Energy Probability Distribution Zeros: A Route to Study Phase Transitions

TL;DR

This paper introduces an iterative method leveraging Fisher zeros of the partition function to analyze phase transitions, providing high-precision critical parameters without prior order parameter knowledge, applicable to various models including classical and quantum systems.

Contribution

The novel approach uses partial Fisher zeros to determine critical behavior, surpassing traditional methods in efficiency and applicability to complex models.

Findings

Accurately determines critical temperature and exponents.

Works effectively for models with different transition types.

Applicable to classical and quantum systems.

Abstract

In the study of phase transitions a very few models are accessible to exact solution. In the most cases analytical simplifications have to be done or some numerical technique has to be used to get insight about their critical properties. Numerically, the most common approaches are those based in Monte Carlo simulations together finite size scaling analysis. The use of Monte Carlo techniques requires the estimate of quantities like the specific heat or susceptibilities in a wide range of temperature or the construction of the density of states in large intervals of energy. Although many of these techniques are well developed they may be very time consuming when the system size becomes large enough. It should be suitable to have a method that could surpass those difficulties. In this work we present an iterative method to study the critical behavior of a system based on the partial knowledge of the complex Fisher zeros set of the partition function. The method is general with advantages over most conventional techniques since it does not need to identify any order parameter \emph{a priori}. The critical temperature and exponents can be obtained with great precision even in the most unamenable cases like the two dimensional $XY$ model. To test the method and to show how it works we applied it to some selected models where the transitions are well known: The 2D Ising, Potts and XY models and to a homopolymer system. Our choice cover systems with first order, continuous and Berezinskii-Kosterlitz-Thouless transitions as well as the homopolymer that has two pseudo-transitions. The strategy can easily be adapted to any model, classical or quantum, once we are able to build the corresponding energy probability distribution.