Pseudopotential SCF–MO studies of hypervalent compounds. I. XeF2 and XeF4

Abstract
The (ab initio) effective potential theory developed by Ewig et al. has been applied to a series of hypervalent compounds with a view to elucidating the anomalous properties of several of the higher fluorides of xenon and iodine. In this initial paper the development of a minimal basis set substantially better than an STO‐4G atom‐optimized set is described. Calculations carried out on XeF2 and XeF4 give valence orbital energies in fair agreement with those obtained with the more flexible, all‐electron SCF–MO calculations by Basch et al. Equilibrium structures of XeF2 and XeF4 provided by the effective potential calculations possess the correct symmetries. Bond lengths, although too long by 0.09 Å, correctly reproduce the contraction observed experimentally upon fluorination of XeF2. Calculated bending and stretch–stretch interaction force constants are in pleasing agreement with experiment, as is the stretching anharmonicity. Stretching frequencies evaluated at the experimental bond length, however, are 25% high. Overall, the ability of the present treatment to give a reasonable account of the structures and force fields of XeF2 and XeF4 justifies its application to the higher fluorides where interpretations of observations are more speculative