Dear all,
I did Rietveld refinement of XRD data with fullprof and now I would like to present the results with the corresponding error. There is somthing about the standard deviation of the refined parameters in the fullprof manual (page 43, equation 3.7), but I still don't know, where I can find the matrix parameters of the equation (pereaps in the output file?).
Another question is, what is the difference between chi2 and "global user-weighted chi2"? And what causes chi2 to be less than 1? (see attached)
The error of each refining parameter are included in the output file.
The Rwp is quite large - may be that is why you have Chi2 less than 1.
Thank you for your answer Mr. Wei Kong Pang.
What value should I expect for Rwp? Chi2 is defined as (Rwp/Rexp)2 , so I should have Rwp>Rexp. This means, that IRwp is too small, or Rexp is too big.
I have 11 different XRD data, and I have the same problem for all the fits. Perhaps I do some mistake in the refinement procedure? But I follow (more or less) the steps in the manual, cap. "The Rietveld method in practice" (page 13), briefly:
If I have chi2
why include preferred orientation?
shape parameter vs FWHM?
From your fit profile, I will say the lattice parameters should be correct.
To refine Biso, atomic coordinates and Occ of elements, you have to be careful. Depending on what phase/symmetry you are dealing with. And may be too many background points???
I include the preferred orientation because I do XRD with Debye-Scherrer geometry (transission through the sample), and the powder is between two scotch tape, which can have an influence on the orientetion of the crystals. Is it better to let it fix?
What you mean withIn "shape parameter vs FWHM "? In Fullprof the shape parameters are Eta and X, and the FWHM parameters are U, V, W (and IG, which I did not refine).
Perhaps, I have really too many background points. I used WinPlotr to find the bgr points (automatically). The results are 76 bgr points.
Thank you for your patience.
from the graphical comparison the fit is very good (the difference curve is almost flat). The preferred orientation should be used only when you can verify an important difference between an experimental and a calculated profile, it is not usual to insert this parameter as a default. The discrepancy factors do not have reference values, the eye is the better indice to decide the quality of the refinement. Be careful with B thermal factors, in many cases they assume meaningless values.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Condensed Matter Physics
- Solid State Physics