Extending the J-walking Monte Carlo algorithm to the isothermal-isobaric ensemble: solid-liquid equilibrium in clusters

The J-walking Monte Carlo algorithm is extended to the isothermal-isobaric ensemble (NPT) in order to reduce the problem of quasi-ergodicity encounter in the standard Monte Carlo procedure. Anomalies in the constant-pressure heat capacity as a function of temperature are identified an associated to the solid-liquid equilibrium. The maximum in temperature in this transition region is used for the construction of P-T phase diagram. The new algorith m reduces the uncertainty in the heat capacity by an order of magnitude whe n compared to standard NPT Monte Carlo. (C) 1998 Elsevier Science B.V. All rights reserved.