High-throughput Discovery of Topologically Non-trivial Materials using Spin-orbit Spillage

TL;DR

This paper introduces a fast, symmetry-independent method using spin-orbit spillage to identify topologically non-trivial materials from large DFT databases, successfully discovering numerous candidates including insulators, semimetals, and crystalline topological phases.

Contribution

The authors develop and validate a new, efficient approach based on spin-orbit spillage for high-throughput topological material discovery without relying on symmetry analysis.

Findings

Identified 1868 candidate topological materials from 4835 analyzed.

Validated the method with Wannier-interpolation on 289 candidates.

Successfully detected various topological phases including insulators and semimetals.

Abstract

We present a novel methodology to identify topologically non-trivial materials based on band inversion induced by spin-orbit coupling (SOC) effect. Specifically, we compare the density functional theory (DFT) based wavefunctions with and without spin-orbit coupling and compute the spin-orbit-spillage as a measure of band-inversion. Due to its ease of calculation, without any need for symmetry analysis or dense k-point interpolation, the spillage is an excellent tool for identifying topologically non-trivial materials. Out of 30000 materials available in the JARVIS-DFT database, we applied this methodology to more than 4835 non-magnetic materials consisting of heavy atoms and low bandgaps. We found 1868 candidate materials with high-spillage (using 0.5 as a threshold). We validated our methodology by carrying out conventional Wannier-interpolation calculations for 289 candidate materials. We demonstrate that in addition to Z2 topological insulators, this screening method successfully identified many semimetals and topological crystalline insulators. Importantly, our approach is applicable to the investigation of disordered or distorted as well as magnetic materials, because it is not based on symmetry considerations. We discuss some individual example materials, as well as trends throughout our dataset, which is available at the websites: https://www.ctcms.nist.gov/~knc6/JVASP.html and https://jarvis.nist.gov/.