Application of reverse Monte Carlo simulations to diatomic molecules. Part 1. Noncomplete radial distribution functions

Abstract

The reverse Monte Carlo (RMC) method generates sets of points in space which yield radial distribution functions (RDFS) that approximate those of the system of interest. Such sets of configurations should, in principle, be sufficient to determine the structural properties of the system. In this work we apply the RMC technique to fluids of hard diatomic molecules. The experimental RDFs of the hard-dimer fluid were generated by the conventional MC method and used as input in the RMC simulations. Our results indicate that the RMC method is only satisfactory in determining the local structure of the fluid studied by means of only mono-variable RDF. Also we suggest that the use of multi-variable RDFs would improve the technique significantly. However, the accuracy of the method turned out to be very sensitive to the variance of the input experimental RDF. © 1995.