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

Abstract

Although proton exchange membrane fuel cells have received attention, the high costs associated with Pt-based catalysts in membrane electrode assemblies (MEAs) remain a huge obstacle for large-scale applications. To solve this urgent problem, the utilization efficiency of Pt in MEAs must be increased. Facing numerous interacting parameters in an attempt to keep experimental costs as low as possible, we innovatively introduce machine learning (ML) to achieve this goal. Nine different ML algorithms are trained on the experimental dataset from our laboratory to precisely predict the performance and Pt utilization (maximum R 2 = 0.973/0.968). To determine the best synthesis conditions, black-box interpretation methods are applied to provide reliable insights from both qualitative and quantitative perspectives. The optimized choices of ionomer/catalyst ratio, water content, organic solvent, catalyst loading, stirring method, solid content, and ultrasonic spraying flow rate are properly made with few experimental attempts under ML results' guidance. Promising Pt utilization of 0.147 gPt kW −1 and a power density of 1.02 W cm −2 are achieved at 0.6 V in a single cell (H 2 /air) at an ultralow total loading of 0.15 mg Pt cm −2 . Therefore, this work contributes to the economy of hydrogen energy by paving the way for MEA optimization with complex parameters by ML.

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

“…The solvent affects the size, viscosity, curing rate, and other properties of the ionomer particles in the catalyst ink. These then affect the newly formed CL structure [ 96 , 97 ]. Most studies choose water or ethanol as solvents; however, these media may not disperse ionomers well [ 98 , 99 , 100 , 101 ].…”

Data-Driven Design of High-Performance Graphene-Based Seawater Desalination Membranes

“…The density functional theory–machine learning (DFT-ML) framework screens structures with desirable target values by the established ML models and further validates these materials employing DFT. This strategy reduces computational effort substantially, shortens design period, and improves efficiency and accuracy, laying the foundation for the rapid discovery of high-precision, efficient, and stable 2D graphene-based materials. − Herein, a high-throughput screening method was developed using DFT-ML framework to search for high-performance 2D-based desalination membranes. We used a novel 2D membrane (named Dadri-C) as the target material and applied particle swarm algorithms to evaluate its salt ion rejection, water flux, mechanical properties, and service life.…”

Phase Engineering of Molybdenum Carbide/Oxide for Highly‐Efficient Electrocatalytic Hydrogen Production

“…Hydrogen energy is considered as one of the ideal clean energy sources to replace fossil fuels due to its high energy density, renewability, and environment friendly. [6,7] Among of many hydrogen productions, water splitting by electrocatalysis is presently the most promising way. [8][9][10] However, it is the key to develop an efficient and stable electrocatalysts for promoting the application of hydrogen evolution reaction (HER) technology from water splitting.…”

“…Therefore, it is the goal of mankind all over the world to develop efficient, clean and economical energy sources, [5] and it is of great significance to achieve carbon peak and carbon neutralization. Hydrogen energy is considered as one of the ideal clean energy sources to replace fossil fuels due to its high energy density, renewability, and environment friendly [6,7] . Among of many hydrogen productions, water splitting by electrocatalysis is presently the most promising way [8–10] .…”