SFB 616


Pictures SFB616

 Project B7:
Ab inito theory of elementary excitations at surfaces

 In project B7, we work towards the goal of developing a quantitative and material-specific theory of energy dissipation at surfaces. This can be achieved by an ab initio approach, starting from the electronic structure of matter, and using only the chemical identity of the atomic species as input. Following this route, we encounter the difficulty that the dissipation processes occur on time- and energy scales differing considerably from those of the overall electronic structure.

 This can be overcome by treating the quantum dynamics of the adsorbates by a separate Hamiltonian describing its coupling to bosonic quasiparticles, with the coupling strengths determined from the ab initio calculations. The bosonic quasiparticles are the elementray excitations of the surface, i.e., electron-hole pair excitations and phonons. While both of them are important for metals surfaces, only the latter play a role for adsorbates on semiconductors.

Carbon monoxide molecules bind with the carbon atom (black) to the silicon surface (orange atoms), with the oxygen atom (red) sticking out. The bond between C and Si is much weaker and softer than the C-O bond, making it difficult for the C-O stretching mode to couple to other vibrational modes.

As an example of energy dissipation at semiconductor surfaces, we perform calculations of the "energy landscape" probed by a CO (cabonmonoxide) molecule adsorbed on silicon. The interest in this system is driven by laser-induced desorption experiments in project B4. In particular, we aim at identifying the dissipation processes that determine the lifetime of the internal stretching mode of the CO molecule. The energy transfer to other modes of molecular vibration (accessible to the diffraction experiments planned in B2), and to surface phonons will be investigated.


 For energy dissipation on metal surfaces, the additional dissipation mechanism of excitation of electron-hole pairs can be studied experimentally by measuring the chemicurrent associated with a chemical reaction on the surface (see project A1). In particular, the violent reactions of halogens, like chlorine, with alkali metals such as potassium produce strong chemicurrents. The energy of both the electrons and holes produced by the chemical reaction are characterized by an energy distribution which is determined mainly by the speed of the reactive particles and by the electronic band structure of the metal surface. In order to predict the energy distribution, we need to take into account both the posibility of multiple excitations and the details of the electronic structure near the Fermi level of the metal.

Left: Electronic band structure of a silver film. Bands below zero energy are occupied by electrons.
Right: Zoom into the region around the Fermi level, taken to be the zero of energy.

 Excitations of electron-hole pairs is possible in several ways, indicated by the arrows, resulting in a different enery distribution to be expected for the electrons and holes.

Eventually, quantum-mechanical correlation effects due to the electrostatic interaction between the electron and the hole will be considered by calculating the electrodynamic susceptitibilty of the metal surface.

Summarizing the goals of this project, theory will enable us to have a glance at the very fast dynamics of molecules on surfaces. Improved understanding of these fast processes could be helpful for detecting chemical reactions electronically, and for steering chemcial reactions at a surface by laser pulses.


Internationaler Workshop 2008

Workshop 2008

B7 Sakong et al.
PDF (2.9 MB)

Begehung 2008

Workshop 2008

B7 Timmer et al.
 PDF (0.9 MB)

Begehung 2005

SFB-Workshop 2007
B7Bücking et al.
 PDF (2.3 MB)