Quantitative Study of Sub-Monolayer growth of Ge(001) Homoepitaxy using Reflection High Energy Electron Diffraction

Abstract

Format

This is a copy of the slides presented at the meeting but not formally written up for the volume.

Abstract

Intensity oscillations of a specularly reflected spot in Reflection High Energy Electron Diffraction (RHEED) contain morphological information about the growth surface and a great deal of film growth kinetics can be learned if it is properly interpreted. However, unlike in x-ray diffraction techniques, the dynamical scattering nature of electron beams, such as the formation of Kikuchi features, have prevented a quantitative analysis of RHEED intensity variations. We have previous demonstrated that the interference of Kikuchi features can be minimized at low incidence angles, where the phase shift of intensity oscillations is absent. Under these diffraction conditions, the specular spot is not influenced by Kikuchi features and can be interpreted with a rather simple model without invoking complicated diffraction calculations. Here we report the study of sub-monolayer (ML) growth of Ge(001) homoepitaxy by molecular beam epitaxy at low temperatures, 100 oC ~ 150 oC, using RHEED intensity oscillations obtained for a range of low incidence angles where the phase of the intensity oscillations does not change. We present a new model for RHEED intensity that includes the diffuse scattering off steps on the surface and the layer interference between terraces of different heights using the kinematic approximation. The model describes very well the measured RHEED intensity oscillations for the entire range of incidence angles used in this work. We find that the strength of diffuse scattering from steps increases with decreasing incidence angles. We determine the evolution of individual layer coverage which shows that at the temperatures investigated, second layer nucleation takes place in the relatively early stage of film growth. It is also shown that the intensity minimum occurs well above 0.5 monolayers of total deposited coverage. Inferring kinetic parameters, such as the step-edge barrier, from the evolution of layer coverage will be also discussed.