270087 Vapor-Liquid Equilibria Using Monte Carlo Wang-Landau Simulations

Monday, October 29, 2012
Hall B (Convention Center )
Tsvetan Aleksandrov, Caroline Desgranges and Jerome Delhommelle, Department of Chemistry, University of North Dakota, Grand Forks, ND

Molecular simulations have become an indispensable tool in the calculation of phase behavior of fluids. Simulation methods clearly do not suffer from the inconveniences of taking measurements at regions of instability of the molecules or in harsh condition at elevated temperature and pressure. The computational determination of vapor-liquid phase equilibria relies on a statistical mechanics sampling that relates the microscopic states along a path that connects the two phases at coexistence. The diculty presented is in the ecient sampling of the interface region with low statistical probabilities. Advancements in the past two decades have allowed the calculation of properties at coexistence of the two phases, a task that is unfit for standard simulation methods such as Monte Carlo (MC) and molecular dynamics (MD). Gibbs Ensemble Monte Carlo (GEMC) method, introduced by Panagiotopoulos1 has gained a signicant popularity in applications regarding vapor-liquid systems at equilibrium, due to its simplicity and robustness. More recently the Wang-Landau (WL) method has been developed to sample uniformly microstates that are characterized with uneven probability distribution. Like the GEMC and histogram reweighting, the method is simple to use, as a single simulation run is necessary to give rise to the properties at equilibrium. Moreover it allows the sampling of very high energy microstates at the interface through an accurate estimation of the partition function. We present applications of this method to a wide range of systems, including metals, linear and branched alkanes.

Extended Abstract: File Not Uploaded