Inverse transitions in the Ghatak–Sherrington model with bimodal random fields

Abstract

The present work studies the Ghatak-Sherrington (GS) model with the presence of a longitudinal magnetic random field (RF) $h_{i}$ following a bimodal distribution. The model considers a random bond interaction $J_{i,j}$ which follows a Gaussian distribution with mean $J_0/N$ and variance $J^2/N$. This allows us to introduce the bond disorder strength parameter $J/J_0$ to probe the combined effects of disorder coming from the random bond and the discrete RF over unusual phase transitions known as inverse transitions (ITs). The results within a mean field approximation indicate that these two types of disorder have complete distinct roles for the ITs. They indicate that bond disorder creates the necessary conditions for the presence of inverse freezing or even inverse melting depending on the bond disorder strength, while the RF tends to enforce mechanisms that destroy the ITs.