No abstract

Growth rates and compositions are reported for GeSi alloy films and superlattices epitaxially grown on both Ge(100) and Si(100) substrates using disilane and germane source gases in an ultrahigh vacuum chemical vapor deposition chamber. Although the growth rate changes rapidly with temperature the composition is nearly independent of it. Specifically, we find that the order of the adsorption reaction for disilane and germane is the same, resulting in the composition being determined by the partial pressures and by the ratio of the adsorption reaction rate constants. This ratio depends very weakly on temperature, if at all, and appears to vary slightly with the layer composition.

We have fabricated strained SixGe1−x/SiyGe1−y multiple quantum wells on Ge(100) substrates and measured the photoluminescence (PL) spectra, observing band-edge emission from the SiGe alloy layers. The emission is due to the recombination of both bound excitons and free excitons in the quantum wells. From the positions of the observed PL lines, we have evaluated the band-gap energies of the strained SiGe alloy layers, and found them to be smaller than those of bulk SiGe alloys. The band-gap energy increases with the Si content of the alloy, reaching a maximum at about 15% Si, and subsequently decreases. These results agree well with the theoretical calculations for strained layers, and suggest a type II band alignment in some cases for SixGe1−x/SiyGe1−y heterostructures on Ge(100).

No abstract

By combining the complementary techniques of x-ray diffraction and transmission electron microscopy we have been able to accurately determine the structure of imperfect GeSi superlattices (SL’s). The samples were epitaxially grown on Ge(001) substrates using Si2H6 and GeH4 source gases. In this report, details of the x-ray experiment and analysis are emphasized. In particular, a model is presented for calculating the diffracted intensity from a SL containing gradients in composition and thickness. Applying this model to the data it is found that between the first and last layers of each SL there exists a roughly 10% increase in the growth rate, without a corresponding change in the alloy composition. This is attributed to a slow increase in the substrate temperature, Tsub, of just a few degrees. A sample grown with a corresponding gradual decrease in the control temperature, TC, was found to be much more uniform.