>>
>>I fully agree with you that sample height is the one parameter to refine
>>and that and that this parameter has a rather different angular
>>dependence. However, in agreement with Rory experience we are never able
>>to refine sample height in Bragg-Brentano geometry on the Siemens D5000
>>with alpha1. We found that the correlation between those parameters was
>>always close to 100%. And yes this is with data collected up to 120 deg 2
>>theta. By the way how do you model your asymmetric peak shapes? I allways
>>find that this parameter in the peak profile correlates highly with all
>>the parameters mentioned earlier, including the lattice parameters.
>>
>>
>>Jaap Vente
>>
>I never refine zero and sample shift because I do believe that zero shift
>has to be 0 on modern diffractometers but I do refine sample shift and
>beam penetrating simultaneously because of low absorption of our
>samples. The asymmetric parameters (I am using GSAS or my own
>software with the same asymmetric function) can be refined simultaneously
>only on the final stages when R-factors are low. I also refine all three
>angular corrections simultaneously but only on LaB6 standard after
>diffractometer's alignment to make sure that ZERO shift is 0. It does correlate
>but is refineable.
The attached GIF file shows 2theta(obs)-2theta(clc) difference for
generated data with -0.1 zero shift calculated as sample shift. You could
see the difference. The ZERO shift 0.01 or less leads to difference greater
than 0.002 only at angles over 120 deg 2th. Therefore, to refine simultaneously
zero shift and sample high one would need more or less resolved peaks
in that range.
Dr. Peter Y. Zavalij University Crystallographer
Materials Research Center, SUNY at Binghamton
Ph/Fax:(607)777-4623 E-mail:zavalij@binghamton.edu
http://imr.chem.binghamton.edu/zavalij/zavalij.html