Abstract

At a given complexity level, one expects Rietveld refinements from synchrotron data being less challenging than from conventional X-ray. Ab initio structure determinations are expected also to be easier by synchrotron powder diffractometry. A few years ago, it was said possible to make Rietveld refinements on much bigger structures than could be determined solely from powder data. Because of recent advances in methods for locating previously known molecules in crystalline cells, this statement was recently modified in "we can now determine structures biggest than we could refine without interatomic distance constraints".

The belief in synchrotron data superiority is unconditional. However, some studies are realized by selecting low-wavelength (down to 0.7Å) and small angular ranges (1-60°2q ). Such choices may correspond to lowest resolution than conventional equipment would provide with CuKa radiation. A modern in-laboratory powder diffractometer may give full width at half maximum as low as 0.04°2q . This is mathematically as good as 0.02° at a synchrotron facility when choosing
l = 0.72 Å. A good example is that of sample 2 (tetracycline hydrochloride) of the Structure Determination by Powder Diffractometry Round Robin [1] for which one participant preferred to solve the structure by using the conventional rather than the synchrotron data (both were provided).

Of course the synchrotron superiority is quite real due to highest resolution, intensity and tunability. Anyway, superb conventional X-ray powder patterns will be shown at this congress, supporting the comparison with 'bad' synchrotron data, apart from the time for obtaining them (3 to 6 days).

1. http://www.cristal.org/SDPDRR/