Electronic structure and phase stability of oxide semiconductors: Performance of dielectric-dependent hybrid functional DFT, benchmarked against $GW$ band structure calculations and experiments

TL;DR

This study evaluates the effectiveness of dielectric-dependent hybrid functional DFT in predicting electronic and structural properties of oxide semiconductors, benchmarking against GW calculations and experiments to validate its accuracy.

Contribution

It demonstrates that dielectric-dependent hybrid functional DFT accurately predicts properties of oxide semiconductors, offering a reliable ab initio approach compared to traditional methods.

Findings

Hybrid functional DFT reproduces experimental band gaps and phase stability.

Good agreement with GW calculations for electronic properties.

Effective in modeling chemical reduction processes in TiO2.

Abstract

We investigate band gaps, equilibrium structures, and phase stabilities of several bulk polymorphs of wide-gap oxide semiconductors ZnO, TiO2,ZrO2, and WO3. We are particularly concerned with assessing the performance of hybrid functionals built with the fraction of Hartree-Fock exact exchange obtained from the computed electronic dielectric constant of the material. We provide comparison with more standard density-functional theory and GW methods. We finally analyze the chemical reduction of TiO2 into Ti2O3, involving a change in oxide stoichiometry. We show that the dielectric-dependent hybrid functional is generally good at reproducing both ground-state (lattice constants, phase stability sequences, and reaction energies) and excited-state (photoemission gaps) properties within a single, fully ab initio framework.