Search

Paper chromatographic analysis of the complex polymetaphosphates, i.e., (Nax-M(1−x)2PO3)n, where M is Ca, Ba, or Cu and x = 1/3 reveals that the observed values of Rf and Rg for the above derivatives are similar to those for the simple sodium polymetaphosphate (NaPO3)n, but are different from those of the phosphate anions viz., orthophosphate, pyrophosphate and trimetaphosphates.

Re-examination of some data which were originally taken to determine diffusion coefficients and their variation with applied stress in electron irradiated poly(bisphenol-A carbonate) and poly(methyl methacrylate) has verified the existence of Kishi induction periods ti for the initial stages of O2 sorption by these polymers. The data conform to the relation . Some new gas sorption data for poly(bisphenol-A carbonate) by Norton are utilized in conjunction with data and equations from our previous paper to estimate initial concentrations of the free radicals associated with the visually sharp diffusion boundary which was the major topic of our previous paper. The value G[oxygen-bleached color-centers/h.e.v.] obtained for these radicals is compared with recent viscometric determinations by Golden and Hazell.

The previous papers in this series contain the experimental results which demonstrated the complexity of the homogeneous, nonterminated, anionic polymerization of methyl methacrylate. In order to explain the observed broad molecular weight distributions, it was necessary to postulate reversible “pseudo-termination” reactions, by which a large fraction of the total anions present in the system became inactive toward further polymerization, yet had a finite probability of recovery of polymerization activity by some process, such as ligand exchange, ring opening of a cyclic complex, etc. In this paper, we have examined the quantitative consequences of a mechanism in which this pseudotermination reaction occurs by a bimolecular exchange of activity between one active and one inactive chain. It has been possible to calculate Mw/Mn, the ratio of weight-average to number-average molecular weights, as a function of the fractional conversion α, the initial monomer to initiator ratio, M0/I0, the fraction of total initiator as active chains, C1/I0, and the ratio of rate of exchange of activity to rate of propagation, kt/kp. The major results of these calculations are: (1) any value of Mw/Mn between 1 and infinity is possible at α = 1; (2) Mw/Mn at α = 1 decreases as kt/kp increases; (3) Mw/Mn at α = 1 increases as M0/I0 increases when kt/kp and C1/I0 are small—under some conditions Mw/Mn at α = 1 increases and then decreases slightly as M0/I0 increases; (4) Mw/Mn goes through a maximum during polymerization if kt ≠ 0; (5) Mw/Mn at any α decreases as C1/I0 increases, for constant M0/I0 and kt/kp. Concentrations of individual chains of any given length as a function of time may be readily calculated for the two special cases kt/kp → 0 or ∞.

Intrinsic viscosities [η] and Huggins' constants k are generally determined by means of one of the following equations: ((1)) ((2)) ((3)) It was proven in an earlier paper that eq. (1) is an insufficient approximation of eq. (2) in the usual range ηr = 1.2–2.0. In this paper it is shown that eqs. (2) and (3) can lead to different Staudinger indexes (difference up to 30%). Huggins' constant k can vary from one equation to the other by as much as 300%. The experimental test of theoretical relationships between [η], k, and other molecular constants can lead to errors of the same magnitude. Values of [η] and k in pure and mixed theta solvents are discussed. It is moreover shown that eq. (2) is an inadequate approximation of eq. (3). Therefore eq. (3) is the only one which should be used.

In a previous paper the end-to-end dimensions of semicrystalline macromolecules were computed by the Markoff chain method, by use of a normalized probability matrix with two probability parameters α and p. The values of these parameters that actually obtain may result from kinetic conditions (as in rapidly quenched specimens) or the values may in some cases be equilibrium values that can be determined by statistical mechanical considerations. In this paper we calculate the overall partition function Q of a semicrystalline macromolecule by a method similar to that employed by Zimm and Bragg for polypeptide chains. Q is computed in terms of segment partition functions fr, fh, and fk. By working out the consequences of this Ising-type model, we are able to relate α and p to fr, fh, and fk. In an appendix we show that the phenomenon of ceiling and floor temperatures in equilibrium polymerization can be regarded from the same mathematical point of view as the helix–coil transition. Moreover, the mathematical results from this approach can be correlated with the results previously obtained by Tobolsky and Eisenberg by direct application of the law of mass action.

The effect of oxygen on ESR spectra of various kinds of irradiated polymers has been examined by using a saturation technique developed in the present investigation. The radiation oxidation of some polymers has been studied by means of ESR as well as infrared spectroscopy. Results for polyethylene (PE) are given in this paper, with some general remarks as introduction. In PE three kinds of radicals are produced on irradiation: CH2CHCH2 (FRVI), CHCHCH (FRVII), and CH(CHCH)nCH2(FRI). On introduction of air, FRVII gave the asymmetric singlet of the peroxide radical which showed anisotropy. Reversible oxygenation of FRVII took place even as low as −113°C. Radiation oxidation of PE proceeds mainly via FRVII, which produced eventually about 12 carbonyl groups and about 5 hydroxyl groups in the course of post-oxidation, the trans-vinylene group decreasing by an amount nearly equal to the amount of initial FRVII. FRVI played only a minor role in the oxidation. An oxidation mechanism involving chain reaction has been proposed. It suggests an intramolecular free-back reaction of hydrogen to the peroxide radical of FRVII. FRI (polyenyl radicals of longer conjugations) is oxidized to form carbonyl and hydroxyl groups, although it does not give the spectrum of the peroxide radical. Polyenyl radicals of shorter conjugations reacted faster.

In this paper an attempt was made to explain the crosslinking mechanism of PVC, when irradiated in the presence of tetrahydrofurane. Taking into account several experimental results, a very likely conclusion seems to be that a bridge between two PVC molecules consists in a THF molecule, of which the ring has been opened by the combined effects of the irradiation and of the chlorine or hydrochloric acid evolved from the PVC.

The combinatorial approach to the problem of linear polymers, developed earlier, is applied in the present paper in order to study the statistical nature of condensation polymers obtained from a large number of very small systems. The results of the customary theory become inapplicable in such cases for sufficiently high degrees of polymerization which should be attainable experimentally. The possibility of realizing homogeneous samples by direct polymerization is discussed from a theoretical standpoint.

This paper is an extension of an earlier work in which thermal pyrolysis gas chromatography and differential thermal analysis were combined to elucidate comonomer distribution in ethylene–acrylate copolymers. In this work consideration is given to the thermal degradation products of ethylene–acrylate and ethylene–vinyl acetate copolymers. The influence of cracking temperature is considered in terms of the utility of the data. The DTA measurements are primarily concerned with the crystallization of the ethylene chains between the branch points. Indeed, a linear relationship between DTA crystallinity and branching is indicated.

The free energy density of a folded-chain crystal varies with the length of the straightchain section between consecutive folds. It is always minimum at infinite length. Another minimum at finite length L appears below a critical temperature T0 which among other things depends on the surface energy of the fold-containing planes. This minimum is a result of the smearing out of the nonharmonic lattice potential energy field by the incoherent vibrations of adjacent chains. In the present paper the simultaneous influence of both longitudinal and torsional vibrations of polyethylene chains in the orthorhombic lattice is considered. Due to the relative smallness of the amplitude of the potential energy for chain translation in comparison to that for chain rotation the calculated equilibrium length L at any temperature does not appreciably differ from the value obtained formerly for the torsional vibrations only. L increases with temperature and is maximum at the critical temperature. With increasing surface energy the equilibrium long period at constant temperature increases, and the critical temperature decreases. The calculated values fit fairly well the experimental data of polyethylene crystals grown from dilute solution if one admits that the surface energy varies slightly from solvent to solvent.

This paper describes the H → D exchange reaction in proteins and polyamides immersed in saturated deuterium oxide vapor. Ten proteins (seven silk fibroins, wool, byssus, and Fibrolane BX) and the nylons 66, 6, and 11 were studied. With all the samples there was an initial rapid exchange followed by a slower exchange, this slow change gradually levelling off to give “equilibrium” values. The extent of the H → D exchange is discussed in relation to order and disorder in the polymer structures. The extent of exchange is shown to be related to the proportions of large side chains in the samples and also to the water sorption values. It is shown that the decrease in the absorbance of the 3300 cm.−1 band of the polyamide films is a reasonably satisfactory measure of the extent of the H → D exchange in these films, provided that changes in band width are taken into account.

Specific diluent effects on the unperturbed mean-square end-to-end distance may be determined in dilute solutions at their Θ-points. Another method to determine these effects is described in this paper. It involves the measurement of the modulus of crosslinked polymers swollen with different diluents and interpretation of the results according to current rubber elasticity theory. As a result of the relatively high concentrations used in this method, combined with the introduction of crosslinks, specific diluent effects on chain dimensions are obtained, unobscured by excluded volume effects. The method is applied to polyethylene diluted with di-2-ethylhexyl azelate, α-chloronaphthalene, n-paraffin, tetralin, and n-hexadecane. Within 10% the value of in n-hexadecane is equal to that in the other diluents used. In view of the similarity of the polyethylene molecule to those of n-hexadecane and n-paraffin this value must then also be equal to that for polyethylene in bulk. It may be concluded that, within experimental error, semi-ordering effects in the melt are absent and that no specific diluent effects exist.

A simple process is described for the preparation of condensation polymers from diamines and diacid halides, which comprises reacting equivalents of the intermediates at about room temperature in an inert solvent together with an acceptor for the by-product acid. This first paper of a series discusses the formation of polyamides from piperazines and aliphatic and aromatic diacid chlorides. Examples are poly(sebacyl piperazine) and poly(terephthaloyl dimethylpiperazine). Such variables are discussed as types of acid acceptors, solvents, reactant balance, interfering reactions, and the mechanism of the reaction. Some of the polymers are described briefly in terms of melt temperature, solubility, and molecular weight-viscosity relationships.

The chain conformation and configuration of some polysorbates and poly-β-styryl-acrylates is discussed in this paper. From x-ray fiber spectra of these polymers we obtained the identity period of 4.80 ± 0.05 A. The conformation of the main chain is the same as that of 1,4 trans polybutadiene and isotactic 1–4 trans polypentadiene. From considerations of the mode of packing of the chains and of the encumbrance of the side groups in respect to the main chain it was possible to assign an erythro configuration in regard to the relative placement of lateral groups. The examined polymers are tritactic and have a trans-erythro-isotactic configuration.

In a previous paper, it was reported that an acetaldehyde polymer of the poly-(vinylalcohol) type was obtained by polymerization of acetaldehyde with sodium amalgam as catalyst. In the present, more detailed investigation, it was found that the catalyst concentration, the reaction time, the reaction temperature, and the solvent concentration have certain effects on the molecular weight and the yield of polymer. At the early stage of this polymerization, the yield is proportional to the square root of the reaction period. So, at this stage, the reaction may be assumed to be diffusion-controlled. Several solvents and the two-stage polymerization have influences upon this polymerization.

Two cases of initiation of vinyl polymerization by radical-ions are considered. In the one case the transfer from radical-ion to monomer is instantaneous and virtually complete. Here the relative amount of free radical propagation as compared to anionic propagation is negligibly small. In the second case a continuous feed of radical-ion initiator is considered. Here the relative amount of free radical propagation as compared to anionic propagation may be considereable. Exact expressions are developed for this relative radical versus ionic propagation as a function of conversion. It is suggested that the results of lithium-initiated copolymerization of methyl methacrylate and styrene may be understood in terms of a model similar to the second case of radical-ion initiation considered in this paper.

In the present paper the effects of substituted derivatives, namely the NH2, OCH3, CH3, Cl, and CO2C2H5 derivatives of s-trinitrobenzene are described on the initiated polymerization of styrene. The investigations were extended to the determination of the relative reactivity k5/k2 of the inhibitor as well as of the value of stoichiometric coefficient. It was shown by the values of k5/k2 that the Hammett equation can be considered valid within a factor of two. Corresponding to the electron-donating properties of the employed substituents, the values of the stoichiometric coefficient in crease from 2.7 to 3.8. This fact as well as other data from literature lead to a mechanism by which these values, which are unexpected in the case of molecular inhibitors, can be interpreted.

Kinetic studies of cyclopolymerization have been reported in several papers to date, but the overall kinetics is not yet established. Therefore, divinylformal was synthesized and polymerized as a difunctional monomer, and the overall rate equation, Rp = k[I]3/4[M]3/2−2, was obtained. To interpret this irregular rate equation, elementary reactions were considered and the theoretical rate equation, Rp = k[I]1/2−1[M]2−1, was obtained. The overall activation energy of polymerization was 27.7 kcal./mole, and this value, together with the result of the measurement of residual unsaturation in the polymer, yielded a value for the difference of activation energy between cyclization and propagation reaction of 2.6 kcal./mole. The results of measurements of intrinsic viscosities and residual unsaturation in the polymer at various conversions suggest that the branching reaction occurs to some extent during polymerization. The polydivinylformal was hydrolyzed to poly(vinyl alcohol) with hydroxylamine hydrochloride, and the 1,2-glycol structures in the poly(vinyl alcohol) were determined. The result showed a higher content of 1,2-glycol structures than that in commercial poly(vinyl alcohol).

On the basis of results presented in previous papers possible mechanisms of polymerization of higher aldehydes are considered. The influence of solvents, temperature, initiator, and monomer upon the polymerization is discussed with particular emphasis on the stereoregularity of the resulting polymer. It is proposed that more than one monomer unit is involved in the propagation step. Both the stereoselective addition of monomer to the growing chain and the chain transfer reaction are thought to take place in a concerted manner. Chain transfer results in termination with the formation of hydroxyl, ester and probably ether end groups; growing chains which do not undergo transfer remain as “living” ends after the monomer supply is exhausted. Some thermodynamic values for carbonyl polymerizations are estimated.

The first paper in this series has shown that when acrylonitrile is polymerized in ethylene carbonate, the transfer-to-monomer reaction is negligible, and the solvent acts only as a diluent and as a transfer agent. Furthermore, the rate of polymerization is first-order to quite high conversions, even though there is a concomitant high viscosity of the polymerization media. These observations permit calculation of the effect of conversion on the average molecular weight of the polymer. The constants of this calculation were evaluated from low conversion polymerizations and applied to the case of high conversion. The measured weight-average molecular weights as a function of conversion agree with the calculation and indicate that transfer-to-polymer reactions are probably very small. Hence, the extent of branching in polyacrylonitrile prepared in solution is also very small even at high conversions.

This paper describes some of the aspects of the thermal oxidation of deuterated isotactic polypropylenes. Several polymers of propylene, deuterated at different sites along the polymer chain, were subjected to a kinetic treatment designed to establish the comparative rates of oxidation at different temperatures. Infrared spectroscopy was used to follow the kinetics over the temperature interval 100–130°C. These experiments have shown the existence of a deuterium isotope effect for the thermal oxidation of polypropylene.

The paper deals with theoretical and experimental data on polymeric products with semiconducting properties. The data show the possibility of producing, on the basis of polymers, particularly polyacrylonitrile, substances characterized by high electroconductivity and exponential dependence of the latter on temperature. Possible changes in electrophysical properties of the materials are discussed. A quasimetallic model of a polymeric semiconductor is described.

The first portion of the paper consists of a brief statement of the statistical procedures involved in obtaining efficient estimates of parameters from experimental data, determining their precision, and selecting optimum experiments. Emphasis is primarily directed toward models in which the parameters appear in a nonlinear fashion. Some of the difficulties and pitfalls which are encountered are discussed; primary among these are the problems arising from linearization schemes and transformations affecting the error structure. References are given which supplement the necessarily cursory discussion given here. Utilizing this background, the second half applies itself to the problem of estimating the reactivity ratios in the copolymer equation. Approximate and exact estimation schemes are described which both guarantee most efficient use of the data and allow objective probabilistic statements to be made about the reliability of the estimates. A graph is presented which allows the experimenter to select the two initial feed ratios for his experiments which provide the most information for the estimation scheme. Use of the graph presumes that order of magnitude estimates of the reactivity ratios can be made in advance. A sequential scheme can be followed when prior information is lacking or is not very precise. When more than two different initial feed ratios are to be used, the optimum ratios for two-point experiments give a good indication of the proper range to cover.

τm, defined as the maximum relaxation time, was shown in previous papers to be one of the elements characterizing the “box” type curve in the rubbery region of a linear amorphous polymer, and a new procedure called procedure X was proposed to obtain τm. The relation among τm, weight-average chain length nw, temperature T, and characterized temperature Ti (glass transition temperature Tg, WLF temperature Ts, or Tobolsky's temperature Td) was indicated by an equation, with which constant values of log A depending on the kind of polymer were obtained for several polymers in the regions of M ≫ Mc. For polymers in the vicinity of Mc, another set of constant values of log A′ was obtained. Finally, the effect of τm on temperatures for a polymer was studied.

Polymers obtained by reaction between formaldehyde and carbon monoxide under anhydrous conditions in presence of Lewis acids are described. Mass balance and elemental analysis show that the CO/CH2O molar ratio in the polymers can reach unity as maximum value. Ammonolysis of the products (A) having a CO/CH2O molar ratio less than 1 gives glycolamide, diglycolamide, and hexamethylenetetramine. Ammonolysis of the products (B) having a CO/CH2O molar ratio about 1 gives glycolamide, diglycolamide, and traces of hexamethylenetetramine. These results show that the chain of polymer A contains the structural units: COCH2O, COCH2OCH2COO, and CH2O, while polymer B contains almost exclusively the structural units COCH2O and COCH2OCH2COO. Infrared spectra of the products are in good agreement with these results and show the presence of ester, ether, and acetal bands in the main chain of the polymers. The polymer structure is described, and a reaction mechanism is proposed.

Linear and branched dextran polyelectrolyte reduced viscosity–concentration curves, obtained by employing a solvent which contained a low concentration of simple electrolyte, were characterized by means of a set of relative expansion ratios, . The sets of for branched and linear dextran polyelectrolyte samples of identical degree of substitution, whose reduced viscosity–concentration curves were determined in the same solvent, did not differ. This is interpreted as indicating that the same expansive mechanism operated in the linear and branched polyelectrolyte systems when they were diluted. Plots of versus absolute concentration for linear and branched polyelectrolyte samples (of identical degrees of substitution, of different degrees of substitution and whose reduced viscosities were determined in solvents of different simple electrolyte concentrations) gave straight lines with a common intercept on the ordinate axis but with different slopes. It is suggested that the slope factor or p values of these lines might be employed as a means of detecting and measuring short branching in polymers convertible to polyelectrolytes.

The theory of Kuhn and Kuhn for the diffusion of polymer chain ends has been applied to the termination mechanism for free radical polymerizations, proposed by Benson and North. By using the dimensions for a Gaussian chain, an expression for this diffusion-controlled termination rate constant in terms of solvent viscosity, polymer chain flexibility, and chain length has been derived. The theory predicts that the termination rate constant is inversely proportional to solvent viscosity for flexible polymer chains and independent of solvent viscosity for very stiff polymer chains. Several previously reported effects of solvent viscosity on the termination rate constant may be explained with the present theory. The effect of polymer chain length on the termination rate constant depends upon the range of chain lengths considered and upon the relative chain stiffness. The necessity of taking into account the frequency distribution of radicals which exists during polymerization is discussed. Calculated values of the termination rate constant have been carried out with the use of values for the various parameters which are believed to be typical of vinyl polymers in solution. These calculated values are in general agreement with those which are usually obtained experimentally.

It is shown that it is possible to predict the reactivity ratios for the copolymerization of a given monomer pair if each monomer has been separately copolymerized with a third monomer, and the product of the reactivity ratios of one of these separate copolymerizations is equal to unity. This latter condition is frequently met in ionic copolymerizations and is sometimes true in free radical copolymerizations. Literature data for five trios of monomer are examined to show the validity of the predictions.

Polymeric material was produced by passing aqueous glucose solutions under a beam of high energy electrons. The yield of polymer increased with dose and dose rate. The weight-average molecular weights of polymers formed at 40 Mrad, measured by light scattering, ranged from 9,800 to 5,000,000. The polymers exhibited very low intrinsic viscosities in water, and are therefore though to be highly branched. Measurement of the number-average molecular weight of one of the polymers by osmometry established an Mw/Mn ratio of 4.9, indicating a fairly broad distribution of molecular weights within the polymer. One polymer was fractionated on Sephadex and the chemical composition of the fractions compared. The carboxyl content increased regularly with molecular weight, suggesting that the carboxyl groups were the result of secondary oxidative attack. Hydrogen and oxygen contents decreased with increasing molecular weight. Elemental analysis and infrared spectral data on the fractions were consistent with the hypothesis that gluconolactone was incorporated into the polymer in the early stages of formation. Chromophoric groups, perhaps of the endiol type, were formed, and oxygen-rich fragments were apparently lost as the molecular weight of the polymer increased. Oxygen contents, on the other hand, decreased with increasing molecular weight. The regularity observed in the fraction analyses suggested that the polymerization followed a pattern, and was not the result of the random joining of unlike fragments. Elemental analysis and infrared spectral data on the fractions were consistent with the hypothesis that gluconolactone was incorporated into the polymer in the early stages of formation. Chromophoric groups, perhaps of the enediol type, were formed, and oxygen-rich fragments were apparently lost as the molecular weight of the polymer increased.

Steric and electronic effects of the β-methyl group on monomer reactivity have been studied by copolymerization of styrenes and β-methylstyrenes. Steric hindrance of the β-methyl group in the transition state of homopolymerization is so great that the rate of the propagation reaction is considered to be depressed by as much as 1/10−1/20 of the corresponding styrene derivatives. In contrast to this, steric hindrance of the β-methyl group in cross propagation reactions with styrene is found to be very small. The monomer reactivity itself of β-methylstyrenes is a little lower than that of corresponding styrenes. This lowered reactivity of β-methylstyrenes is considered to be due to the electronic effect of the β-methyl group.

A turbidimetric titration method for determining solubility distributions (which are closely related to molecular weight distributions) has been developed. In this method, polymer is precipitated from its solution by addition of a nonsolvent of the same refractive index as that of the solvent. The precipitating particles of swollen polymer are allowed to agglomerate until the turbidity reaches a maximum. It is shown by the use of light-scattering theory that under these conditions the concentration of polymer which is precipitated can be calculated from the maximum turbidity on an absolute basis. The effects of particle size and particle size distribution are discussed. The method was tested by using polystyrene, and is shown to be accurate. In principle the method is applicable to most kinds of polymers.

Bisphenol A polycarbonate (Lexan, General Electric) was fractionated from methylene chloride by progressive precipitation with methanol and the fractions were characterized by osmometry, light scattering, and viscometry. Intrinsic viscosities of eight fractions in the molecular weight range 10,000–70,000 were obtained in five solvents at different temperatures to study the molecular expansion of polycarbonate. Light-scattering molecular weights, measured for the three high viscosity fractions, agreed within ±5% with the Mv values. The Mv/Mn ratio of the fractions was between 1.1 and 1.3, as compared to 2.7 for the unfractionated polymer. This indicates a relatively narrow distribution of the fractions. The exponent, a, in the Mark-Houwink equation showed no appreciable change with temperature. Values of d[η]/dT were generally negative. The unperturbed dimensions of polycarbonate, evaluated by the relation [η] = KM½-α2.43, suggest a stiff molecule. Values of K and d log α3/dT indicate increased chain flexibility at higher temperatures. Experimental data demonstrate that the Kurata-Stockmayer-Roig expression, based on the equivalent ellipsoidal model of a linear polymer molecule, is well suited to interpret the molecular expansion of polycarbonate. The Flory relation, α5 − α3 versus M½, is found to be of limited applicability.

Poly-α-deuteroacrylonitrile and poly-α,β,β-trideuteroacrylonitrile were synthesized. α-Deuteroacrylonitrile was prepared by the deuterium exchange reaction of acrylonitrile with D2O in the presence of CaO. α,β,β-Trideuteroacrylonitrile was prepared by the addition reaction of C2D2 and DCN. The infrared spectra of polyacrylonitrile, the deuterated polyacrylonitriles, and the intermediate products of the syntheses were measured in the 4000–400 cm.-1 region and discussed. The polarized spectra of the oriented polymer samples were also measured. The bands of PAN in the CH stretching region exhibit reasonable shifts in the spectra of the deuterated polymers, while the C≡N stretching band shows no appreciable shift. The bands in the region lower than 1400 cm.-1 exhibit complicated behavior in deuteration. The calculation of the normal vibrations of polyacrylonitrile has been made, assuming a syndiotactic planar zigzag chain as a most simple model for calculation. The assignments of the bands are discussed.

The polymerization of methyl methacrylate initiated by 2,2′-azoisobutyronitrile in the presence of halogenated benzenes and naphthalene has been investigated. In each case the order of reaction with respect to monomer is apparently of an order less than unity. Suggestions as to the possible participation of the solvent in the initiation process are discussed and shown to be in agreement with experimental measurements.

Copolymers of p-cyanostyrene and styrene were prepared by heating p-iodostyrenestyrene copolymers with cuprous cyanide in N-methylpyrrolidinone. The p-cyanostyrene copolymers could not be hydrolyzed to the benzoic acid derivative but were converted to the methyl p-vinylbenzoate-styrene copolymer by reaction with methanol and hydrogen chloride. This methyl ester copolymer was partially hydrolyzed to the free acid. A p-cyanostyrene(11%)-styrene copolymer was treated with n-butyl and methyl Grignard reagents to yield the corresponding n-butyl- and methyl-p-styryl-ketone copolymers. In the reaction with n-butylmagnesium chloride, the intermediate n-butyl-p-styrylketimine copolymer was isolated.

The chemistry of three-component coordination catalysts for the polymerization of α-olefins is very complex. The present investigation was initiated to elucidate the chemistry involved in the interaction of third-component compounds such as hexamethylphosphoric triamide, tributylamine, triphenylphosphine, or dimethylformamide with ethylaluminum dichloride. The present chemical and spectral data demonstrate that ethylaluminum dichloride and third-component compounds form stable complexes even at temperatures up to 100°C. in the absence of solvents or in homogeneous solutions. These data are in agreement with the fact that organoaluminum compounds form stable complexes with a variety of electron-donor compounds. The ethylaluminum dichloride third-component complexes in combination with titanium(III) chloride produce highly stereospecific catalysts for the polymerization of α-olefins.

Acrylamide polymerizes at 25°C. in aqueous solution by photosensitization with eosin, in absence of any reducing agent and of oxygen; a strong photoreduction of the dyestuff proceeds simultaneously. The photopolymerization has been followed gravimetrically, while the rate of photoreduction was followed colorimetrically. The rate of photoreduction is proportional to the absorbed light intensity. The rate constants are proportional to the monomer concentration when this concentration is lower than 0.42 mole/l.; above this value they are independent from it. The rate of photopolymerization is proportional to the square root of the eosin concentration when this concentration is ≤ 10−6 mole/l.; above this value, the rate becomes independent of the eosin concentration. With respect to the monomer the order of reaction varies from 1.5 to 1.85 when the eosin concentration is 0.4 × 10−6 and 6 × 10−6 moles/l., respectively. With respect to the light intensity, the order of reaction is 0.35 when the monomer and eosin concentrations are respectively, 2.8 moles/l. and 0.4 × 10−6 moles/l. The kinetics of photoreduction are interpreted on the basis of a reaction between the excited triplet state of the dye and the monomer, and a kinetic scheme is presented. The kinetics of photopolymerization are explained by a termination process between a growing chain and a primary radical, besides the usual termination process between two growing chains.

The nitrosation of o,o-linked linear phenol-formaldehyde polymers was readily affected in aqueous N,N-dimethylformamide to give products containing p-benzoquinone monoxime repeating units. Elemental analyses and infrared spectra indicated the presence of nitrophenol moieties also. Factors influencing the nature of the nitrosation reaction were studied.

It is shown that when methacrylic acid is polymerized in an aqueous system, the stereochemical configuration of the polymer, deduced by observation of the high resolution NMR spectra of polymethyl methacrylate samples prepared by methylation, is markedly influenced by the pH of the system. Syndiotactic propagation is favored at higher pH, presumably because coulombic repulsion between negatively charged monomer and negatively charged growing polymer radicals increases the free energy of activation for isotactic monomer placement.

A concentric cylinder apparatus for the studies of flow dichroism in polymer solutions is described in detail. The outer and inner cylinders are both made of fused quartz which is transparent in visible and ultraviolet region down to 220 mμ. The velocity gradient is produced in solution in the gap between the cylinders by rotating the inner one. The light path is designed to be perpendicular to the axis of rotation of the quartz cylinder. A differential device makes it possible to measure the difference in absorption coefficients, e‖ and e⊥, which are the absorption coefficients measured by plane-polarized light having parallel and perpendicular electric vector with the stream line. The apparatus is found to be useful for studying the regularity of orientation of chromophores in both synthetic and biological polymer chains.

A number of transition metal ions and complexes of the type which effect the stereospecific polymerization of butadiene in aqueous emulsion catalyze the polymerization of other vinyl monomers as well. In contrast to experience with butadiene, the structure of all products was typical of free-radical propagation reactions. The catalysis was somewhat selective, a given transition metal compound being active for some monomers but not for others. A mechanism involving coordinate complex formation between catalyst and monomer, generation of a free-radical initiator by redox reaction within the complex, and noncoordinated free-radical propagation is postulated. Copolymerization studies support such a mechanism; copolymer compositions were always typical of free-radical catalysis, even when the catalyst was active in homopolymerization of only one member of the monomer pair. Butadiene polymerizations using many of the same catalysts clearly proceed by different, little understood mechanisms. Attempted copolymerizations with butadiene gave only polybutadiene; the comonomers had no effect on polymer microstructure but generally reduced polymerization rate.

Diphenyl-p-styrylphosphine (I), prepared by the reaction of diphenylchlorophosphine with p-styrylmagnesium chloride, was converted to the corresponding oxide (II) and sulfide (III) by treatment with tert-butyl hydroperoxide and sulfur, respectively. These substituted styrenes polymerize readily with thermal or free radical initiation to high molecular weight polymers. These polymers are somewhat more thermally stable than polystyrene. Thus their Ti's in air were in the range 320–365°C. and in nitrogen, 375–395°C. Their T10's were in the 390–415°C. range in air and 425–435°C. in nitrogen. Polystyrene in air has a Ti of 263°C. and a T10 of 340°C. The monomers copolymerize readily with styrene and methyl methacrylate, the low conversion polymers always containing more phosphorus than the monomer mixtures. Reactivity ratios were determined, from which Q and e values were calculated. These values for the following monomer pairs are (phosphorus monomer M2): styrene-I, r1 0.52, r2 1.43, Q2 1.34, e2 −0.30; styrene-II, r1 0.42, r2 1.40, Q2 1.34, e2 −0.07; styrene-III, r1 0.43, r2 1.49, Q2 1.37, e2 −0.13; MMA-I, r1 0.32, r2 0.91, Q2 1.58, e2 0.80; MMA-II, r1 0.38, r2 1.46, Q2 1.42, e2 −0.35; MMA-III, r1 0.29, r2 1.22, Q2 1.58, e2 −0.62. Analysis of these data indicates that these phosphorus moieties are activating through strong electronegative effects. The extent of activation is intermediate in intensity between mildly activating groups (e.g. Cl) and strongly activating groups (e.g. CN or NO2). The behavior of diphenyl-p-styrylphosphine is of particular interest in this respect, since its electronegativity points to strong electron delocalization.

The variation of counterion mobilities in aqueous solutions of linear sodium polystyrenesulfonic acid over wide range of concentrations (10−2–0.7N) was studied. The viscosities of these solutions were measured in order to calculate the reduced viscosities at different concentrations. From the data obtained it was possible to conclude that in the range of concentrations 0.08–0.45N, an increase in D was observed. The enhancement of mobilities was attributed to an increase in the degree of apparent dissociation due to purely electrostatic interactions among neighboring chains. As concentration increased and chains approached each other, the above interactions should be accompained by short-range interactions which would produce entanglements among neighbouring chains. This may be responsible for the appearance of a maximum in mobilities as the concentration increased. Evidence was found for a decrease of DNa in solution when N > 0.45. The appearance of entanglements must produce an increase in the reduced viscosity which was observed. Tracer diffusion coefficients and reduced viscosities of the counterions Cs+, Ag+, and Ca++ in the corresponding salts of polystyrenesulfonic acid all at the same normalities were also measured. It was observed that the counterions Cs+ had a slightly larger degree of association than Na+. The behavior of Ca++ could be explained by the very strong association due to its double charge and for silver as partially due to the presence of complexes. For both counterions the reduced viscosity was much smaller than for the sodium salt.

The kinetics of the γ-ray-initiated, gel-forming copolymerization of styrene with the unsaturated ester containing 1.5 double bonds per molecule of number-average molecular weight of 890 (derived from maleic anhydride, phthalic anhydride, and propylene glycol) have been studied with initial ester concentrations ranging from 65% to 2.4%. From experimentally determined gel fractions, rates of disappearance of styrene and ester unsaturation (measured up to complete conversion of ester unsaturation), and the degree of polymerization of the growing chains, a model is suggested based on a free radical copolymerization, assuming a single effective propagation and termination rate constant (kp and kt, respectively), and bimolecular chain termination except between a pair of immobile gel radicals. Three ester concentration ranges were considered. At high ester concentrations (≥ 50%), styrene and ester enter the gel at a constant rate (molar ratio of styrene:ester double bond of 2:1), precluding application of conventional copolymer composition theory. The rate at which overall unsaturation, α, disappears, is given by dα/dt = (kp2/kt)M2° where M2° is the initial molar concentration of ester unsaturation. The rate at which styrene, α1, disappears is given by dα1/dt = 2/3(kp2/kt)M2°(M1° + M2°)/M1°. The rate at which styrene is consumed in the dilute esterstyrene solutions (⩽4.6%), where M2° is negligibly small, is given by dα1/dt = kp(k1/2kt)1/2. Polymerization rates of the intermediate ester concentration ranges (10% to 30%) are higher than predicted from the simple model. An empirical rate expression is presented which describes these systems to 40% conversion of styrene: dα1/dt = (kp2/kt)M2° {[(1 + m)(r1 + M)]/(r1 + 2M)}, where M is the ratio of ester unsaturation to styrene in the starting mixture and r1 is the reactivity ratio for styrene, defined in the usual manner. For initial conversions of ester unsaturation (to 10%) dα/dt = (kp2/kt)M2°.

Intrinsic viscosities have been obtained for polyisobutylene in the series of n-alkane solvents at −12, 24.3, and 57.9°C. Values of the χ parameter have been calculated using [η]/[η]ϑ = α2.43 and the Flory closed expression relating α to χ (as modified by Stockmayer): α5 — α3 = 2 × 0.49 CMM1/2 [(1/2) — χ]. The values found by the first method are in excellent agreement with an approximate expression for χ derived using the Prigogine theory of polymer solutions, viz. Rχ = ArA/T + BT/rA where rA is the number of segments in the solvent molecule. Values of the constants A and B compare well with those found from heats of mixing of polyisobutylene with the same solvents. Values of χ are also obtained using the more recently Kurata-Stockmayer-Roig expression for the α — χ dependence although reasons are advanced for preferring the Flory-Stockmayer equation. Qualitative, but less good quantitative agreement is now found between the χ values and the Prigogine theory.

Poly-(trans-CHDCHD) and poly-(cis-CHDCHD) have been obtained with [Al(Et)3 + TiCl4] and with [Al(i-Bu)3 + TiCl4]. Poly-(trans-CHDCHD) and poly-(cis-CHDCHD) obtained with [Al(Et)3 + TiCl4] give mutually different infrared spectra. The characteristic absorption bands of the former are found at 1335, 1288, and 594, and 586 cm.−1, and those of the latter at 1306, 597, and 590 cm.−1. In order to interpret these spectra, the normal vibration frequencies of the model structures, namely, erythro-diisotactic and threo-diisotactic structures, have been calculated. From the comparison of the observed and calculated frequencies, it has been concluded that poly-(trans-CHDCHD) obtained with [Al(Et)3 + TiCl4] consists mainly of the threo-diisotactic and disyndiotactic portions and poly-(cis-CHDCHD) obtained with the same catalyst consists mainly of the erythro-diisotactic and disyndiotactic portions. The relation between the type of double bond opening and the structure of the resultant polymer has been discussed, and it has been concluded that the cis-opening of the double bond occurs in the polymerization reaction with [Al(Et)3 + TiCl4]. The spectra of poly-(trans-CHDCHD) and poly-(cis-CHDCHD) obtained with [Al(i-Bu)3 + TiCl4] are not so different from each other, but they are different from the spectra of the polymers obtained with [Al(Et)3 + TiCl4]. The polymers obtained with [Al(i-Bu)3 + TiCl4] show characteristic bands at 2919, 2850, 2176, 2095, 1453, 660, 623, 600, 593, 581, 566, and 545 cm.−1. These bands give evidence of the presence of isolated CH2 and CD2 groups. This fact indicates that the hydrogen–deuterium exchange reaction in monomers as well as the polymerization is promoted by [Al(i-Bu)3 + TiCl4].