Document LKnRmp14jGwOwgB2Vyk9bR9kw

L H&tlTji Amer. hi v.-* 335 Immufk*. Med. 1 i3W. W. D,, 4,yJ Toxicity: i id Tolerance Health U: >f Tolerance na, Amer. j, Toxic and Health Effects of Plastics and Resins JOHN A. ZAPP, JR. WILMINGTON, DEL. o< it IiV Wagner, iW, H. E.: ; e Following sl.,ai 138th *1. Society, Twenty-five years ago a discussion of materials. For another, their proper spheres Ac toxic and health effects of plastics and of usefulness were better recognized, so doer Vor- resins would have been relatively easy on that die right plastic was selected to fill a wjtratiooen Aj-ch. Exp. 0. J, and >f Ch-HtO. 2:21, 1961. "V,D., and * 2 counts. There were relatively few syn thetic plastics and resins in common use, and such uses as were common were relatively trivial in comparison`with competitive ma terials. World War II gave great impetus to the given need. is difficult to define plastics and resins in a very meaningful way. Chemically, they are very large molecules, called polymers, formed by the linking up of small molecules, called monomers, into large cohesive chain- r. Studies: Following h-Iiidustr. *e of synthetic plastics and resins. On the like units. If only 1 monomer is involved We hand they were able to replace natural in forming the polymer, it is called a homo* Btcrials, particularly metals and natural polymer. If 2 different monomers are in tie Acute tfitrogep m BedHazard* ng Shop, whber, which were often in short supply: On the other hand synthetic materials, e.g., ylon, turned out to be superior to available ataual materials for certain uses. Following World War II, plastics in evitably assumed a greater role in peacetone applications. Many of the plastics were M suited to these uses, and things made f plastic were often looked down upon as Pr substitutes for the genuine article. This rituation, however, soon worked itself out. For one thing, there were important de*ripments in technology, and the synthetic Pgstics and resins actually became better ^hmitted for publication Nov. 10, 1961. j/Twetrted at 26th Animal Meeting of Industrial Foundation, Pittsburgh, Oct 25-26, 1961. volved, the polymer is called a copolymer; if 3, a terpdymer, etc. Polymers can be classified in several other ways. If the monomers simply link up into long chains by joining bonds, and nothing is eliminated in the process, the polymer is called an addition polymer. An example is shown in Figure 1. If the linking up of monomers is ac complished by the elimination of part of the monomeric molecules, the resulting polymer is called a condensation polymer. An example is shown in Figure 2. If the long-chain polymer molecules are not joined laterally to one another, the plastic or resin is usually flexible, because Haskell Laboratory for Toxicology and Indus- the molecular chains slide against each other ""H Medicine, E. I, du Pont de Nemours & ComMoy. when a deforming .force is applied. This f.\. flexibility can be increased by mixing in i j?.. h, 1962 ,tbrl*o poly*tbylao* Rg- 1.--Addition polymerization. n (io*Jt*COOHj * Hjon*cojn 01 (n-DB^ hydroxy sold polyeator Fig. 2.--foodensatk polymerization. 125 - *$ -v-. Ji.-O-Rv ucc 041087 '75 " ' 'A *5 338 Tails 1.--Thermoplastic, Polymers ARCHIVES OF ENVIROHMEHTAL HE.lLTff* Tails 2.--Thermosetting Polymers Polyethylene Polystyrene Polyecryletes PolymethacryUtes Polyvinyl chloride Polytmld* (nylon) TFE fluorocarbon resins Saturated polyesters ("Dacro"> with the polymer molecules an internal lubricant, known as a plasticizer. It is characteristic of polymers that they soften when exposed to heat and when soft can be made to flow and assume desired shapes. When cooled, djey again become hard. Some polymers, if reheated, become soft again; these are the thermoplastic resins. Other polymers, when heated for the first time, undergo further chemical reactions in which cross links develop be tween polymer chains, holding them rigid in the position assumed as a result of the first heating. These are the thermosetting resins, and they do not soften on reheating like the original polymer. Examples of thermoplastic and thermosetting resins are shown in Tables 1 and 2. When we say that polymers are chains of large molecular weight, no particular degree of largeness is specified. If only a few molecules are linked together, the re sultant molecule is called an oligomer. So, a polymer is larger than an oligomer in that it contains many rather than a few monomer units. It is also apparent that a polymer may be composed of polymer chains of varying length. "When a monomer polymerizes, a chain builds up in length until something terminates the process. Termination may occur sooner in same chains than in others. If the process is random, the final polymer will be a statistical distribution of molecular chains of varying size, and the molecular weight of the resultant mixture can only be expressed in terms of average and stand ard deviation, of normality or skewness of distribution, etc. In general, the average molecular weight of the polymer will be of the order of thousands or millions. Phenol formaldehyde resins Amine formaldehyde resins Epoxy resins Alkyd resins Unsatnreted polyester resins It has frequently been stated of the plastics and resins that the molecules are so big and so chemically inert that they are also physiologically inert. If ingested, tl*y are too big to be absorbed from the gastn>. intestinal tract and hence pass through unchanged. If placed in contact with the skin, they are too big to penetrate or react with the skin. If inhaled as dust, they behave as typically "inert" dusts dojn die lung. These statements are in all probability true for most of the molecules comprising the polymer. The exceptions relate in general to the oligomers, and in somX cases evm to the monomers, which make up a fraction of the total polymer molecules. For example, polyethylene manufactured by the high-pressure process came into me as a container for foods about 1946. Tl* average molecular weight of the paijirer usually ranged from 5,000 to about 40,000. ; Its structure is essentially that of a loqf-` j chain aliphatic hydrocarbon. It is insoluble in almost all solvents at room temperatm. although it swells in contact with hydl*-. carbon solvents like hexane or heptane ] benzene, lubricating oils, and the like. TW swelling is due to the penetration of fit polymer chains by the solvents. It is wonder that the U.S. Food and Dor Administration in 1951 listed polyethylenea a suitable resin for use as a food padogqf film1 on the basis of its structure wA physical properties. By the summer of 1958, however, .k' Food and Drug Administration learned A* commercial polyethylene contained a h* molecular weight fraction that might bro** a component of a fatty food contacting ** polyethylene. When the Food Add*** ,, Amendment of 1958 became effective. * Food and Drug Administration coofct*** the prior sanction status of polyethytor 120 Vol. 4. Mer.-h J* 1 x'f of ucc 041088 ted of the oleculcs hat they ate gested, they the gustmiss through ct^with the at^or react they behave a: the lung. )Ut|lity true tjpiing the m^gnteral cases even up a small nolectilcs. aaufactured ne into use 1946. The ue polymer out 40,000. of a long- insoluble iperature. nth hydror heptane. s like. The ion of 6* .. It is_ no and Drug ethylene a* packaging icture and wever, the ,-arned that ied a low rht become tacting the Additives ective, the confirmed thylene as Vttrch, 1962 CITY OF PLASTICS AND RESINS * dwag (ch2)u . eg *f*Uoo-o*prolot HOPH (CHz)5 eg nOH + (n-D HgO Type 6 Hylon pig. 3.--Type 6 nylon. polyepsiloncaprolaetani. ,jVl packaging resin only with respect use on nonfatty foods. It permitted its Minuet! use in contact with fatty foods wty on an extension-of-time basis pending Mhcr study and evaluation of the data. deliberately supplied in an incompletely polymerized state in order that the user might complete the process at the site of application. Examples of this would be the acrylic denture materials and the poly The sequel to this is that all 13 producers urethane insulating materials or lacquers. J polyethylene in the United States In the former application, molding powder mtmbled data on the chemical identity of composed of polymethyl methacrylate is it low molecular weight fraction in their mixed with liquid monomer, an activator, ttficus polyethylenes and on the amount and color. The monomer acts both as a tftactible by fatty foods. This information plasticizer for the polymer and as a bonding Stilled the Food and Drug Administration agent as it itself is polymerized. Since' establish specifications for polyethylenes methyl methacrylate monomer is a skin Mitabie for contact with fatty as well as sensitizer, a number of dentists and dental wmfatty foods, and a regulation siting technicians have experienced dermatitis^ f*th the specifications was issued in June. problems. The resulting polymerized denture MSI. material, however, has rarely caused any This example merely illustrates the fact fat polymer molecules are not all of the ***** large size and that some are small ough to be diffusible from the plastic or **w>. As with polyethylene, the low "Ocular weight fraction is not necessarily lunnful, but its safety must be evaluated. trouble for the patients. Polyurethane resins are likewise often supplied in "prepolymer" form for final polymerization at the site of application.- A typical application is that of the polyurethane foamed insulation materials.4 Figure 4 shows how a typical polyurethane Some plastics and resins may actually prepolymer (I) is formed by the reaction, in wam unreacted monomer. Type 6 nylon, the absence of water, of a diisocyanate **>>ch is polyepsiloncaprolactam, is an (usually toluene diisocyanate, or TDI) and ***mple (FTg. 3), The epsilon caprolactam a glycol. The glycol now terminates with lnQn0!ner is water soluble, diffusible, and a isocyanate groups (-NCO) and has acquired 'reak skin sensitizer. It can be removed urethane groups (-RNH COOR'-) in the "*** the polymer by washing with hot process. The prepolymer contains a slight **,er* Phenol formaldehyde resins as well excess of TDI. te other formaldehyde resins sometimes If water and a catalyst are added to the fatain free formaldehyde in sufficient con prepolymer, a further reaction takes place Jfa*ation to be detectable by odor. The free as shown in Figure 5. This reaction is ****aldehyde can be kept to safe limits, exothermic. As polymerization takes place "#WtVer, by adequate curing of the resin. It might be well to note at this point that 20SUrc to monomers and low molecular Jw Polymers is much more apt to occur Jrlnng the manufacture of these materials the CQ* evolved ``foams'' the mixture which grows more viscous and rigid under the combined influence of heat and increasing molecular weight and traps the COz, thus giving the final cured foam plastic In the 4,1 during their subsequent use. Thus the lecturer is more liable, to toxicity and 2 0CH-R-NC0 + HO-R'-OH ^*lth problems from monomers than is the ~er f the finished plastic or resin. In some ^Sfs> however, the plastic or resin is OCN-R-MB*CO- 0 -R'-OCO-KH-R-MCO (I) Fig'. 4.--Reactic*! of diisocyanate with polyol. W ucc 041089 i. /v? m & 338 ARCHIVES OF ENVIRONMENTAL HEALTH' z she * * m (i> * %<> --* sensitized, and the safe upper limit was accordingly, revised downward to 0.02 pp^ S3 XBCOXH - - JS2 (IS) * C02t Whether the assumption is valid and the Fig. 5.--Polymerization of polyisocyanate resin. revision justified, I do not know. So far as I have been able to learn, there is no proof process, small amounts of TDI monomer that sensitized workers did not occasionally may be volatilized. receive gross overexposures prior to be Toluene diisocyanate is not a highly toxic coming sensitized. Only time will tell material, but it is extremely irritating to the whether 0.02 ppm will prove to be a mote respiratory tract and may cause asthma-like correct estimate of the safe level than attacks which begin several hours after 0.1 ppm. exposure and dear up without after-effects The important point, however, is that one in a few mor^hours. Repeated exposure to TDI vapors may produce an allergic sensitization of the respiratory tract in should avoid, to the greatest extent possible, exposure to chemicals which are capable of causing allergic sensitization. It is a tricky susceptible individuals. business, for example, to find out just how Since polyurethane resins are being much poison ivy you can pull up with yonformed in {dace on a rather large scale, bare hands before you become sensitized, and there has beeufen unusual amount of interest persons exposed tq^the monomeric diisocya in the definition of the safe upper limit of nates should take all indicated precaution concentration of the monomer in the air, to keep their exposure at a minimum. variously called the Threshold Limit Value, Hygienic Limit, or MAC. Largely based on work in Du Font's Haskell Laboratory,4 the American Conference of Governmental The finished cured polyurethane resins are not sensitizers, because any exposed isocyanate groups would react with water vapor in the air and be destroyed. Industrial Hygienists (ACGIH) in 1956 established a tentative Threshold Limit A second health and toxicity problem is Value of 0.1 ppm for toluene diisocyanate, connection with plastics and resins arises TDI, and this has prevailed until this year from the fact that the commercial materials when the ACGIH lowered the Threshold are seldom just collections of big molecules limit Value to 0.02 ppm. It may be of As was mentioned before, flexibility is current interest to comment briefly on the plastics is a reflection of the ability of respective numbers and the reasons for the individual polymer chains to slide again* change. each other in response to an applied fom Bearing in mind that TDI and( other The cohesive forces between polymer chain similar monomeric diisocyanates are more can be reduced by means of intend irritating than toxic and that they are capable of producing allergic sensitization, the safe level in the atmosphere should be that which lubricants called plasticizers. Usually a gmf plasticizer can be characterized as a ft* solvent for the polymer, which possesses to does not produce primary irritation or induce sensitization of exposed workers. Our volatility and hence is not lost quickly In* the plastic or resin. Sometimes plastic*** a're supplemented by nonvolatile Iiq*A estimate of such a level, based on animal exposures, was 0.1 ppm,, and Ehrlicher in which are not in themselves plasticizer* I* Germany * arrived independently at the same which can supplement the effect of ft* value. Some cases of respiratory sensitization , ticizers by acting as inert diluents. TV* have occurred, however, in some plants in are called softeners. which the TDI concentration was supposedly Plasticity can also be enhanced, hov'rw maintained at or below 0.1 ppm. It was in some polymers by copolymerizat***.' assumed, therefore, that susceptible workers the principal monomer with a small jr w- uv exposed to 0.1 ppm of TDI could become of a second monomer which, while ||v lit -- Vol. 4. *' ucc 041090 TAi HEALTH rjwUnjit/Was, lo 0.02. ppm, valid and the ow. So far as re is no proof t occasionally prior to beme will tell tp; be a more e/. level than ;r, is that one ctent possible, re capable of It Is a tricky out just how up witb your ensitized, and eric dtisocya* l precautions imum. lhane resins any exposed i-vvith water problem in -esins arises ial materials ^femolecules. ^Kbility in ; ability of tide against plied force, ytoer chains of internal tally a good as a foot ssesses low jiddy from plasticizers tile . liquids ticizers but zt of pi*5' nits. These 1, however, rization of all amount tie a fixed March, J962 M toxicity OF PLASTICS AND RESINS 339 part of the polymer chain, acts as an internal components rather than to the basic plasticizer. polymers. Unplasticized polyvinyl chloride, for example, is a rather rigid material and can be used for such applications as pipes for water or process chemicals. When plasticized by the addition of organic phthalate, sebacate, or phosphate esters, however, it becomes quite flexible and can be used for such things as raincoats, shower curtains, and automobile upholstery. Perhaps you will recall an early use of the plasticized copo lymer of vinyl chloride and vinyl acetate for transparent wrist watch bands, garters, and suspenders and will recall the dermatitis which was attributed to the use of dibutyl tin maleate and dibutyl sebacate as plasti cizers.* The example both points up the fact that nonpolymeric cordponents of a plastic can become health or toxicity problems and testifies to the skill with which such toxic components have since been carefully excluded from plastics which are likely to come into prolonged or repeated contact with human skin or with food products. Plasticizers are not the only materials added to plastics and resins to modify their properties. Chemicals are sometimes added to stabilize the polymer against decomposi tion by light or heat. Others may be added to retard oxidation by atmospheric oxygen. Dyes or pigments may be added to impart color. Perhaps this complexity of plastics and resins can be illustrated by reference to the Federal Register of Aug. 8. 1961, establishing a regulation on permissible mgredients for continuous resinous and Polymeric coatings to be used on metal substrates (i.e., can liners) in contact with foods. It amply indicates the fact that Plastics and resins may be more complex than just polymers and that we must con sider molecules other than polymers in evaluating health and toxicity problems eclating to such plastics and resins. Tt is fair to slate that such toxicity and health Problems which have arisen in the past from the use of plastics and resins have been overwhelmingly due to the nonpolvmeric There is another area, however, in which health problems have arisen in connection with plastics and resins, and this concerns not the plastics and resins as such but rather their decomposition products under the influence of heat. One of the earliest indications of hazard from this source came from the Cleveland Clinic disaster of 1929, in which stored nitrocellulose base x-ray film caught fire. Dense brown fumes spread through the hospital, and 125 persons were killed. Death appeared, in most cases, to have been caused by inhalation of carbon monoxide and nitrogen oxides,* both of which were thermal decomposition products of nitro cellulose. * Nitrocellulose is an outstandingly hazard ous material so far as combustion is con cerned, but it is characteristic of organic polymers that they will begin to decompose at some critical temperature and thereafter decompose at an increasing rate as the temperature is raised above the critical temperature. Some, like nitrocellulose, will bum freely if ignited; others, like Teflon TFE-fluorocarbon resin will not bum spon taneously but will nevertheless decompose if sufficient heat is applied externally. As plastics and resins find more extensive use as replacements for older materials, it is natural that there be increasing concern lest exposure to excessive heat or fire result in the production of unusually toxic atmos pheres. It will not be possible to com' the pyroly sis products of all plastics and resins in this discussion, but mention will be made of some general principles pertaining to. the toxicity of the products arising from any conflagration. Specific mention will be made of polyethylene, of the synthetic textile material Orion acrylic fiber, of certain foamed plastics and natural rubber, and of the Teflon TFE fluorocarbon resins. Conflagrations can be divided broadly into 2 classes, those which are well venti lated. add those which are poorly ventilated. Zatp i ucc 041091 m 340 ARCHIVES OF ENVIRONMENTAL HEAL7E& A well-ventilated conflagration is one in which there is free access of air to the fire. In this event the gaseous products of combustion are dissipated rapidly by the strong thermal currents, and injury, if it occurs, is almost always due to thermal effects alone. A poorly ventilated conflagration is one in which the access of air is restricted. In this case, there are several consequences. First, combustion does not go to completion. If the combustible material contains carbon, as it usually does, carbon monoxide will be formed as well as carbon dioxide. Second, there will be a depletion of*oxygen in the air in the vicinity of the fire. Third, there will be heat. And finally,'there will be special pyrolysis products characteristic of the substances being burned. If injury occurs, it will be due to heafc oxygen lack, and carbon monoxide as well as to the special pyrolysis products. It is my belief, which is based on World War II research with flame throwers * that the 3 factors heat, oxygen lack, and carbon monoxide are usually of more overriding importance as toxic factors than the more esoteric special products of combustion. Of course, there may be situations, as in the Cleveland Clinic disaster, where the products of a poorly ventilated combustion may be transported some distance by natural or forced ventilation, mid in these cases, heat ceases to be a significant lethal factor. This does not invalidate the generalization that gaseous products of combustion are of little importance as lethal factors in a wdlventilated conflagration and that carbon monoxide, oxygen lack, and heat are apt to be the most important lethal factors in a poorly ventilated conflagration, regardless of what material is bring burned. Polyethylene may be taken as a partic ularly simple example, in the chemical sense, of a plastic or resin; it is made up of chains of CH2 groups and is therefore a hydrocarbon. It resembles the paraffin waxes but is less of a heterogeneous mixture than the paraffin waxes. It bums slowly in an excess of air, producing CO2 and HaO as the end-products of combustion. If the burning were poorly ventilated, combustion would be incomplete, and CO would be produced as well as CO*. If polyethylene is heated in air to about 250C' it- melts, sublimes, and to a degree decomposes with out burning. Animals exposed to air passed over the heated polyethylene show signs of respiratory distress and may "die. The causative lethal agent or agents were not identified in experiments which lire curried out some years ago, but it was noted that the air containing the decomposition products of polyethylene would decolorize a dilute solution of potassium permanganate if bubbled through it. This would suggest the presence of unsaturated compounds like aldehydes, which are also formed, during the pyrolysis in air of natural hydrocarbons. In summary, polyethylene would prosed' roughly the same health and toxicity prob- ;1j lems on combustion or pyrolysis -as one would expect from paraffin wax or similar natural hydrocarbons. 1. - Orion polyacrylonitrile may be looked { upon as a polyethylene in which 1 of the 4 hydrogens is replaced by a nitrile group (Fig. 6). The presence of the multiple cyanide groups on the polymer, as well as the considerable toxicity of the monomer, 1 raised the question as to whether the com- '* bustion of Orion might not result in the release of dangerous quantities of HCN or of acrylonitrile. The Underwriters' Labora tories undertook an investigation of this point* In one type of experiment Orion fabrics were subjected to direct fame m % the presence of excess or deficient air. In TOi another in an furnace' 600 C. L products Similar 1 of cottor and wool The c Labocate fibers we When tt cess of air of thegaat decomporifc with small corded m deficiency, nets of th Scadde. u hydrocyamc When tfc . thermal <hx of nitrogm clode idrii ammonia. 1 . smaller an . shown m T - Both rid the Orion amounts o the pynoly atmosphen Underwrit The fixe I these acrySc rented by co fabrics of The life h decompcsiliui ties tinder fi to those pa cmnbtatton , woolen fahri Table 3.--P3 CH Acrylonitrile fi & H H C-C-C 1 H> or cb 130 Fig. 6,--Orton poly acrylonitrile. Vot. 4, March, 1962 hbOMUKA MnMiaiaB Nmwbm c Robber l*tec NytlajidMi Via-- Zapfi ucc 041092 '.HEALTH WmXlCITY OF PLASTICS-AND ftESINS 341 |js rhade^tip /therefore a affip waxes ixture than mrly" in an id HjO as m. If the combustion . would be ethyleneis :'it melts, >bs with" '. passed Signs of The re not j^harried 1. that ducts dilute ' ganate if dgjfcest the unds like id during rocarbons. Id-present cxy;: prob$,*s one jx similar jlooked % the 4 Ingroup, multiple s-well as nopbmer, the comltjln the HCNor x>ra- this' tfg6rlon dame in t;air. In another type, the Orion fabrics were heated # an atmosphere of nitrogen in an dectric * {umace which reached a peak temperature of tijOOC. In both types, gaseous decomposition products were collected and analyzed. Similar tests were carried out with fabric of cotton, silk, acetate and viscose rayon, ind wool. The conclusions of the Underwriters' laboratories with respect to Orion acrylic i fibers were as follows: When these acrylic fiber fabrics bum in an ex cess of air (oxygen excess), the chief constituents tl the gaseous products of combustion and thamal decomposition include carbon dioxide, together with small amounts of oocides of nitrogen as re coded in Table V. Under conditions ofmaxygen , deficiency, the combustion and decomposition pred icts of the acrylic fiber fabrics include carbon darkle, and small amounts of carbon monooride, hydrocyanic acid, ammonia, and oxides of njfrogen. When the acrylic fiber fabrics we subjected to thermal decomposition (pyrolysis) in atmospheres of nitrogen (air absent) the volatile products in clude relatively large amounts of hydrocyanic acid, "nnxwia, hydrogm and methane, together with faller amounts of unsaturated hydrocarbons as dw*m in Table VI. Both silk and wool behaved similarly to the Orion fabrics in that relatively large amounts of HCN were evolved only when *he pyrolysis was carried out in a nitrogen ahnosphere. The final conclusions of the Underwriters' Laboratories were as follows: The fire hazards of textile fabrics made from b^e acrylic fibers are in a class with those pre- by cotton, acetate rayon, and viscose rayon hhrics of similar weight and weave The life hazards of the combustion and thermal decomposition products of these acrylic fiber fab"o under fire conditions are judged to be similar 1 those presented by the fumes evolved during totubtistion or thermal decomposition of silk or *ten fabrics. Tails 3.- -Pyrolysis of Elastomeric Foams at 200 C for 6 Hours Foam Fjymathaaa A rwanthua r fynnithaw c toibbtrlutai rtotnyl cblutid* StmpU Wt, Qm, 4.07 4.03 4.07 3.43 5.03 4.70 % Wt. Loot U 4.1 6.1 3.5 Mortality 0/4 0/4 0/4 Oft 0/4 Z/4 This illustrates several things. The syn thetic fiber material Orion polyacrylonitrile, like other plastics and resins, does decompose under the influence of heat; its decomposi tion and combustion products are toxic. But the same statements apply to the natural polymers silk and wool and cotton, and the over-all hazard from the combustion and decomposition of Orion is judged to be similar- to that from silk or wool. The substitution, therefore, of a synthetic plastic or resin for a natural one does not neces sarily increase the hazard that already existed with respect to the natural materials. 2____f A similar question arose about the safety of various synthetic foam plastics as com pared with natural rubber. To answer this question we carried out experiments with foamed polyurethane, polyvinyl chloric^, neoprene, and natural rubber.4 The results are shown in Tables 3, 4, and 5. It can be seen from Table 3 that- polyvinyl chloride- foam was the least thermostable, in that it experienced 43% weight loss and killed 2 of 4 rats at an exposure tempera ture of 200 C Foamed polyurethane, neoprene, and natural rubber all decomposed to a certain extent at 250 C, and all produced some deaths in exposed animals (Table 4). At 560 C decomposition was essentially complete within 10 minutes except for neoprene (Table 5). Since the samples were smaller in this case than in the exposures at 200 C and 250 C, no deaths occurred except for 1 animal exposed to polyvinyl chloride. In all cases death was due to pulmonary congestion and edema regardless of the foamed resin responsible. Again it can be seen that certain synthetic foamed plastics and resins are similar to a Tablk 4.--Pyrolysis of Elastomeric Foams at 250 C for 6 Hours Foam Folyoiatbano A Polyunthano B Polyuathano C Kaopnm Bubbar laws Sampla Wt,.Oni. 6.85 6.S0 6.72 6.67 6.01 %wt. horn 40S .( 26.fi 46.4 160 Mortality 1/4 1ft 0/4 1/4 4/4 -ck, 1962 131 , . -- ' . & Is* -L'ikaii f- '* I* ucc 041093 342 ARCHIVES OF ENVIRONMENTAL NEAL! Table 5.--Pyrolysis of Elastomeric Foams at 560 C for 10 Minutes Foun Polyniothaat A PolynrcthtM B FotnJrethure C Nmchm Rubber lain PotjTloyl flhltfkk Stenpte wt. on. 2.00 2.00 2.00 2.00 2.00 2-00 ?< Wt. Um 100 97.3 100 73.4 93.3 *5.7 Mortality 0/2 0/2 0/1 0/2 0/2 1/2 foamed natural product with respect to hazard from thermal decomposition prod ucts. A history of safe use of the natural product would imply that a similar history of safe use could *be anticipated for the synthetic materials. One should not make the assumption that all synthetic plastics and resins will be as safe as any natural material they might replace. But neither ^hould one assume that the natural materials give harmless pyrolysis products because they have proved to be free from a practical hazard in actual use. In judging the safety of a synthetic resin for a proposed use, the hazard from combustion or thermal decomposition should be com pared under equivalent conditions with that of alternative materials which have, if possible, a history of similar use. Finally, it may be of interest to discuss briefly the hazards of combustion or thermal decomposition of the Teflon TFE fluoro carbon resins, both because the volume of our correspondence indicates a lively current interest in them and because of certain interesting features of the story. Table 6.--Acute Toxicity of Tetrdfluoroethyl^ Structure CF(=CF> AoM IobmUttoo Tortcity lor R, LC.a, ppm * MuOOO DaU from rittdlM at HaaAall Laboratory; 4-honr The Teflon TFE fluorocarbon resins are analogs of polyethylene in which all of the hydrogens of polyethylene are replaced by fluorine atoms. The monomer, tetrafluoroethylene, while more toxic than ethylene, is still quite low in. toxicity, the LC# being 40,000 ppm for 4-hour exposure of rats (Table 6). It polymerizes readily to poly, mers of very large molecular weight, which are characterized by them extreme chemical inertness and thermostability. Male and female weanling rats in our laboratory fed diets containing 25% of finely ground Teflon TFE resins for TO days showed no signs of toxic effects and no pathological changes detectable by gross or microscopic examination of the tissues. Teflon TFE resins are among the most thermostable of all the synthetic plashes and resins, being rated for continuous Use at 260 C. At this temperature the weight loss is from 0.0001% to 0.0006% per hour (Table 7), depending on the type resin. Since the TFE resins soften at 327 C, the melting point of lead, they are not likely to be serviceable for continuous use at or above this temperature, but they can stand brief exposures for short periods of time to temperatures as high as 540 C. The TFE f Table 7.--IVeight Loss and Ventilation Recommendations for Teflon Fabricated Resins Above 450 F lotttal WMght Loo. %/Hr Air I l On. Tt/Mln/Lb 1 Trtll,S, 7* TFKRMtn T*fl<m6,90TPtBMto Tattoo 100FXPBtfln Tofion 1,3,7TPBBiUn Tafton 6,90TFBBota Tridia FZFBahi OuOOOOSto 0-0001 lo ttJom 0.0003 04001 1 04)006 OjOOO* 04)013 1 04006 0.0043 0-0013 0.014 ' (70 0404 M0 04XB 435 003 0422 0.0$ 0.13 ' 03004 0401 0.01 0.02 046 04 0.12 0.21 0.06 1.2 4.2 9 18 223 040 14 34 10 23 72 160 223 OJt 11 SI 160 606 Note: Air raeomnMndsttoo* represent the Tolnma of ilr to redoc* fmaeodl product! to Imll loand nil thrmch Ion* eipfrlCTa wocteriag Teta nriai. 132 Vol. 4, March, M resins do but decor particulate sublimate, carried ou 95% of t monomer, out in thfc is still the at temper? amounts c to 5 C ate have been amounts c perfluorois this gas 380 C.H- Teflon ' flame, sin product^ v flame sour trill not cor ultimate co and HF. detected an products oJ favored by All of ff dude that 1 not present this has be experiment more than best of ou been killed decompositi the Teflon However exposure t health prob before the Sufficiently which whei condition Ji been variou "the sha resembles ir concerned, , "tent or aft *,r at most . Zap/, ucc 041094 'Sit health' W miCITY OF PLASTieS~AND RESINS `rofiuoroethyLn, mins do not liquefy on. exposure to heat "Toifcity tor Rti 7pm hu decompose by giving off gases and particulate matter commonly referred to as sublimate. If thermal decomposition is anted out in the absence of air, more than t: 4-hour expaean, n resins are ich all of the ^replaced by V tetrafluoroWi ethylene, e LC*o being sure of rats <% to poiy.'e^ht, which spe chemical and atory fed way' ground ^showed no pathological microscopic 95% of the decomposition products is the mooomer, tetrafluoroethylene.10 If carried ant in the presence of air, the monomer ji still the major decomposition product, but it temperatures from 200 C to 400 C small mounts of other fluorocarbon gases of 3 to 5 C atoms, HF, and silicon tetrafluoride love been detected. At about 400 C small amounts of a highly toxic C compound, perfhioroisobutylene, begin to appear, but this gas has not been* detected below 380 Teflon TFE resins can be ignited by flame, since the gaseot^ decomposition product?, will bum at 90 C, but once the flame source is removed, the Teflon itself will not continue to support combustion. The ultimate combustion products are COi, CF, >g the most die plastics tinuous use weight loss per hour resin. r327 C, the ot likely to at or above ;tand brief f time to The TFE and HF. No free fluorine gas has been detected among the pyrolysis or combustion products of Teflon, and its formation is nek favored by thermodynamic considerations.11 All of the above would lead one to condude that heated Teflon TFE resins should x* present an appreciable life hazard, and this has been borne out both fay laboratory experiment and by a history of use covering more than a quarter of a century. To the best of our knowledge, no one has ever been killed by exposure to the thermal decomposition or combustion products of w 450 F he Teflon resins. i/UJMn TVScnlOO- WBntn However, the Teflon resins do present on exposure to heat a unique toxicity and health problem which became evident even before the product became commercial. t Sufficiently hot Teflon gives off something *hich when inhaled produces in man a SI cndition like metal fume fever. It has Ml been variously called polymer fume fever11 US ** "the shakes." Like metal fume fever, it resembles influenza so far as symptoms are C0|>cemed, and it passes off without treat- ment or after-effects in a matter of hours '0r at most a day or two. In most cases its occurrence has followed the smoking of tobacco contaminated with dust or particles of Teflon. In the minority of cases it has followed dose proximity to freshly sintered Teflon articles as they have been removed from the sintering oven. In the ordinary course of events one would study polymer fume fever in the laboratory by produdng it in experimental animals breathing air drawn over Teflon resins heated to various temperatures. One would find the lowest temperature at which the syndrome occurred, and one would analyze the pyrolysis products at that^temperature and would identify by the process of elimi nation the causative agent One could then determine the threshold concentration of the agent needed to produce the polymer fume fever and could then estimate ^ Threshold Limit Value or MAC for man. Unfortu nately, experimental animals do not get polymer fume fever any more than they get metal fume fever, so this avenue of approach was dosed. On a priori grounds, one ctiuld assume that the likelihood of getting polymer fume fever would depend (1) on the temperature of the Teflon resin; because this determines the decomposition rate; (2) an die quantity of Teflon being heated, because a large quantity of Teflon at a given temperature might release as much of the causative agent as a smaller quantity of Teflon at a higher temperature, and (3) on the time duration of exposure, since this would determine the amount of contaminated air taken into the lungs. Since industrial processors of Teflon resins may handle many pounds per day, it was felt that a conservative recommenda tion would be that one should provide ventilation if Teflon were heated above the temperature at which pyrolysis is just detectable. This was estimated to be roughly 200 C or 400 F. At or below this tempera ture the quantity of Teflon handled and the exposure time should be irrelevant, since the Teflon simply wouldn't be giving off pyrolysis products. This, then, was the basis for Du Font-label warnings to provide ztP 133 1f . -- if i'"1 a * ucc 041095  : !> i-V [Sitl - 15 , if . t.? '3 ! ;F i 'K 344 ARCHIVES OF ENVIRONMENTAL h^ALTr' Table 8.--Mortality Ratios Resulting from the Toxicity Tests on Various Polymer-Insulated Wires Wlra Insulation Stttaona rubbv Nyian-polyTlnyl chloride Polyslnyl chloride jm-f Teflone Pyrolysis Temperature aooc 250 0 900 C 0/2 0/2 2/2 0/2 0/2 0/2 0/2 0/2 0/2 0,,/3 0/2 0/2 2/2 3/2 ventilation or respiratory protection if the Teflon were heated above 200 C or 400 F. It was directed to people handling large quantities of Teflon in industry. For such purposes the recommendation is adequate but perhaps overconservative, because die 2 conditions under which polymer fume fever is known to occur involve much higher temp^atures, 370-430 C in the sintering operations and perhaps up to 538 C in burning tobacco. On the other hand, the quantity of Teflon consumed in a burning cigarette was bound to be small and yet was capable of evoking the syndrome. In the intervening years little more has been learned about polymer fume fever. We now suspect that the causative agent has a very brief life time and that a short travel path from source to lung, ideally, as in smoking, is an important factor. We have obliged a dog to "smoke" repeatedly through a face mask cigarettes containing Up ^ 200 mg. of Teflon. It produced neither polymer fume fever nor any other observable harmful effect. We conclude from this that there is a wide margin of safety between the quantity of Teflon producing ^ polymer fume fever in man and that which might cause more drastic or lethal effects. We are inclined to suspect the particulate matter evolved during the pyrolysis of Teflon more than the gases, simply because metal fume fever is caused by a particulate It would undoubtedly be much more satisfying if we could both identify the causative factor of the polymer fume fever and assign an MAC value to it. Tie American* Conference of Governmental Industrial Hygienists has, in fact, made 2 attempts to supply at least a number, la 1960 they ^signed a tentative threshold limit value of 0.005 ppm to Teflon pyrolysis products without specifying what products. In the 1961 list the tentative value is OIK mg. per cubic meter "as F."' I know of as scientific basis, however, for either of these numbers. In. a laboratory situation one can, of course, kill animals by exposing them to the pyrolysis products of Teflon resins. Table S compares the toxicity of the pyrolysis products of Teflon' and other resins used as wire insulation materials. The toxicity of the | a F Teflon interest essir.g has ma< ethylent heretoft increase polymer shown ; There and tax which: i than a ] the mid< a machi laminate subseque the rumt puff and and hedi that the was off rtrsponsil taiy inst state tfia company An airrir example, eastern i Force'ins plant on .1 aide to i actual oc Table 9.--Par Cent Mortality * Animals* Exposed to Pyrolysis Products from "TeflouT 6 TPE Resin General, mg statet "AMia Bunds No. 900 Pyraiyaii Temperature, 960 m total Air investigat Service hi OF4 or-T EF-l EF-* 1F4 XF-4 25 0 75 if Bats too 0 Bats f Mica 0 1 BabMtl : f Ontnsapln - 06 0 ot ot "- , 09 j .. M ot too 0 0 Mandated toxicity v r Air Fc Within rtttnor ha tom. It ap 'Icty bul J well as *"d itgain * Unlara othaxiriM lndloaUd, mortality fleuraa are tor rata, t Eipomra tnrinded 10 , S 0 rati; 10 0 mt; 4 d* foisaa piga: Id.n rabbit, t On* oftwo gnlpaa pi* died. OF todtatw production prior to 1000: EF. current production. : s. The c*rlivr i.trs 134 Vol. 4, March, JkJ UCC 041096 'AL JHU.Ai.r7t lining up ii, id neither t!,r |er observable rota this thv. fety between oducing the id that which ethal effects, e particulate pyrolysis of nply because i particulate, much more identify the fume fever to it. The ovemmental ict, made 2 number. In eshold limit 1: pyrolysis it products, lire is 0.0S mow of no er of these e can, of hem to the is. Tabic 8 ^pyrolysis Bs used as ^ity of the 425 JOXtCITY OF PLASTICS AND RESINS 345 resins is not outstanding. It is arresting also that improvement in proci.;ng techniques over the past few years to made it possible to prepare tetrafluororthylene monomer of higher purity than iattofore and has resulted in a significant aenase in the temperature at which the jelymer can kill laboratory animals. This is Aown in Table 9. There is one other item about the health mi toxicity hazards of the Teflon resins riiich illustrates a human frailty rather Ain a property of the resin. Sometime in die middle 1950's there arose a rumor that i machinist had smoked a cigarette con taminated' with a little Teflon and had wbsequently dijd. In its most extreme form Ae rumor stated that the machinist took one piflf and that his lungs filled up with fluid, aad be died within 5 minutes. It is interesting that the vehicle %y which this rumor spread *a$ official safety bulletins issued by mponsible industrial concerns and by military installations. In no case did a bulletin 'fate that the fatality had occurred in the company or installation issuing,the bulletin. An aircraft plant on the West Coast, for Sample, attributed the incident to a certain *stem Air Force installation. Another Air Force installation attributed it to the aircraft P^nt on the West Coast. In no case were we to pin down the alleged event to an *ctusl occurrence. In 1958 the Inspector p'ooral, U.S. Air Force, issued the follow- statement: "Although the rumor appeared in several I0! Air Force publications, a complete Instigation by the Air Force Medical ^vice has proved it to be completely unsub*^tiated. There is no known case of Teflon Atttirity which has occurred in Air Force ** Air Force-Contractor facilities."14 Within the past 12 months the same old ?or has enjoyed a remarkable resurrectK*1- It appeared, for example, in at least 3 **fety bulletins issued in the Pittsburgh area 45 well as in other sections of the country again including some Air Force instaHaThe features were identical with the ^dier cycle. The alleged death never occurred in the facility issuing the bulletin, whose authors copied the story from some one else's bulletin or based it on verbal information from someone who had seen or heard of such a bulletin. One company quite recently copied the story from a bulletin long since disavowed by its origi nators and sent 137 copies of its version to locations in the United States, Canada, and Mexico. And while this company states that each of these locations has since received a retraction, one can predict that the original will live longer than the retraction. On July 13, 1961, and again on Oct. 20, 1961, the U.S. Air Force"reiterated, in a message to all Air Force installations, the fact that the rumor had nevdt been substantiated in spite of thorough investigation. But in spite of all this, the rumor marches on. The latest example, to my knowledge, appears in the Oct 21, 1961, issue of the Canadian Medical Association Journalu as a letter to the editor over the signature, of an industrial physician from British Columbia. This letter contains verbatim chunks from a safety bulletin issued in New York in October, 1960, and rescinded in December, 1960. The author states that his information is derived from a publication of the British Columbia Fire Chiefs' Association, and he apparently was unaware of the ultimate source of his words. The Journal editorialized on the letter, comment ing, "When a hazard is reported... sufficient evidence must be given to allow the reader to judge, on the spot, whether the conclusion is justified. These criteria have been ful filled . . . in . . . [the] letter that appears elsewhere in this issue." The human frailty illustrated by this recital is, of course, our readiness to accept and repeat rumors without checking the facts, simply because the rumor sounds credible. It is not confined to matters of health and toxicity but is perhaps accentuated in that field, to which many popular fads and fancies bear witness. It has been encouraging, however, to find that many industrial hygienists,-- toxicologists, and physicians have been quick to question the 135 i *.* if 5* a ucc 041097 346 ARCHIVES OF ENVIRONMENTAL accuracy of the rumor and haye been of great assistance in securing its retraction. It has not been possible in this discussion to cover systematically the health and toxicity aspects of all of the synthetic plastics and resins, and I have sought instead to illustrate broad areas with a few specific examples. Additional information can be found in the review articles of Schwartz,1 Harris,17 and Wilson and McCormick,1* as well as in the Plastics Safety Handbook18 issued by die Society of die Plastics Industry and the National Safety Council. As a class, the plastics and resins are not as exempt from health and toxicity problems as one might have supposed them to be on the grounds of their large molecular weight * and chemical inertness. There are problems of monomers, of low molecular weight fractions, of adjuvants, and of thermal % decomposition and combustion, products, and ' we must.be alert for them. The recognition of problems, however, is the first requisite for their solution, and the synthetic plastics and resins of today are much safer than their predecessors. It is likely that plastics and resins will continue to find expanding uses in our technology. With sufficient effort it should be possible to make sure that they will be as safe as, or safer than, die older materials which they will be replacing. John A Zapp, Jr., Haskell Laboratory for Toxicology ft Industrial Medicine, E. I. do Pont de Nemours ft Ox, Wilmington, Del. REFERENCES 1. Lehman, A. J.: Chemicals in Foods: A Re port to the Association of Food and Drug Officials on Cuiiout Developments, Bull. Ass. Food Drug Off. 15:8% 1951. 2. Larrick, G. P.: Food Additives, Fed. Reg. 26:5226 (Pt 121. June 10) 196L 3. Moncrieff, R. W.: Artificial Fibers, York, John Wiley ft Sons, Inc., 1954. 4. Zapp, J. A., Jr.: Hazards of Isocyanates ; Polyurethane Foam Plastic Production, .vu l"1 Arch. Industr. Health 15:324, 1937. 5. Ehrlicher, H.: Industriehygienische und ar beiUmedizinische Massnahmcn bei der Verwendim. von Isocyanatcn (`Desmodur'), Arbcitsschutz 7? 276, 1956. . 6. Zeisler, E. P.: Dermatitis from Elasti-Giaj, Garters and Wrist Watch Straps, J.A.M.A 114 2540, 1940. 7. The Hazard of Totxic Gases from CombustMQ of Roentgen-Ray Films, Editorial, J.A.M.A. 9. 1764, 1929. ft Zapp, J. A.. Jr. : The Toxicology of R,, Medical Division Special Report No. 4, Army Chemical Corps, 1951. (JT Underwriters' Laboratories: MisceUatMou Hazard 514ft (Oct. 16) 1960. U). Madorsky, S. L.: Thermal Degradation of Polymers, SPE.J. 17:665 (July) 1961. 11. Am. Ind. Hyg. Assoc., Hygienic Grab Series: Teflon' TFE Fluorocarbon Resins aad Thcjr Decomposition Products, Industr. Hyg t 20:436, 1959. 12. Handling and Use of Teflon* Fluorocart** Resins at High Temperatures, Poiychemkali Department, Wilmington, DeL, E. I. du Pant ft Nemours ft Co., 1961. ' 13. Harris, D. K.: Polymer-Fume Fever, T 2:100ft 1951. 14. Inspector General, U.S. Air Force: Rung of Death Caused by Teflon', TIG Briefs 10:(6) March 17, 195ft 15. Mack, G. J.: Toxicity of Decompositra Products of Teflon,' Canad. Med. Ass. J. 85:955 (Oct 21) 1961. 16. Schwartz, L.: Dermatitis from Synthetic Resins, J. Invest Derm. 6:239, 1945. 17. Harris, D. K.: Some Hazards in the Mamfacturc and Use of Plastics, Brit J. Industr. Uei 16:221, 1959. 18. Wilson, R. H, and McCormick; W. E: Plastics--The Toxicology of Synthetic Restm, Industr. Med. Surg. 24:491, 1955. 19. Plastics Safety Handbook; Society of the Pigsties Industry and National Safety Counft New York, Soc. of Plasties Industry, 1959. 136 Vol. 4, March, Mt Legis Worl \ THEODORE H Workmei enacted in 1 whose legii Of the mor workmen's been fnacte As usual, provide fo: medical, be others affe hearings an< professions, VVorkmen's which have cash benefit include Gal Maryland,'! tana. New 4 sylvania, SI and Wiscor increasing a average wee mum number tion is payal dependents.' the Longsho Act have alsc of benefit ii 10% to appr under the Lc Related to increases m Submitted fc National As: jnnies. Presented ai Meeting, Indu. llrRh,i*ctL^5- UCC 041098