Solute Trapping in Low-Temperature Vapor-Liquid-Solid Growth: A Route to Direct-Gap Ge-Sn Single Crystal Nanowires

Stanford University, Stanford CA

Investigators

Abstract

Non-technical Description: Growing bulk crystals by freezing them from a liquid is a well-established method for making highly perfect semiconductor materials for a host of applications including electronics, and the detection or emission of light. A crystal growth process that is used to produce very small (tens to hundreds of nanometers) diameter semiconductor wires is investigated in this project. The focus is on growth of germanium-tin wires that have the potential to enable exciting devices, such as lasers that can be grown directly on silicon. The principal investigator and his graduate students are studying how unusually large concentrations of tin atoms can be frozen into these semiconductor wires as they grow. This project also integrates undergraduates in the experimental effort by involving research-experiences-for-undergraduates (REU) students in all three years of planned activities. In parallel with their research experiences, the REU students work with the principal investigator to develop simple laboratory demonstrations and presentation materials on semiconductors, light-solid interactions and crystal growth. Finally, this research effort is used as a vehicle for science outreach to low income San Francisco Bay Area high school students, through participation in Stanford RISE summer internship program. Technical Description: This project extends to the nanoscale the study of solute trapping as a means of achieving metastably high semiconductor alloy compositions. In so doing, it exploits the large liquid supersaturations and deep subeutectic temperatures possible with vapor-liquid-solid (VLS) Ge nanowire growth. The project focusses on growth of free-standing Ge-Sn binary alloy nanowires, which are of great interest for their high carrier mobilities and the possibility of achieving a direct band gap for very efficient light absorption and emission. Research on epitaxial thin films suggests that Sn alloying of diamond cubic Ge can decrease the energy of the Gamma valley of the conduction band relative to the L valley. In practice, Sn alloying up to the predicted > 6 at.% composition required to achieve a direct gap in an unstrained thin film is difficult. The equilibrium solubility limit of Sn in bulk Ge is < 1 at.%. Moreover, the lattice mismatch of Ge-Sn to Ge and Si induces compressive biaxial strain in films deposited on these most useful substrates that shifts the sub-band energies in opposition to the effect of Sn addition, thus requiring even larger Sn concentrations. These considerations make clear the potential scientific and technological impact of developing methods to produce free-standing, single crystal Ge-Sn nanowires without coherency strains. A comprehensive experimental approach is used to probe solute trapping in Ge-Sn nanowires grown from either Au or Sn nanoscale catalysts, including energy dispersive spectroscopy in the transmission electron microscope, and local electrode atom probe tomography. In parallel with these studies of VLS solute trapping kinetics and mechanisms, optical probes such as photoluminescence and transient absorption are used to study the effects of Sn incorporation on alloy band structure and carrier recombination dynamics.