[Ifeffit] Ke

[Carlo Segre]

Hi Pushkar:

There are several ways to work out the best background subtraction besides 
just increasing Rbkg.  Making Rbkg as large as 1.2 or 1.4 is dangerous if 
you have an Fe-O distance or something equally short.  Usually it is 
unwise to make Rbkg more than 1/2 of the distance to the nearest neighbor.

I have sometimes been able to get the same results by cutting down the 
spline range to a value smaller than the highest value of k that you have 
in your data set.  At the same time, it is usually necessary to cut your 
Forward FT range down.

Another, but less effective parameter to change is the k-weight for 
background subtraction.  Usually 2 is a good value but sometimes either 1 
or 3 works better (this has only happened rarely for me).

If nothing else works, I sometimes just leave the datqa the way it is and 
I live with the intensity at low R.

The only way to really understand whether you have done it right is to 
take your model and inspect the results to determine if it makes physical 
sense.  Then you use the same model with data which has been treated a 
different way to see how consistent your results are.

Carlo

On Thu, 16 Oct 2014, pushkar shejwalkar wrote:

> Thanks Carlo for that input. I will keep it in mind. Actually I also have
> this Eo>10 problem and I was following this dicussion closely. Thanks Ke
> for raising the problem. I think my problem lies with the 'Not so good
> modelling of data' as my compounds are Fe(II) based and the local structure
> of which is not known. These are new compounds. Although, I want to make
> sure that the basic steps such as E calibration and background removal is
> done correctly.
> I am still facing problem with the background removal. How would you know
> that the RBKG value if set to 1.2 or 1.4 etc is good and the Background
> removal is ok to use from Athena and take it over to Artemis? are there any
> set of parameters we should check, that if ok we can consider the BKG
> removal is good?
> Thank you everyone in advance for your anticipated help.
> Pushkar
>
> On Thu, Oct 16, 2014 at 6:55 AM, Carlo Segre <segre at iit.edu> wrote:
>
>>
>> Either way is fine but I would suggest taht you not use the calibrate
>> function, just read off the zero crossing and put it into the E0 location.
>> Calibrate actually shifts the data and that is not necessarily what you
>> want to do, particularly if you are comparing spectra..
>>
>> Carlo
>>
>>
>> On Thu, 16 Oct 2014, Pushkar wrote:
>>
>>  Dear Ke
>>> Is there a reason you use first derivative?? I thought if you take second
>>> derivative on Athena and then click find zero and then calibrate that value
>>> is the correct way. I am learning XAFS myself for the first time and I am
>>> novice too. So may be I make no sense but just want to confirm if I am not
>>> doing anything wrong. What's your stake in this??
>>> Pushkar
>>>
>>> Pushkar Shejwalkar
>>> Post-doctoral researcher, JSPS fellow,
>>> Catalysis Research Center
>>> Hokkaido University,
>>> Sapporo
>>> Japan-0010020
>>>
>>>
>>>  On १६ ऑक्टो, २०१४, at १२:२७ म.पू., Ke Yuan <keyuan at umich.edu> wrote:
>>>>
>>>> Hello Sin Yuen,
>>>>
>>>> Thanks for forwarding me that study materials on E0.
>>>>
>>>> I use Athena to find the highest peak on the fist derivative of the
>>>> spectra and pick that value as E0. I think I did not put E0 above this
>>>> value. I will try to put it on the high energy side to see what will happen.
>>>>
>>>> Tks!
>>>> Ke
>>>>
>>>> _______________________________________________
>>>> Ifeffit mailing list
>>>> Ifeffit at millenia.cars.aps.anl.gov
>>>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>>>>
>>>
>>> _______________________________________________
>>> Ifeffit mailing list
>>> Ifeffit at millenia.cars.aps.anl.gov
>>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>>>
>>>
>> --
>> Carlo U. Segre -- Duchossois Leadership Professor of Physics
>> Director, Center for Synchrotron Radiation Research and Instrumentation
>> Illinois Institute of Technology
>> Voice: 312.567.3498            Fax: 312.567.3494
>> segre at iit.edu   http://phys.iit.edu/~segre   segre at debian.org
>> _______________________________________________
>> Ifeffit mailing list
>> Ifeffit at millenia.cars.aps.anl.gov
>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>>
>>
>
>
>

-- 
Carlo U. Segre -- Duchossois Leadership Professor of Physics
Director, Center for Synchrotron Radiation Research and Instrumentation
Illinois Institute of Technology
Voice: 312.567.3498            Fax: 312.567.3494
segre at iit.edu   http://phys.iit.edu/~segre   segre at debian.org