Effectively Increasing Pt Utilization Efficiency of the Membrane Electrode Assembly in Proton Exchange Membrane Fuel Cells through Multiparameter Optimization Guided by Machine Learning

Advances in Low Pt Loading Membrane Electrode Assembly for Proton Exchange Membrane Fuel Cells

Hydrogen has the potential to be one of the solutions that can address environmental pollution and greenhouse emissions from traditional fossil fuels. However, high costs hinder its large-scale commercialization, particularly for enabling devices such as proton exchange membrane fuel cells (PEMFCs). The precious metal Pt is indispensable in boosting the oxygen reduction reaction (ORR) in cathode electrocatalysts from the most crucial component, i.e., the membrane electrode assembly (MEA). MEAs account for a considerable amount of the entire cost of PEMFCs. To address these bottlenecks, researchers either increase Pt utilization efficiency or produce MEAs with enhanced performance but less Pt. Only a few reviews that explain the approaches are available. This review summarizes advances in designing nanocatalysts and optimizing the catalyst layer structure to achieve low-Pt loading MEAs. Different strategies and their corresponding effectiveness, e.g., performance in half-cells or MEA, are summarized and compared. Finally, future directions are discussed and proposed, aiming at affordable, highly active, and durable PEMFCs.

Charge Recombination Dynamics in a Metal Halide Perovskite Simulated by Nonadiabatic Molecular Dynamics Combined with Machine Learning

Nonadiabatic coupling (NAC) plays a central role in driving nonadiabatic dynamics in various photophysical and photochemical processes. However, the high computational cost of NAC limits the time scale and system size of quantum dynamics simulation. By developing a machine learning (ML) framework and applying it to a traditional CH₃N₃PbI₃ perovskite, we demonstrate that the various ML algorithms (XGBoost, LightGBM, and random forest) combined with three descriptors (sine matrix, MBTR, and SOAP) can predict accurate NACs that all agree well with the direct calculations, particularly for the combination of LightGBM and sine matrix descriptor showing the best performance with a high correlation coefficient of ≤0.87. The simulated nonradiative electron-hole recombination time scales agree well with each other between the NACs obtained from direct calculations and ML prediction. The study shows the advantage in accelerating quantum dynamics simulations using ML algorithms.