Gast ::
| Hauptseite > Publikationsdatenbank > Struktursymmetrie und Phononen an adsorbatbedeckten Nickel(110)- und reinen Gold(110)- Oberflächen |
| Hauptseite > Publikationsdatenbank > Struktursymmetrie und Phononen an adsorbatbedeckten Nickel(110)- und reinen Gold(110)- Oberflächen |
| Book/Report | FZJ-2018-03208 |
1989
Kernforschungsanlage Jülich, Verlag
Jülich
Please use a persistent id in citations: http://hdl.handle.net/2128/18703
Report No.: Juel-2334
Abstract: High resolution electron energy loss spectroscopy (HREELS) has been used to study the structure and bonding at different metal surfaces. The symmetry of a surface can be determined by a combination of momentumresolved vibrational spectroscopy along the high symmetry directions and the symmetry selection rules for electron scattering. On the unreconstructed hydrogen covered Ni(110)-surface ($\theta_{H}$ $\le$ 1) the hydrogen atoms are adsorbed in a threefold adsorption site with a mirror plane along [001]-direction. The dispersion data are used to determine the polarization of the hydrogen modes. On the reconstructed (1x2)-H surface ($\theta_{H}$ = 1.5) there are two different hydrogen species which reside in different threefold sites. One threefold site is like the one on the unreconstructed surface while the other site is between adjacent Ni rows of the reconstructed surface and has its symmetry plane along [$\overline{1}$10]-direction. The hydrogen induced "pairingrow" reconstruction of the Ni(110)-surface is preceeded by a softening of nickel surface modes at the $\overline{Y}$-point of the surface Brillouin zone. This indicates a change of interactions in y-direction which is the direction of the reconstruction. On the reconstructed surface a rather dispersionless mode is measured above the Rayleigh phonon. Such a mode is also observed on the other reconstructed surfaces studied in this work and can be explained by the backfolding of a phonon present on the reconstructed Ni(110) surface. The Ni(110)-(2x1) CO structure is particulary interesting because of the dense packing of this overlayer. A systematic analysis of the CO-stretch vibration, the CO-metal stretch vibration and the "frustrated" translation mode leads in connection with cluster calculations to an (bent) on-top adsorption site for the CO molecule. For the CO-stretch vibration a Davydov splitting is observed: splitting of the inphase and the out-of-phase vibration of the two CO molecules in the unit cell. The dispersion of the CO-stretch vibration can be described by a simple model of electrostatic dipole-dipole interactions and is sensitive to a structural parameter of the CO-adsorption, namely the lateral displacement from the on-top position along the [001]-direction. On the oxygen reconstructed Ni(110) surface isotope exchange experiments are used to discriminate the oxygen modes which lie partially inside the bulk band of nickel. An azimuth resolved symmetry analysis results in a mirror plane along the [$\overline{1}$10] direction. In combination with a recent dynamical LEED analysis a modified "missing-row" model is obtained. The substrate is reconstructed according to the "missing-row" model, however the oxygen atoms are displaced from the long bridge positions in [$\overline{1}$10] and [$\overline{1}$10] directions, respectively. The dispersion of the surface phonons on the (1x2)- econstructed Au(110)-surface is measured along the high symmetry directions $\overline{\Gamma X}$ and $\overline{\Gamma Y}$. These measurements allow a decision between the contrary predictions of the semiempirical "glue"-model on the one hand and "ab-initio" calculations on the other hand. The measured surface phonons are in good agreement with the "ab-initio" calculations, however the high frequency modes above the bulk phonon band which were predicted by the "glue" model can be excluded experimentally. The "glue"-model seems to overestimate the bonding-strength of the atoms near the surface.
|
The record appears in these collections: |
PDF
PDF (PDFA)