[Ifeffit] Find white line position

[Hiroshi Oji]

Dear Bruce and Carlo,

Thank you very much for your reply.

Now I understand how Athena find the peak position by the Bruce's answer.

I think some kinds of interpolation is made when Athena find the white 
line position because the peak position that Athena tells seems to be 
more precise than the energy step in the measurement. Is that right?

Actually, I use this function often because it is quite useful for quick 
check for white line position.  But I will be careful when I use it from 
now on, following the Bruce' advice.

I agree with the Carlo's suggestion that it is better to use the 
curve-fitting method to find the white line position.

Best regards,

Hiroshi


> Date: Thu, 2 Aug 2018 09:48:02 -0500 (CDT)
> From: Carlo Segre <segre �� iit.edu>
> To: XAFS Analysis using Ifeffit <ifeffit �� millenia.cars.aps.anl.gov>
> Subject: Re: [Ifeffit] Find white line position
> Message-ID: <alpine.DEB.2.21.1808020946550.9741 �� hydride.segre.home>
> Content-Type: text/plain; charset=US-ASCII; format=flowed
> 
> 
> Hi Hiroshi,
> 
> You are probably better off putting a bit more time into the analysis and 
> simply fitting with an arctangent and a gaussian to get the white line 
> position.
> 
> Carlo
> 
> On Thu, 2 Aug 2018, Bruce Ravel wrote:
> 
>>
>> It's not very sophisticated.
>>
>> Starting with the "flattened" spectrum (see 
>> http://bruceravel.github.io/demeter/documents/Athena/bkg/norm.html#the-flattening-algorithm), 
>> it simply finds the highest point in the spectrum.  Literally just the 
>> ceiling function.
>>
>> The flattened spectrum is used rather than the normalized spectrum because 
>> the curvature of the post-edge (or even the curvature of the pre-edge in 
>> fluorescence data with a energy dispersive detector) could result in other 
>> regions being higher in value than the white line.
>>
>> This is a pretty horrible algorithm.  It is, I think, guaranteed to find an 
>> obviously wrong point for a metal.  Basically it works for spectra with a 
>> white line and it will almost certainly fail for spectra without a white 
>> line.
>>
>> Use with caution!
>>
>> HTH,
>> B
>>
>>
>> On 08/02/2018 01:28 AM, Hiroshi OJI wrote:
>>> Dear all,
>>>
>>> I have a question about one of the useful function, "find white line
>>> position" in Athena.
>>>
>>> Could anyone tell me how Athena find the white line position by this
>>> function? In other words, I would like to know the algorithm to find the
>>> white line position in Athena.
>>>
>>> Thank you very much for your help, in advance.
>>>
>>> Best regards,
>>>
>>> Hiroshi Oji
>>>
>>>
>>
>>
>>
> 
> -- 
> Carlo U. Segre -- Duchossois Leadership Professor of Physics
> Interim Chair, Department of Chemistry
> Director, Center for Synchrotron Radiation Research and Instrumentation
> Illinois Institute of Technology
> Voice: 312.567.3498            Fax: 312.567.3494
> segre �� iit.edu   http://phys.iit.edu/~segre   segre �� debian.org

--