Band Gaps and Effective Oscillator Models for Solid Hydrogen and H\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_2$$\end{document}O Ice at High Pressure

The validity ofYim, W.L.simpleShi, H.effectiveLiang, Y. oscillator modelsHemley, R.J. for extracting band gapTse, J.S. energies from frequency-dependent refractive index data of high pressure ice and solid hydrogenSolid hydrogen was assessed through comparisons of theoretical dielectric response functions obtained from Bethe–Salpeter equationBethe–Salpeter equation (BSE) and band gapsBand gap calculated from the GW method. For model structures of solid hydrogen, the ‘effective’ band gap trend with pressure estimated from experimental refractive index correlate well with theoretical predictions. However, this is not the case for ice at high pressureHigh pressure. The success in the former can be attributed to the lack of variation in the optical absorption spectral profile with pressure due to the molecular-like localized electronic excitations. In comparison, the width of the absorption band in ice broadened continuously with compression. Single oscillator models are not expected to be reliable when the spectral profile changes significantly with increasing pressure as we predict for dense phases of solid hydrogenSolid hydrogen, where the molecules exhibit stronger intermolecular interactions than in the lower pressure structures of solid hydrogen.