Edge sites on TiO₂ are photocatalytic active site

Abstract

Although TiO₂ has been widely used as an efficient photocatalyst, the photocatalytic active sites remain ambiguous. Using rutile TiO₂(110) surfaces with well-defined defects, we herein unambiguously identify that edge sites on TiO₂ surfaces with interstitial Ti³⁺ defects are photocatalytically active. On the oxidized TiO₂(110) surface with TiO₂ islands, chemisorbed CO₂ and CO are photo-inactive; on the TiO₂(110) surface with surface bridging oxygen vacancies and bulk interstitial Ti³⁺ defects, CO₂ and CO chemisorbed at the vacancy sites become photoactive; on the TiO₂(110) surface with TiO₂ islands and interstitial Ti³⁺ defects, CO₂ and CO chemisorbed at the edge sites of TiO₂ islands are also photoactive. CO chemisorbed at the surface oxygen vacancies shows the highest photo-induced desorption probability, while CO₂ chemisorbed at the edge sites of TiO₂ islands with interstitial Ti³⁺ exhibits the highest photo-induced desorption probability. Considering their abundance on powder TiO₂ photocatalysts, the edge sites are among the photocatalytic active sites contributing to TiO₂ photocatalysis.

---

Introduction

Photocatalysis has been increasingly demonstrated to be capable of reshaping chemistry in a green and sustainable manner.^1–7 TiO₂ as an earth-abundant, stable, and environmentally-friendly photocatalyst has been extensively studied since the seminal report of UV light-induced water splitting on the TiO₂ electrode,⁸ but the photocatalytic active sites remain ambiguous. This has consequently inspired numerous fundamental studies of TiO₂ photocatalysis using various single-crystal TiO₂ model catalysts to identify photocatalytic active sites and elucidate the underlying reaction mechanisms.^9–16 The results show that the active site on TiO₂ photocatalysts varies with the photocatalyzed reactions.^17–30 Photogenerated holes at point defect sites mediate the desorption of CO and CO₂, such as surface bridging oxygen vacancies (V_O), hydroxyl groups (OH), and Ti_5c sites near Ti interstitials (Ti_int) on various rutile TiO₂ surfaces,^17,19,20 while the interstitial Ti³⁺ site in the rutile TiO₂(001) subsurface also serves as an active site.¹⁷ CO₂ chemisorbed at the double oxygen vacancies on the rutile TiO₂(011)-(2 × 1) surface exhibits photoreactivity.¹⁷ Molecular O₂ adsorbed on the TiO₂ surface acquires excess negative charges at the terminal Ti atoms (Ti_5c on TiO₂(110) and Ti_4c on anatase TiO₂(001)-(1 × 4) surfaces) and at V_O sites to undergo photodesorption and photodissociation, respectively.^17,18,21,31 Methanol undergoes the photogenerated hole-mediated oxidation reaction via the methoxy species at Ti_5c sites on various TiO₂ surfaces,^{22–24,26,28} as well as Ti_4c sites on anatase TiO₂(001)-(1 × 4) surfaces.²⁷ Meanwhile, methoxy species at the Ti_5c site on the reconstructed rutile TiO₂(001)-(1 × 1) surface also exhibited photoreduction reactivity.²⁹ Edge sites are the most common and abundant intrinsic defect on powder catalysts.^32–36 Step edges were found as charge-trapping centers with increased reactivity towards O₂ adsorption on the anatase TiO₂(101) surface³² and as active sites for water and alcohol dissociation on the rutile TiO₂(110) surface.^34–36 Herein, using the rutile TiO₂(110) surface with well-defined defects including surface bridging oxygen vacancies, bulk interstitial Ti³⁺ sites, and TiO₂ islands, we successfully identify the edge sites of TiO₂ islands as a novel type of photocatalytic active sites, suggesting that the widely-existing edge sites are the probable photocatalytic active sites on powder TiO₂ photocatalysts.

Results and discussion

The rutile TiO₂(110) surface is well-known to exhibit a variety of reconstructions and defect structures depending on the sample treatments, as illustrated in Fig. 1.^16,37 Under ultra-high vacuum (UHV) conditions, V_O defects are easily generated on the TiO₂(110) surface during sputtering and annealing, which is often accompanied by the formation of Ti_int defects in the subsurface or bulk.^9,13,16 Upon subsequent oxidation treatments, Ti_int species migrate to the surface and react with oxygen to form TiO₂ islands.^38–41 Sputtering and annealing in UHV at 600–800 K produce TiO₂(110)-(1 × 1) terraces with step bunches, while annealing at 800–900 K forms TiO₂(110)-(1 × 1) flat terraces with single steps. Annealing at 900–1100 K leads to the formation of double-ridge structures on the terraces, arising from the partially reduced oxide.⁴² Further annealing at 1150 K reconstructs the original TiO₂(110)-(1 × 1) surface into a TiO₂(110)-(1 × 2) structure.^42,43 Additionally, flat (1 × 1)-terminated TiO₂(110) surfaces, when annealed in oxygen at 470–660 K, develop irregular networks of rosettes, Ti₂O₃ strands, and small (1 × 1) islands. These rough surfaces can be flattened into regular (1×1) terraces by annealing above 830 K, either in oxygen or under UHV conditions.^44–47

Fig. 1 Schematic illustration of defect structures of the TiO₂(110)-(1 × 1) surface with various sample treatments.

A new TiO₂(110) single crystal was initially cleaned by argon-ion bombardment, resulting in argon embedded in the sample.⁴⁸ Repeated cycles of annealing in oxygen and vacuum annealing effectively removed the residual argon (Fig. S1). The cleaned sample was then used for the subsequent experiments. Three different TiO₂(110) surfaces with various types of defects were prepared through different treatments, as illustrated in Fig. 2A. The clean TiO₂(110) surface was first annealed in oxygen at 700 K for 30 min, followed by vacuum annealing at either 800 K or 1000 K for 10 min. These surfaces are referred to r-TiO₂(110)-800 K and r-TiO₂(110)-1000 K, respectively. The r-TiO₂(110)-800 K surface was fully oxidized by exposure to 50 L of O₂ at 110 K, followed by a flash to 600 K, and is denoted as o-TiO₂(110). This process can be referred to as O₂ oxidation treatment. Fig. 2B and C show the X-ray photoelectron spectroscopy (XPS) of Ti 2p spectra for these three surfaces and the corresponding Ti 2p difference spectra. The characteristic Ti³⁺ 2p_3/2 feature at 457.7 eV verifies the existence of defects on both r-TiO₂(110)-800 K and r-TiO₂(110)-1000 K surfaces (Fig. S2). Additionally, Ti³⁺ species were regenerated at temperatures above 600 K, as indicated by the Ti 2p XPS spectra during O₂ exposure and annealing (Fig. S3).

Fig. 2 (A) Schematic illustration, (B) Ti 2p XPS spectra, and (C) the corresponding difference spectra of r-TiO₂(110)-1000 K, r-TiO₂(110)-800 K, and o-TiO₂(110) surfaces. TPD spectra of r-TiO₂(110)-1000 K, r-TiO₂(110)-800 K, and o-TiO₂(110) surfaces exposed to (D) 50 L O₂, (E) 1 L CO₂, and (F) 1 L CO at 110 K.

Adsorption of O₂, CO₂, and CO molecules at 110 K was studied to probe the structures of various r-TiO₂(110) surfaces. In the O₂ temperature-programmed desorption (TPD) spectra (Fig. 2D), a major desorption peak at 400 K with a shoulder at 440 K and a minor desorption peak at 160 K appeared for r-TiO₂(110)-1000 K, two desorption peaks at 440 and 160 K for r-TiO₂(110)-800 K, and only one desorption peak at 160 K for o-TiO₂(110). The high-temperature O₂ desorption peaks were previously attributed to molecularly adsorbed O₂ species at Vo^49–51 or Ti_int sites.^38–40 We carried out repeated O₂-TPD experiments on r-TiO₂(110)-1000 K (Fig. 3A), during which the r-TiO₂(110)-1000 K surface gradually converted to the r-TiO₂(110)-800 K surface via the reaction of O₂ with Ti_int to form TiO₂ islands.^38–40 It can be seen that the O₂ desorption peak at 440 K grew at the expense of that at 400 K while the O₂ desorption peak at 160 K barely changed. Thus, the O₂ desorption peaks at 400 and 440 K arise from molecularly adsorbed O₂ species at V_O and Ti_int sites, respectively. Previous studies proposed that O₂ chemisorbed at Ti_5c sites near Ti_int species in the near-surface region on reduced TiO₂(110) surfaces¹⁸ or at step sites^31,32 gave low-temperature desorption peaks. As discussed below, the O₂ desorption peaks at 160 K observed herein arise from molecularly adsorbed O₂ species at step sites on the surface.

Fig. 3 (A) O₂ TPD spectra of r-TiO₂(110)-1000 K surface with repeated exposures of 50 L O₂ at 110 K. The O₂ TPD is flashed to 800 K. (B) CO₂ TPD spectra of r-TiO₂(110)-1000 K surface exposed to saturated CO₂ before and after O₂ TPD experiments at 110 K.

In the CO₂ TPD spectra (Fig. 2E and S4), at 110 K (black line in Fig. 2E), a major desorption peak at 144 K and another tiny one at 178 K appeared for r-TiO₂(110)-1000 K, corresponding to CO₂ adsorbed at Ti_5c and V_O sites,^52–54 respectively. For both r-TiO₂(110)-800 K and o-TiO₂(110), the major desorption peak of CO₂ adsorbed at Ti_5c sites remained but the desorption peak of CO₂ adsorbed at V_O sites disappeared, meanwhile, a new shoulder desorption peak at 161 K emerged. Fig. 3B shows the TPD spectra following CO₂ adsorption on the r-TiO₂(110)-1000 K surface subjected to repeated O₂ TPD experiments. As the r-TiO₂(110)-1000 K surface gradually converted to the r-TiO₂(110)-800 K surface, both CO₂ desorption peaks at 144 and 178 K attenuated, and the peak at 178 K eventually disappeared, while the shoulder feature at 161 K emerged and grew and thus could be assigned to CO₂ chemisorbed at the edge sites of TiO₂ islands.

In the CO TPD spectra (Fig. 2F and S5), a major broad desorption peak at 144 K and another tiny one at 180 K appeared for r-TiO₂(110)-1000 K, corresponding to CO adsorbed at Ti_5c and V_O sites,⁵⁵ respectively. Similar to the case of CO₂ adsorption, the desorption feature of CO at Ti_5c sites weakened for r-TiO₂(110)-800 K and o-TiO₂(110) and that at V_O sites disappeared, while a new shoulder feature emerged, corresponding to CO chemisorbed at the edge sites of TiO₂ islands.

Thus, as probed by O₂, CO₂, and CO adsorption, the defects primarily are V_O and Ti_int on the r-TiO₂(110)-1000 K surface, Ti_int and TiO₂ islands on the r-TiO₂(110)-800 K surface, and TiO₂ islands on the o-TiO₂(110) surface. It is worth noting that the O₂ and CO₂ TPD spectra barely changed with repeated exposures of 50 L O₂ on the r-TiO₂(110)-800 K surface (Fig. S6), indicating that the Ti_int defects on the r-TiO₂(110)-800 K surface cannot be oxidized by O₂ adsorption at 110 K followed by heating to a temperature higher than 600 K but they can by O₂ oxidation treatment. Meanwhile, both CO₂ and CO desorption peaks at Ti_5c sites on the r-TiO₂(110)-800 K and o-TiO₂(110) surfaces are smaller than those on the r-TiO₂(110)-1000 K surface, indicating that the Ti_5c density on TiO₂ islands should be less than on the underlying r-TiO₂(110) surface. Probed by O₂ adsorption (Fig. S7), both r-TiO₂(110)-1000 K and r-TiO₂(110)-800 K surfaces can be reproducibly prepared, and so can the o-TiO₂(110) surface.

Photochemistry of O₂, CO₂, and CO molecules chemisorbed on various r-TiO₂(110) surfaces was then examined. As shown in Fig. 4A, upon 20 min UV light illumination, the desorption feature of molecular O₂ absorbed at V_O or Ti_int sites on r-TiO₂(110)-1000 K and r-TiO₂(110)-800 K significantly weakened, suggesting the occurrence of photoinduced O₂(a) dissociation or desorption, in line with previous findings,^{18–21,56–64} however, the minor desorption feature at 160 K remained unaffected. The O₂ adsorbed at Ti_5c sites near Ti_int defects in the near-surface region on the reduced TiO₂(110) surfaces was reported to be photoactive,¹⁸ thus the low-temperature desorption feature at 160 K probably originates from O₂ chemisorbed at step sites.

Fig. 4 TPD spectra of r-TiO₂(110)-1000 K, r-TiO₂(110)-800 K, and o-TiO₂(110) surfaces exposed to (A) 50 L O₂, (B) 1 L CO₂, and (C) 1 L CO without (black) and with (red) 20 min UV light irradiation at 110 K.

As shown in Fig. 4B and C and S8 and S9, UV light illumination barely changed the CO₂ and CO TPD profiles from the o-TiO₂(110) surface, indicating that CO₂ and CO chemisorbed at Ti_5c sites and edge sites of TiO₂ islands on the oxidized TiO₂(110) surface are photoinactive. However, weakening of desorption peaks of CO₂ and CO chemisorbed at Ti_5c sites on r-TiO₂(110)-1000 K and r-TiO₂(110)-800 K were clearly observed upon UV light illumination while other desorption features remained the same. Since CO₂ and CO chemisorbed at Ti_5c sites are photoinactive, the observed apparent decrease of their low-temperature desorption peaks is analogous to the photo-induced desorption of CO reported previously.^17,19,20 The “low-temperature depletion” phenomenon describes the process wherein strongly chemisorbed CO₂ and CO species undergo photo-induced desorption under UV light illumination, with the resulting vacant sites subsequently occupied by weakly-chemisorbed CO₂ and CO species via surface diffusion. Such a “low-temperature depletion” phenomenon leads to the unchanged desorption structures of strongly-chemisorbed CO₂ and CO species and decreased desorption peaks of weakly-chemisorbed CO₂ and CO species in the TPD spectra following UV light illumination. Therefore, CO₂ and CO chemisorbed at V_O sites and edge sites of TiO₂ islands with Ti_int defects are photoactive. As shown in Fig. 5, CO₂ chemisorbed at the edge sites of TiO₂ islands with Ti_int defects and CO at V_O sites exhibit the highest photo-induced desorption probability, meanwhile, CO₂ and CO chemisorbed at the edge sites of TiO₂ islands with Ti_int defects show similar photo-induced desorption probabilities. Herein, the photodesorption probability incorporates multiple factors, including direct photodesorption and thermal/photostimulated diffusion. These observations demonstrate site-specific photocatalysis.^65–68

Fig. 5 Photodesorption probability of CO and CO₂ on the r-TiO₂(110)-1000 K, r-TiO₂(110)-800 K, and o-TiO₂(110) surfaces at 110 K. Photodesorption probability is calculated as (integrated TPD peak without UV light illumination − integrated TPD peak with illumination)/integrated TPD peak without UV light illumination.

The above results have important implications for powder TiO₂ photocatalysis. On one hand, TiO₂ powder photocatalyst generally exhibits edge sites, V_O and Ti_int defects, among which the edge site is significantly more abundant than the V_O and Ti_int defects, therefore, the edge site is probably the dominant photocatalytic active site on the TiO₂ powder photocatalyst. On the other hand, the edge site of the TiO₂ powder photocatalyst generally exhibits stronger adsorption ability than the terrace site, and the photoinactive adsorbed species on the terrace site can contribute to the photocatalytic reaction by surface migration to the vacant photocatalytic active edge site. Notably, while our study with single-crystal model catalysts unambiguously demonstrates the edge sites as one type of photocatalytic active sites, real powder TiO₂ photocatalysts expose various types of crystal facets and defects whose photocatalytic activity need to be studied in order to comprehensively understand the apparent photocatalytic performance.

Conclusions

In summary, we have successfully used surface chemistry and photochemistry of O₂, CO₂, and CO to probe the photocatalytic activity of the oxidized TiO₂(110) surface with TiO₂ islands, TiO₂(110) surface with surface bridging oxygen vacancies and bulk interstitial Ti³⁺ defects, and TiO₂(110) surface with TiO₂ islands and interstitial Ti³⁺ defects. CO₂ and CO chemisorbed at the Ti_5c site are photoinactive, while CO₂ and CO chemisorbed at the surface bridging the oxygen vacancy site and edge site of TiO₂ islands with bulk interstitial Ti³⁺ defects are photoactive. These results for the first time unambiguously identify the abundant edge sites on TiO₂ powder photocatalysts as the photocatalytic active site, which greatly deepens the fundamental understanding of TiO₂ photocatalysis and provides executable guide for the structural design of efficient TiO₂ photocatalysts.

Author contributions

L. W. performed the experiments; Z. W., Z. W., H. X., P. C., and J. S. assisted the experiments; L. W., Z. W., and W. H. wrote and revised the manuscript. W. H. and Z. W. acquired the funding. W. H. supervised all aspects of the work. All authors have given approval to the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5cy01201e.

Acknowledgements

We appreciate the support from the National Key R&D Program of the Ministry of Science and Technology of China (2021YFA1502804), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0450102), the National Natural Science Foundation of China (22250710677, 52261135636, 22202191, 22472160), the Changjiang Scholars Program of the Ministry of Education of China, and the USTC Research Funds of the Double First-Class Initiative (YD9990002026). This work was partially carried out at the Instruments Center for Physical Science, University of Science and Technology of China.

Notes and references

1. Y. Wang, A. Vogel, M. Sachs, R. S. Sprick, L. Wilbraham, S. J. A. Moniz, R. Godin, M. A. Zwijnenburg, J. R. Durrant, A. I. Cooper and J. Tang, Nat. Energy, 2019, 4, 746–760.
2. X. Lang, X. Chen and J. Zhao, Chem. Soc. Rev., 2013, 43, 473–486.
3. Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han and C. Li, Chem. Rev., 2014, 114, 9987–10043.
4. X. Li, C. Wang and J. Tang, Nat. Rev. Mater., 2022, 7, 617–632.
5. G. Liao, G. Ding, B. Yang and C. Li, Precis. Chem., 2024, 2, 49–56.
6. Y. Wu, M. Wu, Y. Sun and Y. Xie, Sci. China:Chem., 2024, 67, 2434–2447.
7. P. Babu, H. Park and J. Y. Park, Surfactant Sci. Technol., 2023, 1, 29.
8. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
9. U. Diebold, Surf. Sci. Rep., 2003, 48, 53–229.
10. M. A. Henderson and I. Lyubinetsky, Chem. Rev., 2013, 113, 4428–4455.
11. C. L. Pang, R. Lindsay and G. Thornton, Chem. Rev., 2013, 113, 3887–3948.
12. Q. Guo, C. Zhou, Z. Ma, Z. Ren, H. Fan and X. Yang, Chem. Soc. Rev., 2016, 45, 3701–3730.
13. W.-J. Yin, B. Wen, C. Zhou, A. Selloni and L.-M. Liu, Surf. Sci. Rep., 2018, 73, 58–82.
14. S. Chen, F. Xiong and W. Huang, Surf. Sci. Rep., 2019, 74, 100471.
15. C. A. Walenta, M. Tschurl and U. Heiz, J. Phys.:Condens. Matter, 2019, 31, 473002.
16. L. Wu, C. Fu and W. Huang, Phys. Chem. Chem. Phys., 2020, 22, 9875.
17. L. Wu, Z. Wang, F. Xiong, G. Sun, P. Chai, Z. Zhang, H. Xu, C. Fu and W. Huang, J. Chem. Phys., 2020, 152, 044702.
18. F. Xiong, L.-L. Yin, F. Li, Z. Wu, Z. Wang, G. Sun, H. Xu, P. Chai, X.-Q. Gong and W. Huang, J. Phys. Chem. C, 2019, 123, 24558–24565.
19. R. Mu, A. Dahal, Z.-T. Wang, Z. Dohnálek, G. A. Kimmel, N. G. Petrik and I. Lyubinetsky, J. Phys. Chem. Lett., 2017, 8, 4565–4572.
20. N. G. Petrik, R. Mu, A. Dahal, Z. Wang, I. Lyubinetsky and G. A. Kimmel, J. Phys. Chem. C, 2018, 122, 15382–15389.
21. Z.-T. Wang, N. A. Deskins and I. Lyubinetsky, J. Phys. Chem. Lett., 2012, 3, 102–106.
22. K. R. Phillips, S. C. Jensen, M. Baron, S.-C. Li and C. M. Friend, J. Am. Chem. Soc., 2013, 135, 574–577.
23. Q. Guo, C. Xu, W. Yang, Z. Ren, Z. Ma, D. Dai, T. K. Minton and X. Yang, J. Phys. Chem. C, 2013, 117, 5293–5300.
24. Q. Yuan, Z. Wu, Y. Jin, L. Xu, F. Xiong, Y. Ma and W. Huang, J. Am. Chem. Soc., 2013, 135, 5212–5219.
25. C. Xu, W. Yang, Q. Guo, D. Dai, M. Chen and X. Yang, J. Am. Chem. Soc., 2014, 136, 602–605.
26. Z. Wang, Q. Yuan, Z. Zhang, F. Xiong, G. Sun, H. Xu, P. Chai and W. Huang, J. Phys. Chem. C, 2019, 123, 31073–31081.
27. F. Xiong, L.-L. Yin, Z. Wang, Y. Jin, G. Sun, X.-Q. Gong and W. Huang, J. Phys. Chem. C, 2017, 121, 9991–9999.
28. L. Wu, Z. Wu, Z. Wang, H. Xu, P. Chai and W. Huang, J. Phys. Chem. Lett., 2022, 13, 2614–2618.
29. L. Wu, Z. Wu, F. Li, Z. Wang, Z. Zhang, X.-Q. Gong and W. Huang, J. Phys. Chem. Lett., 2024, 15, 8481–8486.
30. W. Wang, L. Wu, C. Fu, H. Xu, Z. Wu and W. Huang, Chin. J. Chem. Phys., 2024, 37, 614–618.
31. M. Setvin, J. Hulva, G. S. Parkinson, M. Schmid and U. Diebold, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, E2556.
32. M. Setvin, X. Hao, B. Daniel, J. Pavelec, Z. Novotny, G. S. Parkinson, M. Schmid, G. Kresse, C. Franchini and U. Diebold, Angew. Chem., Int. Ed., 2014, 53, 4714–4716.
33. X.-Q. Gong, A. Selloni, M. Batzill and U. Diebold, Nat. Mater., 2006, 5, 665–670.
34. U. Martinez, L. B. Vilhelmsen, H. H. Kristoffersen, J. Stausholm-Møller and B. Hammer, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 84, 205434.
35. U. Martinez, J. Ø. Hansen, E. Lira, H. H. Kristoffersen, P. Huo, R. Bechstein, E. Lægsgaard, F. Besenbacher, B. Hammer and S. Wendt, Phys. Rev. Lett., 2012, 109, 155501.
36. H. H. Kristoffersen, J. Ø. Hansen, U. Martinez, Y. Y. Wei, J. Matthiesen, R. Streber, R. Bechstein, E. Lægsgaard, F. Besenbacher, B. Hammer and S. Wendt, Phys. Rev. Lett., 2013, 110, 146101.
37. F. Polo-Garzon, Z. Bao, X. Zhang, W. Huang and Z. Wu, ACS Catal., 2019, 9, 5692–5707.
38. S. Wendt, P. T. Sprunger, E. Lira, G. K. H. Madsen, Z. Li, J. Ø. Hansen, J. Matthiesen, A. Blekinge-Rasmussen, E. Lægsgaard, B. Hammer and F. Besenbacher, Science, 2008, 320, 1755.
39. E. Lira, S. Wendt, P. Huo, J. O. Hansen, R. Streber, S. Porsgaard, Y. Wei, R. Bechstein, E. Laegsgaard and F. Besenbacher, J. Am. Chem. Soc., 2011, 133, 6529–6532.
40. E. Lira, J. Ø. Hansen, P. Huo, R. Bechstein, P. Galliker, E. Lægsgaard, B. Hammer, S. Wendt and F. Besenbacher, Surf. Sci., 2010, 604, 1945–1960.
41. Z. Zhang, J. Lee, J. T. Yates, R. Bechstein, E. Lira, J. Ø. Hansen, S. Wendt and F. Besenbacher, J. Phys. Chem. C, 2010, 114, 3059–3062.
42. H. Onishi and Y. Iwasawa, Surf. Sci., 1994, 313, L783–L789.
43. M. Blanco-Rey, J. Abad, C. Rogero, J. Mendez, M. F. Lopez, J. A. Martin-Gago and P. L. de Andres, Phys. Rev. Lett., 2006, 96, 055502.
44. M. Li, W. Hebenstreit and U. Diebold, Surf. Sci., 1998, 414, L951–L956.
45. M. Li, W. Hebenstreit, U. Diebold, M. A. Henderson and D. R. Jennison, Faraday Discuss., 1999, 114, 245–258.
46. M. Li, W. Hebenstreit, L. Gross, U. Diebold, M. A. Henderson, D. R. Jennison, P. A. Schultz and M. P. Sears, Surf. Sci., 1999, 437, 173–190.
47. M. Li, W. Hebenstreit and U. Diebold, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 4926–4933.
48. L. Mohrhusen, J. Kräuter, M. Willms and K. Al-Shamery, J. Phys. Chem. C, 2019, 123, 20434–20442.
49. S. Tan, Y. Ji, Y. Zhao, A. Zhao, B. Wang, J. Yang and J. G. Hou, J. Am. Chem. Soc., 2011, 133, 2002–2009.
50. M. A. Henderson, W. S. Epling, C. L. Perkins, C. H. F. Peden and U. Diebold, J. Phys. Chem. B, 1999, 103, 5328–5337.
51. G. A. Kimmel and N. G. Petrik, Phys. Rev. Lett., 2008, 100, 196102.
52. T. L. Thompson, O. Diwald and J. T. Yates, J. Phys. Chem. B, 2003, 107, 11700–11704.
53. X. Lin, Y. Yoon, N. G. Petrik, Z. Li, Z.-T. Wang, V.-A. Glezakou, B. D. Kay, I. Lyubinetsky, G. A. Kimmel, R. Rousseau and Z. Dohnálek, J. Phys. Chem. C, 2012, 116, 26322–26334.
54. Z. Wu, H. Wang, J. Shi, Z. Li and W. Huang, J. Phys. Chem. C, 2024, 128, 21702–21709.
55. M. Kunat, F. Traeger, D. Silber, H. Qiu, Y. Wang, A. C. van Veen, C. Wöll, P. M. Kowalski, B. Meyer, C. Hättig and D. Marx, J. Chem. Phys., 2009, 130, 144703.
56. I. Sokolović, M. Reticcioli, M. Čalkovský, M. Wagner, M. Schmid, C. Franchini, U. Diebold and M. Setvín, Proc. Natl. Acad. Sci. U. S. A., 2020, 117, 14827–14837.
57. G. Lu, A. Linsebigler and J. T. Yates, J. Chem. Phys., 1995, 102, 4657–4662.
58. C. N. Rusu and J. T. Yates, Langmuir, 1997, 13, 4311–4316.
59. C. L. Perkins and M. A. Henderson, J. Phys. Chem. B, 2001, 105, 3856–3863.
60. T. L. Thompson and J. T. Yates, J. Phys. Chem. B, 2006, 110, 7431–7435.
61. D. Sporleder, D. P. Wilson and M. G. White, J. Phys. Chem. C, 2009, 113, 13180–13191.
62. N. G. Petrik and G. A. Kimmel, J. Phys. Chem. Lett., 2010, 1, 1758–1762.
63. Z. Zhang and J. T. Yates, J. Phys. Chem. C, 2010, 114, 3098–3101.
64. N. G. Petrik and G. A. Kimmel, J. Phys. Chem. Lett., 2011, 2, 2790–2796.
65. C. Fu, F. Li, J. Zhang, D. Li, K. Qian, Y. Liu, J. Tang, F. Fan, Q. Zhang, X.-Q. Gong and W. Huang, Angew. Chem., Int. Ed., 2021, 60, 6160–6169.
66. C. Fu, F. Li, J. Yang, J. Xie, Y. Zhang, X. Sun, X. Zheng, Y. Liu, J. Zhu, J. Tang, X.-Q. Gong and W. Huang, ACS Catal., 2022, 12, 6457–6463.
67. X. Sun, X. Chen, C. Fu, Q. Yu, X.-S. Zheng, F. Fang, Y. Liu, J. Zhu, W. Zhang and W. Huang, Nat. Commun., 2022, 13, 6677.
68. C. Fu, F. Li, Z. Wu, F. Xiong, J. Zhu, X.-Q. Gong and W. Huang, J. Phys. Chem. Lett., 2023, 14, 8916–8921.

---