Developing carbon-based electrocatalysts for highly efficient oxygen reduction reaction

Abstract

The oxygen reduction reaction (ORR) is an important cathode process in fuel cells and metal-air batteries which promises enormous energy storage and conversion capability for portable electronics and electric vehicles. However, this reaction is quite sluggish and therefore requires a highly effective electrocatalyst, such as the state-of-the-art platinum materials, to drive the energy storage devices. The main obstacle for commercial applications lies in the poor electrocatalytic activity of non-platinum materials and the high cost (and scarcity) of platinum-based electrocatalysts. To achieve an ideal combination of high performance and low cost, functional carbon materials like graphene and carbon nanotubes have become more and more predominant in the field of research for their numerous advantages. However, it is essential that the origin of activity in ORR electrocatalysis is understood and the electrode configuration optimized before an effective electrocatalytic cathode can be rationally designed. In this thesis, there are three major sections covering the topics of the optimized structure for a porous carbon electrode, the ORR on metal-free heteroatom-doped carbons and the ORR on metal oxide/carbon hybrid electrocatalysts. In the first section, a study of the electrode configuration of graphene-chitosan composites for ORR is reported. Graphene nanosheets, a two-dimensional carbon allotrope, can gather into various assemblies depending on the amount of introduced chitosan binder. In particular, the assembled composite configuration can have substantial impact on the ORR overpotential. The mass and charge transport inside the graphene-chitosan composite was shown to be most influenced by the assembled structure, due to the distant separation of individual graphene nanosheets in a chitosan medium. Furthermore, a minimal amount of chitosan binder can ensure efficient transport for both mass and charge, provided that the active material itself is electrically conductive. The second part focuses on the ORR performed on oxygen and halogen-doped carbons (and sulfur) as important metal-free electrocatalysts. These attempts were aiming to provide information about the activity origin of metal-free nitrogen-doped carbons; direct study on N-doped carbon was thus not part of the focus. In the study of oxygen-doped carbons, limited activity improvement was found for most of the oxidized carbon nanotubes, although it was determined that the basal plane epoxides could be the most active of the common oxygen functional groups. Other surface oxygen groups (carboxyls and carbonyls) are also responsible for the improved ORR activity. Halogen-doping on reduced graphene oxide showed some small activity improvement for bromine and iodine doping, but not for chlorine. The electronegativity contrast of the dopant does not seem to be a crucial factor for highly efficient ORR. An accidental doping of sulfur in the form of sulfide did not seem to encourage the ORR on the doped graphene. Metal oxide/carbon hybrids have recently attracted attention for their synergistic effect in ORR electrocatalysis. By combining spinel manganese oxide and graphene oxide nanoribbons (GONR), the covalently bound Mn3O4-GONR interface was structurally resolved and determined to be the origin of the synergistic effect. In addition, the functional roles of each electrocatalytic component in the hybrid are really what govern the apparent four-electron ORR. A further study on amorphous manganese-oxygen nanoclusters attached to partially impaired carbon nanotubes showed that the amorphous nanoclusters act like a fully exposed covalent CnOnMn junction in the Mn3O4-GONR hybrid and can deliver similar activity to crystalline Mn3O4-GONR hybrid. Importantly, these manganese-oxygen nanoclusters are the direct active sites for both catalytic O2 and H2O2 electro-reduction.