COMPUTER OPTIMIZATION OF SMALL ATOMIC CLUSTERS

We have successfully identified stable configurations of both rare-gas and NaCl clusters with a new optimization procedure. An initial clust er configuration is prepared in a so-called shoot-and-stay process. It s total energy is then minimized with respect to the atomic coordinate s. To prevent the system from being locked in local minima, the step s ize of each move is chosen as the width of the energy well at a higher level. As the system evolves, the global minimum is contained in the volume bounded by the decreasing value of step sizes. We have also car ried out the optimization of NaCl clusters by the simulated annealing technique, for comparison. The results show that for such heterogeneou s systems, the latter method cannot always find the global minimum, be cause of large energy gaps between different catchment regions in phas e space.