Commentary on Liquid Simulations and Industrial Applications

Abstract

Molecular dynamics and Monte Carlo simulations have proven to be invaluable tools in the study of liquids. Statistical mechanics tells us what to calculate to obtain liquid state properties (equation of state, specific heat, and transport coefficients for example) from a model of the interaction between molecules. Simulations are the tools needed to make those calculations for physically realistic models. While the impact of simulations on the investigation of fluid properties and structure cannot be overstated, there has been very little success in the use of these methods to predict liquid properties outside the range where model potentials have been fit to some property. One result of this is that industrial modelers, who are called upon to estimate properties for various compositions, temperatures, and pressures of liquids, have not found simulation methods to be of much value for their work. Some of the reasons for the lack of use of molecular level simulations in the industrial sector will be examined here. These reasons point to opportunities and challenges for research that could lead to industrially useful molecular simulation methods and practices.