We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Condensed Matter > Materials Science

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

Abstract: Magnetic topological insulators and semi-metals 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 work, we use systematic high-throughput density functional theory calculations to identify magnetic topological materials from 40000 three-dimensional materials in the JARVIS-DFT database (this https URL). First, we screen materials with net magnetic moment > 0.5 {\mu}B and spin-orbit spillage > 0.25, resulting in 25 insulating and 564 metallic candidates. The spillage 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 bandstructures, and Fermi-surfaces to determine interesting topological characteristics of the screened compounds. We also train machine learning models for predicting the spillage, bandgaps, 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.
Subjects: Materials Science (cond-mat.mtrl-sci)
Journal reference: Phys. Rev. B 103, 155131 (2021)
DOI: 10.1103/PhysRevB.103.155131
Cite as: arXiv:2102.00237 [cond-mat.mtrl-sci]
  (or arXiv:2102.00237v2 [cond-mat.mtrl-sci] for this version)

Submission history

From: Kamal Choudhary [view email]
[v1] Sat, 30 Jan 2021 15:09:49 GMT (2118kb)
[v2] Fri, 5 Feb 2021 16:37:35 GMT (2118kb)

Link back to: arXiv, form interface, contact.