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.
This paper considers the problem of designing nonlinear multiport resistive networks for prescribed driving point and transfer characteristics. It is shown that, if the topology of the desired network is given, then the i-v characteristics of the nonlinear elements can be obtained iteratively. This is achieved by implementation of a newly developed steepest descent criterion.
Work on aluminum-alloyed silicon grating cells is continued. Optimization of the geometry (grating line width and spacing) confirms the analysis of such cells. A 1 sq cm grating cell was fabricated and its i-V characteristic was measured under an AMO solar simulator. It is found that the efficiency of this cell would be about 7.9%, if it were covered by the usual antireflection coating. The surface of the cell is not covered by a diffused junction. The response is blue shifted; the current is somewhat higher than that produced by a commercial Si cell. However, the open circuit voltage is low, and attempts to optimize the open circuit voltage of the aluminum-alloy junctions are described. A preliminary X-ray topographic examination of GaAs specimens of the type commonly used to make solar cells is studied. The X-ray study shows that the wafers are filled with regions having strain gradients, possibly caused by precipitates. It is possible that a correlation exists between the presence of low mechanical perfection and minority carrier diffusion lengths of GaAs crystals.
Distortions can be greatly reduced in narrow-band applications by using the second harmonic. The ac behavior of simplified diode amplifier has negative resistance depending on slope of equivalent I-V curve.
Various techniques to increase the open circuit voltage are being explored. It had been previously observed that cells made on CdS deposited from a single source gave a consistently higher V sub oc. Further tests have now shown that this effect may in fact relate to differences in source and substrate temperatures. The resulting differences in CdS structure and crystallinity are being documented. Deposits of mixed CdS and ZnS are being produced and will be initially made into cells using the conventional barriering technique. Analysis of I-V characteristics at temperatures between 25 and 110 C is being perfected to provide nondestructive analysis of the Cu2S. Changes due to vacuum heat treatments and exposure to oxygen are also being monitored by the same technique. Detailed spectral response measurements are being made.
The current collected from a plasma through holes in insulated electrodes was measured. Holes of 0.051- and 2.54-cm diameters in Kapton H film and plasma number densities of 100 and 10,000 electrons/cu cm were used. The current collected by bare electrodes, that is, electrodes with no surrounding insulation, is also presented. For all the samples the current at a given voltage was a function of the surrounding insulator area rather than of the hole size or the underlying electrode size. In addition, at the low plasma density the I-V characteristic showed very steep rises for voltages below 1 kV. In one case the current jumped by a factor of approximately 70 to 200 V. Results are given for positive biases to 10 kV. For negative biases, sparking prevented testing most samples to the 10-kV limit.
The power efficiency curves of photovoltaic solar cells were investigated as a function of the forbidden energy gap (E sub g) and the current-voltage characteristic of the diode. Minority carrier injection, depletion layer recombination, and interface recombination terms were considered in models for the I-V characteristic. The collection efficiency for photons with energy between (E sub g) and an upper energy cutoff (E sub w) was assumed to be 100% and zero otherwise. Results are presented in terms of a single parameter related to the ratio of depletion layer width and minority carrier diffusion length. It was found that increasing depletion layer recombination shifts the efficiency curves to larger values of the energy without changing the shape of the efficiency curve appreciably. It is believed that similar results would be obtained whenever the quality factors in the exponential energy gap and forward bias terms are equal.
P+/p/n/n+ epitaxial silicon solar-cell structures were fabricated with the n-base region intentionally graded over 2 orders of magnitude in a thickness of 35 microns. Solar cells made from these layers exhibited near-ideal I-V characteristics in the practical operating range, with an open-circuit voltage of 636 mV, fill factor of 0.79, and AM1 efficiency of 12.6%. A similar epitaxial structure containing no gradient in the n-base showed good solar-cell performance but lower open-circuit voltage (603 mV).
A theoretical model was used to study the effects of structured electrodes on converter I-V characteristics and results are given. An auxiliary-ion-source triode operated as a plasmatron was used for studying the enhancement distribution and magnetic effects, and results are reported. Design features of the high current-zero power (ZEPO) converter tests are given.
The effects of outdoor exposure were determined by comparing standard I-V data obtained for the as-received modules with similar data obtained after removal from the field and cleaning with detergent solution. All modules measured in this way exhibited nonrecoverable degradation in P sub maximum varying from 4 to 7 percent. One module exposed for 41 days exhibited partial cell discoloration, loss of front surface metallization over the discolored portion, and a decrease in P sub maximum of 7 percent, tentatively attributed to cell damage. Measurements before and after cleaning showed a recoverable degradation due to dirt accumulation. This recoverable loss in power was 11 percent after 245 days in the field for one brand of module, 6 percent after 48 days for another brand, and 4 1/2 percent for the third brand.
The series resistance and diode quality factor are used for evaluation of solar cell performance. These parameters are normally deduced from the terminal I-V characteristics under forward bias. The method produces a smaller series resistance and a larger value for diode quality factor than appropriate for the device when operating as a solar cell. It is suggested that the diode quality factor be determined from measured saturated photocurrent, open circuit voltage characteristics and the series resistance be measured with an r-f bridge or calculated from a measurement of the surface sheet resistance.
A superconducting microbridge small enough to exhibit the Josephson effect has been shunted by a large external capacitance. Surprisingly little capacitive hysteresis is observed in its I-V characteristic. Prominent subgap structures are seen. The resistively shunted junction (RSJ) model fails to explain the observations.
A number of modern solar cell types representing 1976 technology (as well as some older types) were irradiated with 1 MeV electrons (and a limited number with 2 MeV electrons and 10 MeV protons). After irradiation, the cells were annealed, with I-V curves measured under AMO at 30 C. The purpose was to provide data to be incorporated in the revision of the solar cell radiation handbook. Cell resistivities ranged from 2 to 20 ohm-cm, and cell thickness from 0.05 to 0.46 mm. Cell types examined were conventional, shallow junction, back surface field (BSF), textured, and textured with BSF.
Hot-ion hydrogen plasma experiments were conducted in the NASA Lewis HIP-1 magnetic mirror facility to determine how the ion temperature was influenced by the axial position of the cathode tips relative to the anodes. A steady-state EXB plasma was formed by applying a strong radially inward dc electric field near the throats of the magnetic mirrors. The dc electric field was created between hollow cathode rods inside hollow anode cylinders, both concentric with the magnetic axis. The highest ion temperatures, 900 eV, were attained when the tip of each cathode was in the same plane as the end of its anode. These temperatures were reached with 22 kV applied to the electrodes in a field of 1.1 tesla. Scaling relations were empirically determined for ion temperature and the product of ion density and neutral particle density as a function of cathode voltage, discharge current, and electrode positions. Plasma discharge current vs voltage (I-V) characteristics were determined.
A mechanically stable single block mixer design is described utilizing a recessed whisker and beamwidth equalization lens. A stripline I.F. matching section which is an integral part of the mixer is presented. Engineering measurements of wire grids and dielectric transmission loss near one millimeter wavelength are given and an anomolous I-V curve behavior observed during diode whiskering is discussed.
A new metal-oxide-metal device (Ni-NiO-Ni, Edge MOM) which is stable, reproducibly fabricated, and with a 10 to the -10th sq cm tunneling area is presented. Performing detection experiments, the device's nonlinear I-V characteristic is shown to be invariant at audio frequencies, 10.6, 3.39, and 0.6328 microns. Similar devices with 10 to the -8th sq cm tunneling areas perform as well as the Edge MOM's in the visible and the near-infrared range, but deteriorate in performance at the 10-micron range. A dominant competing effect is a thermal-induced signal, which increases with frequency and temperature. Coupling mechanisms at the various regimes are investigated. The device can serve as a broad-band detector and mixer, and might in the future be a basic element of broad-band amplifiers and oscillators.
Gallium arsenide concentrator cells were irradiated with electrons with energies varying from 0.4 to 2.3 MeV, and their electrical performance was measured. The cells are 5 x 5 mm square with a 4-mm diameter illuminated area. At each of four different electron energy levels (0.4, 0.7, 1.0, and 2.3 MeV), three n/p and two p/n cells were irradiated. I-V performance measurements were made prior to irradiation and at several intermediate fluence levels. The final fluence level was 3 X 10 to the 15th e/sq cm. It is concluded that the power degradation is independent of the temperature at which it is measured.
A nonvolatile memory device based on protonic transport in oxides has been proposed. The mobile H+ ions are introduced into the SiO2 layer by annealing Si/SiO2/Si structures in H2 at temperatures greater than 500 deg C. This effect has only been observed for confined oxides that have been annealed at greater than or equal to 1100 C prior to the hydrogenation anneal. This includes buried oxides such as Unibond and SIMOX as well as thermal oxides annealed with a polysilicon cap. An applied field moves the charge within the oxide and the charge stops moving when the field is removed. In a memory device, the hydrogen-annealed oxide is the gate oxide and the position of the mobile charge is sensed by the shift of the I-V curve. Much is still not understood about the motion of the charge across the buried oxide. Previous work has assumed that H+ transport and the time it takes to traverse the oxide is governed by interactions within the bulk of the oxide. Based on parameters that affect the transport time, we conclude that H+ trapping and detrapping at the Si/SiO2 interface are more important than H+ interactions within the oxide bulk. These parameters include the applied field, the H+ concentration and the oxide thickness. One consequence is that projections of device write-time based on the previous assumptions of H+ transport mechanisms may be overly optimistic.
This work investigates the potential induced degradation-polarization (PID-p) of bifacial passivated emitter and rear contact (PERC) modules installed in the field at the Florida Solar Energy Center. 22 modules were stressed over 1 year under various voltage (-1500V, -3500, +1500V) and mounting configurations (close to ground, 2 meters from ground, close to roof). The positively stressed modules were anticipated to have experiences PID-p on their front side. However, no degradation was observed, as the light-induced recovery effect effectively counteracted the degradation. The negatively stressed modules experienced Pmax drops during 2 weeks before stabilizing at 5-6% loss. I-V and EL characteristics suggest a PID-p degradation on the rear side. Lower degradation rates are observed with lower voltage and when the mounting configuration allows more albedo light on the rear side to compensate for the degradation. PID-p is observed to stabilize at similar level for the voltage and mounting configuration tested with slightly lower Pmax in winter. EL imaging reveals that cells are not degrading at the same rate across a module. Some cells get darker and then brighter which corresponds to the silicon/passivation interface going through depletion and then inversion when enough positive charges are accumulated. After full light-recovery, the modules were tested again along with 22 brand-new ones to assess the impact of the module history. Results confirmed the trends observed in the first year with PID-p occurring at voltage as low as -600V and recovered modules degrading faster than brand-new ones.