This report reviews the rationale for selection of the two-step recrystallization process that is being developed, presents the results of the latest first-step recrystallization experiments, and outlines the calculations appropriate to the design of second-step thermal process experiments. Results of a computer run for first-step recrystallization are also presented. The report concludes with an updating of the results obtained from the thermal and pulsed electron beam annealing of shallow implanted GaAs.

The program had two thrusts: (1) based upon electron-beam thermal treatment of deposited silicon films, to increase crystallite sizes to the range thought to be useful for polycrystalline, thin-film cell fabrication; and (2) to explore the feasibility of applying the directed-energy technologies of ion implantation and pulsed electron beam activation, previously developed for silicon cell fabrication, to junction formation in III-V compounds. The culmination of the recrystallization effort was demonstrating grains broader than the 30-.mu.m film in which they were regrown. This proof of principle was accomplished by means of two-step thermal process that consisted of large-area pulsed electron beam melting followed by small-area heating in a moving DC electron beam. The pulsed beam treatment reduced the three-dimensional disorder of the initial submicrometer crystallite silicon film to one characterized by submicrometercross-section, full-film-thickness, columnar crystallites. The swept beam treatment allowed coalesence of these columnar crystallites, through directional freezing, in the melt path of the beam. It is believed that this demonstration is the first evidence of greater-than-film thickness recrystallization of useful thickness silicon films other than by extended heat treatment at greater than 1350°C. The results of the studies on junction formation in III-V materials, while not so dramatic, have shown that low-energy ion implantation is a potentially viable alternative to liquid or vapor phase epitaxy in the fabrication of GaAs solar cells. Further, the technical feasibility of pulsed electron beam activation of ion implanted junctions in GaAs has been demonstrated. Lastly, the concept of forming front-layer windows of GaP and AlGaAs on GaAs by high-dose ion implantation has been shown to be technically feasible.