Perspective: A controversial benchmark system for water-oxide interfaces: H 2 O/TiO 2 (110).

The interaction of water with the single-crystalline rutile TiO2 (110) surface has been the object of intense investigations with both experimental and computational methods. Not only is TiO2 (110) widely considered the prototypical oxide surface, its interaction with water is also important in many applications where this material is used. At first, experimental measurements were hampered by the fact that preparation recipes for well-controlled surfaces had yet to be developed, but clear experimental evidence that water dissociation at defects including oxygen vacancies and steps emerged. For a perfect TiO2 (110) surface, however, an intense debate has evolved whether or not water adsorbs as an intact molecule or if it dissociates by donating a proton to a so-called bridge-bonded surface oxygen atom. Computational studies agree that the energy difference between these two states is very small and thus depends sensitively on the computational setup and on the approximations used in density functional theory (DFT). While a recent molecular beam/STM experiment [Z.-T. Wang et al., Proc. Natl. Acad. Sci. U. S. A. 114(8), 1801-1805 (2017)] gives conclusive evidence for a slight preference (0.035 eV) for molecular water and a small activation energy of (0.36 eV) for dissociation, understanding the interface between liquid water and TiO2 (110) arises as the next controversial frontier.