SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Interface sputter damage effects were measured for Sn-doped (approximately 2x10 to the 17th power/cu cm) 100 oriented GaAs. The sputter surfaces were characterized by Auger and X-ray photoelectron spectroscopy. Gold contacts were then deposited in situ on these characterized surfaces. Results of I-V and C-V measurements are shown in graphs.
The objective of this program is to develop 60 GHz GaAs IMPATT Diodes suitable for communications applications. The performance goal of the 60 GHz IMPATT is 1W CW output power with a conversion efficiency of 15 percent and 10 year life time. During the course of the program, double drift (DD) GaAs IMPATT Diodes have been developed resulting in the state of the art performance at V band frequencies. A CW output power of 1.12 W was demonstrated at 51.9 GHz with 9.7 percent efficiency. The best conversion efficiency achieved was 15.3 percent. V band DD GaAs IMPATTs were developed using both small signal and large signal analyses. GaAs wafers of DD flat, DD hybrid, and DD Read profiles using molecular beam epitaxy (MBE) were developed with excellent doping profile control. Wafer evaluation was routinely made by the capacitance versus voltage (C-V) measurement. Ion mass spectrometry (SIMS) analysis was also used for more detailed profile evaluation.
The effect of processing conditions on the interface state density was evaluated from C-V measurements on metal-oxide-semiconductor capacitors. The optimum processing conditions for the minimum surface state density was found to be related to the postoxidation annealing temperature and time, and was independent of chemical treatments prior to oxidation. Annealing at the optimum condition (i.e., at 350 C for 1 h in either nitrogen or hydrogen gas, with or without an aluminum pattern on the oxide) reduces the fast surface state density by about one order of magnitude. By using a nitrogen/oxygen plasma, the static dielectric constant of the oxide decreased as the N/O ratio was increased, and nitrogen was incorporated into the oxide. In addition, the fast surface state density was reduced as a result of this nitridation process.
Composition-dependent optical and electrical properties of L-CVD (Al2O3)1-c-(AlN)c composite films are examined. The optical properties exhibit a continuous change from Al2O3 to AlN with increasing value of the composition parameter c from zero to one. There is a critical concentration value of about 0.3 below and above which Al2O3- and AlN-like optical properties are observed. The impurity absorption regions are 5.1-6.7 eV for Al2O32 and 4.6-6.0 for AlN, while the energy gap is about 6.7 eV for Al2O3 and 6.0 eV for AlN. The electrical properties of Al-insulator n-Si MIS structures display externally programmable states when bulk-controlled electron emission occurs from the traps. In the high-field region a persistent polarization field is established within the insulator. C-V characteristics of the samples are more sensitive to the concentration factor c than the I-V characteristics. A large clockwise hysteresis is observed for AlN, but a smaller counterclockwise one is observed for Al2O3.
The objectives were to perform surface characterization of high efficiency n+/p and p+/n silicon cells, to relate surface density to substrate dopant concentration, and to identify dominant current loss mechanisms in high efficiency cells. The approach was to measure density of states on homogeneously doped substrates with high frequency C-V and Al/SiN sub x/Si structures; to investigate density of states and photoresponse of high efficiency N+/P and P+/N cells; and to conduct I-V-T studies to identify current loss nechanisms in high efficiency cells. Results are given in tables and graphs.
Phosphorus rich plasma enhanced chemical vapor deposition (PECVD) of silicon nitride and silicon dioxide films on n-type indium phosphide (InP) substrates were exposed to electron-beam irradiation in the 5 to 40 keV range for the purpose of characterizing the damage induced in the dielectric. The electron-beam exposure was on the range of 10(exp -7) to 10(exp -3) C/sq cm. The damage to the devices was characterized by capacitance-voltage (C-V) measurements of the metal insulator semiconductor (MIS) capacitors. These results were compared to results obtained for radiation damage of thermal silicon dioxide on silicon (Si) MOS capacitors with similar exposures. The radiation induced damage in the PECVD silicon nitride films on InP was successfully annealed out in an hydrogen/nitrogen (H2/N2) ambient at 400 C for 15 min. The PECVD silicon dioxide films on InP had the least radiation damage, while the thermal silicon dioxide films on Si had the most radiation damage.
Phosphorus rich plasma enhanced chemical vapor deposition (PECVD) of silicon nitride and silicon dioxide films on n-type indium phosphide (InP) substrates were exposed to electron beam irradiation in the 5 to 40 keV range for the purpose of characterizing the damage induced in the dielectic. The electron beam exposure was on the range of 10(exp -7) to 10(exp -3) C/sq cm. The damage to the devices was characterized by capacitance-voltage (C-V) measurements of the metal insulator semiconductor (MIS) capacitors. These results were compared to results obtained for radiation damage of thermal silicon dioxide on silicon (Si) MOS capacitors with similar exposures. The radiation induced damage in the PECVD silicon nitride films on InP was successfully annealed out in an hydrogen/nitrogen (H2/N2) ambient at 400 C for 15 min. The PECVD silicon dioxide films on InP had the least radiation damage, while the thermal silicon dioxide films on Si had the most radiation damage.
Cubic SIC metal-insulator-semiconductor (MIS) capacitors with thermally grown or chemical-vapor-deposited (CVD) insulators were characterized by capacitance-voltage (C-V), conductance-voltage (G-V), and current-voltage (I-V) measurements. The purpose of these measurements was to determine the four charge densities commonly present in an MIS capacitor (oxide fixed charge, N(f); interface trap level density, D(it); oxide trapped charge, N(ot); and mobile ionic charge, N(m)) and to determine the stability of the device properties with electric-field stress and temperature. The section headings in the report include the following: Capacitance-voltage and conductance-voltage measurements; Current-voltage measurements; Deep-level transient spectroscopy; and Conclusions (Electrical characteristics of SiC MIS capacitors).
A semi-quantitative method to analyze decompression data is described. It possesses the advantage that it allows a graded response to decompression rather than the dichotomous response generally employed. A generalized critical volume (C-V), or stoichiometric time-dependent equilibrium model is examined that relates the constant of the equation P sub i equals m P sub f plus b to variable tissue supersaturation and gas washout terms. The effects of the tissue ratio on gas phase formation indicate that a decreased ratio yields fewer individuals with Doppler detectable gas bubbles, but those individuals still present with Spencer Grade 3 or 4. This might indicate a local collapse of tissue saturation. The individuals with Grade 3 or 4 could be at risk for type 2 decompression sickness by transpulmonic arterialization. The primary regulator of the problems of decompression sickness is the reduction of local supersaturation, presumably governed by the presence and number of gas micronuclei. It is postulated that a reduction in these nuclei will favor a low incidence of decompression sickness in microgravity secondary to hypokinesia and adynamia.
The photovoltaic characteristics of n(+)/p/p(+) homojunction InP solar cells fabricated by organometallic vapor-phase epitaxy (OMVPE) are described. The cells are characterized by I-V, C-V and quantum efficiency measurements, and simulations are used to obtain various device and material parameters. The I-V characteristics show a high recombination rate in the depletion region; this is shown to be independent of the impurity used. It is shown that cadmium is easier to use as an acceptor for the p base and p(+) buffer and is therefore beneficial. The high quantum efficiency of 98 percent at long wavelengths measured in these cells indicates a very good collection efficiency in the base. The short-wavelength quantum efficiency is poor, indicating a high surface recombination.
The properties of heteroepitaxial InP cells were determined both before and after 10-MeV proton irradiations. Numerical values, obtained for the diffusion and recombination components of the reverse saturation currents, were found to be consistent with the distribution of dislocations. The radiation resistance of the heteroepitaxial cells was significantly greater than that observed for n/p homoepitaxial InP cells. The carrier removal rate, obtained by C-V measurements, was 1800/cm for 10-MeV protons compared with 2.2/cm for 1-MeV electrons. The high carrier removal rate was found to have no significant effect on the cell's series resistance. It was concluded that the heteroepitaxial cell performance is dominated by the high dislocation density attributable to lattice constant mismatch. Although the efficiencies of the present cells are low, the recent achievement of 13.7 percent AM0 efficiencies using a GaAs substrate demonstrates the marked improvement that can be attained using more appropriate transition layers.
Models for optimization of planar frequency triplers with symmetrical C-V curves are presented. Role and limitation of various blocking barriers (oxide, Mott, heterojunction) are discussed. Devices with undoped drift regions (BIN) have moderate efficiency but a broad range of power operation, whereas devices with doped drift regions (BNN) have high efficiency in a narrow power window. In particular, an upper power limit of the BNN is caused by electron velocity saturation. Implementations in SiO2/Si and AlAs/GaAs and means for increasing the power of BNN structures are considered.
Theoretical work on Single Barrier Varactor (SBV) diodes, indicate that the efficiency for a multiplier has a maximum for a considerably smaller capacitance variation than previously thought. The theoretical calculations are performed, both with a simple theoretical model and a complete computer simulation using the method of harmonic balance. Modeling of the SBV is carried out in two steps. First, the semiconductor transport equations are solved simultaneously using a finite difference scheme in one dimension. Secondly, the calculated I-V, and C-V characteristics are input to a multiplier simulator which calculates the optimum impedances, and output powers at the frequencies of interest. Multiple barrier varactors can also be modeled in this way. Several examples on how to design the semiconductor layers to obtain certain characteristics are given. The calculated conversion efficiencies of the modeled structures, in a multiplier circuit, are also presented. Computer simulations for a case study of a 750 GHz multiplier show that InAs diodes perform favorably compared to GaAs diodes. InAs and InGaAs SBV diodes have been fabricated and their current vs. voltage characteristics are presented. In the InAs diode, was the large bandgap semiconductor AlSb used as barrier. The InGaAs diode was grown lattice matched to an InP substrate with InAlAs as a barrier material. The current density is greatly reduced for these two material combinations, compared to that of GaAs/AlGaAs SBV diodes. GaAs based diodes can be biased to higher voltages than InAs diodes.
Gallium sulphide (GaS) has been deposited on GaAs to form stable, insulating, passivating layers. Spectrally resolved photoluminescence and surface recombination velocity measurements indicate that the GaS itself can contribute a significant fraction of the photoluminescence in GaS/GaAs structures. Determination of surface recombination velocity by photoluminescence is therefore difficult. By using C-V analysis of metal-insulator-semiconductor structures, passivation of the GaAs with GaS films is quantified.
The back-to-back barrier-N-N(+) (bbBNN) varactor is a nonlinear device being developed for frequency multiplier applications above 100 GHz. Its symmetrical C-V characteristic, low series resistance, freedom from external bias and suitability to planarization make it an ideal choice for high frequency, low power, odd harmonic generation. In this paper, the performance of a 220-GHz waveguide tripler using, for the first time, a planar GaAs bbBNN device integrated on a quartz microstrip circuit is presented. A new split-waveguide block design has been employed to provide the proper embedding impedances to the device at the input and third harmonic output frequencies. A flange-to-flange tripling efficiency of 7% has been obtained at 220 GHz with an output power in excess of 700 mu W. This is believed to be the highest conversion efficiency yet reported for a tripler with an integrated device at this frequency. Theoretical calculations indicate that substantial improvement is possible with modest changes to the device and circuit parameters.
High-efficiency, heteroepitaxial (HE) InP solar cells, grown on GaAs, Si or Ge substrates, are desirable for their mechanically strong, light-weight and radiation-hard properties. However, dislocations, caused by lattice mismatch, currently limit the performance of the HE cells. This occurs through shunting paths across the active photovoltaic junction and by the formation of deep levels. In previous work we have demonstrated that plasma hydrogenation is an effective and stable means to passivate the electrical activity of dislocations in specially designed HE InP test structures. In this work, we present the first report of successful hydrogen passivation in actual InP cell structures grown on GaAs substrates by metalorganic chemical vapor deposition (MOCVD). We have found that a 2 hour exposure to a 13.56 MHz hydrogen plasma at 275 C reduces the deep level concentration in HE n+n InP cell structures from as-grown values of approximately 10(exp 15)/cm(exp -3), down to 1-2 x 10(exp 13)/cm(exp -3). The deep levels in the p-type base region of the cell structure match those of our earlier p-type test structures, which were attributed to dislocations or related point defect complexes. All dopants were successfully reactivated by a 400 C, 5 minute anneal with no detectable activation of deep levels. I-V analysis indicated a subsequent approximately 10 fold decrease in reverse leakage current at -1 volt reverse bias, and no change in the forward biased series resistance of the cell structure which indicates complete reactivation of the n+ emitter. Furthermore, electrochemical C-V profiling indicates greatly enhanced passivation depth, and hence hydrogen diffusion, for heteroepitaxial structures when compared with identically processed homoepitaxial n+p InP structures. An analysis of hydrogen diffusion in dislocated InP will be discussed, along with comparisons of passivation effectiveness for n+p versus p+n heteroepitaxial cell configurations. Preliminary hydrogen-passivated HE InP cell results will also be presented.
The flat-band voltage is the Schottky junction voltage required to shrink the depletion width to zero. At cryogenic temperatures, mixer diodes are generally biased and/or pumped beyond the flat-band condition to minimize conversion loss and noise figure. This occurs despite the presumed sharp increase in junction capacitance near flat-band, which should instead limit mixer performance. Past moderate forward bias, the diode C-V relationship is difficult to measure. A simple analytic expression for C(V) is usually used to model and predict mixer performance. This letter provides experimental data on C(V) at 77 K based on a microwave measurement and modeling technique. Data is also provided on the conversion loss of a singly balanced mixer optimized for 77 K operation. The connection between junction capacitance, flat-band potential, and conversion loss is examined. It is shown that the analytic expression greatly overestimates the junction capacitance that occurs as flat-band is approached.
We investigate the density-gradient (DG) transport model for efficient multi-dimensional simulation of quantum confinement effects in advanced MOS devices. The formulation of the DG model is described as a quantum correction to the classical drift-diffusion model. Quantum confinement effects are shown to be significant in sub-100nm MOSFETs. In thin-oxide MOS capacitors, quantum effects may reduce gate capacitance by 25% or more. As a result, the inclusion or quantum effects in simulations dramatically improves the match between C-V simulations and measurements for oxide thickness down to 2 nm. Significant quantum corrections also occur in the I-V characteristics of short-channel (30 to 100 nm) n-MOSFETs, with current drive reduced by up to 70%. This effect is shown to result from reduced inversion charge due to quantum confinement of electrons in the channel. Also, subthreshold slope is degraded by 15 to 20 mV/decade with the inclusion of quantum effects via the density-gradient model, and short channel effects (in particular, drain-induced barrier lowering) are noticeably increased.