Designing Microporous Layers for Electrolyzers Using Stochastic Approach
Jason Keonhag Lee
Abstract
Electrochemical energy conversion devices, such as water and carbon dioxide electrolyzers, offer significant advantages in achieving net-zero emissions and in mitigating further increases in global temperature. However, their widespread adoption necessitates enhancements in performance and durability. Microporous layers (MPLs) have been gaining attention as a promising means to enhance the performance and durability of membrane-electrode-assembly (MEA) based electrolyzers, but their nontrivial mechanisms and complexity in fabrication pose challenges for optimizing the microporous layer structure experimentally. This study introduces a stochastic model for generating MPLs in application to electrolyzers. The model produces 3D reconstructions of MPLs, with porosity and particle size as input parameters, and is capable of generating biased MPLs by taking the pre-existing 3D reconstruction as an input. The model applies a dilation and erosion algorithm to replicate sinter-necks formed in the MPL during the sintering process, and captures their impact on structural and transport properties. In this work, three types of MPLs are generated by using the presented model, which include single-layer MPLs, MPLs with pore formers, and bilayer MPLs. Surface roughness analysis and pore network simulations on the MPLs highlight the significance of particle size in the MPL design. Using finer particles at higher porosities are favored over using larger particles at lower porosities. Such findings are examples of the valuable insights offered from the presented stochastic model, and the model will guide seminal discovery of next-generation MPLs that will greatly progress the shift toward net-zero electrochemical energy conversion technologies.