At small angles or in case high collimation, the peaks become anisotropic and with a Lorentz addition. The resolution describes correctly only with adding Lorentz and anisatropy parameters. It comes from the convolution beam with collimators, the shape of the peak from the tube, and so on. Just using of the UVW parameters is no longer correct. How to do it right?
First variant. In addition to UVW, one use the anisotropy and the Lorentz contribution from the reference experiment. But then how to calculate strain and so on?
Second variant. It seems more correct to use the resolution file, since HWHM is already included the widths from anisatropy and Lorentz. But the distorted peak shape is still not taken into account.
The sample is a reference Al2O3 or Si with particles larger than a micron. The program is FullProf.
Dear Alexey Bykov ,
That is a critical discussion; some points are already discussed in the paper:
Peak shape and resolution in conventional diffractometry with monochromatic X‐rays, D. Louër J. I. Langford. First Published: October 1988, https://doi.org/10.1107/S002188988800411X.
Also, check the class notes from Frank Girgsdies, Electron Microscopy Group, Department of Inorganic Chemistry, Fritz-Haber-Institut der MPG, Berlin, Germany, Peak Profile Analysis in X-ray Powder Diffraction.
As already stated, the second variant seems more reliable since the correct choice for the anisotropic peak profile shape function accounts for the peak distortion by not rising inconsistency on HWHM from the resolution file neither on UVW obtained in the fitting result.
Best regards,
WNM
I wrote a Matlab script, for automatically generating IRF file from usual resolution parameters. As i know, there isn't any program for it.
There is another one, more accurate way.
This is a deconvolution of wavelength distribution (from measuring the real specter of the tube), instrumental profile (numerical calculate from geometry) and real structure profile (from strain and size). This can be done by TOPAS or bgmn program.
Dear Alexey Bykov ,
For sure, an instrumental profile and an actual structure profile are the right way to deal with it.
The deconvolution of wavelength distribution is a sort of correction that should not be present in a usual X-ray tube in principle. In particular, if an effective wavelength monochromator is used, it shouldn't be necessary. Nevertheless, as a manufactured instrument part, it may happen, and so the unexpected and essential correction can be performed.
Best regards,
WNM
- Badges
- Science topic
- Similar topics
- Correction