The role of alternative geometries in alkali–halide clusters

Abstract

The relative importance of the cubic structures that were proposed to explain magic numbers for alkali–halide cluster ions from cluster sources is examined via total-energy calculations on nine-atom cluster ions of various optimized geometries. The relative energies of the planar, tetrahedral, quasioctahedral, lowest energy nonplanar nine-atom clusters for LiF, LiI, NaI, KI, RbI, CsI, NaF, NaCl, NaBr, and NaI are computed using Martin’s Coulomb plus the Born–Mayer potential model. The most stable structure is invariably a slightly puckered plane. The relative energies of these clusters for LiF have also been tested using Hartree–Fock and density functional theory. Other comparisons are made for NaCl clusters and eight-atom LiF clusters. The computationally more tractable Born–Mayer potentials rather accurately predict the relative energies of the clusters in the ab initio calculations. The largest problem is too strong a repulsion between like atoms which overestimates the energy difference between the planar and quasioctahedral structure proposed by Morgan et al. These calculations suggest a greater population of noncubic structures for the larger and more polarizable alkali–halide cluster ions in beams from cluster sources.