doktorandin0815
Hey guys,
I am a total newbie in Rietveld refinement, especially with Topas.
I am trying to quantify my multi-phase cement samples, but my fit is a catastrophe, although I adjusted all parameters in the right way. The problem might be, that I measured only until 40° 2Theta, as all (for me) relevant phases are within this area and at that time, I didn’t know that I was going to perform Rietveld refinement. However, some second or third peaks of those phases are beyond 40°. Does Topas calculate in consideration of the full 2Theta range and could this be the reason, the refinement doesn’t fit? If yes, can I somehow set the max angle to 40°? I am searching for this option for hours and couldn’t find anything yet. Would be quite annoying to measure everything again up to 70°.
Thanks for helping me!
johnsoevans
Topas will default to fitting the full range of your experimental data. You can change the range fitted using e.g. "start_X 10" and "finish_X 30" in the xdd section of the input file.
Multiphase quantitative Rietveld is always difficult to get correct, particularly with cement samples. I'd strongly recommend you try and find a local expert who could help you get started. If not you should look carefully at something like "The Rietveld Method" by Young or chapters in the Billinge/Dinnebier or Zavalij/Pecharsky texts on powder diffraction. There's a good chapter on quant analysis and potential pitfalls in Billinge/Dinnebier. Those texts will tell you that data to 40 degrees is likely to make your life difficult in terms of information shortage...unless you used a short wavelength.
mfisch
Hi there!
Having quite some experience with cement samples I can assure you that those are among most difficult ones... but it can be done ;-)
Google... You wil find a lot of information by Paul Stutzman at NIST, the Neubauer group at Erlangen, Angeles de la Torre, the MinSocAm cement book, and it might be worth a try to contact Bruker, as they have their own cement-XRD group.
With your range, you almost missed minor phases such as periclase (only one reflection in your range). In general, getting control of peak widths and the background (iron fluorescence, blended cements) is as important... This, you will probably not be able to do with your limited range if you have used Cu radiation.
doktorandin0815
Hey!
Thanks a lot for your answers! With some help of a colleague and the books I got a proper fit of my data.
I am using a pretty old version of Topas, I suppose a cracked one cause I can’t even see what version it is (“Class TPanel not found” pops up when I click "About").
Anyway, there is still a question left. I am working with an internal standard (10% CaF2). Is there any possibility to normalise the data to that standard? I know that this is working with HighScore Plus and probably also with a newer version of Topas. Does anybody have any experience with that in older versions? As mentioned, I unfortunately can’t see what version I am using...
oblivion
Hey!
I am new to TOPAS too. I have been learning it on my own since I couldn't find someone to help me with it, in my university.
I think you can see the version in the "Help" option. Still, the TOPAS version would be shown every time you open the application.
In case you need to know, I have been using the book "Principles and applications of powder diffraction" by Abraham Clearfield. It's a really concise text and gets you up to date with the theoretical aspects of powder diffraction, before you start working on a Rietveld software.
Good luck!