Showing papers in "Macromolecules in 1982"

Effects of annealing and prior history on enthalpy relaxation in glassy polymers. 2. Mathematical modeling

Abstract: A simple four-parameter model reproduces DSC data on the effects of annealing conditions, and thermal history before annealing, on the heat capacity of glassy polymers. The model is an application of the successful treatment of glass transition kinetics, due to Moynihan and co-workers, to thermal histories which include annealing. It is found that a nonexponential relaxation function is essential for the development of sub-T, heat capacity peaks with annealing and that nonlinearity is important in accelerating their development to experimentally accessible aging times. Numerical integration allows accurate predictions of the effects of quench rate, annealing temperature, annealing time, and reheating rate on the magnitude and temperature of heat capacity peaks observed in DSC scans of annealed PVC (part 1) and other polymeric glasses.

Calculation of liquid-liquid phase separation in a ternary system of a polymer in a mixture of a solvent and a nonsolvent

Abstract: A numerical method for the calculation of the binodal of liquid-liquid phase separation in a ternary system is described. The Flory-Huggins theory for three-component systems is used. Binodals are calculated for polymer/solvent/nonsolvent systems which are used in the preparation of asymmetric ultrafiltration or reverse osmosis membranes: cellulose acetate/solvent/water and polysulfone/solvent/water. The values for the binary interaction parameters are taken from literature sources. The effect of a concentration-dependent solvent/nonsolvent interaction parameter is discussed. Although knowledge of the interaction parameters for all compositions in the ternary system is rather poor, fairly good agreement has been found between calculated and experimentally found miscibility gaps when the solvent/nonsolvent parameter is taken to be concentration dependent and the other parameters, the polymer/solvent and the polymer/nonsolvent interaction parameter, are kept constant.

Effects of annealing and prior history on enthalpy relaxation in glassy polymers. 1. Experimental study on poly(vinyl chloride)

Abstract: Endothermal peaks below T. have been observed by differential scanning calorimetry (DSC) in samples of poly(vinyl chloride) (PVC) which were given a variety of thermal, mechanical, and vapor treatments and subsequently aged for varied times (te) at several temperatures (Te). The peaks increase in magnitude and shift to higher temperature with increased te and T. and appear as the well-known \"T5 overshoot\" as T. approaches T.. The magnitude, but not the position, of the peaks is strongly dependent upon the treatment given the sample before aging. The peaks are enhanced by an increased rate of thermal quenching, by tensile or compressive mechanical stressing, and by prior exposure to a swelling vapor (e.g., methyl chloride). Development of the peaks upon aging is suppressed by the presence of a soluble vapor (CH 3C1 or CO 2) during aging. The DSC observations may be qualitatively explained as a consequence of the slow approach toward equilibrium of a system with a broad distribution of relaxation times. Introduction Differential scanning calorimetry (DSC) has revealed enthalpy relaxations occurring near or below Tg in glassy polymers aged after a wide variety of thermal, mechanical, and solvent or vapor treatments. Illersi and Gray and Gilbert2 observed subT. endothermic peaks in annealed PVC samples which had been rapidly quenched and found that the magnitude and temperature of the peaks increased with increasing annealing time and/or temperature between the quench and the DSC scan. Foltz and McKinney3 and Petrie4 demonstrated that the magnitude of the DSC endothermic peak at Tg (\"T, overshoot\") provided a quantitative measure of the enthalpy relaxation which had occurred during prior annealing at temperatures somewhat below Tg. In a recent study of aging in ABS and SAN, Wysgoski5 found no endothermic peaks in samples scanned immediately after a rapid quench, whereas aging at room temperature led to a slow development of small, broad endotherms well below Tg; with increasing aging temperature, the endotherms became more pronounced and occurred at higher temperatures, appearing as the familiar T. overshoot for aging temperatures within about 20°C of Tg. DSC endotherms showing similar dependences upon annealing time and temperature have been observed by Weitz and Wunderlich 6 and others 7-3 in polymers vitrified by cooling through Tg under high hydrostatic pressure. Application of mechanical stress to polymers in the glassy state, through either compression\") or tensile drawing,\" also results in endothermic maxima below or near T g . Matsuokal2 has suggested that dilation under tensile stress increases the enthalpy of the glass, reducing the relaxation time and therefore increasing the rate of enthalpy relaxation during annealing. Compared with these studies of enthalpy relaxations induced by thermal and mechanical treatments, reports dealing with vapor or solvent effects on glassy state relaxations are less extensive or explicit but nonetheless suggestive of closely related effects. Chan and Pau1 13 found that exposure of annealed polycarbonate to high CO 2 pressure reduced the magnitude of the DSC endotherm and suggested that sorption of CO 2 dilated the polymer, increasing its enthalpy relative to the annealed state. Shultz and Young,\" in studying polystyrene and PMMA samples freeze-dried from naphthalene solutions, observed sub-Tg endothermal maxima which intensified and moved to higher temperature with increasing annealing time or temperature. Their observations were attributed to a broad distribution of relaxation times in the very \"open\" glass structure produced by the rapid quench and matrix sublimation. Berens and Hopfenberg 15,16 found that prior exposure to a swelling vapor enhances the sorptive capacity of glassy PVC and polystyrene and suggested this effect may be due to an increase of frozen-in free volume; although DSC data were not included in their study, it seemed likely, in view of the other calorimetric findings, that the vapor-swelling process might also produce an enhancement of enthalpy relaxation effects. Vol. 15, No. 3, May-June 1982 Enthalpy Relaxation in Glassy Polymers. 1 757 The present study was undertaken to confirm the anticipated effects of vapor treatments upon enthalpy relaxations and to directly compare these effects with those of varied thermal and mechanical treatments for a given polymer and similar annealing conditions. The results seem to reveal a common explanation for the effects of the widely diverse treatments. The enthalpy relaxations in all cases closely follow the predictions of a nonlinear, nonexponential relaxation mode120'21 as detailed in part 2. 22 Experimental Section Materials. The polymer used in all experiments reported here was a commercial, suspension-polymerized PVC (Geon 103EP, BFGoodrich Chemical Group), obtained as a free-flowing white powder. Optical microscopy and electron microscopy have shown this resin to consist of roughly 2-gm primary particles firmly agglomerated into porous 100-150-gm grains. Molecular weights given by the supplier are Ain = 65000 and ./t4, = 205000. Portions of this polymer were given the following thermal, vapor, and mechanical treatments: \"Free-Cooled\" Samples. To erase effects of prior history, PVC powder samples were treated 30 min in a circulating-air oven at 120 °C and then allowed to cool to room temperature in the open air. Approximate cooling rates were 100 °C/min from 120 to 100 °C, 60 °C/min from 100 to 80 °C, 40 °C/ min from 80 to 60 °C, and 25 °C/min from 60 to 40 °C. \"DSC-Cooled\" Samples. To provide samples of more closely controlled cooling rate, 15-mg samples in DSC pans were heated 30 mm at 120 °C in the DSC instrument, cooled at the maximum nominal cooling rate of 320 °C/min to -150 °C, and allowed to return to room temperature. \"Liquid-N2-Quenched\" Samples. Portions of the PVC powder (-10 g) were heated 30 mm at 120 °C and then poured directly into a large excess of liquid nitrogen in a Dewar vessel. Boiling of the liquid N2 ceased within 2 s, suggesting an average cooling rate of roughly 104 K/min; assuming exponential cooling, the rate through the glass transition region may have approached 105 K/min. Methyl Chloride Vapor Treatments. For study of the effect of vapor swelling on enthalpy relaxations, CH 3C1 vapor was chosen for its combination of relatively high diffusivity and appreciable solubility in PVC at convenient pressures. Gravimetric sorption measurements by methods previously described 17 showed the diffusivity of CH3C1 in PVC at 30 °C to be 1 X 10 -11 cm2/s and its equilibrium solubility at 13-kPa (0.13 atm) pressure to be 3.6 mg/g of PVC. For the DSC study, samples of the PVC powder freshly free-cooled from 120 °C were evacuated in a stainless steel pressure vessel and then exposed to CH 3C1 vapor at selected pressures (200-500 kPa) at 30 °C for up to 17 h. The vessel was then rapidly vented to atmospheric pressure and evacuated for 1 h at 30°C; the absorbed CH3C1 was essentially all removed, since the half-desorption time under these conditions is approximately 30 s. Hot-Pressed Sheet. Fused PVC sheets approximately 0.3 mm thick X 120 mm diameter were prepared by pressing portions of the powder sandwiched in aluminum foil between flat steel plates at 210 °C for 10 s under 133-kN (30000 lb) force. The sheets were cooled by placing the hot steel plates and PVC sheet in a cold press without applying pressure; the cooling rates measured by a thermocouple were approximately 40 °C/min from 210 to 150 °C, 15 °C/min from 150 to 100 °C, and 7 °C /min from 100 to 50 °C. Samples for DSC measurements were punched from the center of the pressed sheets, where flow orientation was minimum. Cold-Drawn Sheets. Strips about 2 x 10 cm were cut from hot-pressed sheets and drawn at 1.3 cm/min in an Instron tensile tester, forming a stable neck with a draw ratio of about 2:1. DSC samples were punched from this drawn region. Cold-Pressed Powder. Small portions of the PVC powder freshly free-cooled from 120 °C were compressed in a 1.1-cmdiameter piston mold under 517-MPa (75 000 psi) pressure for 10 mm at room temperature, forming opaque, brittle pellets. Fragments of these pellets were used in the DSC measurements. Cold -Re-pressed Sheets. Portions of the hot-pressed sheet described above were cut to ca. 1.1-cm diameter and re-pressed in the piston mold under 517-MPa pressure for 10 min at room T,°C Figure 1. DSC scans on PVC cooled through Tg at 320 °C/min and aged at 20 °C for the indicated times: (—) first scan; (. • .)

Crystalline forms in a copolymer of vinylidene fluoride and trifluoroethylene (52/48 mol %)

Abstract: : The structure of 52/48 mol % copolymer of vinylidene fluoride and trifluoroethylene has been investigated at various temperatures by X-ray diffraction. Melt-solidified samples consist of a mixture of two disordered crystalline phases, one trans-planar, the other 3/1-helical. Samples may be transformed to either phase by appropriate means to reveal a hexagonal (or pseudo-hexagonal) molecular packing. The all-trans phase may be obtained by drawing or poling at low temperatures; both treatments cause a transformation of the disordered mixture of phases into a well-ordered planar zig-zag phase. Isolation of the disordered 3/1-helical phase is achieved by heating to high temperatures, where-upon all samples, irrespective of orientation or polarization, undergo transformation to a poorly ordered helical structure analogous to that of trifluoroethylene homopolymer; upon cooling, the original, disordered mixture of phases is recovered. (Author)

Structural and Dielectric Investigation on the Nature of the Transition in a Copolymer of Vinylidene Fluoride and Trifluoroethylene (52/48 Mol %).

Abstract: : The effect of temperature on the structure and dielectric properties of a 52/48 mol % copolymer of vinylidene fluoride and trifluoroethylene has been investigated at temperatures up to 140 C. Undrawn or unpoled specimens contain an intimate mixture of two disordered crystalline phases, both of which undergo a large increase in d-spacing at 65-70 C, with eventual transformation to a single phase in which the chains assume a disordered 3/1-helical conformation above 90 C. The 70 C transition is accompanied by a dielectric anomaly. High electric fields applied at temperatures below 70 C induce a phase change to a single, well-ordered all-trans conformation, leading to remanent polarization with piezoelectric and pyroelectric coefficients comparable to those of poly(vinylidene fluoride). The changes in crystal phase and dipole orientation upon poling result in a reduction of the dielectric constant at room temperature, a shift of the dielectric anomaly to about 80 C, stability of the all-trans crystal phase to somewhat higher temperatures, and a discrete change in d-spacing to that of the disordered 3/1-helical conformation at the transition region. The loss of polarization in poled specimens at this ferroelectric-to-paraelectric transition is attributable primarily to the molecular change from the polar all-trans conformation to its non-polar, disordered 3/1-helical counterpart, as well as to the onset of rotational dipolar motions leading to the dielectric anomaly. (Author)

Analysis of the mechanism of copolymerization of styrene and acrylonitrile

Abstract: The copolymerization of styrene and acrylonitrile in bulk at 60 °C has been investigated by measurement of the copolymer/comonomer composition relationship and of the monomer sequence distributions using C NMR. Alternative models for the mechanism of the copolymerization have been evaluated in the most general forms, with allowance for nonzero conversions, by deriving reactivity ratios from composition data and then comparing predicted and experimental sequence distributions. The system shows significant deviations from the terminal model. Compositions cannot differentiate between penultimate and complex participation models, both of which give significant improvement over the terminal model. The measured sequence distributions are quite close to the predictions of the penultimate model but substantially different from those of the complex participation model, showing clearly that the penultimate model is the most appropriate of the models considered. We have obtained r = 0.22, r = 0.03, r = 0.63, and r = 0.09. No significant improvement was observed with the antepenultimate model.