Answers are grounded in this paper and cite their sources.
We have studied the structural properties of tensile-strained Si layers grown on polished Si0.6Ge0.4 and Si0.5Ge0.5 virtual substrates as a function of their thickness. Two gaseous precursor chemistries have been assessed for the reduced pressure?chemical vapour deposition of the sSi layers: SiH2Cl2 at 700 ?C and SiH4 at 600 ?C. We have used specular x-ray reflectivity and spectroscopic ellipsometry to gain access to the sSi layer thickness (and the associated sSi growth rate). The surfaces of sSi layers grown at 600 ?C using SiH4 are characterized by a small spatial wavelength (a few hundred nm) roughness. Meanwhile, some lines along the 1?1?0 directions can be observed for thick sSi layers grown at 700 ?C using SiH2Cl2, hinting at the presence of stacking faults. We obtained (for 10 ?m ? 10 ?m atomic force microscopy images) surface root-mean-square roughness (Z ranges) between 0.19 and 0.36 nm (1.8 and 3.9 nm). By comparison, the rms roughnesses (the Z range) associated with 360 ?m ? 368 ?m optical interferometry images are between 0.7 and 1.4 nm (7.1 and 12.1 nm), with some small amplitude but very long spatial wavelength (tens of ?m) surface cross-hatch remaining on Si0.6Ge0.4 VS. The interfaces between sSi and SiGe are very abrupt, as illustrated by high-resolution transmission electron microscopy and by the Ge decay profile in secondary ions mass spectrometry: 0.73 nm/decade for sSi on Si0.5Ge0.5 VS and 1.06 nm/decade for sSi on Si0.6Ge0.4 VS, more or less independently of the sSi growth chemistry. The larger value for sSi on Si0.6Ge0.4 VS is most probably due to some instrumental broadening linked to the small remaining cross-hatch.
No takes yet. Share an insight, caveat, or question.
Hartmann et al. (2007) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: