D. C. Walters

Annealing dynamics of molecular‐beam epitaxial GaAs grown at 200 °C

By separating a 2‐μm‐thick molecular‐beam‐epitaxial GaAs layer grown at 200 °C from its 650‐μm‐thick substrate, we have been able to obtain accurate Hall‐effect and conductivity data as functions of annealing temperature from 300 to 600 °C. At a measurement temperature of 300 K, analysis confirms that hopping conduction is much stronger than band conduction for all annealing temperatures. However, at higher measurement temperatures (up to 500 K), the band conduction becomes comparable, and a detailed analysis yields the donor and acceptor concentrations and the donor activation energy. Also, an independent absorption study yields the total and charged AsGa concentrations. Comparisons of all of these quantities as a function of annealing temperature TA show a new feature of the annealing dynamics, namely, that the dominant acceptor (probably VGa related) strongly decreases and then increases as TA is increased from 350 to 450 °C. Above 450 °C, ND, NA, and [AsGa] all decrease, as is known from previous studies.

Native donors and acceptors in molecular-beam epitaxial GaAs grown at 200° C

Absorption measurements at 1.1 and 1.2 μm were used along with the known electron and hole photoionization cross sections for EL2 to determine deep donor (EL2‐like) and acceptor concentrations ND=9.9×1019 and NA=7.9×1018 cm−3, respectively, in a 2‐μm‐thick molecular‐beam epitaxial GaAs layer grown at 200 °C on a 2‐in.‐diam semi‐insulating wafer. Both lateral and depth uniformities of ND over the wafer were excellent as was also the case for the conductivity. Band conduction was negligible compared to hopping conduction at 296 K as evidenced by the lack of a measurable Hall coefficient.

Deep photoluminescence band related to oxygen in gallium arsenide

Temperature‐dependent photoluminescence and photoluminescence excitation spectroscopy have been used to measure the 0.63‐eV luminescence band present in O‐doped semi‐insulating GaAs. It is shown that the 0.63‐eV band is related to the presence of O. The center responsible for the band forms a deep level similar to the main deep donor EL2. However, spark source mass spectrometry indicates that incorporation of O into GaAs is difficult.

A dominant electrical defect in GaAs

Abstract Several as-grown Bridgman and Czochralski GaAs crystals, with dominant electrical levels from 0.13–0.20 eV below the conduction band, have been studied by the temperature-dependent Hall-effect, spark-source mass spectroscopy, and secondary-ion mass spectroscopy. It is shown that no impurity is of a sufficient concentration to account for these levels, and therefore they are composed of single or multiple defects. The detailed nature of the defects has not yet been established but may involve an As vacancy. It is believed that this is the first time that a dominant electrically active center has been shown to be a pure defect in any as-grown semiconductor.

Photoluminescence determination of effects due to In in In‐alloyed semi‐insulating GaAs

Photoluminescence measurements at 2 and 2–40 K were made to study effects due to In alloying for InxGa1−xAs semi‐insulating substrate materials grown by the liquid‐encapsulated Czochralski method. The neutral CAs bound exciton is a good photoluminescence transition to determine a small variation of In composition in the range of 0≤x≤0.014. The band‐gap reduction ΔEg (eV) can be expressed by −1.59x. The radial nonuniformity of In concentration and photoluminescence intensity were determined. The axial segregation of In was also analyzed with the help of the neutron activation analysis and spark‐source mass spectrometry.

Wafer-level correlations of EL2, dislocation density, and FET saturation current at various processing stages

The neutral deep-donor density [EL2]0, and dislocation density,ρD, are measured on adjacent, semi-insulating GaAs wafers, grown by both high-pressure (HP) and low-pres-sure (LP) liquid-encapsulated Czochralski (LEC) techniques; also, other nearby wafers from each boule are used for low-noise, field-effect-transistor (FET) fabrication. Dense data maps (at least 3500 points per wafer per parameter) are then visually and math-ematically compared for [EL2]0,ρD,Iu, Ir, and Ig where the latter three quantities rep-resent the unrecessed-ungated, recessed-ungated, and gated saturation currents, re-spectively, for ion-implanted, 0.5 ]smm × 300 µm FET’s. For theparticular wafers and processing used in this study, the following conclusions can be drawn: (1) onall of the wafers, materials (EL2 andρD) non-uniformities are correlated with at least some of theIu non-uniformities; (2) onsome of the wafers, materials non-uniformities follow all the way through toIg, but on others, the gate-recess step itself introduces much stronger non-uniformities; (3) the HP-LEC wafers give slightly higherIu’s than the LP-LEC waf-ers; and (4) [EL2]0 is a better predictor ofIu than isρD.

Mechanisms for GaAs surface passivation by a molecular beam epitaxial cap layer grown at 200 °C

A thin, undoped, molecular beam epitaxial (MBE) GaAs cap layer grown on top of an n‐type conductive layer significantly reduces the free‐electron depletion from the latter. By analyzing electron transfer to surface, interface, and bulk acceptor states in the cap, as a function of cap thickness, we show that either (1) the usual EC−0.7 eV surface states are absent, (2) a dense donor near EC−0.4 eV exists or (3) a high donor interface charge (∼5×1012 cm−2) is present. Any of these conclusions constitutes an important new aspect of low‐temperature MBE GaAs.

Infrared Transmission Topography for Whole-Wafer Gallium-Arsenide Materials Characterization

Abstract Infrared transmission topography is shown to be useful for evaluating GaAs wafers. Whole-wafer, half-millimeter resolution plots of EL2 density and dislocation density are shown to correlate with plots of saturation current in MESFET devices at an early stage of fabrication.

Uniformity of 3‐in., semi‐insulating, vertical‐gradient‐freeze GaAs wafers

We have evaluated the uniformity in [EL2], dislocation (or etch‐pit) density (EPD), resistivity, mobility, and carrier concentration for 3‐in., semi‐insulating GaAs wafers grown by the vertical‐gradient‐freeze (VGF) technique. Although slight W or U patterns were observed in [EL2] and EPD along the 〈110〉 directions, for the first time, nevertheless the overall uniformity was excellent, and comparable to that in the best In‐doped and whole‐boule‐annealed ingots grown by the liquid‐encapsulated Czochralski (LEC) technique. Based on results from implant‐activation studies on LEC wafers, it is estimated that the measured nonuniformities in EPD and [EL2] for the VGF wafers would contribute only about 1% to implant‐activation‐efficiency nonuniformities in Si‐implanted wafers designed for field‐effect transistor applications.

Automated and calibrated whole wafer etch pit density measurements in GaAs

A technique for automated measurement of whole-wafer etch pit density (EPD) for GaAs wafers is presented. The technique relies on an infrared transmission experiment similar to that used to measure EL2 concentration. A theoretical relationship between transmission and EPD is established, including effects due to pit size. The new automated and old visual-count methods are compared on a 3“, low-pressure, liquid-encapsulated Czochralski wafer; it is established that the automated method has much better repeatability. An [EL2] map of this same wafer is also presented.