UC Berkeley

Fuel cells are a next-generation, clean energy-conversion technology designed to replace existing internal combustion engines. Their implementation is important in reducing carbon emissions and addressing the world climate crisis. However, many system limitations still need to be resolved before fuel cells can enter widespread use, particularly with regard to transport of chemical species within the fuel cell. Fuel cells are composed of several key components, but paramount among them are the fuel-cell catalyst layers, responsible for the fuel-cell reactions that produce electricity, and the fuel-cell membrane, responsible for transporting ions within the system. Transport of chemical species is irrevocably tied to the performance of each component. Improving fuel-cell efficiency by minimizing gas crossover requires understanding gas transport within fuel-cell membranes. Addressing the issue of fuel-cell flooding and the associated reduction in performance requires study of water transport within the membranes, as well as gas transport within the catalyst layers. This dissertation studies these phenomena and provides guidance on proper measurement techniques as well as the transport properties of current state-of-the-art materials.Following an Introduction in Chapter 1, the dissertation begins with a detailed examination of the use of microelectrodes to study the properties of fuel-cell membranes, particularly gas transport. There is an extensive history of using microelectrodes to study fuel-cell membranes ex-situ, but little standardization of cell design and technique. Different designs available in the literature are discussed as are the results of prior studies. Author recommendations are made for proper use of microelectrode systems to ensure consistent experimental results. Then, a new flow-through microelectrode cell design that ameliorates several of the key issues with prior designs, such as equilibration time, is presented. The cell design is evaluated in several ways, including the impact of applied mechanical pressure, impact of gas flowrate, ability to measure both hydrogen oxidation and oxygen reduction, and minimization of equilibration time. Chapter 2 thus provides a foundation for the study of membrane transport properties in the next two chapters. Chapter 3 provides a comprehensive examination of measuring gas transport using a microelectrode system. The flowthrough cell discussed in Chapter 2 measures the diffusivity, Henry’s constant, and permeability of Nafion and Nafion XL to hydrogen and oxygen gas as a function of water content. Flaws with the existing analytical solutions for analyzing current transients in these systems are discussed, and a 2D numerical model is developed to account accurately for the finite membrane thickness. In addition, the impact of surface roughness at very short times (< 1 s) is quantified and included in the analysis. Finally, a simple multiphase parallel-diffusion model interprets the measured gas-transport parameters. Hydrogen has a higher diffusivity and permeability than does oxygen, but a lower Henry’s constant. Diffusivity and permeability both increase with water content whereas Henry’s constant decreases. This is due to the impact of the hydrophilic phase, as both gases have a higher diffusion coefficient and lower Henry’s constant in the hydrophilic phase compared to the hydrophobic polymer backbone. The parameters presented in Chapter 3 allow for a more accurate picture of gas crossover within fuel cells and assist in creating accurate models of this phenomenon. Electro-osmosis, or coupled ion-water transport, in fuel-cell membranes is the focus of Chapter 4. Once again, the microelectrode cell described in Chapter 2 studies this effect, in both proton- and anion-exchange membranes. Electro-osmosis is examined by measuring the open-circuit voltage as the relative humidity changes within the cell. The necessary background in thermodynamics and transport phenomena is provided to interpret the experimental data. A Nafion membrane is the baseline case and exhibits a higher water transport number than previously reported. However, more focus is given to the measurement of electro-osmosis in anion-exchange membranes. Anion type in the anion-exchange membrane is studied; it is found that the solvation shell of the ions has a significant effect on the measured water transport number, consistent with studies in Nafion. The larger is the solvation shell, the higher is the measured coefficient. Essentially, ions primarily move the water that is directly associated with them. In addition, temperature has little impact on the water transport number in anion-exchange membranes. Finally, a Stefan-Maxwell-Onsager framework and the measured water transport number of Versogen is used to extract the water permeability as a function of water content. Permeability tends to increase with water content, as it is easier for water to move through the membrane when more water is present. Chapter 4 presents all of the water transport parameters necessary to define fully the water balance in fuel-cell membranes. Chapter 5 studies the impact of water droplet growth on platinum catalyst particles within the catalyst layer and whether the transport of oxygen gas to the platinum catalyst is inhibited by drop growth. A moving-mesh numerical model is developed to study this droplet growth. The Navier-Stokes equation captures convection within the water droplet, and Fick’s law models oxygen transport within the expanding drop. Tafel kinetics quantifies the current at the platinum surface. Four different cases are considered: growth of a pinned and advancing drop on a bare platinum surface, growth of an advancing drop on a thin layer of Nafion, and growth of a water layer within a carbon nanopore. In all cases, water droplet growth does not inhibit oxygen transport due to a funneling effect, where the larger gas/water interface compensates for the increasing diffusion length as the drop grows. In the Nafion-layer case, the Nafion membrane is much more mass-transfer resistive than is the droplet, minimizing the impact of the droplet if the platinum is covered in Nafion. In the carbon nanopore, the produced water layer can become limiting, but only at pore lengths much larger than is typically found in porous carbon particles. The formation of local water droplets is thus not mass-transfer limiting in the catalyst layer. Catalyst-layer design should instead focus on reducing the impact of full catalyst-layer flooding or the placement of platinum within the catalyst layer rather than focusing on the formation of water nanodroplets. Overall, this dissertation explores how a microelectrode cell can be used to ascertain critical membrane transport properties including how system geometry plays a key role in the proper measurement of transport properties. The findings quantify the importance of water content on transport properties and how it is the most powerful variable controlling them. Proposed future work includes extending the study to gas transport in novel ion-conducting polymers (ionomers), examining gas and water transport in ionomer thin films, and modeling local bubble growth on platinum nanoparticles in electrolyzer catalyst layers.