Structure and structural evolution of Ag_n (n = 3 - 22) clusters using a genetic algorithm and density functional theory method

Using a genetic algorithm followed by local optimization with density functional theory, the lowest-energy structures of Ag_n clusters in a size range of n = 3 - 22 were studied. The Ag_n (n = 9 - 16) clusters prefer compact structures of flat shape, while the Ag_n (n = 19, 21, 22) clusters adopt amorphous packing based on a 13-atom icosahedral core. For Ag₁₆, two competitive candidates for the lowest-energy structures, namely a hollow-cage structure and close-packed structures of flat shape, were found. Two competing candidates were found for Ag₁₇ and Ag₁₈: hollow-cage structures versus icosahedron-based compact structures. The lowest-energy structure of Ag₂₀ is a highly symmetric tetrahedron with T_d symmetry. These results are significantly different from those predicted in earlier works using empirical methods. The ionization potentials and electron affinities for the lowest-energy structures of Ag_n (n = 3 - 22) clusters were computed and compared with experimental values.