Document a1GOkzGeq021NNYNB6vVeX4ee

( By R.E. Reinke* and C.F. Reinhardt* Fires, toxicity, and plastics The only uniform features in the toxicology of burning plastics are those of the toxicology of fire: heat, carbon monoxide, deficiency of oxygen, other combustion gases, and smoke. The hazard from combustion or thermal decomposition of plastics should be compared under equivalent conditions with that of wood or other appropriate materials. This comparison provides a basis for present assessment of hazard pending a more standardized evaluation anticipated in the future. A technical feature Gordon M. Kline, technical editor In considering the toxicity of the combustion products of plastics, we must consider the toxicology of fire, since this is common to all fires, regardless of the material being consumed. In any real fire situation, the toxicology of fire assumes primary importance. Broadly speaking, toxicity refers to the potential for harm arising from the action of specific chemicals or other agents on living tissue. The major causes of death or incapacitation in fires have been identified by several investigators (1-3)'. In a given fire, more than one or even all of these factors may be in volved: I) heat, by destruction of tissue (direct burns), or thermal shock; 2) carbon monoxide; 3) deficiency of oxy gen; 4) other gases; 5) smoke; 6) panic or emotional shock (indirect). This last factor can be more serious than the ac tual fire, as in a recent evacuation where 31 people were injured, most of them from landing on top of one another (4). The physiologic stress of panic or shock can also ag gravate any injuries from other causes. However, panic or fear is not an intrinsic part of the fire situation, thus it is not primarily involved in the toxicology of fire. Direct burns. Heat that destroys tissue is obvious as a cause of immediate death. Burns Of ovet.50% of the body surface are frequently fatal, especially in children and older people (5), and deep burns of only 10% of the bodycan be disabling (6). However, "it is generally recognized . . .that many, if not a majority, of those who succumb in building fires are not seriously burned and that death or trauma must be due to other factors"(3). Thermal shock. Heat disseminated beyond the area of the actual flames is considered the primary cause of death in well-ventilated fires where there is a good supply of ox ygen. Combustion in excess oxygen generates carbon diox ide (C02), water, and more heal, which in turn promotes more combustion. The amount of heat that develops dur ing a fire is a function of the caloric or heat-yielding value * Du Pom Co.. Djskcll Laboratory for Toxicology and Industrial Medicine. Wilmington. Del. fljsed on a paper presented at a meeting of the National L ire Protection Assoc.. May 1972. I -Numbers in parentheses designate references at end of article. of a given substance, the amount of a combustible mate rial. and the rapidity with which the reaction lakes place (3). Petroleum products, lor example, have a high caloricvalue that can readily result in a lethal generation of heat. Wood has a relatively high level of combustion but is lowin comparison with kerosene or coal. Temperatures in ordinary building tires can quickly climb beyond the usual tolerable limit of 150 to 160 F. In a series of experimental fires started on the first floor, the heat in a second floor bedroom was intolerable within about 2 min. when the bedroom door was open and within 9 to 12 min. when the door was closed (3). In high-rise buildings, where a self-contained tire can reach massive proportions, fires have been reported to reach tempera tures of 1500' F. (7). Available protective equipment is grossly inadequate under such circumstances, and heat is perhaps the most uncontrollable factoFin fires today. Carbon monoxide (CO). In poorly ventilated fires, where the supply of oxygen is not plentiful, combustion is incom plete and the end products are CO, water, and less heat. Also, other toxic gases can be formed depending on the nature of the material being burned. Of all the gases gen erated from natural or synthetic materials, CO produces the most deaths in real fire situations (1). CO is insidiously toxic: it is a colorless, odorless gas primarily absorbed through the respiratory tract, although absorption through uncovered skin has been demonstrated experimentallv. In- Table I:'Physiological response to carbon monoxide CO in atm., % 0.01 0.04-0.05 0.06-0.07 0.1-0.12 0.15-0.2 0.4 1 Response Allowable exposure for several hours No appreciable effect after 1 hr. Just appreciable effect after 1 hr. Unpleasant after 1 hr. Dangerous when inhaled for 1 hr. Fatal when inhaled for less than 1 hr. Fatal when inhaled for 1 min. 4. Table II: Physiological effects of reduced atmospheric oxygen levels 0] In atm., % 21 17 12 9 6 Response None Impaired muscular coordination; increased respiratory rate Dizziness, headache, rapid fatigue Unconsciousness Death in 6 to 8 min. BFG39063 ooooi Table III: Physiolot/ effects of combustion gases (exclusive of CO) Combustion gases' Highlights of toxic hazard halation off). IT concentrations usually produce-- .1 warn in': dizziness or headache, hut IT concentration-. do not produce anv sensors warning before escape is loo late (S). The general response that can he expected from specific atmospheric levels of CO is shown in Table I (8. 9). Physiological response is influenced hv such factors as rate of breathing and individual susceptibility. The rate of breathing will obviously be increased in strenuous phvsical exertion. Age. health, smoking habits, and previous expo sure to CO all have some influence on the total effect pro duced bv any specific concentration of CO. When inhaled. CO combines with the hemoglobin of the blood to form carboxs hemoglobin rather than the ox yhemoglobin which transports oxygen to the tissues. Addi tionally. CO may decrease oxygen delivery to the tissues and hinder remov al of CO- from the blood stream (8). Deficiency foxygen (OThis situation occurs as a nat ural result of combustion. A summarv of the physiologic effects that may be expected at various atmospheric Olevels is given in Table II. Although atmospheric 0> con centration dips of 77c or less have been recorded in various experimental poorly-ventilated fires (8. 10), a 15fc O- level is usually required to support combustion of most bumable materials. Thus, it is possible that man couhd survive where fire could not, if 0-. deficiency by itself were the only serious factor in building fires. However, in a fire situ ation, as the 0> supply decreases, CO will increase and the O-i supply in the blood can then be under a double strain. In the presence of these combined effects and heat, smoke, and other combustion gases. 0-> deficiency mav produce serious and possibly irreversible effects. Other combustion gases. C02 and many other organic and inorganic materials also result from fires (Table III). Smoke. Smoke is the airborne mixture of heated gases, liquid droplets, and solid particles evolved from combus tion. It includes acids, alcohols, aldehydes, and miscel laneous hydrocarbons. Smoke from wood and fat contains the irritant aldehyde acrolein. In some cases, organic acids and aldehydes that are liquid at ordinary temperatures ap pear to condense on the surface of the smoke particles and cause more irritation than the actual particles, which are mostly carbon. The heat in smoke, which can cause tissue injury in itself (II). may enhance the toxic effects of any of the complex products contained in smoke. Smoke can also reduce vision and lead to panic. Smoke obscuration is a definite hazard in itself since exits cannot be located. Relative smoke obscuration time must be con sidered in (he selection of building materials. Plastics Chemically, plastics consist of chain-like substances called polymers, as illustrated below: Carbon dioxide (CO;) Produced in all lires Nitrogen compounds Nitrogen dioxide (NO,) and other oxides of nitrogen Produced in small quantities from fabrics and in large quantities from cellulose nitrate and Celluloid (prepared from cellulose nitrate and camphor). Decreasing use of Celluloid in consumer products today Ammonia (NH,) Produced in combustion of wool, silk, nylon, and melamine. Concentrations generally low in ordinary building fires Hydrogen cyanide (HCN) From combustion of various products such as wool, silk, polyacrylonitrile, nylon, polyurethane, and paper, in varying amounts Halides Hydrogen chloride (HCI) From pyrolysis of some wire insulation materials such as polyvinyl chloride (PVC), also chlorinated acrylics and retardant-treated materials Phosgene (C06l,) From pyrolysis of carbon tetrachloride used as fire extinguisher and, in small amounts, from other chlorinated materials ' Fluorine and bromine compounds From combustion of fluorinated resins or films and some fire-retardant materials Sulfur compounds From pyrolysis of certain high-temperature mechanical parts, rubber, wool, upholstery materials, and some natural polymers. Sulfur compounds are usually not abundant in total combustion sources present at a given site in building fires Sulfur dioxide (S0t) Toxic by itself only in high concentrations, but stimulates respiration; 3% concentration doubles lung ventilation Strong pulmonary irritant capable of causing immediate death as well as delayed injury 'Characteristic odor and unbearable irritating effects at sublethal concentrations would probably give adequate evidence of its presence' (8) A rapidly fatal but flammable asphyxiant poison. No known toxicity in real fire situations Over-ail, 'the HCI generated in fires is probably not a serious threat to life unless one is unable to move from the area of generation' although in one reported instance fire injuries were attributed to exposure to HCI (8) Deaths reported in connection with fire use of carbon tetrachloride. No known effects from small amounts generated from combustion of other sources `In most cases, warning of the presence of a toxic halide would be given by the sharp odor and noticeable irritating action on the respiratory tract' (8) A strong irritant intolerable .well below lethal levels - Polyethylene [CH,-CH,),, Polystyrene fCH-CH} Cellulose (natural polymer--more than !>0% of cotton, 50% of dried wood) fCoHioOi). Hydrogen sulfide (H,S) An irritant capable of causing systemic poisoning. Flammable. Characteristic rotten-egg odor at very low concentrations.' olfactory fatigue possible at higher concentrations ' C.Hj where n = the number of units in the chain. The significance of n is more easily visualized when the a--Alt these gases can be lethal In sufficient concentration. However, in most fire situations, combustion gases except for carbon monoxide would be expected to act as a contributory rather than a primary cause of death or Incapacitation. BFG39064 25200002 r (r chain i> extended, as shown tn the ease of polyethylene: HHHH HHHH !!1I I I II C-C-C-C-............... -C-C-C-C IIII HHHH I IIII HHHH Several hundred or thousand units Combustion of polymers causes the molecular chain to break and decompose into oxidized fragments as illus trated in Table IV. In real lire situations, the lire often in cludes a combination of well-ventilated and poorly-venti lated areas of different temperatures, and the combustion products vary accordingly. Although the toxicology of polymer combustion products is extremely complex, one practical approach is based on experimental animal data, which include comparable testing of natural and synthetic polymers. Data from several of these studies (12-14) are tabulated in Table V. From a flammability standpoint, the most hazardous plastic is cellulose nitrate (15). which thermally degrades into oxides of nitrogen and carbon monoxide. In the case of Celluloid, another toxic factor may be the camphor plasticizer. Consumer use of Celluloid is generally regu lated today in recognition of the Are hazard. Analytical determinations of pyrolysis products provide another means of comparing the combustion hazards from synthetic and natural polymers. An understanding of the toxic mechanism involved can be achieved when these data are coordinated with biological data. Also, it may be possible to assess the hazard from a particular pyrolysis product in terms of both familiar and new parent poly mers. Table VI lists the results obtained when 20-mg. sam ples of various materials were pyrolyzed in a liter flask containing air or nitrogen and the resultant products analyzed for HCN (16). The authors comment that "apparently polyurethane foam does not produce more hydrogen cyanide than do many other natural and syn thetic consumer products. . .(and) that it is difficult to un derstand why cotton and paper produce HCN on pyrolysis unless nitrogen fixation is assumed. If this be the case, apparently some oxidation is necessary for nitrogen fix ation, since less cyanide is produced in atmospheres con sisting originally entirely of nitrogen." In another investigation, experiments were conducted with the combustion products generated from the burning of various building materials and also with varying combi nations of 02, CO, and C02 at different temperature levels (17). When mice were exposed in this experiment, the in- Table IV: Polymer fires Type of fire Well ventilated Poorly ventilated Combustion products Heat 4- carbon dioxide -f water 4- varying complex of oxidized fragments Carbon monoxide 4 heat 4 water 4 varying complex of semi-oxidized fragments Table V: Toxicity results from thermodegradation products of polymers Part 1: Series of 15-min. exposures using 3 g. of foamed materials. 30 g. of all other materials. Pyrolysis temperatures up to 1364" F. (12). Material Mortality for mice* Polystyrene rigid foam (A or B) Phenolic rigid foam Wood-wool cement board Acrylic rigid sheet Wood (cedar) Fire-retardant plywood Melamine laminate Polyvinyl chloride rigid sheet Polyurethane rigid foam 0/10 0/5 1/5 4/5 5/5 5/5 5/5 5/5 5/5 Part 2: Series of exposures (13): No. 1. 6-hr. exposure, 4.7 to 5.5-g. sample. No. 2. 6-hr. exposure, 5.7 to 6.8-g. sample. No. 3.10-min. exposure, 2.0-g. sample.__ Mortality for rats* No. 1, 392* F. No. 2, 482* F. No. 3, 1040* F. Polyurethane A 0/4 1/4 0/2 Polyurethane B 0/4 1/4 0/2 Polyurethane C 0/4 0/4 0/2 Neoprene 0/4 1/4 0/2 Rubber latex 0/4 4/4 0/2 Polyvinyl chloride 2/4 1/2 Part 3: Series of 30-min. exposures, 5-g. samples (14). Mortality for rats* 1 Polystyrene A Polystyrene B or C 4 other PS Expanded cork Rubber Wool Pine wood Felt Leather 572* F. 0/12 0/24 0/48 0/12 0/12 2/12 3/12 6/12 12/12 752* F. 0/12 0/24 0/48 5/18 12/12 12/12 12/12 12/12 12/12 932* F. 0/18 25/42 48/48 12/12 12/12 12/12 12/12 12/12 17/18 1112* F. 11/12 24/24 48/48 12/12 12/12 12/12 12/12 12/12 11/12 Part 4: Series of 30-min. exposures, 5-g. samples (14). Mortality for rats* 392* F. 572* F. 752* F. 932* F. 1112* F. Polyethylene Fir PVC Celluloid 0/12 0/12 0/12 12/12 0/12 13/18 10/12 12/12 12/12 12/12 11/12 12/12 12/12 12/12 12/12 12/12 12/12 12/12 12/12 12/12 a--Mice or rata exposed to products resulting from pyrolysis temperatures indicated. Figures in table show ratio of number of mortalities to number exposed. All temperatures have been converted to *F. BFG39065 V/m if! B9 vs C00002Q2 dicuSn'M xx.is in iikiv t'O in iiu.t-.iII tu\Kil\ lunil (he 10111- I'11'i 10n pi. 'Jik 1 - u'll' p.iktI 10 (Ik- i/uinh 1 n.ii 10ik oi the pre- xnuislv studied v jn.ihk's 1 ( ) . CO. ( ().. .nul hc.ill. \ r,um bet oi qu.dii.iiiu. .iikI qii.intu.iin e .111.1 lx m."> were June in order to deline .md control exposure eonditionv and eeleeted d.tl.i .ire 'how 11 m Table VII. To he re.ih'iie. the toxicologx id' polluter eombustion niN Re I'.i-ed .11 hioioeie.il d.11.1 obtained from eomhus110:1 of the .leti 11I poKmer. Extrapolation of data generoed elsewhere max result in either .111 under- or over estimate of i..\i._ poieiTi d. In .1 real lire situation. the to\ie potent;.u 0 ilie sum. either additive or s\nergtstic. of all ihe xario-j. eatt-e' of death or incapacitation that max be present Thermal degradation of polxmers probably dates from the lir't oad lire xxhen obvious benefit' overshadowed potential harm. The toxicologx of thermal degradation de veloped siovxlv. hul recent vears. particularly the last de cade. hav e seen an increasing awareness of the need for re search support, generation of valid data, and a meaningful estimate of toxicity 1 18-20). The concept that no one test is adequate to characterize the fire hazard of a given material has now been recognized. Unfortunately, standardized tests that include biological evaluation of combustion Table VI: Hydrogen cyanide (HCN) measured in pyrolysis products Material Paper Cotton Wool Nylon Polyurethane foam HCN, ng./g. sample Air Nitrogen 1100 93; 130 6500 780 1200 182 85 5900 280 134 Table VII: Analysis of combustion products generated by burning building materials3 Gas component, parts/million Materials CO CO, HCN NO, COCI, HCI Red oak Douglas fir Red oak Douglas fir Douglas fir Douglas fir Douglas fir PVC rigid pipe Vinyl asbestos lile 13,600 9,400 11,000 -- -- 1,850 1,500 900 93,000 56.000 100,000 9,000 -- 20,000 12.000 10.000 35 10 30 Trace 55 T race 10 0.5 Trace 50 20 10 0 4 0.5 0 0 0 0 0 a -- Materials were burned m nn AS^ M E-84 tunnel furnace under varying conditions of ventilation. Samples from ihe combustion pro ducts stream were Alincollected and then analyzed by colorimetric tube S, or nfrared analwers |hnwhen CO or CO* levels were beyond range of the available indicator tubes 0 0 0 0 0 100 0 produel' .1 re itoi .0. .nl.iMe .11 p' cscut. .ill hone li .1 11111 In Hide ol test' exist 1 - I I I't'oi edin .' nul 1 uteri. 1 ll.r. e bur, pro p. wed I I'I - 3! i I -:'u re w 011. . - it .i.uul.ird test' mu - I iik I tide euilu.itioii. lorni.d .ieeej'1 iiKe. .uni pr.ieiie.il evviieine References I) .1. Autian. .1 l ire and Mammabihtv /. 2-" (Julv 2l I.V I.inborn. "An overview .-f the lUimmaniiitv characteristic'of materials." i'l 1'olv tilers and Materials Conference Series. College of l.ngr.. ( nix. of l tali. Sail Lake Citv. 3). PAY Bieberdorf and C II 'I'uill. "An invesiigation of the hazards of combustion products in building tires.'' SwRI Project No. 3-12.'"'-.-. final Report. I. SPITS Con tract No. PH86-62-208. Southwest Research Institute. San Antonio. Tex t 1963). 4) Fire J z>-/. Nn 6. 4s t 19 oi. 5) M.A. Krupp. M.J. Chatton. S Margen. et al . Cur rent diagnosis and treatment." I.anee Medical Los Alios. Calif. (1971). 6) P. Reich and VAV. Sidel. New England J. Med. 2" No. 2. 86(1967). 7) W.R. Powers. Fire J. 65. No. I. 18 (1971) 8) J.M. Ives. E.E. Hughes, and J.K. Tavlor. "Toxic atmospheres associated with real lire situations." NITS Re port 10807. U.S. Dept. Comm.. Washington. D C. (1972). 9) F.A. Patlv. ed.. "Industrial hvgiene and toxicology." 2nd ed.. Vol. li. Wiley. New York (1963). 10) J.A. Znpp. Jr.. "The toxicology of lire." Mevlical Di vision Special Report No. 4. Chemical Corps. Chemical Center. Md.. PBI43732 <19511 11) B.A. Zikria. "Inhalation injuries in fires." Sym posium on haloiuenated lire extinguishing agents. Apr. 1972. at NAS-NRC. Washington. D.C. *12) K. Kishitani. J l ac. Filer.. I'niv. of Tokxo < B) J/. I (1971). 13) J.A. Zapp. Jr.. Arch. Enxiron. Health 4. No 3. 335 ( 1962). 14) H.Th. Holfmann and H. Oettel. Modern Plasties 46. 94 (Oct. 1969). 15) C.J. Hilado. "PTammahilitx handbook for plasties." Technomic. Stamford. Conn, t 1969). 16) W.H. McDermott and F E. Critchtield. "Thermal degradation of polvuretlutne tv> hvdrogen exanide." I 111011 Carbide Plasties Co. ( 19611: cited in R.E. Dufour. "Survey'' of available infomiation on the toxicitv of the combustion and thermal decomposition products of certain building materials under lire conditions." Bull. Res. No. 53. Under writers Laboratories. Chicago ( 1963). 17) A.J I'rxor. I) I Johnson, anil N.N. Jackson. ards of smoke and toxic gases produced in iirh.m "Haz tires." ft SwRI Project No.- (13-24(12. Fmal Report. CK I) i ontract C/1 No. DAI i( 2it-7(l-( -0212. Southwest Research Institute. ft San Antonio, lex . Al)697s34( |96'>). O IS) R |.. Detour, rctcieiuc i(> 19) ( IF 3 mil. Fire J No. 3.47| I'UJi o o 20) J Autian IomcoIo-.w coiisiileratiotis ol 111 a 11 made 1^5 m.iieimis vine n tires " st't ''"lb -\ntec. M.iv "U2. ( hi- eae'o. 21 I S Stems: iscr. .1. I rv .nul I l.imin.il'uhl * 1 I1'?.1) I \I> Kcprmi cd 1 n in 1I11.` 1 ci>l (1,11 v 197 1 Is uc nl Mi; 'dern 1 ':i *Wc 1 1 If A nicn M'x- \1 / \ hv x (.f '< 11 ill I n BFG39066