Re: [sdpd] (unknown)

[leosol...@chat.ru]

>I▓m working with excellent program FULLPROF refining XRPD output, but 
>sometimes I have some difficulties refining temperature factors (Biso), 
>and so I must use references datas without refining them.
>
>		Luca Mollica

Dear Luca,

Problems in refining temperature factors may result from correlations
between them if the powder data are not sufficiently informative for refinement
or if the profile fit is not good enough. In this case constrains should be set
on temperature factor parameters and they should be refined only at the very
last stage of Rietveld refinement when the rest structural and profile
parameters are properly adjusted.
Temperature factors may also correlate with background and profile parameters,
then these parameters should be either constrained or refined separately.
The finite sample thickness effect in Bragg-Brentano geometry and peak
overlapping in the high angle area may cause unrealistically big values of Biso 
which should be considered carefully.
In some cases, when the powder data quality does not correspond to the structure
complexity, temperature factors of light atoms can't be controlled at all and 
one has to set them manually. 

Sincerely,
Leonid Solovyov

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/