Unusually large discrepancy exists between the reported value of the experimental geometries of hexafluorocyclobutene determined from microwave spectroscopy (MW) and those from electron diffraction (ED). In an effort to explain the origin of the difference, the geometries of cyclobutene, 1,2-difluorocyclobutene, 3,3,4,4 -tetrafluorocyclobutene, hexafluorocyclobutene and 1,2-dichloro-3,3,4,4 -tetrafluorocyclobutene are optimized by ab initio calculations at the HF, MP2 and CISD levels of theory, and further refined for the higher correlation effects through the use of model calculations of ethane,ethene, fluoroethane and fluoroethene. Comparison of the calculated results with available MW and ED parameters imply that MW geometrical parameters are probably more accurate for hexafluorocyclobutene than ED ones. The difficultly associated with vibrational corrections for hexafluorocyclobutene appears to be responsible for the unusual pattern of the difference between ED and MW structures.
For hetero-substituted molecules MXY (M = Ca, Sr, Ba ; X, Y = H, F, Cl, Br, I), the equilibrium geometries of the ground states have been optimized at the HF, MP2, and CISD levels of the theory using effective core potentials and extended valence basis sets. The bondings, bending potentials and structures characteristic of hetero-substituted MXY molecules are presented in this paper. Core polarization (influence of the subvalence) and s-d hybridization (influence of the valence shell), to varying extents, control the bending in $MX_2$ and MXY systems. The bending potentials of MXY compounds, however, are relatively flatter than those expected from $MX_2$ and $MY_2$, leading to all linear forms of CaXY and linear SrFBr, SrFI, SrFAt, SrClBr, SrClI, and SrClAt of SrXY.BaXY are all in the bent forms but their bending potentials are very flat. Natural Atomic Orbital and Natural Population Analysis are suitable tools for explanation to this phenomena.
