[Ifeffit] EXAFS signal from dopant atoms

[Carlo Segre]

Hi Eugenio:

This only partly depends on the actual concentration of your dopant atom. 
Just becuaseit is a low concentration does not mean the signal is 
necessarily in the noise.  If you set up your experiment correctly and 
collect sufficient data you can get quite a good signal from even a low 
concentration.

What you will see in the EXAFS, if the structure is well-ordered, is the 
fact that the C atom is surrounded by a bunch of A atoms at whatever 
distance will be typical for a dopant atom of this kind.  You will also 
see further "shells" which will likely be mostly B atoms.

As you increase the concentration of C atoms, you might not see too much 
difference in the near neighbor A atom environment but you will be seeing 
a second shell with more and more C atoms replacing the B atoms.

There are many more nuances to this but it is not a great idea to put a 
lot of presumptions into your structural model since you might bias the 
results of the analysis.

Good luck

Carlo

On Thu, 17 Sep 2015, Eugenio Paris wrote:

> Dear all,
>
> I have a very general question about EXAFS analysis.
> Let's take as example a binary system AB in which i am substituting the
> atom B with another atom C so that we have AB(1-x)C(x). Let us imagine that
> we know the structure of binaries AB and AC with Rab and Rac atomic
> distances and we know that in AB(1-x)C(x) the substitutional site will be
> 100% the B-site.
>
> If I measure the EXAFS spectrum of atom C as a function of "x" what can I
> expect for the EXAFS oscillations? My idea is that when "x" is very small
> the EXAFS will be highly noisy and damped, while as "x" increases the
> oscillations will become more evident above the noise level.
>
> How to model this situation to get information on C-A bond? In Athena I can
> start with the .cif file of AC structure with full occupancy and obtain the
> distance and the DWF for the A-C bond but.. how to treat the
> signal-to-noise ratio? I mean, as "x" decreases I think there must be a
> leaking out of the EXAFS signal, should this effect be introduced in S0^2?
> Or can it affect the effective number of neighbors?
>
> Hope I explained myself,
> thank you!
>
> Eugenio Paris
> PhD student, University of Rome (Italy)
>

-- 
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