A new algorithm for Reverse Monte Carlo simulations

Abstract

We present a new algorithm for Reverse Monte Carlo (RMC) simulations of liquids. During the simulations, we calculate energy, excess chemical potentials, bond-angle distributions and three-body correlations. This allows us to test the quality and physical meaning of RMC-generated results and its limitations. It also indicates the possibility to explore orientational correlations from simple scattering experiments. The new technique has been applied to bulk hard-sphere and Lennard-Jones systems and compared to standard Metropolis Monte Carlo results. (C) 1998 American Institute of Physics.