Interstitial Defect Thermodynamics Drive Phase Separation in Lead Halide Perovskites

Lead halide perovskites show unrivaled performance and versatility in optoelectronic devices; however, the issues of photo- and electrochemical stability and halide phase segregation remain a challenge to take full advantage of halide perovskites. Here, we describe our proposed model to explain how homogeneously mixed iodide (I):bromide (Br) perovskite alloys (e.g. MAPbBr_xI_1-x) phase separate into I-rich and Br-rich regions under bias or illumination. The model predicts iodine +1 interstitials play a critical, intrinsic role giving rise to unequal fluxes of halide species – the origin of voltage- or photo-induced compositional instabilities. We briefly review our recent experimental results testing the model and discuss in detail our recent density functional theory (DFT) computations comparing the relative formation enthalpies of I_i⁺ versus Br_i⁺ interstitial defects in different halide compositions. We find that I_i⁺ defects are energetically favorable by 0.1-0.4 eV over Br_i⁺ under most conditions, and also show that I_i⁺ is further stabilized when surrounded by iodide first- and second-nearest neighbors. In other words, I_i⁺ will be most stable and favorably migrate towards iodide-rich regions making them more iodide-rich. Thus, we confirm our model’s core assumption and elucidate a major fundamental, thermodynamic driving force for halide phase separation following oxidation and creation of I_i⁺ defects. Overall, this body of work brings a level of clarity to interstitial defect chemistry and physics in halide perovskites, advancing our understanding a step closer to that of more conventional semiconductors.