Structure-activity investigations of mixed Ni-Fe oxides as catalysts for electrochemical water splitting

A balanced utilization of energy requires new approaches in catalysis, to reach sufficient levels of production of renewable energy. Specifically, there is a need of efficient catalysts for energy utilization, conversion, and storage. Research concerning electrochemical water splitting has attracted attention due to the possibility to utilize water in integrated solar-to-water splitting devices. One challenge in utilizing these devices is catalyzing the sluggish oxygen evolution reaction (OER). Of recent interest have been non-precious metal catalysts, which at high pH have shown competitive activities in comparison to noble metal based catalysts. Specifically, Ni-Fe based catalysts show superior OER activity to a variety of both non-precious and noble metal catalysts. The origin of this high activity is not completely understood, and elucidating this is one of the goals of the present work. This dissertation focuses on establishing a structure-activity correlation of mixed Ni-Fe oxide based materials for the oxygen evolution reaction. In Chapter 3, we present a study of mixed Ni-Fe oxyhydroxides (OOH) in attempt to establish a structure-activity correlation between catalyst composition, OER activity, local atomic structure, and catalyst redox states. Ni-Fe oxides - specifically oxyhydroxides - show catalytic activity extending to the bulk of the catalytic film, meaning there is no passive oxide core. This enables all metal sites to participate in the reaction, which facilitate the use of X-ray methods that are sensitive to the bulk. The catalysts we investigated were prepared by a solvothermal method, which is up to date less investigated than electrodeposited Ni-Fe catalysts. We established activity-composition trends, following the Sabatier principle, with optimal activity around 50 % Fe. We identified the catalytically active state using quasi-in situ X-ray absorption spectroscopy (XAS), and combined the study using differential electrochemical mass spectrometry (DEMS) to evaluate the faradaic charge efficiencies toward O2 formation. With this approach, we could determine how much of the faradaic current that was directed into products and how much that was used to change oxidation states of the catalyst. Our results in addition revealed that Fe stabilizes Ni atoms in its low-valent +2 state, which is so far in contrast to earlier studies. We justified this by proposing a mechanistic hypothesis that a kinetic competition between the metal oxidation step (i) and the rate of OER catalysis leading to O2 formation (ii) prevail. Subsequently, if process (ii) outweighs that of process (i), Ni atoms will not accumulate in its high-valent +4 state. This would justify why discrepancies regarding the oxidation states of Ni and Fe have been reported in literature. In Chapter 4, we extended the OER activity measurements to a wider range of pH (pH 7-13), where we focused on a carbon supported Ni-Fe/C oxyhydroxide catalysts. We could establish a pH dependent activity trend, where we confirmed an extensive corrosion (leaching) of Ni at pH 7. In addition, we could show on morphological changes occurring after exposure to catalytic potential, proposed as the result of a leaching-recrystallization process during potential cycling. In Chapter 5, we investigated a series of physical mixtures of Ni+Fe(OOH) catalysts, prepared by mixing the parental Ni(OH)2 and Fe(OOH) catalysts. The physical mixtures showed as high OER activity as the pre-synthesized catalysts. To estimate the degree of mixing, XAS measurements were correlated to electrochemical measurements and EDX elemental mapping. In Chapter 6, we focused on a comparison of OER activities of unsupported and carbon supported Ni-Fe/C oxyhydroxide catalysts, where the focus was to investigate the origins of the enhanced catalytic activity observed on carbon support. We used XAS and DEMS to deduce the local atomic structure and metal redox state contribution. We also looked at the influence of varying the catalyst loading and the electrolyte pH. In Chapter 7, we investigated shape selected Ni-Fe oxide nanocatalysts and looked into how to obtain a narrow size distribution in the range of 10 nm. We compared OER activities at different oxide surfaces in order to relate the activity at a confined oxide surface to the activity of a Ni-Fe oxyhydroxide catalyst with catalysis extending to the bulk. We extend the discussions to the importance of electrocatalytically active area surface area, the influence of the oxide phase, size, and surface composition. The vast number of investigations of Ni-Fe OER catalysts continue to contribute to knowledge regarding the origin of the high catalytic activity, which will eventually help to establish a complete understanding of these catalysts. However, many areas remain unexplored up to date. With this work, we aim to contribute to a deeper insight into some aspects.