Document 659DBpj0Rn9xXO5d1BnqboJ1d

JOUHNAI. ()!' .'OI.YMlilt NClIvYClit I'ART A-1 VOI,, it, ItUtT-lllW (1970) Ilij'li-TcnijK*ralui*c pyrolysis of I*oly(vinyl chloride); Gas Clu'oiiiatogi*a]>iiic-Ma.s^ Sjtcclromclric Analysis of llic Pyrolysis Products from PVC ; Kcsin and PiaslisoU \ ` MICIIAKJ. ^1. O'.MAHA, It. h', (iwHirirh Chemical Cmh/kihii, \ Amm ImIc, Ohio 440tJ Synopsis A pyruiymii gun rliroMiiiiutfrnpliii* uun* iy in'lnmii'ilit* tt'i'liui.pii* IIir Itiiiuyxmg lira pyrolysis pi-mlurl* Irom imlyiiii'i* in mi inrrl uliiitti.|>lu-n* in iIi-'ti'iImiI, 1 nil in) stuilii-* i*hi*iniipiiMfiiiiit tin* pyrolysis iit |n>iv(vnivl rlilmiili*) liiinio|nilvmrr nun u siu-ir* ol l'VC plnslisols (hnn*tt on i>-j ill l luiinU! ikIi'im) liuvr prnviiu'.l u ininjtii'lr 1,i IIIi i I III IVV milt *1**1IIitpimiiitnlivi* tmiilysis ol lira pyrolysis ptishit'is tiiwi llu*sc iimioriiils, l'VC It**iii yii'ltls n suries ul ;1111111111 ir unit nroiiml ii* liytlroriirisMis w in'll pyrolysis I nl CiOO'C; l in* miiuiiul ul urihiinI ii- |irmiiii'i m in gu-ulor limn llic iini.miii ul iili|iluiiii* pnslnrl*. Beinonr is llic iniijiir nr|;mui* liruriiiliiliini im'ihIiii'I. A ly|iirl l'VC plnsl ini11 I I'Vl '/o-liilM-lyI plillinlnlo (100. 00)| yirl.i-i, upon jiyrulysis, pririliit'ti* I lint Itro riiiiriifti'nni ii* ol limli llic l'VC mnlrix mul llie pliilinliiio jiliunirim*r. Tlnj pyrolysis produi'ls liimi i Ira plnsi iriicr ililuio llmso lr< mi lint l'VC purtiim ul ilu* phisiisoi mul lire, in lurii, 11il* inujiii* ilrgriuliidoii produrl*. There nre im m'griululion prouurts rouilinn Irniii mi iiiioriii'hon ul tliu l'VC ivilli (lie plnstisol. The pyrogrsims rrsiilliug front pyrolysis ul iho various pinstisoi* ul l'VC i*uii he iimmI lur piirjuimH ul "liogcipriniiiig," Itleiililieniiun ul lira major pcnics in n typirol pinslisol pyrogrmn providw information lending in n precise iilaiitificnlioii ol lira plutlciitcr. The pyrolysi* dnln from thin study were rohued tu n spodal com ol flammability nnd toxicity. INTRODUCTION Tito mechanism of the thermal dchydrochlorhmtiou of jx>ly(vinyl chlo ride) 1ms Ix'en the subject of many studies in the general area of jxdynicr degradation. A number of the important contributions in this amt have been reviewed liy Jn;vv inJavtLexcellent Tito present pajier deals with the identification of the volatile, hydrocarbon pyrolysis product* resulting froin the thermal degradation of l'VC at dOO'C in un inert (heiintn) atmouplieiv. The objective of the current study was to obtain u complete profile of tho degradation products from l'VC resin and l'VC piastisols. This study bus acted as u starting ixjiut in an overall study of thermal degradation, com bustion, and flammability of cldoriuated irolymors. 1IU17 Q 11)70 liy Joint Wiley it Sunn, lira. I ltuut M. At. O'AJAiVA PREVIOUS WORK Stroinbcrg' Itaa xtudicd the thermal dcgi'mlntimi of WC in vacuum by (tot atripping tlui renin of IIGI nt tent] tern tn rex up U> UnO'O An* 30 min. * Tho rcaiilkitig ronidtio wax then beaked tit >100*0 for 30 milt and Uto volatile hydrocarbon producta rcauiting front Utia aecond heating wore identified vitv i mnaa apoeirometry, In Una particular study, Stromberg identified iqt* proxinutloly tin hydrocarbon producta. Obtain unit lliiknwn4 pyrniyzcd i'VG ut i2'*0 in a nitrogen nunoxpitcro and studied kite resulting degradation product* by infrared nod ultraviolet ... spectroscopy. Their findings allowed that: (0 tlto pyrolysi* products (other thou HOI) consisted of aliphutio mid aromatic hydrocarbons, and (2) tho t.vpcN of pyrolysis product* from l'VC were a function of tlto stereo- -- : regularity (tnoticity) in tho undegraded polymer. Additionid (piniitntivo or quantitative measurements were not made. NofTz <it itl.* degraded i'VG in it high-frequency pyrolyxcrmtd separated the dogmdntiou products by Kit" chromatography. Their findings rovetded tluvt tho hydrocarbon degradation products consisted of only uronuvtic compoumla; n volatile aliphatic hydrocarbon fraction wius not identified. Coleman and Thomns* have studied tho combuation of l'VC by ther mally degrading tlto polymer in an exec** of oxygen. Under the ex;tenmental condition* employed, tlto itydrooarbott* formed tv* a result of polymor degradation were completely oxidized to carbon oxides. In addition to HC1, CO, atvd COj, a very tnvco mnount (5 ppm) of carbonyl chlorido we* ' alao identified.* In a moro recent atudy of tho degradation of l'VC in tvir, Boottner ot id.*'* found that when l'VC ltomnpolymcr ivml comitoutvd* were heated iiv ait* from ambient to (iOO'C, approximntely ik`% of tlto I*VC wax reacted to IICl, CO, ivitd CO* (tvo pltoagcne above 0.1 ppm). Tlto 5% hydrocarbon fraction consisted of IS aliphatic and ivromatic hydrocarbon*. Analogous degradation oxi>erimcnta of l'VC plaatiaoi* (l'VC plus a phthalato cater) revealed that the hydrocarbon fraction of tlve pyrolyzato incrcivacd duo to product* from tlvcrmal breakdown of the plasticizer. Characteristic degradation producta dorivablo from tlto plaxticizor aiono Were not formed. " INSTRUMENTATION AND PYROLYSIS TECHNIQUES 'litormoiyaia of l'VC rcain (Coon 103, Cl - <">7.4%) was carried out by two general technique*. Tlto first method involved heating tho rcain in tlto > heated (32d*C) inlet of our max* spectrometer (CMC ntodol 21-103C) in > order to obtain a maaa apectrum of tlto total pyrolyzate. Tho accond, moro detailed method consisted of degrading the ream in a pvrolyaia-gaa chromatograph-mass spectrometer (PGM) tandem ayatem. Briefly, ,our PGM unit cmiaixtx of an 1' & M Model 700 temperature* . . progfitirtmcd gax chromatograph interfaced to tlto nuiax xpoctrometer through a molecular onriehor.M A pyroiyxi* utiit consisting of a-radiant !**!#; ! Srhcii (helium) lino; hlltf VftiVft Ai<<! (. CritP tho iJileeM typo furnace \ 1* A M gtia sa the gaa aamp> citromatograp unit. Titova tlto body of tit way, carrier g. chamber ami directed to oi diagram of th . valve ia presen Tlto results of tiic maaa a chromatograpi chromatog.'api aeparated mt a accnaplitltalen CRG) coltimo was alao uacd 1 The piastiaola of plasticizer (i Sample* of i go* how of GO--I Since a atuih heated at tillO* The problem o lowing mmmei chromatograpi uoc 037503 PYROLYSIS OP POLY(Vb\YL ClILOIUDli) 1W 1' Pit;- ! SrltcmiUie nf |iyhilyxui null. illasi ruling rolnlivo imsilimi til (,-l) currier gus (helium) limn; (II) hunt Mi oil11totiU Jiiity*; ((/) iivcii Mmlniiihig; (/>) P niul M giM-suntp- vnlvo and (A) ipiurU pynhysiu duuulKtr null Yiulbull heal pyrelyxer.--Arrowi iitdis cut a ilia Oiractiim of gus (low. t.vjwf furnace (Sargent A *<., .Model S-dOlOO), quarts chamber, and heated l'' t M gas sampling Valve (.Model (1V-U) was constructed. Function of tho gu* sampling vnlvo wits to interface tho pyrolysis chamlter to tho gns chromatograph and to permit on-line or off-lino ojteratiou of tlio pyrolysis unit. The two nan chromatographic currier gns linos were "looped" outside the body of tlie chromatograph through Tygott tubing connectors. In this way, currier gas could be directed from tiio chromatograph to tho pyrolysis chamber and tlie diluent lino from tlie pyrolysis chamber could then bo directed to either of the injection ports of tlie gns chromatograph. A diagram of the pyrolysis chandler, gas chromatograph, and gits sampling Valve is presented in Figure 1. The results from the pyrolysis experiments carried out in tho heated inlet of tho mas* spectrometer indicated that in order to obtain a complete chromatographic separation of tho pyrolysis products from PVC, two chromatographic colupins would bo required. Light gases (CrCO woro separated on an S-ft I'ompnk QS while tlio higher-boiling components (Ct to ncenaphthaicno) were analysed on a 20-ft X */u in. SI0n2 (10% on 80/100 CUC) column. This same combination of gas chromatographic columns was also used for tho analyse* of degradation products from PVC plastisols. The plastisols wero prepared by fusing 100 parts of PVC resin with 00 putts of plasticiser (i.e., o-diootyl phthaJate) at &>0oF for .*> min. Samples of PVC resin and plastisols were pyrolysed at 000*C in u carrier gas How of 00-70 ml/min; samplo sixes ranged from 10 to 20 mg. HC1 EVOLUTION DURING PYROLYSIS Since a stoichiometric amount of IICI is released ('>8.3%) from PVC when heated at <iOOsC, over half of the degradation products, by weight, is IICI. The problem of chromatographic analysis of IICI wits dealt with in tlto fol- lowing manner. It was found that for niudyscM performed on tho SMo2 chromatographic column* tito roteutiou timo of tlto IICI peak cortespotulod 1I1VU M. ,Vi. O'JIAIIA approximately to tho dead time of tho column. Thin ]>cuk ww fairly sym metrical and did not exhibit excessive bleeding. However, with the I'orupuk column, IIC'l \e;i* idund to bleed excessively, In order to correct thin, the pyrolyr.nte wax scrubbed of liCi by [illicitly 4 A molecular move* in Urn heated (lAO'C) diluent line (Fig. 1, IS) to the chro matograph. Thi* eliminated the II Cl mid did not *ccm to remove any of the other ga*e*. 11 won found, a* a result of otlier work, that the nievos did have tut adhiity for abort eluun chlomalkanox under UteNO conditions ITik. X Pymgnun nf l'VC mwii from 2<i ft X *A. in. i<>% SIWtl <m *0/100 CllW dim. mutiiKrn|>liir column. Ientered [xuiku refer lo iilonlifiivuioiui iu Yublo I, Since tlicfto tyjwat of compound* arc not evolved during tiie pyrolysis of i'VC, titis selectivity for ehloroalkunc* did not interfere with tho result* of tiie analysis. In n typical pyrolysis experiment, IG-aO mg of I'VC resin were precisely weighed into a o.O mm X 10 mm porcelain boat and loaded into the pyroly sis chandler along with a small magnet. The chamber was cloned, the gait sampling valve waxno positioned tJiat;tiio pyrolysis unit wax on-line with tJio chronuitograpli, and the furmico wax turned on, Onco tlio temperature of tho furnace readied 0Q0*C (calibrated), tiie boat won moved, witli tho aid of i ii the magncl typical pyr. After thi positioned While ti chumhcr (i weighed, calculated, material wl I'yi'oi.VMremoval oi resulting c 3-4% of t'n 1V.1 idciitilii' A b C )) ii V G n J K I, M N' P It s T U V w tfopunu Aside f. of a sent ehlorinati componci A typi compone; ie i i ` PYUOLA'SIS W POl.YfViiSYI. CUUNUOU) I, 1UVI the magnet, into the hot mum of tho pyrolysis chamber. To this way a typical pyrogrnm was recorded. After tho remains of the sample cooled down, tho gas sampling Volvo won positioned for otf-iine alteration, While tho gas-chromatographic analysis was being completed, tho chamber (now off-line) wns opened, and tho sample boat was removed and \ weighed. This oi>eration permitted tho percentage char formation to bo ' calculated. Tho measurement was not corrected for minor amounts of ^material which condensed at the end of tho pyrolysis chamber. ' v.. 1`YItOI.YSIS OF PVC lUiSIN Pyrolysis of 1'VC at (KX)aC in a helium atmospheru results in the complete removal of Cl as 1101 (f>N.3%) front tho hackltono of tho poiyntor. Tho resulting carbonaceous ash, dovoid of chlorino, represents approximately 3-4% of tho original polymer. TABUS I Identification of Components iit Pynwmn (Fig. 2) of PVC Peak idonl ideation Component* A CIt.,* CO,* CO,.* C.IU* C,Uf *b ItCl, 0,11**0,11.* . 'c , ' Untune,* butene,* butadiene,* diacotylonc* : it ( Ci nnd Ct uiiphutio and oiofiuio nydroearbotut 1 ii ' Uuinutne v; `Toluene , 1 j- c CltlnroUenseiM , 31 , Xylano ' . 1 '' . *. J \ AllylbeiwMts .' . V, ' . K ! - C.J!,, i:' , I. j C,1I,, M Indane . j ; X; Indene, eiltylUtlnetto ' J* jtlclliyliiidnno - .* It . Mnlliylimluno* . S Xnploliiilnmi T Dimel bylindane u V Metliylnnplulinleuo Mel liylimpltl liak'ite, tteeuaplnIndene w 1 timet It)'Inapln Ituleno Separated Ami identified on an K-ft Pornpak (IS. Aside from a stoichiometric amount of KCi, the PVC pyroiyaute consists of a series of aliphatic, olefinic, and aromatic hydrocarbons. Only ono chlorinated hydrocarlttin, citlorolx'iuene, tvas found, anti tbo yield of tltis component Wits well Itelow 1 wt-%. A typical pyrogram of PVC. (Sl'lnU column) is presented in Figure 2; continent idoutilicutions aro summarised in Tablo I. A similar analysis 1|W2 M, M. O'.UAHA \vn* ulxo carried out on the l'orupnk Q.S column. Tho component* uopnruted oiul identified in Uiw nnnly*ix are noted in Table I. Tlo major component* reuniting from tho pyrolyxi* of l'VC ore IIC1, bciuone, toluene, mid naphthalene. In addition to tlicso major products, , no liomolngou* aerie* of aliphatic mid olelinic hydrocarbon* muring from Ci to C* m'o formed. Tni* finding in in direct cnntraat to tho renuitx obtained by Noff. This difference may bo attributed to differenced in technique* ainco Xoffx degraded hi* Kamplo* with rf energy. GAS CIZIiOMATOGltAniXC ANALYSIS OF IIC1 Tho cliromatograiiliie peak from IIC'l can be uxed to quantify Clio amount of 1101 released during the pyrolyai* of l'VC if tho pyrolynin and cltromntograpliic *y*tcm i firot "ncnnoncd" with a whig of HC1 go*. l'or example, in ono experiment a aerie* of l'VC *nmplca (10-20 mg) were pyroJyxed at G00*C and the gm^cliromatograplne arena of tho resulting IIC1 peaks wero rccordod. (Tlio gas chromatography column and pro-column were maintained at ambient temperatures throughout this experiment Usually six to sevou pyrolyses could bo oarried out before bonsono began to PVJIOLYWS OK POi.YfVlNYh CJII-OIUDW) 1093 bleed from tho column. Bleeding of minor amounts of other more volntilo hydrocarbon* did not interfere with the analyses). A plot of I IC'l content (peak area) venom weight of renin pyrolysod nt (iOO*G in presented in Figure H. Aside from the find run, nil subsequent analyse* provided n linear reintionnhip lx;tween the two variables of I1CI nron uud nmount of renin pyrolyaed. 'Hm find nin of IICl through the column seems to condition the apparatus hy coating the wall* with gaseous 1KH. In subsequent rmm, nil of the 11(11 formed during pyrolysis, reaches the detector with little or no adsorption on the chromatographic walk*. Thin impact of tho pyrolysis of I'VC in l>oing further investigated in tho direction of optimising tho technique for metmuriug tho amount of liCl produced whou l'VC conqumuds utul other chlorinated polymors aro thermally degraded. .i j '! 1 ' PYROLYSIS REPR0DUCI3ILITY Tito demonstration of reproducibility in pyrolysis work is an important aspect of this technique. In order to meet this requirement, three pyrolysis TAUUi II Composition {Gas-ChromatoKraphio Area) of PVC Hydrocarbon PyrolyxAta Aron, % Component Hun 1 Hun 2 Hun 3 coMethane ___ . ..... - 0.0 "8.8*"- . 2.2 7.8 0.8 : 0.3 CO, ' 0.4 ! 0.3 "7 0.2 * . Ethylene Ethane Propylon# - 3.8 i 4.8 ' 2.0 ' 3.8 3.8 8.1 r>.5 , 2.0 : 3.0 . Propane lint one, bitlone 8.0 V ' . 2,3 . 2.8 ' 3.1 ' a*i llutadicna, diarolylou# 1 C, and C* hyilroaeriianM Ilcnxciio Tnlnone ' )' Chlnrobenxene Xylene " - : , AllyllieiMene C,Ih, C,II,, Tndnno Indcna . 1.0 , At .2 8,;i ` l.l , 1.0 ' i.:t 0,4 0.2. 0.2 0.0 t .8 ,1.3 ' . 80.8 ' 80.1 8.0 8.8 1.1 ; , *. .1.3 2.1 2.3 1.4 1.0 0.8 0.8 0.2 0.3 . 0.3 ' 0.3 1.4 ` 0.0 I'Pliyliiilnrim Methylindnna Miilhylindcno* Xaplilhidcne Dimothylindano Methylnaplillwdene Mothyhmpht halene Aconaphthaleno 0.8 I.t 4.8 0.7 0,8 1.2 0.8 1.3. . 4.0 0,8 0.7 .. 1.0 , ii t 0.8 |1.0 4.f) 0.7 lei * i 1 PWt M. M. O'M MU i , ox}>orimcnt were carried out in which various PVC (remit) sumj ilex were pyrolyxed nt <KX)*C. (A dual nnalyxix wax carried out in ouch ettxa, one for anxlyxix on the Slv`i- chromnlogriiphio column, the other for analysis on tho v I'or.'tpnk Q>S coi\nnn.) The resulting gnx-ohromntogruphic peak areas of nil ` components were recorded Tor the three pyrolysis csiierimculx. These three analyses, summarized in Table il, demonstrate the quantitative and ' qualitative reproducibility of the technique. , The gas chromntogrophic peak iuviim reported in Tn'ble II cloudy reflect " the mole per cent contents, thus nn approximate quantitative analysis of tha > pyrolyznto from thermnlly degraded PVC ix poxxil>le. It should he noted v 't that the total hydrocarbon fraction com*x|)on<lx to approximately 3<S% V (weight) of the original l`VG xamplo; the remaining (12% consists of HC1 '|tml u carhonaccoux, cldorino-frco axh. Uciucno ix the component of ' greatest concentration in the entire hydrocarbon fraction. , ' i 1 : PYROLYSIS OP POLY(VINYL CHLORIDE) PLASTISOLS ( Initial pyrolyxix of l'VC plaxtixolx centered alxmt a plaxtixol prepared by fusing 100 partx of PVC and (10 partx of u-(2-cthylhcxyl) phthalato (DOP) at 3oO*P for o min. The xeopo of this initial study wax thon expanded to 1 include a number of other PVC plaxtixolx. The plaxticiaorx uxod through' out thix study woro dorived from e-phthalic acid: ' 1' 'i 0 1 OR R-diphAtic t c The rexultx of the initial xtudy (PVC/DOP) revealed that tho degrade tion productx reflected both jmlymer and plasticizer thermolysis. Thus, tho pyrogram could l>o \ixod ax a finger printing tool for identifying the specific plasticizer prexcut in tho plaxtixol. v , Tho pyrolyxix of the I'VC/DOP plaxtixol wax carried out at (100C in a ' helium atmoxphera; tho rexuiting pyrogram (Sli`>2 gux chromatography ' column) from this pyrolyxix ix prexented in Figuro 4. Peak identifications , are summarised in Tnhlo III. Ax previouxly mentioned, the pyrolyxix products reflect polymer degrada tion (HC1, benzene, toluene, etc.) and piaxtioixer degradation (C hydro- _ carbon, C aldehyde, and phtludic anhy<lrido). It ix tho presence of thoxo componcntx due to DOP fragmentation that provides tho basis for using the plaxtixol pyrogram ax u "fingerprinting" aid. , Tltix finger])rinting capttcity of pyrogramx from typical PVC plaxtixolx ix . 1 illustrated in Figure 4. A xnricx of xcvcn pyrogramx rexuiting from tho " pyrolyxix of tho following PVC plaxtixolx ix prexented: (If) o-dihexyl* phtluduto (DIIP),` (O') o-di-2-eth,ylhoxyl*phthalata (DOP), (D) o-dicapryl. * phtlialute. (DCP), (A1) o-diisodeeyl phthniuto (DDP), and (F) tho mix*ed ,*** ester butyl lx that of the re Itxhouid be t Ciinij; Dei CO Mm hum CO, El hyleiu Kthnne Prnpylci Pmpnne Bui.me, Ct nmi C Cdl,. Iwiprcitd 2-Mcihy iJ-Pl'ItlCtl Ct olefm Benzene C, olefia Tninene C, i iiciin Chhirobt Xylene AUvllxn c,i[,, C,Wu 2-Eiliyl Iml.ino Iniicne Kihvitiii' Mcihyli Meiityli Xapatlt llimeili Methyl Arenap Methyl Phthali Aw rrlal k Appiu-mil pyrograms settings on of ditYercncc* ucc I'YUOI-YM OK I'OlA'fYINYf, Cltl/)1\10W) 195 uaUn* butyt iioiixyl ii'pliUmUtu (littlUl'). Tito (ii'xl pyrtijtmm in Kijturo 4 in that of tlto fontn without [ilnxtixol .-tail i* itinluiM for ]tiir|>ot0M of Coinpiuixoit. Itxhottid l>o ttoU.il Unit modi of Uto iloutil (i.o., minor component*) of Utcw) ' TAHU4 111 CotpnrUh of (itti*CltromuliiKi`iipliic Aren* ii/.l>fKnuliUiim I'nwliicttt from 1*VC litwtt #ml l'VO 1`lutiUMtl (1)01*, 100/00)* t "---------------------------\ Aron, % Dctfriululum prtMluat v > \ Moilmito * \co, Kl li.vlcito Kilmno Propylene Propiine . Biitnno, Imtone : Oi nml Ct Jiydrorwlmiui (toUl)* * Ojlt* iMiprcno 2-Mcihyl-l-btiteite . 3-Pcntcn-l-yiia Ct nlotin (totol) Bcnxcno Cj olefin* 1. , Toluene , i .. Cl nlcfilM . ClilomOciwene Xylene '' \ Allylbenaene C,lt,, c,tr,, t 2-KUtyl hexeiutl ,1 . Intltmo ' Indcne 1 Klliylinluene .` Mcthyiiiuimie rUoiltvlimloiio . Xiipltihaieno ( Dimoiliylimluite MeOtyhiephlhnlene t Arotixplitlmloiio 1 IWciliyluapliiliuk'itol Phihitlic wiltydriilo J IUmii 0,(1 , ,s..s 0.4 3.3 4,H ' 2.0 3.0 j.r " " 1.0 -- -- `-- 81.3 --, 3.3 -- 1.1 1.0 1.3 0.4 0.3 -- 0.3 -- ' 0.0 0.3 1.1 4.8 o.r 0.3 ' 1.3 -- 1101' Plwittwtl 1 3.3 0.3 . 2.3 4.0 4.7 1.3 * f>.2 -- 1.0 2.1 2.0 0.4 1.0 11.8 / 1.3 2,0 ` , 24.7 0.4 0.7 , ; --' #. -- . ' . 1.4 1" -- 0.3' *-- 0.8 ' 1.2 --* 10.Of (minor) \ (major) Arc* related to * Applimiiie only to dagriidHiioii of renin. jiyro^nutM ittttt liocit omitted tiiroti(jii n judicial!* choice of uttciiutitioii xettinjpt on tlto KttH-clifOntitUtufttitliic recorder.' Thin won (lotto for rotutoiM of Minpiicity tut tlto ttmjor jwitkit in ouch of Uto pyrognutu* providod odotpiAto diiTcroneo* for purpowo of futnorpriiithig. 5; ii i: im .I I 1i 3: \ , k, \ .ii V - \ \ \ \ \ M. M. 0`MM\K Pytvfv* * ! L .1a | " <! 1., * ; i* ,t r? 1! Lj L^J s_ ' ( % l& l t L Ki*. i. l>.rrom of (.i) l'VC remit; (H) 1`VCAwlilwxyl ; !i*i liyiliexyl plulieloie; U>) PVC/eiliciipryl phUwUw; () 1 VC/emiueodeayl phU^ 1 i*n; O') l'VC/e-butyl bon*yl phdtAloto. Dcl.'iil pllistislpi, llUVI! ill'l Till! (i till! iViilj iKlJllltllil coiitiiiii.'i phthaiic The pyr dihoxyt the nritr the ololi pltistlciz Olefin unit Mil phihiiln The i terest in phthiiiii the thri ulrciuiv ployed fleets tl these t' olefins ; pyrolvs plnstici. if it cizers v chroma hydrol.' ysis pr The of an u the rest A Ci present vidcs a . The benzyl pyrolyi sruph * IrflV t Tin ivpt tin reilectit uCC H37511 1Y0I,Y*I8 Oh* PW,Y(ViNYb ClIhOIUDK) 1A97 Detailed qualitative mudynon of oil degradation product* from tho live plontinoln (/f-/1*) went not curried nut; only Uto major degradation product* hove been identified. Tint first pyrognmi in Uto series (Kin, '0 indicates tint relative locution of tint nmjor degradation front I'VO resin (i.o., IIOI, benzene, toluene mid naphthalene),* Tito second pyrogruni (/f), i`VC/o*dilio.\yl phthuiutc, contain* tliono [teaks representing thermolysis of the plasticizer (hexene mid phlhnlie anhydride) itt mlilitiou to those from degradation of Uto polymer, Tl.epyrogrmn of u-diisodeeyl pliUiulnte/l'VC! (10 in analogous to Unit of tlio dihcxyl plttlmlnto pyrogruni*: the presence of ]ththulic anhydride identifies the itroutntie portion of the plasticizer while tlio jmnitiou (mid presence) of the olefin moiety (CM in thin cone) identilien the uliphutio jmrtioli of tlio plasticizer. . Olefinio degradation product* result from tlio thermal dc-csterificution mid subsequent dehydration of ihu plunticixer. In one enno (diisodocyl phthulnle), nn ulcohol (0,) won found among the degradation product*. The pyrogrunm of tint two (!,, plnntinoln (DOI* mid DCI`) provide mi in* teresting cone. Until pyrogruni* contain iv C. nldehyde peak, the expected phthnlic mthydride peek, nml tlinte (1* olelin itenkn. However, the rutin of the three olefin jieukn in different in eueli of tlio two cunen. Mince it hit* itlreudy iieen demoimtriitcil Hint the reproducibility of pyrolynin, un em ployed in thin ntudy, in excellent, thin difference in inomcr distribution re flect* the difference in tiio two 0 plasticizers. Kurthcr investigation of these two plnstinols directed toward tlio identification of tlio specific C olcfinn in cuch enno won not curried out. Tiio important point was that pyrolysis-gun chromatography could bo uned to distinguish tlio two isomerio plusticizow. If it lmd been ncccnnitry to completely characterize tlicno two C> plonticixorn with renpoct to tlio nliphutio jxirtion, it in obvioun tliut pyrolynin-gun chromatography offem many advantages over a technique involving acid hydrolysis of tlio pliwticixcr followed by an mudynin of tho alcoholic hydrolyni.s product*. The pyrogmm of tlio mixed enter (BiiIlzP) plantwol reveals tlio prcnonco * of an unexpected secondary reaction occurring during pyrolynin or during tlio resulting chromntogrophic mudynin. A Ci olefin duo to fragmentation of tlio butoxy portion of the enter in present, an cxjwctcd. Pyrolysis of tlio -OCITiCH portion however, providen a uniquo enno in that olefin formation via dehydration in not possible. Tho pyrolynin product* from thin fragment (--Oil/.) were tolucucf and benzyl chiorido. Independent work in our laboratory, involving tlio pyrolynin of dlbcnxyl ether (HCl not rigorously excluded from gan-cliromato- * graph linen or pyrolynin chamber) showed that thin material produces tiio * Lower-chain nlipluoic prtshim* (CII,, ole.) arc Imricd mwlcr the IIC1 paik. f Tho no i f lioiMMioo/iiJiMBic (pcok height;punk bright) in nil of tho pyrugram*, ex cept Unit of j)Uip, wan Gai 0.4; la die liulUP pyrogram die ratio wee 3.0, thereby reflecting zit increase ia ioIuoim. iw r M. M. O'M MU same pyrolysis products. These data suggest that dibcrotyi other may ho formed from the pyrolysis of butyl benzyl phihuluto. Secondary pyrolysis of this in termed into fn!Iowl by reaction with lU'l would give theobserved pyrolysis products, On the other hand, this ether may not l>e an inter* mediate, in wldciraisb'froorudicals formed from the pyrolyi* 0/ BuBaP I may scuvongo IIC1, tlioroby forming Uto observed product; ^ ` 0 1' V* + c,h# + cut*--cn^ou cd-d--ca,a 0 + [C,K,OHj f CACH,* + K* |-iw> C,H, The question u* to whether or not dihciuyl ether is nn intermediate in the pyrolysis of hutvl bciuyl |ihthninto is not within the scope of this paper; it is noted, however, that a secondary reaction of the pyrolysis products from Bull/.!' with HC1 is in operation and it is this reaction that is the source of tho benzyl chloride! SEMIQUANTITATIVE ANALYSIS OP PLASTISOL PYROLYSIS PRODUCTS ; The gas-chromatographic analysis of tho pyrolysis products from a I'VC/* .1)01' (100/dO) plustisol arc presented in Table III. Included in this sum* mury is a similar analysis for the 1'VC homojtolymcr (resin) in tho absence of plasticiser. The pyrolysis product* from the plasticiser (DOT) constitute tlio major portion of the total products, thereby exhibiting a dilution effect on those products solely from tho resin. The exception to this is tho increase in the ' amount of COj in tho plustisol pyrolyzute. Tins results from a secondary reaction of the plasticizer--loss of CO*; the major degradation path of DOP 'is however, directed toward phthalic anhydride formation. Tho increase in Cj and C* hydrocarbon content (relative to tho resin) and the appearance of Ct olefin in the plustisol pyrolyzute is duo to secondary . crocking of the C* moiety. The precursor of this material is again tho DOP ' ..plasticizer. Additional experiments were carried out in which tho pyrolysis of tho ' DOT plustisol was performed under static ( no helium How) conditions. Tins was accomplished by placing the gas sampling valvo in tho "off-lino" . posi tiom Aftor a five minute pyrolsysis timo, tho valvo was returned to tho "nii-iiiic" posi t to tin: chromi 1 i compicte cra< Boettner* 1 programmed addition to a ular weight Tlie overall t atid IlCi, the The preset is applicable the plastic i undergoes ti Titus, in till phthalic an) ticizcr's deg not grossly; ular quan tit. ctcr. Boettner' as the plast may Same ( The findi bility arc of Tins (iuUkn Dnnuis, 1 tie i tho tcrliatcnl t. 1). \V. ami Vlin'iiKii ami Torii. I a j. n. a. Thermal Oxi iilghmi-o for 3. i'tram i. S. Oil! 1. i). .vi . k. n. 1 7. it. .s. a. 1). K. A. in. J.T. n. u. n Clmrtiineri* Received Revised I 1 ucc I PYIIOI.Y.SIS <)! l'OI.Y(VINYI. CIIUIHIDU) urn "<m-lino" position ami tho pyrolysis products were swept from tho chamber to tho chromatograph. Under those conditions, tho C* olefins underwent complete cracking to haver chain olefins. TOXICITY 01-' DEGRADATION PRODUCTS Boettner* has shown that when plasticized i'VC is degraded in air under programmed temperature conditions a number of organic molecule* (In addition to a stoichiometric amount ofllCl), very similar to the lower molcc. ttlur weight hydrocarbons identified in tho present study, are released. ; The overall toxicity, however, is related to the gross quantity of CO, C()i \ | and 1101, the first of those having tho greatest effect.'1 ^ Tin* present pyrolysis work, having heeu carried out in tho nlwcnco of air, -is applicable to only ono particular set of fire conditions; that case where , tho {ilasiio is subjected to "shock" heating conditions where plasticizer undergoes thermal degradation rather than evaporation and oxidation. Thus, in tins particular instnnee, the degradation products would contain 1 phthulie anhydride, the ievel of winch would he dependent upon the plus- : timer's degradation to evaporation/nxidatiou ratio. This product would ' not grossly affect the toxicity of tho degradation products since tho partic ular quantity of CO*, HC1, and most imjxntantiy CO determines this param; ctcr. tj Uocttnor's work most closely simulates actual burning conditions, that is, ` as the plastic is heated, IIC1 and plasticizer are boiled off, tho plasticizer 1 may damo (depending on the HC1 content) and CO and CO* are produced. The findings of tho present study representing a special cose of fiamma1 bility are offered as a corollary to Bocttnor's work. Tlio author wishes In acknowledge lira technical communis of L. U. Cridor and II. L. 1 Daniils, lira instrument modification* suggested by It. A. Youat oad J, Dakman, aad ( lira lorhuical assistance of J. A. Nikorn, i References ... I, |), \V. Iravi, "lateraluro fhtrvoy mi Thermal Degradation, Thermal Oxidation, , and Thermal Analysis of iiigii Polymers," June, HIGH, CJeariaglioase for Pod. Scion. , ami Tech. Info., U.i*. Dept, of Coimnunxi, AlH'J.'tVIO. , 2. I). A. Teaisei ami i). \V. Jravi, ``[.ileraiure .Snrvoy nn Tiiormal Degradation, Thermal Oxidation and Thermal Analysis of High Polymer* II," January 10M, Claar* ingiionso for Perl. .Scion, and Turin Into., U.S. Dept. of Commorco, AiJufilGoA. ' .'h .Stnimlrargetal., J, I'olym.Sf.i., Jo.ttVi (10.10). . ' 4. S, Ohlani and T. lahiknwa, Kogm Kugak* Xuumki, GS, 11117-22 (10G2), ' <1, I). N'oiTx, W. Poo*, and \V. Pfali, Z. Anal, Clirin., 211. 121 (lOuS). (1. K. I [.Coleman and C. II. Thomas, J, Ap/il, Chrm,, 4,1170 (10.VI). ' 7, !;. i I, Coleionn, /`luKtirx, 21, -11 (i (J 0.10). , X. K. A. iloulliter noil ii. Weiss, .1 mrr./ntl. Hull. .1 **ne..J,, 2$,-12.1 (1007). , !l. !;. A. linutliicr at nh, J, I'ait/m. AW., 1.1. ',',17 (ItalO). *'r , ( II). J.T. WnlSnnanilK. llicninliu,.!(/.f.Virnt,, 30. lln.1 (11)04). *' 11. If. II. Cornish, paper |>resenlod at Polymer Conferentxt Series on Plarontabikiy ) Cliararteristirsof Polymeric Materials, Univorsiiy in Dotmit, Jtura 10,1000. 'i Received December 1, 10(10 Itoviscd Decomber 31, 1030 1