Vapor Liquid Equilibrium Calculation for Binary Mixtures Using Molecular Dynamics Simulation

This work has been aimed to develop a computational method based on the molecular dynamics technique to predict the local composition (LC) interaction parameters (Wilson, NRTL, and UNIQUAC) for VLE calculations, and check validity of LC models. Five binary mixtures of water-acetonitrile, water-isopropanol, methanol-chloroform, acetone-cyclohexane, and meta xylene-benzene were simulated (each at five mole fraction). MD simulations were performed using the COMPASS force field. Although concept of LC models consider only nearest-neighbor layer, but in this study all range of radial distribution function (RDF) were imposed. Results show that interaction parameters are composition and radius dependent. Although effective radius is where the fluctuation of RDF propagates, but optimum radius for interaction parameters were determined using experimental data. The findings demonstrate that interaction parameters must be composition dependent so that those were best fitted to a third-order polynomial equation. In addition, the interaction parameters were determined though experimental data optimization and were compared with the present MD results. It turned out that Wilson and UNIQUAC models are more consistency with the MD results. Finally, results of UNIQUAC model are better than Wilson and NRTL models.