Towards rational design of solid oxide fuel cell electrodes through surface modification
MetadataShow full item record
Solid oxide fuel cells represent a scalable energy generation technology capable of operating at high efficiencies on multiple fuel sources. However, wide-spread implementation of SOFCs has been limited by the high degradation rate at current operating temperatures of 800-1000°C. Lowering the operating temperature to an intermediate range of 500-700°C will decrease the degradation phenomena, but will also decrease the catalytic activity of the electrodes. Modifying the surface of the electrodes is one method to increase the catalytic activity at these relatively low operating temperatures. This dissertation seeks to understand the role of surface modification on solid oxide fuel cell electrodes through conformal and non-conformal coatings. The first part of this dissertation demonstrates an asymmetric cell testing platform that is used to better understand the effects of conformal film deposition. Depositing a conformal thin film into a porous cathode is nontrivial and requires exhaustive optimization of either solution or gas phase deposition techniques. Even then, if the backbone material and the coating material aren’t very similar (in crystal structure, thermal expansion, etc), then the film will no longer be conformal after reaching SOFC operating temperatures. The asymmetric testing platform in this work was designed to focus on the effect of the thin film modification, which was accomplished by depositing a dense LSCF cathode on one side of an SDC electrolyte support with an accompanying porous LSCF counter electrode. Because of the high surface area of the counter electrode, the polarization resistances measured were dominated by the dense LSCF thin film. The planar dense film allows for precise control over the modification with conformal thin films via RF sputtering. The first part of the dissertation describes the fabrication and electrochemical characterization of this testing platform, which demonstrated the ORR activity was the dominant feature in the impedance spectra. The second part of the dissertation describes the surface modification with undoped ceria and samarium doped ceria. First, infiltration was used to modify the surface and it was seen that a change in morphology influenced the ORR activity. More specifically, for the undoped ceria, a more conformal morphology as opposed to a more dispersed, nanoisland morpohology lead to lower impedance for the ORR. Using the asymmetric testing platform and sputtered ceria, it was found that the thickness of the conformal ceria influenced the ORR. Thinner films showed an increase polarization resistance, while thicker films showed a decreased polarization resistance. The increase in polarization resistance for the thinner films was explained by an increase in vacancy concentration as demonstrated through comparison of the impedance behavior under bias to a doped ceria thin film. Second, it was found infiltration with samarium doped ceria decreased the polarization resistance. Interestingly, the performance increase was independent of the mol% of the samarium doped into ceria. This goes against the conventional thinking that increasing ionic conductivity (by increasing samarium mol %) will lead to increasing surface exchange properties. Thin film conformal deposition of 20SDC demonstrated an overall increase in polarization resistance with increasing resistance correlating to film thickness. These last two results suggest that the ionically conducting surface modification reduces the oxygen through a surface mediated process that requires high surface nanoparticles The third part of the dissertation describes the work using praseodymium doped ceria as the modification material to better understand the role of ionic and electronic conductivity in the ORR catalytic activity. Doping praseodymium into ceria increases both the ionic and electronic conductivity. Through infiltration, it was found that the optimal performance occurs at 50 mol% praseodymium in ceria even though 70 mol % exhibits higher electronic and ionic conductivities. Through XPS and TGA, it was found that amount of Ce3+ (i.e. reduced ceria) changes non-linearly with praseodymium dopant concentration. The 50 mol% doped ceria showed more Ce4+ available relative to the 30 and 70 mol% praseodymium concentration. Thus, it was found that oxygen ion vacancy concentration and electronic conductivity are not the only material properties relevant to increasing ORR activity. Instead, the results indicate a more nuanced view of oxygen reduction reaction and the correlation to bulk material properties. In the end, this work describes a platform for the characterization of conformal thin film surface modification and demonstrates the potential to increase material performance beyond bulk material properties. Importantly, this work has shown the nuanced performance enhancement beyond traditional correlations to ionic and electronic conductivity.