Three-dimensional Image-based Modeling of Microscale Phenomena in Polymer Electrolyte Fuel Cells

The main scope of this thesis is to analyze the electrochemical processes in polymer electrolyte fuel cells (PEFC), focusing on the 3D geometry of electrode materials for performance and durability improvement. Specifically, the thesis includes new electrochemical modeling methods on real material-based structures and idealized representative geometry. To obtain the 3D structures of platinum group metal-free (PGM-free) cathode materials, the nano-scale X-ray computed tomography (nano-CT) technology is applied, and the 3D structures from the imaging are directly applied in a new transport-reaction physics modeling framework. In addition to the nano-CT scale modeling, this thesis also includes a work on performance and durability modeling at the PEFC catalyst particle scale. The effects of the heterogeneity of electrodes on the performance and durability are analyzed by using an idealized geometry and geometry generated from images obtained by high-resolution scanning transmission electron microscopy-computed tomography (STEM-CT). For most of the model methods presented in this thesis, we made an effort to take advantage of open-source, free of charge software packages for modeling flexibility and scalability. The core part of the modeling frameworks has been built on top of open-source, high-performance finite element simulation framework to accelerate the modeling process considering the inevitable large problem size in the 3D image-based modeling approach. This approach also should contribute to the availability of the models to other researchers in the research field. By conducting simulations on different types of materials and different conditions, it has been possible to infer the relationships between the electrode nanostructures and fuel cell performances. Remarkably, the PGM-free cathode model suggests that a uniformly distributed ionomer structure may enhance the total current density at a high electrode potential region by reducing the ohmic losses within the catalyst aggregates. Modeling of platinum (Pt) dissolution and re-deposition at the carbon support scale has provided us insight as to the mechanisms for Pt to migrate from the interior of the support to the exterior. The STEM-CT model of the carbon supported Pt catalyst aggregates suggests that larger catalyst particles inside a carbon support may have larger current density by increasing the proton concentration on the surface. In addition, we find that the particle spatial distribution inside carbon support has little impact on the performance due to thin electric double layers on the catalyst surfaces and short transport length scales. These results are particularly useful to understand the underlying electrochemical mechanisms in the electrodes and evaluate the performance in various fuel cell operating conditions. The insights obtained from this work should be beneficial for high-performance electrode design and more broad applications of PEFC.