[Ifeffit] amorphous metal alloys EXAFS spectra to be fitted with Artemis

[Matt Newville]

Hi Piotr,

On Aug 24, 2013 4:19 PM, Piotr J. Bardziński <bardzinski at poczta.fm> wrote:
>
> Dear Sir or Madam,
>
> I got a normalized Fe K-edge EXAFS spectrum for my amorphous 7-component
alloy sample, for which I know its atomic percent composition.
>
> I need to find out the interatomic distances between the Fe and the other
atoms present in the alloy (identify the peaks in R-space) as well as how
many atoms of each element is surrounding the absorbing iron atom
(coordination number).
>
> To meet my goal, I think I need to fit the R-space Fourier transform of
my EXAFS spectrum in a program like Artemis. Unfortunately, I haven`t found
any tutorial discussing how to deal with amorphous sample with unknown
organization of the atoms. When you are about to prepare an atomic model in
Artemis for fitting, you need to build a model with an initial interatomic
distances, atomic coordination number, etc, but all that is obviously
unknown to me. I know nothing about the short range order in my material
and I want to learn it from the experiment itself (note that it  can be and
most probably is different from the one present in crystalline structures).
So, my question is, could somebody tell me how to deal with such system
using the Artemis, what type of initial models for fitting should be
prepared?
>
> All of your help will be greatly appreciated.
>
> Best regards,
> Piotr Bardziński,
>
>
> _______________________________________________
> Ifeffit mailing list
> Ifeffit at millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

If you really "know nothing" about the local structure around iron, you
might be in for a challenging introduction to EXAFS analysis.  You might
actually have some idea about the local structure from the XANES, say for
example if the iron is oxidized or is it still really in a metallic state.
Even if only that, it is a place to start....

For EXAFS with artemis, you'll want to start with guesses of the local
environment.  They don't need to be good guesses, a suitable crystal
structure will do, and you really only need to get the first shell
distances close, because you'll refine these distances.  You will probably
need to do some real work in trying lots of candidate structures, but if
you know it's a metallic glass, you probably actually know a little bit....

Cheers,

--Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20130824/9b0ccdd3/attachment.html>