Document RgQ9LGLno0BMRVrprQkdbRYk

/c rv c j Liquid-Phase Degradation of Poly(Vinyl Chloride) K. S. MINSKER, M. I. ABDULLIN, and G. E. ZAIKOV Bashkirian State University, Ufa, USSR Institute of Chemical Physics USSR Academy of Sciences Moscow, USSR The main kinetic regularities of the thermal and thermal-oxidative degradation of polyvinyl chloride) (PVC) in the liquid phase have been analyzed in comparison with the solid-phase degradation ofPVC. The thermal (in an N, atmosphere) degradation of PVC in the form of dilute solutions is characterized by a number ofessential distinctions, the primary peculiarity consisting ofa lower dehydrochlorination rate due to the retarded reaction of the formation of ^C--C^ bond polyconjugated systems. The same is observed for the thermal degra dation of PVC plasticized with di- and polyesters. On the contrary, in an oxygen-containing atmosphere the solvents promote PVC decom position. Accelerated PVC degradation under such conditions is due to the solvent oxidation that causes the appearance, within the system, of products activating the PVC macromolecular decomposition. The in tensified degradation processes are accounted for, in the first place, by an increased reaction rate ofthe statistical (by the random law) detach ment of HC1 from normal macromolecular units. In general, the kine tics of the thermal-oxidative liquid-phase degradation of PVC is de termined by the partial pressure ofoxygen in the reaction zone, by the quantity of the solvent introduced into the polymer, as well as by the oxidative stability of the solvent. It has been shown that an effective stabilization of PVC in the liquid phase, particularly in the systems highly plasticized with esters, can be achieved through stabilizing the solvent, first of all, rather than the polymer, against oxidative decom position. In this case the PVC dehydrochlorination rate decreases sharply and may reach an essentially lower value than under similar conditions of the solid-phase thermal degradation ofPVC. PVC stabili zation with respect to the reaction of HC1 elimination achieved through the solvent stabilization against thermal-oxidative decomposi tion has been called the effect of "echo-stabilization" of PVC. INTRODUCTION Up to the present the most important kinetic data on the degradation and stabilization of poly(vinyl chloride) (PVC) have been obtained through the study of the polymer decomposition in the solid phase (powder, films). The process ofPVC degradation in solution has far less been con sidered, practically no investigations on the prob lem of the phase influence on the PVC dehydrochlorination process being under way. Mean while, this is an interesting problem from the theoretical and practical point of view with refer ence to the wide application of PVC materials ob tained on the basis of strong (plasticates) or rela tively dilute (fiber) solutions. The main reaction in PVC degradation is the hy drogen chloride detachment from the polymeric macromolecules, the PVC dehydrochlorination process being complicated and including: a) statis tical (by the random law) HC1 elimination from normal sequences of vinyl chloride units proceed ing at the rate Vr and accompanied by the formation of single internal C=C bonds within the mac- /\ romolecules; b) propagation of polyconjugated sys tems of ^C="C^ bonds (at the rate V,,) due to activation of the reaction of HC1 elimination by labile conjugated carbonvlallvl groups structured as --C(O)--CH--CH--CHC1--CHj- (1). The rate of the PVC brutto-dehydrochlorination VHCi in an inert atmosphere (or under vacuo) is described by the equation: VHa * V, + kp7t (1) and is determined by the content of carbonylallyl group y,, within the PVC macromolecules (2*4). Introduction of oxygen into the reaction zone accelerates PVC degradation. In an oxidizing atmo sphere, the PVC dehydrochlorination process pro ceeds in accordance with the equation (5, 6): ucc 037427 JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981, VOL. 3, NO. 4 Liquid-Phase Degradation of PolyiVtnyl Chloride) Vhci -- (Vir + V,p) + (V3r + Vjp) = (fc,, + *.,.) a,, + (UpOJ* + A-3pI^[p02])a,, (2) where Jttr and fc,, are the rate constants of a noncatalytic reaction of random HC1 elimination and of the formation of polyconjugated systems of ^C=C^ bonds; k3r and k3p are those for a catalytic (under the influence ofO*) reaction; [pOt] denotes a partial pressure oxygen in the zone of thermal deg radation; a, stands for the content of HC1 in PVC before the polymer dehydrochlorination started. The degradation kinetics of PVC undergo a sub stantial change during the transition from the solid phase to the dilute solutions ofthe polymer in inert solvents (o-dichlorobenzene, decalin, benzyl al cohol, cyclohexanone and others), namely, the brutto-rate of HCI elimination (in an inert atmo sphere) shows a considerable drop (Fig. 1) which is an essential peculiarity ofPVC degradation (7,8). A decrease in the rate of PVC brutto-dehydrochlorination is clearly observed at various tempera tures as well as with a number of solvents and PVC samples (Table 1). The division of the brutto-process of PVC dehy drochlorination Vgci into the reactions V, and V, has made it possible to establish that the values of the random detachment of HCI during the polymer de composition in solution and in the solid phase are close to each other (Table I). A decrease in the rate Fig. 1. Kinetics of PVC brutto-dehydrochlorination (175'C, in the atmosphere of S\j: 1--powder; 2--in o-diehlorobent*ne; 3--in benzyl alcohol; 4--in decalin; 5--powder (150`C), 6--in cyclohexanone (lSO^C). VHCI in the liquid phase is due to inhibition of the reaction of polyconjugated systems propagation (the rate constant ofthe process k,, decreases during the transition to the polymeric solutions), although no chemical interaction with carbonylallyl groups \ or polyconjugated systems of / bonds is observed. As a result, the value of VHCI for the "PVC + solvent" system approximates that of the rate V,, This is another peculiarity of the liquid-phase thermal degradation of PVC. The third distinction consists in a decreased ac tivation energy of the PVC dehydrochlorination process in solution as compared with the polymer decomposition in the solid phase (Table 2). The activation energy of the brutto-dehydro chlorination process of powdered PVC samples depends on die contribution from the activation energies of the random HCI detachment and of the formation reactions of polyconjugated systems \/ of ^CTMC^ bonds. It is to be noted that the polyene formation rate is connected with the con tent of carbonylallyl groups in the polymeric mac romolecules. The higher is the magnitude ofy,, the greater is the value of the ratio V^Vr. During degra dation of PVC dissolved in an inert solvent, th contribution from the reaction of the random HCI elimination, which proceeds at a lower activation energy (Farr * 23 kcal/mol) than the polyene forma tion (Ebci 35 kcal/mol) (4, 9, 10), to the bruttoprocess ofPVC dehydrochlorination increases, i.e,, the ratio V^/V, becomes about one order less (Table 1). The latter results in a decrease in the observed activation energy of the brutto-process of HCI elimination (Table 3). The above facts suggest that with PVC decomposing in inert solvents the corre lation between y, and Eaft (2-4) is apparently re tained. Noticeable is the fact that the introduction of some solvents, such as benzyl alcohol, o-dichlorobenzene and the like (Fig. 2), into PVC in concentrations up to 1 g of the solvent per 1 g of th polymer does not change the rate of HCI hruttoelimination, i.e., the introduction of the above type of solvents into PVC in relatively small quantities does not affect the degradation kinetics of PVC. The inhibition effect of the PVC dehydrochlorination reaction is only observed in dilute solutions of the above agents. Another matter is the use of a different class of solvents which are well compatible with PVC un like the above compounds. Such solvents readily form solutions with PVC even at the "solventpolymer" ratios below one. In contrast to the sol vents that are non-compatible with PVC, this seems to be the case of a plasticizer dissolved in PVC rather than the polymer solution in the solvent medium. The introduction of low-molecular plas- JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981, VOL. 3, NO. 4 UCC 037428 231 Table 1. K inetic P aram eters o t T herm al (In th e A tm o sp h e re o l N J D e g ra d a tio n o f PVC In V arious S olvents K. S. Mtnsker. M. 1. Abdullin, and C. E. Zuikov *-4 k |y S 3 A 03 3 IA tA > ooo || i T *1 ^ rt> 1 tt o tA co o o o r) d 03 tA ** o> d 0> - - tA tA iA <0 O (0 M O T > > lA tA tA IA IA CM d d ev f* Io S 3 tA 3 03 11 1 M4 d d o d d o b A 3 fm A CM 03 03 M- s CM CM 03 03 3 M4 d d ri d d o d d d 1 # tA pm lA Pm tA Pm PIAm o CM 11 | 6oood d y sso v* 3 3 3 3 CD <0 CM 3 M o d d d o d dd o 4 > tA A IA 3 3 o || | *" d d d * Pm p- CM 3 3 s SS 3 > q o o d fm CM d pi pi d E ** > PtA- (PAm tA Pm tA Pm 3 CM oooood E 11 |1 ss> > Ame 3 d 3 d 3 o 3 d CD d 3 d dd 3 d PX > oi CM *M| 3 A o 3 CM d d t- CM d e * tA lA O O o Q e z d V* d CM d d d d o > Table 2. Activation Energy Values for Brutto-Dehydrochlortnatlon of PVC Dissolved In o-Diehlorobenzene (175*0; in the Atmosphere of Ni) To 10*. M. mol/PVC mol 57,000 70,000 102.500 157,000 1.54 1.53 1.52 1.04 ' DiyidiiMt In E,,,, kcal/mol 33 = 1 33 c 1 33 1 31 1 E',,,, kcal/mol 28 r 1 28 1 28 = 1 25 = 1 Table 3. Kinetic Parameters of PVC Thermal Degradation (175*C; In the Atmosphere of Ni) in the Presence of Ester Plasticizers Plas ticizer DOP DOA DOS PAS-22 PDEA-4 _Concentration of plaatleizer, g/PVC g 0.03 0.06 0.10 0.20 0.60 0.01 0.03 0.10 0.30 0.50 0.03 0.10 0.30 0.50 0.07 0.18 0.37 0.45 0.60 0.03 0.05 0.10 0.20 0.30 0.40 0.50 V, 10T V,. 10T HCI mol PVC mol a 0.80 0.80 0.75 0.74 0.74 0.75 0.78 0.78 0.77 0.70 0.77 0.79 0.80 0.78 0.78 0.80 0.77 0.77 0.76 0.77 0.80 0.78 0.76 0.77 0.78 0.76 0.78 7.30 4.40 3.65 3.66 3.46 3.25 5.42 5.04 4.43 3.81 4.13 5.81 4.90 4.22 4.12 5.20 4.23 4.23 4.64 4.73 7.10 6.22 6.04 5.78 6.02 5.94 6.02 setessssss H 4 c* i Jli*a 9 -1 A lA 40 lA lA tA O < ** d d CM 1 <S* * ii|ssiiiii3?C K niuS il II ils s n il I! 5 3 S 3 3 3 33 3 . E ii 1 La 5 lA pm Si tA | I ^ P* NN tA iA IA lA I id l/} W VAV4 >*S.* ticizers, such as di(2-ethylhexyl)phtalate (DOP), di(2-ethylhexyl)adipate (DOA), di(2-ethylhexyl)sebacate (DOS), into PVC even in the quantities ranging from 0.05 to 1.0 g/PVC g results in a notice able decrease in the HC1 elimination rate (Fig. 2) though to a less extent than in the case of the poly mer decomposition in dilute solution (11, 12). The same is observed for PVC degradation in the medium of oligomer polyesters, namely: PDEA-4 (the product of dibutvladipate reesterification with diethylene glycol), PAS-22 (the reesterification Eroduct of a blend of dibutyladipate and diutvlsebacate). The presence ofa plasticizing agent in the system has hardly any effect ofthe formation rate ofrandom single ^0C^ bonds (Table 3). Thus the thermal degradation of PVC in the form of dilute and strong (in plasticizers) solutions re- 232 JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1961, VOL. 3, NO. 4 ucc 037429 Liquid-Phase Degradation of PolytVmyl Chloride) critical concentration and the content of labile structures (carbonyl allyl groups) within the mac romolecules in expected to exist. Indeed, it can be seen from Fig. 3 that the greater value of7, calls for the higher content of the ester plasticizer to cause the brutto-dehydrochlorination rate VHCt to reach a constant value. As a result, there exists a linear relation between the parameters 7, and [DOP]^ (Fig. 4). The kinetic mechanism of PVC degradation in the liquid phase changes considerably if an inert atmosphere is replaced by one containing oxygen (14). No retardation of PVC dehydrochlorination Fig, 2. Effect ofdi- and polyester plasticizers on the rote ofthe thermal (175'C, in the atmoephere of NO bruttadehydrochlorination of PVC: i --OOP; 2--DOA; 3--DOS; 4--PAS-22 (); 5--PDEA-4 (U); 6--benzyl alcohol, veals a number of basic distinctions as compared with the polymer decomposition in the solid phase. First of all, a considerably reduced dehydrochlorination rate is observed, the decrease in the rate of HC1 detachment from the polymer mac romolecules during the transition to the liquid phase occurring mainly due to the reaction of polyene sequences formation. ' The observed effect may be accounted for by two factors: 1) a difference between the PVC structure on the supermolecular (powder) and molecular (so lution) levels; 2) a catalytic influence of hydrogen chloride on the polyene formation during PVC de gradation in the solid phase. At the moment of the detachment ofanother HC1 molecule, the [polymer HC1] complex forms in situ to activate dehydrochlorination of the neighbouring vinyl chloride unit in the PVC macromolecule. It is known (13) that hydrogen chloride does not affect the rate VP, while it accelerates the reaction V,. Apparently, in the case of the polymer degradation in solution, such a catalytic action is eliminated due to the sol vation of macromolecules by the solvent As can be seen in Fig. 2, the PVC dehydrochlori nation rate is falling ifthe plasticizing additive con centration increases only to a certain critical value ca. 13 wt percent. The existence of a critical con centration for a plasticizer is evidently connected with the creation of such conditions under which the groups that initiate the reaction of the polyene system formation find themselves "blocked" due to their solvation by the plasticizer's molecules, thus making difficult in a certain way the growth of polvene sequences. The introduction of plasticiz ers in concentrations above the critical value no longer affects the degree of "shieldedness" of the polyene growth labile sites. Therefore the exceed ing of the critical concentration is not accompanied by any further reduction in the PVC degradation rate. As a result, a certain correlation between the , t/TTC g Fig. 3. Dependence ofthe rate ofPVC brutto-dehydrochlorina(ion on the content of dif2-ethylhexyl)phthalate for polymer samples with various content oflabile ctirbonylaUyl group* 7, (mol/PVC mol); 1--0.82 10-; 2--1.02 10-; 3--1SS 10-; 4--1.80 10- ^Di ( ^ , (/FTC g Fig. 4, Dependence of the critical concentration of di(2ethylhexyliphthalate [DOPt, on the content of carbonylallyl groupt 7. in PVC responsible for initiating the formation of bonds polyconjugated systems. JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981, VOL. 3, NO. 4 233 ucc 037430 K. S. Minsker, M ! Abdullin, and G. . Zadarp occurs in this case. The HC1 elimination rate in creases rapidly. Moreover, the greater concentra tion of the plasticizing agent introduced into the polymeric composition causes a higher rate of the PVC thermal decomposition at a permanent partial pressure of O*, i.e., in an oxidizing medium, ester plasticizers accelerate significantly the decomposi tion of the polymeric macromolecules rather than inhibit it (Fig. 5), the kinetic dependence of PVC dehydrochlorination acquires an autocatalytic character (Fig. 6). It should be pointed out that the plasticizer is only observed with the concentrations of the latter above some critical value [Plasticizer]rr. fi*e, Bln Fig. 6. Kinetic* of HCl brutto-elimination during thermaloxidatioe degradation (175X1, in the atmosphere ofOJ ofFVC containing DOS (1.4), DOA (2,5). DOF (3,6): 1-3-03; 44-0.1 gIFVCg. Another special feature of the PVC thermaloxidation degradation, as compared with the ther mal one, consists of the fact that the plasticizers produce an unequal effect on the polymer dehydrochlorination kinetics. Specifically this manifests M itself in the values of the critical concentration of the plasticizeT which differ for the investigated es ters, the value [Plasticizer]^ being dependent on the partial pressure of oxygen in the reaction zone (see Table if). The above peculiarities of the influence of ester plasticizers on the PVC thermal-oxidative decom position can obviously be explained in the follow ing way: first, the replacement of an inert medium by the oxygen-containing atmosphere gives rise to various oxidation reactions involving the plas ticizer molecules, where the plasticizer decompo- (hi Fig. 5. Dependence of the hydrogen chloride quantity evolved within 60 min An,i on the concentration of the etter plasticiser during thermal degradation of plasticized FVC (175*0 in the atmosphere of oxygen (a) and air (b): l,l`--DOS; 23'--DOA; 3,3--OOP. table 4. Values of Critical Concantrations of Estar Plaatlclzars Obtained during Thsrmal-Oxldatlva Dahydrochlorinatlon f PiatUcizad fvc at 17B*C Eatar OOF DOA DOS OOP OOA DOS Atmosphere air air air oxygan oxygen oxyQ*n [Ptasdclxarln, g/PVC g 0.20 0.10 0.05 - 0.06 0.15 0.06 - 0.07 0.01 - 0.02 JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981, VOL. 3, NO. 4 uec 037431 Liquid-Phase Degradation of PolyCVinyl Chloride) sition is accompanied by the formation of products accelerating the degradation of the polymeric mac romolecules. This is also substantiated by the fact that the value VHC, grows with the increasing con tent of the plasticizing agent in the composition (Fig. 5). Secondly, judging by the degradation kine tics of plasticates, the investigated plasticizing ester additives to thermal oxidation possess differ ent stability. This means that the PVC decomposition rate in the presence of plasticizers presubjected to thermal-oxidative exposure should be higher than with the use of the parent esters due to a higher content of their decomposition products which are active towards PVC degradation. Indeed, one can see from Table 5 that the rate VHCt increases with the prolongated thermal pretreatment of the plas ticizer. A comparison of these results with the data on the thermal oxidative degradation kinetics of plasticates shows, however, that the introduction of a preoxidized plasticizer into PVC does not accelerate the decomposition of the polymeric molecules as significantly as one could expect. These results make it possible to suggest that the accumulating products of the plasticizer oxidative decomposition, such as acids, ketones, etc., within the system are not the basic and only reason for the intensification of HC1 elimination during the PVC thermal-oxidative decomposition in the medium of esters. Another reason lies apparently in the ap pearance of free radicals in the plasticate in the course of its oxidative degradation due to an inter action between ester molecules and oxygen. The forming radicals attack the polymeric mac romolecules, thus increasing the rate VHCl, A kinetic study of the initiated (by di-tertiary butyl peroxide) oxidation of DOP, DOA, DOS has , shown (15) that the process rate is directly propor tional to the ester concentration and the square root of the initiator concentration (Fig. 7). These ex perimental results indicate that a thermal-oxidative attack on the plasticate causes an intense oxidation of esters, the reaction proceeding according to the radical-chain mechanism with a quadratic termina tion at the peroxide radicals. The oxidation rate is described by the equation: o >. !, u I, ). [iimmwiJ , w/l Fig. 7. Dependence of the oxidation rate W,,(I35'C) ofDOS (1, 4), DOA (2, S), DOP (3, 6) on the concentration of the eater plo*ticizer(l-3;[ditertiary butyl peroxide], "5 10~*mol/l)and ditertiary butyl peroxide (4-6; [platticixer] molll: 4--2.35; 5 --2.70; 6--2.56). W,, = k,[RHJROO ] k, *,-[RHjVWi (3) where [RH] denotes the ester concentration; jfc, and k, are the rate constants for the propagation and termination of the reaction chain, Wt stands for the initiation rate. The dependence of W,, on the square root ofthe initiator concentration (Fig. 7) has made it possibl to determine the parameter k, fclf* (Table 6) characterizing the plasticizer oxidizability. As one can see from Table 6, the investigated ester plas ticizers of PVC form the following sequence according to their tendency to oxidation: DOS > DOA > DOP. The above results are in good agreement with the data on the kinetics of the plasticized PVC thermal-oxidative dehydrochlorination. The high est value of the polymer brutto-decomposition rate is observed for the thermal-oxidative degradation ofPVC plasticized with DOS, while the lowest one has been obtained with DOP used as the plas ticizer. An essentially lower rate of PVC dehy drochlorination in the presence of DOP, as com pared with that in the presence of DOS or DOA, is apparently the result of the fact that it is the alcohol group of the ester molecule only that is subject to oxidation in DOP. Table S. Effect of PreoxMatlOA of Eatar Plasticizers (175*C, Air) on Kinetic Parameters of PVC Thermal Degradation Duration of Plasticizer, thermal treat 0.3 g/PVC g ment, h DOA DOS 0 1 2 4 6 0 2 4 6 Vmc. - IQ* V 10' V, 10' NCI mol/PVC mol ' 0.82 1.10 1.21 1.40 1.70 0.65 0.86 0.98 1.30 0.80 0.80 0.81 0.79 0.78 0.80 0.79 0.77 0.78 7.22 5.41 10.19 11.31 16.20 5.70 7.81 9.03 1232 Table 6. Value* of the Parameter Mcj'1 for tome Ester Plasticizers of PVC Temperature, "C Plasticiser MU'* 10>, i/mol a 115 DOP 115 DOA 115 DOS 125 DOP 125 DOA 125 DOS 135 OOP 135 DOA 135 DOS 145 DOP 145 OOA US DOS 0.58 1.03 2.00 1.31 2.68 3.38 1.81 3.70 4.90 1.99 4.74 7.66 JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981. VOL. 3. NO 4 ucc 037432 K. S. Minsker, M. I. Abdullin, and C. E, Zaikor A studv on the autooxidation of ester plasticizers at 140-2i0C, i.e., under the conditions similar to those of the thermal-oxidative degradation of plas ticized PVC, has indicated that the interaction of esters with oxygen is accompanied by the accumulation of hydroperoxide groups within the plasticizer, the process autoaccelerating (Fig. 8, curves 1-7) due to the reaction of a degenerated chain branching: ROOH RO- + HO- (4) while chains originate (at the rate W^) according to the reaction: RH + 0*-R- + HOO- (5) The rate of ROOH accumulation is determined through the expression: d[ROOHJ/df - WROO-IRH] = kt *f`[RH](Wi. + fcJ[ROOH])1/1 (6) (a) i * k * u I (b) Fig. 8. Kinetic curvet for the accumulation of di(2ethylhexylladipate hydroperoxide in the autooxidation duty <l-7) and the relative straight line* (l'-7'); a) 1,1'--14<FC; 2,2'--150X2; 3.3'--162"C; 4,4--175'C, b) 5.5 --186X2; 6,6'-- 198X2; 7,7'--210X2. with Win k3 ROOH the integration of Eq 6 yields: ROOH* * = -j- (7) In accordance withEq 7 the kinetic dependences of the hydroperoxide group formation straighten out in the coordinates "ROOH1,1 - time" (Fig. 8, lines 1-7), which is evidence ofthe first order for the hydroperoxide concentration in the reaction of de generated chain branching. On the basis of these dependences we have determined the value ktk^mksn and extrapolated the values ktk;1,1 found from the results of an initiated oxidation of esters onto the temperature range used for the study of autooxidation. Further the decomposition rate con stant for hydroperoxides Jt3 has been calculated (Table 7). The maximum point of the kinetic curves for the accumulation of hydroperoxides (Fig. 8) can be ex pressed as: WJU - - k, = kt ki'^RHy^kJiAOOHi (8) where Wftrm and WJ* are the rates of the hy droperoxide formation and degradation respec tively. The values of the maximum hydroperoxide con centrations, the respective initiation rates Wf and the maximum rates of the hydroperoxide decom position are summarized in Table 8. The compari son of the rates Wf and indicat s that the brutto-decomposition of hydroperoxides proceeds much more rapidly than their decomposition via the monomolecular mechanism. This is apparently associated with the growing rate of the radical- initiated decomposition ofthe hydroperoxide as its concentration grows. The reactions of hydro peroxides with the molecular products of their de composition might also contribute to a certain ex tent. It should be pointed out that the maximum con centrations of the hydroperoxides accumulated Table 7. Values of Decomposition Rata Constanta for Hydroperoxides of Ester Plasticizers, X, Temperature, *C Plasticizer It, 10\ a- ISO ISO 150 162 182 182 175 175 175 188 188 188 . 210 210 210 DOP OOA DOS DOP OOA DOS DOP DOA DOS DOP DOA DOS DOP DOA DOS 0.20 0.29 0.48 1.21 2.48 2.69 1.44 2.59 9.60 8.41 8.50 13.65 11.36 12.50 26.80 JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981. VOL. 3, NO. 4 ucc 037433 jrjirf-Phase Degradation of PolyCVmyt Chloride! Tbl . Rat** of Initiation and Brutto>D*corpo*itlon of Eatar Hydroperoxide* at ttiolr Maximum ConcantraUona ([OOP], - 2.56; [DOA], * 2.70; [DOS], - 2-3$ molt) t. C 140 ISO 162 175 186 198 210 fftOOHI__ mol/l DOP io* w; io* w-,,, 10' mol/l t _ 17.0 15.4 14.0 10.0 9.4 0.34 1.86 _ 11.78 9.35 10.68 -- 1.1C 3.49 __ 14.SS 15.96 20.14 DOA [ROOHJhu io*. wr - io W 10* mol/1 moll * 12.0 9.4 4.5 5.0 6.0 -- 2.95 2.44 3.82 5.31 7.50 _ -- 11.19 12.48 22.50 31.41 44.40 f*oXOl mol/l DOS 10*. wr 10* W*. 10* mol/l 12.3 20.8 9.1 7.0 6.2 8.3 -- 0.59 5.60 8.74 9.55 14.52 22.27 -- 5.58 25.41 42.65 49.10 83.50 13.35 reach fairly high values (Table 8) which ensures high initiation rates. This means that ester plas ticizers ofPVC are active generators offree radicals due to the reaction ofdegenerated chain branching proceeding at 180-210C, i.e., under the conditions that correspond to the treatment temperature of plasticized PVC compositions. The free radicals that form during the hydro peroxide decomposition are able to propagate the oxidation chains of the plasticizer as well as to at tack the PVC macromolecules, thus sharply accelerating the polymer degradation. One could expect the radicals to intensify, first of all, the ran dom elimination of HC1 from the polymer mac romolecules Vr, since the initiators of free-radical processes are known to accelerate the reaction (16). With reference to the thermal-oxidative degrada tion of ester-plasticized PVC this means that the higher formation rate of hydroperoxides and the numerical value ofthe constant k3 for the plasticizer under investigation, should bring about a higher formation rate of random single C--C bonds. /\ Indeed, the steady of ester plasticizers' effect on the reaction rate of HC1 random elimination has shown that an essential increase in the rate Vr(Fig. 9a, lines 1-3) is observed in the sequence: DOP < DOA < DOS, i.e., in a strict accordance with the decreasing constant ks. The end products of the plasticizer oxidation in clude various oxygen-containing compounds, such as acids, ketones, etc., which are known (1) to be able to accelerate the formation reaction of poly- conjugated systems of ^G=C^ bonds (this is also products grows high enough, the catalytic effect of these compounds on the rate V, comes to play a decisive role. As a result, during the PVC thermaloxidative degradation in the medium of an ester plasticizer, the dependence of the formation rate of ^C="C^ bond polyconjugated systems on the plasticizing agent content essentially differs from that for Vr(Fig. 9a, curves 4-6), and the dependence of the rate ratio V/Vp on the plasticizer concentra tion has an extreme character (Fig. 9b). A study ofthe liquid-phase degradation ofPVC in ester plasticizers with regard to the reactions Vrand V, makes it possible to explain why, in a number of cases, the process of PVC brutto-dehydrochlorination does not depend on the presence ofester plas ticizers in the polymer if their content in the poly mer is below the critical value (Fig. 5). The reason for it lies in the fact that, although the introduction evidenced by the fact that, as stated above, the introduction of a preoxidized plasticizer into PVC accelerates the polymer degradation due to ah in crease in the rate V, only). On the other hand, plas ticizers themselves inhibit the polyene formation (during the plasticized PVC degradation in an inert medium). In the range of low concentrations of a plasticizer (0.05-0.20 g/PVC g depending on the nature of the ester plasticizer), the inhibition of the process prevails; while for high concentrations, when the rate of accumulation of the oxidation JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981, VOL. 3, NO. 4 ucc 037434 237 K. S. Minsker, M. 1. Abdullin, and G. E. Zattor (b) Fig. 9. Dependence 0/ ihe quantity 0/internal bonds formed within 60 min (1975*C, in air) b.(l-3) and HCl evolved from PVC due to the formation reaction of polyconjugated sequence> ofdouble bonds within macromolecules bp (4-6) <a), as well as the dependence ofthe ratio ofthese parameters (b) on the content of DOS (1,4,8), DOA (2,5,9), DOF (3, 6, 7) in the polymer. the rate of HCl random detachment, in some cases (depending on the partial pressure of O, and the nature ofthe ester) the increase is compensated for by a decrease in the growth rate ofpolyene systems. As a result, VHCi might be independent of the plas ticizer concentration in the system, since the pro cess of PVC brutto-dehydrochlorination includes both reactions. However, exceeding the critical concentration gives rise to the violation of the com pensation due to a rapid accumulation, within the polymer composition, of the ester oxidation prod ucts that catalyze the reaction ofthe polyene forma tion. Therefore, in the case of a reduced concentra tion, an increase in the rate of PVC bruttodehydrochlorination is observed. Other evidence for the above results is the fact that the rate ratio VJVP varies continuously over the whole range of the plasticizer concentrations investigated, includ ing those below the critical one (Fig. 9b). Thus ester plasticizers promote the thermaloxidative decomposition of PVC, while in an inert atmosphere the decomposition of PVC in the medium of plasticizers is inhibited. The accelera tion of PVC degradation in an oxygen-containing atmosphere is accounted for by the plasticizing agent oxidation leading to the appearance of inter mediate and end products that activate PVC de composition in the system. The intensity of the process depends on the nature of the plasticizer used and, first of all, is determined by the rate constants ktkivt and k3 characterizing the oxidiza bility of the plasticizing agent. The principal way to PVC stabilization is nowa days associated with the chemical modification of carbonylallyl groups that are to be found in the polymeric macromolecules and are responsible for a high rate of PVC decomposition. Howev r the above approach proves ineffective for PVC so lutions, specifically for highly plasticized composi tions. One can see from Fig. JO that the rate of the thermal-oxidative dehydrochlorination of the ini tial PVC and the PVC with pre-deactivated car bonylallyl groups (for instance, through the interac tion of labile structures with triphenyl phosphite (17)) in the medium of DOS is practically the same, though the modified PVC exhibits much higher stability in an inert atmosphere and in the absence of the solvent (Fig. 10, cf. lines % and 3). It follows, from the above stated, that effective stabilization of PVC in solution, particularly in ester-plasticized systems, can be achieved through the solvent stabilization against oxidative decom position in the first place, and not by way ofstabiliz ing the polymer. In this case, the PVC dehy drochlorination rate decreases sharply and can ulti mately reach the value VHCt observed during the PVC thermal degradation in solution, i.e., an essen tially lower value than the one obtained under simi lar conditions of PVC decomposition in the solid phase. Thus in this particular case, an effect takes place that we have called the effect of " chostabilization" of the PVC dehydrochlorinati n pro cess. It means that an effective stabilization ofPVC with respect to the reaction of HCl elimination achieved through the plasticizer stabilization against the thermal-oxidative decomposition. The Fig. 10- Kinetic dependences of thermal-oxidative and thermal (2,3) dehydrochlorination of initial PVC (1 (9) and 2) and PVC with predeactivated carbonylallyl groups (l (X) and 3). 238 JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981, VOL. 3, NO. 4 Lufuid-Phast Degradation of PoiytVmyl Chloride) stabilization of a plasticizing agent causes auto matically the PVC stabilization due to the specific feature ofthe solvent to retard the thermal degrada tion of the polymer. REFERENCES 1. K. S. Minsker and G. T, Fedoseeva, "Destruktsiya i stabilizatsiya polivinilkhlorida," p. 14, "Khimiva," Moscow (1979). 2. K. S. Minsker, Al. Al. Berlin, and V. V, Lisitsky, Vysokomolek. Soed., B18, 54 (1976). 3. K. S. Minsker, Al. Al. Berlin, V. V. Lisitsky. and S. V. Kolesov, Vytokomolek. Soed., A19, 32 (1977). 4. K. S. Minsker, Platte Kouttehuk, 24, 375 (1977). 5. K. S. Minsker, Al. Al. Berlin, and M. 1. Abdullin, Vy- sokomolek Soed., B16, 439 (1974). 6. K. 5. Minsker and M. I. Abdullin, in: "Degradation and Stabilization ofPolymers," Akzo Research Laboratories and Institute of Chemitral Physical Acad, of Sci. USSR. 7. K. S. Minsker, M- I. Abdullin, V. I. Manushin, L. I. Malvshev, and S. A. Arzhakov, Doklady AN SSSR, 242,366 (1978). 8- K. S. Minsker, M. I. Abdullin, S- A. Arzhakov, V. I. Man ushin, A. V. Matvkin, and V. A. Kraikin, Vytokomolek. Soed., B21, 194 (1979). 9. K. S. Minsker, Al. Al. Berlin, V. V. Lisitsky, and S.V. Kolesov, Platt. Matty, No. 10, 69(1976). 10. K. S. Minsker, Al. Al. Berlin, V, V. Lisitsky. S. V. Kolesov, and R. S. Korneva, Doklady AN SSSR. 232, 93 (1977). 11. M. I. Abdullin, V. I. Manushin, L. B. Zhikharevich, and K. S. Minsker. Vytokomolek. Soed., B22, 229 (1980). 12. K. S. Minsker, M. I. Abdullin, and V. A. Kraikin, Platt. Matty, No. 3. 31 (1980). 13. K. S. Minsker, V. P. Malinskay, M. I. Arsis, S. D. Razumovski, and G. . Zaikov, Doklady AN SSSR, 223,138 (1975). 14. K. S. Minsker, M. I. Abdullin, V. G. Kalashnikov, V. P. Birukov, G. M. Trosman, and A- V. Salnikov, Vytokomolek. Soed.. A22, 2131 (1980). 15. V. S. Martemvanov, M. I. Abdullin, T. E. Orlova, and K. S. Minsker, Neftekhimiya, 21, 123 (1981). 16. K. S. Minsker, M. I. Abdullin, and G. A. Garipova, Doklady AN SSSR, 218,851 (1974). 17. K. S. Minsker, N. A. Mukmeneva, Al. Al. Berlin, D. V. Kazachenko, S. I. Agadzhanyan, and P. A. Kirpichnikov, Doklady AN SSSR, 226 1088 (1976). JOURNAL OF VINYL TECHNOLOGY, DECEMBER 1981, VOL. 3, NO. 4 ucc 037436 239