Search all library resources

Core-shell germanium/germanium-tin nanowires

Germanium-tin is a promising material for novel devices for light sources and optical sensing in the mid-IR region. For sufficiently high Sn compositions, the material has a direct band-gap near 0.5 eV, and could have applications either as a detector or as an emitter. The main challenge to growth of high-quality single crystals is the large lattice mismatch of the system (~14% for diamond cubic Sn on Ge), the low equilibrium solubility of Sn in Ge (~1 at%), and compressive misfit strain counteracting the transition to a direct band-gap induced when growing thin films on Si (001) or Ge (001) substrates. We demonstrate that core-shell Ge/Ge1-xSnx nanowire structures take advantage of an elastically compliant, small-diameter nanowire substrate for high quality single crystal growth. Ge1-xSnx growth can be compatible with complementary metal-oxide semiconductor (CMOS) processing techniques. Therefore, there is potential for monolithic integration of Ge1-xSnx light sources onto silicon for photonic applications. In this thesis, we demonstrate growth, characterization, and optimization of core-shell Ge/Ge1-xSnx nanowire structures. First, growth of core-shell Ge/Ge0.96Sn0.04 using a low temperature chemical vapor deposition (CVD) process is demonstrated using typical industrial precursors, GeH4 and SnCl4. In contrast to Ge1-xSnx epitaxial thin films, free-standing nanowires deposited on misfitting Ge or Si substrates can avoid compressive elastic strains that inhibit formation of a direct gap. The Sn incorporation is several times greater than the equilibrium solubility limit of Sn in Ge. Structural characterization by lab source x-ray diffraction and transmission electron microscopy are performed along with optical characterization by photoluminescence. Second, core-shell Ge/Ge0.96Sn0.04 nanowires are cross-sectioned and characterized extensively to analyze the interplay between the composition distribution and the core-shell strain and how optical properties are affected. The nanowire cross section reveals six Sn-poor radial spokes approximately 60 degrees apart in the Ge1-xSnx shell. Phase field simulations provide an estimate of the expected strain, which matches well with experimental results and explain the stability of the Sn-poor spokes by considering their effect on the elastic strain energy. There is a two-fold synergistic effect on the optical properties produced by the core-shell nanowire geometry: the Ge core acts as an elastically compliant substrate for growth of an axially lattice-matched epitaxial Ge1-xSnxshell, which facilitates growth of high-quality single crystal Ge1-xSnx having intense photoluminescence; at the same time, the tensile misfit strain in the Ge core serves to decrease its direct gap transition energy with respect to the indirect gap transition energy, thus enhancing its optical emission. Finally, we examine the parameters affecting Ge1-xSnx shell growth and optimize the CVD parameter space for high Sn incorporation in the shell. Ge1-xSnx CVD chemistry with GeH4 and SnCl4 precursors undergoes a transition from growth at higher temperatures to etching at lower temperatures, in keeping with the entropies of the respective chemical reactions. Also, Sn composition increases when temperature is decreased because Sn at the growth front can be kinetically trapped into the Ge1-xSnx shell. We demonstrate that the degree of axial and radial growth of Ge/Ge1-xSnx core-shell nanowire heterostructures can be controlled by varying precursor to H2 partial pressure ratio during CVD growth, with the SnCl4:GeH4 partial pressure ratio fixed. By increasing the SnCl4 partial pressure, radial growth rate increases and axial growth rate decreases. This is consistent with SnCl4 disruption of H-passivation of Ge sidewall facets. Examining shell thickness variation with shell growth time showed slow initial growth that approached a constant volumetric growth rate, which is consistent with precursor mass transport-limited shell growth. Controlling nanowire density by the vapor-liquid-solid (VLS) catalyst loading per unit substrate area as a means to probe the effect of SnCl4 mass transport on Ge1-xSnx shell growth, we found that very sparse nanowires tend to be decorated by Sn precipitates while very dense nanowires exhibited bending induced by Sn composition variation across the wire circumference caused by local SnCl4 depletion. Thus, mass transport of SnCl4 plays an important role in the Ge1-xSnx shell growth. It is likely that a balance between Sn precursor flux and the available surface sites for Sn incorporation is required to prevent these undesirable effects. We are also able to achieve different Sn compositions up to 14 at% by varying SnCl4 partial pressure and growth temperature. With control over geometry, morphology, and composition of Ge1-xSnx heterostructures, a wide range of potential device architectures can be achieved.