Extending first principles spectroscopy to disordered materials : a study on amorphous and crystalline aluminas

Advancing the next generation of materials for solid-state devices requires an understanding of their underlying electronic structure; yet the disordered nature of many materials hinders progress in creating an atomic-level model. Although X-ray diffraction (XRD) is used to match crystalline models with experiment, XRD fails to characterize amorphous structures due to their lack of periodicity. However, nuclear magnetic resonance (NMR) and X-ray absorption spectroscopy (XAS) are two locally sensitive techniques that do not rely on periodicity. To compare amorphous models to experiment, it is then necessary to extend first principles calculations of NMR and XAS to amorphous materials. In this thesis, I apply first principles methods for calculating NMR and XAS spectra, to both crystalline and amorphous phases of alumina (Al₂O₃). First, I analyze the Al K-edge XAS for crystalline α- and γ-Al₂O₃ at finite temperature. Phase transitions from Al s to mixed s-p states are symmetry forbidden at 0 K, but incorporating the effects of temperature reproduces these transitions. Applying this method to γ-Al₂O₃ reproduces a peak at 1568 eV in the XAS which is not observed in the first principles calculations at 0 K. I then address the question of how to calculate these spectra for an amorphous phase. By using representative configurations of am-Al₂O₃ from ab initio molecular dynamics (AIMD), I calculate averaged NMR and XAS spectra for am-Al₂O₃. I show that both the experimental ²⁷Al isotropic chemical shifts and quadrupolar lineshapes are successfully reproduced using this method. Further, by calculating XAS spectra for each individual Al atom in this model, I am able to deconvolute the experimental XAS into site-specific spectra from four-, five-, and six-fold coordinated Al. Finally, I examine how the structure of am-Al₂O₃ is affected by hydrogen fluoride incorporation in Li-ion batteries, by calculating ²⁷Al NMR on protonated and fluorinated amorphous aluminas. These results are in line with experimental ²⁷Al NMR evidence for a pathway of HF attack involving the formation of AlO(6−x)Fx species during battery cycling.