Search

The primary objective of the Silicon Sheet Task of the Flat-Plate Solar Array (FSA) Project was the development of one or more low cost technologies for producing silicon sheet suitable for processing into cost-competitive solar cells. Silicon sheet refers to high purity crystalline silicon of size and thickness for fabrication into solar cells. Areas covered in the project were ingot growth and casting, wafering, ribbon growth, and other sheet technologies. The task made and fostered significant improvements in silicon sheet including processing of both ingot and ribbon technologies. An additional important outcome was the vastly improved understanding of the characteristics associated with high quality sheet, and the control of the parameters required for higher efficiency solar cells. Although significant sheet cost reductions were made, the technology advancements required to meet the task cost goals were not achieved.

A Workshop on Crystal Growth for High-Efficiency Silicon Solar Cells was held December 3 and 4, 1984, in San Diego, California. The Workshop offered a day and a half of technical presentations and discussions and an afternoon session that involved a panel discussion and general discussion of areas of research that are necessary to the development of materials for high-efficiency solar cells. Topics included the theoretical and experimental aspects of growing high-quality silicon crystals, the effects of growth-process-related defects on photovoltaic devices, and the suitability of various growth technologies as cost-effective processes. Fifteen invited papers were presented, with a discussion period following each presentation. The meeting was organized by the Flat-Plate Solar Array Project of the Jet Propulsion Laboratory. These Proceedings are a record of the presentations and discussions, edited for clarity and continuity.

The state of the art in the growth of silicon crystals for high-efficiency solar cells are reviewed, sheet requirements are defined, and furture areas of research are identified. Silicon sheet material characteristics that limit cell efficiencies and yields were described as well as the criteria for the ideal sheet-growth method. The device engineers wish list to the material engineer included: silicon sheet with long minority carrier lifetime that is uniform throughout the sheet, and which doesn't change during processing; and sheet material that stays flat throughout device processing, has uniform good mechanical strength, and is low cost. Impurities in silicon solar cells depreciate cell performance by reducing diffusion length and degrading junctions. The impurity behavior, degradation mechanisms, and variations in degradation threshold with diffusion length for silicon solar cells were described.

Minority carrier lifetime distributions in silicon web dendrites are measured. Emphasis is placed on measuring areal homogeneity of lifetime, show its dependency on structural defects, and its unique change during hot processing. The internal gettering action of defect layers present in web crystals and their relation to minority carrier lifetime distributions is discussed. Minority carrier lifetime maps of web dendrites obtained before and after high temperature heat treatment are compared to similar maps obtained from 100 mm diameter Czochralski silicon wafers. Such maps indicate similar or superior areal homogeneity of minority carrier lifetime in webs.

Technique yields better mixing of impurities and superior qualiity crystals. Accellerated motion stirs melt which reduces temperature gradients and decreases boundary layer for diffusion of impurities near growing surface. Results better mixing of impurities into melt, decrease in tendency for dendritic growth or cellular growth and crystals with low dislocation density. Applied with success to solution growth and Czochralski growth, resulting in large crystals of superior quality.

The distribution of slip dislocations in silicon dendritic web ribbons due to plastic deformation during the cooling phase of the growth was studied. The results show the existence of two distinguishable stress regions across the ribbon formed during the plastic deformation stage, namely, shear stress at the ribbon edges and tensile stress at the middle. In addition, slip dislocations caused by shear stress near the edges appear to originate at the twin plane.

The open-circuit voltage and short-circuit current of large-grain (1 to 10 mm grain diameter) polycrystalline silicon solar cells is determined by the minority-carrier diffusion length within the bulk of the grains. This was demonstrated by irradiating polycrystalline and single-crystal (Czochralski) silicon solar cells with 1 MeV electrons to reduce their bulk lifetime. The variation of short-circuit current with minority-carrier diffusion length for the polycrystalline solar cells is identical to that of the single-crystal solar cells. The open-circuit voltage versus short-circuit current characteristic of the polycrystalline solar cells for reduced diffusion lengths is also identical to that of the single-crystal solar cells. The open-circuit voltage of the polycrystalline solar cells is a strong function of quasi-neutral (bulk) recombination, and is reduced only slightly, if at all, by grain-boundary recombination.

The quality of silicon cast by present techniques is limited by the presence of dislocations and grain boundaries in unseeded growth and by cellular structures with dislocation networks in the case of the seeded growth. To address these concerns, a new method of directional solidification called the oscillating crucible technique (OCT) is developed. During growth, a carbon crucible is oscillated to provide for effective stirring of the melt. This growth technique (seeded growth only), along with material characterization and solar-cell fabrication and testing, is described. Solar-cell efficiencies of up to 13 percent at 100 mW/sq cm area obtained in the single crystalline areas. Minority-carrier diffusion lengths exceeding 100 microns are measured even in the polycrystalline areas of the wafers. Limitations of the present setup and possible future improvements are discussed.

The Heat Exchanger Method (HEM) is a promising low-cost ingot casting process for material used for solar cells. This is the only method that is capable of casting single crystal ingots with a square cross section using a directional solidification technique. This paper describes the chemical, mechanical and electrical properties of the HEM silicon material as a function of position within the ingot.

Most of the methods which have been developed for the measurement of the minority carrier diffusion length of silicon wafers require that the material have either a Schottky or an ohmic contact. The surface photovoltage (SPV) technique is an exception. The SPV technique could, therefore, become a valuable diagnostic tool in connection with current efforts to develop low-cost processes for the production of solar cells. The technique depends on a knowledge of the optical absorption coefficient. The considered investigation is concerned with a reevaluation of the absorption coefficient as a function of silicon processing. A comparison of absorption coefficient values showed these values to be relatively consistent from sample to sample, and independent of the sample growth method.