Comment on “Imaging of the Hydrogen Subsurface Site in RutileTiO2”

Abstract

The adsorption of hydrogen on oxides leading to the formation of OH species and the subsurface diffusion of H atoms are topics of pronounced interest with regard to understanding oxide surface chemistry. In a recent Letter by Enevoldsen et al. [1], the authors use ab initio density functional theory (DFT) calculations to determine the activation energy for subsurface diffusion of H atoms. The value obtained by them, 2.4 eV, is rather high and would imply that this process can be basically neglected at room temperature. In fact, this value lies more than 1 eV above the results reported by us in a previous publication [2], 1.11 eV. Enevoldsen et al. explain this difference by the different H-atom coverage in the two cases. We would like to point out that this explanation is not correct; the rather large difference arises from the fact that in [1] an inadequate diffusion mechanism has been considered. The path taken by the H atom to enter the bulk considered in [1] starts at a surface bridging site O b and from there leads directly to a subsurface position O sub , whereas the optimum path described in [2] starts from a threefold surface oxygen O 3f site. The two different paths are depicted in Fig. 1. We repeated the DFT calculations for the nonoptimal path considered in [1] and were able to recover a value of 2.6 eV very close to that of 2.4 eV, reported by Enevoldsen et al. As described above, this value is unphysically high and is also not consistent with He-atom scattering experiments [3] where it was found that the hydrogenated surface undergoes a structural rearrangement at a temperature of 388 K. This temperature would correspond to an activation energy of $0:97 eV.Recent DFT calculations reported by Kowalski et al.[4] also support the threefold to subsurface diffusion path proposed by us [2]. Table I compares the values reported for H diffusion paths from similar DFT calculations. The barriers calculated for O b -O sub diffusion are in the order of 2.4-2.56 eV, while the O 3f -O sub barriers are lower, 0.93-1.11 eV. The surface diffusion O b -O 3f is 0.63 eV. The hydrogen coverage , considered by the authors of [1] to play a crucial role, has a rather limited impact on the calculated barriers, in the case of the direct diffusion 2.4 eV ( ¼ 1=8, Ref. [1]) versus 2.56 eV ( ¼ 1, Ref. [4]).In conclusion, to correctly obtain a theoretical value for the activation energy governing subsurface diffusion of H atoms, the minimum-energy path has to be considered. For H atoms adsorbed on rutile TiO 2 ð110Þ, this path starts from a threefold hollow site, possibly after an easy diffusion O b -O 3f , and not from a bridge site. For this minimumenergy path the activation energy amounts to 1.1 eV. This value is fully consistent with previous experimental data and clearly demonstrates that subsurface diffusion of H atoms is a process occurring at room temperature, with important implications for H-atom doping and for hydrogenation processes occurring at metal oxide surfaces [5].COST D-36 and D-41 actions are acknowledged.

The Subsurface Diffusion of Hydrogen on Rutile TiO2 Surfaces: A Periodic DFT Study

“…However, the higher barrier (1.00 eV) makes this process less likely to occur compared to other terminations. The value of 1.00 eV is consistent with the previous reports (1.11 eV [13], 0.93 eV [47]). The small difference can come from the different coverage, layers and sizes, and the use of GGA+U in the present work.…”

“…In this process, the calculated ∆E is -0.08 eV, and the activation barrier of H migration is 0.79 eV, indicating that H diffusion to the subsurface region is an activated process, thermodynamically slightly favorable. For (110) it was shown previously that the direct migration O2C to subsurface was more energetic [47]. Fig.…”

Hydrogen Loading of Oxide Powder Particles: A Transmission IR Study for the Case of Zinc Oxide

“…[4,5] Finally, in a recent paper by Yin et al [6] it has been found that heating of a hydroxylated single crystal TiO 2 (110) substrate does not lead to the expected desorption of H 2 or H 2 O molecules, but instead to a diffusion of H atoms into the bulk of the crystal. [7] In theoretical calculations this surprising behaviour was rationalized by the large activation energy of 2.52 eV for recombination of H atoms to form dihydrogen at the TiO 2 surface. This value considerably exceeds that for diffusion into the bulk, for which an activation energy of 1.1 eV has been reported.…”

“…This value considerably exceeds that for diffusion into the bulk, for which an activation energy of 1.1 eV has been reported. [6][7][8] Recently published STM images have provided direct evidence for the population of such subsurface sites by H atoms on TiO 2 substrates. [9] In case of ZnO single crystal surfaces it has been recently demonstrated that exposure of the mixed-terminated ZnO(10 " 10) to atomic hydrogen leads to the metallization of the surface, [10] whereas exposure of the oxygen-terminated OZnO(000 " 1) surface leads to the occupation of bulk interstitial sites.…”

Understanding the Role of Rutile TiO2 Surface Orientation on Molecular Hydrogen Activation

“…It is well known that the surface properties strongly vary with different crystallographic orientations, which can greatly affect their reactivity [42,43,44,45,46]. For rutile TiO 2 , the main exposed low energy surface is the (110) surface, which is also the most studied [23,28,47,48,49,50,51].…”