Engineering topics
Publications and source records attributed to Weinberg, Irving.
Transient photoluminescence is used to measure the minority carrier lifetime on n-type and p-type InP wafers. The measurements show that unprocessed InP wafers have very high minority carrier lifetimes. Lifetimes of 200 ns and 700 ns were observed for lightly-doped p- and n-type material respectively. Lifetimes over 5 ns were found in heavily doped n-type material.
Using a detailed simulation model of p(+)nn(+) and n(+)pp(+) indium phosphide (InP) homojunction solar cells, extensive parametric variation computer simulation runs are conducted to help arrive at near-optimum designs of these two solar cell configurations. Values of all the geometrical and material parameters corresponding to the near-optimal designs of both these configurations are presented. For each configuration, results are given for parametric variation runs showing how the performance parameters JSC, VOC, and eta vary with each of the cell parameters for the near-optimally designed cell.
An etchant, namely (o-H3PO4)u: (HNO3)v: (H2O2)t: (H2O)1-(u+v+t) is developed for thinning, after contacting, the p(+) emitter layer of p(+)n InP structures made by thermal diffusion. Varying u, v, and t, reproducible etch rates of 5 to 110 nm/min are obtained. After thinning the 0.6- to 2.5-micron thick p(+) InP layer down to 60-250 nm, specular surfaces are obtained at up to 80 nm/min etch rate. Due to its intrinsic qualities the resisual P-rich oxide after thinning the emitter layer provides surface passivation of p(+) InP surfaces and can also serve as a first-layer AR coating.
The current status of indium phosphide solar cell research is reviewed. In the NASA research program, efficiencies of 18.8 percent were achieved for standard n/p homojunction InP cells while 17 percent was achieved for ITO/InP cells processed by sputtering n-type indium tin oxide onto p-type indium phosphide. The latter represents a cheaper, simpler processing alternative. Computer modeling calculations indicate that efficiencies of over 21 percent are feasible. Relatively large area cells are produced in Japan with a maximum efficiency of 16.6 percent.
The characteristics of InP cells processed from thin layers of InP heteroepitaxially grown on GaAs, on silicon with an intervening GaAs layer, and on GaAs with intervening Ga(x)In(1-x)As layers are described, and the factors affecting cell efficiency are discussed. Under 10 MeV proton irradiations, the radiation resistances of the heteroepitaxial cells were superior to that of homoepitaxial InP cells. The superior radiation resistance is attributed to the high dislocation densities present in the heteroepitaxial cells.
The effects of various surface preparation procedures, including chemical treatment and anodic or chemical oxidation, closed-ampoule diffusion conditions, and post-diffusion surface preparation and annealing conditions, on the passivating properties of InP have been investigated in order to optimize the fabrication procedures of n(+)p InP solar cells made by closed-ampoule diffusion of sulfur into p-type InP. The InP substrates used were p-type Cd-doped to a level of 1.7 x 10 to the 16th/cu cm, Zn-doped to levels of 2.2 x 10 to the 16th and 1.2 x 10 to the 18th/cu cm, and n-type S-doped to 4.4 x 10 to the 18th/cu cm. The passivating properties have been evaluated from photoluminescence (PL) and conductance-voltage (G-V) data. Good agreement was found between the level of surface passivation and the composition of different surface layers as revealed by X-ray photoelectron spectroscopy (XPS) analysis.
Solar cells made from gallium arsenide (Gaas), with a room temperature bandgap of E(sub g) = 1.43 eV have exhibited the best measured open circuit voltage (V sub oc) of 1.05 V at 1 AM0, 25 C. The material InP is in many ways similar to GaAs. A simple calculation comparing InP to GaAs then shows that solar cells made from InP, with E(sub g) = 1.35 at 300 K, should exhibit the best measured V sub oc of approximately 950 mV at 1 AM0, 300 K. However, to date, the best measured V sub oc for InP solar cells made by any fabrication method is 899 mV at AM1.5, 25 C which would translate to 912 mV at 1 AM0, 25 C. The V sub oc of an n(+)pp(+) InP solar cell is governed by several factors. Of these, some factors, such as the thickness and doping of the emitter and base regions, are easily controlled and can be adjusted to desired values dictated by a good performance optimizing model. Such factors were not considered. There are other factors which also govern V sub oc, and their values are not so easily controlled. The primary ones among these are (1) the indirect or Hall-Shockley-Read lifetimes in the various regions of the cell, (2) the low-doping intrinsic carrier concentration n(sub i) of the InP material, (3) the heavy doping factors in the emitter and BSF regions, and (4) the front surface recombination velocity S(sub F). The influence of these latter factors on the V sub oc of the n(+)pp(+) InP solar cell and the results were used to produce a near-optimum design of the n(+)pp(+) InP solar cell.
The flight performance of the NASA Lewis Research Center's indium phosphide homojunction cell module on the LIPs III spacecraft is presented. Four n+-p diffused junction cells, an early product of an InP development program, were flown. The voltage range of the data prohibits determination of the open-circuit voltage or the maximum power point. However, analysis of the 32 months of short-circuit current data reveals a slight increase, not the 4 percent decrease expected from radiation tolerance studies. This increase may be due to continual cleaning of possible prelaunch dust contamination or changes in data acquisition system calibration. For all cells, the average short-circuit current remains below prelaunch values.
Progress, dating from the start of the Lewis program, is reviewed emphasizing processing techniques which have achieved the highest efficiencies in a given year. To date, the most significant achievement has been attainment of AM0 total area efficiencies approaching 19 percent. Although closed tube diffusion is not considered to be an optimum process, reasonably efficient 2cm x 2cm and 1cm x 2cm InP cells have been produced in quantity by this method with a satellite to be launched in 1990 using these cells. Proton irradiation of these relatively large area cells indicates radiation resistance comparable to that previously reported for smaller InP cells. A similar result is found for the initial proton irradiations of ITO/InP cells processed by D. C. sputtering. With respect to computer modelling, a comparison of n/p homojunction InP and GaAs cells of identical geometries and dopant concentrations has confirmed the superior radiation resistance of InP cells under 1 MeV electron irradiations.
Researchers have calculated the expected performance dependence of near-optimally designed shallow homojunction n+pp+InP solar cells on incident intensities up 200 AM0 and temperatures up to 100 C (373K). Both circular and rectangular cells have been considered, the former for use in a Cassegrainian concentrator array at 100 AM0, 80 to 100 C and the latter for use in a Slats type concentrator array at 20 AM0 80 to 100 C. Calculation of the temperature dependence of the performance parameters I sub sc, V sub oc, FF and eta was done by first verifying that the use of the measured temperature variation of I sub sc, of the best published value of the temperature dependence of the bandgap of InP, and of the temperature dependences of the lifetimes and mobilities of electrons and holes the same as in equivalently doped GaAs, gave calculated results that closely matched measured data on the temperature variation of I sub sc, V sub oc, and FF of four existing InP cells at 1 AM0. It was then assumed that the same temperature dependences of I sub sc, the bandgap and lifetimes and mobilities would hold in the near-optimally designed cells at the higher concentrations.
Progress, in the past year, in InP solar cell research is reviewed. Small area cells with AMO, total area efficiencies of 18.8 percent were produced by OMCVD and Ion Implantation. Larger area cells (2 and 4 sq cm) were processed on a production basis. One thousand of the 2 sq cm cells will be used to supply power to a small piggyback lunar orbiter scheduled for launch in February 1990. Laboratory tests of ITO/InP cells, under 10 MeV proton irradiation, indicate radiation resistance comparable to InP n/p homojunction cells. Computer modeling studies indicate that, for identical geometries and dopant concentrations, InP solar cells are significantly more radiation resistant than GaAs under 1 MeV electron irradiation. Additional computer modeling calculations were used to produce rectangular and circular InP concentrator cell designs for both the low concentration SLATS and higher concentration Cassegrainian Concentrators.
Performance data for the NASA Lewis Research Center indium phosphide n+p homojunction solar cell module on the LIPS 3 Flight Experiment is presented. The objective of the experiment is to measure the performance of InP cells in the natural radiation environment of the 1100 km altitude, 60+ deg inclination orbit. Analysis of flight data indicates that the performance of the four cells throughout the first year is near expected values. No degradation in short-circuit current was seen, as was expected from radiation tolerance studies of similar cells. Details of the cell structure and flight module design are discussed. The results of the temperature dependency and radiation tolerance studies necessary for normalization and analysis of the data are included.
The authors calculated the expected performance dependence of near-optimally designed shallow homojunction n+pp+ InP solar cells on incident intensities up to 200 AM0 and temperatures up to 100 deg C (373 K). Both circular and rectangular cells were considered, the former for use in a Cassegrainian concentrator array at 100 AM0, 80-100 deg C and the latter for use in a Slats concentrator array at 20 AM0, 80-100 deg C. With efficiencies near 22 percent at 80 deg C, both the circular and rectangular InP shallow homojunction solar cells compare very favorably to GaAs cells of the same design and may be preferable to the GaAs cells for space applications because of the superior radiation tolerance of the InP cells.
Performance data for the NASA Lewis Research Center indium phosphide n+p homojunction solar cell module on the LIPS 3 flight experiment is presented. The objective of the experiment is to measure the performance of InP cells in the natural radiation environment of the 1100 km altitude, 60+ deg inclination orbit. Analysis of flight data indicates that the performance of the four cells throughout the first year is near expected values. No degradation in short-circuit current was seen, as was expected from radiation tolerance studies of similar cells. Details of the cell structure and flight module design are discussed. The results of the temperature dependency and radiation tolerance studies necessary for normalization and analysis of the data are included.
Key issues and opportunities in space photovoltaic research and technology relative to future NASA mission requirements and drivers are addressed. Examples are given of space missions and/or operational capabilities on NASA's planning horizon presenting major technology challenges to the use of photovoltaic power generation in space. The status of cell R and D and the performance goals to be met by space photovoltaic power systems to remain competitive are described.
Recent developments and progress in indium phosphide solar cell research for space application are reviewed. Indium phosphide homojunction cells were fabricated in both the n+p and p+n configurations with total area efficiencies of 17.9 and 15.9% (air mass 0 and 25 C) respectively. Organometallic chemical vapor deposition, liquid phase epitaxy, ion implantation and diffusion techniques were employed in InP cell fabrication. A theoretical model of a radiation tolerant, high efficiency homojunction cell was developed. A realistically attainable AMO efficiency of 20.5% was calculated using this model with emitter and base doping of 6 x 10 to the 17th power and 5 x 10 the the 16th power/cu cm respectively. Cells of both configurations were irradiated with 1 MeV electrons and 37 MeV protons. For both proton and electron irradiation, the n+p cells are more radiation resistant at higher fluences than the p+n cells. The first flight module of four InP cells was assembled for the Living Plume Shield III satellite.
Using a fairly comprehensive model, researchers have done a parametric variation study of the InP n+p homojunction solar cell for AMO, 25 C operation. The results of this study are presented. These results indicate that an efficiency of about 25 percent should be realistically possible in a shallow homojunction InP solar cell with near-optimum design.
State-of-the-art n/p and p/n heteroface GaAs cells, processed by metal organic chemical vapor deposition, were irradiated by 1 MeV electrons and 37 MeV protons and their performance determined as a function of fluence. It was found that the p/n cells were more radiation resistant than the n/p cells. The increased loss in the n/p cells was attributed to increases in series resistance and losses in the p-region resulting from the irradiation. The greater loss in fill factor observed for the n/p cells introduces the possibility that the presently observed superiority of the p/n cells may not be an intrinsic property of this configuration in GaAs.