Chemical Stability of Perovskite Semiconductors: Kinetics of Degradation and Machine Learning Models of Device Lifetime

Understanding the chemical reactions that hybrid organic-inorganic halide perovskite (HP) semiconductors undergo in the presence of moisture, oxygen, and light are essential to the commercial development of HP solar cells and optoelectronics. The presentation will cover recent experiments that have determined the dominant degradation mechanism, developed quantitative kinetic rate laws, and demonstrated machine learning models that utilize the kinetic expressions are capable of forecasting device lifetime. The core dataset for the kinetic model is based on in-situ optical absorbance measurements during 105 degradations conducted over 41 unique environmental conditions. The data are used to quantify the kinetics of methylammonium lead iodide (MAPbI₃) degradation in response to combinations of moisture, oxygen, and illumination over a range of temperatures. We discover and report a new chemical reaction pathway referred to as a water-accelerated photo-oxidation (WPO) pathway, which is the dominant chemical degradation pathway when light, oxygen, and humidity are present (and much faster than other dry photooxidation pathways, DPOs). The overall rate of degradation is very sensitive to small amounts of water, which behaves effectively like catalyst. The WPO process has significant implications for perovskite stability and encapsulation, which will be discussed. In addition, the presentation will include recent stability results on the broader (FA, MA, Cs)Pb(I,Br)3 space and details of a machine learning approach that uses the kinetic rate expressions as features in a predictive model of perovskite solar cell lifetime (T80).