MANIPULATING AND SIMPLIFYING THE INTERMOLECULAR INTERACTIONS IN LIQUID MIXTURES

Abstract

Long ranged intermolecular interactions have significant influence on the structure of the liquid and present serious challenges for computer simulations. In particular, the long ranged tail of Coulomb interaction usually needs to be calculated using Ewald summation or related techniques in computer simulation, which can be too time consuming to be carried out for large systems. Local Molecular Field(LMF) theory has been developed to simplify long-ranged Coulomb and Van der Waals interactions for nonuniform liquids by approximating these long ranged interactions as effective static single-particle fields. Despite the success LMF theory made in describing the structure of nonuniform liquids, it is not appropriate to use LMF theory to describe the structure of uniform liquid mixtures, since the dynamically moving unbalanced forces produced in mixture can not be captured by the framework of LMF theory. In this thesis, we propose a new framework which approximates the unbalanced forces produced in a mixture as effective intermolecular interactions. This new framework can simplify the long ranged intermolecular interactions and produce a mimic system with short ranged solvent-solvent interactions, which is much easier to simulate or analyze. Based on this framework and other techniques introduced in this thesis, we have constructed a "Short Solvent Model", which has noticeable advantages compared to the explicit solvent model and implicit solvent model. This framework has also been used to simplify the interactions of phase-separating mixtures. The impact of using this framework on the diffusion dynamics of the solutes has also been discussed. Possible application of this framework and the Short Solvent Model to biopolymers folding problems is briefly discussed.