A CRITICAL-REVIEW OF THE EXPERIMENTAL VALENCE CHARGE-DENSITY OF GAAS

The valence charge and difference densities of GaAs have been calculat ed without previous refinements of a charge density model using six di fferent data sets of X-ray structure amplitudes published until now. S ince the data sets have been measured by means of different experiment al methods and due to the different data treatment, the individual str ucture factors differ on the absolute scale. Furthermore, different te mperature factors have been published. In order to bring the data to a common level, we used the same two harmonic temperature factors and t he same algorithm for correcting the different sets of experimental da ta for anomalous dispersion. Because of the non-centrosymmetry of the zinc blende structure, these procedures are not strictly model-indepen dent. A simple bond charge model was used to obtain phases of the stru cture amplitudes and to perform the above-mentioned corrections. In ge neral, the details of 'experimental' charge densities depend sensitive ly on the balanced ratio among the structure factor moduli used. A smo oth density map is only obtained if all F have the same high level of accuracy [delta(F)/F less than or equal to 1%] and if 'outliers' are o mitted. Only four of the six data sets describe the covalent bond and the partial charge transfer between neighboring atoms, in qualitative agreement with our expectation based on the results of pseudo-potentia l calculations. However, some quantitative discrepancies remain, parti cularly in the height of the charge density maximum between nearest ne ighbours and in some details outside the bonding region.