GW approximation with LSDA+U method and applications to NiO, MnO, and V$_2$O$_3$
S. Kobayashi, Y. Nohara, S. Yamamoto, and T. Fujiwara

TL;DR
This paper introduces a new GW approximation method called U+GWA that starts from LSDA+U wave functions, applied to transition metal oxides, achieving accurate band gaps and spectra, with discussions on its limitations and criteria for application.
Contribution
The paper proposes the U+GWA method that improves GW calculations by incorporating localized wave functions from LSDA+U, and applies it to NiO, MnO, and V$_2$O$_3$ with detailed analysis.
Findings
Excellent agreement of band gaps with experiments for MnO, NiO, and V$_2$O$_3$
Narrower d-band widths in V$_2$O$_3$ than observed, indicating complex multiplet effects
GWA/U+GWA ineffective in paramagnetic V$_2$O$_3$, with reasons discussed.
Abstract
A GW approximation (GWA) method named U+GWA is proposed, where we can start GWA with more localized wave functions obtained by the local spin-density approximation (LSDA)+U method. Then GWA and U+GWA are applied to MnO, NiO, and VO in antiferromagnetic phase. The band gaps and energy spectra show an excellent agreement with the experimentally observed results and are discussed in details. The calculated width of d-bands of VO is much narrower than that of the observed one which may be a mixture of t multiplet and single electron t level. GWA or U+GWA does not work also in the paramagnetic phase of VO and the reason for thois is clarified. The method of the unique choice of on-site Coulomb interaction is discussed in details. The criterion for whether we should adopt GWA or U+GWA is discussed and is assessed with the help of the off-diagonal…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
