Deconvoluting Transport and Kinetics on Ionic Liquid-Modified Fe Catalysts for Oxygen Reduction

Abstract

Introduction Hydrogen fuel cells could play a key role in the decarbonization of the energy sector. However, their commercialization is hindered by the sluggish kinetics of the oxygen reduction reaction (ORR) and by the requirement of platinum, which is expensive and unsustainable. Alternative catalysts based on transition metals have proven promising, but their activity and durability is still far from that of platinum.Recently, thin layers of ionic liquids have been successfully used to improve both the durability and activity of noble-metal free ORR catalysts[1]. As shown in Figure 1, the presence of the ionic liquid layer can influence the reaction kinetics through multiple effects: (i) O2 transport, (ii) water transport (iii) proton transport and (iv) binding to the reaction intermediate.These competing effects are generally convoluted. To that end, we have recently developed models to quantify the effect of the ionic liquid on both oxygen transport and reaction kinetics. Effect of Ionic liquids on reaction kinetics The ionic liquids tested for oxygen reduction are hydrophobic and their degree of water uptake is largely controlled by the cation. By decreasing the concentration of water at the active sites, they can cause the dehydration of the *OH intermediate, ultimately weakening the *OH bond[2]. For those catalysts such as Pt, which sit on the strong binding side of the volcano, this weakening of OH causes an increase in activity[3].To the contrary, we have now shown that ionic liquid layers decrease the activity of catalyst on the weak binding side of the volcano, such as iron phtalocyanine (see Figure 2). By changing ionic liquid, we can control hydrophobicity and by monitoring the position of the voltammetric peak for *OH adsorption we can probe the strength of *OH binding. In this way, we are able to confirm the earlier hypothesis that hydrophobicity controls *OH binding[2]. We hence provide a new lever for tailoring the reaction kinetics of oxygen reduction for any transition metal catalyst using ionic liquids. Oxygen Transport in ionic liquid layers The ideal ionic liquid should transport oxygen quickly, while featuring high oxygen solubility. However, it has been shown that O2 solubility and diffusivity cannot be independently tuned, as, for example, fluorination of the anion improves oxygen concentration, at the expense of diffusivity[4]. The optimum balance between these two parameters has been so far unclear. To that end, we have developed a way to deconvolute the effect of oxygen solubility and diffusivity.Using gravimetric, volumetric and electrochemical techniques, we reliably characterized oxygen transport in ionic liquids, and we were able to experimentally distinguish between oxygen activity and concentration. By using Fick’s diffusion and the Faraday law, we obtained a prediction for the diffusion-limited current in a rotating disc-electrode, as a function of the nature of the ionic liquid, its thickness and the rotational speed. As shown in in Figure 3, the diffusion-limited current density correlates to the permeability of oxygen in the ionic liquid layer; with our experimental data confirming the model.When looking at the diffusion-limited region, this model leads to the trivial conclusion that oxygen transport can be improved by maximizing oxygen permeability in the ionic liquid and increasing the thickness to that of the diffusion layer. However, real fuel cell devices are operated at a potential where both oxygen transport and kinetics are rate-limiting.In this case, the reaction rate is proportional to oxygen activity in the ionic liquid. Using the model developed, we were able to conclude that oxygen activity should be maximised over transport and that a single monolayer is the ideal ionic liquid coverage.Finally, we used a sorption analyser to study the ionic liquid distribution on the surface of the catalyst. We observed that rather than forming the ideal monolayer, the ionic liquid fills up the smallest pores first and tends to block bigger pores. Our current work is focusing on improving catalyst wettability by the ionic liquid to approach the ideal distribution. Conclusion In conclusion, the cation and anion of ionic liquids can be independently tailored to control hydrophobicity on one side and O2 activity on the other. This can ultimately allow to optimize the binding energy of key intermediates and maximize the reactant concentration at the active site. Our approach provides a rational method to improve the activity of noble-metal-free oxygen reduction catalysts. [1] Favero et al., Adv. Energy Sustainability Res. 2021, 2, 2000062 [2] Casalongue et al., Nat. Commun., 2013, 4, 2817 [3] Huang et al., J. Electrochem. Soc., 2017, 164, F1448 [4] Vanhouette et al., RCS Adv., 2018, 8, 4525 Figure 1