Melting of Rare-Gas Crystals: Monte Carlo Simulation versus Experiments

TL;DR

This study uses Monte Carlo simulations with Lennard-Jones potential to analyze the melting points of rare-gas crystals, revealing discrepancies with experimental data and proposing parameter adjustments for better accuracy.

Contribution

It provides a detailed comparison of Monte Carlo simulation results with experiments and introduces modified parameters for improved melting temperature predictions.

Findings

Simulated melting temperatures are 13-20% higher than experimental values.

Finite-size and cutoff effects significantly influence simulation accuracy.

Modified Lennard-Jones parameters improve agreement with experimental melting points.

Abstract

We study the melting transition in crystals of rare gas Ar, Xe, and Kr by the use of extensive Monte Carlo simulations with the Lennard-Jones potential. The parameters of this potential have been deduced by Bernardes in 1958 from experiments of rare gas in the gaseous phase. It is amazing that the parameters of such a popular potential were not fully tested so far. In order to carry out precise tests, we have written a high-performance Monte Carlo program which allows in particular to take into account correctly the periodic boundary conditions to reduce surface effects and to reduce CPU time. Using the Bernardes parameters, we find that the melting temperature of several rare gas is from 13 to 20% higher than that obtained from experiments. We have throughout studied the case of Ar by examining both finite-size and cutoff-distance effects. In order to get a good agreement with the experimental melting temperature, we propose a modification of these parameters to describe better the melting of rare-gas crystals.