Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Magnetic Structures Database from Symmetry-aided High-Throughput Calculations

Hanjing Zhou, Yuxuan Mu, Dingwen Zhang, Hangbing Chu, Di Wang, Huimei Liu, Xiangang Wan

TL;DR

The paper tackles the difficulty of predicting magnetic ground states by introducing a symmetry-guided approach based on Landau theory, which constrains candidate magnetic orders to irreducible representations of the parent space group via Wyckoff positions. Ground states are then selected through first-principles calculations, enabling high-throughput magnetic-structure prediction. Benchmarking on 302 MAGNDATA structures yields ~70% accuracy, and application to 7,520 ICSD compounds builds a magnetic-structure database of 2,901 materials, facilitating systematic exploration of magnetic topology and altermagnetism (identifying 1,070 magnetic topological materials and 392 altermagnets). This framework provides a scalable route to map magnetic order to physical properties across large material spaces, with potential extensions to two-dimensional magnets and additional physical properties.

Abstract

Magnetic structures, which play a central role in determining their physical properties, are known for only very limited compounds. Traditional theoretical methods based on first-principles calculations are fundamentally limited by the need to calculate a large space of input magnetic configurations. Here we introduce a symmetry-aided strategy based on Landau's phase transition theory. By utilizing the crystallographic space group and the Wyckoff positions of magnetic ions, we narrow down the initial magnetic configurations to a limited number of candidates via the analysis of the group representations. The magnetic ground state is subsequently determined by the lowest energy of those well-seleted magnetic configurations via first-principles calculations. Benchmarking calculations were performed on a subset of the MAGNDATA database with wave vector q=0 and fewer than 40 atoms per unit cell, comprising 302 materials. Our method successfully identified the magnetic structures for 212 of these materials. We further apply our highly efficient method to 7,520 stoichiometric transition metal compounds with fewer than 20 atoms per unit cell in the Inorganic Crystal Structure Database, and establish a magnetic structure database containing 2,900 magnetic materials. To demonstrate the utility of our database, we apply it to the systematic exploration of magnetic topological phases and altermagnets, leading to the identification of 1,070 and 392 instances, respectively.

Magnetic Structures Database from Symmetry-aided High-Throughput Calculations

TL;DR

The paper tackles the difficulty of predicting magnetic ground states by introducing a symmetry-guided approach based on Landau theory, which constrains candidate magnetic orders to irreducible representations of the parent space group via Wyckoff positions. Ground states are then selected through first-principles calculations, enabling high-throughput magnetic-structure prediction. Benchmarking on 302 MAGNDATA structures yields ~70% accuracy, and application to 7,520 ICSD compounds builds a magnetic-structure database of 2,901 materials, facilitating systematic exploration of magnetic topology and altermagnetism (identifying 1,070 magnetic topological materials and 392 altermagnets). This framework provides a scalable route to map magnetic order to physical properties across large material spaces, with potential extensions to two-dimensional magnets and additional physical properties.

Abstract

Magnetic structures, which play a central role in determining their physical properties, are known for only very limited compounds. Traditional theoretical methods based on first-principles calculations are fundamentally limited by the need to calculate a large space of input magnetic configurations. Here we introduce a symmetry-aided strategy based on Landau's phase transition theory. By utilizing the crystallographic space group and the Wyckoff positions of magnetic ions, we narrow down the initial magnetic configurations to a limited number of candidates via the analysis of the group representations. The magnetic ground state is subsequently determined by the lowest energy of those well-seleted magnetic configurations via first-principles calculations. Benchmarking calculations were performed on a subset of the MAGNDATA database with wave vector q=0 and fewer than 40 atoms per unit cell, comprising 302 materials. Our method successfully identified the magnetic structures for 212 of these materials. We further apply our highly efficient method to 7,520 stoichiometric transition metal compounds with fewer than 20 atoms per unit cell in the Inorganic Crystal Structure Database, and establish a magnetic structure database containing 2,900 magnetic materials. To demonstrate the utility of our database, we apply it to the systematic exploration of magnetic topological phases and altermagnets, leading to the identification of 1,070 and 392 instances, respectively.
Paper Structure (8 sections, 1 equation, 4 figures)

This paper contains 8 sections, 1 equation, 4 figures.

Table of Contents

  1. Introduction
  2. Method
  3. Magnetic Topological Materials
  4. Magnetic topological insulators
  5. Magnetic topological semimetals
  6. Altermagnets
  7. Outlook
  8. Acknowledgements

Figures (4)

  • Figure 1: Magnetic and topological classification of our magnetic structure database. (a) Magnetic ground-state classification. Among the 2,901 identified magnetic materials, 49.78% exhibit ferromagnetic (FM) ordering, 11.58% are ferrimagnetic (FiM), and the remaining 38.64% are antiferromagnetic (AFM). The FiM materials are further classified into 7.65% collinear (Col) and 3.93% noncollinear (NoC) states according to the relative orientation of magnetic moments. Within the AFM materials, 20.85% exhibit collinear (Col) order, 4.27% exhibit noncollinear (NoC) order, and 13.51% are classified as altermagnets (Alter). (b) Topological classification of magnetic materials. In total, 38% of the magnetic materials in the database are identified as magnetic topological materials, including 4% magnetic topological insulators (TIs) and 34% magnetic topological semimetals (TSMs).
  • Figure 2: AFM axion insulator FeSe. (a) magnetic structure (b) band structure. The green and yellow balls are Se and Fe ions, respectively. The magnetic moments of Fe are along the $b$-axis ([010]), as indicated by red arrow.
  • Figure 3: Topological magnetic semimetals. (a)-(c) Magnetic structure, band structure, and distribution of Weyl points of the non-collinear AFM Weyl semimetal V$_3$Ga$_2$. The magnetic moments of V lie in the $xy$ plane, as indicated by red arrow. (d)-(f) Magnetic structure, band structure, and distribution of Dirac points of the AFM Dirac semimetal MnTe. (g)-(i) Magnetic structure, band structure, and distribution of nodal rings and Weyl points of the FM nodal-line semimetal Co$_3$SnC. The red and blue dots in (c) and (k) denote Weyl points with topological charges +1 and $-1$, respectively, while the green lines represent the nodal rings.
  • Figure 4: (a) Crystal structure and (b) band structure of CsV$_2$S$_2$O.