Caroline,
This is a very thorny issue. In general, displacement [nee thermal]
parameters correlate highly with occupancy. With x-ray data, while
refinement tricks may allow both to be refined, there is not typically
enough data to allow both to be determined. With neutrons, the problem
is not as severe but is still present. With neutrons one can sometimes
refine both Uiso and the occupancy, but even then sometimes one needs to
fix or group the displacement parameters. It helps to have a feel for
what reasonable Uiso values are, but this is a function of the material
type, temperature and the atomic mass.
My recommendation to perform a sensitivity analysis at the end of the
refinement to see if the occupancy values are significant. I typically
take the final Uiso and fix it at the final value divided by 2 and
multiplied by two and refine the occupancy each way. If, with both
extreme Uiso values, I see occupancy values similar to my refined value,
then I am willing to accept that the occupancy value is likely to be
"real" rather than an artifact of the fitting procedure.
Brian
"C.A.Kirk" wrote:
>
> Dear All,
> I have a question regarding the correlation of thermal parameters
> and occupancies. If anyone could give some tips on achieving a
> satisfactory conclusion/minimum during a refinement, it would be
> greatly appreciated.
>
> At the moment I refine the atomic positions then the Us until a
> minimum is reached. If a U is particularly high, for instance, then I
> check the occupancy of this site; sometimes partially occupied
> sites are expected in the phases we study.
>
> However, as the U is fixed at the refined value then this willeffect
> the occupancy. As I understand it the refined U is for the
> occupancy fixed during the refinement. Sometimes a change in the
> occupancy is observed on refining. I then will alternate between
> refining the occupancy and U until a minimum is reached.
>
> Is a better strategy to fix the U at a value for the same atom type in
> the phase and look at the effect this has on the occupancy?
>
> Explainations, suggestions, tips, further reading welcome!
>
> Caroline
>
> Caroline Kirk
> Solid State Research Group
> Department of Engineering Materials
> University of Sheffield
> Sir Robert Hadfield Building
> Mappin Street
> Sheffield
> S1 3JD
>
> Telephone: 0114 222 6013
> E-mail: [EMAIL PROTECTED]
--
********************************************************************
Brian H. Toby, Ph.D. Leader, Crystallography Team
[EMAIL PROTECTED] NIST Center for Neutron Research, Stop 8562
voice: 301-975-4297 National Institute of Standards & Technology
FAX: 301-921-9847 Gaithersburg, MD 20899-8562
http://www.ncnr.nist.gov/xtal
********************************************************************
