Loading…

Sym4state.jl: An efficient computation package for magnetic materials

Exploring magnetic configurations of magnets often involves utilizing the four-state method to obtain the magnetic interaction matrix, and Monte Carlo method to simulate spin textures and phase transition processes. However, computing the interaction matrix between magnetic atoms using the four-stat...

Full description

Saved in:
Bibliographic Details
Published in:Computer physics communications 2024-10, Vol.303, p.109283, Article 109283
Main Authors: Wan, Guolin, Li, Yuhui, Lai, Ting, Li, Peixuan, Zhu, Yongqian, Yang, Jingyu, Zhang, Yan-Fang, Pan, Jinbo, Du, Shixuan
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Exploring magnetic configurations of magnets often involves utilizing the four-state method to obtain the magnetic interaction matrix, and Monte Carlo method to simulate spin textures and phase transition processes. However, computing the interaction matrix between magnetic atoms using the four-state method requires plenty of individual calculations. Despite manual simplifying the number of individual calculations based on material's symmetry is possible, there remains a necessity for an automated approach to streamline the process for high-throughput screening of magnetic materials. Meanwhile, the traditional sequential Monte Carlo simulation encounters challenges of low efficiency and long time consuming in dealing with large systems. Furthermore, the prior parallelism in the Heisenberg model was limited to parallel computation of the system's energy or run several replicas in parallel. Hence, in our pursuit of comprehensive parallelization for the Heisenberg model, we have introduced a novel adaptation of the checkerboard algorithm, enabling a fully parallelizable simulation of the Heisenberg model. To address these problems, we have developed Sym4state.jl, a program specifically designed to simplify the computation of magnetic interaction matrix and simulate spin textures under various environmental conditions. This program, available as a Julia package, can be freely accessed at https://github.com/A-LOST-WAPITI/Sym4state.jl. Program title: Sym4state.jl CPC Library link to program files:https://doi.org/10.17632/s6dkmgrjfw.1 Developer's repository link:https://github.com/A-LOST-WAPITI/Sym4state.jl Licensing provisions: MIT Programming language: Julia Nature of problem: Employing the four-state method to calculate magnetic interaction matrix for magnetic materials can be simplified based on material symmetry, however, there is a lack of automated approach to streamline the simplification. Additionally, the commonly used Metropolis method for simulating magnetic texture can only make parallel computation of the system's energy or run several replicas in parallel, which could hardly boost the performance when simulating the large-scale magnetic textures. Solution method: We simplify the four-state method calculations by utilizing the principles of energy invariance under symmetry operations and time reversal operations. To enhance the efficiency of the Metropolis algorithm, we have designed a strategy to divide the entire 2D lattice into multiple domains.
ISSN:0010-4655
1879-2944
DOI:10.1016/j.cpc.2024.109283