Replica-exchange algorithm and results for the three-dimensional random field Ising model

The random field Ising model with Gaussian disorder is studied using a diff erent Monte Carlo algorithm. The algorithm combines the advantages of the r eplica-exchange method and the two-replica cluster method and is much more efficient than the Metropolis algorithm for some disorder realizations. Thr ee-dimensional systems of size 24(3) are studied. Each realization of disor der is simulated at a value of temperature and uniform field that is adjust ed to the phase-transition region for that disorder realization. Energy and magnetization distributions show large variations from one realization of disorder to another. For some realizations of disorder there are three well separated peaks in the magnetization distribution and two well separated p eaks in the energy distribution suggesting a first-order transition.