Dresden 2017 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
O: Fachverband Oberflächenphysik
O 53: Nanostructures at Surfaces: Other Aspects
O 53.12: Poster
Tuesday, March 21, 2017, 18:30–20:30, P1C
Considering the Convergent Beam in a Quantitative LEED Analysis — •Tilman Kisslinger, Pascal Ferstl, M. Alexander Schneider, and Lutz Hammer — Lehrstuhl für Festkörperphysik, Universität Erlangen–Nürnberg, D–91058 Erlangen, Germany
In a conventional LEED experiment the primary electron beam is not parallel, but slightly convergent (opening angle typically 2 × 0.6∘), in order to focus the beam on the screen. Thus, experimental LEED intensity spectra are in fact a superposition of spectra belonging to all angles within that cone of incidence, therefore, model calculations performed for the central cone angle are just an approximation of the experimental situation.
For nominally normal incidence of the primary beam, which is the common experimental alignment, as it preserves a maximum of symmetry elements of the surface, we performed a pseudo–experiment, in which for a Rh(100)–(2×2)–O surface intensity spectra were calculated for various angles within a cone of Θmax = 0.75∘, weighted according to their corresponding solid angle and averaged. A comparison with the set of spectra calculated for normal incidence yields a RP of 0.036. It is further shown that using an average cone angle Θav = Θmax/√2 instead fits the pseudo-experiment with high accuracy (RP = 0.0003). Fitting the (average) angle of incidence for real experimental data of various and structurally very different systems taken nominally at normal incidence always led to an improvement of the R-factor by about 0.02 - 0.03. Consistently, all fits result in the same value for Θav, because it is solely determined by the experimental geometry.