Electronic properties of the silver-silver chloride cluster interface.

The objective of this study was to gain insight into the electronic structure of silver-silver chloride cluster composites and especially into the metal-semiconductor interface. For this purpose a theoretical study of (AgCl)(n) (n=4, 32, 108, 192, and 256), of Ag(m) (m=1-9, 30, 115, 276, and 409), and of the cluster composites Ag(115)-(AgCl)(192) and Ag(409)-(AgCl)(192) has been carried out. Density of levels (DOL), local density of levels (l-DOL), and projection of surface states, as well as projection of properties of individual atoms or groups of atoms obtained in molecular orbital calculations, are shown to be powerful tools for gaining deep insight into the properties of these large systems. The Ag(115)-(AgCl)(192) aggregate, consisting of a cubic Ag(115) cluster without corner atoms on top of a cubic (AgCl)(192) cluster, was found to be remarkably stable with a cluster-to-cluster distance of about 280 pm, and a geometry in which the number of bonding interactions between the silver atoms of Ag(115) and the chloride ions of (AgCl)(192) is at its maximum. A sharp jump in charge distribution occurs at the Ag(115)-(AgCl)(192) composite interface. The first AgCl slab picks up negative charge from the two adjacent silver slabs, so that in total the silver cluster is positively charged. In addition, the core of the silver cluster is positively charged with respect to its outermost layer. The main reason for the charge transfer from the silver cluster to the silver chloride is the newly formed MIGS (metal induced gap states) in the energy-gap range of the silver chloride and the MIdS (metal induced d states) in the d-orbital region. Their wave functions mix with orbitals of the silver cluster and with both the orbitals of the silver and the chloride ions of the silver chloride. The MIGS and the MIdS are of a quite localized nature. In them, nearest neighbor interactions dominate, with the exception of close-lying silver chloride surface states-which mix in to a large extent. We conclude that especially the MIGS not only influence the photochemical properties of silver chloride, but that their existence might be probed by appropriate spectroscopic measurements.