Document 2jRzrBVz0ggz56G0XpJrRBNKa

S/' l~ ! Georr Vinyls TECHNICAL SERVICE REPORT POLYVINYL CHLORIDE COMBOSTION TOXICITY: A COMPARATIVE REVIEW World-wide studies by aeadeaia, governaent and industry have shown that the ooabustion toxicity and hasard of PVC is not unique or extreae, but ia siailar to that of aany other aaterlals, including wood. In fact, the ooabustion products of aany coaaon aaterials can cause incapacitation or death in test aniaals faster or by burning less aaterial than PVC. Robert X. Hinderer, Ph.D. Senior Toxicologist and N. M. O'Hara, Ph.D. Vice President of R*D Noveaber 1, 1982 rc; tJ ^ O h* Supersedes June 1, 1981 report TW>^aMOaaOiwnr,ftnmlHftr/9100Oc*T-- 8>v<l.CWvWond<Xo*4131 3 9V < / Chwnteal Gioup : ABSTRACT For 20 years, considerable research has been conducted to identify and study the effects of combustion products from polyvinyl chloride (PVC;. More recently, efforts have been made to compare the combustion toxicity of PVC and other materials. Although universal test methods have not been developed, many investigators have compared the combustion toxicity and hazard of numerous materials. These world-wide studies by industry, academia, and government have shown that the combustion toxicity and hazard of PVC is not unique or extreme but is similar to that of many other commonly used materials. Such research has established that even common building materials such as wood will yield numerous irritating and corrosive substances which can cause pulmonary edema. INTRODUCTION Over the past several years, there has been considerable speculation about the toxicity of PVC combustion products relative to those of other materials. Me have seen assertions that PVC combustion products are much more dangerous than combustion products from other materials, and PVC comboston products lead to "early knock down", or combustion products from alternative materials are safe or safer than PVC, with little if any factual basis. Reports of real fires, perhaps out of ignorance, sometimes indicate effects common to combustion products of "*any materials, as "unique" to PVC. While it is often mentioned that *e combustion of PVC products produces Irritating and corrosive HC1, uae fact that other materials such as wood also produce combustion products which are corrosive and can cause pulmonary edema is rarely mentioned (Xikria et al, 1977). Some people believe that one can predict the relative toxicity of various materials by comparing the type and levels of combustion products given off. This is not correct. There are many variables including the physical properties of the material which affect toxicity. The most meaningful and feasible way to assess the relative toxicity of the materials is to evaluate the combustion toxicity using an animal model. la 1982, the Rational Bureau of Standards--through a select committee of "academic, government, and industry axperts--published a test method to evaluate the combustion toxicity of materials used in buildings and transportation (Lavin, NBS, 1982). Although there is still no world-wide concensus, the past 20 years of testing has provided us with a very expansive and extensive body of knowledge relating to the toxicity of combustion products from a host of materials. This document summarises research comparing the combustion toxicity of many materials including polyvinyl chloride. RESULTS AMP DISCUSSION Although a considerable amount of combustion research has been conducted in the last 20 years, only a few studies have tried to compare the combustion toxicity of various materials. Furthermore, most of these comparative studies have been conducted within the last six years. These recent studies have been extremely valuable because they have provided the ability to compare materials under identical test conditions. The data provided in the following discussion are excerpts from these studies. Because of protocol differences, the data in one study may not be directly comparable to another. However, the data in each study does enable a comparison of the combustion toxiclties of the materials tested. The following discussion compares the effect of numerous materials on various parameters of toxicity and hazard. MORTALITY Studies qf the lethality of combustion products of various materials in mice have been conducted by Alsrie and Anderson (1979) and by LeMoan and Chaigneau (1977). The results of the studies are presented in Tables 1 and 2. Because the methodologies in these studies differ, only the materials within each study can be compared. Table 1, the data shows that the combustion products from pvc rank among the least toxic of materials tested*. Only Douglas fir and fiberglass were numerically higher (less toxic).Similar results are evident la the Trench study (Table 2). Again, PVC combustion products rank among the least toxic materials. More recently, the National Bureau of Standards published the results of interlaboratory evaluations of its new combustion toxicity test method (KBS, 1982). In these evaluations, groups of animals were exposed to the combustion products of PVC and numerous other materials for 30 minutes followed by a 14 day observation period. A 10 minute exposure period was also used to identify rapid acting toxicants. The lethality (LC50) of these materials was compared with that of wood. Any ^material with an (LCgo) value mete than one order of magnitude lower 'than that of wood was considered significantly more toxic than wood. In these studies, the combustion toxicity of PVC was found to be similar to *that of wood (Table 3). Furthermore, the combustion produets of PVC did mot produce rapid intoxication. ' RBSPIBATORT ETTBCTS Alarleaaid Anderson (1979) studied potential respiratory effects of the combustion products of various materials in mice. The data in Table 1 shoes the Mount of material required to produce a 50% reduction in respiratory fats and a Stress Index of 100 (200 - maximum stress). For both parameters, PVC is in tha mlddls range of those materials tested. 3 C u K* & O 2 Asphyxiation, collapse, and incapacitation are terms which describe a r general toxic response to short-tern (acute) exposures to chenicals -- in this case, conbustion products. These parameters are Important not only because they provide evidence of toxicity, but because they reflect the relative levels of incapacity. Many studies have determined the amount of material required to cause asphyxiation and collapse, while others have measured the time required to cause incapacitation and/or death. Xn all cases, these have been used to evaluate and compare the toxicity and hazard of various materials. Studies by Alarie and Anderson (1979) and Kishitani and Yusa (1979) compared the amount of material required to cause asphyxiation and collapse, respectively. The results in Table 1 show that PVC combustion products are among the least asphyxiating of those materials tested. In the latter study, PVC also ranks among those which required the most amount of material to cause collapse (Table 4). Numerous other investigators have evaluated the hazard of various materials by determining the time required to produce incapacitation or death when samples are combusted at various temperatures ranging from low temperature non-flaming to high temperature flaming (Hilado, Lopez, and Damaat, 1977} Hilado and Brauer, 1979} Spurgeon, 1978} Crane, Sndecott, Sanders, and Abbott, 1978). Here again they have found that PVC ranks among those which require the greatest time to cause incapacitation and death (Tables 5-8). OVERALL CLASSIFICATION O Although there is presently no universal teat method that enables the classification of materials based on their combustion toxicity, at least one study attempts to classify numerous materials (Alarie and Anderson, 1981). Based on time-response and concentration-response data, the authors placed the materials into one of three categories: Class A--as toxic as wood} Class 3--more toxic than wood or Class C--much more toxic than wood. . Alxteen of the 23.evaluated fell into the Class B category. These included PVC (92% bomopolyaer), ABS, urea formaldehyde foam, polyester resin-acrylic.modlfled (commercial), wool, pvc-a (46% haaopolymer), ^modacrylie ^polyester reain-styrenated halogen modified, cellulose^-flexible polyurethane foam, flexible polyurethane foam with <*.- fire retardant, high resilience polyurethane foam, high resilience polyurethane foam with fire retardant, polystyrene expanded, polystyrene expended with a fire retardant, and phenol formaldehyde-phenol resin, Ja. , expanded with blowing agent. This shows that PVC combustion toxicity is comparable with that of the majority of materials tested and is less than that of some (Class C). 4 COHCLOSION Although there has been auch speculation and misinformation regarding the eoabuation toxicity of PVC, experimental data generated In world-wide studies by industry, academia, and government do not support the contention that the hasard froa PVC coabustion products is unique or extreme. The data discussed above and presented in Tables 1-8 clearly show that the coabustion toxicity and hazard of PVC is coeparable with any other natural and synthetic building materials. Purtheraore, studies conducted by Zikria et al (1977) show that wood, not just pvc, contains numerous irritating and corrosive substancee which can cause pulmonary edeM. When one considers that the combustion toxicity of pvc is siailar to that of other aaterials, that PVC flaae spread is low, and that PVC comprises only a saall portion of building aaterials, it is difficult if not impossible to develop a scenario that suggests that PVC presents a unique or unacceptable hasard. TABLE 1 A COMPARISON OF THE EFFECTS OF THE THERMAL DECOMPOSITION PRODUCTS OF VARIOUS MATERIALS ON SEVERAL TOXICOLOGICAL PARAMETERS WITH MICE MATERIALS RD50* (eg) S1100b (g) LC50c . (g) Flexible urethane foaa #1 Flexible urethane foaa #2 Flexible urethane #1 .Flexible urethane #2 Rigid urethane foaa #1 gid urethane foaa #2 Rigid urethane foaa #3 7 Rigid urethane foaa f4 Isocyanurate #1 Isocyanurate #2 Polystyrene tl % Vo^styreM #2 ,,Fbenbl foraaldehyde booglaafir flrea foraaldehyde 3.7 3.3 9.2 8.6 124.0 v 93.0' 92.0 85.0 106.0 218.0 22.0 i4.o 765.0. 34 To - 20.0 17.0 * 16.0 - ' *. gHDOJ/;V. .V'.; x 1. -- -- 1.5 2.0 1.3 2.0 0.7 ... 0.7 0.9 0.5 0.5 0.7 4.0 13.0 4.0 4.0 ' 0.3 d 1.0 -- -- -- 13.0 10.4 8.3 14.4 10.4 8.2 7.5 8.0 6.4_ 6.1 5.8 10.0 6.3 63.8 2.5 0.64 15.2 35.7 11.9 3.0 Asphyxia' Rar.g< (g) 2-15 2-10 7-10 2-10 2-8 2-8 2-7 2-6 2-5 0.8-2 5.3-10 8-13 3-6 55-95 0.4-1.25 0.3-1.0 8-20 -- -- ** ^V^Vr-2* V Mple weight.associated with a 50% decrease in respiratory rate, * ^ .. .5 fftress indexr 200 is aaxiaua value. **C*Ae''mouvn- *t>.o f > Material required to produce mortality in 50% of the calculated T.v*'. '*(<!' boOOpoXyaeri incorrectly reported as 52% hoaopolyaer) . (Alarig and Anderson, 1979) .L'.- . .r - ,vTi - -yp* C u p ** o 05 1 r'. (tit! n*u6T0 P maul) SOO'O ceo*o oecro 8E0*0 350-0 rro n-o 091*0 (satvao) AXIlViHOW %0S HOd davIOOSH 31dMVS 40 XN00HV tWliA MtltfOldCfOd a*q)2nxa* .A... a*i^o ixod i-.'s-y 3A* T T 3* juoiAjtD*T<4 TYIHStXYW T3o008) xzxt xv aaxsnawoD sz aidwvs zsx nshm SDZH N1 3XYH AXZ1VXH0H 0S V 83QO0M OX awmow TviHaxvw ao inoowy asx ao Nosnrvdww 3 318VX r TABLE 4 COMPARISON OP THE AMOUNT OP MATERIAL REQUIRED TO PRODUCE COLLAPSE IN MICE IN 2.5 to 6.0 MINUTES WHEN COMBUSTED AT SELECTED TEMPERATURES 'Untreated plywood wi.'1......... Fir* retardant-treated \- plywood -'Melamine resin-impregnated sheet #1 Melamine resin-impregnated sheet 2 AMOUNT OP SAMPLE REQUIRED FOR COLLAPSE (Grams) 1022F (5S0C) 1562P (850C) 5.50-8.32* 3.72-5.83* 6.13-9.11* 3.47-5.21* 3.09-4.46* 4.35-6.98b 1.31-2.19* 3.42-4.91* 2.20-3.55* 0.60-1.00* 2.47-3.72* 0.64-0.97* 0.61-0.88* 0.38-0.55* 0.25-0.39* 10.7-15.20b 0.24-0.33* 0.17-0.25* 0.41-0.60* 0.24-0.33* 17.7-24.60 0.58-0.87* 2.61-4.02* 34.5-48.60 1 TABLE 5 r \ COMPARISON OF TIMES TO DEATH IN MICE EXPOSE*TO THERMAL DECOMPOSITION PRODUCTS OF MATERIALS <1.0g) COMBUSTED AT TEMPERATURES RANGING FROM 392F (200C) T01472F (800oC) MATERIAL Tda (MINUTES) V t * Wool fabric Cotton Rad oat Polypropylene Nylon 6 " ' Douglas fir Polyethylene PVC-nylon PTC *2 PVC#1 Polychloroprene #1 .... Nylon fabric ABS Polyatyrene CPVC #1 owe #2 .. . ; **v-* . * -V Polychlorprene #2 ... . .. ".'-ufe'- V: 7.6 10.3 11.5 13.0 ,, >3.5 13.6 14.0 15.3 16.4 16.8 17.4 iili\ 19.3 20.0 21.8 22.7 27.5 1 V .. "\S$jh:i# -f A vi. t* , *TilM -to i-&. (Bilado at al>lt77lV * ** v - ... . J*. t -7 i&r' * * -- i jzr'-.'-zzti- r %-- *' 6r-WyV%*7 ' *10 - * 41 f* v > TABLE 6 COMPARISON OP TIMES TO DEATH IN MICE EXPOSED TO THERMAL DECOMPOSITION PRODUCTS OP MATERIALS (l.Og) COMBUSTED AT TEMPERATURES RANGING PROM 392<>P (20QOQ TO 1472F (800C) MATERIAL Td* (MINUTES) Polyathlana #2 Polypropylene Polystyrene Polyoxysethylene ABS #19 Nylon (121) ~. . Polychloroprene #12 Polyurethane #6 Polystyrene #2 PTC #U . Polyamide #9 Polychloroprene #15 Red Oak Douglas fir i * ' - J V Time-art- o.* dt'eath (Hilado at al 1979) 11.5 10.4 22.1 2.6 7.1 6.8 12.2 8.7 15.4 14.5 2.7 16.3 6.1 7.0 O il - 11,- flp r-; rar.it COMPARISON OP TIMES TO INCAPACITATION AND DEATH IN MICE EXPOSED TO THERMAL'DECOMPOSITION PRODUCTS OP MATERIALS (0.7SQq> COMBPSTED AT 11120? (600C) FOAMS Ti (MINUTES) PR polyurethan* "7.^4. a; PR PET polyurethane 4.8 PR polyurethane 5.0 PR PS polyurethane S.l PRpolyethylene '4 V * yy-.^** ^ -^rraCS ' .w""- ' - - ---" -ppc-v: v-- PR-PR urethane 'v -"r.*'rj^ * r* * * PR .urethane............ ' - > :;3i5 t 5.6 .. 7.6 PR PR . > vurethane ' .... - Modncryllc f' " -ST*" " PR wool-~ .` - <. .:- ... : PolyaulOe 1 * PR wool-nylon Poljiiiae;^. v.f*c vf?0 - weol/PVC tih'riimSti "fr* #*r ult MAMgm . Bf--gzz. -V: i.i - *>. . :V'2.0 : i . : v^'v' * '* 2.1 2.2 .^ V ** wr*. 4* -- 't. - ru^;3.V r* . 4.2 S'.?; :* - *->. ts - ^ * 6.1 4.2 4.6 6.1 S.l 10.2 14.4 "aBl t w... 4r2f,; .*5ai -- *<**--. - TABLE 8 COMPARISON OP TIMES TO INCAPACITATION AND DEATH IN RATS EXPOSED TO THERMAL DECOMPOSITION PRODUCTS OF MATERIALS (l.Og) COMBUSTED AT TEMPERATURES RANGING PROM 887Qp (4750c) TO 1S62P (8S0P) ELECTRICAL INSULATION MATERIAL Silicon* rubber, glass braid. Mylar Silicon* rubber, x-polyethylene Polyethylene, aluainua, PVC F9C Polyethylene EPR-neoprene Cross linked polyethlene bane (cross-linked polyolefin) Teflon BPR/Bypelon Proprietary conpound/neoprene alar Tefsel TIME TO INCAPACITATION (MINUTES) LTNPa 21.9 19.3 LTPb - HTPC - 17.7 8.2 12.1 7.3 21.2 12.0 7.3 6.8 10.6 7.3 9.2 9.4 7.4 13.2 15.7 7.2 13.8 9.7 7.0 6.7 11.2 8.2 13.5 13.2 6.3 10.8 s.a 9.0 5.9 7.9 4.7 10.8 10.4 4.5 7.0 7.7 4.5 Low tenperature nonflaaing b&0W'tenperature flaning llgh tenperature flaalng (Crane et al, 178) - 13 - REFERENCES Alarie, Y. and Anderson, R.C. (1979). Toxicologic and acute lethal hazard evaluation of thermal decomposition products of syn thetic and natural polymers, Toxicol, Appl. Pharmacol. 51i 341-3*2 Alarle, Y. and Anderson, R.C. (1981). Toxicologic classification of thermal decomposition products of synthetic and natural polymers, Toxicol. Appl. Pharmacol. 57: 181-188. Rirky, M.M., Levin, B.C., and Paabo, M. (1981). Development of recommended test method for toxicological assessment of inhaled combustion products. The Toxicologist 1: 7. Crane, C.R., Bndecott, B.R., Sanders, D.C., and Abbott, J.K. (1978). Electrical insulation fire characteristics, Volume II: Toxicity, Civil Aeromedical Institute, Federal Aviation Administration prepared for the O.S. Department of Transportation, Report No. DMTA-MA-06-0025-79-2, II. Rilado, C.J. and Brauer, D.P. (1979). Concentration-time data in toxicity tests and resulting relationships, J. Combust. Toxicol. 6: 136-149. Rilado, C.J., Lopes, N.T., and Damant, G.R. (1977). Relative toxicity of pyrolysis products from some upholstery fabrics, J. Coated ^ Fabrics 6: 155-175. rRishitani, R. and Tusa, S. (1979). Study of evaluation of relative feoxicities of combustion products of various materials, J. ' ' Faculty Engineer., D. of Tokyo XXXV: 1-17. LeMoan, G. and Chaigneau, M. (1977). Toxicite des produits de combustion des natieres plastigues. III. Methods rapide d'evaluation par determination de la DLxg chez la Souris, Annales Pharmaceutiques Francaise, 461*464. Levin, B.C. and Birky, N.M. (1981). An interlaboratory evaluation for the Rational Bureau of Standards test method for measuring . v ~ eonbustion products in rats. The Toxicologist 1: 8. ``Levins B.C, et al. Further Development of a Test Method for the .`Assessment of the Acute Inhalation Toxicity of Combustion Froduets. NBSXR 82--2532. U.S. Dept, of Commerce, National Bureau of Standards, Center for Fire Research, Washington, DC ^ 20234, June, 1982. Levin, B.C. Workshop on Combustion Product Toxicity, National Bureau of Standards. September 10, 1982. C - 14 - REFERENCES CONT. Spurgeon, J.C. (1978). The correlation of animal response data with yields of selected thermal decomposition products for typical aircraft interior materials. National Aviation Facilities Experimental Center, Federal Aviation Administration, prepared for the O.S. Department of Transportation, Report No. FAA-RD-78-131. . .? ..... , , . .. ^ -.Y- Rikria, .A.,~Ferrer, J.M., and Flock, H.F. (1972)'. The chemical factors contributing to pulmonary damage in "smoke poisoning", Surgery 71: 704-709. 0021s/bjl <*ac - ^7iw..wr . ^ ... A v't; - -* atyv.v ;r rV"""**. W--; -TL. IT*- r- r. r4 1