Document qdrJDm5Qayr59JEDzerwJjLXx

1 Stf/N ^ fH-LRGy NGUJS ftPRli-* 777^5" Vinyl Chloride Tied to Workers' Skin Problems International Medical News Service Vinyl chloride, the industrial monomer that has been implicated in the etiology of angiosarcoma in plastics workers, can also cause skin problems including minor irri tation and acro-osteolysis. Dr. Marcus Key told Skin & Allergy News. Irritation can be caused by the vinyl chloride evaporating from the skin, result ing in a small local freeze. Acro-osteolysis, which has been found mainly in workers who clean the reaction kettles in which vinyl chloride is poly merized. may be a result of both the mechanical trauma of the cleaning and ex posure to vinyl chloride, said Dr. Key, former director of the National Institute of Occupational Safety and Health. Dr. Harold Magnuson. one of the first students of the relationship between vinyl chloride and acro-osteolysis. said in an in terview with this newspaper that the pre cise agent causing the condition is not known. Vinyl chloride is strongly impli cated. but one of several other reaction products that may remain in the lank could cause acro-osteolysis. The incidence of X-ray abnormalities caused by acro-osteolysis was almost 12*/? in a group of 5.011 men who worked in plants that manufacture polyvinyl chlo ride. while a control group of 2.407 men had an incidence of only 3.169?- for such X-ray changes, said Dr. Magnuson, of the University of Michigan School of Public , Health. Ann Arbor. A nodular thickening of the hands and the forehead and blanching of the fingers associated with Raynaud's phenomenon were the dermal manifestations found in patients with acro-osteolysis who worked at the B. F. Goodrich Company plant in Louisville, Ky., Dr. John L. Creech Jr., who has seen most of these patients, told Skin & Allergy News. Dr. V. K, Rowe, of the Dow Chemical Company, Midland, Mich,, told this newspaper that he feels that under present conditions workers in vinyl chloride plants do not face a very great hazard of either angiosarcoma or skin disease. The observations that are now being made reflect the high levels of exposure these workers were subjected to prev iously. Exposure levels are now much lower in most plants making polyvinyl chloride, he said. There have been no significant vinyl chloride related health problems at the Dow plants, where operations have been maintained at relatively low levels of ex posure. Apparently, the protective mea sures already in effect have been adequate to prevent angiosarcoma, acro-osteolysis. and scleroderma, he said. Dr. Creech, who was the first man to recognize the association of the liver tumors in vinyl chloride workers agreed that there is probably little danger for present workers. Both the acro-osteolysis and the an giosarcomas wgre found mainly in work ers who cleaned out the reaction kettles, where the levels of vinyl chloride were in the range of hundreds of thousands of parts per million. Workers in the same job may now face no exposure whatsoever, because they work with a self-contained respiratory system that eliminates all at mospheric particles of vinyl chloride, he said. None of the patients with angiosar comas had skin diseases that could be re lated to their vinyl monomer exposure. Of eight cases of acro-osteolysis observed by Dr. Creech, no patients had any detecta ble liver abnormalities. Becaue of the observed carcinogenic.activity of vinyl chloride and the association between the plastic and various skin dis orders, it seemed likely that an increased incidence of skin tumors might be found in exposed workers. However, no such in crease was found and there were no skin tumors in any of the patients with an giosarcomas, liver changes, or splen omegaly that could be attributed to vinyl chloride. Dr. Creech said. Dr. Rowe f GENC 01H17 v/ 272 journal or thi aoac (Vol. 58, No. 2,1975) INDUSTRIAL CHEMICALS Gas-Liquiri Chromatographic Determination of Vinyl Chloride in Alcoholic Beverages, Vegetable Oils, and Vinegars DAVID T. WILLIAMS and WALTER F. MILES Health, Protection Branch, Health and Welfare Canada, Ottawa, Oh farm, Canada KlA OLS Vinyl chloride in food* is determined by gatliquid chromatograph*' either by direct injee* tion of vinegars and alcoholic beverages or by headspace analysis of vegetable oils. The lower limit of detection it 10--15 ng/ml for direct Injection and 5--10 ppb for headspace analysis. Confirmation by gas chromatography*mas spec trometry, single ion monitoring at nt/e 62, is possible at 50 ppb for cither method. The levels of Wnvl chloride found in food?* packaged in |in]vtiny| chloride Imttlo %cr* 0.0-- l.C) .iS/ml for ulroliolir betcraev-. 0.0--8. i j.g/rnl for *inc* gars, and 0.3--3.3 |>pm for peanut oil. The detection of 10-20 ppm vinyl chloride in alcoholic beverages led to a proposed ban, in the United States, on the use of polyvinyl chloride (PVC) materials in contact with alcoholic foods: however, this ban did not extend to other foods ((1974) Fed, Regist. 38(95), 12931; 38(134), 18684). In order to assess the full extent of this problem, we have developed methodology for and surveyed foods available in Canada which are packaged in PVC bottles. Experimental Reagent* and Apparatus Use distilled-in-glass solvents and verify absence of vinyl chloride by gas-liquid chromatography (GLC). (a) Cat cylinder of vinyl chloride, (b) Gas light tyringe.--25 or 5.0 ml. (cl Reactmnls.--25 ml, equipped with screw caps and Teflon-faced septum! (Pierce Chemical Co., Rockford, IL 61105). (d) Constant temperature water bath,--Should maintain temperature to within ctl'C, (e) Gas chromatograph. -- Varian-Aerograph Model 2100, with flame ionisation detector and 1 mv pen recorder. Operating conditions: tempera tures (C)--detector 270, injection port 220, col umn ambient until vinyl chloride peak has been recorded, then raised to ca 150 until solvent has eluted from column: 6`X Vi' od glass rolumns Received Autust IS, IfTS. packed as follows; column A, 5% OV-101 on 100- 120 meeh Chromosorb W (HP); column B, 5% 0V- 210 on 100-120 mesh Chromosorb W (HP); col umn C. 107c Apiezon M on 60-60 mesh Chromo sorb W (acid-washed). Column D, 6'X 2 mm id glass column packed with 80-100 mo-h Porapak Q, was operated at 130*C. In all cases carrier gas wa nitroccn and flow rate was adjusted so that vine! chloride eluted at ca SO sec. (ft Gar ch,ninnii,g,ni,),-inn ~ r;i{r/,nnir/rr.-- Hf.T'hi Pcrkiu-Fl'i'cr tv.--- r-iinm. 'it (Oupled with Perkm-Elmer Modi I POO gas chro matograph fitted mlh flame ionization detectors Operating conditions: temperatures (*C)--detector 270. injection port 240. column ambient until vinyl chloride peak has hcen recorded, then raised to ca 150 until solvent has eluted from column; carrier gas helium 10 ml/min; 6' X 'A" stainless steel column packed as follows: column E, 37r XE-60; column F, 37< OV-101; both on 100-120 mesh Chromosorb W (HP). Analysis Vinegars and alcoholic beverages.--Invert bottle several times to mix contents, remove cap. take 5 yl aliquot, and recap bottle immediately. Inject aliquot onto column A and if peak is obtained with same retention time as vinyl chloride verify by injecting another aliquot onto column B or C. Quantify by comparison of peak height or peak area with standard solutions of vinyl chloride. Confirm by GLC-mass spectrometry (MS), col umn E, using single ion monitoring (m/e 62). Vegetable oils.--Invert bottle several times to mix contents, remove cap, and quickly transfer 10 ml oil to tared 25 ml Reaetivial. Seal vial and bottle as quickly as possible and determine quan tity of oil transferred by weighing vial and con tents. Heat flask 20 min in constant temperature water bath (75*1*0. Take 1 ml sample of headspace vapor and analyze on column A. Verify positive samples by injecting 1 ml headspace vapor onto column B, C. or D. Confirm by GLC-MS, column E or F, using single ion monitoring (m/e 62). Quantify by comparison of peak height or peak area with control peanut oil spiked with ethanol solution of vinyl chloride. WTT.l.lAl Prepar Vimji t (a) 1 taxed s mcreur^ through with secbkjridi invert content vinyl c lion wt solutior ards b\ (b) i mg) by of etha not vi; i pre=>".r vinyl ' c' me) y : ethane. Determ weight Repb prepare centrati weight t ration: Head ehloridt Etbano volume 0-27r of axd soh gas tigl with ae access tions st dark hi Vine analyst the san ehites 1 lowing senativ as low tion). ; ses on polarit; agree * ng/ml I, HOTEL I Jrrrr.il - 2. pp/Y) I GEHC 01M18 'o, 2,1975) ide 2 ,01 on 100B, 5% OV,'HP); colii Chromo< 2 mm id h Porapak carrier gaa ed so that rometer.-- ectrometer gas chrOdetectors. --detector until vinyl lised to ca in ; carrier iloss steel 7r XE450; -120 mesh et t bottle cap. take ly Inject lined with \* lift lu ll nr C. ' or peak chloride. MS), cole 62). times to transfer ! vial and ine quanand connperature of headV Verify ace vapor jf-C-MS, ing (m/e Height or ked with WILLIAMS A MILES: OLC OF VINYL CHLORIDE 273 Preparation of Standard Solution* of Vinyl Chloride (a) Weigh known volume (12 ml) of ethanol into fared aeptum-eealed 15 ml vial containing 2 ml mercury. Bubble vinyl chloride slowly into ethanol through syringe barrel and needle and vent vial with second syringe needle. When ca 100 mg vinyl chloride has dissolved (5-10 min) remove needles, invert vial to form mercury seal, weigh vial and contents, and determine accurate weight of added vinyl chloride. Dilute aliquots of this stock solu tion with ethanol to give appropriate standard solutions and confirm concentration of these stand ards by GLC comparison with stock solution, (b) Prepare solutions of vinyl chloride (ca 100 mg) by bubbling vinyl chloride into known weight of ethanol (12 ml) in septum-mercury sealed vials not vented to atmosphere. (Cnulion: Avoid excess pressure buildup.) Determine accurate weight of vinyl chloride added by weight difference. (c) Using 10 mi gas syringe, inject 20 ml (ca 55 mg) pure vinyl chloride gas into known weight of ethanol (6 ml) in septum-mercury sealed vial. Determine weight of vinyl chloride added by weight difference. Replicate vinyl chloride stock solutions were prepared by above 3 methods. Vinyl chloride con centrations of these solutions as determined byweight difference agreed within 1% with concen trations determined by GLC analysis. Headspace volume in vials containing vinyl chloride solutions should be kept to a minimum. Ethanolic vinyl chloride solutions with headspace volume to solution volume ratio of 1:10 have ea 0.29k of the vinyl chloride in the heniNpacp. Stand ard solutions of vinyl chloride must be siored in gas light containers. T",sn of mnrniiy si-al i-uupled with septum muj both gs- light seal \nd nadv access to solution Ethanolic \ inyl chloride solu tions stored in septum-mercury sealed vials in the dark have remained stable for at least 8 weeks. ples when they were analyzed on the more polar column (column B), but this did not occur on the less polar columns (columns A and C). The mass spectrum of vinyl chloride (Fig. 2), as expected for a monochloro compound, shows 2 molecular ions in the ratio of ca 1:3 lor the "Cl (m/e 64) and ,JCI (m/e 62) vinyl chloride. GLC-MS can be used for confirmation of vinyl chloride but a full mass spectrum, distinguish able from background, could only be obtained from samples containing >80 pg/ml vinyl chloride (ca 400 ng injected). However, with single ion monitoring at ra/e 62 vinyl chloride could be confirmed at 50 ng/ml (ca 025 ng injected) (Fig. 3). GLC-MS, with single ion monitoring, can be used semiquantitatively and confirmation of vinyl chloride by this method should give values that agree within 10-15% with the GLC determinations. Analysis of 22 samples of the 5 types of alcoholic beverages available in Canada in PVC bottles showed levels of vinyl chloride ranging tft Results and Discussions Vinegars and alcoholic beverages can be analyzed for vinyl chloride by direct injection of the sample onto the GLC column. Vinyl chloride elutes before the solvent peak (Fig. 1C, D), al lowing the chromatograph to be set at maximum sensitivity and a detection limit for vinyl chloride as low as 10-15 ng/ml (3-4% full scale deflec tion). Positive samples are confirmed by analy ses on 2 or more GLC columns of different polarity and the results on each column should agree within 10%, Trace interference (<25 ng/ml) was occasionally detected in blank sam time, mim FIG. 1 <Ui chromatogram* (column A) of vinegar* and alcoholic beveragei; A. control gin; 8, control apple eldar vinegar; C, gin containing 0.5 jig/ml vinyl chloride. 2.5 ng vinyl chloride injected; O, apple cider vinegar con taining 0.1 *tg/ml vinyl chloride, 0.5 ng vinyl chloride in jected. | Indicates vinyl chloride peak. GENC 01141? - 274 journal or thx aoac (Vol. 58, No. 2,1975) 27 B h > i ...i n i . i, l.i, C i.t _____ i-l-j_____i. i m/e i. FIG. 2--GLC'MS spectra of vinyl chloride run on column E; A, vinyl chloride; B, carbon tetrachloride extract of apple eldar vinegar, ca. 400 ng vinyl chloride Injected; C, background spectrum Immediately prior to obtaining pactrum K. from 0 to 1.6 /ig/ml (Table 1). Samples of similar beverages packaged in glass bottles showed no detectable vinyl chloride. Analysis of 28 samples of vinegars packaged in PVC bottles showed levels of vinyl chloride from 0 to 8.4 /tg/ml (Table 1). Vinegars pack aged in polyethylene containers showed no de tectable levels of vinyl chloride. Extraction of the vinegar, in a septum-sealed bottle, with onefortieth its volume of carbon tetrachloride gave a concentration of vinyl chloride in the carbon tetrachloride approximately 16 times that orig inally found in the vinegar. Those vinegars with >5 fig/ml vinyl chloride when concentrated by this method and the.carhon tetrachloride extract analyzed by OLC-MS gave mass spectra which, after subtraction of the background spectrum, were identical with that from vinyl chloride (Fig. 2). Vinyl chloride is best determined in vegetable oils by headspace analysis. Heating the vege table oil to 75C gave about 18TC of the vinyl chloride in the headspace; a change of 1C at this temperature gave a variation of mO.3% in the concentration of the vinyl chloride in the Table l. Levels of vinyl chloride In alcoholic beverages, peanut oil, and vinegar* contained In PVC bottle* Sample Type of PVC No. of bottle samples Range, /.g/ml* Av., PS/m 1 TIME, MIN FIG. 3--GLC-MS spectra (mingle ion monitoring at m/o 2): A. alcoholic beverage, 0.2$ ng vinyl chloride Injected; B, control alcoholic beverage. 1 indicate* vinyl chloride peak. Alcoholic beverages Gin Martini Beaujolais Cognac Sherry Vegetable oil Peanut Vinegar* Apple cider Malt Matt Matt Salad Red wine A A B C D H E E F G 0 G 4 0.21-0.65 0.44 4 0.86-1.60 1.37 4 0.15-0.84 0.60 4 <0.025 -- 6 0.38-0.98 0.66 10 0.3-3.29* 2.16* 13 0.56-8.40 3.49 4 0.16-2.28 1.86 _7 0,`1.5* 1 0* _2 O' 1 O' -- * Analyzed on column A, average of duplicate injec tion*. * Value* expressed a* ppm. * No vinyl chloride detected, detection limit 0.01 *g/ml. * Single positive, 6 negative. williams * m; A FIG. 4 Gas ch headspace anai* containing 0-32 ride inject headspace. E< reached and in the beadsp: at periods fre from 50 ppb of vinyl chloi for the head' (5-50 pd) dc cenlage of vi determinatioi with the sam alcoholic bev Porapak colt the 1 ml in somewhat br headspace sa vinyl chlorid parable vines GENC 011420 1975) leg of lottles kaged iloride packto deioD of h one; gave i arbon origt a with J :ed by t xtract which, :trum, i nloride l f getable ! vege- e vinyl 1C at .3% in in the Stic Av.. Hg/ml , 0.44 I 1.37 I 0.60 I 0.66 3* 2.16* : I 3.49 i I l.K t injec- .01 Mt/tnl. WILLIAMS * MILES: GLC OF VINYL CHLORIDE 275 FTO. 4-- 4n chromitofranit (column A) btolnod from htd*tci mal/tii of poanvt oMt A* control; B. pttnut oil containing 0-32 ppm vinyl chlorldo, c 23 ng vinyl chlo- rldi (nfoctod, i Indicate* vinyl chloride peek. headspace. Equilibration of the system is rapidly reached and the concentration of vinyl chloride in the headspace remains constant when analyzed at periods from 5 min to 2 hr. Peanm oil spiked from 50 ppb to 20 ppm with ethanol solutions of vinyl chloride give a linear calibration graph for the headspace analysis. The ethanol added (5-50 fill does not detestably affect the per centage of vinyl chloride in the headspace. GLC determination and confirmation are carried out with the sime column used in the vinegar and alcoholic bcver.ure analysis and. in addition. a Porapak column can also be used. Because of the 3 ml injection volume the GLC peak i somewhat broader (Fig. 4B) but since the 1 ml headspace sample contains about 14 times more vinyl chloride than a 5 /d sample from a com parable vinegar the actual limit of detection is slightly better (5-10 ppb). GLC-MS (single ion monitoring m/e 62) can be used down to the 50 ppb level (about 3.5 ng injected) but GLC-MS (full spectrum) is not very useful since the air present in the headspace sample tails into the vinyl chloride peak and interferes with the mass spectrum below m/e 40. A scan over the range m/e 50 to m/e 70 can be used but is at least an order of magnitude less sensitive than single ion monitoring. Peanut oil is the only vegetable oil packaged in PVC bottles in Canada and analysis of 10 samples showed levels of 0.3-33 ppm vinyl chloride (Table 1). Analysis of a control sample of peanut oil shojved a trace peak (about 15 ppb) with a retention time similar to vinyl chloride (Fig. 4A). Other vegetable oils (corn, soybean, papeseed, olive), packaged in glass bottles, contained no detectable levels of vinyl chloride except for samples in glass bottles with caps containing a PVC liner. These samples showed trace peaks (about 5-10 ppb) with the same retention time as vinyl chloride but the levels were too low to confirm by GLC-MS. The problem of vinyl chloride contamination is complicated by the use of a different type of PVC resin by each bottle manufacturer. Eight types of PVC bottles, varying in shape, source of resin, and/or formulation were encountered in this survey but in no case were 2 different foodstuffs packaged in the same type of bottle (Table 1). It is, therefore, not posrihle to assess which of the food-tuffs is most susceptible to vinyl chloride contamination. However, it ran hr seen that the problem is not re-trieted m alco holic foodstuffs packaged in PVC containers. Acknowledgment We thank B. J. Blanchfield for technical assistance. r<r=^) THE JO U R N A L O f COM M ERCE _______ W a d n e td a y , A p ril 16, 1975 , Plastics Engineers *To Hear FDA Director Dr. Q*rks P. Jellaek, di rector o* the U. 6. Food end Drug Atknkri*tratkm'* Dlvi- ion of Cbonical Tednolocr, rill be a feature speaker at a special session' dealing with v 1 o y 1 chloride monomer (VOf) at the Society of Fla*tics Engineers' Annual Tech nical Conference (ANTEC). The ANTEC will be held at the Atlanta Marriott Hotel in At lanta, May 5*8, 1975. Dr. JeHnekwS speak during a 3 p.m. panel discussion on Tuesday, MayC. At the ANTEC, represents- ; tive* Atnh the industry will also be on the panel with Dr. Jelinefc andwill discuss the ef fects of (be regulations on present and future FVC mar ket*. Stringent monomer-ex posure level*, promulgated by the Department of Labor's Occupatkxial Safety and Haalth Administration (OSHA), went into edtect on April L The low worker-exposure levels aifeet companies manufacturing VCM or processing it Into pol yvinyl chloride (FVC), and those who further process FVC into products. The ettitroversy over the regulations was recently fueled by Dr. Herbert E. Stockinger, chief of the tox icology branch of the National Institute for Occupational Safety and- Health (NIOSH) and1 the agency that estab lishes industrial exposure lev els for OSHA. Dr. Stockinger has described the govern ment's regulation that work ers be exposed to virtually no vinyl chloride a* "an irration al decision," and that he was "consternated" at NIOSH's "essentially z e r o-tblerance" recommendations to OSHA. OcctlPATtONPU. SfiFJST'i f- #fU.TH RBPORTRF. '///7/7& California CAL/OSHA APPEALS BOARD OVERTURNS VC HAZARD CITATION FOR LACK OF PROOF SAN FRANCISCO -- (By an OSHR staff correspondent i -- In what is believed to be the first contested case in L'ahtornia involving alleged vinyl chloride exposure, the state Oc cupational Safety and Health Appeals Board overturned a citation because of lack of proof that harmful exposure ac tually existed. The case involved Safeway Stores. Inc., Store =210. Sacramento (Docket No. 8861. and the board unanimously adopted the proposed decision of presiding hearing officer Paul F. Dauer in the matter. Safeway was cited for an alleged nonserious violation of 8 CAC 5142 ic) for failure to provide ventilation adequate to disperse vinvl chloride fumes generated bv a meat wrapping operation. The Division of Industrial Saletv enpinppr whii made the inspection testified that he observed fumes rising through a meat wrapper's breathing zone. He said the fumes arose when the wrapper severed the wrap with heat wire. However, no_mgasurements were taken to determine Hie concentration of the Himes or even that they were, m iarl. vinyl chloride fumes. Safeway told Dauer that an exhaust fan did exist four to five feet above the meat wrapping machine. The firm also noted that there have been no complaints received Iroin the meat wrapping workers of any discomfort experienced at the work station. Dauer noted that the division failed to provide any compe tent, objective, or scientific evidence of any harmful ex posure to the fumes. Not only does the division's case fail to establish the presence of harmful quantities of fumes, it fails to establish that any fumes that were present were harmful, he said. The violation was set aside. No penalty was proposed by the division. GENC 011422