Accelerated Electrocatalyst Degradation Testing by Accurate and Robust Forecasting of Multidimensional Kinetic Model with Bayesian Data Assimilation
Miao Wang, Akimitsu Ishii, Ken Sakaushi
Abstract
High Resolution Image Download MS PowerPoint Slide Degradation tests represent a significant bottleneck in electrochemical technology development, occasionally requiring tens of thousands of hours. Thus, reliable degradation forecasting in a short time frame is a game-changer in accelerating the establishment of future electrochemical devices. Herein, we show a multidimensional kinetic model for electrocatalyst degradation by quantifying the relationship among potential, current, and time, applicable under various conditions. Aiming to predict reliable degradation behaviors in shorter experimental timeframes and inspired by modern weather forecasting methods, we integrated Bayesian data assimilation with our model to expedite multidimensional parameter optimization. Consequently, we achieved accurate and robust forecasting of electrocatalyst lifetime by employing oxygen evolution reaction as a representative system: it takes just 300 h to obtain the final lifetime of close to 1000 h even with environmental noise. This data-driven approach can accelerate our understanding of the microscopic electrochemical mechanisms and simultaneously directly bridge this understanding to develop next-generation energy technologies.