Search

The SAFT1 equation of state captures the melting retrograde behavior for a ternary system of poly(ethylene-co-octene-1) + isooctane + ethylene. The melting temperature minimum depends on the polymer branch density, solvent composition, polymer concentration, and polymer molecular weight, in decreasing order of sensitivity. Furthermore, the effects of the solvent/antisolvent ratio and polymer concentration on the melting temperature are nonmonotonic, that is, their sign may change in going from low to high pressures. Finally, the sign of the slope of the melting curve depends only on the difference between the partial molar volume of polymer in solution and the molar volume of pure solid polymer.

Cloud and spinodal loci have been obtained as functions of pressure (P), temperature (T), polymer segment number (r), segment fraction (ψ), and H/D substitution on solute (zD) or solvent (yD) for polystyrene-h/propionitrile-h (PSh/PNh) solutions, polystyrene-h/propionitrile-2,2-d2 solutions (PSh/PNd), and polystyrene-d/propionitrile-h solutions (PSd/PNh). A light scattering technique was employed. An increase in pressure or an increase in the H/D ratio in the solvent (or decrease in the polymer H/D ratio) increases the region of miscibility. The solutions show upper and lower branches which may join at hypercritical points depending on the pressure, molecular weight, or solvent H/D ratio. A least-squares technique to fit experimental data to the continuous mean-field thermodynamic description of polymer−solvent demixing developed by Luszczyk, Rebelo, and Van Hook (Macromolecules 1995, 28, 745) is described and then illustrated by application to the PS/PN data.

Cloud and spinodal loci have been obtained as functions of pressure (P), temperature (T), polymer molecular weight (MW), polymer fraction, and H/D substitution on solute (z D ) or solvent (y D ), for acetone-polystyrene and methylcyclopentane-polystyrene solutions. A light scattering technique was employed. The isotope effects and their pressure dependences are large. An increase in pressure, an increase in H/D ratio in the solvent (or decrease in the polymer H/D ratio), or a decrease in polymer molecular weight increases the region of miscibility. The solutions show both upper and lower branches, which, for acetone, join at a hypercritical point. The cloud point diagrams are highly distorted in the hypercritical region but smooth as one moves away by appropriately varying pressure, H/D ratio, or solute molecular weight. In the vicinity of the multiple critical point, hypercritical distortion for polystyrene/ (CH 3 ) 2 CO solutions causes a highly unusual homogeneous one-phase «hole» in the (T, W ps ) cloud point diagram. This «extra» one-phase region appears close to the hypercritical point, and just inside the two-phase region which forms immediately after the system collapses into the «hourglass» configuration. The effect is not seen in (CD 3 ) 2 CO or mixed (CH 3 ) 2 CO/(CD 3 ) 2 CO solutions, but those solutions also show marked distortion in the (T, W ps ) plane. H/D substitutions (on either polymer or solvent) cause large differences in cloud point and spinodal temperatures, especially in the vicinity of the hypercritical point. Isotope shifts as large as 40 K were observed on the cloud points of the polystyrene/acetone system. A mean-field formalism to interpret cloud point and spinodal data on liquid-liquid demixing of polymer-solvent solutions exhibiting both upper and lower consolute branches and hypercritical points is developed. The dependence of the phase equilibria on pressure, temperature, segment number, polydispersity, and H/D substitution of either or both solvent and polymer are explicitly considered. Calculations of cloud and shadow curves are reported. The method is based on a Flory-Huggins type continuous thermodynamic representation of free energies in polydisperse polymer solutions due to Ratzsch and co-workers. The computer algorithm PHASEEQ r written in QUICKBASIC r has been coded for the case of Schulz-Flory segment distribution functions. Calculations interpreting demising data for polystyrene/acetone and polystyrene/methylcyclopentane solutions over wide ranges of pressure, concentration, molecular weight, and H/D substitution are reported. The isotope effects, which are large, are discusaed using Bigeleisen-Van Hook-Wolfsberg condensed phase isotope effect theory. Along the critical line multidimensional classical or nonclassical scaling equations permit economical representation of the data. The scaling approach is developed and compared with the mean-field based interpretation