Monte Carlo method for obtaining the ground-state properties of quantum spin systems

Abstract
A Monte Carlo method to find the ground-state properties of quantum spin systems is presented. Transforming a quantum spin Hamiltonian in a matrix with non-negative elements, we set up a Markov process whose stationary probability is dominated by the leading eigenvector of this matrix. From the simulation of the Markov process, by means of a Metropolis algorithm, we obtain the properties and the energy of the ground state. The method is applied to the spin-1 isotropic, Heisenberg antiferromagnet chain. © 1996 The American Physical Society.