SCFMO calculations have been made on lithium fluoride clusters both within the crystal and as isolated species. Calculations have been made with different bases in order to separate exchange, charge-transfer and induction energies. Calculations confirm the conclusion from perturbation theory that charge-transfer is an important contribution to three-body energies and gives a small four-body energy. The three-body energies in the crystal environment are much smaller than for the isolated clusters.
I. INTRODUCTIONThe considerable interest in the static and dynamic properties of ionic crystals, both perfect and defective lattices, and ionic melts, provides a continuing incentive to obtain accurate interionic potential functions. There has indeed been no lack of effort on this front; the first alkali halide potentials which gave a good fit to the lattice energy were obtained by Huggins and Mayer in 1933 [1].The total potential for a finite cluster of ions can be written as a many-body expansion. The first term will be the sum of pair potentials, the second we refer to as the three-body term, etc. The pair potentials contain the Coulomb energy, which is dominant at large distances and the exchange energy which provides the short range repulsion. There will be other smaller terms, some long range and some short, such as dispersion, induction and charge transfer, whose relative importance depends on the internuclear separation. There are however problems relating to whether these energies make separate and additive contributions to the two-body energy [2]. For polar molecules and particularly for ions it has generally been considered that the inductive energies are the most important contribution to the higher n-body terms. These energies are associated with the polarization of one atom or molecule in the total electric field arising from its neighbours. As the component fields must be combined vectorially the polarization depends on the relative positions of the neighbours.