Germanium: the good, the bad, and the ugly, how d-orbitals can ruin materials or create new opportunities

Abstract

We examine from a high-level perspective the chemical basis that has limited use of Ge in modern technology to niche applications. From the measured optical properties [n,k] of c-Ge, GeTe, and Ge₂Sb₂Te₅ (GST-225), we examine their sum-rule and superconvergence (SC) properties (to effectively measure the total N_e/atom) and thereby demonstrate the significance of d-orbitals. We find that while the crystalline form (fcc) of GST-225 only makes use of s- and p-orbitals, the amorphous form has a greater overall N_e/atom ratio, which can only occur if d-orbitals also participate. Similar results are found for GeTe. We interpret these N_e/atom densities by a simple valence model, which places rather strict limitations on the type and concentrations of different local atomic environments. Moreover, we show that since the SC is closely related to electric susceptibility, quantitative changes in the SC between the crystalline and amorphous forms imply removal (suppression) of Te’s lone pair orbitals (existing in the crystalline form) by the d-orbitals (existing in the amorphous form). With this, it now appears that we have a more comprehensive understanding of how d-orbitals play important roles in the switching behaviors in both optical and electronic devices. We propose that this understanding can eventually be extended to the ultimate “single-atom” man-made electronic devices.

Acknowledgements

D.V. Tsu dedicates this paper to his father R. Tsu, who has always been a source of amazement and clarity illuminating the complexities of physics. He is also inspired by the deep love of chemical bonding of Stan Ovshinsky in inspiring the development of new materials.