[Ifeffit] negative e0 value

[mauricio.maps]

Hello Bruce, thanks so much for your kind response... Now i'm okay :-)
Cheers
mauricio


2007/10/9, Ravel, Bruce <bravel at bnl.gov>:
>
>
>
> > Dear all,
> > I´m trying to deal with exafs data of a Ni-phosphate compound, using
> > Athena/Artemis. This is a simple structure (I think), with the
> > hexacoordinated ion surrounded  by water molecules on the first shell.
> The
> > question I have concerns the negative value for e0 I´m having.
> > Is this a bad thing?
> > Maybe I´m doing something wrong?
> > I´ve tried some issues, as those proposed on the Shelly Kelly´s
> tutorial,
> > but without success.
> > Can you help me, please?
> > Attached are an example of the Athena and Artemis projects i´m using.
> > Thanks a lot.
> > Mauricio
>
>
> Hi Mauricio,
>
> I think you might not be understanding the purpose of the e0 shift in
> EXAFS analysis.  When we process data, we choose an edge energy which
> sets the zero of wavenumber.  When we make a Feff calculation, it
> somehow chooses its interstitial level, setting its zero of
> wavenumber.  There is no reason to expect that these two wavenumber
> grids will be aligned.  The purpose of the e0 shift in EXAFS analysis
> is to align the k-grid of the data with the k-grid of the theory.
> Just as you cannot expect those k-grids to be aligned a priori, you
> don't know which way they will be misaligned.  It is just as likely
> that you will need a negative value of e0 as a positive one.  As long
> as you have not made any mistakes processing the data (such as a very
> poor choice of the edge energy) or in making the theory (such as
> computing for the K edge when you actually measured the L3 edge), then
> the absolute value of the e0 parameter should be small -- a few volts
> or less.
>
> There is nothing obviously wrong with the e0 shift in the artemis
> project you attached -- it was only about -2 volts.
>
> HTH,
> B
>
> PS.  Sometimes you see fits with more than one e0 parameter.  That
> might be justified if you have some reason to think that one component
> of your fit is electronically dissimilar from the rest of your fit.
> Typically, we use a single e0 parameter which then has the
> interpretation I gave above.
>
>
>
> --
> Bruce Ravel  ----------------------------------- bravel at bnl.gov
>
>
> My homepage:    http://cars9.uchicago.edu/~ravel
> EXAFS software: http://cars9.uchicago.edu/~ravel/software/exafs/
>
>
>
>
> _______________________________________________
> 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/20071016/1ff1ad26/attachment.html>