Parameter-free calculation of single-particle electronic excitations inYH3

Abstract

In recent optical experiments for YH 3 , a large band gap of almost 3 eV has been observed. In contrast, calculations within the local density approximation ͑LDA͒ result in a semimetallic band structure with a band overlap near the Fermi level of more than 1 eV. This unusually large discrepancy between the LDA results and experiment has led to suggestions that strong local correlation effects, similar to those found in transition metal oxides, are important in YH 3 . Here we conclude from parameter-free quasi-particle calculations within the GW approximation that YH 3 is essentially a conventional semiconductor. We argue that the experimental results can be fully understood in band structure terms without invoking strong local correlations. The unusually large error in the band structure made by LDA is traced to its poor description of the electronic structure of the hydrogen atom. Our GW results predict a fundamental band gap of only 1 eV and an optical gap of 2.6 eV, the difference being due to vanishing matrix elements for optical transitions at lower energies. Our prediction of a small fundamental gap could be experimentally confirmed by a combination of photoemission and inverse photoemission experiments.

Electronic structure and optical properties of lightweight metal hydrides

“…͑2͒ 46,47 We will benchmark the effects of the pseudopotential approximation on the quasiparticle gaps of hydrides in the next section.…”

Fermi-surface topology of rare-earth dihydrides

“…Van Gelderen 11,12 pointed out, that the failure of LDA can be traced back to an incomplete LDA description of the H ground state. This artifact directly stems from deficiency of the local exchange-correlation potential.…”

“…8,9 Recently GW calculations were able to reproduce the experimental gap and it was claimed, that the insulating trihydride is similar to a conventional semiconductor. [10][11][12] From a practical point of view it seems that the GW calculations only lead to a shift of LDA bands with no significant modifications in terms of bandwidth or band topology. It is at present still unclear, where the limitations of LDA are and if there is a general recipe how to correct LDA results.…”

First-principles study of the optical properties ofMgxTi1−xH2

“…[1][2][3][4][5][6] The metals are reflective, but after hydrogenation become insulators, and hence in most cases transparent. [7][8][9][10] Especially if an alloy with high hydrogen mobility is used and applied in thin films, the optical switching can be fast, reversible, and robust. 11 Recently, metastable thin films composed of various composition ratios of magnesium and titanium have been shown to exhibit remarkable optical properties which could be especially useful for smart solar cell coatings and hydrogen sensor applications.…”