Excitation Intensity-Dependent Terminal Halide Photosegregation\nStoichiometries in Formamidinium/Cesium Lead Iodide/Bromide [FACsPb(I<sub>1–<i>x</i></sub>Br<sub><i>x</i></sub>)<sub>3</sub>]

A thermodynamic\nband gap-based model for mixed-halide perovskite\nphotosegregation currently explains numerous features of the phenomenon.\nThis includes excitation intensity (<i>I</i><sub>exc</sub>) thresholds for photosegregation as well as <i>I</i><sub>exc</sub>-dependent photosegregation rates and rate constants. An\nintriguing prediction of the model involves <i>I</i><sub>exc</sub>-dependent terminal halide stoichiometries (<i>x</i><sub>terminal</sub>), following photosegregation. Rather than suggest\na common terminal value, e.g., <i>x</i><sub>terminal</sub> = 0.2, for methylammonium lead iodide/bromide [MAPb(I<sub>1–<i>x</i></sub>Br<sub><i>x</i></sub>)<sub>3</sub>], the\nmodel predicts <i>I</i><sub>exc</sub>-dependent <i>x</i><sub>terminal</sub>. This, in principle, allows for the\ncontrolled tuning of mixed-halide perovskite photoresponses. More\nimportant, though, is the opportunity to study this response to develop\ndeeper insight into the origin of nearly ubiquitous photosegregation\nin lead-based, mixed-halide perovskites. Here, we demonstrate <i>I</i><sub>exc</sub>-dependent <i>x</i><sub>terminal</sub> in formamidinium/cesium, mixed-halide [FACsPb(I<sub>1–<i>x</i></sub>Br<sub><i>x</i></sub>)<sub>3</sub>] perovskites.\nWe show that modifications to theory, which account for photosegregated\ndomain subpopulations and photocarrier funneling efficiencies, lead\nto good agreement between measured and predicted <i>I</i><sub>exc</sub>-dependent <i>x</i><sub>terminal</sub> values.\nI-rich phase fractions increase with <i>I</i><sub>exc</sub> and result in asymptotic <i>x</i><sub>terminal</sub> versus <i>I</i><sub>exc</sub>. This addresses an open discrepancy between\nexperiment and theory to advance a detailed understanding of light-induced\ninstabilities in mixed-halide perovskites.