Questions about Wyckoff Positions and Refining Free Site Coordinates in JEdit.

Ramon13

Hi all,

During John Evans' great JEdit TOPAS tutorial videos he reviews the refining of space group parameters and relates the ability to refine certain parameters to the atom's Wyckoff positions (etc.). Refer to the link and timestamp attached for the exact site positions and use of the num_posns function to get the Wyckoff positions: . at 47:16.

From that, he goes to the ISOSPACEGROUP site and finds what of those positions can be refined. The process and logic is a bit unclear to me, so I was wondering if someone could potentially help me out with some related questions please? Questions are:

When selecting the spacegroup name on ISOSPACEGROUP, John refers to the spacegroup number, 206. I don't know these numbers off by heart (does anyone?), they're usually not in .cif/.str files and I haven't been able to find an easy way to relate the spacegroup names to the numbers to use on ISOSPACEGROUP for convenience and am usually left scrolling a lot to find my spacegroup. Could someone please advise the easiest way to relate a spacegroup name to the number to use on the site?
John decides to refine the x coordinate on the 24d site. I understand why the 8b site was not used as there's no free variable, but why not the x, y or z on the 48e site? Can you refine both? Why/why not?
How often and when is it a good idea for the user to refine site positions?
BONUS QUESTION 4: When doing Phase ID for a sample, is it always a good idea to strip Ka2?

Thanks for reading and for the amazing site, resources and videos!

Kind regards,
Ramon (PhD student, Western Australia).

johnsoevans

A few answers:

I thought everybody knew space group numbers :-). The best source is to look in International Tables for Crystallography. Or there's a list on wikipedia. You could also try Jeremy Cockcofts online tables (though he uses a non-standard diagram). the Bilbao server is anothe great symmetry resource.
I'm not sure exactly where you were in the tutorial. In general you can refine x, y and z for an atom on 48e in Ia-3. At this point I may have just been refining Y coordinates as its contribution to the pattern is higher. In generaly you can refine any coordinate that has an x, y or z label. Mike Glazer's book "space groups for solid state scientists" is another good resource.
This is a text book question, and too involved for a forum! The throw away answer is that you should refine the site positions when your data quality is good enough to do so, and when their values are important for your science. In some cases (e.g. quantitative analysis on multi-phase samples) it might be better to fix coordinates at well determined literature values rather than refine them.
I personally don't think Ka2 stripping is necessary if you keep your wits about you and remember it's present in the data.

Ramon13

Thanks for the answers! 🙂

With regards to question 2, I was more asking the logic as to why some sites are refineable and others are not for different spacegroups.

johnsoevans

You should find the answer in any crystallography text on symmetry. You could try the teaching edition of international tables.

Ramon13

johnsoevans I've since done a lot of reading on this and am much better at QPA now as I've got a better understanding. I haven't found too much on occupancy sites though, any rules of thumb here? I've recently had to do a QPA with a Goethite-V5+ bearing and it was quite difficult, especially when comparing the analytes to the bulk XRF as a sort of "practical calibration". Would love to see more videos on this too.

You've mentioned previously that some of my questions aren't suitable for this forum. If so, I can try and find another forum and ask the more technical questions? Are there general forum rules I'm missing?

johnsoevans

Hi Ramon. There aren't any specific rules on content beyond normal sensible etiquette and posts being on Rietveld refinement or powder diffraction.

We just need to remember that people who answer posts are generally doing so outside their normal working hours to try and help the community. Questions on topics such as which parameters can/can't be refined take a long time to answer properly, and are covered in lots of widely available text books. It's always best to consult those first as the explanations will have been written more carefully and in a more didactic style than on a forum. I personally found "Space Groups for Solid State Sciences" very useful when I was learning.