Worm algorithms for the 3-state Potts model with magnetic field and chemical potential

TL;DR

This paper introduces worm algorithms for simulating the 3-state Potts model with external field and chemical potential, overcoming the complex phase problem via a flux representation with dimers and monomers.

Contribution

The paper presents two novel generalizations of the Prokof'ev-Svistunov worm algorithm for efficient Monte Carlo simulations at arbitrary chemical potential.

Findings

Flux representation enables overcoming the complex phase problem.

Two worm algorithms are proposed and evaluated for performance.

Algorithms facilitate simulations of the Potts model under external field and chemical potential.

Abstract

We discuss worm algorithms for the 3-state Potts model with external field and chemical potential. The complex phase problem of this system can be overcome by using a flux representation where the new degrees of freedom are dimer and monomer variables. Working with this representation we discuss two different generalizations of the conventional Prokof'ev-Svistunov algorithm suitable for Monte Carlo simulations of the model at arbitrary chemical potential and evaluate their performance.