Check what has been said at https://www.researchgate.net/post/Why_is_the_Rexp_value_of_my_Rietveld_refinement_so_high

Hii,

Rwp value does not matter here with fullprof. Only the gof= square root of chi square will be less as much as possible but not less than 1.Your results are good and reportable. Even the blue line is approximately flat. Rwp value differs for different software and depends upon the algorithm of minimization. You can check my paper. I have reported the same kind of values for Rwp.

10.1016/j.physb.2020.412735

Gaurav Gupta Thanks sir for giving valuable information.

Andreas Leineweber Sir I have read post in which it is said that if third column(expt error) will be added in addition to angle 2theta & counts then Rwp value will be reduced. but sir i am not able to understand in which format of data file can be put in fullprof software. Please explain.

Gaurav Gupta : No, Rwp matters. It may be ok, but it need not be ok. Priya Dhuria , I suggest that you really read the old discussion and that you do not rely on the suggestion that everything is ok. The discussion I hinted at refers to a frequent flaw upon using diffraction data for Rietveld refinement.

I think that I did not say that ("...which it is said that if third column(expt error) will be added in addition to angle 2theta & counts then Rwp value will be reduced.") in this way. The consideration of the 2theta-Intensity (XY) data in a Rietveld refinement considers also the standard deviation of the Intensity. If you are counting single quanta ("counts") according to Poisson statistics the standard deviation of the intensity is (intensity)1/2. Typical Rietveld software (I do not know the details for Fullprof) interprets "intensities" in terms of "true" intensities in counts, IF NOT STATED DIFFERENTLY. Depending on data conversion or even output of your intensity data may lead to XY data where Y is not a true intensity in counts, while the program may incorrectly assume sigma(I)= I1/2. This will lead to meaningless Rwp, standard deviations of refined parameters etc. That problem can be considered by using an XYE (X,Y,Sigma) format. Than you can scale you data as you wish (where you scale the Sigma in the same way) and even consider variable counting times. As said, I cannot help you with the specifics of the Fullprof formats (but they are likely documented). However, be also sure about the meaning of your own data as you provide them to the software.

Andreas Leineweber thank u so much sir for guiding me. I will take care of your comments.