Tin oxide (SnO2) is an emerging electron transport layer (ETL) material in halide perovskite solar cells (PSCs). Among current limitations, open‐circuit voltage (VOC) loss is one of the major factors to be addressed for further improvement. Here a bilayer ETL consisting of two SnO2 nanoparticle layers doped with different amounts of ammonium chloride is proposed. As demonstrated by photoelectron spectroscopy and photophysical studies, the main effect of the novel ETL is to modify the energy level alignment at the SnO2/perovskite interface, which leads to decreased carrier recombination, enhanced electron transfer, and reduced voltage loss. Moreover, X‐ray diffraction reveals reduced strain in perovskite layers grown on bilayer ETLs with respect to single‐layer ETLs, further contributing to a decrease of carrier recombination processes. Finally, the bilayer approach enables the more reproducible preparation of smooth and pinhole‐free ETLs as compared to single‐step deposition ETLs. PSCs with the doped bilayer SnO2 ETL demonstrate strongly increased VOC values of up to 1.21 V with a power conversion efficiency of 21.75% while showing negligible hysteresis and enhanced stability. Moreover, the SnO2 bilayer can be processed at low temperature (70 °C), and has therefore a high potential for use in tandem devices or flexible PSCs.

Hybrid perovskites have attracted much attention as promising photovoltaic materials in the past few years. However, the fundamental understanding of their crystallization behavior lags far behind the pace of empirical solar cell efficiency improvement. Methylammonium iodide (MAPbI 3 ) is a widely studied reference compound whose solar cell performance can be improved by chloride addition (e.g., in the form of PbCl 2 ) during the thin-film preparation. Because of the large difference in the ionic radii of both halides, no mixed perovskites MAPbI 3−x Cl x are formed and generally only minute amounts of chlorine can be detected in the final MAPbI 3 thin films. Here, we demonstrate by means of a variety of complementary X-ray diffraction (XRD) techniques that, unexpectedly, the formation mechanism proceeds via an initial MAPbCl 3 layer, which subsequently transforms to MAPbI 3 in an anion exchange reaction during the thermal annealing step, completing the thin-film preparation. The perovskite lattice is highly strained along the process, much more than what is expected from the sole effect of the difference between the thermal expansion coefficients of the perovskite and the substrate. At room temperature, the existence of a double [hh0]/[00l] texture is explained by the ferroelastic character of the cubic/tetragonal transition of MAPbI 3 , which induces the formation of twins. The relative population of these domains is correlated to their strain level. Although strain is known to weaken the stability of the MAPbI 3 phase, our results unambiguously show that it also favors the reproducibility of the thin-film microstructure. When used as active layers in solar cells, the dependence of the cell efficiency and stability on the annealing time is in striking accordance with the formation kinetics of MAPbI 3 , as revealed by the XRD measurements. Therefore, the understanding of the crystallization behavior achieved with the present approach, applicable also to other types of metal halide perovskites, allows for the rational optimization of the device performance and long-term stability.