[Ifeffit] Basic questions on XANES

[Carlo Segre]

Hi Ashley:

I think you are struggling with one of the most difficult aspects of XAS 
analysis.  There is no single, automatic way of setting these parameters 
although Athena and Ifeffit do a good job of initial guesses.

It is rather trite to say that it is a question of experience but that is 
exactly what it is.  The important thing is consistency throughout a data 
set and this is much more difficult if there are large changes in the 
absorption as a function of sample or measurement conditions.

So, it is difficult to answer you in the abstract.  It would be good for 
you to send a project file to the list and then you would be able to get 
some direct feedback on _your_ data.  I am sure that you will get several 
comments from the experts on this list.

Cheers,

Carlo

On Sat, 14 Jul 2007, Ashley Harvey wrote:

> Dear ifeffit users,
>
> This posting is with quite basic questions on XANES data processing.
> I am a somewhat new XANES experimentalist, in a group that does not
> typically use XANES, so I am lacking some basic knowledge.
>
> For normalization of XANES, my impression is that people generally
> normalize the spectra from ca. 50 eV up to 100 eV or more after the
> edge onset.  Is there a standard acceptable normalization range for K
> edges?  I have noticed that when I change the start of the
> normalization window by even a few eV, this can make substantial
> differences in the XANES when comparing two similar spectra (for
> example, change in peak and pre-edge peak intensities between XANES
> of one sample at two temperatures).  I keep second-guessing my choice
> of normalization range.
>
> Second, I am also looking at the effect of the various ways to set
> E0.  When I start with ifeffit's default (how is this determined?)
> and then proceed to set E0 to the zero-crossing of 2nd derivative,
> I'll get value X.  When I change it to a set fraction of the edge
> step (is this legitimate?) I typically get a lower value, and then
> when I proceed again to set it to the 2nd derivative zero-crossing, I
> get a value different from the first time it was set to the zero-
> crossing.
> For example, the E0 at an Fe K edge:
> ifeffit default = 7128.9 eV
> 2nd derivative zero = 7128.8
> set fraction = 7125.1
> 2nd derivative zero = 7125.5
> The value given by the 2nd derivative zero-crossing is dependent on
> the previous E0 setting.
>
> I am having a hard time to move on to the actual interpretation of
> the data while I am still uncertain about such details.
> Thanks for your patience with some simple questions.
>
> Best,
> Ashley
>
> _______________________________________________
> Ifeffit mailing list
> Ifeffit at millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>

-- 
Carlo U. Segre -- Professor of Physics
Associate Dean for Special Projects, Graduate College
Illinois Institute of Technology
Voice: 312.567.3498            Fax: 312.567.3494
segre at iit.edu    http://www.iit.edu/~segre    segre at debian.org