New implicit solvation models for dispersion and exchange energies

Area of Science:

• Computational Chemistry
• Physical Chemistry
• Molecular Modeling

Background:

• Implicit solvation models efficiently estimate solvation energies, with dielectric continuum models commonly used for long-range electrostatic interactions.
• Existing models may inadequately represent short-range interactions like dispersion and exchange, necessitating specialized approaches.
• Adapting gas-phase intermolecular force treatments offers a promising avenue for developing improved implicit solvation models.

Purpose of the Study:

• To develop novel implicit solvation models specifically designed to capture short-range dispersion and exchange interactions between solutes and solvents.
• To formulate these models based on solute charge densities, utilizing a minimal number of adjustable parameters.
• To evaluate the performance of the proposed models in predicting solvation free energies for various solutes in nonpolar solvents.

Main Methods:

• Formulation of new implicit models for dispersion and exchange interactions, adapted from gas-phase intermolecular force methodologies.
• Models are expressed in terms of solute charge densities and incorporate three adjustable parameters.
• Electronic structure calculations were performed for numerous solutes in two nonpolar solvents to assess model performance.

Main Results:

• The new models, after empirical parameter optimization, achieved an average error of approximately 0.4 kcal/mol for solvation free energies.
• This level of accuracy surpasses previous methods that often require a larger number of parameters.
• The models effectively capture the balance between attractive dispersion and repulsive exchange forces in nonpolar solvents.

Conclusions:

• The proposed implicit solvation models provide a highly efficient and accurate means to account for short-range solute-solvent interactions.
• These models represent a significant advancement over existing methods, offering improved predictions with fewer parameters.
• The developed approach holds promise for enhancing the accuracy and efficiency of molecular modeling and computational chemistry applications.