Document DOyj0rbJ0Qg2kn2za3XrbRaN

_ ........ 1 ,,1<;rvs.;w-'i'* ~ . .'&*?; i ./, *- J,,, ./'>s'-r,i;:t<_ ----vii i->. - --#-< 7'?:'- 5?? ijyTv; ^ /i. iA j?.. /<*.-, s.ts~^U-J j. r>'X=,w5,4ffiBSBEt/?.,j>^ *~vi,. 4 'rk. 1976. \ 7429777. Hub. 0. 11 252! 7S0. u 830689. Nam. I / fr/c* M's*)4fd C-Z^dtsLs'Qg' Chapter 3 V" 7 v / ./ / .; Cu ;m. K. K. Japanese 2608078. 779861. York. L. I. Nass, ! 17. . 25. SIR Modem d S40600. . S7. i on P VC. Lvons. 79547. 22553 TOXICrTY AND VINYL CHLORIDE .2 H. M. Clayton British Industrial Plastics Ltd, Darlington, UK I.Y ,, f*' -f SUMMARY I Alter nearly 50 years oj commercial production of PVC, n/nV chloride monomer was found to be a human carcinogen. The following chapter describes how contact with vinyi chloride both in the work place and in consumer applications has been reduced to a level where no major hazard exists to health. Data are presented showing how great the progress has been with the problem since it was first recognised at the end of 1973. The technology used to bring about such results is reviewed. Further progress will be significantly less dramatic. AIthough the health precautions taken have led to an increase in the manufacturing costs of PVC, it remains an essential component oj present-day civilisation. There is, in fact, an increased awareness of its remarkable versatility. 11 **A-k * I! 1 8S5.5- it fi f: i *. " 4ylf'ir*.C--jf""** s' ^ 3.1 INTRODUCTION Some future historian of the chemical industry will, in retrospect, point to the vinyl chloride (VCM) problem as a landmark in the progress of its olTshoot, the plastics industry. Long periods of exposure to a substance hitherto regarded as relatively medically safe led to the appearance of a number of occupationally related diseases, including a rare form of incurable liver cancer--angiosarcoma of the liver. The intensive investigations set in train by this discovery' are by no means complete--nor is the development of improved manufacturing methods and procedures to reduce still further the traces of VCM escaping into the 43 ) fe-Y'XTT . at-zv.evu -- f i*.zrysgif. I : -fi fi-c'hkkrj *. v'.rWd*r.V j. iT 1 * `.i * V' .T- ' * ' it xzZrtfefVi fir. fsYkfrtTk:- II. M. O.AVIUN ;i[mos]iMc tin polymerisation anti fabrication plants) or, migrating into foodstuffs (alt>cit in minute quantities). Medical research programmes continue into impmved and early diagnosis of VCM-rdaled diseases as well as into understanding liie mechanisms of its various creels. In spite of the Tact that any report on the problem must, to some extent, be regarded as interim, the PVCimhislry can point to considerable progress made during the last three years in solving the various technical problems encountered in reducing any possible hazard to a socially acceptable level. The author lias been actively involved in one of the UK Working Parties dealing with the VCM problem and lias first-hand experience of the extensive co-operation among companies (both national and international), trade unions and government agencies in order to bring about a speedy solution to the various problems. In spite of the ditlerenccs . of detail, bioadly similar approaches lo problem resolution have been employed around the world. There are useful accounts of the problem seen t hrough German eyes,J 7 and through American eyes:'1 the la iter includes a chronology of key events. t ! ' , ' I 3.1.1 Background in the Problem Latgc scale production of PVC resins by the polymerisation of VCM Mailed in the l93Us. Although the hazards of fire, explosion and narcosis were well recognised, there was no evidence to suggest that VCM constituted a health baza rd; indeed it bad been suggested that it would have been adopted as an anaesthetic, but for the fire and explosion hazard. The first work suggesting a potential for carcinogenicity came from studies carried out by Viola and eo-workers in an attempt to reproduce aei o-usteolysis in rats.1 (Suggestions that Russian workers had observed liver damage amongst polymerisation workers as early as 1949 arc unfounded ' The study ielated, in fact, to 48 workers compounding PVC wuh chlorinated diphenyl). Aero-osteolysis, Raynaud's Syndrome and selciodenna had been observed amongst some PVC polymerisation workers, particularly those involved in the manual cleaning of deposits from autoclave interior walls. The act of removing deposits of encrusted PVC resin releases unrcacted VCM and it is this VCM which presumably causes diseases such as acro-ostcolysis -a vascular disturbance. 1 he clinical findings may be, for example: I. Increased sensitivity lo cold, 2 I'ins am! needles and other pai aeslhesia, and 3 Changes m the colour o! the skin ol the digits. 99IS00 TOXKT1V AND VINVt. Cttt.OHlUf: 45 These tire the signs of a disturbance known as Raynaud's Syndrome. Another feature of theacro-osteolysis may be changes in the bone pattern of the terminal phalanges, often accompanied by deformation of the nails. Scleroderma is the medical term for the thickening of the skin which can occur as a result of contact will) VCM, and it aw lead to gross deformation of, lor example, the thumbs. In studies reported in 1971, Viola showed that exposure of rats lo atmospheric concentration as high as 30 000 ppm (parts per million) failed to produce acro-ostcolysis. However, malignant tumours were found. The level of exposure of the rats was so high that its significance lo human beings was not clear and further woik was necessary. A consortium of European interests, including ICI Limited (UK). Rhone-PoulencPolymcrcs (h'runcc), Montedison (Italy) and Solvay (Belgium) commissioned Professor Ccsare Maitoni to carry out further studies. Cancers were confirmed in animals at exposures as low as 250 ppm, including angiosarcoma of the liver.1 Multoni's findings led to an epidemiological sun'ey being conducted in the USA oil men who luid previously worked in a PVCpolymerisation plant (D. P. Goodrich, Louisville, Kentucky). This study revealed a cluster of cases of the exceedingly rare form ofcancer, angiosarcoma ofllic liver. The results of this study group (by Creech and Johnson) were transmil led to the National Institute of Occupational Safety & Health (USA) on 22 January 1974, the Department oi'thc Lmploymenl (UK) on 23 January 1974 and the Acting ChicT employment Medical Adviser (UK) on the next day. Other national governments were notified during (lie same period. In the UK on 29 January 1974, ICI issued a press statement announcing that government departments, the TUC, its own workers and customers, were being informed of the facts available so far concerning the urgent investigations into the cause of death ofa seventy-year old retired autoclave worker. Ii was subsequently confirmed that the worker, who had died in 1972, was sulfering from angiosarcoma of the liver. During his working life lie had undergone twenty years exposure lo VCM. In February 1974 the Chemical Industries Association in the UK formed a Vinyl Chloride Committee to study the whole problem and institute necessary programmes of work through a series of specialised working parties. Similar arrangements were made in other countries of the world. In P.urope international co-operation was ensured through Oil IC (Conseil Luropcen de Federations dc Vlndustrie Cliimique) which I'ouned its own Vinyl Chloride Committee with national representatives taking decisions on behalf of their colleagues. 4 b II. M. Cl.AVmN 3.1.2 KxIcitI of the Vinyl Chloriile Problem Al the end of 1976 the number of notified angiosarcoma deaths was fiftyfive. The distribution of deaths is shown in Table I. The average length of exposure before death was 15-20 years. Almost without exception the deceased had worked on polymerisation plants and bad been engaged in the manual cleaning of autoclaves during part of their service. Certain plants show a clustering of deaths.6 TADLb I NUMiti.n or Nonni.o deaths thou ANGIUSAKCOMA (ur TO Till: HMD OP 1976) Country Number of deaths Jlclgitmi Norway UK Italy Sweden Pastern llurupe [`ranee Germany _ Canada USA I I 2 J 3 4 6 7 10 |8 Champion'1 and other authors'1 describe other disorders caused by VCM. These include acro-osteolysis (approximately sixty-five cases in the world all autoclave cleaners), associated Raynaud's Syndrome and scleroderma. ami liver dysfunctions. This latter group of disorders has sometimes been referred to as TVC Disease'.1 Mom of the workers mi lie ring from this group of disorders were polymerisation workers and particularly those manually cleaning autoclaves. Whilst some disagreement exists on the severity of the elfecls of the other groups of disorders, angiosarcoma of the liver is invariably fatal. Studies on the carcinogenicity of VCM. whether by inhalation or ingestion, will not be complete loi some eonsideiable time. Until a medically pi oved acceptable exposme level is established, the object of 91.500 'I OMIT I V AND VINY1. Cl It.(tit 1 lit- 47, much of the work has, during the last three years, been lu ieduce possibleexposure to VCM to as near zero ns possible. lu the meantime, epidemiological surveys of polymerisation workers have been carried out in the UK, the USA and elsewhere to establish ihe possible extent of the pmblem. Other work has been directed at improved diagnostic methods to give advanced warning of problem s.J-M Yet further work lias been aimed at understanding the mechanism of the way in w liicli VCM causes angiosarcoma of the liver. Tentative hypotheses have been advanced involving* the trole of metabolites of VCM, such as cbloicthylene oxide and modification of DNA molecules.'' Although no limit conclusions can yet he drawn from the medical work can icd out, some authorities have attempted an assessment of the risk from VCM in the light of recent performance. The results arc encouraging.y'lo Whilst there must inevita bly be an element of speculation in assessing all medical work, the results of work by the PVC industry are demonstrable. These will be described in the following pages. 3.1.3 Sources of Interface between Unman Beings and VCM To umleisland the problem and the progress made in containing it, a brief account of the processes fui the manufacture of PVC resins, compounds and finished articles is necessary. VCM is a gas boilingat -- 1 3-5C. It is therefore necessary to polymerise it under pressure. Commercial resins arc made generally by polymerising 70-9U/' of the VCM. This leaves a quantity of unreacted gas within the autoclaves as well as within the particles of the PVC resins. Complete reaction of the monomer is not Feasible. If made by the suspension or emulsion processes (accuunling between them for the majority of world PVC production) it is necessary to separate I lie water present, for example by centrifuging and drying. All these processes lead to the release of I'm ther quantities of VCM. Finally, the lesin itsclfwill still contain small quantities of unreactcd monomer, particularly tiiosc resin types with dense glassy particles used Tor direct powder blend processing, c.g. in the extrusion of water pipe or extrusion blow moulding of bottles. Subsequent processing operations involving the use of heat will lead to further releases of VCM. Small quantities remai/iing in the Finished articles will be released by slow dill'usion. In the case of packaging materials, very small quantities of VCM may be transferred into foodstuffs and beverages. There is also the possibility of transfer of very small traces of VCM into drinking water supplied through PVC pipes. 48 II. M. CLAYTON 3.2 ^^VEI.OPMLNTS IN REDUCING CONTACT WITH VINYL CHLORIDE I lie preceding section identified the possibilities ofexposurc to VCM in the following areas: 1. Polymerisation and associated operations, 2. Storage ami handling of PVC resins, 3. Hot processing of PVC compounds, 4. fabrication of finished products, and 5. Dillusion Irom finislicd products, including migration into loodslulls and diinking watei. llcfoie considering the developments that have taken and still are taking place, some estimate ol exposure levels is appropriate. Before the TAItl.H 2 VCM LI.vi LS AlUlWND 1*01.VMliRISA LION I'l AMIS Year Typu at average Icve! (ppm) 1945 1955 1955-1960 I960 1970 M k!-]973 Mid-1975 bud 1976 500 -1000 400-500 300-400 150 5 2-5 realisation that c .ccssive exposure to VCM could lead to crippling or even fatal illnesses, little ichahle data on atmospheric VCM levels around polymerisation plants was available. I able 2 gives such inlormation, and is based on mine recent measurements and earlier estimates. Levels cncounteied in Ihe labiicnliun and end use seclois ate given in Table 3. TAUl.b 3 Pi 1`lt'Al. KI.C1.N I Vt M I f.VLI.S (n-Mj / aeatiais Jaanttre 1974 July 1975 December 1976 In [il.uii atmo\pheres In I'W buttles In bcsct.tgcs 2 I5 -50 |) 1 <2 < 1 <1 0 01 <. 0 005 89I.S00 S$H TOXICITY ANO VINYI. Cm.ORIIMi 4<j; Estimates of the intake of VCM have also been made;4, * these are shown - in Table 4. As expected, the figures show that lho intake of VCM by polymerisation plant workers was by lar the most significant. As will be shown, many of the steps taken to reduce the exposure of polymerisation workers to VCM led to improvements in downstream activities. TAOLE 4 ten IM.VILS Ol: IMILY VCM INTAKE Type of person Daily VCM intake {gfkg body weight) Polymerisation worker at 1 000 ppm 0-36 500 ppm 018 5 ppm 0 001 8 2 ppm 0000 7 Rubrication plant operator 1 ppm 0 000 4 European cithen- through ingestion of food 0 000000002 -0 000000008 i 3.2.1 Polymerisation Plant Developments The level of VCM permitted in the atmosphere about polymerisation plants, the absolute quantities which may be emitted into the atmosphere, the quantities which may be present in waste water and in some cases the quantities remaining in PVC resins, arc the subject oTvarious compulsory national regulations or voluntary codes of practice and hygiene standards They vary in detail and in the methods of calculating certain required numerical data, in particular average and peak personnel exposure results. These apparent did'erenees obscure the general similar level of achievement in most countries of the world. Typical average levels of atmospheric VCM about PVC polymerisation plants arc now 2-5 ppm; this has been achieved by improving the operating procedures in a number of ways: I. More cllicicnt degassing of the PVC resin, either in the polymerisation autoclave or downstream from the autoclave, icsulting in lower losses at the drier stage. Much eilort has been 50 II M. Ct AVION expended on llic development of improved degassing systems ami a number of patents have appeared.UJ'33-3'3 J-33 The D.F. Goodrich process for emmter-eunent stripping of resin slurry has been licenced to a number of other PVC resin producers. Development of automated high pressure washing methods for autoclaves so as lo avoid the necessity of entry it ml manual cleaning. Development of autoclave suiface-treatment systems which reduce the tendency ul crusts or skins lo lor in, so reducing the amount of cleaning necessary. *3-1 1 " i11-2 -4 ] he .Shin-htsu Companv of Japan and the B.b. Goodrich Company of the USA have licenced their technology to a number of other PVC resin producers. Ollier companies have icpoiled (lie use ol additives in polymerisation recipes which reduce encrustation.34 Remote operation ofautoclaves which may often be controlled by a computer. Ibis technique will grow in importance as a result of recent spectacular explosions, such as that at Flixborough. 1 he use ol larger autoclaves, so i educing the number of flanges and \abes. etc. lor a plant of given output. Such developments were under way belore the health problem of VCM was appreciated. I he largest reported commercial autoclaves arc (hose operated by C hemischc Werke iluls --200m3. B.b. Goodrich operates a plant at Louisville with autoclaves ol 62 m3. I be Sliin-Ftsu Company of Japan uses autoclaves intermediate in size at 130 m3. All three companies have licenced their larger autoclave technology. Development or recipes and agitation systems yielding more uniloim resin particles. Uniformity of resin particle structure is a rccpiiremcnt or the industry, irrespective of the VC?M problem. Added impetus has been given lo lhe work since a more uniform panicle st incline will lead lo more uniform degassing. Fspecial pnonly Iras been given to investigations into the production of resins lor such applications as waiter pipe and bottles which normally start from a powder blend. Processing rales ate iullueneed bv bulk density and particle shape. The problem is to produce resins vviih a regular shape and porous structure and at the same lime an acceptable bulk density. Some progress lias been made towards this objective. Use ol icspiiaiors by pci sound when carrying out operations in aieas with unacceptably high atmospheric levels of VCM. Development ol automatic sensitive monitoi ing systems scanning a It w ot king aieas 691.S00 tnxn tiY and viNYt. cm.oittni: 51 Outstanding problems still exist, notably in reducing losses from certain downstream stages after the autoclaves. The Environmental Protection Agency in the USA lias set very stringent requirements which m:iy prove diHicult to achieve.3*1 Abramovvitz31' gives a useful account of progress in mass polymer plants. Some quality problems have inevitably arisen as a result of producers finding themselves forced to bring about rapid improvements with existing equipment or incompletely developed modifications. These considerations, as well as long delivery limes of some equipment, have forced companies lo use ad hoc interim arrangements. For example, reductions in residua! VCM have been^-Heclcd by the expedient of more severe drying conditions leading to poorer resin colour and stability, poorer How in hoppers (due to static electricity Gleets) and reports of dillerent fusion characteristics (probably attributable to subtle surface changes due to the higher drier temperatures}. The cost of modifying some plants has resulted in their closure. Reduction of approximately 10 of plant productivity has occurred in many cases, and production costs have increased by tin average of 10-15 On the other band, VCM utilisation efficiencies tire markedly improved. A further benclil is that (lie reduced necessity lo enter autoclaves for cleaning has improved working conditions. 3.2.2 Reductions of VCM Exposure in Handling, Processing tint! Fabrication The preceding section refers lo steps taken to reduce residual VCM levels in PVC resins. Some resin particles arc more diiVicult to degas elfeclivcly than others. For rigid extrusion, extrusion blow moulding and injection moulding operations (using a powder blend feed rather than a fully compounded grannie), dense regular particles arc required in order to obtain good powder How and ellicient packing of screws. Such PVC resins retain VCM more tenaciously than those with an open particle structure used for the manufacture of plasticised compounds and plasticised finished products. In early llJ7'l it was common for such resins to contain 1000 2000 ppm residual VCM. In some cases even more was retained, e.g. in homopolymers made by the mass polymerisation route, and acetate copolymers. VCM is slowly released from the polymer at room temperature. At elevated temperatures such as those encountered during processing the i dense is extremely rapid. In the USA, there are reports of VCM accumulating in the free head space of tail cars until the lex els reached aic 52 II. M. Cl.AVION poientil^^capablc ofexplosion. In Europe, {lies have occurred at the vents of cxti'udcis processing rigid powder blends.16 In the early pari of 1974, extensive surveys of more than 100 companies were carried oul in the UK, in order to establish the levels of VCM encountered during processing and handling operations. The findings17 arc shown in Table 5 TAIiLE 5 AIMOSI'JICIUC VCM IIIVKU AROUND RRiK issiNt; and fadkica'iion operation v/v) IWC Locution Range Auerage Polymer warehouses Cold mixing equipment 1 ligh speed mixing' Danbury lyjvc mixer Two-roll mill compoundiing Paste mixing Calendering 1 nbrie and paper coaling Pipe extrusion Cable extrusion Injection moulding PVC welding Vacuum forming tioule blowing Extrusion of flexible film Extrusion of rigid sheet <2-17 <2-3 <2-53 <2-13 <2-2 <2-16 <2-3 <2-5 <2-7 <2-4 <2-4 <2 <2 < 1-2 <1-2 < 1-4 2 <2 3 2 <2 <2 <2 <2 <2 <2 <2 <2 <2 <t <1 <2 * .Single rcstili of 53 will) high acctale copolymer al Inn filling point. Most of the results wcie of a low order, averaging 2-3 ppm. However, certain unacceptable practices came to light. High speed mixing areas were particularly prone to unacceptably high atmospheric levels. This was not surprising since al temperatures in excess of IKCC, PVC resins release VCM very rapidly. Smpioveil operating procedures, such as the installation of localised extraction, the venting of extruders am! mixers to the outside of buildings, and good general ventilation in warehouses etc., resulted in levels below 2 ppm by caily 1 'J75. Ity the latter part of 1976 the clients of inducing 0ZIS00 toxin I Y AND VlNYl- ntLOtUDU S3 residua! VCM in PVC resins have resulted in levels generally below I ppm with many resulls at 0*1 0 2 ppm; near the limit of detection of (he analytical methods used. 3.2.3 l'ola hie Water Supply Virtually all potable water pipe is produced by extrusion from powder blend. Although some dillercmes in formulation practice have developed with, for instance, Europe favouring l tic use of solid lead salts as stabilisers and the USA favouring the use ofliquid tin compounds, a common feature of the pipe-making industry was the increasing use of high bulk density resins to increase extruder outputs. As reported earlier, such resins tend to retain residual VCM more tenaciously, and levels of 1000 ppm or more were not uncommon. As well as the. necessity to reduce residual VCM levels in new pipe production, measurements on drinking water passing through existing pipe systems were required. At one location in the UK, the Chemical Industries Association was able to carry out measurements on drinking water supplied entirely through a rigid 1'VC pipe network manufactured by one company. The pipes were 102 mm {4 in) am! 152 mm (bin) in diameter. Even in those parts of the network where PVC resins containing 1000 ppm residual VCM had been employed to manufacture the pipe, no VCM could be delected in the drinking water. Eurther studies were carried out on houses connected by small bore PVC and chlorinated PVC pipe, i Icre.a more unfavourable suifaee/volume ratio could be expected to increase the possibility of extraction of VCM; however, only at one location could any VCM be detected. This was in a home built some nine or ten months previously and unoccupied since completion. In the small volume of water which had stood in contact with the PVC pipe. lM)l5ppm of VCM was delected.15 The above results were extremely encouraging. Nevertheless, three lines of attack were vised to reduce VCM levels in rigid PVC pipe* 1. Inci eased use of hot (i.e. > 1 !0C), high speed mixing, combined with aeration, in the preparation of the powder blends. Some companies had originally only mixed to relatively low lempeiatnres, i.e. 70 XtP'C. ` 2. Impioved degassing techniques during PVC resin production. Typical average residual VCM levels foi pipe resins are now in the 5 50 ppm range. 5-1 II. M. ( I AVION J. ^development of suspension resins of increased porosity. A judicious selection of suspending agent combinations, or other appropriate formula modifications in tltcca.se of mass polymer, can produce resins with regular particle shape, significantly increased porosity and only a marginal sacrifice in bulk density, Further developments can be expected by this approach. A combination of the above approaches has led lo significant reductions in the residual VC M levels in rigid I'VC potable water pipe and some companies are now achieving average levels of less than I ppm. Huih the Ethyl Corporation in tiic USA and ICI in the UK. have deleimined the diffusion eoellieieiit of VCM from rigid I'VC pipe. ']'he results have been used lo predict VCM levels in th inking water supplied lo typical households, in the USA anil the UK, using pipe willi 1 ppm residual VCM.16 For die UK, the level of VCM in the waiter is ptcdicicd to lie between 0(106 ppb and 0 022 ppb (b = 1000 million). In the USA the predicted v alues aie 0 01 0 I 4 ppb. Not surprisingly, the use of liquid tin stabilise!s in the USA in place of solid lead salts in the UK gives a higher eoellieieiit of dillnsion. Since most water consumed is used in cooking or in the preparation of hot beverages, with allcmlant furlher loss of VCM, the average levels in consumed water ate decreased even further. .1.2.4 1 tttnl Packaging Any review of the use ol PVC lor food packaging must, in the light of the VCM problem, lake into account both the political background of legislation and the variations in usage in different count/ics. In the USA, Die Pood and Drugs Administration (PDA) is constrained by prcccdcnlsscl in the interpretation of the so-ealled Delaney Clause ofthc l inttl. Drug and Cosmetic Act of 195,5 which states 'No additive shall he deemed to he sale it it is found, aflci tests which arc appropriate for the evaluation of die safety of (ooil additives, lo induce cancer in man or annual . . .' A i the lime ol its introduction, the above clause was clearly in tended lo be intetpreled in the light of testing, probable consumption levels, safety maig'ms and reasonable certainty about conditions of use. In practice, precedent has established the clause as intending to mean a total ban on any additive suspected of causing cancer under any conditions. As a tesull ol this. A met lean legislators are not hue lo make a truly scientific assessment of a possible caictnupenic hazard. Any regulation ioxk iiv and vinvi. nu.oium: 55 which falls short of a lota! prohibition is, in fact, a politically inlluenced decision. In this situation the PDA produced a new form of words in a proposed regulation issued in September 1975. 1 lie proposals permitted the continuing use of I'VC in contact with foodstuffs'where the potential for migration of vinyl chloride is diminished to (lie extent that it may not reasonably be expected to become a component of food.' Against this definition (he PDA proposed to allow the continuing useofflexibie film and hose. Rigid and semi-rigid foils and bottles would have their status of prior sanction withdiawn. Any producer of the latter group of products would then be obliged to demonstrate that they met the PDA criteiion. At the time of writing no decision lias been taken on the proposals and it is extremely difficult to predict the outcome. In the United States the most important outlet for PVC in food packaging is in plasticised films such as meat wrap. On the other hand, the use of rigid I'VC is relatively less important and in particular the use of rigid I'VC bottles for food packaging has scarcely developed. Therefoie. whatever the decision its economic ell'ect on the United Slates packaging industry will he relatively small. The P PC is slowly moving towards a unified approach on food packaging legislation, and no matter how much European countries may wish to display their independence of events in the USA, in practice, packaging legislation in the hitler country has previously influenced events in P.uropc. In the case of PVC and VCM, however, the EEC produced a draft directive at the end of 1976 which takes into account the realities of the situation in Europe.Since a `no effect' level has yet to be established for VCM, the proposed directive takes into account the best practical achievements of European industry in reducing VCM levels in I'VC food packaging, before discussing these achievements, some account oT the variations in European food packaging practices is necessary to understand the basis of the proposals. Whilst plasticised PVC lilms arc important for food packaging in Europe, the use of rigid I'VC bottles or foils is especially important. Certain countries in Europe show marked variations in the pattern of usage. In France lor instance the manufacture of rigid I'VC bottles for mineral water is a major industry, accounting for 100000 tonnes of PVC compound in 1976, whereas in Germany, rigid PVC foils account for the major proportion of rigid PVC used in food packaging'. ' In the UK a major proportion of rigid I'VC bottles is used for the packaging of fruit squashes. Such squashes are mainly consumed by young children, a section of the population for whom iheie is special concern. ui-soo s'sa 56 !!. M. CI.AVHJN '1 hciWfre also some significant differences in formulating practice in F.mope. In France ami countries subjected to French influence, most bottles are based on tin-free formulations. Such formulations contain significantly higher levels of other liquid additives to help processing. In Germany particularly, tin stabilisers tire preferred. Whilst there is a significant usage of tin-free bottle formulations in the UK, the mnjotity are based on tin stabilisers. Most rigid PVC foils in 1 iurope tire based on liomopolymers, but in the UK a significant percentage o!' production (circa 75%) is based on acetate copolymers. TAilt.t- 6 VCM lliVMS IN I'K-MANUrAClUUl.O t'VC PACK ACilNd MA1I.HIAI.S rc\U Mention VCM concentration (j>ptn) January 1974 March 1975 December 1976 Uattlei Level in polymer al lime of use Typical level m powder blend al lime ol rmumfaeliiie T\ pival level in tioule wall 1'tnt Level in polymei al time of use Typical level in knl i U` \<f'ic Jiltn [r \ tru\ioti hlt/wti) Level in polymer al lime of use Typical level in film 500 -1000 50 50 100-250 5 .1 800-1 500 80 100-250 8 400-1 000 1 50-150 1 15 -50 1 <! 50 100 3 5 30 <1 table 6 slums the residual VCM levels in UK-manufacimed PVC packaging materials doling the lluce-ycar period January 1974 lo Ikacmhei PUC It will lie noted that in the case of flexible film, residual VCM levels were very lovv.cxen tit the beginning of the period under review, i his is explained by the fact that the piesence of plasticiser aids the release ol VCM during hot processing. 1 he highei levels of residual VCM in rigid foil computed with bottles is explained by the use of eopolymeis in the manufacture of the former mnletial. the relatively low softening point of copolymers makes it extiemely dilliculi to achieve low VCM levels either at the resin-making stage or dm ing mixing Fur live Imescenble future copolymer loiIs aie likely locomain slightly Inglici lesidual VCM levels than bottles or lulls based on 2ZLS00 SSd lOXft fiV ANtJ VINVt. em.OKfDt: 57: liomopolymers. Since most such foils are either used for the packaging of solid foddslullsor malerinls of restricted shclflilc, (he slightly higher level in the package does not normally lead to higher levels in foodstulfs. The results shown in Table 6 are also generally typical of the position in Europe as tt whole with two provisos: ], Where liomopolymers are used to manufacture rigid foils, residual VCM levels ate lower ami more in line with UK results for bottles; and 2. Where tin-free stabiliser systems are used in bottle formulations, the presence of extra liquids and low melting point additives assist live displacement of VCM during processing, and residual levels in such bottles will tend lo be lower than in bottles based on tin stabilisers. Some important conclusions can be drawn from 'fable 6: 1. Flexible film lias low residual VCM levels even at the beginning of 1974. Subsequent improvement has only been marginal; 2. Rigid PVC lias shown a marked improvement, especially during the period January 1974-March 1975; .1. Since current USA achievement docs not differ significantly ('torn that in Europe, the logical course for the Food and Drug Administration would be to accord rigid and semi-rigid PVC the same status as flexible film, i.e. continuing prior sanction. 3.2.5 VCM in Packaged Food Numerous studies have been carried out on the migration of VCM from the package into the contents. In the UK regular surveys have been carried out on random samples from retail outlets. The results of this large body of continuing work may be summarised as follows: 1. Under ambient conditions rigid PVC packages lose part of their residual VCM to the atmosphere; after a year approximately 3(1 has been lost in (his way;- 2. VCM reaches an equilibrium concentration in the contents of packages alter several months; 3. Titis equilibrium concentration allows typical partition concentrations to be calculated. For foodstulfs commonly packaged in PVC in ilie UK the concentration ol vinyl chloride remaining in the PVC is 200 700 limes that present in the content v. 4. Only very small proportions oflhe VCM in a package may transfer lo the con ten is and for a wide range of Ibodst nil's tin: levels of extraction arc similar; and 5. Under certain circumstances small numbers of packages may remain on relnil shelves for far longer than the recommended shell' life of the contents. The use of the partition data enables calculations to he made of expected maximum VCM contents of foodslulfs packaged in rigid PVC. Intensive retail surveys have provided firm support for such predictions. At the end ol 1976, VCM levels in foodslulfs packaged in the UK were in the range 0 003 0 (HIS ppm or less. Work carrier! out both in the UK and in Continental Europe lias shown that the equilibrium partition concentrations with tin-free bottle formulations such as those typically used in France arc not as favourable as those with tin stabilised formulations. For some foodslulfs the ratio of concentration of vinyl chloride in the bottle to that present in the contents may be as low as 100. This disadvanlage is, to some extent, offset by the fact that VCM lends lo be displaced more readily during hoi processing, giving slightly lower initial levels in the bottles.7(1 Gilbert"'* also lias suggested that at low initial concentrations in bottles and other rigid containeis, extraction of VCM by foodstuffs will be less than expected because ol the presence of active sites in (lie matrix of (he PVC which retain VCM. A lurthcr factor inlluencing the lil:C proposals is the stale of development of suitable analytical techniques. With some foodstuffs problems of interference exist. Also, even in the hands of highly skilled research analysts, it is very dillicull to gel ngiecmcnl between dill'eient laboratories. finally, the manulact in er ol the PVC container or fnodsiulf must be able lo observe tivuds in the VCM control of his products and take corrective action, for this to be possible, the maximum allowable VCM must he rather more than that at the limit of analytical detection. Against this background the PEC has produced its draft directive. I lie ' main tvaunex of the directive are. 1. A limit ol OOStlmg'kg in the loodslulf (i.c. 5l)ppb); 2. A limit of I ing/kg in the container (i.c. I ppm); and 3. In (he case of containeis made from copolymers, a limit of 5 mg kg, piocided that they are not used lor liquid foodsiulls. UlXItllY AND VINVt. niitmiiii; 59 It is too early lo predicl the success ol the proposals. However, similar requirements or recommendations already exist in some European countries. For further information on VCM and PVC packaging the reader is referred to the publications of GECOM (Grimpc d'Etudc Pour ie Condilionnement Modcnie). GECOM was founded in France and is an association of IWC producers and the manufacturers of packaging and foodslulfs, including water, oil and wine. It was set up to study the various problems in the packaging field raised by the discovery of (he toxicity of VCM. The address ol'GECOM is 11 Rue Marguerilte, Paris 17. Oncoftfie GECOM publications gives a very useful account of the problems of analysis of foodslulfs containing vinyl chloride.is 3.3 ANALYSIS AND DETECTION OF VCM The progress described in the preceding pages would not have been possible without significant advances in the methods of detecting VCM. Developments have taken place in three major areas: 1. The monitoring of atmospheric levels of VCM in the vicinity of polymerisation planls; 2. Analysis of VCM, often present only in minute traces, in PVC.' resins and compounds, food packaging and foodslulfs; and 3. Small portable detectors. 3.3.1 Monitoring of Atmospheric Levels Prior to 1974 the measurement of VCM was a time-consuming process and the methods required development so that they were capable of yielding accurate results at great speed and were suitable for the sequential scanning of working aieas on polymerisation plants. Two methods of analysis have gained particular favour for the task: 1. Gas chromatography combined with a llame ionisation detector; ami 2. hi fra led absorption. In the first method samples of airure drawn through a suitable column which sepmates VCM from other possible aiibornc contaminants. 3 lie VCM fi action is burnt in the presence of pure hydrogen, forming ions which are then measured by an electrical technique. 'I his met boil is very sensitive and can measure fractions of I ppm of VCM. I he second method depends upon the fact that VCM absorbs inflated radiation at a tmmhei of wavelengths which tire characteristic both of the 1.900 SSd 60 ]!. M. CLAYTON imsaluration and of llie carbon chloride bond. The timounl of radiation absoibed depends on the amount of VCM in 1 he beam of radiation. In tire equipment used in PVC plants, lire oplicnl system is designed so that the beam of radiation is repeatedly reflected between a pair of mirrors to give an absorption path length of more than 20 metres. Tins means that the equipment is of reasonable size, yet it can still detect concentrations of less than I ppm. Doth methods are in widespread use in PVC polymerisation plants, drawing in samples of air from a large number of points which are scanned sequentially every few minutes. Such instruments are often used in conjunction with specially designed computet s which permit t apid calculation of shift averages and analysis of excursions above hygiene standards. As an example of what is being achieved by such techniques, see paragraph 37 of the `Vinyl Chloride Code of Practice for Health Precautions' issued by the Health and Safety [executive in Fcbruaiy 1975. 3.3.2 Detection iu Foodstuffs VCM may occur in both PVC resins and articles fabricated from PVC whete the concentration may range ft om fractions of I ppm to hundreds of ppm. However, in the case of foodsUill's the range is parts per billion. Quick, reliable methods of analysis were required to allow control of PVC manulaclming processes as well as to assess the progress in reducing the minute traces migrating into 1'oodsl nil's. The method which has gained general acceptance for such work is the so-called headspace method.36 In this method, llie sample is dissolved in a suitable solvent and allowed to equilibrate in a small vial. Gas from tbe headspace, consisting of a mixture of (lie solvent and VCM, is passed through a gas liquid cluomalographic column and the peak due to VCM is measured. Systems are available which permit the automatic measurement of pre-loaded samples in a typical measurement time ofthirty minutes. A number of solvents are used, such as lelrahydt o In ran and dime I by Incut a inkle. In iliecase of PVC resins and objects, tin: method iscapablc of measuring iiaclions of I ppm of VCM. With some foodst tills the method will tided a lew pails pet lutliou. whereas with otheis there are serious inta Terence problems. 3.3.3 Portable Detectors A number ol mmiatuie detedurs have been developed which may he worn by plant pcisiuiiicl so Unit persona! exposure levels can he ineasined TOXK'IIY AND VINYL. CIILOIlllJl: 61 directly. All types depend on the oxidation of VCM to hydrocldoricaeid or chlorine. These substances then stain specially treated paper, and the intensity of l lie slain gives an indication of the intensity oft lie concentration of VCM. Such instruments will measure from fractions of I ppm up to 40 ppm, depending on lltc equipment. A Chemical Industries Association publication36 gives details of reliable analytical and detection methods. 3.4 CONCLUSIONS Due to the long latency period of angiosarcoma of the liver, it is inevitable that some further cases will occur. However, there is no doubt that the major hazards associated with VCM have been corrected. Some further improvement in reducing atmospheric levels and the levels of VCM remaining in PVC resins will occur, but the improvements will be relatively marginal compared with what has already been achieved in so short a time. PVC continues to remain an essential component of present day technology with an improved awareness of its remarkable versatility and usefulness. ACKNOWLEDGE M LNTS The author gratefully acknowledges the help and encouragement of numerous colleagues in the PVC industry, both in the UK and overseas. UEFEUENCES !. Viot.A, P. L., Ihc.oni, A. and Covota, A. Cancer Research, 37 (1971). 51b. 2. Maltoni, C., Cursi'i, M. & liuitcu, P. J. 11. E.wcrpta Merlieo International Congress, Series No. 275, 1973. 3. Hakni.s, A. W. National and international aspects of die VCM healili problem. Pari It. Paper given at die UPVjPHI Joint Conference. Vinyl Chloride anti Safety uf Work, 28 May 1975. 4. Pi ot ceilings oj the Royal Society of Metheine Set lion of Occupational Mctlicniv, 69. April 1976. 5. Vinyl thlorhle toxicity anti the use of PVC for packaging foodstuffs. A presentation by the ITTIC Committee for the Toxicity of Vinyl Chloride. February 1976. 6. CitAMrioN.lt. 1 AssesMiiei . of toxic dsks to woikcrs in die plasties imlusity, UPf'IPHI Joint Conferent e Vinyl Chloride tnul Safety at ttWA, 2K May 1975. S'SH Vr/? Sfc&*e in rT-- in o o 0) 08 CC .,<- _ -*S*, y s; *3j>&+{ -+ ** '.5 4,.< ;VTi irM *&S >^1^5 _ - r- j**/^Li 25H 62 H. M. CLAM US 7. 1'C PIC. dn Example of a Problem Resowed, Verband Kunststoifcrzeugcnde Industrie e V Frankfurt am Mam. West Germany, I; September 1975. 8. Williams. D. M. J., ei at. Brit. J. Indus Med. 33, I Aug. 1976) pp. 152-7. 9. Wilson. R. Risk benefit analysis far toxic chemicals' rim/ chloride. Paper Given at the 172rid ACS National Meeting. San Francisco, 25 August 1976. 10. Schlatter. Ch. 'Gelahrdung von Arbeitnenmer und Konsuuicni durch Vinyl Chlond.' Press Release by Lonza AG. 29 October 1975. 11. `PV'C and Health.' A background statement issued by the Society of Plastics Industry Inc., Neu York, April 1976. 12. British Patent 1 291 145. 13. British Patent 1 421 71S. 14. British Patent 1 444 360. 15. Federal Register, 21 October 1976. 41. No. 205. Environmental Protection Agency. National Emission Standards for Hazardous Air Pollutants, Standard for Vinyl Chloride. 16. Private communication. 17. Clayton. H. M. Guide lines for PVC processors. Paper given at the BPFPP.I Joint Conference Vinyl Chloride and Solely al IVork, 28 May 1975. IS. Belgian Patent S33 558. 19. Belgian Patent S3I 964. 20. Belgian Patent 833 041. 21. British Patent 1 407 665. 22. Belgian Patent 837 070. 23. British Patent 1 446 5S3. 24. US Patent 3 669 946. 25. Moffitt, T. W. Private communication. 26. Moffitt, T. W. Private communication. 27. Document 1565/VI/76-F. 28. Tester, D. A. and Moffitt, T. W. Vinyl chloride monomer in food packaging. Paper given at the BPFjPRI Joint Conference liny! Chloride and Sajety at Work, 28 May 1975. 29. Gilbert, $. G. The migration of minor constituents from food packaging materials.' J. Food Set., 41, 1976. 30. Abramowitz.R. J. `Bulk polymerisation plants.' Paper given at the |72/;</.4CS National Meeting, San Francisco. 2 September 1976. 31. Unexamincd Japanese Patent 47 076 (1976). 32. Unexamined Japanese Patent 47 081 (197C). 33. Unexamined Japanese Patent 17 950 (1976). 34. Unexamincd Japanese Patent 17 287 (1976). 35. Food and Cosmetics Toxicology, )4, No. 5. (1976) 498. 36. The Determination of Vinyl Chloride. A Plant Manual published by the Chemical Industries Association, Alembic House, 93 Albert Embankment, London S1 7TU. 37. VCiPVC: Health Protection Measures. Verband K unstoSerzcugcnde Industrie e V Frankfurt. 12 November 1974. 38. Colloquc Scientifique sur !e Chlorure de t Inylc Monomere, tenu le mardi 2 Mars 1976, a Hnstitut du Radium Fondation Curie, Paris. Yarstev . In this chapter at directions of the atldirices. riz. phi in a similar u t/r n ciz. polymeric mm UV siabil^rs. The sn^Bsizc approacltinon eti many new technic, to associated ecm the effects of ado The role of addii functions more d1 are the factors hr materials used as and/or proporur tillers. The field h current state of it factors, highly sic The treatment is t R&s 005176 * (3)8 A3b ork, 1976. it 2429777. Huls, X). it 2521780. u 830689. Nara. em. KK, Japanese 2608078. . 779861. 117. i, 25. SIR Modem it 840600. . 87. n on PVC, Lyons, 29547. 22553. ^ A. dtA*. & t'iopfi be f //4 *~vi, . H. M. Clayton British Industrial Plastics Ltd, Darlington, UK SUMMARY Ajter nearly 50 years oj commercial production of PVC, vinyl chloride monomer was found to be a human carcinogen. The following chapter describes how contact with vinyl chloride both in the work place and in consumer applications has been reduced to a level where no major hazard exists to health. Data are presented showing how great the progress has been with the problem since it was first recognised at the end of 1973. The technology used to bring about such results is reviewed. Further progress will be significantly less dramatic. A Ithough the health precautions taken have ted to an increase in the manujacturing costs of PVC, it remains an essential component of present-day civilisation. There is, in fact, an increased awareness of its remarkable versatility. 3.1 INTRODUCTION Some future historian of the chemical industry will, in retrospect, point to the vinyl chloride (VCM) problem as a landmark in the progress of its offshoot, the plastics industry. Long periods of exposure to a substance hitherto regarded as relatively medically safe led to the appearance of a number of occupationally related diseases, including a rare form of incurable liver cancer--angiosarcoma of the liver. The intensive investigations set in train by this discovery are by no means complete--nor is the development of improved manufacturing methods and procedures to reduce still further the traces of VCM escaping into the 43 44 H. M. CLAYTON R&S 005177 atmosphere (in polymerisation and fabrication plants) or, migrating into foodstuffs (albeit in minute quantities). Medical research programmes continue into improved and early diagnosis of VCM-related diseases as well as into understanding the mechanisms of its various effects. In spite of the fact that any report on the problem must, to some extent, be regarded as interim, the PVC industry can point to considerable progress made during the last three years in solving the various technical problems encountered in reducing any possible hazard to a socially acceptable level. The author has been actively involved in one of the UK Working Parties dealing with the VCM problem and has first-hand experience of the extensive co-operation among companies (both national and international), trade unions and government agencies in order to bring about a speedy solution to the various problems. In spite of the differences of detail, broadly similar approaches to problem resolution have been employed around the world. There are useful accounts of the problem seen through German eyes,37 and through American eyes:11 the latter includes a chronology of key events. 3-1.1 Background to the Problem Large scale production of PVC resins by the polymerisation of VCM started in the 1930s. Although the hazards of fire, explosion and narcosis were well recognised, there was no evidence to suggest that VCM constituted a health hazard; indeed it had been suggested that it would have been adopted as an anaesthetic, but for the fire and explosion hazard. The first work suggesting a potential for carcinogenicity came from studies carried out by Viola and co-workers in an attempt to reproduce acro-osteolysis in rats.1 (Suggestions that Russian workers had observed liver damage amongst polymerisation workers as early as 1949 are unfounded.3 The study related, in fact, to 48 workers compounding PVC with chlorinated diphenyl). Acro-osteolysis, Raynaud's Syndrome and scleroderma had been observed amongst some PVC polymerisation workers, particularly those involved in the manual cleaning of deposits from autoclave interior walls. The act of removing deposits of encrusted PVC resin releases unreacted VCM and it is this VCM which presumably causes diseases such as acro-osteolysis--a vascular disturbance. The clinical findings may be, for example: 1. Increased sensitivity to cold, 2. Pins and needles and other paraesthesia, and 3. Changes in the colour of the skin of the digits. TOXICITY AND VINYL CHLORIDE 45 These are the signs of a disturbance known as Raynaud's Syndrome. Another feature of the acro-osteolysis may be changes in the bone pattern of the terminal phalanges, often accompanied by deformation of the nails. Scleroderma is the medical term for the thickening of the skin which can occur as a result of contact with VCM, and it can lead to gross deformation of, for example, the thumbs. In studies reported in 1971^Viola showed that exposure of rats to atmospheric concentration as high as 30 000 ppm (parts per million) failed to produce acro-osteolysis. However, malignant tumours were found. The level of exposure of the rats waf so high that its significance to human beings was not dear and further work was necessary. A consortium of European interests, including IC1 Limited "(UK), Rhone-PoulencPolymeres (France), Montedison (Italy) and Solvay (Belgium) commissioned Professor Cesare Maltoni to carry out further studies. Cancers were confirmed in animals at exposures as low as 250 ppm, including angiosarcoma of the liver.1 Maltoni's findings led to an epidemiological survey being conducted in the USA on men who had previously worked in a PVC polymerisation plant (B. F. Goodrich, Louisville, Kentucky). This study revealed a cluster of cases of the exceedingly rare form of cancer, angiosarcoma of the liver. The results of this study group (by Creech and Johnson) were transmitted to the National Institute of Occupational Safety & Health (USA) oh 22 January 1974, the Department of the Employment (UK) on 23 January 1974 and the Acting Chief Employment Medical Adviser (UK) on the next day. Other national governments were notified during the same period. In the UK on 29 January 1974, ICI issued a press statement announcing that government departments, the TUC, its own workers and customers, were being informed of the facts available so far concerning the urgent investigations into the cause of death of a seventy-year old retired autoclave worker. It was subsequently confirmed that the worker, who had died in 1972, was suffering from angiosarcoma of the liver. During his working life he had undergone twenty years exposure to VCM. In February 1974 the Chemical Industries Association in the UK formed a Vinyl Chloride Committee to study the whole problem and institute necessary programmes of work through a series of specialised working parties. Similar arrangements were made in other countries of the world. In Europe international co-operation was ensured through CEFIC (Conseil Europeen de Federations de I'lndustrie Chimique) which formed its own Vinyl Chloride Committee with national representatives taking decisions on behalf of their colleagues. 46 H. M. CLAYTON 3.1.2 Extent of the Vinyl Chloride Problem At the end of 1976 the number of notified angiosarcoma deaths was fiftyfive. The distribution of deaths is shown in Table 1. The average length of exposure before death was 15-20 years. Almost without exception'the deceased had worked on polymerisation plants and had been engaged in the manual cleaning of autoclaves during part of their service. Certain plants show a clustering of deaths.6 TABLE 1 NUMBER OF NOTIFIED DEATHS FROM ANGIOSARCOMA (UP TO THE END OF 1976) Country Number of deaths Belgium Norway UK Italy Sweden Eastern Europe France Germany Canada USA I 1 2 3 3 4 6 7 10 18 Crampton6 and other authors* describe other disorders caused by VCM. These include acro-osleolysis (approximately sixty-five cases in the world--all autoclave cleaners), associated Raynaud's Syndrome and scleroderma, and liver dysfunctions. This latter group of disorders has sometimes been referred to as'PVC Disease'.7 Most of the workers suffering from this group of disorders were polymerisation workers and particularly those manually cleaning autoclaves. Whilst some disagreement exists on the severity of the effects of the other groups of disorders, angiosarcoma of the liver is invariably fatal. Studies on the carcinogenicity of VCM, whether by inhalation or ingestion, will not be complete for some considerable time. Until a medically proved acceptable exposure level is established, the object of 005178 TOXICITY AND VINYL CHLORIDE 47 much of the work has, during the last three years, been to reduce possible exposure to VCM to as near zero as possible. 1 In the meantime, epidemiological surveys of polymerisation workers j have been carried out in the UK, the USA and elsewhere to establish the 1 . possible extent of the problem. ; : Other work has been directed at improved diagnostic methods to give I advanced warning of problems.4,6 | Yet further work has been aimed at understanding the mechanism of the i way in which VCM causes angiosarcoma of the liver. Tentative hypotheses have been advanced involving the role of metabolites of VCM, such as chlorethylene oxide and modification of DNA molecules."15 Although no final conclusions can yet be drawn from the medical work ri; carried out, some authorities have attempted an assessment of the risk from VCM in the light of recent performance. The results are encouraging.9,10 Whilst there must inevitably be an element of speculation in assessing all medical work, the results of work by the PVC industry are demonstrable. These will be described in the following pages. 3.1.3 Sources of Interface Between Human Beings and VCM To understand the problem and the progress made in containing it, a brief ; account of the processes for the manufacture of PVC resins, compounds j and finished articles is necessary. / | VCM is a gas boiling at - 13-5 C. It is therefore necessary to polymerise it under pressure. Commercial resins are made generally by polymerising 70-90% of the VCM. This leaves a quantity of unreacted gas within the autoclaves as well as within the particles of the PVC resins. Complete reaction of the monomer is not feasible. If made by the suspension or emulsion processes (accounting between them for the majority of world PVC production) it is necessary to separate the water present, for example 1 by centrifuging and drying. All these processes lead to the release of further : quantities of VCM. Finally, the resin itselfwill still contain small quantities of unreacted monomer, particularly those resin types with dense glassy particles used for direct powder blend processing, e.g. in the extrusion of ! water pipe or extrusion blow moulding of bottles. ; Subsequent processing operations involving the use of heat will lead to 1 further releases of VCM. Small quantities remaining in the finished articles : wilt be released by slow diffusion. In the case of packaging materials, very j smalt quantities of VCM may be transferred into foodstuffs and beverages. There is also the possibility of transfer of very small traces of VCM into drinking water supplied through PVC pipes. i 48 H. M. CLAYTON R&S 3.2 DEVELOPMENTS IN REDUCING CONTACT WITH VINYL CHLORIDE The preceding section identified the possibilities ofexposure to VCM in the following areas: 1. Polymerisation and associated operations, 2. Storage and handling of PVC resins, 3. Hot processing of PVC compounds, 4. Fabrication of finished products, and 5. Diffusion from finished products, including migration into foodstuffs and drinking water. Before considering the developments that have taken and still are taking place, some estimate of exposure levels is appropriate. Before the TABLE 2 VCM LEVELS AROUND POLYMERISATION PLANTS Year Typicalaverage level (ppm j 1945-1955 1955-1960 1960-1970 Mid-1973 Mid-1975 End 1976 500-1000 400-500 300-400 150 5 2-5 realisation that excessive exposure to VCM could lead to crippling or even fatal illnesses, little reliable data on atmospheric VCM levels around polymerisation plants was available. Table 2 gives such information, and is based on more recent measurements and earlier estimates. Levels encountered in the fabrication and end use sectors are given in Table 3. TABLE 3 TYPICAL RECENT VCM LEVELS (PPM) Location January 1974 July 1975 December 1976 in plant atmospheres 2-15 <2 <1 In PVC bottles -- 50 ~2 <! In beverages 01 001 <0 005 TOXICITY AND VINYL CHLORIDE 49 Estimates of the intake of VCM have also been made;4'5 these are shown in Table 4. As expected, the figures show that the intake of VCM by polymerisation plant workers was by far the most significan t. As will be shown, many of the steps taken to reduce the exposure of polymerisation workers to VCM led to improvements in downstream activities. TABLE 4 ESTIMATES OF DA* ILY VCM INTAKE Type of person Daily VCM insake {glkg 'body weight) Polymerisation worker at 1000 ppm 036 500 ppm 0-18 5 ppm 0 001 8 2 ppm 00007 Fabrication plant operator ' 1 ppm 0 0004 European citizen--through ingestion of food 0-000 000 002-0 000 000 008 // 3.2.1 Polymerisation Plant Developments The level of VCM permitted in the atmosphere about polymerisation plants, the absolute quantities which may be emitted into the atmosphere, the quantities which may be present in waste water and in some cases the quantities remaining in PVC resins, are the subject of various compulsory national regulations or voluntary codes of practice and hygiene standards. They vary in detail and in the methods of calculating certain required numerical data, in particular average and peak personnel exposure results. These apparent differences obscure the general similar level of achievement in most countries of the world. Typical average levels of atmospheric VCM about PVC polymerisation plants are now 2-5 ppm; this has been achieved by improving the operating procedures in a number of ways: 1. More efficient degassing of the PVC resin, either in the polymerisation autoclave or downstream from the autoclave, resulting in lower losses at the drier stage. Much effort has been H. M. CLAYTON R&S 005180 expended on the development of improved degassing systems and a number of patents have appeared.19-23'31'32-33 The B.F. Goodrich process for counter-current stripping of resin slurry has been licenced to a number of other PVC resin producers. Development of automated high pressure washing methods for autoclaves so as to avoid the necessity ofentry and manual cleaning. Development of autoclave surface-treatment systems which rediice the tendency or crusts or skins to form, so reducing the amount of cleaning necessary.12''3-18-20-21'24 The Shin-Etsu Company of Japan and the B.F. Goodrich Company of the USA have licenced their technology to a number of other PVC resin producers. Other companies have reported the use of additives in polymerisation recipes which reduce encrustation.14-12-34 Remote operation ofautoclaves which may often be controlled by a computer. This technique will grow in importance as a result of recent spectacular explosions, such as that at Flixborough. The use of larger autoclaves, so reducing the number of flanges and valves, etc. for a plant of given output. Such developments were under way before the health problem of VCM was appreciated. The largest reported commercial autoclaves are those operated by Chemische Werke Hiils--200 m2. B.F. Goodrich operates a plant at Louisville with autoclaves of 62 m3. The Shin-Etsu Company of Japan uses autoclaves intermediate in size at 130 m3. All three companies have licenced their larger autoclave technology. Development of recipes and agitation systems yielding more uniform resin particles. Uniformity of resin particle structure is a requirement of the industry, irrespective of the VCM problem. Added impetus has been given to the work since a more uniform particle structure will lead to more uniform degassing. Especial priority has been given to investigations into the production of resins for such applications as water pipe and bottles which normally start from a powder blend. Processing rates are influenced by bulk density and particle shape. The problem is to produce resins with a regular shape and porous structure and at the same time an acceptable bulk density. Some progress has been made towards this objective. Use of respirators by personnel when carrying out operations in areas with unacceptably high atmospheric levels of VCM. Development of automatic sensitive monitoring systems scanning all working areas. TOXICITY ANO VINYL CHLORIDE 51 Outstanding problems still exist, notably in reducing losses from certain downstream stages after the autoclaves. The Environmental Protection Agency in the USA has set very stringent requirements which may prove difficult to achieve.28 Abramowitz30 gives a useful account of progress in mass polymer plants. Some quality problems have inevitably arisen as a result of producers finding themselves forced to bring about rapid improvements with existing equipment or incompletely developed modifications. These considerations, as well as long delivery times of some equipment, have forced companies to use ad hoc interim arrangements. For.example, reductions in residual VCM have been effected by the expedient of more severe drying conditions leading to poorer resin colour and stability, poorer flow in hoppers (due to static electricity effects) and reports of different fusion characteristics (probably attributable to subtle surface changes due to the higher drier temperatures). The cost of modifying some plants has resulted in their closure. Reduction of approximately 10% of plant productivity has occurred in many cases, and production costs have increased by an average of 10-15%. On the other hand, VCM utilisation efficiencies are markedly improved. A further benefit is that the reduced necessity to enter autoclaves for cleaning has improved working conditions. / 3.2.2 Reductions of VCM Exposure in Handling, Processing and Fabrication The preceding section refers to steps taken to reduce residual VCM levels in PVC resins. Some resin particles are more difficult to degas effectively than others. For rigid extrusion, extrusion blow moulding and injection moulding operations (using a powder blend feed rather than a fully compounded granule), dense regular particles are required in order to obtain good powder flow and efficient packing of screws. Such PVC resins retain VCM more tenaciously than those with an open particle structure used for the manufacture of plasticised compounds and plasticised finished products. In early 1974 it was common for such resins to contain 1000-2000 ppm residual VCM. In some cases even more was retained, e.g, in homopolymers made by the mass polymerisation route, and acetate copolymers. VCM is slowly released from the polymer at room temperature. At elevated temperatures such as those encountered during processing the release is extremely rapid. In the USA, there are reports of VCM accumulating in the free head space of rail cars until the levels reached are 52 H. M. CLAYTON potentially capable ofexplosion. In Europe, fires have occurred at the vents of extruders processing rigid powder blends.16 In the early part of 1974, extensive surveys of more than 100 companies were carried out in, the UK, in order to establish the levels of VCM encountered during processing and handling operations. The findings11 are shown in Table 5. TABLE 5 ATMOSPHERIC VCM LEVELS AROUND PVC PROCESSING AND FABRICATION OPERATION (PPM V/V) Location Range Average Polymer warehouses <2-17 Cold mixing equipment <2-3 High speed mixing* <2-53 Banbury type mixer <2-13 Two-roll mill compounding <2-2 Paste mixing <2-16 Calendering <2-3 Fabric and paper coaling <2-5 Pipe extrusion <2-7 Cable extrusion <2-4 Injection moulding <2-4 PVC welding <2 Vacuum forming <2 Bottle blowing <1-2 Extrusion of flexible film <1-2 Extrusion of rigid sheet <1-4 2 <2 3 2 <2 <2 <2 <2 <2 <2 <2 <2 <2 <1 <1 <2 * Single result of 53 with high acetate copolymer at bin filling point. Most of the results were of a low order, averaging 2-3 ppm. However, certain unacceptable practices came to light. High speed mixing areas were particularly prone to unacceptably high atmospheric levels. This was not surprising since at temperatures in excess of 110C, PVC resins release VCM very rapidly. Improved operating procedures, such as the installation of localised extraction, the venting of extruders and mixers to the outside of buildings, and good general ventilation in warehouses etc., resulted in levels below 2 ppm by early 1975. By the latter part of 1976 the effects of reducing R&S 005181 TOXICITY AND VINYL CHLORIDE 53 residual VCM in PVC resins have resulted in levels generally below 1 ppm with many results at 0I-0-2ppm; near the limit of detection of the analytical methods used. 3.2.3 Potable Water Supply Virtually all potable water pipe is produced by extrusion from powder blend. Although some differences in formulation practice have developed with, for instance, Europe favouring the use of solid lead salts as stabilisers and the USA favouring the use of liquid tin compounds, a common feature of the pipe-making industry was the increasing use of high bulk density resins to increase extruder outputs. ! As reported earlier, such resins tend to retain residual VCM more tenaciously, and levels of 1000 ppm or more were not uncommon. j: As well as the necessity to reduce residual VCM levels in new pipe production, measurements on drinking water passing through existing pipe systems were required. At one location in the UK, the Chemical Industries Association was able to carry out measurements on drinking water supplied entirely through a rigid PVC pipe network manufactured by one company. The pipes were 102 mm (4in) and 152mm (6in) in diameter. Even in those parts of the network where PVC resins containing 1000 ppm residual VCM had been employed to manufacture the pipe, no VCM could be detected in the drinking water. Further studies were carried out on houses connected by small bore PVC and chlorinated PVC pipe. Here, a more unfavourable surface/volume ratio could be expected to increase the possibility of extraction of VCM; however, only at one location could any VCM be detected. This was in a home built some nine or ten months previously and unoccupied since completion. In the small volume of water which had stood in contact with the PVC pipe, 0-015 ppm of VCM was detected.15 The above results were extremely encouraging. Nevertheless, three lines of attack were used to reduce VCM levels in rigid PVC pipe: 1. Increased use of hot (i.e. > 110 C), high speed mixing, combined with aeration, in the preparation of the powder blends. Some companies had originally only mixed to relatively low temperatures, i.e. 70-80C. 2. Improved degassing techniques during PVC resin production. Typical average residual VCM levels for pipe resins are now in the 5-50 ppm range. 54 H. M, CLAYTON R&S 005182 3. Development of suspension resins of increased porosity. A judicious selection of suspending agent combinations, or other appropriate formula modifications in the case of mass polymer, can produce resins with regular particle shape, significantly increased porosity and only a marginal sacrifice in bulk density. Further developments can be expected by this approach. A combination of the above approaches has led to significant reductions in the residual VCM levels in rigid PVC potable water pipe and some companies are now achieving average levels of less than 1 ppm. Both the Ethyl Corporation in the USA and ICI in the UK have determined the diffusion coefficient of VCM from rigid PVC pipe. The results have been used to predict VCM levels in drinking water supplied to typical households, in the USA and the UK, using pipe with 1 ppm residual VCM.26 For the UK, the level of VCM in the water is predicted to lie between 0-006ppb and 0-022ppb (b = 1000 million). In the USA the predicted values are 0 01-0-14ppb. Not surprisingly, the use of liquid tin stabilisers in the USA in place of solid lead salts in the UK gives a higher coefficient of diffusion. Since most water consumed is used in cooking or in the preparation of hot beverages, with attendant further loss of VCM, the average levels in consumed water are decreased even further. 3.2.4 Food Packaging Any review of the use of PVC for food packaging must, in the light of the VCM problem, take into account both the political background of legislation and the variations in usage in different countries. In the USA, the Food and Drugs Administration (FDA) is constrained by precedents set in the interpretation of the so-called Delaney Clause of the Food, Drug and Cosmetic Act of 1958 which states `No additive shall be deemed to be safe if it is found, after tests which are appropriate for the evaluation of the safety of food additives, to induce cancer in man or animal___ ' At the time of its introduction, the above clause was clearly intended to be interpreted in the light of testing, probable consumption levels, safety margins and reasonable certainty about conditions of use. In practice, precedent has established the clause as in tending to mean a total ban on any additive suspected of causing cancer under any conditions. As a result of this, American legislators are not free to make a truly scientific assessment of a possible carcinogenic hazard. Any regulation TOXICITY AND VINYL CHLORIDE 55 which falls short of a total prohibition is, in fact, a politically influenced decision. In this situation the FDA produced a new form of words in a proposed regulation issued in September 1975. The proposals permitted the continuing use of PVC in contact with foodstuffs `where the potential for migration of vinyl chloride is diminished to the extent that it may not reasonably be expected to becorpe a component of food.' Against this definition the FDA proposed to allow the continuing use of flexible film and hose. Rigid and semi-rigid foils and bottles would have their status of prior sanction withdrawn. Any producer of the latter group of products would then be obliged to demonstrate that they met the FDA criterion. At the time of writing no decision has been taken on the proposals and it is extremely difficult to predict the outcome. In the United States the most important outlet for PVC in food packaging is in plasticised films such as meat wrap. On the other hand, the use of rigid PVC is relatively less important and in particular the use of rigid PVC bottles for food packaging has scarcely developed. Therefore, whatever the decision its economic effect on the United States packaging industry will be relatively small. The EEC is slowly moving towards a unified approach on food packaging legislation, and no matter how much European countries may wish to display their independence of events in the USA, in practice,^packaging legislation in the latter country has previously influenced events in Europe. In the case of PVC and VCM, however, the EEC produced a draft directive at the end of 1976 which takes into account the realities of the situation in Europe.21 Since a `no effect' level has yet to be established for VCM, the proposed directive takes into account the best practical achievements of European industry in reducing VCM levels in PVC food packaging. Before discussing these achievements, some account of the variations in European food packaging practices is necessary to understand the basis of the proposals. Whilst plasticised PVC films are important for food packaging in Europe, the use of rigid PVC bottles or foils is especially important. Certain countries in Europe show marked variations in the pattern of usage. In France for instance the manufacture of rigid PVC bottles for mineral water is a major industry, accounting for 100 000 tonnes of PVC compound in 1976, whereas in Germany, rigid PVC foils account for the major proportion of rigid PVC used in food packaging. In the UK a major proportion of rigid PVC bottles is used for the packaging of fruit squashes. Such squashes arc mainly consumed by young children, a section of the population for whom there is special concern. 56 H. M. CLAYTON There are also some significant differences in formulating practice in Europe. In France and countries subjected to French influence, most bottles are based on tin-free formulations. Such formulations contain significantly higher levels of other liquid additives to help processing. In Germany particularly, tin stabilisers are preferred. Whilst there is a significant usage of tin-free bottle formulations in the UK, the majority are based on tin stabilisers. Most rigid PVC foils in Europe are based on homopolymers, but in the UK a significant percentage of production (circa 75%) is based on acetate copolymers. TABLE 6 VCM LEVELS IN L K-MANUFACTURED PVC PACKAGING MATERIALS VCM location VCM concentration {ppm) January 1974 March 1975 December 1976 Bottles Level in polymer at time of use Typical level in powder blend at time of manufacture Typical level in bottle wall Foil Level in polymer at time of use Typical level in foil Flexible film (extrusion blown) Level in polymer at time of use Typical level in film 500-1000 50 50 100-250 5 3 800-1500 100-250 80 8 400-1000 1 50-150 1 15-50 l <1 50-100 3 5-30 <1 Tabic 6 shows the residual VCM levels in UK-manufactured PVC packaging materials during the three-year period January 1974 to December 1976. It will be noted that in the case of flexible film, residual VCM levels were very low, even at the beginning of the period under review. This is explained by the fact that the presence of plasticiser aids the release of VCM during hot processing. The higher levels of residual VCM in rigid foil compared with bottles is explained by the use of copolymers in the manufacture of the former material. The relatively low softening point of copolymers makes it extremely difficult to achieve low VCM levels either at the resin-making stage or during mixing. For the foreseeable future copolymer foils are likely to contain slightly higher residual VCM levels than bottles or foils based on R&S 005183 TOXICITY AND VtNYL CHLORIDE 57 homopoSymcrs. Since most such foils are cither used for the packaging of solid foodstuffs or materials of restricted shclflife, the slightly higher level in the package does not normally lead to higher levels in foodstuffs. The results shown in Table 6 are also generally typical of the position in Europe as a whole with two provisos: 1. Where homopolymers are used to manufacture rigid foils, residua) VCM levels are lower and more in line with UK results for bottles; and 2. Where tin-free stabiliser systems are used in bottle formulations, the presence of extra liquids and low melting point additives assist the displacement of VCM during processing, and residual levels in such bottles will tend to be lower than in bottles based on tin stabilisers. Some important conclusions can be drawn from Table 6: 1. Flexible film has low residua! VCM levels even at the beginning of 1974. Subsequent improvement has only been marginal; 2. Rigid PVC has shown a marked improvement, especially during the period January 1974-March 1975; 3. Since current USA achievement does not differ significantly from that in Europe, the logical course for the Food and Drug Administration would be to accord rigid and semi-rigid PVC the same status as flexible film, i.e. continuing prior sanction. 3.2.5 VCM in Packaged Food Numerous studies have been carried out on the migration of VCM from the package into the contents. In the UK regular surveys have been carried out on random samples from retail outlets. The results of this large body of continuing work may be summarised as follows: 1. Under ambient conditions rigid PVC packages lose part of their residual VCM to the atmosphere; after a year approximately 30 % has been lost in this way;- 2. VCM reaches an equilibrium concentration in the contents of packages after several months; 3. This equilibrium concentration allows typical partition concentrations to be calculated. For foodstuffs commonly packaged in PVC in the UK the concentration of vinyl chloride remaining in the PVC is 200-700 times that present in the contents; 58 H. M. CLAYTON R&S 005184 4. Only very small proportions of the VCM in a package may transfer to the contents and for a wide range of foodstuffs the levels of extraction are similar; and 5. Under certain circumstances small numbers of packages may remain on retail shelves for far longer than the recommended shelf life of the contents. The use of the partition data enables calculations to be made of expected maximum VCM contents of foodstuffs packaged in rigid PVC. Extensive retail surveys have provided firm support for such predictions. At the end of 1976, VCM levels in foodstuffs packaged in the UK were in the range 0 003-0 005 ppm or less. Work carried out both in the UK and in Continental Europe has shown that the equilibrium partition concentrations with tin-free bottle formulations such as those typically used in France are not as favourable as those with tin stabilised formulations. For some foodstuffs the ratio of concentration of vinyl chloride in the bottle to that present in the contents may be as low as 100. This disadvantage is, to some extent, offset by the fact that VCM tends to be displaced more readily during hot processing, giving slightly lower initial levels in the bottles.18 Gilbert29 also has suggested that at low initial concentrations in bottles and other rigid containers, extraction of VCM by foodstuffs will be less than expected because of the presence of active sites in the matrix of the PVC which retain VCM. A further factor influencing the EEC proposals is the state of development of suitable analytical techniques. With some foodstuffs problems of interference exist. Also, even in the hands of highly skilled research analysts, it is very difficult to get agreement between different laboratories. Finally, the manufacturer ofthe PVC container or foodstuff must be able to observe trends in the VCM content of his products and take corrective action. For this to be possible, the maximum allowable VCM must be rather more than that at the limit of analytical detection. Against this background the EEC has produced its draft directive. The main features of the directive are: 1. A limit of 0 050 mg/kg in the foodstuff (i.e. 50ppb); 2. A limit of 1 mg/kg in the container (i.e. 1 ppm); and 3. In the case ofcontainers made from copolymers, a limit of 5 mg kg, provided that they are not used for liquid foodstuffs. TOXICITY AND VINYL CHLORIDE 59 It is too early to predict the success of the proposals. However, similar requirements or recommendations already exist in some European countries. For further information on VCM and PVC packaging the reader is referred to the publications of GECOM (Groupe d'Etude Pour le Conditionnement Moderne). GECOM was founded in France and is an association of PVC producers and the manufacturers of packaging and foodstuffs, including water, oil and wine. It was set up to study the various problems in the packaging field raised by the discovery of the toxicity of VCM. The address of GECOM is 11 Rue Margueritte, Paris i 7. One of the GECOM publications gives a very hseful account of the problems of analysis of foodstuffs containing vinyl chloride.28 3.3 ANALYSIS AND DETECTION OF VCM The progress described in the preceding pages would not have been possible without significant advances in the methods of detecting VCM. Developments have taken place in three major areas: 1. The monitoring of atmospheric levels of VCM in the vicinity of polymerisation plants; 2. Analysis of VCM, often present only in minute traces, in PVC resins and compounds, food packaging and foodstuffs; and 3. Small portable detectors. 3.3.1 Monitoring of Atmospheric Levels Prior to 1974 the measurement of VCM was a time-consuming process and the methods required development so that they were capable of yielding accurate results at great speed and were suitable for the sequential scanning of working areas on polymerisation plants. Two methods of analysis have gained particular favour for the task: 1. Gas chromatography combined with a flame ionisation detector; and 2. Infrared absorption. In the first method samples of air'are drawn through a suitable column which separates VCM from other possible airborne contaminants. The VCM fraction is burnt in the presence of pure hydrogen, forming ions which are then measured by an electrical technique. This method is very sensitive and can measure fractions of I ppm of VC M. The second method depends upon the fact that VCM absorbs infrared radiation at a number of wavelengths which are characteristic both of the 62 H. M. CLAYTON 7. VCjPVC: An Example ofa Problem Resolved, Verband Kunststofferzeugende Industrie e V Frankfurt am Main, West Germany, 15 September 1975. 8. Williams, D. M. J,, et al. Bril, J, Indus, Med. 33, (Aug. 1976) pp. 152-7. 9. Wilson, R. Risk benefit analysisfor toxic chemicals: vinyl chloride. Paper given at the 112nd ACS National Meeting, San Francisco, 25 August 1976. 10. Schlatter, Ch. `Gefahrdung von Arbeitnehmer und Konsument durch Vinyl Chlorid.' Press Release by Lonza AG, 29 October 1975. 11. `PVC and Health.' A background statement issued by the Society of Plastics Industry Inc., New York, April 1976. 12. British Patent 1 291 145. 13. British Patent 1 421 718. 14. British Patent 1 444 360. 15. Federal Register, 21 October 1976, 41, No. 205. Environmental Protection Agency, National Emission Standards for Hazardous Air Pollutants, Standard for Vinyl Chloride. 16. Private communication. 17. Clayton, H. M. Guide lines for PVC processors. Paper given at the BPF/PRI Joint Conference Vinyl Chloride and Safety at Work, 28 May 1975. 18. Belgian Patent 833 558. 19. Belgian Patent 831 964. 20. Belgian Patent 833 041. 21. British Patent 1 407 665. 22. Belgian Patent 837 070. 23. British Patent 1 446 583. 24. US Patent 3 669 946. 25. Moffett, T. W. Private communication. 26. Moffitt, T. W. Private communication. 27. Document 1565/VI/76-F. 28. Tester, D. A. and Moffitt, T. W. Vinyl chloride monomer in food packaging. Paper given at the BPF/PRI Joint Conference Vinyl Chloride und Sa/etv at Work, 28 May 1975. 29. Gilbert, S. G. `The migration of minor constituents from food packaging materials.' J. Food Sci., 41, 1976, 30. Auramowitz, R, J. `Bulk polymerisation plants.' Paper given at the 112nd ACS National Meeting, San Francisco, 2 September 1976. 31. Unexamined Japanese Patent 47 076 (1976). 32. Uncxamined Japanese Patent 47 081 (1976). 33. Unexamined Japanese Patent 17 950 (1976). 34. Unexamined Japanese Patent 17 287 (1976). 35. Food and Cosmetics Toxicology, 14, No. 5, (1976) 498. 36. The Determination of Vinyl Chloride. A Plant Manual published by the Chemical Industries Association, Alembic House, 93 Albert Embankment, London SE1 7TU. 37. VC/PVC: Health Protection Measures. Verband Kunstoflerzeugende Industrie e V Frankfurt. 12 November 1974. 38. Cotloque Scientifiquesur le Chlorure de Vinyte Monomere, tenu Ie mardi 2 Mars 1976, a l'lnstitut du Radium Fondation Curie, Paris. Yarsley I In this chapter an directions of the additives, viz. phi' in a similar way to viz. polymeric mo. UV stabilisers. /. The sheer size approach; Itoneo many new technic, to associated cent, the effects of add. The role of add it functions more d: are the factors lar materials used as and/or proportio fillers. The field is current state of it factors, highly sig The treatment is t CONFIDENTIAL REGISTER OF CASES OF ANGIOSARCOMA OF THE LIVER DUE TO EXPOSURE TO VINYL CHLORIDE - REVIEW DATE 1 JANUARY 1986 PREFACE : Acknowledgments, significant changes and. comments 1. The new cases in the Register are the following : France Germany UK 18 23, 2k, 25, 26, 2? 11, 12 2. There are amendments from the previous Register against the following cases: Germany 21, 22 US 36, 38 3. I would like to thank Dr De Sloover, Dr D H J Williams and Dr Friedel for informing me of new cases. k. For the case corrections I would like to thank Dr Friodol, Hr J T Barr and Hr R W Frase. 5. For other corrections and information, thanks to Dr D Forman, Mr J T Barr, Mr T H Sutherland and Mr N F Humphry. 6. I should be very grateful if all recipients of the Register could continue to keep me fully informed of any new cases, or, if they should hear of other Companies who have acquired a case of ASL, it would he much appreciated if I could be notified by the appropriate person. It is extremely helpful if new information is supplied In tabulated form, using the headings In the Register as far as possible. 7. I would like to thank the many people who have suggested research topics, and should be grateful if I could be kept up to date. 8. I shall be happy to supply toany recipient ofthe Register a copy ofthe Paper "Exposureto Vinyl Chloride andAngiosarcoma of theLiver: a report of theRegister of cases" by Dr David Forman,myself, Dr John Stafford and Sir Richard Doll, published in the British Journal of Industrial Medicine, November 1985 (k2; 750-753) Dr Brian Bennett Medical Department Imperial Chemical Industries pic Hillhouse Site P 0 Box Thornton-Cleveleys Blackpool FY5 *tQD England February I986 r R&S 00518^ a() Index IMPERIAL CHEMICAL INDUSTRIES PLC LIVER ANGIOSARCOMA CASES (REVIEW DATE - L JANUARY 1986) WPX/ASL_2 - LANGIOCASES ,4 Table Sequence, Belgium Canada Czechoslovakia France Germany Italy Japan Norway Sweden UK USA Tugoslavia Incidence of Cases per annum (Table 1) Distribution of Cases by Country (Table 2) Page 1 2-3 4 5-8 9-14 15 16 17 18 19-21 22-28 29 30 31 ASL.Case Numbers by year of Death (Table 6) ASL Case Numbers by year of first exposure (Table 7) Western Europe PVC Producers - Start-up Dates vs ASL Cases (Table 8) USA PVC Producers - Start-up Dates vs ASL Cases (Table 9) Canadian PVC Producers - Start-up Dates vs ASL Cases (Table 10) East European PVC Producers - Start-up Dates vs ASL Cases (Table 1L) Japanese PVC Producers - Start-up Dates vs ASL Cases (Table 12) South & Central American PVC Producers - Start-Up Dates (Table 13) Australasian PVC Producers - Start-up Dates (Table 14) Page 42 43 44 45+46 47 4B 49 50 51 Pattern of Cases by Factory (Table 3) 32 Latency Periods (Table 4A) 33 Distribution of Latency Periods (Table 4B-4H) ' 34-40 Annual Geographical Incidence (Table 5) 41 R&S 005187 Page 30 Incidence of ASL Cases Per Annum TABLE 1 (Review Date - I Jan 1986) By date of death 1955 ......................................................... 1 6 ......................................................................................... 0 7 ......................................................................................... 1 8 ............................................................... ....................0 9 ........................................ 0 1960 ................................................................................... 0 1 ....................................................... ................................ 1 2 ........................................................................................ 1 3 ......................................................................................... 0 4 ........................................ 2 5 ................................. 0 6 ........................................................................................ 0 7 .......................... 1 8 ......................................................................................... 5 9 ........................................ 3 1970 ......................................................................................... 2 1 ......................................................................... 3 2 .............................. 5 3 ......................................................................................... 8 4 ................................................. 5 5 ................................................................ 11 6 ..........................................................................................11 7 ......................................................................................... 9 8 ..........................................................................................10 95 1980 ........... i0 1 ......................................................................................... 5 2 ....................................................... ................................ 2 3 ......................................................................................... 9 4 ........... 5 5 .......................... j_2 Total 117 N.B. 1 Does not Include Italian 01 but does include German 03. ,fZ 2 The 1977 total Includes US case 24 which is a Cholangiosarcoraa 3 The 1982 total does not Include US 33 and F 15 who are alive. 4 The 1983 total does not include US 36 who is alive. H)'X/ASL_2 - J,ANCIOCASES,35 R&S 005188 \ \ \ \ Page 31 Distribution of ASL Cases by Country TABLE (Reylew Pate - 1 Jan 1986) 2 USA W Germany F ranee UK Canada Sweden Yugoslavia Italy Czechoslovakia Japan Belgium Horway 35 *** 26 ** 18 *** 11 10 5 **** 4 ***** k* 2 2 2 1 Tota 1 120 * Omit ItOl - atypical case of PVC bag extruder. ** Includes CE03 aerosol can filler. *** Includes case US 2k which was diagnosed as Cholangiosarcoma & US 33, and 36; and F 15 who are alive. **** Includes one monomer worker. ***** includes two monomer workers. l.e Western Europe North America Rest of World 67 45 8 WI'X/ASL_2 - LANCIOCASES.36 Total 120 -- ` R&S 005189 \ Page 32 Pattern of ASL Cases by Factory TABLE 3 (Review Date 1 Jan 1986) (a) W Europe Of JtUe 67 cases in W Europe, 57 have occurred in clusters at 11 factories. Of the 26 German cases, 23 have occurred in clusters at A factories. ASL Cases Dynaroit, Troiadorf, Germany Chloe Chiraie (Formerly RP), St Fons, France BP, Barry, UK Wacker, Burghausen, Germany Kemanord, Sundsvall, Sweden PCUK, Brignoud, France CUU, Marl, Germany ICI, Hillhouse, UK Solvay, Tavaux, France Lonza, Waldshut, Germany Solvay, Jemeppe, Belgium 11 8 8 7 5 5 3 3 3 2 2 57 NB: There are 32 pre 1962 PVC plants in W Europe; 18 of these have had ASL cases (Ref Table 8) (b) H America Of the 45 cases in N America, 38 have occurred at four factories out of a total of 54 factories!- In N America 43 plants have not had ASL cases. BFG, Louisville, US BFC, Shawinigan, Canada UCC, S Charleston Coodyear, Niagara ASL Cases 14 10 10 4 38 MB: There are 28 pre 1962 PVC plants in N America; 11 of these have had ASL cases (Ref Tables 9 & 10) (c) Rest of World The 8 ASL cases are believed to be in l cluster of 4 and 2 clusters of 2 caseB. WPX/ASL_2 - LANGIOCASES,37 R&S 005190 Page 33 TABLE 4A Latency Periods (Years from first Exposure to Diagnosis) (Review Date - 1 Jan 1986) N Europe N America Country Belttiura France Germany Italy Norway Sweden UK No of Cases -2 18 26 4 1 5 11 67 Total Years 35 459 495 74 22 122 254 1461 Average 17.5 25.5 * 19.04 * 18.5 22 24.5 23.09 * 21.80 \ N Rest of World US Canada 35 886 25.31 * 10 204 20.4 * 45 1090 24.22 Summary Czechoslovakia Japan Yugoslavia 2 2 4 8. 30 15.0 39 19.5 85 21.2 154 19.2 W Europe 67 1461 21,80 N America 45 1090 24.22 Reat of World 8 120 154 2705 19.2 22.54 * Of the countries with below world average > 10 cases, France, UK and USA have an average latency period above world average, Germany and Canada WF L_2 - LANGIOCASES,38 f R&S 0051