[Ifeffit] (no subject)

Scott Calvin SCalvin at slc.edu
Thu Feb 15 06:59:25 CST 2007

Hi Ruwn,

Using a different S02 for each path effectively means that you're 
using a different coordination number for each path. In some cases, 
that may be appropriate--for example, small nanoparticles with 
unknown morphology. It should be noted, however, that S02 often 
correlates highly with sigma2, and a fit that prefers different S02's 
may be getting "mixed up" with sigma2's...although in a case like 
that Artemis/Ifeffit will give high uncertainties to the S02's it finds.

More fundamentally, just because a fit matches the data well, it 
doesn't mean it's "good." The choice of parameters to fit must also 
be physically justifiable. If you can justify each path having a 
different coordination number than in your model, you can try 
different S02's; otherwise, your fit is unconvincing.

--Scott Calvin
Sarah Lawrence College

At 02:31 AM 2/15/2007, you wrote:

>Dear Bruce
>We have used Athena and Artemis software to analyses the EXAFS. When 
>we do the analysis, we used different S2o for each path for fitting 
>of Cu2O data. I succeeded to obtain good fitting by this way. 
>Documents related to percent EXAFS analysis have been mentioned S2o 
>is one value for all paths. What I want to know is our fittings are 
>correct or not. I am looking forward early response.
><?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />
>About 10 years ago we had being using another program made by 
><?xml:namespace prefix = st1 ns = 
>"urn:schemas-microsoft-com:office:smarttags" />Tokyo University in 
>Japan. In this program, the refined parameters also included S2oj 
>and Nj to the each cells
>Ifeffit mailing list
>Ifeffit at millenia.cars.aps.anl.gov

More information about the Ifeffit mailing list