Continuum estimates of rotational dielectric friction and polar solvation

Dynamical solvation data recently obtained with the probe solute coumarin 153 are used to test the reliability of dielectric continuum models for estimating dielectric friction effects. In particular, the predictions of the Nee-Zwanzig theory of rotational dielectric friction are examined in some detail. The analysis undertaken here uncovers an error made in virtually all previous applications of the Nee-Zwanzig formalism. The error involves neglect of the solvent's electronic polarizability when calculating dielectric friction constants. In highly polar solvents the effect of this neglect is shown to be minor, so that the results of past studies should not be appreciably altered. However, in weakly polar and especially in nondipolar solvents, the proper inclusion of electronic polarizability terms is essential. The equivalence between the Nee-Zwanzig theory of dielectric friction and more general continuum treatments of polar solvation dynamics is also demonstrated. This equivalence enables the use of solvation data to test the reliability of the Nee-Zwanzig description of electrical interactions between a solute and solvent that form the core of this and related continuum theories of dielectric friction. Comparisons to experimental data show that, with the important exception of nondipolar solvents, such continuum treatments provide reasonably accurate (+/-40%) predictors of time-dependent solvation and/or dielectric friction. (C) 1997 American Institute of Physics.