Exact calculations of fluid-phase equilibria by Monte Carlo simulation in a new statistical ensemble

Research output: Contribution to journalArticlepeer-review

102 Scopus citations

Abstract

A recently proposed method, Monte Carlo simulation in the Gibbs ensemble, allows the prediction of phase equilibria from knowledge of the intermolecular forces. A single computer experiment is required per coexistence point for a system with an arbitrary number of components. The new technique has significant advantages relative to free-energy methods that have been used for phase equilibrium calculations is the past. In this work, a variation of the Gibbs method appropriate for calculations in mixtures with large differences in molecular size is developed. The method is applied for the calculation of high-pressure phase equilibria in two mixtures of simple monatomic fluids, the systems argon-krypton and neon-xenon. Pairwise additive potential functions of the Lennard-Jones type are used to describe the intermolecular interactions. Agreement with experimental results is generally good over a wide range of temperatures and pressures, including the fluid-fluid immiscibility region for the neon-xenon system. Results from the Van der Waals one-fluid theory are compared with experimental data and computer simulation predictions. Agreement is excellent for the mixture with small differences in size (argon-krypton), but the theory fails to describe the coexistence curve for the highly asymmetric system neon-xenon.

Original languageEnglish (US)
Pages (from-to)447-457
Number of pages11
JournalInternational Journal of Thermophysics
Volume10
Issue number2
DOIs
StatePublished - Mar 1 1989
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Condensed Matter Physics

Keywords

  • Lennard-Jones
  • Monte Carlo
  • computer simulation
  • intermolecular potential functions
  • phase equilibria
  • vapor-liquid equilibria

Fingerprint Dive into the research topics of 'Exact calculations of fluid-phase equilibria by Monte Carlo simulation in a new statistical ensemble'. Together they form a unique fingerprint.

Cite this