I find that the main effect of the multi-electron excitations is to make a pair of bumps in an unfortunate region of the spectrum, which need to be subtracted out before EXAFS analysis can begin. The positions and amplitudes of these bumps are known for Ce3+ and Ce4+, so if you start with post-edge normalized (XANES-type) data, you can subtract these out and proceed with EXAFS analysis per normal. mam On 7/14/2016 11:05 AM, Neil M Schweitzer wrote:

Thanks for the responses. When I had originally tried fitting the CeO2 data, the fit was awful, but I think it was because the Eo was really far off. I set it close to where it should be and was able to find a common feature in Io to calibrate the rest of the samples to.

That being said, fitting CeO2 is still quite difficult. I found a reference online (EXAFS analysis of the L3 edge of Ce in CeO2: effects of multielectron

excitations and final-state mixed valence: J. Synchrotron Rad. (1999). 6, 34-42) that, from my understanding, essentially suggests duplicating all the paths and using a second delta_Eo value for the duplicated paths while also weighing the SO2 of each set of paths based on the degeneracy of the excitations. This does tend to make the fit better (as one would expect from doubling the number of paths!), but does anyone know if this is a generally accepted correction for fitting ceria materials? Does anyone have a different reference that might help me? Thanks for all the help!

Neil

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit