Answers are grounded in this paper and cite their sources.
We have investigated the elastic strain relaxation in Si 1-x Ge x layers grown by the molecular beam epitaxy (MBE) technique and in situ controlled with RHEED. Up to ≈0.8% critical lattice mismatch (about 20% Ge) uniform strained and flat layers were grown both on (111) and on (001) Si substrates. Calculations of the elastic constants evidenced a tetragonal distortion about 50% higher on (001) than on (111) in the same experimental conditions. At higher misfits (and/or thicknesses) a growth instability was evidenced only on (001) Si substrates. Si 1-x Ge x layers there displayed a surface layer undulation. On the contrary, Si 1-x Ge x layers grown on (111) Si substrates remained smooth throughout the growth up to the plastic relaxation of the layers. To determine stress fields in the Si 1-x Ge x layers, a high spatial resolution convergent beam electron diffraction (CBED) experiment was performed with a field effect analytical microscope. The CBED technique was applied to two typical cases: totally strained layer and undulated dislocation-free layer. In the latter case, CBED patterns recorded on nanometer scale areas of an undulation crest (cross-section sample) showed a gradual elastic relaxation mainly directed along the growth axis (z). Moreover a triclinic distortion of the unit cell was pointed out. These results were confirmed on a plane view sample. In conclusion, our results show that the driving force for the undulation is not the in-plane elastic relaxation since CBED experiments proved an important elastic relaxation of the (001) Si 1-x Ge x layers along the z axis. This was in agreement with the calculations of the elastic constants. We think that this could be at the origin of the undulation.
No takes yet. Share an insight, caveat, or question.
Berbézier et al. (1998) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: