MODELING ALUMINUM CLUSTERS WITH AN EMPIRICAL MANY-BODY POTENTIAL

An empirical two-plus-three-body atomistic potential, derived by fitti ng experimental data pertaining to bulk aluminium, has been applied to study the structures and growth patterns of small aluminium clusters. The high dimensionality of the nuclear configuration space for cluste rs results in an extremely large number of isomers - local minima on t he potential energy hypersurface. Global optimisation (i.e. searching for the lowest energy structure) was carried out, using Random Search and Monte Carlo Simulated Annealing methods, for Al-2-Al-20. The resul ts of random searching have been used to put lower bounds on the numbe r of minima for these nuclearities and the efficiency of the Monte Car lo Simulated Annealing approach has been demonstrated. Detailed result s using both search methods are presented for Al-19. Comparisons are m ade with the results of previous calculations - using electronic struc ture and empirical potential methods and good agreement is generally o bserved. While many of the global minima correspond to structures (mos tly based on icosahedral growth) which are also global minima for Lenn ard-Jones or Morse clusters, a number of new structures have been iden tified for Al-N clusters - notably for N = 9, 16, 17, 18 and 20. (C) 1 998 Elsevier Science B.V. All rights reserved.