A cluster algorithm for Monte Carlo simulation at constant pressure

Abstract

We propose an efficient algorithm to sample the volume in Monte Carlo simulations in the isobaric-isothermal ensemble. The method is designed to be applied in the simulation of hard-core models at high density. The algorithm is based in the generation of clusters of particles. At the volume change step, the distances between pairs of particles belonging to the same cluster do not change. This is done by rescaling the positions of the center of mass of each cluster instead of the position of each individual particle. We have tested the performance of the algorithm by simulating fluid and solid phases of hard spheres, finding that in both cases the algorithm is much more efficient than the standard procedure. Moreover, the efficiency of the method measured in terms of correlation ”time” does not depend on the system size in contrast with the standard method, in which the sampling becomes rapidly inefficient as the system size increases. We have used the procedure to compute with high precision the equation of state of the face-centered-cubic phase of the hard sphere system for different system sizes. Using these results we have estimated the equation of state at the thermodynamic limit. The results are compared to different equations of state proposed in literature