Best powder data for structure solution

[Arianna]

Hi forum!
I am about to dive in the intriguing waters of structure solution from XRPD data, with a couple of presumably simple molecules.
Since I have no experience with that, and don't know directly someone who has it, but I am strongly determined to get those structures, I would like to ask you if you can recommend HOW a powder pattern should be collected (with a lab diffractometer) to increase the success probability.
Everything I've been told so far is "you should collect the best you can do".
I happen to have 3 or 4 different instruments around Europe where I could collect my data and so it would be nice to know which minimum characteristics should a diffractometer feature to be suitable to my purpose (or at least not inadvisable), eg: transmission (implying oil and polymer foils) or reflection geometry? minimum stepsize? 2theta range I guess it's recommendable up to at least 80-90° degrees... What about the required optical setup and slit-sizes?
I am also looking for some book/article to read on this topic, but sometimes the experience can help more effectively, that's why I also ask.
Thank you very much for helping,
Arianna

[jchrist]

Hi Arianna

I haven't done structure solution from powder my self, but here are some links that should be helpful.

From an old workshop on structure solution from powder diffraction, contains some tutorials and links to much of the software available.
http://www.fkf.mpg.de/xray/downloads/AK13_DGK/Bericht9.pdf

Another software homepage: the CCP14 project:
http://www.ccp14.ac.uk/solution/powder_struct_solution/index.html
And their tutorial:
http://www.ccp14.ac.uk/solution/powder_structure_solution_pathways/

Finally what the IUCr has some instructions:
http://www.iucr.org/resources/commissions/powder-diffraction/projects/organic-datasets

[Rudi]

In addition to Jeppe's suggestions, I recommend reading of some recent research articles, where structures were determined ab initio from X-ray powder data. The experimental conditions may vary somewhat, depending on the nature of your compounds (e. g. inorganic solids, molecular crystals, coordination polymers).

Best of luck.

[Arianna]

Grazie mille!!
Arianna

[Beamtime Knitter]

Hi Adrianna,
The best advice I have it that as well as 'The best you can do' also the 'most uncomplicated' you can do. I find that as soon as I have any contaminant peaks this makes structure solution much more challenging. In terms of set up transmission geometry will always be better than reflection because of the better resolution of your peak shapes, I'd even say the smaller the step size the better as the more information you have on the peak shape, the better. What software are you thinking of using? I have had quite a bit of success with simulated annealing in FOX, and the wiki page is very informative http://vincefn.net/Fox/ (but simulated annealing really only works if you know your unit cell contents pretty well).
I should also maybe suggest (and given where I'm from you knew I was going to say this) that if you get no luck with the Lab data then maybe thing about a synchrotron experiment?
Good luck!

[Arianna]

Thanks a lot, I am trying to study something and looking forward to try. Synchrotron would be to good to be true! And thank you for this link to fox, I never heard about it!