Ex ante life cycle assessment of GaAs/Si nanowire–based tandem solar cells: a benchmark for industrialization

Abstract

Abstract Purpose The goal of this study is to perform an ex-ante life cycle assessment (LCA) of the emerging gallium-arsenide nanowire tandem solar cells on silicon (GaAs/Si) and to provide a benchmark for the commercialization of the technology. The environmental impacts and energy payback time (EPBT) of the GaAs/Si modules are compared with those of the incumbent single-Si modules. Parameters and efficiencies most relevant to be optimized in order to commercialize the technology are identified and discussed. Methods Two production routes for GaAs/Si solar cells are being up-scaled: the growth of GaAs nanowires on a native substrate, peel-off, and transfer to a silicon substrate (transfer route) and the direct growth of GaAs nanowires on a silicon substrate with assistance of a silicon-dioxide (SiO2) nanotube template (direct growth route). Two ex-ante LCAs for the different manufacturing routes and an LCA for the incumbent single-Si technology were conducted. Environmental impacts of the GaAs/Si technology were assessed and compared with the incumbent. Various scenarios regarding sensitive parameters and processes were modeled—such as modeling several industrial scale tools, the energy consumption of sensitive processes, the number of substrate reuses, the frequency of re-polishing the wafer, and benchmarking the scale of improvement of major impact drivers. Results and discussion The analysis showed that, if expected process efficiencies are achieved, a 28% efficient GaAs/Si module performs 5 to 20% better (transfer route) and 20 to 30% better (direct growth route, except the ozone depletion impact) compared with an 18% efficient single-Si module, for all impact categories assessed—climate change, land use, acidification, ozone depletion, freshwater, marine, terrestrial ecotoxicity, eutrophication, human toxicity, and photochemical oxidation. Critical hotspots identified include the use of gold, trifluoromethane (CHF3), and a GaAs wafer. The EPBT of the GaAs/Si nanowire tandem module is in between 1.37 (expected process efficiencies achieved) and 1.9 years (worst case scenario), while the EPBT of the single-Si module is 1.84 years. Results can be considered as a benchmark for the successful commercialization of the technology. Conclusions If 28% efficient GaAs/Si nanowire tandem modules are developed, expected process efficiencies are achieved, and at least 100 reuses of the GaAs substrate (transfer route) are realized; then, the GaAs/Si modules perform better compared with an 18% efficient single-Si module for most impact categories assessed. Conclusions from the ex-ante LCA are conditional (if-then) and can be used as a benchmark, allowing to quantify the efficiencies that need to be achieved to commercialize the technology.