Kinetics of the Hydrogen Evolution Reaction at Rh-Doped SrTiO3 Driven By Visible Light: A Study By Photoelectrochemical Impedance Spectroscopy

Abstract

Because of the gradual exhaustion of carbon-based fuels and the problematic issues associated with their combustion, modern human societies are urged to find alternative, sustainable and renewable energy sources.1 The enormous amount of radiative energy arriving to earth from sun appears as an excellent alternative source of primary energy. In this context, the use of semiconducting materials that can absorb visible light and transform solar radiations into clean fuels is a topic of significant scientific interest. In particular, the hydrogen evolution reaction (HER) that produces hydrogen from earth-abundant water is a widely targeted reaction.2 In electrochemistry, the HER is a multistep process for which a large variety of efficient electrocatalysts, operating in different media and at different pH, have already been developed. But the kinetics of water photo-reduction at photocathodes involves different reaction steps and still requires improvement. In this communication, we report on the kinetics of water photo-reduction using Rh:SrTiO3 as light-absorbing semiconductor. The charge transfer (kt ) and recombination (kr ) rate constants have been determined in order to unravel the heterogeneous mechanism for the HER that prevails at Rh:SrTiO3 / electrolyte interfaces. Rh:SrTiO3 samples were prepared by calcinations of appropriate precursors. The X-ray diffraction pattern of the semiconducting powder confirmed the perovskite structure of SrTiO3 as a pure phase. Though, the presence of Rh as the dopant could not be put in evidence by XRD, an indication that the substitution of Rh4+ atoms by Ti4+ in the crystal lattice did not distort appreciably the structure of the solid. Further characterization was performed by X-ray photoelectron spectroscopy (XPS), and the presence of Rh in the sample was clearly established. The band structure of the photoelectrode was characterized by UV-Vis and Diffuse Reflectance spectroscopies, suggesting a band gap of 2.7 eV as reported before.3 Linear sweep voltammetry at different levels of illumination (λ > 450 nm) showed an increase in cathodic currents upon illumination, which expressed the role of the device as a photocathode. Subsequently, the photoelectrode was characterized using photoelectrochemical impedance spectroscopy (PEIS). The shape of PEIS-diagrams was found to change upon potential cycling. This may be due to either the accumulation of reactive intermediates at the surface of the photoelectrode or to the reduction of Rh4+ to Rh3+ in the material that causes the catalytic activity of the material. PEIS-diagrams then attained a stable shape which permitted to perform a full kinetic analysis. It was observed that kt was independent of light intensity and that kr increased slightly with increasing light power. Conversely, kt increased exponentially while the electrode potential became more negative. Such fact indicated the presence of surface states that caused Fermi level pinning. As a result, an important part of the applied potential dropped in the Helmholtz layer. The Tafel plot of ln kt vs. applied potential exhibited a slope of 93 mV/dec (α = 0.28), which suggested that water discharge at the surface (Volmer reaction, H2O + e- + M = M-H + OH-) was the rate determining step. Overall, a thorough kinetic determination of the HER, photo-assisted by visible light, has been performed and details will be reported during the Meeting.