenic Bioassay of Oialkyl Phthalate," Technical Report Series 242, Bethesda, Md. , PB83-200824. 57. Monsanto Chemical Co., 'eport to EPA Office of Toxic Substances, as reDorted by Pesticide aid Toxic Chemical News, October 12. 1934, p. 12. ! no SPI-00584 66. Expert Panel, Am. College of Toxicology (1983), ,L Anr Col 1 ege Toxicol 2 35. 67. National Safety Council (1983), "Work Injury and Illness Rates, 1983 Edition," Chicago, IL Stock No. 125.63. 68. Lampert, B. L. and Edwa rds, R. G. (1983), Am. Ind. H yq. Assoc. J., 44 394. 69. Riko, K. and Alberti, P . W. (1983), J. Occup. Med., 25 523. 70. Withey, J. R. (1977), Origins of Human Cancer. Bock A, 219, Cold Spring Harbor Laboratories. 71. Usmani, A. M. (1981), ot-q. Coat. Plast. Chem, . 44 357. 72. Keller and Heckman (198? ), "Report on Food Contact Usage of PVC Plastics," prepared for the SPI, August 17, Washington, DC. 73. Chang, S. S., Senich, G A., and Smith, L. E. (1982), "Migration of Low Molecular Weight Additives in Polyolefins and Copolymers," NBSIR82-2472, National Bureau of Standards, Washington, DC. 74. Berens, A. R. and Danie s, C. A. (1976), Pol . Eng. Sci . , 1_6 552. 75. Berens, A. R. (1977), Pclymer, 13 697. See also Polymer Preprints (1974), 15 (2) 203. 102 SPI-00586 85. Havender, W. R. (1984) "Of Mice and Men: The Benefits and Limitation of Animal Cancer Tests," March, Am. Council on Services and Health, Summit, NJ 07901. 86.- -Report, of-the-Ad- Hoc- f2-4ne 1 on Chemical Carcinogenesi s-Testing and Evaluation of the Nati dinal Toxicology Program Board of Scientific Counselors, National Toxicology Program, 1984. 87. Aim, A. L., "Accuracy In Risk Assessment," memo dated Jan. 25, 1934. U.S. EPA, Washington, DC. 3ee also Ruckleshaus, W. D. (1983), -L A i r Pollution Control Assn. , 33 731. 88. Hoel , D. G.-, Kaplan, N. 1., and Anderson, M. W. (1983), Science, 219 1032. 89. Environmental Protecti cn Agency (1984), "Health Assessment Document for Trichloroethylene," EPA -600/8-82-006, 1984; W. R. Havender, Regulation, 3 (1) 13. 90. European Council of Manufacturers1 Federation (CEFIC) (1976), "Vinyl Chloride Toxicity and tie Use of PVC for Packaging Foodstuffs," Brussels, Belgium, February. 91. Chemical and Engineering News (1983), "Plastic Bottles Due for Strong Growth," August 1, p. 104 SPI-00588 100. Chemical Week (1983), More Flap Over PVC Pipe Safety," January 12, p. 18. 101. Chemical Week (1983), Plastic Makers Answer Back on Fire Safety, January 26_,-p_-2ij-------- 102. Flax, S. (1983), "The Cjubious War on Plastic Pipe," Fortune, Feb. 7, p. 68. See also Chemic al Week, March 17, 1983, p. 52 and Industry Week, May 31, 1982, p. 76. 103. Rawls, R. L. (1983) , " Flire Hazards of Plastics Spark Heated Oebate, Chemical Eng. News January 3, p. 9. 104. Love, K. (1982), "Plast ic Pipe: War Shifts to Safety Front," Los Angeles Times, Feb. 16, p. 1. 105. BNA Chemical Regulation Reporter (1983), Current Report section, p. 280, May 6. 106. Vensor, Inc. (1983), "H 'pothetical Assessments of Exposures to Incidentally Produced Pit B's," report to the EPA under contract 68-01-6271 , Task 21, Feh . 13, Springfield, VA 22151. 107. Montgomery, James M. (1 !l80), Consulting Engineers, Inc., "Solvent Leaching from Potable W ter Plastic Pipes," prepared for the California Dept, of Health Service , Pasadena, Ca. 106 SPI-00590 115. National Sanitation Foundation (1983), "Proposed Organohalide Leachate Testing Protocol for Pf! a s t i c Pipe," Ann Arbor, MI, April. 116. McClelland, N. (1983), as quoted on p. 9 of the June 27, issue of Food - - - -Chemi cal- News-.---------- --- 117. Cassady, M. J., Cole, . F., 3ishop, T. A., and Pfau, J. ?. (1983), "Phase One Report On E' aluation of the Permeation of Organic Solvents Through Gasketed Joints d and Unjointed Poly (Vinyl Chloride), Asbestos Cement, and Ductile Ire n Water Pipes," Battelle, for the SPI, Columnus OH, October 28. 118.. Schuuring, C. (1981), 1 Dutch Dumps," Nature, 289 340. 119. Environmental Protectio n Agency (1975), "Scientific and Technical Assessment Report on Vi nyl Chloride and Polyvinyl Chloride," EPA-600/6-75-004, June. 120. Dressman, R. C. and McF irran, E. F. (1978), J. Am. Waterwork Assn., 70 29. 121. Sierra Club, Los Angele Chapter (1934), "Toxic Metals in Drinking Water," Los Angeles, Ca 122. Muggett, C. (1985), "Coiribustion Conditions and Exposure Conditions for Combustion Product Toxic ity Testing," J^ Fire Sciences, 323. 103 SPI-00592 133. California State Fire Marshall (1983), "Assessment of Test Methods to Determine Combustion/ lloxicity of All Materials," ACR-146, Sacramento. CA, April 30. 134. Kaplanf H; L., Grand, L F., and Hartzell, G.- E. (1983), "Combustion Toxicology," Technomic Pub. Co., Lancaster, Pa. 135. Algleeff, G, V. and Pa: khaun, S. C. (1984), "Evaluation of Smoke Toxicity Using Concentration-Ti ne Products," .L Fire Sciences, 2 362. 136. Rawls, R. (1984), Chem Eng, News, 62, pg. 20, July 9, 1984. 137. Snell, J. E. (1984), F re J^, 18 (5) 69, September. 138. Bukowski, R. W. and Jones W. W. (1985), Fire J., 79 (2) 24. 139. Jaeger, R. J., Skornik W. A., and Heimann, R. (1982), Anu Ind. Hyg. Assn. J^_, 43 900. 140. Belles, D. W. (1982), uiId. Official Code Administr,, Sept./0c:. p. 17. 141. Brown, E., Krasny, J. ., Peacock, R. D., Smith, G. F., and Stolte, A. ( 1982), J^_ Consumer Prc duct Flamm., 9 (4) 167; Gad, S. C. and Smith, A. C. (1983), J Fire Sciences, 1 465. 142. Hindener, R. K. (1984) J. Fire Sciences , 2 82. 110 SPI-00594 152. Hi 1 ado, C. J. ( 1982), ' Flammability Handbook for Plastics," 3rd ed. Technomic Pub. Co., Lar caster. Pa. 153. Ikeda, G. K. (1983), " Cjxygen Index Tests to Evaluate the Suitability of a Given Material for Oxy c en Service," in "Flammability and Sensitivity of Materials in Oxygen-Enr|iched Atmospheres," ASTM STP812, 8. Werley, ed.. ASTM. 154. Dickens, E. D., Jr. (1984), J. Fire Sciences, 2 123. 155. Peterson, B. A. and King, R. R. (1980), Fire Technol., 16 143. 156. Birky, M. (1982), Testi nony Before the New York State Finance Fire Safety Subcommittee, May 6, 193 2, transcript, p. 36. 157. Hilado, C. J., Murphy, . M., and Casey, C. J. (1978), Fi re Technol . , 1_4 113; Hilado, C. J. (19711 ), Fire J., 73 (6) 69. 158. Hilado, C. J., Murphy, (;. M. , and Huttlinger, P. A. (1980), F i re Technol., 16 273. 159. ASTM (1980), "Proposed lest Method for Heat and Visible Smoke Release Rates for Materials," AS TM Standards, Part 18, p. 1382. 160. Landrock, A. H. (1983) "Handbook of Plastics Flammability and Combustion Toxicology," Noyes Publiications. Park Ridge,.N-J- 112 SPI-00596 171. Hall, J. R., Jr. (198!>), "A Decade of Detectors: Measuring the Effect Fire J., 79 (9) 37. 172. Avento, J. M. (1981), Modern Plastics Encyclopedia, 1981-1982, McGraw-Hi 11. 173. Offensend, F. L. and Martin, S. B. ( 1 982), Fi re Techno]., 13 (1) 5. 174. National Toxicology Pr ugram (1984), "Annual Plan for Fiscal Year 1984," NTP-34-023, February. 175. Hilado, C. J., Cumming H. J., and Machado, A. M. (1973), Mod. Plastics 55 61 . 176. Powell, P. (1984), Fir4 J., 78 (1) 17; Karter, M. J., Jr. and Gancarski, J. L. (1984 , ibid, 78 (5) 48. 177. Jannerfeldt, E. (1982), Health Hazard Evaluation No. HETA-82-114-1097, Fire Dept., Houston, Tx ., National Institute of Occupational Safety and Health, Cincinnati, Oh. , PB 83-193739. 178. Ibid. (1981), Health Haz ard Evaluation No. HETA-81 -1 39-990, Federated Fire Fighters of Nevada , Las Vegas, Nv., National Institute of Occupational Safety and Health, Cincinnati, Oh., November 1981, PB 83-199919. 179. Lewis, S. S., Bierman, II. R., Faith. M. R. (1983), Cancer Mortality in Los Angeles City Fire F ghters, AACR Abstracts. No. 743. 114 SP1-00598 188. Uchida, S., Kamo. H., .[Kubota-, H.. and Kanaye, K. (1983), Ind. Eng. Chem. Process Des. Dev. . 22. 144. 139. Grummett, W. 6. (1982), "Environmental Chlorinated Dioxins from Combustion -- The Trac? Chemistries of Fires Hypothesis," in Pergamon Series on EnvironmentaI Sc rence, 5 253, Pergamon Press. -190. Hoffman, D., Patrianokos, D., Brunnerman, C., and Gori, K. G. (1976) Anal. Chem.. 43 47. 191. Bumb, R. R. (1980), Sc: ence, 210 385. See also, "Interim Evaluation of Health Risks Associatec With Emissions of Tetrachlorodioxins from Municipal Waste Resource Recovery Facilities," Environmental Protection Agency, November 1981, and Czuczwa, J. W. and Hites, R. A. (1984), Environ. Sci. Techno]., 18 444. 192. Schaub, W. M. and Tsang , W. (1933), Environ. Sci. Technoi., 17 721. See also, Karasck, F. W., Viau, A. C., Guiochon. G., and Gronnard, M. F. (1984), Chromatog., 170 227, and Schaub, W. M. (1984), "Technical Issures Concerned with 3CDD and PCDF Formation and Destruction in MSW Fired Incinerators," Na:ional Bureau of Standards NBSIR 84-2975, Gaithersburg, Md., November. 193. Karasee, F. W., Viare, A. C., Guiochon, G., and Gounord, M. F. <1933 >, J Chroma tog., 270 227. spl.00600 116