Abstract

Published X-ray powder measurements on diamond and three different sets of published X-ray data on silicon have been re-analysed for bonding features using full-matrix least-squares refinement combined with a statistical analysis of the results obtained. In all cases a highly significant improvement in the fit between observed and calculated structure factors was obtained by introducing a tetrahedral distortion of the spherical `prepared' charge distribution, but subsequent introduction of a fourth-order cubic distortion proved to be highly significant only for the diamond data and one set of the silicon data. Hartree–Fock calculations gave a better fit to the diamond measurements than calculations based on Hartree–Fock–Slater wavefunctions. The necessity for placing restrictions on the form of the radial functions associated with the non-spherical distortions, the large estimated standard deviations of the distortion parameters and the dependence of the parameter values on the set of basis wavefunctions chosen to describe the spherical `prepared' charge distribution indicate the need for exercising caution in analysing the experimental measurements for bonding features.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call