Simulation of catalyst layer and membrane durability in polymer electrolyte membrane fuel cells under real-world vehicle operation

Addressing climate change has become a prominent focus, leading to the implementation of regulations aimed at reducing greenhouse gas emissions by fossil fuel vehicles. While the polymer electrolyte membrane fuel cells (PEMFCs) have been proven as a promising candidate to replace internal combustion engines, their commercialization has been impeded by high costs and uncertainties related to their durability in vehicle applications. In this thesis, degradation models for membrane and cathode catalyst layer of PEMFCs are developed separately due to their distinct degradation mechanisms to study the possibilities of enhancing the overall fuel cell lifetime with respect to fuel cell cost and hydrogen consumption. Widely calibrated with experimental data from accelerated stress test cycles, the catalyst degradation model is utilized to study the degradation of a fuel cell transit bus under use-level conditions in the city of Victoria, B.C., Canada. The model is coupled with the performance degradation by calculating the voltage degradation and updating the polarization curve. The recorded bus drive cycle is converted to a cell voltage profile considering fuel cell stack specifications and vehicle dynamics. Next, the membrane degradation is modeled considering the fibrillar morphology of the membrane and is calibrated with experimental stress test results. Coupled with the catalyst degradation model to include the voltage drop impact over time, the membrane degradation model is used to estimate the lifetime under the same drive cycle. The reduction in the stack size and temperature were found to significantly increase the fuel cell lifetime. Next, a number of degradation mitigation strategies are evaluated such as lowering the cell temperature and hybridizing with the battery. A multi-objective optimization is implemented using the genetic algorithm to optimize fuel cell lifetime and mitigate the rise in hydrogen consumption resulting from hybridization. It is discovered that hybridization leads to more durable stacks and reducing their cost; however, it comes at the expense of increased hydrogen consumption. Finally, following a comparison between the results of the present modeling framework and a realistic transit bus operation, a favorable agreement is observed. The modeling framework can be used to estimate the fuel cell lifetime and hydrogen consumption and optimize stack based on these considerations.