Engineering topics
Publications and source records attributed to Baraona, C. R..
To fabricate 50 microns thick, coplanar back contact (CBC) silicon solar cells, wraparound junction design was selected and proved to be effective. The process sequence used, the cell design, and the cell performance are described. CBC cells with low solar absorptance have shown AMO efficiencies to 13%, high cells up to 14%; further improvements are projected with predictable optimization.
Results of computer calculations using a one dimensional model of the silicon tandem junction solar cell with both front and back current collection are presented. Using realistically achievable geometrical and material parameters, the model predicts that with base widths of 50 micrometers and 100 micrometers and base resistivities between 1 ohm/cm and 20 ohm/cm, beginning of life efficiencies of 14% to 17% and end of life efficiencies of 12% to 14%, after about seven years in synchronous orbit, can be obtained.
A one-dimensional theoretical model of the tandem junction solar cell (TJC) with base resistivity greater than about 1 ohm-cm and under low level injection has been derived. This model extends a previously published conceptual model which treats the TJC as an npn transistor. The model gives theoretical expressions for each of the Ebers-Moll type currents of the illuminated TJC and allows for the calculation of the spectral response, I(sc), V(oc), FF and eta under variation of one or more of the geometrical and material parameters and 1MeV electron fluence. Results of computer calculations based on this model are presented and discussed. These results indicate that for space applications, both a high beginning of life efficiency, greater than 15% AM0, and a high radiation tolerance can be achieved only with thin (less than 50 microns) TJC's with high base resistivity (greater than 10 ohm-cm).
Positive and negative electrical contacts are on back surface of wraparound-contact solar cell. With both terminals on nonilluminated side, cells can be connected back-to-back, and interconnection of many cells can be automated by using printed-circuit techniques. Cells are made by screen-printing layer of dielectric around edge of cell and extending top contact over dielectric to back surface. Wraparound also facilitates application of transparent covers and encapsulants. Efficiencies of cells are in excess of seventeen percent.
The performance of silicon solar cells with p-n junctions on the nonilluminated surface (i.e., upside-down or back-wall cells) was calculated. These structures consisted of a uniformly shaped p-type substrate layer, a p(+)-type field layer on the front (illuminated) surface, and a shallow, n-type junction on the back (nonilluminated) surface. A four-layer solar cell model was used to calculate efficiency, open-circuit voltage, and short-circuit current. The effect on performance of p-layer thickness and resistivity was determined. The diffusion length was varied to simulate the effect of radiation damage. The results show that peak initial efficiencies greater than 15 percent are possible for cell thicknesses or 100 micrometers or less. After 10 years of radiation damage in geosynchronous orbit, thin (25 to 50 micrometers thick) cells made from 10 to 100 ohm cm material show the smallest decrease (approximately 10 percent) in performance.
The glass resins were deposited by several techniques on 200 micron thick cells and on 50 microns thick wafers. The covered cells were exposed to ultraviolet light in vacuum to an intensity of 10 UV energy-equivalent solar constants at air mass zero for 728 hr. The exposure was followed by a single long thermal cycle from ambient temperature to -150 C. Visual inspection of the samples indicated that all samples had darkened to varying degrees. The loss in short-circuit current was found to range from 8 to 24%, depending on the resin formulation. In another test over 40 glass resin-coated silicon wafers withstood 15 thermal cycles from 100 to-196 C in one or more of the thicknesses tested. Several of the resin-coated wafers were tested at 65 C and 90% relative humidity for 170 hr. No change in physical appearance was detected.
Solar cells with wraparound contacts provide the following advantages in array assembly: (1) eliminate the need for discretely formed, damage susceptible series tabs; (2) eliminate the n gap problem by allowing the use of uniform covers over the entire cell surface; (3) allow a higher packing factor by reducing the additional series spacing formly required for forming, and routing the series tab; and (4) allow the cell bonding to the interconnect system to be a single-side function wherein series contacts can be made at the same time parallel contracts are made.
The paper describes the development of wraparound contact solar cell technology. Future development trends are distinguished. The current method of module fabrication with wraparound contact cells is briefly reviewed. The prospects for automated cell assembly are discussed.
Silicon solar cells and silicon wafers coated with a heat-curable resin consisting of alternating Si-O atoms were subjected to three tests to evaluate the potential utility of this coating in space environments. These included UV irradiation in vacuum at an intensity of 10 air mass zero UV energy-equivalent solar constants for 728 hours followed by a long thermal cycle; 15 thermal shock cycles between 100 C and minus 196 C; and high temperature and humidity (65 C at 90% relative humidity). The UV tests resulted in a 8 to 24% loss in short-circuit current and darkening of the covers. Modification of the resin to provide a better match between the coefficients of expansion of the resin and silicon improved resistance to thermal shock, but also increased the darkening effect under UV irradiation. Silicon wafers coated with the resin were not adversely affected by the temperature/humidity test.
Alteration of surface of silicon solar cell with chemical etchant prior to application of antireflection (AR) coating and encapsulant FEP Teflon, reduces surface reflectivity to 2%. Reduced reflection achieved occurs at all wavelengths resulting in improved short circuit current and spectral response. Improved collection efficiency is also expected.
The performance of epitaxial drift field silicon solar cell structures having a variety of impurity profiles was calculated. These structures consist of a uniformly doped P-type substrate layer, and a P-type epitaxial drift field layer with a variety of field strengths. Several N-layer structures were modeled. A four layer solar cell model was used to calculate efficiency, open circuit voltage and short circuit current. The effect on performance of layer thickness, doping level, and diffusion length was determined. The results show that peak initial efficiency of 18.1% occurs for a drift field thickness of about 30 micron with the doping rising from 10 to the 17th power atoms/cu cm at the edge of the depletion region to 10 to the 18th power atoms/cu cm in the substrate. Stronger drift fields (narrow field regions) allowed very high performance (17% efficiency) even after irradiation to 3x10 to the 14th power 1 MeV electrons/sq cm.
Explore the source record for details and available documents.
Silicon solar cells with macroscopic V-shaped grooves and microscopically texturized surfaces have been made by preferential etching techniques. Various conditions for potassium hydroxide and hydrazine hydrate etching were investigated. Optical reflection losses from these surface were reduced. The reduced reflection occurred at all wavelengths and resulted in improved short circuit current and spectral response. Improved collection efficiency is also expected from this structure due to generation of carriers closer to the cell junction. Microscopic point measurements of collected current using a scanning electron microscope showed that current collected at the peaks of the texturized surface were only 80% of those collected in the valleys.
Open-circuit voltages measured in silicon solar cells made from 0.01 ohm-cm material are 150 mV lower than voltages calculated from simple diffusion theory and cannot be explained by poor diffusion lengths or surface leakage currents. An analytical study was made to determine whether high doping effects, which increase the intrinsic carrier concentration, could account for the low observed voltages and to determine the limits on voltage and efficiency imposed by high doping effects. The results indicate that the observed variation of voltage with base resistivity is predicted by these effects. A maximum efficiency of 19% (AMO) and a voltage of 0.7 volts were calculated for 0.1 ohm-cm cells assuming an optimum diffused layer impurity profile.
The significance of the heavy doping effects (HDE) on the open-circuit voltage of silicon solar cells is assessed. Voltage calculations based on diffusion theory are modified to include the first order features of the HDE. Comparisions of the open-circuit voltage measured for cells of various base resistivities are made with those calculated using the diffusion model with and without the HDE. Results indicate that the observed variation of voltage with base resistivity is predicted by these effects. A maximum efficiency of 19% (AM0) and a voltage of 0.7 volts are calculated for 0.1 omega-cm cells assuming an optimum diffused layer impurity profile.
Silicon solar cells with macroscopic V-shaped grooves and microscopically texturized surfaces were made by preferential etching techniques. Various conditions for potassium hydroxide and hydrazine hydrate etching were investigated. Optical reflection losses from these surface were reduced. The reduced reflection occurred at all wavelengths and resulted in improved short circuit current and spectral response. Improved collection efficiency is also expected from this structure due to generation of carriers closer to the cell junction. Microscopic point measurements of collected current using a scanning electron microscope showed that current collected at the peaks of the texturized surface were only 80 percent of those collected in the valleys.
The conventional surface photovoltage (SPV) method is extended to the measurement of the minority carrier diffusion length in diffused semiconductor junctions of the type used in a silicon solar cell. The minority carrier diffusion values obtained by the SPV method agree well with those obtained by the X-ray method. Agreement within experimental error is also obtained between the minority carrier diffusion lengths in solar cell diffusion junctions and in the same materials with n-regions removed by etching, when the SPV method was used in the measurements.
Recent use of alloying techniques for rear contact formation has yielded a new kind of silicon solar cell, the back surface field (BSF) cell, with abnormally high open-circuit voltage and improved radiation resistance. Several analytical models for open-circuit voltage based on the reverse saturation current are formulated to explain these observations. The zero surface recombination velocity (SRV) case of the conventional cell model, the drift field model, and the low-high junction (LHJ) model can predict the experimental trends. The LHJ model applies the theory of the low-high junction and is considered to reflect a more realistic view of cell fabrication. This model can predict the experimental trends observed for BSF cells.