A Simplified Treatment of Solvent-Solute Dispersion^†

The polarizable continuum model (PCM), particularly its integral equation formalism variant (IEF-PCM), is one of the most widely used methods for computing electrostatic reaction fields, in which the standard implementation models the solvent as an isotropic dielectric medium. This approach is effective in treating electrostatic interactions, which often dominate for charged and polar solutes in high-dielectric-constant solvents; however, it becomes increasingly critical to accurately capture solvent-solute dispersion interactions in low-dielectric-constant solvents, where their contribution is significantly enhanced and can become the dominant component, particularly for nonpolar solutes. The solvent-solute dispersion energy is typically computed using the Caillet-Claverie atom-atom pair potential method, which conventionally employs a universal coefficient of cf⁽⁶⁾ = 0.2140 kcal mol^-1 for all atomic pairs. In this work, we optimized this dispersion coefficient against a benchmark set of 1031 experimental solvation energies from the extended solvation model with state-specific polarizability-(SMSSP) data set, comprising 29 low-dielectric-constant solvents. This parametrization yielded an optimized value of cf⁽⁶⁾ = 0.1650 kcal mol^-1. Subsequently, this refined parameter was applied to calculate solvation energies for solvents containing large electronegative atoms. Evaluation across 1054 data points spanning 61 solvents demonstrated a substantial improvement in accuracy compared to results obtained with the conventional parametrization.