Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Silicon materials task of the low-cost solar-array project. Effect of impurities and processing on silicon solar cells. Final report

Technical Report ·
DOI:https://doi.org/10.2172/5358007· OSTI ID:5358007
The object of the program has been to investigate the effects of various processes, metal contaminants, and contaminant-process interactions on the properties of silicon and on the performance of terrestrial silicon solar cells. The study has encompassed topics such as thermochemical (gettering) treatments, base-doping concentration, base-doping type (n vs. p), grain boundary-impurity interaction in polycrystalline devices, and long-term effects of impurities and impurity impacts on high-efficiency cells, as well as a preliminary evaluation of some potential low-cost silicon materials. The effects have been studied of various metallic impurities, introduced singly or in combination into Czochralski, float zone, and polycrystalline silicon ingots and into silicon ribbons grown by the dendritic web process. The solar cell data indicate that impurity-induced performance loss is caused primarily by a reduction in base diffusion length. An analytical model based on this observation has been developed and verified experimentally for both n- and p-base material. Studies of polycrystalline ingots containing impurities indicate that solar cell behavior is species sensitive and that a fraction of the impurities are segregated to the grain boundaries. HCl and POCl gettering improve the performance of single-crystal solar cells containing Fe, Cr, and Ti. In contrast Mo-doped material is barely affected. The efficiencies of solar cells fabricated on impurity-doped wafers is lower when the front junction is formed by ion implantation than when conventional diffusion techniques are used. For most impurity-doped solar cells stability is expected for projected times beyond 20 years. Feedstock impurity concentrations below one part per million for elements like V, or 100 parts per million for more benign impurities like Cu or Ni, will be required.
Research Organization:
Westinghouse Research and Development Center, Pittsburgh, PA (USA)
DOE Contract Number:
NAS-7-100-954331
OSTI ID:
5358007
Report Number(s):
DOE/JPL/954331-82/1; ON: DE82009390
Country of Publication:
United States
Language:
English