Development of III-V/Si Multijunction Space Photovoltaics

The goal of this research is to develop multijunction III-V/Si solar cells for space application. This has been approached via iterative cycles of various studies comprising epitaxial growth of lattice-mismatched III-V/Si materials, fundamental analysis and characterization of material defects, photovoltaic (PV) device design and characterization. Relative to all-III-V multijunction technologies, which represent the state-of-the-art for space PV, the use of a Si subcell can reduce manufacturing costs. However, to realize this cost advantage, AM0 efficiencies of III-V/Si operating in space environments must be increased. To this end, we have realized >5% absolute AM0 efficiency gains (up to 22% as demonstrated Q2 YR4) over the course of this project in demonstrated monolithic dual-junction (2J) III-V/Si photovoltaic devices. Furthermore, a straightforward path to 27% AM0 efficiencies is now enabled by a recent advances in epitaxial growth processes that afford a substantial increase in achievable material quality, an issue that had otherwise been the incumbent limiter of monolithic III-V/Si performance. The material quality metric of greatest interest in this context is the threading dislocation density (TDD) in the III-V top cell layers, which are largely the result of inherent difficulties in the epitaxial integration of dissimilar III-V with Si. To address this issue, this project partially focused on thedevelopment of a new level of fundamental understanding and control of the introduction and evolution of dislocations, a detrimental form of crystalline defect, during epitaxial growth of III-V/Si materials. This effort consisted of intensive epitaxial growth and characterization campaigns, with rapid feedback on dislocation populations provided by electron channeling contrast imaging (ECCI) to inform metalorganic chemical vapor deposition (MOCVD) growth process studies. As a result, we have achieved an order of magnitude reduction in TDD in the GaAs0.75P0.25¬ top cells on Si, down to <3×106 cm-2, which is among the lowest TDD values for III-V/Si reaching this lattice constant. At this TDD level, minority carrier diffusion length and open-circuit voltage can be drastically improved beyond the that of the 22% demonstrated devices, leading to potential 2J efficiencies of >27% AM0, provided commensurate optimization of cell design and components at this recently achieved low-TDD. Full tandem devices have been produced in multiple iterations over the course of this project by leveraging collaborations with Si PV experts at UNSW and III-V PV experts at SolAero Technologies, in addition to our own extensive expertise and capabilities. Both partners have demonstrated industrially viable fabrication processes compatible with this PV technology – specifically Si subcells with diffusions and optical structures tailored to a GaAsP-filtered spectrum (UNSW) and full 4” wafer fabrication and antireflection coating of completed 2J devices (SolAero). In addition to this successful integration with industrial processes and tooling, epitaxial growth processes were all developed on a commercial MOCVD reactor at OSU using 4” wafers. Preliminary space readiness of III-V/Si devices was assessed in collaboration with NASA Glenn Research Center, including aspects of thermally dependent performance, thermal cycling, and thermal loading in LEO conditions. This effort has culminated in sending a 2J device on MISSE-13. Once returned, this year of spaceflight will yield valuable information for future III-V/Si PV development regarding thermal cycle and radiation exposure. This research serves to impact low-budget, short duration missions, where cost and area constraints for PV power generation are important factors. In addition, this work provides knowledge transfer to other high-efficiency space photovoltaics. Lastly, a broad swath of technologies stand to benefit from the III-V/Si materials and device advances that we have demonstrated through this work.