At 12:04 19/01/99 +0100, Lubo wrote:
>Hi Armel,
>I am sure you will not be satisfied by this answer, but my solution to
>this problem is : USE transmission geometry. We have been doing it for
>years getting nice results for kaolinites, micas, vermiculites and
>smectites, not to
>speak of some organic plates. No need to add -osils (what's that, BTW).
>Lubo
>
>P.S. I am not adding the type of diffractometer, but if you wish I can
>scan a sample for you ....
>
After some scans, and up to now, the technique of mixing the
sample with a solid filler material (like Cab-osil M5) or dusting
a sample through a sieve are the clear (less expensive) winners
of this private round robin, provided the mixture is inserted in
a side-loading sample holder and never packed. Moreover a
mixture fifty-fifty in volume is recommended with the amorphous
solid filler technique.
Transmission geometry favours of course preferred orientation
directions at 90� from the reflection geometry one. With
platelet habit clay minerals and 00l orientation by reflection
geometry (and say, a March-Dollase parameter which can
be easily as low as 0.5 when a sample is packed), one obtains
with transmission geometry the reverse : a March-Dollase
parameter as high as 1.5, selecting the same 00l orientation
direction by the Rietveld method (no orientation corresponding
to MD=1.0). The problem of course is that Rietveld refinements
are of quite low quality when the MD is so different from 1....
Best,
Armel Le Bail - Universite du Maine, Laboratoire des Fluorures,
CNRS ESA 6010, Av. O. Messiaen, 72085 Le Mans Cedex 9, France
http://www.cristal.org/
