Wiley

Values for A′ and C_tr,m for styrene polymerization between 0 and 90° are presented. These are important for accurate calculation of rates of initiation.

Benzoyl, anisoyl, veratroyl, p‐bromobenzoyl, p‐phenylbenzoyl, p‐benzoylbenzoyl, p‐methylsulfonylbenzoyl, p(N,N‐dimethylsulfonamido)‐benzoyl and ‐naphthoyl disulfides are promoters of the emulsion copolymerization of butadiene and styrene, but exhibit little effect on the polymer properties. Furoyl, phenyl, p‐chlorophenyl and p‐bromophenyl disulfides are neither promoters nor modifiers.

The infrared spectra of oriented films of valonia cellulose and of ramie and bacterial cellulose crystallites have been observed in the 3 μ region. Polarization properties of the bands have also been determined. The differences between the polarized spectra of bacterial and ramie crystallites in this region are attributed to per cent crystallinity and orientation effects. Two new bands in the CH stretching region have been observed. With this new information the CH₂ symmetric and antlisymmetric stretching modes are assigned to parallel and perpendicular bands, respectively, requiring a specific orientation of the CH₂OH group. From the observed polarization of the bands in the OH stretching region, a system of hydrogen bonding in the crystal structure of cellulose I is proposed. This involves a change in conformation of the cellobiose unit to permit an intramolecular hydrogen bond between the C₃ hydroxyl and the ring oxygen of contiguous glucose units. Two sets of intermolecular hydrogen bonds are proposed: in the 101 plane the C₆ hydroxyls of the antiparallel chains are joined to the bridge oxygens of the adjacent parallel chains; in the 101 plane the C₆ hydroxyls of the parallel chains are hydrogen‐bonded to the bridge oxygens of the adjacent antiparallel chains.

The electrolytic dissociation of several polymeric acids and bases in aqueous solutions has been investigated. The potentiometric behavior is well described by the following equations, For polyacids: and for polybases: pK₀ is the intrinsic dissociation constant of the monomeric unit, α and β are the degrees of ionization of the polyacid and the polybase respectively, ψ₀ is the electrostatic surface potential of the polyion. It is shown that ψ₀ is equal to (1/ϵ) (δF_e/δv) for polyacids and (1/ϵ) (δF_e/δζ) for polybases where F_e is the electrostatic energy of the polyion and v or ζ the number of negative or positive ionized groups respectively. Equations for the calcualtion of ψ₀ in the cases of randamly kinked and stretched polyelectrolytes are given and discussed. As the potential ψ₀ can be obtained independently from electrophoretic measurements, the above equations correlate potentiometry and electrophoresis. The potentials of polymethacrylate ions were calculated from theory and obtained from electrophoretic and potentiometric measurements. The potentials obtained by the three methods agree within 3% thus justifying the use of the combined potentiometric and electrophoretic measurements for the evaluation of pK₀. Application of this method to the potentiometric analysis of pectic acids gave good agreement with experiment and lead to pK₀ = 3.40. Combined potentiometric‐electrophoretic analysis for polyaspartic acid gave pK₀ = 3.53 and for polylsine pK₀ = 10.44.

A procedure is described for the measurement of number‐average molecular weight M̄_n on a specimen in bulk form, utilizing a dynamic torsion pendulum. For both atactic and isotactic polystyrenes it has been found that the minimum value of the damping logarithmic decrement, δ_m, (above the glass temperature) is related to an measured osmometrically in dilute solution by M̄_n = δ × 10⁵. The M̄_v/M̄_n ratio can also be estimated directly from the δ vs. temperature curve. A four‐element spring‐and‐dashpot model yields similar 6 vs. T curves when the Williams‐Landel‐Ferry temperataure dependence is assigned to the viscosity of the dashpots.

The Gordon‐Taylor equation relating the glass transition temperature θ of a copolymer to the glass transition temperatures θ₁ and θ₂ of the homopolymers is equivalent to where c₁ and c₂ are the weight fractions of the constituents and A₁ and A₂ are constants. It can be recast into the following forms suitable for linear plots and where k = A₂/A₁. Data from the literature on 10 copolymer systems, including butaciene‐styrene copolymers, give linear plots, verifying the equation within experimental error. However, the observed value of k is in most cases significantly smaller than the ratio of the differences of the volume‐temperature coefficients for each homopolymer in the rubbery and glassy states, as required by the derivation of Gordon and Taylor. The glass transition temperature (in °C.) for a butadiene‐styrene copolymer prepared by emulsion polymerization at 50°C. may be calculated from the weight fraction c₂ of bound styrene as and for a similar 5° copolymer as

Low pressure polyethylene fractions of molecular weight above 100,000 have been studied by the light scattering method. The presence of long chain branches in fractions of molecular weight above 300,000 was indicated by the data. The molecular weight‐intrinsic viscosity relationship for linear polyethylene chains was found to be [η] = 4.60 × 10⁻⁴ M^0.725 when tetralin at 130°C. was used as the solvent for the intrinsic viscosity measurements. This relationship gives molecular weights about 1.15 times higher than those found previously by the number‐average molecular weight determination of the fractions.

Improved methods for measuring weight‐ and number‐average molecular weight of polymers soluble only at elevated temperatures are applied to selcted samples of polyethylene. For two samples of essentially linear polyethylene, number‐average molecular weights by cryoscopy are 11,500 and 13,000, and weight‐average molecular weights by light scattering are 144,000 and 175,000. For samples of branched Fawcett‐type polyethylene, number‐average molecular weights are in the range 10,000–19,000 and weight‐average molecular weights are between 300,000 and 700,000. For one sample each of branched and essentially linear polyethylene it is demonstrated that light scattering measurement in three solvents gives the same molecular weight: no evidence is found for the “association” of these samples of polyethylene in thermodynamically poor solvents. It is thus demonstrated that association is not observed for all samples of polyethylene in dilute solution in α‐chloronaphthalene.

A procedure and apparatus are described for the large scale (50‐gram) fractionation of high‐density polyethylene by Desreaux's fractional extraction technique. Two samples of Ziegler polyethylene were fractionated by this technique and the intrinsic viscosity in decalin at 135°C. and molecular weight by light scattering were measured on the fractions. The relationship [η] = 4.6 × 10⁻⁴ M̄_w^0.73 was found to hold for molecular weights between 25,000 and 640,000. The samples differed in polydispersity, one having a broad distribution and the other a rather narrow distribution. Evidence for the sharpness of the fractions obtained by the new technique was provided by refractionating of one fraction and the resolation of two fractions which were combined and refractionated.

Relationships, as reported in the literature, between intrinsic viscosity and weight‐average molecular weight for high‐density polyethylene are in poor agreement. Experimental work in this laboratory, aimed at resolving similarly conflicting data, suggests that likely causes for these discrepancies are: polydispersity, incorrect dissymmetry correction, and inadequate solution clarification. These factors are discussed in some detail, and results on both fractionated and unfractionated samples support these conclusions. It is apparent that fairly narrow fractions must be studied to obtain a valid intrinsic viscosity‐weight‐average molecular weight relationship.