Investigation of transport phenomena in polymer electrolyte membrane fuel cells

Polymer electrolyte membrane fuel cell performance is governed by electrochemical reaction and transport phenomena. The transport process in fuel cells is complicated by phase changes and porous media. The objective of this dissertation is to advance the understanding of the transport phenomena in fuel cells with a focus on transport in porous gas diffusion layers (GDLs).;First, this dissertation evaluated macro level models used to simulate the flow in porous media. As a result, the Brinkman equation was recommended because the inertial term was found negligible in GDLs. A micro structure model was then developed based on the reconstructed geometry from GDL Scanning Electron Microscope (SEM) images. The model predicted a reasonable cell performance.;Second, a one way fluid-structure-interaction simulation was conducted to model the two phase flow in the deformed GDL due to assembly compression. Both assembly compression and liquid accumulation reduce effective permeability and porosity, thus resulting in reactant shortage and performance decrease. Additionally, a three dimensional model was developed to investigate the channel-to-channel crossover in a serpentine channel fuel cell. Oxygen concentration was found to shift from the channel center to adjacent channels. The local current density showed a similar pattern. GDL anisotropic permeability and thermal conductivity were found to influence the current density and temperature distribution.;Last, a numerical model was developed for the polybenzimidazole (PBI) membrane fuel cell. Using the adjusted Arrhenius Law, the influences of relative humidity, phosphoric acid doping level and temperature on cell performance were studied. It was found that increasing the doping level, temperature or relative humidity improved the cell performance. The cell performance was more sensitive to phosphoric acid doping level.;This dissertation studied different aspects of the transport phenomena inside fuel cells and contributed to the fundamental understanding of its transport mechanism. It provided directions for GDL models selection. The influences of nonhomogeneous and anisotropie GDL properties on the transport and cell performance were identified. The numerical models incorporated several novel features, thus contributing to the PEM fuel cell design and optimization.