Home > Research > Publications & Outputs > Calculation of the melting point of NaCl by mol...

Links

Text available via DOI:

View graph of relations

Calculation of the melting point of NaCl by molecular simulation

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published
Close
<mark>Journal publication date</mark>8/01/2003
<mark>Journal</mark>Journal of Chemical Physics
Issue number2
Volume118
Number of pages8
Pages (from-to)728-735
Publication StatusPublished
<mark>Original language</mark>English

Abstract

We report a numerical calculation of the melting point of NaCl. The solid-liquid transition was located by determining the point where the chemical potentials of the solid and liquid phases intersect. To compute these chemical potentials, we made use of free energy calculations. For the solid phase the free energy was determined by thermodynamic integration from the Einstein crystal. For the liquid phase two distinct approaches were employed: one based on particle insertion and growth using the Kirkwood coupling parameter, and the other involving thermodynamic integration of the NaCl liquid to a Lennard-Jones fluid. The latter approach was found to be significantly more accurate. The coexistence point at 1074 K was characterized by a pressure of -30+/-40 MPa and a chemical potential of -97.9+/-0.2k(beta)T. This result is remarkably good as the error bounds on the pressure enclose the expected coexistence pressure of about 0.1 MPa (ambient). Using the Clausius-Clapyron relation, we estimate that dP/dTapproximate to3 MPa/K. This yields a melting point of 1064+/-14 K at ambient pressure, which encompasses the quoted range for the experimental melting point (1072.45-1074.4 K). The good agreement with the experimental melting-point data provides additional evidence that the Tosi-Fumi model for NaCl is quite accurate. Our study illustrates that the melting point of an ionic system can be calculated accurately by employing a judicious combination of free energy techniques. The techniques used in this work can be directly extended to more complex, charged systems. (C) 2003 American Institute of Physics.