UC Berkeley

Electrolytic devices are energy-conversion technologies that can assist in decarbonizing essential industries that are hard to decarbonize. These systems use electricity to drive chemical transformations and generate valuable products, such as hydrogen gas. One promising device is a water electrolyzer that splits water into hydrogen and oxygen gases. These systems use membrane-electrode assemblies (MEAs) that are comprised of an ion-conducting membrane as the solid electrolyte and catalyst layers (CLs) where the electrochemical reaction proceeds. Each of these MEA layers has its own set of design parameters and can influence the device’s performance. Specifically, the catalytic activity of the anode or oxygen-generating electrode of the MEA depends not only on the electrocatalyst, but also the reactant and product mass transport to/from the reaction sites. While one could increase performance via increasing catalyst loading, this is not cost-effective since in acid systems costly iridium oxide is the most common catalyst. Thus, to increase performance, we examine optimizing catalyst utilization via understanding multiphase and multiscale transport in the CL. Studying the multiscale transport helps determine how the reaction conditions in an MEA can dictate the performance of the catalyst and its surrounding microenvironment. This dissertation focuses on exploring and elucidating the transport phenomena in electrolytic cells to diagnose their inherent issues and develop pathways to ameliorate them. This dissertation starts by investigating liquid-fed systems in the first two chapters and the rest of the chapters explore vapor-phase conditions. Chapter 2 investigates the transport within the anode CL of a proton-exchange-membrane water electrolyzer (PEMWE). The local transport within the anode CL depends strongly on its structure and how the species (i.e. water and oxygen) are transported throughout it. The structure of the CL depends on the catalyst ink from which it is formed. Correlating the ink properties, such as solvent ratio and aggregate size of the catalyst particles and ionomer, to the structure of the CL can help guide others in fabricating new CLs and elucidate how the physical structure changes the transport phenomena in porous electrodes. In Chapter 2, the large aggregates found in the alcohol-rich ink result in a denser CL (porosity = 20%, compared to a baseline porosity = 40%) and hence high mass-transport overpotentials, which govern cell performance. The best performing CL had double the porosity (40%), small aggregates (210 nm), and anisotropic tortuosity, in which the through-plane tortuosity was higher than in-plane. The structure plays a significant role in minimizing overpotentials. In addition to the PEMWE device, other electrolyzer motifs are being explored to allow for a wider selection of catalysts and different operating strategies. Investigating different conditions, such as pH, can identify the differences in the kinetic mechanism of the oxygen evolution reaction and how species are transported throughout the cell. Chapter 3 uses a microkinetic model and experimental data to explore the kinetics of the oxygen-evolution reaction near the surface of the catalyst for three different pHs: 1, 9, and 13. Chapter 3 introduces liquid electrolytes to achieve the desired conditions and investigates the performance of the system both in a rotating disk electrode (RDE) and in an MEA. From the RDE experiments, the kinetic data can be inputted into the model and can identify how the ion species are influenced by different current densities and pH values, and provide the behavior at the catalyst surface. The local transport results in very different pH conditions at the catalyst surface compared to the bulk, which greatly impacts the reactivity at near-neutral pHs. MEA performance links these kinetic findings to relevant systems at high operating current densities (> 100 mA cm-2) and reveals even more challenges for these systems when investigating ion transport, particularly large concentration gradients must be overcome. Since the MEA systems are complex, balancing multiple species and phases, we can focus on a simpler system, where a single-phase water electrolysis can be examined. Chapter 4 explores vapor-fed PEMWEs through coupled experiments and mathematical modeling. While the literature showed low activity for water-vapor PEMWEs, Chapter 4 shows the best performing vapor-fed system to date within the literature, achieving >100 mA cm-2 performance at < 1.7 V. However, this system exhibited major limitations due to membrane dehydration. Thus, the decreased membrane water content at higher current densities is exacerbated by lower local relative humidity, resulting in poor CL utilization. Water is not only essential for the oxygen-evolution reaction, but also for ensuring good ion conductivity in the polymer electrolyte. Finally, the last two chapters of this dissertation connect the findings of Chapter 4 to specific applications. Chapter 5 explores water-vapor-fed PEM unitized regenerative fuel cells (PEMURFCs). Vapor-phase operation simplifies the physics of the system, as shown in Chapter 4, and Chapter 5 explores the impact of different PEMs, feed gases, and relative humidity on URFC performance and durability. By tailoring operating conditions and membrane chemistry, the vapor-URFC achieves a roundtrip efficiency of 42% and a lifetime of 50,000 accelerated-stress-test cycles for fully humidified feeds. Chapter 6 builds on the simple vapor-fed PEMWE in Chapter 4 and introduces another reaction: methane partial oxidation. Using an electrolytic device for electrochemical synthesis directly can revolutionize and decarbonize the way chemical products are synthesized. However, there are significant challenges to overcome, including an electrocatalyst with high conversion and selectivity, high efficiency, and sufficient transport of the multiple reactants and products. This chapter explores the literature, the current status of the field, and recommendations for designing these systems for methane partial oxidation. Unfortunately, using the vapor-fed MEA and the promising catalysts, no methanol was found in the system and more work needs to be done within the electrocatalysis space. In summary, electrolytic devices can be engineered better by modulating the transport of the reactants, products, and ion species to and from the reaction site. This dissertation uses experiments and simulations to explore species transport within various electrolytic devices at different length scales. The results provide guidelines and identify open questions to improve these devices and determine how to start using them for other electrochemical-conversion systems.