Are Mixed-Halide Ruddlesden-Popper Perovskites Really Mixed?

Mixing bromine and iodine within lead halide perovskites is a common strategy to tune their optical properties. This comes at the cost of instability, as illumination induces halide segregation and degrades device performances. Hence, understanding the behavior of mixed-halide perovskites is crucial for applications. In 3D perovskites such as MAPb(Br x I 1- x ) 3 (MA = methylammonium), all of the halide crystallographic sites are similar, and the consensus is that bromine and iodine are homogeneously distributed prior to illumination. By analogy, it is often assumed that Ruddlesden-Popper layered perovskites such as (BA) 2 MAPb 2 (Br x I 1- x ) 7 (BA = butylammonium) behave alike. However, these materials possess a much wider variety of halide sites featuring diverse coordination environments, which might be preferentially occupied by either bromine or iodine. This leaves an open question: are mixed-halide Ruddlesden-Popper perovskites really mixed? By combining powder and single-crystal diffraction experiments, we demonstrate that this is not the case: bromine and iodine in RP perovskites preferentially occupy different sites, regardless of the crystallization speed.