Importance of ligand on-site interactions for the description of Mott-insulators in DFT+DMFT

TL;DR

This paper highlights the critical role of ligand p-state corrections in DFT+DMFT calculations for Mott insulators, demonstrating that including these states or applying empirical corrections leads to more accurate insulating behavior.

Contribution

It shows that accounting for ligand p states and applying empirical corrections improves the accuracy of DFT+DMFT in describing Mott insulators, addressing limitations of previous approaches.

Findings

Including ligand p states yields correct insulating behavior.

Empirical corrections to p states improve interaction parameters.

Realistic corrections enable accurate DFT+DMFT descriptions of LaTiO3, LaVO3, and RNiO3.

Abstract

Calculations combining density functional theory (DFT) and dynamical mean-field theory (DMFT) for transition metal (TM) oxides and similar compounds usually focus on improving the description of the TM $d$ states. Here, we emphasize the importance of also accounting for corrections of the ligand $p$ states. We demonstrate that focusing exclusively on an improved description of the TM $d$ states results in difficulties to obtain the correct insulating behavior for a variety of materials, and requires to use values for the local interaction parameters that are inconsistent with values obtained using, e.g., the constrained random phase approximation (cRPA). Importantly, these considerations not only apply to cases where the $p$ states are explicitly included in the DMFT low-energy subspace, but also to cases which only include so-called frontier bands with dominant TM $d$ character. We demonstrate that, to a large part, these inconsistencies arise from the use of local/semi-local DFT as starting point for computing interaction parameters within cRPA, and we show that applying a simple empirical correction to the O $p$ and other low energy states not included in the correlated subspace results in improved values for the interaction parameters that then allow to obtain the correct insulating behavior. For the cases where the ligand states are included in the DMFT subspace, we show that even an approximate but realistic Hartree-Fock-like correction applied to the O $p$ states leads to a correct and, most importantly, also a quantitatively consistent DFT+DMFT description of typical Mott insulators such as LaTiO$_3$, LaVO$_3$, or the perovskite rare-earth nickelates, $R$NiO$_3$.