Hi Dave,

I'm cc'ing this to the Ifeffit mailing list, as it may be of interest to others as well.

At 03:14 PM 3/22/2006, you wrote:
I recently read your Ifeffit Wiki page concerning using Artemis to model doped crystals and as a new user to Artemis, I have two questions to ask of you. 
 
1.  When you define IronPercent in the second method you have it defined as 1 - MolyPercent and you have MolyPercent defined as 20.  Is this a simple typo, should the definition actually be 100 - MolyPercent (or MolyPercent = 0.20, for that matter).

It's a "typo" (I love that euphemism for "mistake"). It should be MolyPercent = 0.20. Thanks for finding it--I've fixed it on the Wiki.

 
2.  I know the S02 term needs to be in the range of 0.7 to 1, but if using a material (say Moly) that is doped with a small percentage of the Fe, is it possible to achieve an S02 term in the appropriate range for both the Moly and the Fe.  It seems that if the amp term is 1 (for simplicity's sake) then S02 for Moly would be 0.8 and S02 for Fe would be 0.2.  Or is the amp term more important for the S02 in these types of models?

I'm not certain I understand what you're asking, but I'll try to give you an answer--and let me know if it's not what you were looking for!

The "S02" we're putting into Ifeffit is not the real S02 with these methods. We're using a trick: in the EXAFS equations,  N and S02 have exactly the same effect for a given path...e.g. doubling coordination number or doubling S02 has exactly the same effect on the spectrum. Somewhere way way back in the Ifeffit lineage, it was decided that N would not be a parameter that could be varied, and S02 could. (There are actually very good reasons for this, but I don't think I need to go into them right now.) So if we want to fit coordination number, we actually vary the "S02" parameter in Ifeffit.

So here's a typical way of making it look nice. Suppose path 1 uses a FEFF calculation with a coordination number of 6, but we think the coordination number might be a bit different. We could do the following:

Guess N1 = 6
Set Amp = 0.90

path(index = 1,
                          file     = feff0001.dat,
                          S02      = Amp*N1/6)

If we have more than one path, we could also guess Amp...with only one path, it won't work, because there's no way for Ifeffit (or anyone for that matter) to know the difference between Amp and N1.

So the upshot of all this is that "S02" (the ifeffit path parameter) could very well come out to be, say 0.60, or even 1.20, if the coordination number is very different from 6. But if we defined things in the way I've suggested, Amp would be the "real" S02, and should hopefully be in the 0.70 to 1.0 range.

Doping is handled similarly.

Hope that helped.

--Scott Calvin
Sarah Lawrence College

 
Thank you very much.
 
Dave
 
 
David C. Schmitter
Dept. of Physics and Astronomy
University of Nebraska Lincoln
dschmitt@unlserve.unl.edu
(402) 472-5094