This example uses example Cimetidine data from the EXPO software
Now assuming you have an acceptable peak profile list, it is time to try
and index the powder diffraction pattern.
As possibly repeated elsewhere, for organics, it can be best to scope the range of possible results by running ALL the indexing programs (ito, treor, treor, taup, kohl, lzon and fjzn). In this example, this might seem quite contrived as the result is quite obvious (we already know the true answer), but in most cases, being to lax on scoping the possible range of solutions could put you in a world of pain! For example, on standard laboratory instruments where alignment and sample preparation are of high quality, you might still be lucky to get Figure Of Merits greater than 10 on large cell organics. As per the Crysfire tutorials, run XF2CRYS to convert the XFIT.TXT file into a Crysfire *.CDT file. (The wavelength in this case is 1.52904 Angstrom) For brevity of this tutorial (though you can be running Chekcell concurrently with Crysfire), run All 7 elligible programs (ito, treor, treor, taup, kohl, lzon and fjzn). Whem prompted, a rule of thumb is to use around 25 peaks, but everyone has their favourite values when these is an excess of peaks to choose from)
|
In this example, it is obvious which is the "most likely" cell but in a larger cell, you may have a number of different cells with close FOM's such that closer examination will be necessary. Le Bail whole profile fitting might help you and at worse case, you may have to try to solve the structure using the different cells to see which gives the most reasonable answer. A few things to look out for include doubling and halving of cell volumes as an indicator of possible "true" cells.
|
Now run Chekcell and we can graphically go through this mass of results. When prompted, the wavelength used for the peak positions is 1.52904 Angstrom
|
Performing a spacegroup assignment using the Best Group option in Chekcell suggests that P 21/A could be the best spacegroup. Thus the result is:
This automatica spacegroup assignment should only be used as a guide and the other spacegroups manually browsed through to see the differences in observed and calculated lines.
|
Also note, as per the peak profiling page, that those low angle shoulders are most likely due to something weird in the XRD geometry and sample preparation.
The cell also calculates a HKL at a peak we missed in the peak profiling. (minor confidence booster)
The conditions used for assigning the spacegroup are viewable as well if you wish.
|