This is the first comprehensive review of the impact of water impurities on PEM, AEM, and alkaline water electrolysers.

Understanding what controls the reaction rate on iridium-based catalysts is central to designing more active and stable electrocatalysts for the water oxidation reaction in proton exchange membrane (PEM) electrolysers. Here, we quantify the densities of redox active centres and probe their binding strengths on amorphous IrOx and rutile IrO2 using a combination of operando time-resolved optical spectroscopy, X-ray absorption spectroscopy (XAS) and time of flight secondary ion mass spectrometry (TOF-SIMs). Firstly, our results show that although IrOx exhibits an order of magnitude higher geometry current density compared to IrO2, the intrinsic rates of reaction per active state, on IrOx and IrO2 are comparable at a given potential. Secondly, we establish a quantitative experimental correlation between the intrinsic rate of water oxidation and the energetics of the active states. We use density functional theory (DFT) based models to provide a molecular scale interpretation of our data. We find that the *O species formed at water oxidation potentials have repulsive adsorbate-adsorbate interactions, and thus increasing their coverage weakens their binding and promotes the rate-determining O-O bond formation. Finally, we provide insights into how the intrinsic water oxidation kinetics can be increased by optimising both the binding energy and the interaction strength of the catalytically active states.

In many electrochemical systems, such as carbon dioxide reduction, batteries, and supercapacitors, hydrogen evolution reaction (HER) is an undesired competing reaction. Herein, we investigate the factors controlling the HER on seven different commercial carbon materials commonly found in many of these systems. The electrochemical HER response was determined by rotating disk electrode potential hold measurements in acidic media, and correlated with the physical characteristics of the carbon materials determined nitrogen adsorption/ desorption, as well as previous experiments on the same materials. An on-chip electrochemical mass spectrometer was used to probe the gaseous products produced at the electrode in situ, which allowed HER to be distinguished from other competing reaction and the onset of the reaction was established to be -0.38V vs RHE. The results indicate that carbons with low amount of metal impurities have the lowest H2 evolution rates.

While PEM electrolyser catalyst cost may not be a significant portion of system costs1 it does represent a bottleneck for the ability to generate TW level of H2. This is primarily because of the reliance on IrOx as a stable oxygen evolution catalyst in order to meet future green H2 needs either replacement or reduction of iridium loading of at least 50 times is needed while maintaining a high level of stability2. IrOx based materials are the only oxygen evolution catalysts combining activity and stability under PEM electrolysis conditions; even so, they are insufficiently stable. In the current work, we tailored the activity of IrOx catalysts synthesised by a variant of the Adams fusion reaction3 using decomposition of Iridium nitrate and varying temperature of synthesis to generate a series of catalysts with differing crystallinity and surface area. We benchmarked their stability using both accelerated degradation electrochemical measurements (30k cycles 1.2-1.7VRHE @ 500 mV s-1) and inductively coupled plasma-mass spectrometry(ICP-MS), both in rotating disk electrode(RDE) measurements and in a single cell PEM electrolyser. We have compared several different methods for probing electrochemical surface area, including BET, double layer capacitance from cyclic voltammetry, adsorption capacitance using impedance spectroscopy and CO stripping using ultrasensitive on chip electrochemical mass spectrometry. The results from the RDE measurements are shown in figure 1; they show that while the high surface area amorphous IrOx catalysts demonstrate higher activity normalised to geometric area, when normalised to specific activity the difference is insignificant. In addition to electrochemical performance losses, the amorphous IrOx shows an order of magnitude increase in iridium dissolution, determined via ICP-MS. Future studies will look at the ability to overcome the limitations of aqueous model studies for stability testing and utilising testing to select OER catalyst candidates that meet both activity and stability required for long term operation in PEM electrolyser systems. 1 L. Bertuccioli, A. Chan, D. Hart, F. Lehner, B. Madden and E. Standen, Study on development of water electrolysis in the EU, Fuel Cells and hydrogen Joint Undertaking, 2014, vol. 1. 2 P. S. Alexis Grimaud, Jan Rossmeisl, Research nees towards sustainable production of fuels and chemicals, Section 1: Water splitting and sustainable H2 Production, 2019. 3 D. F. Abbott, D. Lebedev, K. Waltar, M. Povia, M. Nachtegaal, E. Fabbri, C. Copéret and T. J. Schmidt, Chem. Mater., 2016, 28, 6591–6604. Figure 1

State-of-the-art proton exchange membrane (PEM) electrolysers employ iridium-based catalysts to facilitate oxygen evolution at the anode. To enable scale-up of the technology to the terawatt level, further improvements in the iridium utilisation are needed, without incurring additional overpotential losses or reducing the device lifetime. The research community has only recently started to attempt systematic benchmarking of catalyst stability. Short term electrochemical methods alone are insufficient to predict catalyst degradation; they can both underestimate and overestimate catalyst durability. Complementary techniques, such as inductively coupled plasma - mass spectrometry, are required to provide more reliable assessment of the amount of catalyst lost through dissolution. Herein, we critically review the state of the art in probing degradation of iridium-based oxide catalysts. We also highlight considerations and best practices for the investigation of activity and stability of oxygen evolution catalysts via short term testing.