Journal of Physical Chemistry B, Vol.106, No.2, 418-433, 2002
Thermodynamic expressions relating different types of preferential interaction coefficients in solutions containing two solute components
In aqueous solutions that contain a macromolecular solute (charged or uncharged) and molecules or ions of a smaller solute, differences between solute-macromolecule and water-macromolecule interactions have solute-concentration-dependent thermodynamic effects that are characterized quantitatively by "preferential interaction" coefficients. Of primary interest in this paper are Gamma(mu1) and Gamma(mu3), defined as partial derivatives that specify the dependence of the molality of the smaller solute on the macromolecular molality at fixed temperature and pressure and the chemical potential indicated by the subscript (mu(1) for water and mu(3) for the smaller solute). Coefficients of the type Gamma(mu3) (but not Gamma(mu1)) are direct gauges of thermodynamic effects due to the preferential interactions of a small solute with a macromolecule. Although individual values of Gamma(mu3) cannot generally be obtained directly by any experimental method, corresponding values of Gamma(mu1) are directly measurable for involatile solutes (0.01 less than or similar to m(3) less than or similar to 3 m) by an accurate, efficient method based on water vapor pressure osmometry [Courtenay et al. Biochemistry, 2000, 39, 4455]. In that study, alternative approximate expressions, for which we present generalized derivations here, were used to calculate The resulting values differ significantly from the corresponding values of Gamma(mu3) for the interactions of various small biochemical solutes with a common globular protein. To identify the general physical origins and thermodynamic implications of numerical differences between Gamma(mu3) and Gamma(mu1) we derive an exact thermodynamic relationship linking these coefficients and examine how it is affected, for various types of systems and ranges of conditions, by contributions from the ideal mixing entropy of components 2 and 3 and from the nonideality due (primarily) to interactions of the macromolecule with the smaller solute. This analysis shows why, in general, Gamma(mu3) differs significantly from Gamma(mu1) and why Gamma(mu3) congruent to Gamma(mu1) in the exceptional cases where components 3 and 2 have a common ion present in large excess over the macromolecular species.