Toward Quantitative Calculation and Analysis of X-Ray Absorption Near Edge Spectra

Toward Quantitative Calculation and Analysis of X-Ray Absorption Near Edge Spectra Joshua Kas Chair of the Supervisory Committee: Professor John Rehr Physics X-ray absorption spectroscopy (XAS) has been used extensively to determine local electronic and structural properties. While quantitative analysis of extended x-ray absorption fine structure (EXAFS) has been available for over two decades, theories of x-ray absorption near edge structure (XANES) remain qualitative at best. In addition, analysis techniques for EXAFS can be unstable, and quantitative XANES analysis codes are not widely available. Here we present several developments in the theory and analysis of x-ray absorption that address these shortcomings. Our goal is to provide a set of theoretical and analysis tools which enable quantitative analysis of XANES spectra for a wide variety of materials, with a focus on biological molecules. Bayesian-Turchin analysis techniques are applied to XANES as well as EXAFS. Theoretical developments include an efficient many-pole model of inelastic losses, and full potential effects in real space multiple scattering XANES calculations. These theoretical developments are implemented within the FEFF multiple scattering code for calculating x-ray absorption and related spectroscopes. The Bayesian fitting method is implemented by extending the FEFFIT EXAFS analysis code. These developments significantly improve the theory and analysis of XAS. Our many-pole model agrees well with first principles calculations of the quasi-particle self-energy, as well as with experimental XANES spectra. Our full potential calculations of the density of states give improved agreement with those of other electronic structure codes. Our Bayes-Turchin analysis methods are compared with standard analysis techniques and are shown to stabilize the fitting procedure. Results are presented for a variety of materials including mono-atomic solids, perovskites, simple molecules, and metalloproteins.