Rietveld assymetry parameter

Charles Witham ( ckw@hyperfine.caltech.edu )
Thu, 22 Jan 1998 13:49:33 -0700

I've been writing a peak-fitting package for the application Igor that fits
XRD data to Voigt peaks modified to include Kalpha1 and Kalpla2 peaks and
using Rietveld's semi-empirical assymetry parameter:

Y = I exp[-b (2Th -2Th )^2] * {1-P(2Th -2Th )^2*sign (2Th -2Th )/tan(Th )}
i k k i k i k i k k

2Theta values in radians. The assymetry parameter (P) and the Kalpha
amplitude ratio are varied as pattern parameters (20 - 30 peaks in a
pattern). If I understand correctly, this is the right correction to use
for X-ray peaks, while the sum of 5 peaks would be used for neutron
diffraction peaks.

My problem is that the assymetry parameter is taking on extreme values,
usually 4000-7000.

This happens when I reverse my logic as well: The difference between peaks
with an assymetry of 0 and an assymetry of 1 is on the order of 10exp-4.

Am I implementing the assymetry correction incorrectly? Should I expect
the assymetry value to be so large?

thanks,

chuck

ps, Thanks to everyone for the information about stacking fault analysis,
all of which I have not yet had the time to pursue. If anyone is
interested in the above fitting package, please ask.

Charles Witham

Graduate Research Assistant, Materials Science
California Institute of Technology
Mail Stop 138-78 ckw@hyperfine.caltech.edu
Pasadena, CA 91125 (818) 395-3627
fax -2940