In a study of (100)Si growth by molecular-beam epitaxy, we have observed an interesting phenomenon associated with the breakdown of (100)-oriented growth at low temperatures (<∼400 °C). Electron microscope imaging revealed that the (100) surface no longer remains planar but develops a series of cusps with {111} oriented facets. At the base of each cusp a small (∼3 nm) cylindrically shaped defect region is formed, but this is morphologically unstable and breaks down to form a linear array of spherical defects trailing behind the growing interface, aligned parallel to the growth direction. The similarities observed between these structures and those found in the unidirectional solidification of rod eutectoids or monotectoids are striking, although the scale of the microstructure in the present case is much smaller. From a series of electron microscopy experiments using both cross-sectional and plan-view specimen geometries it has been found that the cylindrical–spherical defects are microvoid regions. Furthermore, the microvoid defect structure was found to be stable as the epitaxial Si matrix transformed to the polycrystalline and amorphous states beyond an epitaxial critical thickness.
No takes yet. Share an insight, caveat, or question.
Perovic et al. (1991) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: