Improved R-space resolution of EXAFS spectra using combined regularization methods and nonlinear least-squares fitting

A number of methods have proven to be useful for extended x-ray-absorption fine-structure data analysis of disordered systems. Regularization methods, which describe the radial distribution function nonparametrically, have the advantages of eliminating bias caused by modeling the distribution with a specific functional form, properly including spherical wave effects, and allowing the imposition of physical constraints in a systematic way. In this work we find empirically that the regularization method reliably resolves subshells that are unresolved by Fourier transforms and provide unbiased starting models for nonlinear least-squares fitting. The method is demonstrated with experimental and simulated data.