Hi Fuxiang, Further to Bruce's note, I looked at your file. Your model splits the nearest oxygens by only 0.01A. Is that the magnitude of splitting you anticipate? Are you familiar with the resolution criteria for EXAFS? (k-range [deltak]) * (R-resolution [deltaR]) ~ pi/2 for identical near neighbours This is also discussed in the archives, e.g.: http://millenia.cars.aps.anl.gov/pipermail/ifeffit/2007-November/003414.html Looking at your chi(k), it seems usable, charitably, up to about 8 A-1, not to 12 A-1 as your window suggests you have used. Claims of being able to resolve small splittings with this data will be viewed with skepticism. -R. On 4/18/2017 11:34 AM, Bruce Ravel wrote:
On 04/18/2017 11:10 AM, Fuxiang Zhang wrote:
Hi, all, I am in studying the EXAFS of perovskite and clearly see the split of the first peak after ion irradiation. It should come from a symmetry change, so I use a tetragonal model (instead the original cubic) which contains two independent O sites. However, after running Atoms and Feff, all paths are contributed by only one O atom and nothing related to another Oxygen atoms (see attached file). As a result it may be difficult to get a good fit of the first split Ta-O peaks. I am not sure if it is true or something wrong. Thank you very much
A fairly common question here on the mailing list. This is a *feature* of Artemis, not a bug.
http://bruceravel.github.io/demeter/documents/Artemis/extended/fuzzy.html
If you want to /model/ a split in distance, you can drag that path into the path list twice and adjust the delta R and amplitude parameters accordingly.
B