Ab initio theory of electronic excitations at surfaces
During the last years there has been constant progress in the area of non-adiabatic effects, especially at surfaces. Experimental results propose a measurable electronic excitation during many processes involving a metal surface. Theoretical results support these measurements. The present thesis investigates these excitations. A theory is developed, which is aimed to provide an understanding of electronically nonadiabatic processes, especially at a metal surface, using the methods of time-dependent perturbation theory in combination with density functional calculations. The developed theory is then applied to the case of gas particle adsorption on Schottky diodes. The model systems of hydrogen adsorption on aluminum and potassium, as well as the case of magnesium epitaxy and adsorption of a chlorine molecule on a potassium surface, are investigated. Special attention is paid to the so-called isotope effect, where particles of different mass lead to different excitation spectra of electrons and holes. In the chapter treating hydrogen adsorption on aluminum several alternative theories and extensions of the perturbative approach are investigated. Special attention is paid to the so-called spin transition, where the spin polarization of the adsorbing hydrogen atom vanishes when it approaches the surface. For hydrogen adsorption on potassium it is found that electronic excitations are more important than was believed by many researchers up to now. It is also found that the density of states may influence the excitation spectra. The case of magnesium epitaxy deserves special attention, since theoretical results contradict simple expectations, when it is found that hot electrons and holes in notable quantities are excited. In the case of chlorine adsorption on potassium the range of validity of the perturbative approach is left. However, important conclusions may be drawn from calculations of a sample adsorption trajectory. It is found that the perturbative approach is able to describe excitations of electrons and holes, and describes the isotope effect within experimental error bars. Limits for the range of validity of the perturbative approach are also found. Generally, further work on the topic of electronic excitations is required, but the present work leads to a better understanding of electronically nonadiabatic effects. It is able to shed light on the topic of the so-called spin transition. Methods and conclusions of previous approaches are challenged.