Search

The polymerization of styrene in benzene solution was studied at 60°C using a catalyst composed of α-titanium trichloride and [14C]triethylaluminium. The results obtained showed that the use of this system to measure the number of active sites on the catalyst is an unreliable method likely to over-estimate the number manifold. Isotope dilution experiments indicated that the contamination was due to the formation of both free and bound [14C]polyethylene. Use of tritiated methanol as a quenching agent to measure active centres indicated that this method yielded results much nearer to the true values. Evidence was also obtained to show that the main reaction governing polymer molecular weight was chain transfer with aluminium alkyl. Die Polymerisation von Styrol in Benzollosung wurde bei 60°C und mit einem Katalysator aus α-Titantrichlorid und [14C]-Triathylaluminium untersucht. Die dabei erhaltenen Ergebnisse zeigten, das es unzulassig ist, die Anzahl der aktiven Zentren an den Katalysatoten auf diese Weise zu bestimmen, da hierbei die Anzahl mannigfach uberschatzt werden kann. Versuche mit Isotopenverdunnungen zeigten, das die radioaktive Verunreinigung auf die Gegenwart von freiem und gebundenem [14C]-Polyathylen zuruckzufuhren war. Die Anzahl der aktiven Zentren kann besser bestimmt werden, wenn die Polymerisation mit tritiiertem Methanol abgebrochen wird. Die dadurch erreichten Werte sind viel realistischer. Es wurde auch bewiesen, das die entscheidende Reaktion die das Molekulargewicht des Polymers bestimmt, eine Ubertragungsreaktion mit Aluminiumalkyl ist.

Vinyl chloride has been polymerized by a modified Ziegler-Natta type catalyst VOCl2 · 2THF–Al(i-Bu)3–THF, and the kinetics of the polymerization reaction have been investigated in some detail. The present kinetic analysis has demonstrated that the generation of active centers is directly proportional to the catalyst concentration, and that their depletion is by a second-order decay process. Triisobutylaluminum has been found to contribute to the effective removal of active centers through complex formation.

The 220-MHz proton magnetic resonance and infrared spectra of stereoregular polypropylenes polymerized with a number of Ziegler-Natta catalysts and isotactic polymers of low molecular weight obtained by thermal degradation of a highly isotactic polypropylene were measured in an attempt to obtain some information on the local regularity. The fraction of thermally degraded polymer soluble in diethyl ether shows stereorandomness (tactie sequence length is quite short), and the portion soluble in n-pentane has stereoblock character. The results so obtained provide strong evidence that racemic dyads of whole polymer consist of two models of racemic dyad isolated and racemic dyads in groups. The polymers prepared with vanadium catalyst systems show stereorandom character and these polymers have groups formed by two propylene units in a tail-to-tail linkage. Syndiotactic polypropylene has head-to-head and tail-to-tail arrangements of two propylene units and this is the origin of randomness of syndiotacticity.

The influence of triphenylphosphine and organoaluminium additives of various composition on the activity of titanium dichloride in polymerization of ethylene and propene as well as on the structure of the polypropylene obtained is studied. When these additives are introduced into highly active samples of TiCl2, the polymerization rate decreases, whereas the stereospecificity of the catalytic system increases. On the basis of these and earlier data, conclusions are drawn about the composition of active centers and about the role of organometallic cocatalysts in formation of active centers, in chain propagation, and in formation of stereoregular polymer. Der Einflus von Zusatzen von Triphenylphosphin und aluminiumorganischen Verbindungen verschiedener Zusammensetzung auf die Aktivitat von Titandichlorid bei der Polymerisation von Athylen und Propen sowie auf die Struktur des erhaltenen Polypropylens wird untersucht. Wenn diese Verbindungen zu hochaktiven Titandichlorid-Proben zugesetzt werden, nimmt die Polymerisationsgeschwindigkeit ab, wahrend die Stereospezifitat des Katalysatorsystems zunimmt. Ausgehend von diesen und fruheren Ergebnissen wurden Schlusfolgerungen uber die Zusammensetzung der aktiven Zentren und die Rolle der metallorganischen Cokatalysatoren bei der Bildung von aktiven Zentren, beim Kettenwachstum sowie bei der Bildung des stereospezifischen Polymeren gezogen.

The activating effect of triethylamine in the VCl 3 AlR 3 system has been explained in terms of a previously proposed kinetic scheme. Activation is observed at low amine concentrations and is thought to occur by removal of adsorbed aluminium alkyl species from the catalyst surface. Deactivation occurs at higher donor concentrations and is believed to be due to competitive adsorption of donor with monomer. These proposals are consistent with the rate, molecular weight and polymer tacticity findings of other workers in Ziegler-Natta systems where the donor is capable of interaction with both the transition metal and metal alkyl components.

A kinetic scheme is postulated for Ziegler-Natta polymerization where propagation is considered to occur between adsorbed monomer and an active centre which has been formed by interaction of the metal alkyl with the transition metal halide. Chain transfer is considered to occur with adsorbed metal alkyl and also with adsorbed monomer. Rate, molecular weight and active centre dependencies are derived which are in qualitative agreement with the results of previous studies in Ziegler-Natta polymerization.

The kinetics of the polymerization of 4-methylpentene-1 by the Ziegler-Natta system VCl 3 AlR 3 have been examined at 30°C and 40°C. The dependence of the overall rate of polymerization on the concentration of metal halide, aluminium alkyl concentration, monomer concentration, temperature, and nature of the aluminium alkyl has been established. The energy of activation of the overall polymerization process has been found to be 16·6kcal/mol in benzene and 13·7kcal/mol in heptane. The results have been examined in relation to a previously described kinetic scheme.

The crystal structure of trans-1,4-polybutadiene has been reported by Natta et al. to be pseudo-hexagonal below its solid-phase transition temperature. In the present X-ray structure analysis, it is revealed that the crystal belongs to the monoclinic system with the space group P21/a. The unit cell, with the lattice constants a = 8.63 A, b = 9.11 A, c = 4.83 A, and β = 114°, includes four molecular segments.

Considerable progress has been made in the past 20 years in the synthesis, characterization and fabrication of plastics. Previous SPE Award winners, such as Mark, Natta, and Marvel dealt largely with synthesis; Flory with characterization; Alfrey and Du Bois with fabrication. One of the still unsolved problems lies in the realm of relating mechanical properties, such as impact strenght and creep to molecular structure. The design enginner who wishes to use a plastic part is concerned primarily with how some property such as impact strenght varies with temperature, speed of test, test method, etc. The polymer scientist must know why. Through knowing why, he may be able to design better plastics. This paper attempts to survey some of the world-wide progress made in this area in the past 10 years. The ultimate goal is to understand these mechanical properties in terms of internal molecular motions which occur in solid polymers. Internal motion can be detected by electrical, electromagnetic and dynamic mechanical measurements. When these three methods are applied on a given polymer over a range of temperatures, insights can be gained as to the variation of impact strength and other properties with temperature and frequency for that same polymer. These three fundamental methods, which require very small samples (less than 50 grams), can provide insight into the practical behavior of plastic materials over the wide range of temperatures and frequencies encountered in the real world.

The initial rate and the aging effect of ethylene polymerization with TiCl4- and TiCl3-AlEt3 catalyst were investigated. The initial rate, r, was expressed by the equation; r=k1PE+k2PE2, where pE is the ethylene partial pressure and where k1 and k2 are constants. The explanation of the equation is as follows:&(active site)+M\xrightarrowk_1(polymerization site), rate, k_1P_E&(polymerization site)+M^*\xrightarrowk_2(propagation), rate, k_2P_E^2M shows the ethylene monomer which is liberated by the catalyst formation or which is introduced into the system. M* is the monomer which is introduced into the system. The initial rate decreased in the TiCl4-AlEt3 system with the aging-time interval; on the other hand, in the TiCl3-AlEt3 system the rate showed its maximum at about half an hour’s aging. The optimum aging temperature was found to be 20°C. The decrease in the rates between the two systems may be ascribed to the reduction of the Ti atom in the initial valence state of the reaction to the lower valence.

The polymerization activity of the Ziegler-Natta catalysts is significantly affected by the aluminium alkyls in the liquid phase. These may be studied provided the solid phase of the catalyst remains unchanged. Part of these aluminum alkyls is adsorbed on the surface of the solid phase. It is possible to displace one type of alkyls by another one or to elute it from the adsorbed layer. This last affects the polymerization rate.

This work is devoted to the infrared study of the products of the copolymerization of propylene and vinyl chloride with the modified TiCl3–Et3Al catalyst. Thus, in our view the data given demonstrates two important features of the copolymerization process of propylene and vinyl chloride on the modified Ziegler-Natta catalyst: 1. that real copolymerization takes place in the case of this monomer pair; and 2. that copolymerization procedes on the stereospecific active sites of the heterogeneous catalyst, i.e., on the same sites which are responsible for the stereoregular polymerization of olefines. Infrared study allows one to estimate the distribution of the propylene units in the copolymers. The distribution is very typical of the random copolymers obtained with the stereospecific catalysts.

Dilatometry was used to measure the rates of copolymerizations of styrene with hex-1-ene, hept-1-ene, 4-methylpent-1-ene and 4-methylhex-1-ene, in the catalyst system, AlEt3-αTiCl3 in toluene. The copolymerizations were found to be first order with respect to TiCl3 and independent of Al/Ti ratio. Increased temperature produced a rapid increase in rate; variation of monomer feed, over the whole range, resulted in a gradual change from one homopolymerization rate to the other. The nature of the α-TiCl3 sample was found to have considerable effect on the rate of copolymerization. The order of reactivity, both from rate measurements and reactivity ratios, implies that, for the polymerization of α-olefins by heterogeneous Ziegler-Natta catalysts, steric effects outweigh electronic consideration.

1. A. I. MAREI, G. Ye. N0VIKOVA and N. P. KUZNETSOV, Kauchuk i rezina, No. 10, 7, 1964 2. Ye. V. KUVSHINSKII and Ye. A. SIDOROVICH, Zh. tekh. fiz. 27: 702, 1957 3. V. M. GORELIK, A. I. MAREI, lYl. F. BUKHINA, G. Ye. NOVIKOVA and B. A. POMIRCHAYA, Kauchuk i rezina, No. 11, 13, 1964 4. G. NATTA, G. CRESPI, G. GUZZETTA, S. LEGHISSA and F. SABBIONI, Rubber and Plastics Ago 42: 42, 1961 5. J. L QUINNDEN and F. W. SHIPLEY, J. Polymer Sci. 34: 77, 1959 6. G. S. TRICK, J. Polymer Sci. 31: 529, 1958

The stereospecific polymerization of propylene with Natta catalyst under ordinary pressure has been investigated and discussed. Propylene of 99.5% purity was polymerized with a metal-organic complex catalyst, formed from A1(C2h5)3 and TiCl3, giving a white or greyish stereopolymer which contained ca. 70% isotactic polypropylene. The optimum conditions of polymerization were as follows: The mole-ratio of Al(C2H3/TiCl3 was 8–15; the polymerization-temperature was 70°C, and the concentration of TiCl3 was 0.6–0.8g per 120ml of n-heptane. The maximum polymerization rate reached the value of 6g C3H6/hr./g TiCl3 and the apparent activation energy as calculated from an Arrheniusplotting line, was 12.1 kcal/mole.

The X-ray diagrams of polypropylene specimens of widely varying properties have revealed the existence of three modifications besides the normal modification studied by Natta and co-workers. One of these modifications has been described in the literature. For this modification a modified tentative structure is proposed which is more consistent with the available data than the tentative structure suggested in the literature.

I. A. I. KITAIGORODSKII, Tetrahedron 9: 183,1960; 14: 230, 1961' A. I. KITAIGORODSKII and K. V. MIRSKAYA, Kristallografiya 9: 174, 1964; A. I. KITAIGORODSKII, Acta Crystallogr. 18: 585, 1965; A. I. KITAIGORODSKII, J. Chem. Phys. 63: 9, 1966; V. G. DASHEVSKII, Yu. T. STRUCHKOV and Z. A. AKOPYAN, Zh. strukt, khimii 7: 594, 1966 2. A. I. KITAIGORODSKII and K. V. MIRSKAYA, Kristallografiya 9: 634, 1964; 10: 162, 1965 3. G. NATTA, P. CORRADINI and P. GANIS, J. Polymer Sci. 58: 1191, 1962; P. De SANTIS, E. GIGLIO, A. M. LIQUORI and A. RIPAMONTI, Nuovo cimento 26: 616, 1962; J. Polymer Sci AI: 1383, 1963; M. IWASAKI, J. Polymer Sci. AI: 1099, 1963 4. P. De SANTIS, E. GIGLIO, A. M. LIQUORI and A. RIPAMONTI, Nature 206: 456, 1965; C. RAMAKRISHNAN and G. N. RAMACHANDRAN, Biophys. J. 5: 909, 1965; S. I. LEACH, G. NEMETHY and H. A. SCHERAGA, J. Biopolymers 4: 887, 1965 5. T. M. BIRSHTEIN and O. B. PTSITSYN, Konformatsii makromolekul (Makromolecular Conformations). Science Publishing House, 1964 6. T. L. KHOTSYANOVA, A. I. KITAIGORODSKII and Yu. T. STRUCHKOV, Zh. fiz. khimii 27: 781, 1341, 1953; A. TULINSKY and J. G. WHITE, Acta Crystallogr. l h 7, 1958; G. GAFNER and F. H. HERBSTEIN, Acta Crystallogr. 13: 702, 1960 I. N. STREL'TSOVA and Yu. T. STRUCHKOV, Zh. strukt, khimii 2: 312, 1961; I. N. STREL'TSOVA and Yu. T. STRUCHgDV, Izv. Akad. Nauk SSSR, Otd. khim. n. 250, 1961

Results of light scattering as well as viscosity measurements on a series of isotactic polystyrene fractions, and on a sample of unfractionated isotactic polymer in toluene at 30°C. are reported. The results are in agreement with the conclusion of Natta, Danusso, and Moraglio that no difference exists between the viscometric behavior of atactic and isotactic polystyrene; the same seems to hold also for the molecular dimensions. The ratio A2M/[η] for isotactic polystyrene is lower than for atactic fractions, and according to the results of Danusso and Moraglio, the second virial coefficient, A2, is probably responsible for the observed difference. The data obtained with the unfractionated sample can be explained in terms of an unusually high Mw/Mn ratio.

水-アルコール系(30:70容量比)および酢酸-水-アルコール系(30:30:40容量比)溶媒中でアクリロニトリル-塩化ビニリデンの共重合反応を行なつたところ,アルコールの種類によりモノマー反応性比が系統的に変化することがわかった。使用したアルコールはメタノール,エタノール,イソプロピルアルコール,tert-ブチルアルコール,n-プロピルアルコール,イソブチルアルコール,n-ブチルアルコール,イソアミルアルコールの8種で,ほぼ上記の順序に従って,水-アルコール系ではアクリロニトリルを第1モノマーとした場合r1は0.63より0.65まで,r2は1.7より0.49まで,また酢酸-水-アルコール系ではr1は0.7から0.44まで,r2は1.8より0.40まで漸次変化した。この原因は,アルコールの種類が変われば,モノマーに対する溶媒和が変わり,このためにモノマーの共重合体粒子上への吸着量が影響をうけ,したがって見掛けの反応性比が変わることにあるのではなかろうかと考えた。

Some features of the molecular weight distributions found by experiment for low pressure polyethylenes are examined; an analytical interpretation of them is suggested which is in agreement with the polymerization mechanism resulting from kinetic investigations of Natta and co-workers. Such interpretation is based on the consideration of the distribution of the individual times of growth (or of the “lifetimes”) of the single macromolecules. It is assumed that (a) these lifetimes tend to be distributed according to the Gauss formula, such as a series of events tending to be distributed at random about a mean; and (b) the growth of any single macromolecule proceeds at a rate which can be considered as the rate of a chain reaction. Possible deviations from such simplified scheme are taken into account. A distribution formula of a generalized Lansing-Kraemer type is deduced, giving an explanation of all the observed features of the molecular weight distribution.

Trans-1,5-polypentenamer can be produced with a high degree of uniformity by solution polymerization using complex catalysts of the Ziegler/Natta type. The state of order of the polymer indicates a rubber with good low temperature properties and crystallization behavior similar to that of natural rubber. There is therefore a fundamental difference between this material and other synthetic rubbers of the general purpose type. Trans-1,5-polypentenamer is easy to process, can be heavily loaded, and is therefore inexpensive to compound. Even when small amounts of the crosslinking agents are used, a high cross-linking yield is obtained and this makes it possible to produce vulcanizates that are resistant to dynamical stresses. As high green strength and good tack are present simultaneously, the material is suitable for use in the tire industry, particularly in carcass production.

トリエチルアルミニウム- 塩化チタン(III)触媒によるプロピレンの重合反応において,各種の不飽和炭化水素化合物を加えてその影響を比較検討した。これらの研究から,モノオレフィン化合物は顕著な影響を与えないが,ジエン類,アセチレン類は微量の添加で著しく重合を阻害することがわかった。阻害作用の大きさはプロパジエンが最も大きく,メチルアセチレン, アセチレンがこれにつぎ, ブタジエンがこれらの中で最も小さい。これらの阻害作用は一時被毒であり,回復する。この阻害作用の大きさと速度の回復の速さとの間には相関性はなく,各成分に固有である。一方,添加成分によって重合が実質的に禁止されているような阻害の大きな条件下で,微量生成する重合体はプロピレンを主体とするものであることが認められた。これは重合の阻害が添加成分の独占的な重合によるものでないことを意味する。これらの重合の阻害の機作を共重合反応の概念で説明しようと試みたが,生長反応の速度恒数の関係からは十分解釈できないものと考えられた。

A mechanism for the polymerization of olefins is proposed. An essentially octahedrally coordinated ion of a transition element with empty or nearly empty t2g orbitals carrying in its coordination sphere one alkyl group and having one vacant octahedral position is supposed to be the active center. In this model the monomeric olefin is coordinated in the vacant position through a “π bond.” It is shown that theoretically one may expect the transition metal-to-carbon bond to become more susceptible to radical breaking at the very moment the π bond between metal ion and olefin is formed. Based on this theoretical argument, the proposed mechanism: involves to a first approximation an electronic rearrangement and only small nuclear displacements. This mechanism may account for the low activation energy of the propagation step. It further relates catalytic activity to the ionization potential for a d electron of the transition metal and the nature of the negative ions used in the transition metal compound. It also provides a basis for the explanation of the formation of isotactic polymers when special solid catalyst systems are used.

The reaction of styrene with the following 5 complexes has been examined. I was a complex of empirical formula C4H12O2TiAlCl2, which contained OH and alkyl groups, but not C5H5. II was NATTA'S complex (C5H5)2TiCl2Al(C2H5)2 and IIa the diisobutyl homologue. IIIa was the SHILOV complex which has the reputed structure (C5H5)2TiH2Al(i-C4H9)2, and III was its diethyl homologue. I, II, and IIa, or their derivatives, initiate styrene polymerisation. III and IIIa do not polymerise styrene and inhibit radical reactions. The effect of additives on polymerisations initiated by Complex II was examined. Oxygen accelerates the polymerisation. Water and methyl methacrylate inhibit polymerisation and destroy the catalyst. — The kinetics and mechanism of these polymerisations are discussed.

The properties of stereoregular polypropylene have been modified by the introduction of controlled branching into the polymer chain backbone. Branching is achieved by copolymerizing propylene with certain long-chain 1-olefins by using a standard Ziegler Natta low pressure catalyst. Physical and mechanical properties of the copolymers are related to the degree and length of branching. It was found that ultimate strength properties of moderate molecular weight polymers could be improved at some expense in short-range deformation properties by adjusting the degree of branching to a suitable level. The copolymers did not show a good combination of low temperature brittleness characteristics along with higher temperature strength properties. It is suggested that the branching technique described will allow the tailoring of polypropylene resins for specific end use applications.

プロピレンの定常重合速度の実験値と三塩化チタン触媒の実測表面積とから得られる単位表面積あたりの頻度因子の値が, プロピレンの気体運動論的表面衝突のそれにきわめて近いことを見出した。この事実を手がかりとして, 可能な反応モデルを絶対反応速度論的に考察した。最もわかりやすいモデルは, あたかも気相から表面に衝突したように, 溶解プロピレンが分子状で, 全表面をおおっている重合中心に衝突するTwigg-Rideal型である。重合中心への吸着プロピレンのアタックを考える説は, きわめて特殊な吸着を考えないかぎり実験と合わないことなど, 通常考えられているモデルと実験との比較検討も行なった。

トリエチルアルミニウム- 塩化チタン(III)触媒によるプロピレンの重合反応におけるジェン類・アセチレン類の重合阻害作用について検討した。プロパジエンを添加した重合で, まずプロピレンとの添加順序の影響について明らかにした。また温度の影響から阻害は高温ほど小さくなることが認められ,阻害の機作が可逆的な物理過程にあることが示唆された。一方,各種ジエン類・アセチレン類の添加濃度と阻害作用の大きさの関係について検討し,阻害の機作は阻害モノマーとプロピレンの共重合あるいはMaxted型の被毒の式によって説明できないことが認められた。それに対し,Langmuir型の吸着により阻害が起こるとすると関係式がよく適合することが明らかになった。そして,これまでの実験結果はいずれも添加成分の活性点への選択的な吸着により阻害が起こるとしてうまく説明されることがわかった。各成分に対する:Langmuir式の恒数を求め阻害の大きさを比較するとともに,塩化チタン(III)の表面積との関係について考察した。また,イオウ化合物の阻害作用についても同様な吸着の機作が考えられ,興味が持たれる。

The number of active centres in the polymerization system VCl3/AIR3/4-MP-1 has been shown to lie in the range 2·3–6·1 × 10−4mol/mol VCl3 for a series of aluminium alkyls, and to vary with the nature of the aluminium alkyl used. Chain transfer with aluminium alkyl has been demonstrated. The propagation rate constant has been evaluated and found to be independent of the nature of the metal alkyl. The location of, and nature of, the active centre is discussed. The validity of the tritium labelling procedure is tested by comparison of calculated and experimentally determined molecular weights.

The surface product formed in the reaction between TiCl3 and Al(CH3)3 has been studied. Stoichiometric data, CH3/CD3 exchange, and infrared spectra permit the conclusion that the surface product is essentially a compound having the formula A model structure is proposed for this compound, valid for the 001 face of α-TiCl3. In it the titanium and chlorine atoms maintain the positions which they occupy in the α-TiCl3 lattice. One of the methyl groups protrudes from the surface whereas the other occupies the chlorine vacancy created during the reaction in the chlorine surface layer. A different sterism of the methyl groups is compatible with the experimental result that half of the methyl groups are very easily exchanged whereas the other half are not touched by the exchanging agent. According to this model it has to be assumed that the titanium atoms in the 001 plane, by far the largest face of the α-TiCl3 crystal, are not accessible. A similar model, loading to equivalence conclusions is proposed for β-TiCl3. The infrared spectra of Al(CH3)3, Al(CD3)3, AlCl(CH3)2, AlCl(CD3)2, AlCl2CH3, AlCl2CD3, TiCl3CH3, TiCl3CD3, Hg(CD3)2, and Zn(CD3)2 are discussed. Spectra of surface products formed on interaction of some of these compounds with TiCl3 are given.

The following reactions, carried out in the absence of solvents, has been studied: α-TiCl3 + Al(CH3)3 at 20°C., β-TiCl3 + Al(CH3)3 at 65°C., α-TiCl3 + Al(CH3)2Cl at 20 and 65°C., and α-TiCl3 + Al(C2H5)3 between 30 and 65°C. It appears that a general reaction mechanism, such as discussed in the preceding paper of this series, applies to all these reactions between TiCl3 and aluminum alkyls. The differences in overall stoichiometry between some of these systems may be linked to differences in stability of the intermediate TiC bonds. In the case of α-TiCl3 + Al(CH3)2Cl, alkylation is probably accompanied by fixation of the AlCH3Cl2 on the nonvolatile product.

TiCl3 (AA)-(C2H3) nAlCl3-n (n=3, 2, 1.5) 系触媒によってプロピレンを重合するとき, 重合前に触媒を95-155℃で加熱処理してその効果を比較した。加熱処理しないときの活性はnの大きいものほど大きいが, 加熱処理による活性増大の比率はnの小さいものほど大きくなった。n=3のときは, 加熱処理の結果活性はかえって低下するが, TiCl3 (AA) の代わりにTiCl3 (H) を用いると活性が増大した。熱処理に伴って液中のアルキルアルミニウムの全濃度は変化しないがアルキル基が塩素でおき換えられる。このような実験事実を三塩化チタンのアルキル化反応と結びつけて説明した。重合反応の活性中心はアルキルチタン二塩化物とアルキルアルミニウム化合物との錯合体であると考えると実験事実の説明が容易である。

The kinetics of isoprene polymerization catalyzed by VCl3 and Et3Al were studied by measuring fractional conversions, polymer composition, and molecular weight distributions at a series of reaction times and temperatures. The rate of polymerization plotted against temperature shows an inflection point with a minimum and maximum in the 60–90°C range. The isomeric composition of the polymer changes with temperature but not with reaction time, while the molecular weight distribution undergoes substantial change with both of these variables. The rate of polymerization at sites producing low molecular weight polymers was measured, and the activation energy calculated to be about 10 kcal/mole. The active sites were found to deactivate at different rates. The results support the hypothesis that several species of active sites are present in the system and that these exhibit characteristic polymerization behavior.