Search

Halide perovskites are promising photovoltaic (PV) materials with the potential to lower the cost of electricity and greatly expand the penetration of PV if they can demonstrate long-term stability under illumination in the presence of moisture and oxygen. The solar cell service lifetime, as quantified by T80 (the time required for the power conversion efficiency to drop to 80% of its starting value), for utility, commercial, or residential PV systems needs to be several decades in order to yield low-cost electricity, and thus it is not practical to directly measure it. It would be useful if T80 could be predicted from the initial dynamics of a solar cell's performance, but until now no models have been developed to do so. In this work, we report the development of machine learning models to predict T80 of ITO/NiOx/CH3NH3PbI3/C60/BCP/Ag solar cells operating at maximum power point under 1-sun equivalent photon flux in air at varying temperatures and relative humidities. Efficiency losses are driven by short-circuit current and fill factor, indicating that photochemical reactions with O2 and H2O are a major contributor to degradation. Spatial patterns evident from in situ dark field optical microscopy also suggest that the electric field gradient at device edges plays a significant role in perovskite decomposition. Models are trained using a menu of features from three distinct categories: (i) measurements of the initial rates of change of device parameters, (ii) ambient conditions during operation (temperature & partial pressure of H2O), and (iii) features based on underlying physics and chemistry. We show that a theory-based physiochemical feature derived from a model of the chemical reaction kinetics of the rate of degradation of CH3NH3PbI3 is particularly valuable for prediction and was selected as the most dominant feature in the best performing models. With a dataset consisting of 45 degradation experiments with T80 values ranging over a factor of almost 30, the model predicts T80 with an average accuracy of about 40% on samples not used in training. This hybrid ML approach should be effective when applied to other compositions, device architectures, and advanced packaging schemes. [ABSTRACT FROM AUTHOR]