Hi Moritz,

You have to start the refinement first by refining the scale factor (all the other parameters fixed). This is particularly important for two phase analysis. Then detector position comes and so on. One thing that is very important for Rietveld is that you have to do the measurment with the idea that you are going to do refinement afterwards. For example, one very important refinement parameter is peak shape and asymmetry parameters, at best you have to know them from your instrument resolution, sometimes you can upload a resolution file. For me, most of the time the refinement is fine until refining these parameters and without having a good knowledge about them, your fit goes wrong at this point. Send me your email address for seding the documents.

Yes, I guess you can DL the manual from their website. If you search for fullprof-intro in google, you can find a bunch of useful presentations and tutorial.

In your editor of the pcr files

you can add several phases

on the website, you have examples and tutorials

http://www.ill.eu/sites/fullprof/php/tutorials.html

Hi I am having the same problem. There is not really a example out there of two cif files combined into one pcr file. Maybe someone could post a pcr file. Eg. I tried to create one for Fe and Fe3O4 but it would not work...

To do a fine rietveld using fullprof, you should have a pretty clear and background corrected pattern. Does the pattern distinct and signal to noise ratio high enough? another important point is that you have to a step by step refinement, this means that some parameters are much more important than the others. If you like I can send you some procedure.

Hi Aidin,

thanks for your post. SNR is good and it is a distinct pattern. I am not sure what you mean by background corrected pattern. Do you correct before fitting and then make a already corrected dat file? I though by choosing a 6 coefficient polynomial background in fullpof this account for background correction with regards to peak intensities and fitting during the actual fit? I am doing step by step fitting. I start with lattice parameters, the first two background coefficients and scaling factors and then go from there.

I have solved my previous pcr file problem. I created one in MAUD and exported it with two phases. Not sure why this one worked compared to the one posted above.

However, I cannot reach a good convergence. I have one sample someone did for me using TOPAS (fitted perfectly with low Rwp), but I cannot reproduce this in fullprof and therefore cannot fit the rest of my data either.

If you have some guidelines I am more than happy to get a copy, maybe I am missing out on something.

Thanks.

-Moritz

Hi Aidin,

Thanks for your reply. It would be very helpful if you send me some procedure. I reach a convergence with good values of Rp, Rwp, Rexp, che2 etc. but the final fitting profile is not so much satisfactory. If you give your valuable suggestion then it would be helpful for me.

Thanks

Satish

Hi Moritz,

You have to start the refinement first by refining the scale factor (all the other parameters fixed). This is particularly important for two phase analysis. Then detector position comes and so on. One thing that is very important for Rietveld is that you have to do the measurment with the idea that you are going to do refinement afterwards. For example, one very important refinement parameter is peak shape and asymmetry parameters, at best you have to know them from your instrument resolution, sometimes you can upload a resolution file. For me, most of the time the refinement is fine until refining these parameters and without having a good knowledge about them, your fit goes wrong at this point. Send me your email address for seding the documents.

Hi Satish,

If you got such good fit evaluation parameters as you mentioned, probably you do not need to care much about how perfect measured and simulation patterns are matched. But I am wondering if the measured and simulated patterns are not fit to each other, how do you get a small Chi2? consider playing with the backgroud correction. do you correct you backgroud before running refinement, or you upload a background file?

The below youtube links are about two-phase structural refinement of Powder X-ray diffraction data using FulProf Suite Software. These videos are for beginners and present one of the easiest way to do two-phase structural refinement using FullProf Suite. Highlighting the Steps involved in two-phase crystal structure refinement.

Part 1: https://www.youtube.com/watch?v=PzUJzHAqPTg

Part2: https://www.youtube.com/watch?v=yd-aSOySUiA&t=50s

Part3: https://www.youtube.com/watch?v=PMKe21o-3AM

Dear Satish,

Sorry for delay response to your question. I think your problem has been resolved. Yet I want to add a simple and easy procedure for the refinement of the two phases Rietveld refinement using FullProf Suite. The procedures are as follows:

1) Fit Le-Bail profile fit for 1st phase

2) Then fit Le-Bail profile fit for 2nd phase (by adding second phase) by fixing all the refining parameters of 1st phase

when you come to the end of best Le-Bail fits. Then go to the next step.

3) Try Rietveld refinement of 1st phase fixing refining parameters of the 2nd phase.

4) Then try Rietveld refinement of 2nd phase fixing refining parameters of the 1st phase.

5) After that you can refine required parameters of both phases simultaneously and reach to the best fit.

Note that within the refinement please save the pcr file with different name to avoid the failure of refinement. When one pcr file get corrupted then you can use previously saved file.

I hope it will be helpful for you and the new coming researchers in this field of crystallography.

Thanking you,

Dinesh Kumar