Hello, I have been reading what I can find about the use of spherical harmonics in correcting for preferred orientation in PXRD data. I have read that the number of terms which are appropriate depends on the sample symmetry, but cannot find any specifics which would allow me to determine how many terms are physically justifiable (I have a needle-like sample with P21/c symmetry). If someone has a helpful reference or other recommendation, I would greatly appreciate it. Thanks!
Hi John,
In my experience, addition of spherical harmonics terms beyond a certain number not only does not help, but makes the fit more unstable. I have also had samples with needle-like shapes and I found that beyond more than somewhere between 10-15 terms, fit became unstable without any further improvements.
Best,
Arman
There is a nice paper which explains e.g. the number of parameters that has to be calculated (i.e. refined) in dependence of spherical harmonics order, see attachement
You can try the refinement from a lower n (such as n=2) at the beginning, increase the order of spherical harmonics gradually, until you satisfy the results.
- Badges
- Science method
- Similar topics
- Methodology
- Crystallography
- X-ray Diffraction