High-throughput search for magnetic topological materials using spin-orbit spillage, machine learning, and experiments

• Published in: Physical review. B 2021-04, Vol.103 (15), p.1, Article 155131
• Main Authors: Choudhary, Kamal, Garrity, Kevin F., Ghimire, Nirmal J., Anand, Naween, Tavazza, Francesca
• Format: Article
• Language: English
• Subjects: CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND SUPERFLUIDITY; Density functional theory; electronic structure; Fermi surfaces; high-throughput calculations; Machine learning; Magnetic moments; Magnetic properties; Materials selection; Metalloids; Quantum computing; Spillage; Spintronics; Topological insulators; topological phases of matter
• ISSN: 2469-9950; 2469-9969
• DOI: 10.1103/PhysRevB.103.155131

Magnetic topological insulators and semimetals have a variety of properties that make them attractive for applications, including spintronics and quantum computation, but very few high-quality candidate materials are known. In this paper, we use systematic high-throughput density functional theory calculations to identify magnetic topological materials from the ≈40 000 three-dimensional materials in the JARVIS-DFT database. First, we screen materials with net magnetic moment > 0.5 μB and spin-orbit spillage (SOS) > 0.25 , resulting in 25 insulating and 564 metallic candidates. The SOS acts as a signature of spin-orbit-induced band-inversion. Then we carry out calculations of Wannier charge centers, Chern numbers, anomalous Hall conductivities, surface band structures, and Fermi surfaces to determine interesting topological characteristics of the screened compounds. We also train machine learning models for predicting the spillages, band gaps, and magnetic moments of new compounds, to further accelerate the screening process. We experimentally synthesize and characterize a few candidate materials to support our theoretical predictions.