[Ifeffit] 4+2 model for distortion

Carlo Segre segre at iit.edu
Thu May 17 03:11:34 CDT 2018

```Hi Abdul:

Here is what I do under these circumstances, when the distances are very
close.

I first select one path only, set the value of N to the total number of
neighbors (in your case 6) and then fir using only this one path.  In your
case, if you choose, F1, the initial value of N should be 4 and you change
it to 6.

I then look at the outcome of the fit, in particular, the value of
sigma-squared.  If this is physically reasonable, then I make the
determination that it may be impossible to separate the two paths.  So
what is a physical value?  Gnerally, if this is a room temperature
measurement, I would expect a value of between 0.002 and 0.005 for a first
shell Ni-F bond.  If the value of sigma squared is higher than that, it
could be legitimate to include the second path.  In this case, I owul dset
the value of N back to the original number (in your case 4) and fit again
but constrain the sigma squareds of the two paths to be the same
initially.  Again, if the result is significantly better in terms of chi
squared and the Real/Imaginary part of the data/fit and the sigma squared
is physically reasonable, then I might decide to keep the two path fit.
Remember that you wil have more free parameters and as such the
chi-squared will usually be smaller for the two shell fit.

The bottom line is that you need to decide what you are trying to extract
from the data.  Are you measuring the asymmetry of the octahedron as a
function of temparature or some other parameter?  If so, then perhaps
using two paths is the right solution.  basically, you need to try
difrernt fits and then look at the results to make sure you have a good
reason to choose one over the other.  There is no magic bullet, I am
afraid.

Carlo

On Wed, 16 May 2018, Abdul Ahad wrote:

> Hi Prof. Segre
>
> I have used .cif file for the K2NiF4 type structure.
> In this structure the first shell and first single path shows 6
> coordination number. and the path is Ni to F1 with 2 leg. My concern is to
> use 4+2 model. Do i change the CN number or i have to add one more site. I
> cant visualize the situation. Becuase here the F have 2 sites i.e F1 and
> F2. And the path is only for the F1. While the diffrence between the bonds
> Ni-F1 and Ni-F2 is about 0.1 A. which lies under the resulution limit of my
> experiment via delta r and delta k relation.​
>
>
>
> *--------------------*
> Junior Research Fellow
> *Dept of Physics*
> *Aligarh Muslim University*
> *Aligarh 202002 (U.P.), India*
>
> *http://www.researcherid.com/rid/G-8198-2018
> <http://www.researcherid.com/rid/G-8198-2018>*
>
> On Wed, May 16, 2018 at 10:30 PM, <ifeffit-request at millenia.cars.aps.anl.gov
>> wrote:
>
>> Send Ifeffit mailing list submissions to
>>         ifeffit at 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 at millenia.cars.aps.anl.gov
>>
>> You can reach the person managing the list at
>>         ifeffit-owner at millenia.cars.aps.anl.gov
>>
>> than "Re: Contents of Ifeffit digest..."
>>
>>
>> Today's Topics:
>>
>>    1. Re: 4+2 model for distortion (Carlo Segre)
>>
>>
>> ----------------------------------------------------------------------
>>
>> Message: 1
>> Date: Tue, 15 May 2018 16:00:08 -0500 (CDT)
>> From: Carlo Segre <segre at iit.edu>
>> To: XAFS Analysis using Ifeffit <ifeffit at millenia.cars.aps.anl.gov>
>> Subject: Re: [Ifeffit] 4+2 model for distortion
>> Message-ID: <alpine.DEB.2.21.1805151558480.17852 at hydride.segre.home>
>> Content-Type: text/plain; charset=US-ASCII; format=flowed
>>
>>
>> Hi Abdul:
>>
>> Are you strating with a cif file for a particular structure or are you
>> simply working with a quick first shell theory?
>>
>> Carlo
>>
>>
>> On Wed, 9 May 2018, Abdul Ahad wrote:
>>
>>> hi all
>>> How to include this model in atom that is 4+2 or 2+4 for 2short or 4 long
>>> bond length of octahedra
>>> thanks
>>>
>>>
>>> *--------------------*
>>> Junior Research Fellow
>>> *Dept of Physics*
>>> *Aligarh Muslim University*
>>> *Aligarh 202002 (U.P.), India*
>>>
>>> *http://www.researcherid.com/rid/G-8198-2018
>>> <http://www.researcherid.com/rid/G-8198-2018>*
>>>
>>
>> --
>> 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 at iit.edu   http://phys.iit.edu/~segre   segre at debian.org
>>
>>
>> ------------------------------
>>
>> Subject: Digest Footer
>>
>> _______________________________________________
>> Ifeffit mailing list
>> Ifeffit at millenia.cars.aps.anl.gov
>> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit
>>
>>
>> ------------------------------
>>
>> End of Ifeffit Digest, Vol 183, Issue 14
>> ****************************************
>>
>

--
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 at iit.edu   http://phys.iit.edu/~segre   segre at debian.org
```