[Ifeffit] Multiple data set fit

[Leandro Araujo]

Dear Debora,

we have done some temperature-dependent EXAFS measurements on nanoparticles
(NPs) and analyzed the data through multiple dataset fits, as published in:
L.L. Araujo et al., Physical Review B 78, 094112, 2008.
R. Giulian et al., Journal of Physics: Condensed Matter 21, 155302, 2009.
P. Kluth et al., AIP Conference Proceedings 882, 731, 2007.
I also strongly recommend you take a look at the references therein, like
Anatoly Frenkel et al papers on Pt (and possibly other) NPs. I believe more
people in the mailing have also done some work on this area and hopefully
they will provide you with more references.

As for the second part of the question, you will notice that for the NPs we
have restricted our analysis to first NN shell only, where we are sure
correlations between overlapping single-scattering and multiple-scattering
paths are not affecting the MSRD or sigma2. We have compared the MSRDs
obtained by fitting each temperature individually with the ones obtained by
the multiple dataset fit forcing them to follow the Einstein model and the
results are virtually the same within error bars. For our first shell
neighbours there is a physical bond between the atoms and thus realting the
Einstein frequency to an effective interatomic potential, its force
constants and the mean vibrational frequency is physically plausible, as
demostrated in several references such as PRB 48, 585, 1993 and PRB 70,
174301, 2004 (among others, obviously).

My personal opinion is that for higher shells you can still restrain the
MSRDs to follow the Einstein model, but you should be very careful with the
physical interpretation of the obtained Einstein frequencies. I say that
because we actually discuss this very issue in a paper with exafs
measurements in bulk InP, which has just been accepted for publication in
PRB. If you are interested, I can send you a copy of the manuscript.
My final advice would be looking for articles in bulk materials as well
(there are many of them, I can assure you) because this issue is approached
in some of them, like in Paolo Fornasini's papers, for example.
Just keep in mind that, as it happens often with exafs, different authors
may have different points of view. Hopefully you will get more answers here
in the mailing list.

Hope this helps,
Leandro


On Sat, Apr 18, 2009 at 12:23 AM, Debora Meira <debsmeira at hotmail.com>wrote:

>  Dear all,
> I'd like to know a reference that has multiple set data fit as a function
> of temperature in nanoparticles. Is the relation of Debye Waller factor (s
> 2 = ss2 + sd2) valid for the higher shells? In particular, can I use the
> Einstein model for sd2?
>
> Thanks in advance, Débora
>
> Débora M Meira
> Universidade Federal de São Carlos
> Pós-Graduação em Engenharia Química
> Laboratório de Catálise
> Tel: (16) 92461134 / (16) 34125935
>
>
>
>
> ------------------------------
> Descubra quais produtos Windows Live tem mais a ver com você! Faça o
> teste! <http://www.windowslive.com.br>
>
> _______________________________________________
> Ifeffit mailing list
> Ifeffit at millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20090418/4656f5eb/attachment.html>