Core-shell bimetallic nanoparticles are effective in enhancing activity and durability of catalysts for oxygen reduction and hydrogen oxidation reactions in fuel cells. It is less clear about using this approach to develop advanced metal oxide nanocatalysts for oxygen evolution reaction (OER), partly due to insufficient understanding of the more complicated OER that involves multiple intermediates adsorbed on both metal and oxygen active sites. In this study, we deduced a phase diagram of the dominant adsorbates based on experimental results and density functional theory (DFT) calculations.

We prepared RuO₂ and RuO₂@IrO₂ core-shell nanocatalysts via a facile method using carbon support as a template without and with H₂IrCl₆ as the Ir precursor. During calcinations in air for 1 h, Ru catalyzes a complete burning out of the carbon, which facilitates the formation of small dioxide nanoparticles. The catalysts’ weights after syntheses were consistent with the masses of pure metal dioxide - no cleaning procedure needed.  

We compare voltammetry signatures (Fig. 1) and polarization curves (Fig. 2) of the two synthesized and commercial IrO₂ catalysts. The currents were normalized to the oxide surface area calculated from the average particle sizes and catalyst loadings. The integrated charges between 0.4 and 1.23 V are all close to 1e per metal atom for the rutile (110) surface, supporting a 1e surface reaction over a broad potential range. RuO₂ exhibited higher current, but larger Tafel slope at high potentials than IrO₂ did, consistent with literature results.¹ The OER activity increased on RuO₂@IrO₂ core-shell nanocatalyst from that of IrO₂ while the Tafel slope remained similar to that of IrO₂, which bodes well for performance  in water electrolyzers operating at ~1.8 V.

To gain fundamental understanding of the OER mechanism on these most active catalysts, we derived a kinetic equation using a simplified model with one major adsorbed intermediate, *O, and two activation barriers (See Fig. 2 caption).

Connecting the fitted free energies with the most stable adsorbates determined by DFT calculations for metal oxidation state from 3 to 7, we deduced the phase diagrams. The one for IrO₂ is shown in Fig. 3b. Water dissociative adsorption is highly facile over the coordination unsaturated site (cus) of metal and bridge oxygen (Fig. 3a), resulting in an adsorption phase of *OH H*. Ir oxidation state remains 4 because changes by +2 for each *O and -1 for each *H cancel out. In the OER potential region above 1.4 V, metal oxidation state rises with proton removal - first from adsorbed OH to the *O *H phase and then to the *O phase. It is followed by second water dissociation occurring over the *O-covered site and bridge oxygen site with one proton removed to the *OO *H phase.  Below 1.2 V, *OH desorbes from M_cus, yielding a *H phase with H adsorbed on bridge oxygen site. In situX-ray absorption near edge spectra (XANES) support a change of metal oxidation state from 4 to 3 with decreasing potential from 1.3 V to 0.5 V (Fig. 3c).

Comparison of DFT-calculated and experimentally-derived free energy diagrams for the OER will be presented to shed lights on how a RuO₂ core enhances the OER activity on the IrO₂ shell.

Acknowledgements

This research was supported by the US Department of Energy, Division of Chemical Sciences, Geosciences and Biosciences Division under contract DE-SC0012704.

References

(1)         Lee, Y.; Suntivich, J.; May, K. J.; Perry, E. E.; Shao-horn, Y. J. Phys. Chem. Lett. 2012, 3, 399.