> If
>somebody suggested to combine Rietveld with MC ray-tracing, this may be
>new, but I don't think we are there yet (not, perhaps, we need to).
>
Paolo,
this job was done by Joerg Bergmann about 5 years ago, and an actual
program exists and is available. The program GEOMET inside the BGMN package
simulates device profiles by MC raytracing depending on a lot of geometric
input parameters like sample dimensions (for irradiating and penetrating
the sample),linear attenuation coefficient of the sample, formula
describing the ADS divergence,a misfit angle between line focus and slit
edge,focus dimensions and all the known (since WILSON) divergence
influences from the slits and collimators. Besides Bragg-Brentano you can
choose also flat sample transmission and capillary geometry. So, the
instrumental profile is predefined and not to refine within Rietveld
calculation. This was the aim of Joergs approach. We do not believe that a
refinement of some of these parameters can be done together with real
structure parameters without correlation problems.
And you are rigth, all the models used contain a simplification, e.g. a box
shape for intensity distribution inside the focus and so on. May be, this
is equal to Armels "empirism" problem?
But, in my experience,the input of primary (fundamental) geometric
parameters is more problematic. Who knows
- the effective (not nominal) divergence of the collimators depending from
missetting by the manufacturer or the operator
- the angle between line focus and slit edges (always 0 degree?)
- the effective width of line focus and the intensity distribution emitted
(exact box shape versus double peak and feets)
- the effective penetration depth/linear attenuation of a packed powder
sample in reflection geometry
- the thickness of the prepared sample in transmission geometry
- the effective sample roughness
and so on?
To check BGMNs profile model, we try to produce somewhat extreme profile
shapes and will present the results at EPDIC-6.
Cheers
Reinhard Kleeberg
University of Mining and Technology
Mineralogical Institute
Brennhausgasse 14
D-09596 Freiberg
Germany
http://www.mineral.tu-freiberg.de/mineralogie/bgmn