CCP14
Methods, Problems and Solutions
GSAS (General Structure Analysis System) Rietveld powder diffraction and Single Crystal software
Difficulty in profiling Powder XRD data in GSAS, Peak Cutoff values and the influence of Powpref
(Thanks to Rory Wilson of Queen Mary University of London for the hints behind this tutorial)
The CCP14 Homepage is at http://www.ccp14.ac.uk
[The reference to use for GSAS in any resulting publications is:
A.C. Larson and R.B. Von Dreele, "General Structure Analysis System
(GSAS)", Los Alamos National Laboratory Report LAUR 86-748 (1994).]
[The reference to cite in any resulting publications for using EXPGUI is:
B. H. Toby, EXPGUI, a graphical user interface for GSAS, J. Appl. Cryst. (2001). 34, 210-213]
|
The following is performed using Brian Toby's EXPGUI which can be a very user friendly
way to control GSAS refinements. It is written in Tcl/Tk so works in Windows and UNIX, is
freely available, and the source can be freely modified and added to.
EXPGUI Homepage with download and installation information on EXPGUI for Windows and UNIX
Download FTP Site and Mirrors
|
Note: The following is done with refinements that are not yet fully converged
If you find you are having difficulty fitting laboratory powder XRD data, this could be due to
a combination of problems: 1) having to high a Peak Cutoff value (0.01 instead of 0.001 - 1% instead of 0.1%)
and 2) not running POWPREF after resetting the Peak Cutoff value. (if you are doing Le Bail extraction, turn off
all refining parameters before running Powpref (which resets all the Le Bail intensities to zero);
then do a few cycles of Genles to let the Le Bail intensities get to their approximately correct
value before turning on the refined parameters again)
Following is an example of the type of non-fitting peak shape with an 0.01 Peak Cuttoff value which
you may be up against. (though other cases may be more subtle)
|
|
The first part of this is to set the Peak Cutoff to a lower value (in this case going from
0.01 to 0.001 (1% to 0.1%)).
Though if you then do further Genles cycles, the profile will not improve!! For the new
Peak Cutoff to have any effect, you must then run POWPREF.
(if you are doing Le Bail extraction, turn off all refining parameters
before running Powpref (which resets all the Le Bail intensities to zero); then
do a few cycles of Genles to let the Le Bail intensities
get to their approximately correct value before turning on the refined parameters again)
|
|
Now run Genles and you should see immediate improvements in the profile
as per the following screen dump.
|
|
Now continue on to get the best fit you can to the data.
|