Cloud-point curves by DSC measurements at different scanning rates on solutions of two samples of poly(vinyl methyl ether) in water show two minima in a temperature−composition plot. The significance of the data was confirmed by static measurements of the coexisting-phase compositions and phase-volume ratios. A thermodynamic analysis in terms of a strongly concentration-dependent interaction parameter leads to the conclusion that the system H2O/PVME exhibits so-called type III behavior. For polymers of infinite molar mass, such behavior is characterized by the occurrence of two off-zero critical concentrations, in addition to the usual zero critical concentration marking the ϑ state.

Liquid-Liquid Phase Separation in Multicomponent Systems 637 improvements in selectivity. Such increases in selectivity must reflect a consolidation of the membrane skin region.Coagulation behavior for the aromatic polyamides (polymers III and II) is similar. Rate of coagulation, as evidenced by the development of an opaqueness, occurs more rapidly for the aromatic polyamides than for the polyamide-hydrazide. ConclusionsTypical flux-rejection data for the membrane systems described, with various feed solutions, are summarized in Table V. These aromatic polyamide membranes are characterized by their good mechanical strength, as illustrated in Figure 7.Here a comparison of flux-pressure response between polymer I and cellulose acetate is made. A straight line plot for flux is observed up to pressures of 1800 psi, with some departure from linearity evident after 1 hr. The deflection in the flux line for cellulose acetate is seen to coincide with a decrease in rejection, indicating development of a membrane defect.Typical mechanical properties for asymmetric membranes prepared from these polymers are shown in Table VI. In addition to their good mechanical properties, these aromatic polyamides show remarkable chemical resistance, especially in the high-pH range. This is in marked contrast to the cellulose acetate systems.1 23456789 Membrane reverse osmosis performance for aqueous salt rejections appears unaltered after 1 week of storage in 1 A solutions of H2S04, HC1, HNOs, CHsCOOH, NaOH, andNH4OH.Selectivity of these membranes to solutes other than NaCl is quite good, as illustrated by results for polymer III in Table VII.

Equilibrium swelling of networks capable of separating into a highly swollen and a collapsed phase is discussed on the basis of a concentration-dependent pair interaction parameter. Classic rules, to be obeyed by the various binodals and their metastable and unstable extensions, rationalize the experimentally accessible portions of the phase diagram and are covered by a simple model yielding acceptable parameter values when fitted to swelling data on the system water/polyW-isopropylacrylamide). The description of the solvent/network binodal is good, but the predicted location of the LCST miscibility gap for linear chains deviates considerably from the measured cloud points. It is probable that linear and branched chains do not differ much in enthalpic contributions to the interaction parameter but deviate appreciably in the entropic terms. This feature offers an acceptable explanation for the discrepancy.

It is demonstrated that the collapse of gels of radiation cross-linked poly(vinyl methyl ether) swollen in water is a truly discontinuous process obeying classic thermodynamic principles. The phenomenon can further be shown to be related to the relevant type of limiting critical demixing of solutions of the corresponding non-cross-linked polymer. There are three such types, including “classic” ϑ behavior, and they can also be recognized in the swelling behavior of the cross-linked analogue.

Theoretical and experimental evidence is put forward to prove that the determination of the phase‐volume ratio as a function of temperature and concentration is a sensitive and simple means of determining the liquid–liquid critical state. Knowledge of the critical conditions permits very accurate calculations of the interaction parameter g in the free‐energy function. In experiments with polystyrene–cyclohexane, g was found to depend on the concentration. The value of g and its concentration dependence agree very well with the results of osmotic measurements by Rehage and Palmen. In experiments with polyethylene–diphenyl ether, g proved to be independent of concentration in the range of measurement. The temperature function was found to be: g = −0.6086 + 482.2/T (at 137–148°C.). Gibbs' expressions for the critical conditions were worked out for a free‐energy relation in the form of an extended Flory‐Huggins function.