ABSTRACT
Extensive synchrotron (28 K) and conventional sealed-tube (9 K) X-ray diffraction data have been collected on tetrakis(dimethylphosphinodithioato-S,S')thorium(IV), [Th(S(2)PMe(2))(4)]. The use of very low temperatures, well below those obtained with liquid-nitrogen cooling, is crucial for the accuracy of the data. This is due to minimization of temperature-dependent systematic errors such as TDS and anharmonicity, and extension and intensification of the data in reciprocal space. Comparison of structural parameters derived separately from the sealed-tube data and the synchrotron data shows good agreement. The synchrotron data are markedly superior when comparing refinement residuals, standard uncertainties (s.u.'s) of the data and s.u.'s of the derived parameters. However, the study suggests that there are still small uncorrected systematic errors in the data. The very large extent [(sinstraight theta/lambda)(max) = 1.77 Å(-1)] of the synchrotron data and the very low temperature at which they were collected makes it possible to separate anharmonic effects from electron-deformation effects even with only an X-ray data set at a single temperature. The electron density shows a large polarization of the outer Th core of d-type symmetry. This deformation is successfully modelled with contracted multipolar functions, which are only slightly correlated with anharmonic expansions in reciprocal space when using the full extent of the data. In the data collection more than a factor of 100 in speed is gained by use of image-plate area detectors at the synchrotron source compared with conventional sequential measurements. Thus accurate, very low temperature synchrotron-radiation diffraction data can now be measured within days, which makes electron-density studies of compounds beyond the first transition series more frequently within reach.
