I have a XRD data with a number of overlapping peaks. Hence I am not able to go below GOF 2 and Rwp 10 .
What Goodness of Fit and Rwp could be considered to be the best for Rietveld Refinement in XPert High Score Plus for a material with overlapping peaks?
I have to disagree with Rajeev on this. The higher quality data may give you better Rexp values, but there is no guarantee that this will result in better GoF or Rwp values; it is more difficult to obtain better Rwp and therefore GoF with data that would result in a very low background. Poorer data with high backgrounds can give you better GoF and Rwp - not that you should collect bad data! it is just the way the values are calculated. Also Gof and Rwp values depend on the model rather than the raw data. A low GoF does not guarantee that the model is correct.
While improving resolution may help with peak separation there are two things to consider here - the Rietveld full pattern fitting is not affected very much by overlapping peaks as long as the models used are good. (But the better resolution would allow a more accurate qualitative analysis). Secondly as peaks become sharper there appears a greater difference between the raw data and model as so increasing the GoF and Rwp values.
Once again I would recommend the paper by Brian H Toby.
Article R Factors in Rietveld Analysis: How Good is Good Enough
A GoF value of 2 is not very good, the ideal value is one. The Rwp depends on the statistics of data and so there is not a reference value. If you post a graphical comparison maybe we can better help you.
A GoF value of two is not so good, the ideal value is one. The Rwp value depends on the statistics of data so there is not a reference value. Is you post the graphical comparison maybe we can better help you with some suggestion.
This topic has been covered before. Have a look at the answers in the links.
(I have never, ever seen a GoF of 1!! - not in the real world anyway)
https://www.researchgate.net/post/In_Rietveld_refinement_what_is_the_value_of_Rp_and_Rwp_for_an_excellent_fit
https://www.researchgate.net/post/Which_is_better_in_determining_the_goodness_of_fit_Rwp_or_Chi2
In fact one is the "ideal" value. Me too, I never got one, but in some cases values very near to one can be obtained
Nibedita, to get good GOF and Rwp values the majority of your input goes into data collection phase. The lower your Rexp value, the better your GOF and Rwp values would be. If you are getting overlapping peaks try to reduce the step size and increase time per step, that will help you in separating out closely spaced peaks.
I have to disagree with Rajeev on this. The higher quality data may give you better Rexp values, but there is no guarantee that this will result in better GoF or Rwp values; it is more difficult to obtain better Rwp and therefore GoF with data that would result in a very low background. Poorer data with high backgrounds can give you better GoF and Rwp - not that you should collect bad data! it is just the way the values are calculated. Also Gof and Rwp values depend on the model rather than the raw data. A low GoF does not guarantee that the model is correct.
While improving resolution may help with peak separation there are two things to consider here - the Rietveld full pattern fitting is not affected very much by overlapping peaks as long as the models used are good. (But the better resolution would allow a more accurate qualitative analysis). Secondly as peaks become sharper there appears a greater difference between the raw data and model as so increasing the GoF and Rwp values.
Once again I would recommend the paper by Brian H Toby.
Article R Factors in Rietveld Analysis: How Good is Good Enough
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials