In the past decade, hydrogen evolution from photocatalytic alcohol oxidation on metal-loaded TiO2 has emerged as an active research field. While the presence of a metal cluster co-catalyst is crucial as a H2 recombination center, size and coverage effects on the catalyst performance are not yet comprehensively understood. To some extent, this is due to the fact that common deposition methods do not allow for an independent change in size and coverage, which can be overcome by the use of cluster sources and the deposition of size-selected clusters. This study compares size-selected Ni and Pt clusters as co-catalysts on a TiO2(110) single crystal and the resulting size- and coverage-dependent effects in the photocatalytic hydrogen evolution from alcohols in ultrahigh vacuum (UHV). Larger clusters and higher coverages of Ni enhance the product formation rate, although deactivation over time occurs. In contrast, Pt co-catalysts exhibit a stable and higher activity and size-specific effects have to be taken into account. While H2 evolution is improved by a higher concentration of Pt clusters, an increase in the metal content by the deposition of larger particles can even be detrimental to the performance of the photocatalyst. The acquired overall mechanistic picture is corroborated by H2 formation kinetics from mass spectrometric data. Consequently, for some metals, size effects are relevant for improving the catalytic performance, while for other co-catalyst materials, merely the coverage is decisive. The elucidation of different size and coverage dependencies represents an important step toward a rational catalyst design for photocatalytic hydrogen evolution.

The suitability of the Thin-Film RDE (TF-RDE) technique to rigorously evaluate stability measurements for the oxygen evolution reaction (OER) was recently questioned. The main issue was the inability to deconvolute bubble blockage of catalytic active sites from catalyst dissolution using the TF-RDE technique. It is also possible that the low-loading of TF-RDE OER catalysts exacerbates the effect of bubble blockage. In this work, the modified rotating disk electrode (MRDE) is used with commercial catalyst coated membranes (CCMs) to evaluate catalyst stability. The MRDE may be better suited for stability measurements, since the CCM samples used can better avoid experimental artifacts and can explore much higher current densities than a TF-RDE. Thicker catalyst layers have good adhesion to the membrane, making experimental artifacts less pronounced in stability measurements. Three different stability protocols are used to study the effect of cycling, lower/upper potential limits, and regeneration. The protocol which induced the most irreversible degradation was the square-wave voltammetry (SWV) cycling between 0.05–2.0 VRHE. This irreversible degradation is likely the result of catalyst dissolution. The importance of differentiating between irreversible and reversible degradation is highlighted as a potential future standard for stability evaluation.

Pt catalyst particles on reducible oxide supports often change their activity significantly at elevated temperatures due to the strong metal-support interaction (SMSI), which induces the formation of an encapsulation layer around the noble metal particles. However, the impact of oxidizing and reducing treatments at elevated pressures on this encapsulation layer remains controversial, partly due to the 'pressure gap' between surface science studies and applied catalysis. In the present work, we employ synchrotron-based near-ambient pressure X-ray photoelectron spectroscopy (NAP-XPS) to study the effect of O2 and H2 on the SMSI-state of well-defined Pt/TiO2(110) catalysts at pressures of up to 0.1 Torr. By tuning the O2 pressure, we can either selectively oxidize the TiO2 support or both the support and the Pt particles.Catalyzed by metallic Pt, the encapsulating oxide overlayer grows rapidly in 1x10 -5 Torr O2, but orders of magnitudes less effective at higher O2 pressures, where Pt is in an oxidic state.While the oxidation/reduction of Pt particles is reversible, they remain embedded in the support once encapsulation has occurred.