Skip to main content.
 

Abstract's Details

Towards a Unified Theory of X-ray Absorption Spectroscopy?
Abstract IDW:XANES-09 
PresenterFrank  de Groot
Presentation TypeXANES Workshop
Full Author ListF. M. F. de Groot (1)
Affiliations(1) Utrecht University
CategoryInstrumentation/Development
AbstractAt present there is no unified theory of X-ray absorption spectral shapes. Some of the issues that have not been resolved will be highlighted:
  • Oxygen (and C, N, S) K edges can be described from DFT based codes, but several issues still remain unclear. To mention a few: (a) how to treat the core hole potential (Z+1, transition state, relaxed final states), (b) is it necessary to include the matrix element or can one just calculate the DOS in the region of the core hole, (c) are effects from correlations and magnetic order visible in the 3d-band?, (d) why do mixed oxides, for example SrTiO3, often give wrong results with pseudo-potentials?, (e) Will high-resolution oxygen K edge results (if feasible experimentally) necessitate improved theoretical descriptions?
  • Metal K edges are traditionally measured with low resolution due to their long lifetime. High resolution electron and fluorescence detection is increasingly showing better resolved K edges for the 3d metals and L edges of rare earths and 5d metals. Soon the L edges of 4d metals will be measured with improved resolution. Will these better resolved spectra show new features? In case of the HERFD cobalt K edges, the non-local peaks are very clearly visible for low-spin Co3+ oxides, while they are absent for high-spin Co2+ oxides and isolated Co3+ ions. Some issues are: (a) how should one combine DFT and multiplets to describe the complete pre-edge region of these spectra? (b) Will the high-resolution spectra of the main edge structures better reveal the presence of major charge transfer channels? (They should be there but seem ‘invisible’ as yet).
  • Metal L edges have the additional complications due to major multiplet effects and charge transfer channels. Here we are obviously much further from ab-initio-based calculations. However, efforts are made to couple ab-initio DFT codes to multiplet codes, for example by analyzing the corresponding ground states. Another approach is the ab-initio calculation of all parameters that are used in the semi-empirical multiplet codes. Although I think such approach should work, the results obtained as yet are not very satisfactory.
Footnotes 
Funding Acknowledgement