Abstract
It has been well established that gas—liquid chromatography (glc) can give accurate thermodynamic data on binary solutions where the components differ considerably in volatility or molecular weight. The substance of lower molecular weight (component 1) is injected into the moving gas phase and dissolves at effectively infinite dilution in the stationary liquid phase. This is formed by the higher molecular weight material, for example, squalane, biphenyl, dinonyl phthalate, glycerol, or the higher n-alkanes such as C16, C24, C36, etc. The convenience of the technique is such that activity coefficient data have already been obtained for hundreds of systems. In contrast, activity data are available for far fewer high polymer systems, in part certainly because of the need to use the laborious vapor sorption technique. While that technique gives activity data as a function of concentration, it would still be desirable to have data at infinite dilution for a variety of systems in order to test contemporary theories of polymer solution thermodynamics. Recently Guillet and coworkers have applied the glc technique to systems in which the stationary phase is a high polymer. (J. E. Guillet and coworkers refer to the gas-phase component as the molecular “probe”. This avoids the glc teminology in which that component is the solute and the stationary-phase polymer would be the solvent. This terminology is confusing to polymer chemists used to solutions where the polymer is the solute, being present at low, rather than high, concentrations.) Their primary interest has been to demonstrate the versatility of the technique in determining first- and second-order phase transitions, degrees of crystallinity, and other physical characteristics of the polymer, while the present communication considers the determination of thermodynamic quantities. It has been prompted by comments from several workers who have noted the difficulty of applying the usual thermodynamic equations of glc which yield γ1∞, the activity coefficient of component 1 at infinite dilution [Equations (5) and (6)]. The equations require an exact value of the molecular weight of component 2, making difficult their use for polymer systems. Our main objective is to resolve this problem. However, we also wish to stress the utility of the technique in providing data with which to test contemporary theories of polymer solution thermodynamics. We therefore comment on equations which directly relate experimental glc data to the interaction parameter, χ, of polymer solution thermodynamics.