ALBERT

Notes: Infrared absorption and photocurrent measurements have been applied to study the photoresponse below the band gap of indium gallium arsenide (In0.53Ga0.47As) grown lattice matched to Fe-doped semi-insulating indium phosphide (InP) substrates by various epitaxial growth techniques, including molecular beam epitaxy, liquid phase epitaxy, and metalorganic chemical vapor deposition. It is found that Fe at the InGaAs/InP interface is responsible for exciton-like and polarization sensitive absorption peaks. Both electron and hole emission into the conduction and valence bands, respectively, were observed, and a deep Fe level was identified 0.37 eV below the conduction band edge of bulk Fe:InGaAs. Lowering of the local crystal symmetry due to the interface electric field is proposed to be the mechanism that describes the dipole-allowed interband absorption of 3d transition metal impurities in narrow band gap III-V compounds like the Fe2+:InGaAs used in this study. The ambiguity in distinguishing InGaAs quantum well intersubband absorption signals from the Fe interband absorption signals is also addressed.

Notes: Infrared photoluminescence has been used to study the band-gap energy of InAs1−xSbx digital superlattices and band alignment of InAs1−xSbx/AlSb quantum wells at 5 K. It is found that the InAs1−xSbx digital alloys have a smaller effective band gap than InAs1−xSbx random alloys. In addition, the valence band offset between type-II InAs/AlSb is determined to be 130 meV. This number reduces as the Sb mole fraction in InAs1−xSbx is increased, and the alignment between InAs1−xSbx/AlSb becomes type I when x〉0.15. © 2000 American Institute of Physics.

Notes: Peak current densities of InAs/AlSb/GaSb/AlSb/InAs resonant interband tunneling diodes (RITD) grown by molecular beam epitaxy have been measured as a function of the growth temperature. The growth procedures were designed to produce nominally identical AlSb tunneling barriers. The variations observed in the peak current for positive bias are consistent with the barrier on the substrate side of the RITD becoming effectively thicker for diodes grown at high temperatures. Plan-view in situ scanning tunneling microscopy (STM) measurements indicate that smoother AlSb barriers are grown at high temperature. The growth temperature dependence of the peak current density and STM results are consistent, because tunneling is highly dependent on barrier thickness. While the high and low temperature growths were designed to have the same barrier thickness, the large current flowing through the thin areas of a rough barrier result in an effectively thinner barrier compared to the smooth one. © 2001 American Institute of Physics.

Notes: The surface and interface electronic structure of mismatched InxAl1−xAs epitaxial layers grown by molecular beam epitaxy on InP have been investigated using surface photovoltage spectroscopy. The crystalline structure of the epilayers was also examined by double-crystal x-ray diffraction. For coherently strained layers, only a few gap states are identified. Highly strained samples with inferior crystalline quality, as judged by a broadened x-ray diffraction peak and the absence of Pendellosung fringes, display a rich spectrum of states in the band gap. Some of the states are close to the surface while others are located in the vicinity of the InAlAs/InP heterointerfaces. © 1995 American Institute of Physics.

Notes: We have examined the initial interface formation and subsequent film growth for Fe films on the As-terminated (2×4) reconstructed surface of GaAs(001) in an effort to correlate the initial adsorption sites and film growth with the magnetic anisotropy. The growth and surface studies were performed in a four chamber ultrahigh vacuum system which incorporates UHV sample transfer between the chambers for III–V semiconductor growth, metal growth, photoelectron diffraction (PED), and scanning tunneling microscopy (STM). After coating with a thin film of Au (40 A(ring)) at room temperature, the samples were removed from the system for postgrowth characterization, which included ferromagnetic resonance and magnetometry. The GaAs(001) surface was prepared via homoepitaxial growth by MBE, with the growth terminated in a manner which resulted in a well-ordered As-dimer terminated (2×4) reconstructed surface as revealed by reflection high energy electron diffraction and STM. Fe deposition was performed in a second MBE chamber for coverages which ranged from 0.1 ML to several tens of monolayers at a substrate temperature of 175 °C.The sample was then transferred under UHV to either the PED or STM chamber to determine the initial adsorption sites, growth mode, and evolution of film structure with increasing coverage. For the lowest coverages studied (0.1 ML), we obtain evidence for preferential adsorption at As-dimer sites. At a coverage of 1 ML, two-dimensional islands form approximately 20 A(ring) in size, but elongated along the missing dimer rows with an aspect ratio of 2:1. At high coverages ((approximately-greater-than)40 A(ring)), we observe three-dimensional mounds or clusters with an average in-plane diameter of approximately 100 A(ring) and a narrow size distribution. Even at this coverage, the monolayer terraces of the GaAs substrate are clearly visible in the STM image, with the Fe film uniformly growing on each terrace. Photoelectron diffraction in the forward scattering regime (high electron kinetic energy) is utilized to determined the growth mode. At low kinetic energy, information on the adsorption site is obtained from Auger electron diffraction and compared with the results of the real space STM images.

Notes: Spin injection into a semiconductor is expected to utilize a ferromagnetic material as the electrical contact. One system of potential interest consists of Fe contacts to a GaAs-based device structure. We have used photoreflectance spectroscopy to measure the Schottky barrier height formed for Fe films on GaAs(001) as a function of the GaAs surface reconstruction. The samples were grown by molecular-beam epitaxy in a multichamber UHV facility which includes chambers for photoelectron diffraction and scanning tunneling microscopy. These techniques were used to obtain a detailed atomic model of the Fe/GaAs interface, enabling us to correlate the measured barrier heights with the interfacial structure, and further discuss the results in the framework of several models for Schottky barrier formation. For the As-rich 2×4 and c(4×4) surfaces, we conclude that the interface is formed by Fe bonding with an As monolayer after dissolution of the surface As dimers and some Ga into the Fe film. The samples were removed from the UHV system after coating with 50 Å of Au for the photoreflectance measurements. While the pure Schottky model predicts a value of 0.43 eV, we find nearly identical barrier heights of 0.8 eV for both the 2×4 and c(4×4) surfaces. This is consistent with the similar interfacial structure resulting from Fe growth on each surface, and the value of 0.8 eV is in agreement with that predicted by the effective work function model assuming predominantly Fe–As bonding. We compare these results with similar data for growth on the Ga-rich 4×6 reconstruction and on ZnSe(001) surfaces.

Notes: Two different approaches, a photoconductive response technique and a correlation of lasing thresholds with theoretical threshold carrier concentrations have been used to determine Auger lifetimes in InAs/GaInSb quantum wells. For energy gaps corresponding to 3.1–4.8 μm, the room-temperature Auger coefficients for seven different samples are found to be nearly an order-of-magnitude lower than typical type-I results for the same wavelength. The data imply that at this temperature, the Auger rate is relatively insensitive to details of the band structure. © 1998 American Institute of Physics.

Notes: We have used atomic force microscopy to investigate the surface morphology of ZnSe films grown on GaAs(001) by molecular beam epitaxy. We report the observation of nanometer scale surface clusters 400–1200 Å in diameter and 60–200 Å in height. The clusters form ex situ as the result of the initial exposure of the ZnSe to atmosphere, and undergo Ostwald ripening at room temperature. Our observations, combined with the relevant literature, suggest that the cluster composition is SeO2. We propose that the oxidation of the ZnSe epilayers produces a thin layer of SeO2 which migrates to form surface clusters on a stable, zinc-related oxide surface. These findings should facilitate greater control of surface morphologies during ZnSe based device fabrication processes.

Notes: We have investigated the transport properties of a field-effect transistor (FET) with a composite quantum well, which consists of two adjacent semiconductor quantum wells, GaSb and InAs, sandwiched by AlSb barriers. This FET shows a novel V-shaped transfer characteristic, which is a direct result of the switching between electron and hole channels.

Notes: The peak current density of InAs/AlSb/GaSb/AlSb/InAs resonant interband tunneling diodes has been enhanced by replacing the AlSb barriers with Al1−xGaxSb that has a narrower band gap. The devices were grown by molecular beam epitaxy and tested at room temperature. Diodes with nominally identical 7-ML-thick ternary alloy barriers with x=0.35 are found to have peak current densities three times larger than those with AlSb barriers. The peak-to-valley current ratio decreases by only one third from 18 for the AlSb diodes to 12 for diodes with the ternary alloy barriers.