Inhomogeneous low temperature epitaxial breakdown during Si overgrowth of GeSi quantum dots

Abstract

Low temperature epitaxial breakdown of inhomogeneously strained Si capping layers is investigated. By growing Si films on coherently strained GeSi quantum dot surfaces, we differentiate effects of surface roughness, strain, and growth orientation on the mechanism of epitaxial breakdown. Using atomic force microscopy and high resolution cross-sectional transmission electron microscopy we find that while local lattice strain up to 2% has a negligible effect, growth on higher-index facets such as {113} significantly reduces the local breakdown thickness. Nanoscale growth mound formation is observed above all facet orientations. Since diffusion lengths depend directly on the surface orientation, we relate the variation in epitaxial thickness to low temperature stability of specific growth facets and on the average size of kinetically limited growth mounds.

Addition of Mn to Ge quantum dot surfaces—interaction with the Ge QD {105} facet and the Ge(001) wetting layer

“…The protective function of Si capping layers exceeding a thickness of 1.5-2 nm for Mn-nanowires on Si(001) has been demonstrated recently [59], and δ-doping of Ge devices with P has also been achieved [34,35]. Conformal growth of Si on a Ge QD array has been demonstrated with an epitaxial breakdown thickness [60] exceeding by far the thickness required to achieve protective function and the different surface orientations are therefore not an obstacle. These observations strongly indicate that the Mn-surface islands can be preserved and trapped as δ-doped structures at the interface between the Ge QD and WL and a Si or Ge capping layer.…”

Epitaxial Nanotemplates for Strain Modulated Carrier Confinement in Silicon

“…To provide a detailed microscopic view of the defective epitaxial breakdown interface, we performed cross-sectional TEM and EELS analysis. For the 12 nm thick low temperature Si capping layer, for which RHEED indicates partial amorphization, we observe defect free epitaxial growth above the (001) wetting layer and the {105}-faceted pyramids [90], as shown in Figure 4 Finally, we observe two abrupt increases in angle of the h 1 interface above the dome: Bratland, et al [47], linked nucleation of the amorphous phase to the formation of growth mounds having a critical aspect ratio. The critical ratio was related to a peak-tovalley distance that is larger than the temperature-dependent diffusion length.…”

“…In this study, we use three different capping strategies: (1) all low temperature (300°C), (2) elevated temperature (400°C), and (3) two-step capping, 5 nm at 250°C followed by 45 nm at 500°C. Typically, low capping temperatures are chosen to ensure conformal Si growth [104], however this may be at the expense of crystalline quality.…”

“…Unlike the experiments reported by Guise et al[103], the Ge/SiC features were encapsulated with 50 nm Si grown at 250-300°C. The low capping temperature was chosen to ensure conformal Si growth whilst maintaining single crystalline epitaxy[104].High resolution, cross-section TEM (HRTEM) micrographs of two Si-capped, Ge/SiC/Si (001) nanodots are shown inFigure 5-10. These nanodots have basal diameters of 7 nm and 15 nm, respectively, and heights of 1.5 -2 nm.…”

“…To ensure conformal Ge overgrowth, it will be required to deposit Ge at, or near, room temperature. Fortunately, Ge epitaxy is known to be possible at these low temperatures[47,104]. All films investigated here have a Ge interlayer, which does not coat the SiC nanotemplates.…”