Systematic Improvement on the Classical Molecular Model of Water

Abstract

The theoretical and computational modeling of water requires striking a balance between a detailed description of the physical interactions and the numerical simplicity needed for sampling the condensed phase. Here we report the iAMOEBA classical polarizable water model, which is a simplified version of the AMOEBA model from ten years ago. iAMOEBA uses a or or first-order approximation to describe electronic polarizability, which reduces the computational cost relative to a fully polarizable model such as AMOEBA. The model is parameterized using ForceBalance, a systematic model optimization method which simultaneously utilizes training data from experimental measurements and high-level ab initio calculations. We show that iAMOEBA is a highly accurate model for water in the gas and condensed phases with the ability to predict a comprehensive set of properties outside of the training data. The increased accuracy of iAMOEBA over the fully polarizable AMOEBA model indicates the validity of the direct polarization approximation in effectively capturing polarization effects in water. Our work also demonstrates ForceBalance as a method which allows the researcher to systematically improve empirical models by optimally utilizing the available reference data.View Large Image | View Hi-Res Image | Download PowerPoint Slide