Chemical nature of the passivation layer depending on the oxidizing agent in Gd(2)O(3)/GeO(2)/Ge stacks grown by molecular beam deposition

In Ge-based metal oxide semiconductor technology, the insertion of a passivation layer seems to be crucial in unpinning the Fermi level at the interface and in reducing the amount of interface defects. GeO(2) was obtained by atomic oxygen (AO), molecular oxygen or ozone chemisorption. Atomic or molecular oxygen was used in the deposition of Gd(2)O(3). Gd(2)O(3) thin films were grown by molecular beam deposition directly on (1 0 0) Ge or on a GeO(2) interlayer. The chemical nature of the Gd(2)O(3)/Ge interface was characterized by time-of-flight secondary ion mass spectrometry depth profiles. Without GeO(2) layer Gd and Ge interdiffusion is observed and the concomitant formation of Ge-O-Gd bonds is also supported by X-ray photoelectron spectroscopy energy shift at the Ge 3d peak and by a singularity in the interface defect energy distribution at similar to 0.48 eV. Further, depending on the GeO(2) formation process, the profile shape of Ge and O related secondary ions at the GeO(2)/Ge interface can be related with a defective Ge region close to the GeO(2)/Ge. In particular, considering the ratio between Ge and GeO(2) related secondary ion signals, the interlayer passivated using AO turns out to be comparatively enriched in Ge, while the use of ozone for GeO(2) formation leads to a Ge deficient layer.