Resonant Two-Photon Photoemission from Ti 3d Defect States of TiO2(110) Revisited
Adam Argondizzo,Shijing Tan,Hrvoje Petek
DOI: https://doi.org/10.1021/acs.jpcc.6b04517
2016-01-01
The Journal of Physical Chemistry C
Abstract:Recently three groups reported on resonant two-photon photoemission (2PP) from Ti 3d defect states on rutile TiO2(110) surface. The first publication (Argondizzo, A.; Cui, X.; Wang, C.; Sun, H.; Shang, H.; Zhao, J.; Petek, H. Phys. Rev. B 2015, 91, 155429) to appear assigned the resonance enhancement of the two-photon absorption processes to excitations from the nominally t(2g) symmetry Ti3+ 3d defect states of reduced bulk TiO2 through a one-photon resonance with nearly degenerate pair of e(g) symmetry intermediate states, before absorbing another photon into the photoemission continuum. This resonance occurs for h nu similar to 3.6 eV, and has pronounced dependence on the alignment of the electric field vector of the excitation light with the in-plane crystalline axes of the TiO2(110) surface. Subsequently, two other groups reported similar 2PP spectra of reduced TiO2 surfaces, with one claiming a single d-d transition of OH terminated TiO2 at a lower energy than in the primary report (Wang, Z.; Wen, B.; Hao, Q; Liu, L.-M.; Zhou, C.; Mao, X.; Lang, X.; Yin, W.-J.; Dai, D.; Selloni, A.; Yang, X. J. Am. Chem. Soc. 2015, 137, 9146-9152), and the other attributing the resonance to wet electron states of low density surface OH, rather than the bulk d-d transitions (Zhang, Y.; Payne, D. T.; Pang, C. L.; Fielding, H. H.; Thornton, G. J. Phys. Chem. Lett. 2015, 6, 3391-3395). Both groups made bold claims that this UV resonance of reduced TiO2 could confer photocatalytic activity in the visible region below the fundamental band gap at 3.0 eV. Here we reinvestigate the resonant 2PP spectra of reduced TiO2(110) surface with wavelength tunable ultrafast laser pulse excitation for different surface preparations, temperatures, and modifications by molecular adsorbates. We conclude that the original assignment to predominantly bulk t(2g)-e(g) transitions entirely explains the new data, and the surface OH makes no contribution to the observed spectra. Moreover, the t(2g)-e(g) transition is inconsequential for sub-band-gap photocatalytic activity, but rather illuminates the nature of trapped carriers in TiO2 and their interactions with chemisorbed molecules.