Performance comparison of III-V-on-Si tandem solar cells in the 2-terminal, 3-terminal and 4-terminal configurations

Abstract

Tandem (or multijunction) solar cells are the only photovoltaic (PV) devices that have demonstrated efficiencies above 30%. Hybrid tandem solar cells consist of dissimilar PV materials, each layer optimized to absorb a specific portion of the solar spectrum. Tandem cells are often designed with two terminals (2T), requiring current-matched subcells connected in series. They can, however, be designed with additional terminals, to circumvent the need for current matching, which enables greater design flexibility. This research investigates the performance and device physics of III-V-on-Silicon (III-V//Si) tandem solar cells operated in the 2-terminal (2T), 3-terminal (3T) and 4-terminal (4T) configurations. Previously, record efficiency 4T III-V//Si tandem cells were fabricated at NREL with top cells grown via metalorganic vapor phase epitaxy (MOVPE). Techno-economic analysis, however, shows that costs could be substantially reduced by using a lower-cost III-V deposition technique, such as hydride vapor phase epitaxy (HVPE). To explore tandem cell performance using HVPE-grown top cells, 4T GaAs//Si tandem cells were fabricated and analyzed using the same top cell structure grown by each method. Optical modeling was used to simulate possible performance improvements. Simulations show that 3T designs can also extract surplus current from mismatched subcells, yet 3T tandem designs avoid losses from intermediate conductive layers required for 4T tandems. To examine this experimentally, the performance of a GaInP//Si tandem with minimal current mismatch was compared to that of a GaAs//Si tandem with substantial mismatch, both operated in the 2T and 3T configurations. Both tandems cells exhibited comparable efficiencies when operated in the 3T configuration, demonstrating that the third terminal expands top cell material options, similar to 4T designs. The operational complexities of a 3T tandem cell will also be addressed. Future research priorities for III-V//Si tandem cells will also be discussed, with particular emphasis on reliability of the 3T and 4T cell designs and how they can be interconnected at the module level. This thesis will provide an introduction to a project currently underway to test a 4T GaInP//Si cell on the International Space Station, to study degradation mechanisms that effect current mismatched subcells with sustained exposure to the space environment.