Dear Rana, sorry for really late response, but I was cut off from the internet for a while... if I understand correctly you created a model of the structure and you fit it to the first shell (oxygen?), what is the expected difference between different path? with my humble experience you can try to clarify following problems: - is your model really correct and cover all possible scenarios? - are you sure that in your sample are no other (unwanted) phases? with Mn, Fe it can happen... - have you tried to increase the number of variables? (to free the fit from boundaries) I am only thinking about it (with an examples at Fe site): you have 4 positions for Fe and if there is a mixing between Fe and Mn sites additional 3, it means maximally 7 sets of (e0, delr, and ss2 - N you can probably estimate, calculate). that means you have more variables than independent points. you should try, in the first approximation, to free the delr parameter and keep others the same. I havent seen good example of fitting such samples, but if you start from different points, and during fitting you will minimize no. of variables with the same result (with the same fit) should be ok. good luck darek -----Wiadomość oryginalna----- Od: ifeffit-bounces@millenia.cars.aps.anl.gov w imieniu Jatinkumar Rana Wysłano: Śr 2010-06-16 12:57 Do: ifeffit@millenia.cars.aps.anl.gov Temat: Re: [Ifeffit] Ifeffit Digest, Vol 88, Issue 12 Dear Darek, Sorry for the late response. To fit the "first shell" the No.of independent points and No.of variables which i use are as follows: I use deltaK=11, deltaR = 1.1 so No.of independent points are 8 and No.of variables are 4 (i.e. N, e0, delr, and ss2) I set s02=0.9. Best regards, Rana On 14.06.2010 19:00, ifeffit-request@millenia.cars.aps.anl.gov wrote:

Send Ifeffit mailing list submissions to ifeffit@millenia.cars.aps.anl.gov

To subscribe or unsubscribe via the World Wide Web, visit http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit or, via email, send a message with subject or body 'help' to ifeffit-request@millenia.cars.aps.anl.gov

You can reach the person managing the list at ifeffit-owner@millenia.cars.aps.anl.gov

When replying, please edit your Subject line so it is more specific than "Re: Contents of Ifeffit digest..."

Today's Topics:

1. XAFS fitting with multiple atomic sites of absorbing atom (Jatinkumar Rana) 2. ODP: XAFS fitting with multiple atomic sites of absorbing atom (Zajac, Dariusz A.)

----------------------------------------------------------------------

Message: 1 Date: Mon, 14 Jun 2010 12:00:27 +0200 From: Jatinkumar Rana To: Ifeffit users Subject: [Ifeffit] XAFS fitting with multiple atomic sites of absorbing atom Message-ID:<4C15FDBB.1050203@helmholtz-berlin.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Dear users,

I have measured XAFS spectra on Mn and Fe edge on my material. The material contains 3 different atomic positions for Mn and 4 different atomic positions for Fe.

Initially, I tried fitting Fe-edge/Mn-edge spectra using indiviual Fe/Mn atomic position one-by-one (i.e. first take one Fe/Mn position, calculate Feff paths, consider certain paths to fit specific R-range, then take another Fe/Mn position and do the same) but the fit was not good visually and in terms of goodness of fit parameters.

Then, i tried fitting Fe-edge/Mn-edge spectra, taking into account all Fe/Mn atomic position together i.e. I separately calulated Feff paths for each Fe/Mn atomic position and summed up related paths from each Feff calculation for fitting a specific R-range. (The assumption being, the experimental Fe-edge/Mn-edge XAFS spectra has varying contribution from all Fe/Mn atomic position in crystal structure).

Anybody in the group could comment about my approach to fit such problem ?

Is there any specific strategy to fit such problem ?

I would be thankful if you could suggest some literature available on such problems.

Thanks in advance.

Best regards, Rana

------------------------------

Message: 2 Date: Mon, 14 Jun 2010 13:34:18 +0200 From: "Zajac, Dariusz A." To: "XAFS Analysis using Ifeffit" Subject: [Ifeffit] ODP: XAFS fitting with multiple atomic sites of absorbing atom Message-ID: <35E004AD6290A7438FCA34BBF325F41601206741@ADXV2.win.desy.de> Content-Type: text/plain; charset="iso-8859-2"

Dear Rana, could you specify how many independent data points and variables you have and you use for fitting? thnx darek

-----Wiadomo?? oryginalna----- Od: ifeffit-bounces@millenia.cars.aps.anl.gov w imieniu Jatinkumar Rana Wys?ano: Pn 2010-06-14 12:00 Do: Ifeffit users Temat: [Ifeffit] XAFS fitting with multiple atomic sites of absorbing atom

Dear users,

I have measured XAFS spectra on Mn and Fe edge on my material. The material contains 3 different atomic positions for Mn and 4 different atomic positions for Fe.

Initially, I tried fitting Fe-edge/Mn-edge spectra using indiviual Fe/Mn atomic position one-by-one (i.e. first take one Fe/Mn position, calculate Feff paths, consider certain paths to fit specific R-range, then take another Fe/Mn position and do the same) but the fit was not good visually and in terms of goodness of fit parameters.

Then, i tried fitting Fe-edge/Mn-edge spectra, taking into account all Fe/Mn atomic position together i.e. I separately calulated Feff paths for each Fe/Mn atomic position and summed up related paths from each Feff calculation for fitting a specific R-range. (The assumption being, the experimental Fe-edge/Mn-edge XAFS spectra has varying contribution from all Fe/Mn atomic position in crystal structure).

Anybody in the group could comment about my approach to fit such problem ?

Is there any specific strategy to fit such problem ?

I would be thankful if you could suggest some literature available on such problems.

Thanks in advance.

Best regards, Rana _______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

------------------------------

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

End of Ifeffit Digest, Vol 88, Issue 12 ***************************************

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit