Dear All,
I wonder if anyone is using Fullprof option to create file CODFILn.hkl
where n is a phase number. If you do (as I did for magnetic phase),
there seems to be a problem with some values written to this file and
I couldn't find any previous information that it may happen.
1. You may find there incorrect multiplicities of some peaks
(satellites in my case) and some peaks will be missing. This happens
for reflections like 551 and 711, which coincide in d-spacing in cubic
groups.
In my case the satellites of 711 have incorrect multiplicity
h k l n Mult
7 1 1 2 24
7 1 1 1 24
while it should be 16 like in
9 1 1 2 16
and the satellites of 551 are absent. They are "incorporated" into 711s.
This is also present in tetragonal setting eg. for 431 and 501. I
haven't check further but it looks like all d-spacing coincidences may
be affected.
This does NOT affect the refinement since all peaks and multiplicities
are correctly written into other files (CODFIL.hkl, *.out).
2. The CODFILn.hkl may be configured to contain F2 but for some
unknown reason they are multiplied by the multiplicity of reflection.
This creates confusion since they are still called F2.
Here's the example.
First correct output to CODFIL.hkl
CODFIL.hkl
h k l Mult sinT/lamb 2T Fwhm F2 sf2
0 0 0 1 0.02116 4.8519 0.22361 82.962 0.0000
0 0 0 1 0.02116 4.8519 0.22361 82.962 0.0000
-1 -1 1 2 0.07235 16.6406 0.22361 10.635 0.0000
-1 1 1 2 0.07235 16.6406 0.22361 10.635 0.0000
1 -1 1 2 0.07235 16.6406 0.22361 10.635 0.0000
1 1 1 2 0.07235 16.6406 0.22361 10.635 0.0000
And the same section in CODFILn.hkl (this is from simulation of
magnetic phase so F2(0bs)=F2(calc)).
CODFILn.hkl
h k l n Mult F2(obs) F2(calc)
0 0 0 2 2 165.925 165.925
1 1 1 2 8 42.541 42.541
I would expect the Mult to be present in the intensity but not in the
F2 (can it be still called F2?).
Anyway, the FP manual says that the files CODFILn are optional but in
case someone else is using it - be aware of those problems.
When I looked through the rietveld_l archive, I found similar issue
reported for CODFIL.fou files by Armel Le Bail on 17th May 1999.
http://www.mail-archive.com/[email protected]/msg00443.html
Is this the same "Bug or not bug", just for different file?
Best
Pawel
--
dr Pawel Zajdel
Division of Physics of Crystals
Institute of Physics
University of Silesia
Uniwersytecka 4
40-007 Katowice
(currently on a leave but back soon!)
