Re: [Ifeffit] Problems with feff and crystallographic data (Now with bug report)
Hi Hashem,
Atoms doesn't handle the occupancy data in the CIF file. See this page for
more info:
http://cars9.uchicago.edu/iffwiki/FAQ/FeffitModeling#head-b273ecfd19542ec82e...
I assume you see a message like this one that I see:
This warning indicates a problem in the Atoms list. For some
reason you have atoms that are separated by than 0.93 Angstroms
(or 1.75 Rydberg).
If you look on the Atoms tab in Artemis when you open the CIF file, you
should see that the co-ordinates
for Mg and Al are all 0.00000 ; i.e., all Atoms sees is that they are in the
same place.
Bruce, here's two bug reports:
1. The warning message above is missing the word "less", as in "separated by
_less_ than 0.93 Angstroms"
2. It's an extensive message but doesn't mention this kind of problem.
If you want to drive yourself mad with interface stuff, you could catch and
highlight these errors on input and disable the controls they affect until
the input is corrected. A tricky part of that is making sure the user knows
what the highlighting means and has a clue about how to fix the error
without - say - throwing up a talking paperclip. A sort of "heavy tooltip"
("lead balloon"?) could be shown - placed so as not to obscure the relevant
controls - with a brief message, a way to get to the more help, and
dismissal button. An automatic fade out might be nice, not sure. So long as
the bad input exists, there should be a way to get the message back, like
mousing over the highlighted input or the disabled controls. I don't know of
any cross-platform toolkit that makes that kind of thing easy.
Cheers,
Greg
On Thu, Dec 18, 2008 at 10:04 PM, Hashem Stietiya
Hello!
I was using Artemis to run feff and generate paths from hydrotalcite. However, it failed to run and an error message indicated that there is something wrong with the crystallographic data. I used the crystallographic data of hydrotalcite from different sources, but encountered the same problem. Can you please help identify the problem and run feff?
Please find attached the crystallographic data for hydrotalcite.
I appreciate your help.
Regards,
Hashem Stietiya
Louisiana State University
_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Hi Hashem: As Gregory mentions below. The problem is partial occupancies. In diffraction, it is OK to put two different atom species in the same (or close) crystallographic position. This makes no sense in analyzing EXAFS because you cannot calculate scattering from a fractional atom. For EXAFS, you need to decide what the specific local environment of your target atom will be. If you have two different kinds of environments, then you have to do two (or many more) separate FEFF calculations and then fit to both of them simultaneously to get the effect of partial occupancy. Cheers, Carlo On Fri, 19 Dec 2008, Gregory Merchan wrote:
Hi Hashem,
Atoms doesn't handle the occupancy data in the CIF file. See this page for more info:
http://cars9.uchicago.edu/iffwiki/FAQ/FeffitModeling#head-b273ecfd19542ec82e...
I assume you see a message like this one that I see:
This warning indicates a problem in the Atoms list. For some reason you have atoms that are separated by than 0.93 Angstroms (or 1.75 Rydberg).
If you look on the Atoms tab in Artemis when you open the CIF file, you should see that the co-ordinates for Mg and Al are all 0.00000 ; i.e., all Atoms sees is that they are in the same place.
On Thu, Dec 18, 2008 at 10:04 PM, Hashem Stietiya
wrote: Hello!
I was using Artemis to run feff and generate paths from hydrotalcite. However, it failed to run and an error message indicated that there is something wrong with the crystallographic data. I used the crystallographic data of hydrotalcite from different sources, but encountered the same problem. Can you please help identify the problem and run feff?
Please find attached the crystallographic data for hydrotalcite.
I appreciate your help.
Regards,
Hashem Stietiya
Louisiana State University
_______________________________________________ Ifeffit mailing list Ifeffit@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@iit.edu http://www.iit.edu/~segre segre@debian.org
participants (2)
-
Carlo Segre
-
Gregory Merchan