At 05:51 PM 7/3/01 +0200, Luca Lutterotti wrote:
> - only one type of peak is inside the program but the asymmetry is
> handled in a much better way than in other programs (is a real
> convolution as it should be) so I never found a pattern that I was not
> able to fit very well with my function (from high resolution synchrotron
> to neutron TOF). This is the reason I never added another function.
The peakshape you're using sounds very interesting... is it written up
somewhere? Sounds like a programmers dream. I'm on the lookout for
something (a function) which could fit hkl dependant peak asymmetry,
possibly due to stacking faults. Ideally I'd just like to get hold of a
functional form for the hkl dependence and asymmetry, without starting off
with a model for the stacking faults. Is there a simple model out there
I've missed?
Thanks in advance,
Jon
---
PS: Apologies for historical nitpicking but there's been a peakshape in
CCSL doing low angle asymmetry effects by convolution, apparently, for more
than 15 years.
[See: Eddy MM, Cheetham AK, David WIF, Powder Neutron-Diffraction Study of
zeolite NA-ZK-4 - An Application of New Functions For Peak Shape and
Asymmetry,
Zeolites (1986) 6, 449.]
ToF peakshapes are also handled by convolution, as are lorentzian and
gaussian contributions to whichever instrument type. In both cases my
limited understanding is that the functions are evaluated directly in
fourier space, then an FFT is used to put them back into twotheta or ToF,
which saves doing one of the two FFT's needed for numerical convolution. It
also keeps my understanding limited ;)
