Peaks in the electronic density of states are not strong predictors of Superconductivity

TL;DR

This study investigates whether peaks in the electronic density of states, associated with flat bands near the Fermi energy, reliably predict superconductivity, revealing complex correlations that challenge their use as universal indicators.

Contribution

The paper critically evaluates the predictive power of density of states peaks for superconductivity across different crystal symmetries, highlighting limitations in current screening approaches.

Findings

Anticorrelation between density of states peaks and superconductivity in tetragonal materials.

Density of states peaks are more common in hexagonal superconductors than non-superconductors.

Highlights the need for standardized digital resources for materials discovery.

Abstract

Electronic bands near the Fermi energy with minimal energy dispersion in k-space, i.e. "flat bands", are often said to be an important characteristic of superconducting materials. When extending over a significant region of the k-space, this leads to peaks in the electronic density of states. If one could systematically search the calculated electronic structures of materials, already available through the Materials Genome Initiative, for such peaks, then the rate of new superconductor discovery could be accelerated greatly. Here we test whether this characteristic actually distinguishes known superconductors from known non-superconductors. We find that there is an anticorrelation for tetragonal symmetry materials, but that hexagonal symmetry superconductors exhibit such peaks more frequently than hexagonal non-superconductors. More generally, our study's limitations underscore the importance of standardizing digital resources for achieving the accelerated materials discovery paradigm promised by the Materials Genome Initiative.