Accurate DFT simulation of complex functional materials: Synergistic enhancements achieved by SCAN meta-GGA

Abstract

Complex functional materials are characterized by intricate and competing bond orders, making them an excellent platform for evaluating the newly developed strongly constrained and appropriately normed (SCAN) density functional. In this study, we explore the effectiveness of SCAN in simulating the electronic properties of displacive ferroelectrics (BaTiO3 and PbTiO3) and magnetoelectric multiferroics (BiFeO3 and YMnO3), which encompass a broad spectrum of bonding characteristics. Due to a significant reduction in self-interaction error, SCAN manifests its improvements over the Perdew–Burke–Ernzerhof (PBE) method in three aspects: SCAN predicts more accurate ionicity, produces more compact orbitals, and better captures d-orbital anisotropy. Particularly, these synergistic enhancements lead to notable phenomena in calculating the bandgap of YMnO3: while the PBE+U simulation may suggest a strong correlation appearance attributed to high Hubbard-like U values (∼5 eV), the value is dramatically lower (∼1 eV) in the SCAN+U method. Furthermore, we provide an intuitive analysis of SCAN’s operational principles by examining the complex electron densities involved. These insights are theoretically intriguing and have practical implications, potentially encouraging wider adoption of SCAN in the computational modeling of complex functional materials.