This paper describes preparation of a new class of Ge 1−x−y Si x Sn y direct-gap semiconductors grown on Ge-buffered Si substrates via depositions of trigermane (Ge 3 H 8 ), tetragermane (Ge 4 H 0 ), tetrasilane (Si 4 H 10 ), and stannane (SnD 4 ) hydride precursors. These react at ultralow temperatures 320−290 °C to produce thick (∼500 nm) monocrystalline films with concentrations closely reflecting the gas phase molar ratio of the coreactants. A series of Ge-rich samples with a fixed 3−4% Si content and progressively increasing Sn content in the 4−10% range are grown to explore the possibility of obtaining direct gap materials for the first time in this semiconductor system. The resultant films also exhibit residual compressive strains that are largely relaxed via rapid thermal annealing between 550 and 700 °C depending on composition. These temperatures are 50−75 °C above the thermal decomposition threshold of Ge 1−y Sn y /Ge materials with same Sn content, indicating that the ternaries are significantly more robust than the binary analogues. The compositional dependence of the strain relaxations is similar to that found for Ge 1−x Si x alloys grown directly on Si. Cross hatch patterns are formed on the surface of these materials as a consequence of interface misfit dislocations generated due to the mismatch with the underlying Ge buffer. The annealed films exhibit low defect densities and atomic scale chemical uniformity as indicated by STEM and element-selective EELS mapping, allowing a meaningful study of the optical properties as a function of Sn concentration. The results show that the separation of the direct and indirect edges can be made smaller than in Ge even for the non-negligible 3−4% Si content, confirming that with a suitable choice of Sn compositions the ternary Ge 1−x−y Si x Sn y reproduces all features of the electronic structure of the binary Ge 1−y Sn y , including the sought after indirect-to-direct gap cross over. The alloys produced in this study represent an attractive alternative to Ge 1−y Sn y for applications in IR optoelectronic technologies requiring high thermal budget processing and harsh operating conditions due to their enhanced thermal stability conferred by the incorporation of Si in the diamond-like structure.

No abstract

Crystalline Al1–x B x PSi3 alloys (x = 0.04–0.06) are grown lattice-matched on Si(100) substrates by reactions of P­(SiH3)3 and Al­(BH4)3 using low pressure CVD. The materials have been characterized for structure, composition, phase purity, and optical response by spectroscopic ellipsometry, high-resolution X-ray diffraction, high-resolution transmission electron microscopy, electron energy loss spectroscopy, and energy dispersive spectroscopy, which indicate the formation of single phase monocrystalline layers with tetrahedral structures based on AlPSi3 parent phase. The latter comprises interlinked AlPSi3 tetrahedra forming a cubic lattice in which the Al–P pairs are imbedded within a diamond-structured Si matrix as isolated units. Raman scattering of the Al1–x B x PSi3 films supports the presence of substitutional B in place of Al and provides strong evidence that the boron is bonded to P in the form of isolated pairs, as expected on the basis of the AlPSi3 prototype. The substitution of small size B atoms is facilitated by the stabilizing effect of the parent lattice, and it is highly desirable for promoting full lattice matching with Si as required for Si-based solar cell designs. The substitution of B also increases the bond-length disorder leading to a significantly enhanced absorption relative to crystalline Si and AlPSi3 at E < 3.3 eV which may be beneficial for PV applications. Analogous reactions of As­(SiH3)3 with Al­(BH4)3 produce Al1–x B x AsSi3 crystals in which the B incorporation is limited to doping concentrations at 1020 atoms/cm3. In both cases the classical Al­(BH4)3 acts as an efficient delivery source of elemental Al to create crystalline group IV–III–V hybrid materials comprising light, earth abundant elements with possible application in the fields of Si-based technologies and light-element refractory solids.

We report a new doping protocol of pure Ge films grown on Si and related Si/ Sn materials based on S delivered from high reactivity hydride molecules S(MH 3 ) 2 (M = Si,Ge). The new doping strategy targets next generation semiconductor applications requiring enhanced IR optical performance as well as high-mobility field effect transistors fully integrated with silicon. To explore this paradigm, we first developed a practical and straightforward synthesis approach, which avoids the use of toxic starting materials and yields viable quantities of the title compounds. These were then used to carry out proof-of-concept low-temperature depositions of Ge/Si(100) and GeSn/Si(100) films, doped with "double donor" S atoms for the first time. These systems are characterized using standard materials science techniques via RBS, XTEM, SIMS, and XRD for structure, composition, and crystallinity, and their electrical properties are measured by the Hall method. Thermally robust dopant levels <10 18 are systematically obtained using a range of process protocols, which can be further optimized for practical applications. Complementary first-principles density functional theory simulations were then used to study the stability and strain associated with the incorporation of S within the parent lattice as molecular fragments such as Ge−S−Ge, Ge−S, Si−S−Si, and Si−S derived from the S(SiH 3 ) 2 and S(GeH 3 ) 2 molecular sources. Completely incorporated Ge−S−Ge or Ge−S units in Ge in which S resides in either substitutional (tetrahedral) or near-substitutional (3-coordinate) sites are predicted to be strongly bound (−2.36 eV and −1.49 eV, respectively) relative to interstitial S and isolated Ge vacancies, while the corresponding binding in Si− S−Si and Si−S analogs is slightly enhanced. By considering the induced lattice strain, binding energy, and the structural accommodation of molecular core bonds, plausible defect-clusters are identified and tentatively used to correlate sulfur concentrations and carrier concentrations trends. In the case of the S(GeH 3 ) 2 grown films, a mixture of deep and shallow donor centers must be invoked to account for the observed carrier concentration enhancement.