[Ifeffit] ATOM and GDS

[Carter Abney]

Dr Li,

If I'm understanding the questions properly, there should be some fairly straightforward fixes.  Simplest first:  in GDS, there should be a button on the right side that says "Add GDS."  Clicking on it will give you additional variables (or definitions, restraints, etc.) in GDS.  Otherwise, you could right-click on the numbers to the left of the "type" column.  You'll get options like "insert blank line above (or below) current row."  This is convenient for keeping your variables grouped intelligently while fitting.  To the best of my knowledge, there is no maximum number of lines in GDS, though your number of variables should be limited by 2/3 the number of independent points available from your data.  Additionally, you're only allowed a maximum of 10 restraints at any time in a fit.

Regarding Atoms, it sounds like you're trying to manually add elements to the coordination sphere of your absorbing element.  There should be a button, "add a site" which will allow you to add more than 8 elements.  That said, I've never prepared my scattering pathways in this fashion (this probably says more about my inexperience than anything else).  Most of the time, I call Atoms from Artemis, input a cif file, run atoms, run feff, and have a list of scattering pathways afterwards.  If your system isn't crystalline, you can run several Feff calculations with different cif files which have the correct absorbing atom and an appropriate scattering atom.  (You have to add each of these Feff calculations in Artemis individually.)  Then, use the generated scattering paths to model your data.  

Also, Atoms allows a person to modify the input file (I believe they're referred to as "cards") as a text document from within the Atoms program.  So, if I have a model with no cif file, I can run atoms with any other cif file to generate some sort of card for feff, then I can edit the card appropriately by copy/pasting the Cartesian coordinates and defining the appropriate ipot value for each element.  If you have the card already generated from another cif file, it's fairly straightforward to edit it to represent your model.

- Carter


Message: 2
Date: Mon, 14 Jul 2014 16:25:40 -0400
From: Dien.Li at srs.gov
To: ifeffit at millenia.cars.aps.anl.gov
Subject: [Ifeffit] ATOM and GDS
Message-ID:
	<OF5F2EDED2.B07C6B62-ON85257D15.006FF11D-85257D15.007036E1 at srs.gov>
Content-Type: text/plain; charset="us-ascii"

HI, Bruce and Matt

For the Demeter version I am using, I found that I could just enter 8 atoms in ATOM, and define 12 variables in GDS. Could I enter more atoms in ATOMS, and define more variables in GDS? and How? Thanks.


Dien Li, Ph.D.
Environmental Sciences and Biotechnology Savannah River National Laboratory Aiken, South Carolina, 29808
Tel: 803-725-7520
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20140714/c3a6fc91/attachment-0001.htm>