AVS 45th International Symposium
    Surface Science Division Friday Sessions
       Session SS1-FrM

Paper SS1-FrM10
Xe Adsorption Sites on Metal Surfaces

Friday, November 6, 1998, 11:20 am, Room 308

Session: Surface Structure and Strain
Presenter: M. Caragiu, Pennsylvania State University
Authors: M. Caragiu, Pennsylvania State University
Th. Seyller, Pennsylvania State University
R.D. Diehl, Pennsylvania State University
P. Kaukasoina, Tampere University of Technology, Finland
M. Lindroos, Tampere University of Technology, Finland
Correspondent: Click to Email
Based on an adsorbate-substrate potential which consists of the attractive van der Waals interaction and a repulsive interaction due to wave function overlap, the equilibrium site for physisorbed noble gases would be expected to be a high-coordination site. The presumption which arose from this expectation, that physisorbed atoms prefer high-coordination sites, has been a great hindrance to the development of physisorption potentials since it has delayed experiments to measure the adsorption geometries of physisorbed atoms. Several years ago, a top-site geometry was proposed for Xe/Pt(111) based on He-atom diffraction intensities from an incommensurate phase, although this assignment was disputed by a later SPLEED study. Nevertheless, a density-functional theory cluster calculation suggested that a preference for top sites in this case may arise from the hybridization of the Xe 5p electrons with the Pt 6d states. Recently it has been shown that Xe on Ru(0001) adsorbs in the top-site geometry in the (@sr@3x@sr@3)R30° phase. If hybridization with substrate d-states is the origin for top-site adsorption, then Xe would not be expected to occupy the top sites on Cu surfaces where the d-states are several eV below the Fermi energy. We present the findings of LEED I(E) studies of Cu(111)-(@sr@3x@sr@3)R30° and Pt(111)-(@sr@3x@sr@3)R30° which were carried out to test this hypothesis and to resolve the disagreement on the adsorption site for Xe on Pt(111). These LEED studies indicate that Xe occupies the top site in both cases. Since it is unlikely that Xe atoms hybridize appreciably with the deep d-levels in Cu(111), we propose a new model for Xe adsorption on metal surfaces in which the hybridization occurs between the occupied part of the excited Xe 6s resonance (which extends below the substrate Fermi energy) and the unoccupied substrate orbitals near the Fermi level.