Increasing the efficiency of solar cells further is now the most direct avenue to make solar electricity even cheaper and ease the energy transition. This involves designing solar cells that achieve efficiencies beyond the practical limit of 27% of crystalline silicon (c-Si) solar cells, the technology dominating the photovoltaics market. In addition, this solar cell of tomorrow should be processed at low costs, employ earth-abundant elements and exhibit a long life time in the field. One emerging cell design may meet these criteria: multi-junction solar cells combining metal halide perovskites and c-Si. The tuneable bandgap, soft processing conditions and high single-junction performance of perovskite cells indicate that a c-Si solar cell could be upgraded with one or more perovskite top cell(s) to reach efficiencies >30% through a few extra process steps and hence at low additional process costs.

Fulfilling this potential is first a processing challenge. For maximum performance and compatibility with existing c-Si process flows, the perovskite solar cell should be deposited directly on the textured front side of the c-Si solar cell, a texture that improves light management in the c-Si. But this pyramidal texture is not compatible with standard perovskite solution-based deposition protocols as these lead to non-conformal coatings and electrical shunt paths. A hybrid evaporation/solution perovskite processing route was developed at EPFL PV-Lab to enable the conformal deposition of perovskite top cell(s) on the pyramidal texture of c-Si cells to achieve high photocurrents (Panels a of TOC). This contribution will discuss how a careful analysis of the device structure on the nanometre scale enabled the identification of performance-loss mechanisms and helped guiding processing efforts. High-efficiency (>25%) perovskite/c-Si tandems were demonstrated on different n- and p-type c-Si technologies [1,2], notably by identifying i) optimal bottom cell contact structures [3], ii) crystallographic and chemical features enabling the recombination junction to quench shunts [4], iii) hole-selective contact instabilities depending on recombination junction and perovskite crystallisation conditions (Panel b). This contribution will then elaborate on the next performance-loss mechanisms that should be addressed to achieve an efficiency of >30%.

The second challenge, and likely the most difficult one to tackle before any commercialisation of the technology can be envisaged, is related to the instability of perovskite solar cells. Degradation pathways triggered by reverse voltages, which may appear when a module becomes partially shaded, or during long-term operation at maximum power point at various temperatures will be discussed [5]. These emphasise the dynamic nature of the perovskite nanostructure (ionic migration within the absorber and also into the contacts (Panel c), volatilization of species, crystallographic phase change/decomposition, shunt formation due to metal migration) depending on the external stimuli and its influence on the solar cell performance. Overall, these findings demonstrate that opaque solar cell architectures commonly employed by the community are particularly unstable in reverse bias, prompting the need for urgent research efforts.