Development of a methodology to compute solvation free energies on the basis of the theory of energy representation for solutions represented with a polarizable force field

We have developed a method of molecular simulations utilizing a polarizable force field in combination with the theory of energy representation (ER) for the purpose of establishing an efficient and accurate methodology to compute solvation free energies. The standard version of the ER method is, however, based on the assumption that the solute-solvent interaction is pairwise additive for its construction. A crucial step in the present method is to introduce an intermediate state in the solvation process to treat separately the many-body interaction associated with the polarizable model. The intermediate state is chosen so that the solute-solvent interaction can be formally written in the pairwise form, though the solvent molecules are interacting with each other with polarizable charges dependent on the solvent configuration. It is, then, possible to extract the free energy contribution δμ due to the many-body interaction between solute and solvent from the total solvation free energy Δμ. It is shown that the free energy δμ can be computed by an extension of the recent development implemented in quantum mechanicalmolecular mechanical simulations. To assess the numerical robustness of the approach, we computed the solvation free energies of a water and a methanol molecule in water solvent, where two paths for the solvation processes were examined by introducing different intermediate states. The solvation free energies of a water molecule associated with the two paths were obtained as -5.3 and -5.8 kcalmol. Those of a methanol molecule were determined as -3.5 and -3.7 kcalmol. These results of the ER simulations were also compared with those computed by a numerically exact approach. It was demonstrated that the present approach produces the solvation free energies in comparable accuracies to simulations of thermodynamic integration (TI) method within a tenth of computational time used for the TI simulations.