Computational study of the surface properties of aluminum nanoparticles

We calculate the surface energy, surface stress, and lattice contraction of AI nanoparticles using ab initio density functional and empirical computational techniques. Ab initio calculations are carried out using the SIESTA pseudopotential method combined with the generalized gradient approximation. Empirical calculations are conducted using the embedded atom method. The ab initio density functional approach predicts the surface energies of Al nanoclusters to be in the range of 0.9-2.0 J/m(2). These values are consistent with the surface energy of bulk aluminum and are close to the surface energies of silver nanoparticles calculated in our previous study. In contrast to our previous results for Ag nanoparticles, we found a significant discrepancy between the theoretical values of surface energy and stress for Al nanoclusters. This result could be explained by a greater degree of surface reconstruction in Al clusters than in Ag clusters. (C) 2009 Elsevier B.V. All rights reserved.