Defect‐mediated carrier recombination at the interfaces between perovskite and neighboring charge transport layers limits the efficiency of most state‐of‐the‐art perovskite solar cells. Passivation of interfacial defects is thus essential for attaining cell efficiencies close to the theoretical limit. In this work, a novel double‐sided passivation of 3D perovskite films is demonstrated with thin surface layers of bulky organic cation–based halide compound forming 2D layered perovskite. Highly efficient (22.77%) mixed‐dimensional perovskite devices with a remarkable open‐circuit voltage of 1.2 V are reported for a perovskite film having an optical bandgap of ≈1.6 eV. Using a combination of experimental and numerical analyses, it is shown that the double‐sided surface layers provide effective defect passivation at both the electron and hole transport layer interfaces, suppressing surface recombination on both sides of the active layer. Despite the semi‐insulating nature of the passivation layers, an increase in the fill factor of optimized cells is observed. The efficient carrier extraction is explained by incomplete surface coverage of the 2D perovskite layer, allowing charge transport through localized unpassivated regions, similar to tunnel‐oxide passivation layers used in silicon photovoltaics. Optimization of the defect passivation properties of these films has the potential to further increase cell efficiencies.

Mixed‐dimensional perovskite solar cells combining 3D and 2D perovskites have recently attracted wide interest owing to improved device efficiency and stability. Yet, it remains unclear which method of combining 3D and 2D perovskites works best to obtain a mixed‐dimensional system with the advantages of both types. To address this, different strategies of combining 2D perovskites with a 3D perovskite are investigated, namely surface coating and bulk incorporation. It is found that through surface coating with different aliphatic alkylammonium bulky cations, a Ruddlesden–Popper “quasi‐2D” perovskite phase is formed on the surface of the 3D perovskite that passivates the surface defects and significantly improves the device performance. In contrast, incorporating those bulky cations into the bulk induces the formation of the pure 2D perovskite phase throughout the bulk of the 3D perovskite, which negatively affects the crystallinity and electronic structure of the 3D perovskite framework and reduces the device performance. Using the surface‐coating strategy with n‐butylammonium bromide to fabricate semitransparent perovskite cells and combining with silicon cells in four‐terminal tandem configuration, 27.7% tandem efficiency with interdigitated back contact silicon bottom cells (size‐unmatched) and 26.2% with passivated emitter with rear locally diffused silicon bottom cells is achieved in a 1 cm2 size‐matched tandem.

We present a low-temperature post-processing module, utilizing polyimide as a sacrificial layer and novel materials such as PECVD SiC and metals (sputtered aluminium and titanium) as structural layers. The use of spin-on polyimide allows an all-dry final release step overcoming stiction problems often encountered in wet sacrificial etching processes. The spinning and curing procedure has been tailored to the specific needs of the IC-compatible post-process module. For the patterning of the polyimide, thin films of aluminium, PECVD silicon oxide or silicon carbide are employed as a mask layer. Anisotropic etching of the mask film and of the polyimide layer is accomplished by RIE. After patterning the structural layer, sacrificial etching of the polyimide is done using an isotropic dry etch process in high-density oxygen plasma. An underetch rate of 4 μm min−1 is achieved. Compatibility with different structural materials is tested and test structures are designed and realized in a fully post-processing surface micromachining module.

No abstract

Incorporating 2.5% Cs in FA0.8MA0.2Sn0.5Pb0.5I3 improves the photo-stability of the low-bandgap perovskite solar cells. The champion device with power conversion efficiency of 18.9% maintain 92% of its initial efficiency after 120 min MPP tracking.