Ceria-based thin film electrodes for sustainable hydrogen production

The human race is currently enjoying an unprecedented level of wealth, characterized by an ever increasing rate of technological developments and scientific discoveries. However, under the looming menace of the consequences of anthropogenic global warming, we need to develop efficient and scalable ways to store energy to facilitate the transition to a sustainable, carbon neutral society. High temperature Solid Oxide Electrolysis Cells (SOEC) are particularly promising, as part of the energy needed to drive the reaction can be supplied as heat, resulting in very high efficiency. The state-of-the-art devices typically employ yttria-stabilized-zirconia electrolyte, a Lanthanum Strontium Manganite (LSM) oxygen electrode and a porous Nickel-YSZ cermet as a fuel electrode. Developing new materials could increase the lifetime and current density of SOEC, allowing their widespread adoption in society.

Cerium oxide is very promising, for its high electrocatalytic activity towards important reactions such as water splitting and carbon dioxide reduction, its tolerance for impurities and its mixed ionic and electronic conductivity, which expands the active area to the whole surface.

In this project, we focused on optimizing cerium oxide’s properties as a water splitting electrocatalyst via doping and surface termination engineering. We then performed a systematic investigation of the technologically relevant Ni-CGO cermet, to unravel the impact of the triple-phase-boundaries (TPB) between metal, oxide and gas, to guide future efforts to optimize its microstructure.

We fabricated model cells to study the effects of the surface termination on the properties of 10% and 20% Gd-doped ceria (CGO10 and CGO20). Through the use of model SOEC cells with high quality, epitaxial thin films in the (100), (110) and (111) orientations, we highlight the superior performance of the less commonly studied (110) and (111) facets, which offer lower activation energies and an up to doubled catalytic activity compared to the (100) termination. We reveal that the dependence of the polarization resistance and chemical capacitance on the steam, hydrogen and oxygen pressure do not strongly depend on the doped ceria facet.
The properties of the (100) orientation appear to be strongly influenced by the test atmosphere, as it offers excellent performance in high steam and hydrogen contents, and is then significantly slowed down by exposure to CO-CO2 mixtures. Understanding this surprising behavior could hold the key to the development of high performance fuel electrodes.

In the second main study we explored the effects of inserting doping in the lattice of ceria. We investigated the effect of an ionic mismatch, through 10% doping of the isovalent Zr, and of acceptor doping, via a 10% content of thetrivalent Gd cation, whose ionic radius is very close to the one of cerium. Finally, we explored the effects of codoping with Pr and Bi, which were identified as promising in a recent computational investigation by our colleagues at DTU Energy. Our findings show a clear impact of cationic substitution on the reducibility of ceria, as a result of the mismatch in valence and ionic radii, which are observable from the different power-law dependence of the chemical capacity of our electrodes versus the oxygen partial pressure. The polarization resistance of (111) oriented model cells does not appear to be strongly affected, leading to similar dependencies on the steam, hydrogen and oxygen activities for all the electrode compositions.