Modeling TiO2 surfaces for hydrogen generation from water
Titanium dioxide (TiO2), versatile and widely used for photocatalysis, is a promising material for air and water remediation with broad applications in coating technologies and maintenance-free surfaces.1 Using both experimental approaches and theoretic modeling, we investigated the prospects of using Niobium (Nb)-doped TiO2 as an active photocatalytic.2 At fairly high doping concentrations (10-20 mol %), we showed that this material is a poor solar-active photocatalyst compared to TiO2 despite the fact that Nb doping extends optical absorption into the visible spectrum and uses a larger portion of sunlight.
Using the interplay between measurements by Fourier transform spectroscopy (FTIR) and density functional theory (DFT) calculations, we also investigated in detail the importance of cation coordination and bonding of intermediate reaction products for the total photocatalytic oxidation rate.
Cost-effective and sustainable hydrogen production makes photo-induced decomposition of water on TiO2 electrodes a topic of special interest.3 The possibility that water can be photoelectrochemically split by illuminating TiO2 was first demonstrated thirty-five years ago.4 However, the high intrinsic band gaps of the two different structures of TiO2 (the anatase and rutile bandgaps are 3.2 eV) allow only absorption of the ultraviolet part of the solar irradiation. Doping with anionic species (e.g., N, C, S, F) or cationic species (transition metals), has shown promise in reducing the TiO2 band gap, which would thereby extend optical absorption to the visible light region of the solar spectrum.
Although water adsorption on TiO2 has been been extensively investigated in both theory and experiment, the type of adsorption remains a matter of controversy. Experiments suggest that, for the most part, H2O adsorbs associatively. Dissociation may occur at low coverages (<15%) associated with defective sites.5 By contrast, theoretical modeling predicts various degrees of both associative and dissociative water adsorption.6
The effect of cationic and anionic dopants on the water adsorption process for both anatase and rutile structures of TiO2, has also been investigated on a theoretic basis by our research group (see Figure 1). For non-doped surfaces, we found anatase to be more reactive towards water adsorption than rutile. This tendency was greater when using nearby N substituents within the surface. However, neighboring H2O adsorbates were required in order to to improve the probability of disrupting the water bond. The situation was totally different when using Nb dopants. Here Nb was found to largely affect H2O adsorption on rutile, and to clearly affect the tendency for H2O dissociative adsorption at higher surface coverages.
Karin Larsson is a Professor in Inorganic Chemistry at the University of Uppsala. Her major research focus is theoretical investigation of surface processes on semiconducting materials.
