[Ifeffit] High Correlation betn SO2 and ss

[Carlo Segre]

Hello Shan:

The high correlation between SO2 and the Debye-Waller parameter is not 
surprising at all since they both act to change the intensity of the peaks 
in Chi(R).  With an estimated standard deviation of +-0.3, a change of 0.5 
in the coordination number is quite reasonable too since ESDs are not 
really "error bars" but tend to be lower.  Unless you have other 
information which can help to pin down one of the two parameters, I am 
afraid that you are stuck with this ambiguity.

You have not said what the ss value does when you do not restrain it. 
That might help you to figure out an interpretation.  If it goes way out 
to a non-physical value then you _have_ to restrain it and make sure that 
you report the fact.  In any case the CN is always one of the most 
difficult values to pin down.

Without actually seeing the data, it is difficult to give you more than 
generalities.

Cheers,

Carlo


On Tue, 19 Dec 2006, Shantanu Behera wrote:

> Hi Folks
>
> I have a small doubt. I work on the structural features of Y that are
> segregated to Alumina grain boundaries (very dilute amounts of Yttrium
> in polycrystalline alumina).
>
> In some of my fits, I find a correlation ~0.85 between SO2 and ss. If
> I restrain the ss, from 0.004 to 0.007, the CN changes by ~0.5 (not
> considering +/-0.3). I would appreciate if someone can explain me how
> to interpret the results from a structural point of view.
>
> Warm regards
> Shan
>
>

-- 
Carlo U. Segre -- Professor of Physics
Associate Dean for Special Projects, Graduate College
Illinois Institute of Technology
Voice: 312.567.3498            Fax: 312.567.3494
segre at iit.edu    http://www.iit.edu/~segre    segre at debian.org