SLIC (solvation-layer interface condition)

Abstract

Implicit solvent models have employed continuum theory to describe the electrostatic interactions between biomolecules and the surrounding solvents for years. These models are simpler to implement and orders of magnitude faster than explicit-solvent molecular-dynamics (MD). Macroscopic continuum theory, however, exhibits sub- stantial deviations from experimental results when applied to systems of very short (atomistic or near-atomistic) length scales. The most significant errors, in particu- lar, are due to the poor treatment of the first solvation shell of solvent molecules in continuum electrostatics theory (the first layer of the solvent molecules surrounding the solute). The first chapter of this work overviews the major shortcomings of the continuum dielectric theory and the second chapter presents the standard continuum electrostatics theory. Third chapter will focus on a simple modification to the classical boundary condi- tion that has been shown to make predictions of solvation thermodynamics remark- ably more accurate. The standard continuum model and a boundary-integral formu- lation of the problem, along with some details on numerical implementation will be presented. The major physical mechanisms that are incorporated in the new model to remedy the inaccuracies of the standard continuum theory will also be demonstrated. A novel nonpolar model to capture different nonpolar components of solvation will also be demonstrated. Finally, an assessment of the new model's accuracy will be illustrated in chapter 4 by comparing the model predictions of solvation thermodynamics of monovalent ions and neutral small molecules in water and multiple ionic liquids to experimental data. MD simulations. The thesis concludes with a discussion in chapter 5.