Gibbs Project - Abstract to AIChE 1996

Molecular Modeling of Phase Equilibria

V. I. Harismiadis, J. Vorholz, J. Errington, A. Z. Panagiotopoulos
School of Chemical Engineering, Cornell University, Ithaca, NY 14853-5201, USA

Reliable data for phase equilibria of mixtures are necessary for the proper design of chemical processes. Experimental measurement of the phase behavior for complex mixtures can be an expensive and time-consuming procedure, even when specialized equipment and personnel are available for this task.

Recent advances in molecular simulation methods now enable the routine calculation of the phase behavior of fluid mixtures. These techniques include the Gibbs ensemble Monte Carlo method combined with configurational-bias sampling. The main constraint limiting the application of molecular methods to problems of industrial importance is the lack of appropriate intermolecular potential models valid over a large range of temperatures, pressures and compositions.

We have initiated a project to develop software tools that will allow the rapid optimization of intermolecular potential parameters for real mixtures. In this presentation, we will describe the key elements of the project and illustrate its initial applications. Results for pure components and mixtures involving small polar molecules (such as water and methanol), light gases (such as carbon dioxide and ethane), and electrolytes will be presented. Issues related to the selection of intermolecular potential models for polar systems will be addressed.