(669f) Proton Transfer and Acidity at the IrO2-Water Interface from Deep Potential Molecular Dynamics Simulations

Iridium oxide (IrO₂), a prototypical electrocatalyst for the oxygen evolution reaction (OER), operates via inner-sphere processes that are highly sensitive to the composition and structure of the oxide-water interface. One fundamental characteristic of oxide surfaces that controls the surface composition and effective surface charge is the point of zero charge (pH_PZC). Experimentally, this can be used as a tuning knob to optimize the OER kinetics through, e.g. the appropriate selection of electrolytes and/or surface modification of the electrocatalyst. However, central to the atomistic design of the oxide electrocatalyst, is the estimation of the acid dissociation constants (pK_a) of the different surface sites, which are directly related to the pH_PZC but still not accessible using current experimental techniques. Computationally, their estimation has also been limited by the length and time scales accessible through ab-initio molecular dynamics simulations (AIMD).

In this presentation, I will discuss our approach on using a deep neural network potential, trained on accurate first-principles data to represent the potential energy surface of the rutile IrO₂(110)-water interface. We use this to provide fundamental and hitherto unreported insights on the hydration structure, water dissociation fraction, and proton transfer mechanisms by performing long (nanosecond) timescale deep potential molecular dynamics (DPMD) simulations of the interface. Subsequently, we combine DPMD with enhanced sampling methods to perform an efficient exploration of the free-energy surface of the acid-base reactions and obtain quantitative estimates of the pK_a of the different surface sites and the pH_PZC, found to be in very good agreement with experiments.