Free energy and phase equilibria for the restricted primitive model of ionic fluids from Monte Carlo simulations

Gerassimos Orkoulas, Athanassios Z. Panagiotopoulos

Research output: Contribution to journalArticlepeer-review

214 Scopus citations


In this work, we investigate the liquid-vapor phase transition of the restricted primitive model of ionic fluids. We show that at the low temperatures where the phase transition occurs, the system cannot be studied by conventional molecular simulation methods because convergence to equilibrium is slow. To accelerate convergence, we propose cluster Monte Carlo moves capable of moving more than one particle at a time. We then address the issue of charged particle transfers in grand canonical and Gibbs ensemble Monte Carlo simulations, for which we propose a biased particle insertion/destruction scheme capable of sampling short interparticle distances. We compute the chemical potential for the restricted primitive model as a function of temperature and density from grand canonical Monte Carlo simulations and the phase envelope from Gibbs Monte Carlo simulations. Our calculated phase coexistence curve is in agreement with recent results of Caillol obtained on the four-dimensional hypersphere and our own earlier Gibbs ensemble simulations with single-ion transfers, with the exception of the critical temperature, which is lower in the current calculations. Our best estimates for the critical parameters are T c*=0.053, ρc*=0.025. We conclude with possible future applications of the biased techniques developed here for phase equilibrium calculations for ionic fluids.

Original languageEnglish (US)
Pages (from-to)1452-1459
Number of pages8
JournalThe Journal of chemical physics
Issue number2
StatePublished - 1994
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • General Physics and Astronomy
  • Physical and Theoretical Chemistry


Dive into the research topics of 'Free energy and phase equilibria for the restricted primitive model of ionic fluids from Monte Carlo simulations'. Together they form a unique fingerprint.

Cite this