Vielteilcheneffekte in der elektronischen Struktur von Adsorbaten auf Halbleiteroberflächen
Final Report Abstract
This theory project was centered around studies of self-organized nanostructures formed by adsorbates at semiconductor surfaces. These systems offer a unique playground for controlled investigations of a series of important physical phenomena in two and one dimensions (nanowires at surfaces), such as, e.g., metal-insulator transition, onset of charge-density waves, spin effects, to mention only a few. These phenomena are interesting from both, fundamental and technological points of view. Our approach was based on a joint application of the density functional theory and modern many-body methods, which allow a detailed insight into the electronic correlations and thermodynamic phases in these systems. A testcase for our approach was the Sn/Si(111)-(√3 x √3)R30° surface. Using many-body approaches on top of the DFT results, we could confirm the existence of a metal-insulator transition at this surface in accordance with the existing experimental evidence. The complementarity of the used approaches (dynamical mean-field theory, variational cluster approach, dual-fermion method) allowed a detailed characterization of the properties of this system. It was shown that the Sn/Si(111) surface is dominated by short-range correlations and the second order metal-insulator transition takes place for Hubbard-U parameter close to 0.65 eV. The calculated spin susceptibility, as a function of the interaction strength U, points to a more complex magnetic order of this system, than the previously much-discussed 120°-AFM state. We elaborated on the important role of the next nearest-neighbor electronic hoppings for the density of states and the magnetic order. Other systems studied in detail were monolayers of noble metals on the (111) surfaces of silicon and germanium. Most of our work there was devoted to the Au/Ge(111) surface. These studies have been done in close cooperation with the experimental group of Claessen/Schäfer and the comparison between theoretical and experimental results contributed to a significant improvement of our methodology. It turned out, for example, that the energy position of the Au-5d states is important for the correct picture of surface bands. Since their energy is underestimated by the standard local-density approximation, as well as by semi-local corrections to LDA in form of the GGA approximations, we had to go beyond these methods. It turned out that a significant improvement of the agreement with experiment can be achieved including the self-interaction corrections (SIC) for the Au-5d levels. The Au/Ge(111) surface is one (of a few known to date) of metallic surfaces at a semiconductor substrate which displays very large spin splittings of the metallic surface bands. This is an interesting and technologically relevant property. In the theoretical and experimental studies, the Fermi surfaces have been characterized with special emphasis on their spin character. One of our findings is that, although the spins at constant-energy cuts show a characteristic (Rashba) vortex behavior, a standard Rashba scenario is not prevailing in this system (for example, a sizeable perpendicular spin component is also present which alternates its direction). The above studies are currently continued and are also extended to other systems.
Publications
- Electronic band structure of the twodimensional metallic electron system Au/Ge(111). Phys. Rev. B 83, 235435 (2011)
P. Höpfner, J. Schäfer, A. Fleszar, S. Meyer, C. Blumenstein, T. Schramm, M. Heßmann, X. Cui, L. Patthey, W. Hanke, and R. Claessen
- Geometrical frustration and the competing phases of the Sn/Si(111)√3 × √3R30° surface systems. Phys. Rev. B 83, 041104(R) (2011)
G. Li, M. Laubach, A. Fleszar, W. Hanke
