| This thesis develops models for the design and control of Dead-Ended Anode (DEA) fuel cell systems. Fuel cell operation with a dead-ended systems anode reduces fuel cell system cost, weight, and volume because the anode external humidification and recirculation hardware can be eliminated. However, DEA operation presents several challenges for water management and anode purge scheduling. Feeding dry hydrogen reduces the membrane water content near the anode inlet. Large spatial distributions of hydrogen, nitrogen, and water develop in the anode, affecting fuel cell durability. The water and nitrogen which cross through the membrane accumulate in the anode during dead-ended operation. Anode channel liquid water plugging and nitrogen blanketing can induce hydrogen starvation and, given the right conditions, trigger cathode carbon oxidation leading to permanent loss of active catalyst area. Additionally, the accumulation of inert gases in the anode leads to a decrease in cell efficiency by blocking the catalyst and reducing the area available to support the reaction. Purging the anode uncovers the catalyst and recovers the available area, but at the expense of wasting hydrogen fuel.;To understand, design, and control DEA fuel cells, various models are developed and experimentally verified with plate-to-plate experiments using neutron radiography and gas chromatography. The measurements are used to parameterize dynamic models of the governing two-phase (water liquid and vapor) spatially distributed transport phenomena. A reduced order model is developed that captures the water front evolution inside the gas diffusion layer and channels. A second model captures the nitrogen blanketing front location along the anode channel. The reduced order models are combined to form a complete description of the system. They require less computational effort, allow efficient parameterization, and provide insight for developing control laws or designing and operating DEA fuel cells. |
