J. Chem. Soc., Faraday Trans. I, 1982, 78, 869-879 Alternating Copolymerization of Conjugated Dienes with Methyl Acrylate Part 2.-Isoprene BY CLEMENT H. BAMFORD* AND XIAO-zu H A N ~ Department of Inorganic, Physical and Industrial Chemistry, University of Liverpool, Liverpool L69 3BX Received 22nd April, 198 1 The alternating copolymerization of methyl acrylate (MA) and isoprene (Ip) in the presence of ethylaluminium sesquichloride (Al,Et,Cl,) has been studied. Photoinitiation (3, = 436 nm) was effected by the system Mn,(CO),, + CCl,. Alternation occurs over a wide range of reactant composition. Copolymerization competes with a rather rapid Diels-Alder addition yielding a mixture of 3-methyl- and 4-methyl- (methyl cyclohex-3-ene- 1 -carboxylate). Evidence is adduced indicating that the rate of copolymerization is proportional to [MA- --all [MAfree]' [Iplo9# where MA- --a1 represents the methyl- acrylate-Lewis-acid complex and 9 is the rate of initiation.All the kinetic data are consistent with the cross-propagation mechanism of alternation. The propagation process -1p' +MA is relatively slow in the absence of Lewis acid, but is markedly accelerated (ca. 88-fold) by Al,Et,Cl, under the conditions used, with resulting alternation. This enhancement in rate is attributed mainly to weakening of the C=C bond in MA which accompanies complexation of the monomer with the Lewis acid. Chain-transfer to CBr, has been observed and the appropriate kinetic parameters evaluated. In the previous paper1 we have reported an investigation of the alternating copolymerization of methyl acrylate and butadiene in the presence of ethylaluminium sesquichloride, with photoinitiation by a transition-metal derivative + organic halide system [Mn,(CO),, + CCl,].We concluded that the kinetic data are compatible with the simplest alternation mechanism, namely that arising from the predominance of cross-propagation reactions. A similar conclusion was reached earlier for the methyl acrylate/styrene alternating copolymerization.2 In both cases it appears that com- plexation of methyl acrylate with the Lewis acid greatly increases the rate of reaction of this monomer with propagating chains carrying terminal units derived from the hydrocarbon monomer. This paper describes an extension to our work to the methyl acrylate/isoprene copolymerization in the presence of ethylaluminium sesquichloride.Furukawa et aL3 have reported the synthesis of an alternating copolymer of methyl acrylate and isoprene by photoinitiated copolymerization in the presence of aluminium chloride and vanadyl butoxide, and Akimoto and Ohtsuru, have investigated the influence of water on the alternating copolymerization of the same monomers with Et,AlCl,-, + VOCl, as catalyst. However, there is not much mechanistic information in these works. EXPERIMENTAL MATERIALS Isoprene (Ip) of research grade (Phillips Petroleum Co) was distilled and dried over calcium hydride; before use, it was distilled on the vacuum line. t Present address : Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun, 869 China.870 COPOLYMERIZATION OF CONJUGATED DIENES Methyl acrylate (MA), manganese carbonyl [Mn,(CO),,] and carbon tetrabromide (CBr,) Ethylaluminium sesquichloride (A1,Et3Cl3, 25 % toluene solution) and carbon tetrachloride Toluene (A.R.) was dried over sodium wire.The apparatus and techniques used were similar to those described in the previous paper.' All kinetic experiments were carried out in a laboratory illuminated by inactive (sodium) light. Reactions were initiated by light of wavelength 436 nm, the optical system being the same as previously de~cribed.~ Mn,(CO),,, CCl,, A1,Et3C13 (in toluene solution as required), MA and Ip were introduced by pipetting the necessary volumes into the Pyrex glass reaction vessel in the order stated. The reaction mixture (total volume 10 cm3) was then thoroughly degassed by the conventional freeze-thaw technique. When the vessel had been sealed it was irradiated in a thermostat at 25 0.1 OC for the required reaction time.The reaction mixture was then poured into 400 cm3 methanol containing a small amount of phenyl-8-naphthylamine to prevent oxidation of the copolymer. Precipitated copolymers were filtered off, dried and weighed ; they were rubbery solids free from gel. Samples for analyses were reprecipitated from methanol. Random copolymers of MA and Ip required for spectral observations were prepared by copolymerizing the two monomers ([MA] = 6.5 mol dm-3, [Ip] = 2.5 mol dm-3) in toluene solution using benzoyl peroxide (0.008 mol dm-3) as initiator at 40 OC. The conversion was 1.7% after 26 h.Number-average molecular weights of copolymers were measured osmometrically with a Hewlett-Packard 503 high-speed membrane osmometer, with toluene as solvent. Infrared absorption spectra of copolymers were recorded by a Perkin-Elmer 5 17 grating spectrophotometer using films cast from chloroform solution. Compositions of the copolymers were determined by elemental analysis and the extent of alternation was estimated from the 60 MHz n.m.r. spectra in, deuterochloroform. The Diels-Alder adduct of MA and Ip in the reaction mixture was estimated on a Pye 104 gas-liquid chromatograph using a 3 m column at 160 OC, packed with 3% OV-22s on Supelcoport (100-120 mesh). The solution for g.1.c. analysis was made by pouring the reaction mixture into 90 cm3 methanol containing 5 cm3 water, to facilitate separation of copolymer and destroy AI,Et,Cl,.RESULTS AND DISCUSSION were purified as previously reported.' (A.R.) were used as supplied without further purification. APPARATUS AND TECHNIQUES COMPOSITION AND STRUCTURE OF THE COPOLYMER Elemental analyses showed that the composition of the copolymer does not depend on the initial concentration of MA in the range 0.4-3.2 mol dmb3; the 1 : 1 composition is maintained except when the concentration of MA is so low ( e g . 0.1 mol dm-3) that homopolymerization of isoprene to cyclic polymers occurs (see later) (fig. 1). Fig. 2. shows the n.m.r. spectra of an alternating and a random copolymer of MA and Ip. The peak near 2 6 (ppm relative to TMS) in the spectrum of the random copolymer arises from adjacent methylene groups in isoprene-isoprene diads and the absence of this peak from the spectrum of copolymer (b) confirms the alternating structure.s The peak at 5.1 6 arises from -CH= protons in isoprene units and the peak at 3.65 6 from -OCH3 protons in methyl acrylate units.The ratio of the above two peaks, estimated from the integral line of spectrum (b), is close to 1:3 and is therefore consistent with the 1 : 1 composition. The peak at 1.55 6 originates from -CH3 protons in isoprene units. Spectra of copolymers of MA and Ip, both alternating and random, are very similar to those of methyl methacrylate and isoprene. The infrared spectra of alternating and random copolymers of MA and Ip are shown in fig. 3. The characteristic absorption band^^?^ for cis-1,4 and trans-1,4 isomers ofC.H. BAMFORD A N D XIAO-ZU H A N loo I 00 h E - 4 6 0 - - 5 0 a O0 4 0 - .r( d E - / P S n 0 / I I I I -d 1 I 1 I 87 1 0 Q0 1.6 2.4 [MAlo in reaction mixture/mol dm-3 FIG. 1 .-Dependence of copolymer composition on [MA]. Initial concentrations/mol dm-, : Ip 0.8, Al,Et,Cl, 0.2, Mn,(CO),, 2 x CCI, 0.1 ; 25 OC, 1 = 436 nm, reaction time 10 min. FIG. 2.-N.m.r. spectra of MA/Ip copolymers in CDCI,. (a) Random copolymer. Initial concentrations/mol drn-,: MA 6.5, Ip 2.5, benzoyl peroxide 8 x lo-,; toluene solution, 40 OC, 26 h. (b) Alternating copolymer. Initial concentrations/mol dm-3 : MA 1.6, Ip 0.8, Al,Et,Cl, 0.2, Mn,(CO),, 2 x CCl, 0.1; toluene solution, 25 OC, 10 min, 1 = 436 nm.872 COPOLYMERIZATION OF CONJUGATED DIENES 2000 1000 1600 1400 1200 1000 000 600 wavenumber/sm-' FIG.3.-Infrared spectra of MA/Ip copolymers: (a) random, (b) alternating. Copolymers prepared as described for fig. 2. isoprene units at 840,1131 and 1 152 cm-l overlap absorptions which arise from methyl acrylate units, so that no significant difference between them is observed. KINETICS In general, the concentrations of MA and Ip were both 0.8 mol dm-,, and the concentrations of Al,Et,Cl,, Mn,(CO),, and CCl, were 0.2,2 x lo-, and 0.1 mol dm3, respectively. The rate of initiation, 9, is based on those derived from calibration experiments in which the homopolymerization of methyl methacrylate was photoinitiated by Mn,(CO),, + CCl, with similar light intensity and concentrations of Mn,(CO),, and CCl, as described in ref.(1). In our experiments Y = 2.07 x lo-' mol dm-3 s-' for short reaction times, except in the case of the chain-transfer experiments. To determine the rate of copolymerization we used the gravimetric method, which involved weighing the copolymer formed at low conversion : d w 1000 rate of copolymerization o = --([MA] + [Ip]) = -- dt 77.1 lot where Wis the copolymer weight (in g), t the reaction time (in s) and 77.1 is the average molecular weight of MA and Ip. The plot of w against [Mn,(CO),,]b was found to be effectively linear over a wide range of [Mn,(CO)l,]O (fig. 4), consistent with a conventional free-radical mechanism. The dependence of o on [Al,Et,Cl,], is shown in fig. 5. The presence of Al,Et,Cl, leads to greatly enhanced rates of copolymerization, as observed in the copolymeri- zation of methyl acrylate and butadiene.l An approximately linear relation is followed for [Al,Et,Cl,], between 0.025 and 0.2 mol dm-, when [MA], = 0.8 mol drn-,.C. H.BAMFORD AND XIAO-ZU HAN 873 0 1 2 3 4 5 [ Mn2 (CO), I i/ 1 (T2 rnol&drn-# FIG. 4.-Dependence of rate of copolymerization, w, on [Mn,(CO),,]~. Initial concentrations/mol dm-, : MA, Ip 0.8, Al,Et,Cl, 0.2, CC1, 0.1 ; 25 OC, L = 436 nm. 25 2 0 - 'Y) IE 15 rn a z 10 - \ 3 5 o 0.1 0.2 0.3 a4 [Al, Et, Cl,]/mol dm-, FIG. 5.-Dependence of o on [Al,Et,Cl,]. Initial concentrations/mol drn-,: MA, Ip 0.8, Mn,(CO),, 2 x CC1,O.l; 25 OC, 1 = 436 nm. In the presence of Al,Et,Cl,, MA and Ip undergo a Diels-Alder reaction (1) forming a mixture of methyl 4-methyl- and 3-methyl-3-cyclohexene- 1 -carboxylates (MMCC)O (95%) COOCH, (1) H3c'Q COOCH3 (5%) H3cx+ ( < H3 c -0- COOCH, 29 FAR 1874 COPOLYMERIZATION OF CONJUGATED DIENES m 'E 2 a - --..u u E E 41 rr .- x reaction time/min FIG. 6.-Variation of yield of MMCC with [Ip],. Initial concentrations/mol dm-3 : MA 0.4, Al,Et,Cl, 0.2, Mn,(CO),, 2 x lo-,, CCl, 0.1; 25 OC, 1 = 436 nm. [Ip],/mol dm-3: 0, 0.8; 0, 0.4; 0, 0.2. Curves calculated from eqn (24, (4) and (5). 0 60 120 180 reaction time/min FIG. 7.-Variation of copolymer yield with [Ip],. Conditions as for fig. 6. Ap represents a unit of methyl acrylate in the copolymer. This reaction has a relatively high rate so that it is difficult to disentangle the kinetics of the competing copolymerization.Fig. 6-8 show the progress in time of the two types of reaction. For a given reaction time, the yield of MMCC increases with [Ip], at constant [MA], (fig. 6) and increases with [MA], at constant [Ip], (fig. 8). These results are consistent with the reportlo that the rate of the adduct formation is proportional to [MA---all [Ip]. On the other hand, for a given reaction time, the yield of copolymer falls with increasing [Ip], at constant [MA], under the conditions of fig. 7. We believe this is a result of the rapid consumption of MA at the higher Ip concentrations by reaction (1). Thus, in the experiments of fig. 7, for a fixed reaction time [MA] falls below [all, to an extent which is greater for greater [Ip],; consequently, if the rate of copolymer formation is proportional to [MA---all [Ip], we should expect the co- polymer yield to decrease as [Ip], increases.Under the conditions of fig. 8, the copolymer yield at constant [Ip], increases rather sharply with [MA],. This situation arises because the lower [MA], (0.4 mol dm-3) is initially just equal to [all and falls below the latter as the reaction proceeds, while with [MA], = 0.8 mol dmb3 the methyl acrylate unreacted exceeds [all for most of the reaction. Consequently under the latter conditions [MA- - -all, and hence the rate of copolymerization, is always higher.C. H. BAMFORD A N D XIAO-ZU HAN 875 -.. 0.6 . 0.5 - 'E -0 0.1 1 /-* 60 120 1 80 reaction time/min 0.2 * I€ a 0 0.1 E - 1 c-.l Y 4 0 FIG. 8.-Variation of yields of MMCC and copolymer with [MA],.Initial concentrations/mol drn-,: Ip 0.8, AI,Et,CI, 0.2, Mn,(CO),, 2 x lo-,, CCl, 0.1; 25 OC, 1 = 436 nm. [MA],/mol dm-3: 0, 0.8; 0, 0.4. Fig. 6-8 demonstrate that the conversion of monomers into MMCC and copolymer can approach 100%. The dependence of the rate of copolymerization on the initiai monomer concen- trations for relatively short reaction times (10 min) is presented in fig. 9 and 10. According to fig. 9, w passes through a maximum for rather low [Ip],, then decreases monotonically with increasing [Ip],. An interpretation of the latter (decreasing) part of the curve has already been given, namely the progressive depletion of MA with increasing [Ip], attributable to the rapid Diels-Alder adduction (1). It seems clear from 2 ) 0 0.8 1.6 2 A [ Ip], /mol dm-' 3.2 FIG.9.-Dependence of o on [Ip],. Initial concentrations/mol drn-,: MA 0.8, AI,Et,CI, 0.2, Mn,(CO),, 2 x lo-,, CCI, 0.1 ; 25 OC, 1 = 436 nm. 29-2876 COPOLYMERIZATION OF CONJUGATED DIENES 10 0 0.8 1.6 2.4 3.2 4.0 4.8 [MA],/mol dmW3 FIG. 10.-Dependence of o on [MA],. Initial concentrations/mol drn-,: Ip 0.8, AI,Et,Cl, 0.2, Mn,(CO),, 2 x lo-,, CCl, 0.1; 25 OC, 1 = 436 nm. The dashed curve represents cyclopolyisoprene formation. this result that the rate of copolymerization shows a lower dependence on [Ip] than does reaction (1) (unity); over the range we are considering the order is probably zero. If the copolymerization is similar mechanistically to the methyl acrylate/styrene and methyl acrylate/butadiene reactions we should expect that at sufficiently low [Ip] the rate would be proportional to [Ip]: this is apparent in fig.9. The existence of the maximum in co is therefore understandable. Fig. 10 shows that at high [MA],, the rate of copolymerization is effectively independent of [MA],. Under these conditions the concentration of MA is sufficiently high to maintain complete complexation with A1,Et3C13 throughout reaction. This finding is therefore consistent with the view that co is proportional to [MA---all but is independent of the free methyl acrylate concentration. At lower [MA],, when complete complexation cannot be maintained throughout the reaction, mainly on account of the depletion of methyl acrylate by reaction (l), the rate falls off (fig. 10). The latter figure also demonstrates another complication, namely the existence of a fast reaction at very low (or zero) [MA].A white powdery solid product is formed which, from the n.m.r. spectrum, appears to be cyclized polyisoprene.ll9 l2 This reaction does not seem to be significant for [MA], > 0.1 mol dm-3; below this concentration some cyclized polyisoprene may be formed as well as the alternating copolymer, We have concluded above that the results presented are qualitatively consistent with the rate expressions below, = k,[MA- - -all [Ip] d[MMCC] dt which are essentially the same as those applicable to the methyl acrylate/butadiene system. In eqn (2b) Ap, Bp represent methyl acrylate and isoprene units, respectively, in the copolymer.C. H. BAMFORD AND XIAO-ZU HAN 877 We have attempted to quantify the kinetic treatment by applying the integrated forms of eqn (2a) and (2b) to the results in fig.6 and 7. In the previous paper it was pointed out that the aluminium derivative may complex with the copolymer and the Diels-Alder adduct as well with methyl acrylate; in the latter case complexation is strongest, but no reliable values for any of the equilibrium constants exist at present. It is therefore difficult to estimate [MA- - -all reliably. We thought that calculations could be carried out most usefully for the case [MA], < [all since in these circumstances no recycling of aluminium to MA from other species can occur, so that knowledge of the equilibrium constants is not required provided MA and the aluminium derivative are strongly complexed. According to the kinetic treatment set out in the previous paper [see eqn (7a)], k in eqn (2b) is equal to kbakrib 94 (MA = A, Ip = B), where 9, the rate of initiation, is given by13 kd being the first-order rate coefficient for photodecomposition of manganese carbonyl.Thus we obtain from eqn (2b) 9 = 2kd[Mn,(CO)10] = 2kd[Mn2(CO)1010 exp ( - kd t> (3) dm = 1/2kba k& k\[Mn2(CO>,,]i[fMA- - -all exp ( -ikd t ) . dt (4) We also have the stoichiometric relations [MA- - -all = [MA- - -allo - [Ap] - [MMCC] [Ip] = [Ip], - [MA- - -all0 + [MA- - -all. ( 5 ) Values of [Ap] and [MMCC] have been computed from eqn (2a), (4) and ( 5 ) for a series of starting conditions. The parameters used are shown below: k, = 2.86 x mol-l dm3 s-l kd = 5.17 X 10-4S-' (6) k,,k& = 0.366 m o l t dmt s t .Of these, kd was evaluated from eqn (3), the rate of initiation being determined by calibration with methyl mechacrylate as described in the previous paper. The other parameters were obtained from the initial slopes of the curves for [MA], = [Ip], = 0.4 mol dm+ in fig. 6 and 7. There is a reasonable degree of agreement between the calculated and observed data which strengthens our confidence in kinetic relations of the form shown in eqn (2a) and (2b) and the kinetic mechanism proposed for the alternating copolymerization. CHAIN-TRANSFER AND VALUES OF REACTION PARAMETERS Chain-transfer in alternating copolymerization is of mechanistic interest on account of the information it provides about the character of the propagating species. As a rule, transfer to active agents such as halides and mercaptans occurs much less readily in alternating copolymerizations in the presence of Lewis acids than would be expected from data on homopolymerizations.For example, in the methyl acrylate/styrene system,14 the rate coefficient for transfer to CBr, has a value 441 mol-1 dm3 s-l, ca. 30-fold less than that for the (uncomplexed) radical wv MA-St * . The difference has been attributed to complexation of the radical with the Lewis acid. Results for methyl acrylate/isoprene are presented in table 1. The presence of carbon tetrabromide leads to copolymers of lower molecular weight but has little effect on878 COPOLYMERIZATION OF CONJUGATED DIENES TABLE 1 .-CHAIN-TRANSFER TO CARBON TETRABROMIDE IN THE COPOLYMERIZATION OF METHYL ACRYLATE AND ISOPRENE AT 25 O C Concentrations/mol dm-3: MA 0.8, Ip 0.4, Al,Et,Cl, 0.2, Mn,(CO),, 2 x lo-,, CCl, 0.1 ; 10 min, A = 436 nm, 9 = 1.57 x lo-' mol dm-, s-l.0 2 5 10 20 1.37 1.34 1.33 1.31 1.25 1380 1205 960 758 566 I 1 1 I L 0 4 0 12 16 2 0 [CBr, 1 / 1 (r4 mol dm-3 FIG. 1 1 .-Plot of 1 /E against [CBr,]. Conditions as in table 1. the rate of copolymerization. The plot of l / E against [CBr,] is satisfactorily linear (fig. 1 l), with a slope 0.525 mol-1 dm3. Following the procedure outlined in the previous paper' [see eqn (lo)] we find kfb - 0.384. The rate data in table 1 allow kbak;&, to be evaluated as 0.43 mol3 dmi sf [see ref. (l)], hence we obtain from eqn (8) -- kba k,, k<tb = 0.17 molf dmi s-4. (9) Thus this parameter has closely similar values in the alternating copolymerizations of methyl acrylate with styrene, butadiene and isoprene.The value for kbak;jb quoted above is different from that in eqn (6); in theC . H. BAMFORD AND XIAO-ZU HAN 879 experiments of table 1 the ratio [MA],/[al] was greater than that in the runs of fig. 6 and 7, so possibly the difference implies incomplete complexing in the latter. No estimates of ktbb are available, but it seems unlikely that this is less than the corresponding coefficient in the methyl acrylate/styrene system, uiz. 6 x los mol-1 dm3 s-l. Assuming this figure, we find kb, x 1053 mol-l dm3 s-l. From the data of Morton et aZ.15 we may estimate that, in the homopolymerization of isoprene at 25 O C , k , (i.e. kbb) x 9.3 rno1-l dm3 s-l; hence, with the aid of the reactivity ratio data of Ida et alls (assuming that their values at 50 O C are not very different from those holding at 25 "C) we obtain kba x 12 mol-l dm3 s-l (without Lewis acids).As with the corresponding styrene and butadiene systems the great enhancement in k,, brought about by the Lewis acid is evident. In a similar fashion we find that kab x 4933 mol-1 dm3 s-l; analogy with the MA/St system suggests that when Lewis acid is present k,, may have a larger value than this. Both cross-propagation coefficients are therefore sufficiently large to ensure effective alternation at normal monomer concentrations in the presence of Lewis acid. In conclusion, the kinetic data on all three systems MA/St, MA/Bd and MA/Ip are consistent with the cross-propagation mechanism of alternation.If no Lewis acid is present the propagation rn B' +A (B = hydrocarbon monomer) is relatively slow and effective alternation does not occur. We believe the enhancement in the rate coefficient of this process, which is the crucial feature in achieving alternation, arises essentially from the weakening of the carbon-carbon double bond which accompanies complexation of monomer A with Lewis acid. We are indebted to the Government of the People's Republic of China for financial support for one of us (X-z.H.). C. H. Bamford and X-Z. Han, J . Chem. Soc., Faraday Trans. 1, 1982, 78, 855. C. H. Bamford, S. N. Basahel and P. J. Malley, Pure Appl. Chem., 1980, 52, 1837. U.S. Patent, 3,840,449, 1974 (to J. Furukawa, E. Kobayashi, Y. Iseda and T. Yukuta). A. Akimoto and M. Ohtsuru, J . Polym. Sci., Polym. Chem. Ed., 1975, 13, 549. C. H. Bamford and S. U. Mullik, Polymer, 1973, 14, 38. 0. Eizo, T. Akiji and S. Tadamasa, Kobunshi Kagaku, 1973, 30, 22. W. S. Richardson and A. Sacher, J. Polym. Sci., 1953, 10, 353. J. C. Binder and H. C. Ranshaw, Anal. Chem., 1957, 29, 503. T. Inukai and T. Kojima, J . Org. Chem., 1966, 31, 1121. lo T. Inukai and T. Kojima, J. Org. Chem., 1967, 32, 872. l1 M. A. Golub and J. Heller, Tetrahedron Lett., 1963, 2137. l 3 C. H. Bamford in Reactivity, Mechanism and Structure in Polymer Chemistry, ed. A. D. Jenkins and l4 C. H. Bamford and P. J. Malley, J . Polym. Sci., Polym. Lett. Ed., 1981, 19, 239. l5 M. Morton, P. P. Salatiello and H. Landfield, J. Polym. Sci., 1952, 8, 279. l6 F. Ida, K. Uemura and S. Abe, Kagaku To Kogyo (Osaka), 1965,39, 565; cJ C.A. 1966,64, 3695a. I. Kossler, M. Stolka and K. Mach, J. Polym. Sci., Part C, 1964, 4, 977. A. Ledwith (John Wiley, London, 1974), chap. 3. (PAPER 1/645)