Mixed cation metal halide perovskites with increased power conversion efficiency, negligible hysteresis, and improved long term stability under illumination, moisture, and thermal stressing have emerged as promising compounds for photovoltaic and optoelectronic applications. Here, we shed light on photoinduced halide demixing using insitu photoluminescence spectroscopy and synchrotron Xray diffraction (XRD) to directly compare the evolution of composition and phase changes in CH(NH 2 ) 2 CsPbhalide (FACsPb) and CH 3 NH 3 Pbhalide (MAPb) perovskites upon illumination, thereby providing insights into why FACsPbhalides are less prone to halide demixing than MAPbperovskites. We find that halide demixing occurs in both materials.However, the formed Irich domains accumulate strain for the case of FACsPbperovskites but readily relax for the case of MAPbperovskites. The accumulated strain energy is expected to act as a stabilizing force against halide demixing and may explain the higher Br composition threshold for demixing to occur in FACsPbhalides. In addition, we find that while halide demixing leads to a quenching of the high energy photoluminescence emission from MA 2 perovskites, the emission is enhanced for the case of FaCsperovskites. This behavior points to a reduction of nonradiative recombination centers in FACsperovskites arising from the demixing process. FACsPbhalide perovskites exhibit excellent intrinsic material properties, with photoluminescence quantum yields that are comparable to MAperovskites. Since improved stability is achieved without sacrificing electronic properties, these compositions are better candidates for photovoltaic applications, especially as wide bandgap absorbers in tandem cells. , and high photoluminescence quantum yields 2,4 . Their general crystal structure is described by ABX 3 , typically comprising a monovalent organic cation A (e.g. Despite the importance of overcoming halide demixing for achieving stable perovskitebased photovoltaic devices, there remains uncertainty about the underlying mechanism(s) and most studies have focused on MAperovskites. Currently, strain or carrierinduced lattice distortion, 6 compositional inhomogeneity, 7 defectmediated halide migration, 6,8,9 and crystal domain size 10 are actively considered as contributing to halide segregation. In particular, Bischak et al. propose that halide demixing is a consequence of localized strain generated from the interaction of charge carriers with the lattice (polaron formation).
6In this respect, they find that the combination of 4 mobile halides, long charge carrier lifetimes, and significant electronphonon coupling are prerequisites for halide demixing. 6 In a different study, Barker et al. suggest that defectassisted halide ion migration away from the illuminated surface, with a slower hopping rate of iodide and a potential dependence on charge carrier generation gradients, results in formation of Irich regions at the surface. In this explanation, they argue that halide segregation in a single cr...