• Sørensen, Jakob Ejler

Abstract

The study of catalysts is motivated by giving a short overview of the huge and growing energy demand which, at present, is based mostly on a polluting and limited resource. A transition to sustainable energy is therefore needed and catalysis plays an important role in balancing out the intermittent nature of the truly renewable energy sources.<br/>This thesis deals with model systems of catalysts in order to elucidate the mechanisms responsible for their activity. Two submethods have been implemented to improve the method of preparing model system catalysts, and two catalytic systems for splitting water have been investigated with isotope studies.<br/>Rastered depositions: It is shown how the distribution of nanoparticles can vary quite a lot across the surface after a deposition. To fix this, stepper motors now control movement of the sample during depositions so it can berastered. This smooths out any variations in the beam of nanoparticles, but introduces an uncertainty in the calculation of loading. Simulations are used to show how this uncertainty can be understood and minimized.<br/>Isotope studies: Electrocatalytic water splitting is a sustainable way of storing energy as chemical bonds, but the reaction is expensive. Nickel-iron based catalysts are commonly used in alkaline electrolyzers because of the stability and activity, but the nature of the active site is still under debate. Through isotope studies, we show that lattice oxygen is not exchange during oxygen evolution from which we conclude that the active sites are only located on the surface of the nanoparticles. Thereby, our well-defined 5.4 nm Ni₃Fe nanoparticles have a turnover frequency of 6:2 ± 1:6 s^-1 at 300 mV overpotential, which is the highest reported in the literature. For polymer electrolyte membrane electrolysis, thin film electrodes of RuO₂ are compared to IrO₂ in an ongoing investigation to provide mechanistic insights for oxygen evolution. We find evidence of oxygen evolution down to 60mV overpotential on a RuO₂ foam, which is experimentally unprecedented. Comparing to thin films show that there is nothing special about the intrinsic activity, but the record owes to the high surface area of the catalyst and the sensitivity of the measurement setup. Initial results from isotope studies indicate that lattice oxygen exchange is much lower than dissolution processes on RuO₂ and that the exchanged oxygen remains close to the surface. On hydrous IrO₂, however, the lattice oxygen exchange is larger than in the RuO₂ system. The dissolution processes are on the same scale as the oxygen exchange. These results are only based on a small subset of the data, though.<br/>Liquid nitrogen-cooled TPD: LN₂-cooled temperature-programmed desorption has been implemented so CO-TPD can be performed on copper nanoparticles. This is to have a characterization method which probes the reactivity of the prepared model systems for a more complete characterization. The technique is illustrated by three examples: 1) A copper foil provides a reference that is comparable with the literature. 2) Overlapping 5nm mass-selected copper nanoparticles show a new peak, which may be due to unique sites from overlapping nanoparticles, or simply an up-concentration of natural under-coordinated sites. 3) Intermetallic charge transfer is shown to weaken the reactivity of palladium atoms in palladium-tin bimetallic alloys. Slightly more than 50% tin content leaves the palladium sites with binding energies similar to copper.

Topics

• nanoparticle
• Deposition
• impedance spectroscopy
• surface
• polymer
• nickel
• thin film
• simulation
• Oxygen
• Nitrogen
• copper
• iron
• intermetallic
• tin
• palladium