Influence of the molecular adsorption of CO₂, CO and NO on the stability of oxygen vacancies on the anatase TiO₂ (101) surface

Abstract

The impact of the molecular adsorption of CO₂, CO and NO on the stability of oxygen vacancies at the anatase TiO₂ (101) surface is studied through first-principles calculations. Our findings reveal that the adsorption of CO₂, CO, and NO stabilizes the surface oxygen vacancy relative to the subsurface vacancy, with total energies being 0.08 eV, 0.32 eV, and 1.58 eV lower, respectively. This suggests that the adsorption of these molecules can thermodynamically reverse the relative stability between the surface and subsurface oxygen vacancies, with the surface oxygen vacancy surface becoming the most stable. Additionally, we investigate the kinetic effects of oxygen vacancy interactions with small molecules. The diffusion barriers for oxygen vacancies on surfaces with adsorbed CO₂ and CO are found to be 0.68 eV and 0.36 eV, respectively—significantly lower than the diffusion barriers on the clean surface by 0.16 eV and 0.50 eV, respectively. These results suggest that CO adsorption can effectively promote the diffusion of oxygen vacancies. Overall, this study highlights the crucial role of molecular adsorption in modulating the stability and interaction of oxygen vacancies on the anatase TiO₂ (101) surface, providing insights into the photocatalytic activity of this material.

---

Introduction

Titanium dioxide (TiO₂) has been extensively studied due to its promising applications in heterogeneous catalysis, solar cells, gas sensors, photocatalysis, and environmental cleaning.^1–9 Although rutile TiO₂ is the thermodynamically stable phase, the (110) surface has been widely used as a model surface.^10–12 However, the anatase phase exhibits greater stability relative to rutile when nanoparticle dimensions are reduced below 11 nm (ref. 13), while also demonstrating superior surface reactivity.¹⁴ Among anatase surfaces, the (101) surface is the most stable.¹⁵ The interactions between small molecules and the TiO₂ (101) surface have been studied extensively in recent years.^16–19

Surface oxygen vacancies (V^sur_O) are key defects on the TiO₂ (110) surface and play a crucial role in surface reactions.^8,11,20–27 In contrast, studies have shown that subsurface oxygen vacancies (V^sub_O) are more stable than surface vacancies on the anatase TiO₂ (101) surface.^26,27 As a result, there has been limited research on the interaction between small molecules and defective anatase (101) surfaces.^17–19 Notably, Setvin et al. reported that O₂ adsorption could reverse the relative stability of V^sur_O and V^sub_O.¹⁷ Additionally, other studies have shown that water and methanol can facilitate the migration of oxygen vacancies (V_Os) from the subsurface to the surface on the anatase (101) surface.^18,19 We previously studied the interaction between H₂S and oxygen vacancies on the anatase TiO₂ (101) surface and found that H₂S adsorption could reverse the stability of V^sur_O and V^sub_O.²⁸ Upon the adsorption of small molecules such as H₂O, H₂S and methanol, V_Os on the anatase TiO₂ (101) surface undergo a stabilization transition from the subsurface to the surface layer. These V_Os can now function as accessible active sites, as they are no longer obstructed by steric hindrance or electronic screening from the surface molecular adlayer. Consequently, the decomposition of adsorbed small molecules including H₂O, H₂S, and methanol is significantly promoted, leading to an improved surface photocatalytic performance.^18,19,28

Several important reactions occur on the anatase TiO₂ (101) surface, such as the photocatalytic reduction of CO₂ to methanol for fuel production,^16,29 low-temperature CO oxidation and CO hydrogenation reactions,^30–33 and the oxidation of NO to HNO₃ to reduce nitrogen oxide emissions from vehicles and industrial sources.^34,35 However, studies on these reactions typically overlook the stability of oxygen vacancies in the subsurface, and the effect of adsorbed molecules on the stability of oxygen vacancies is not well understood in such works. Additionally, the role of oxygen vacancies in various photocatalytic reactions remains unclear, which limits the practical application of anatase TiO₂ as a catalyst.

In this study, we systematically investigate the impact of molecular adsorption (CO₂ CO, and NO) on the stability of V_Os at the anatase (101) surface. Our results show that after CO₂, CO, and NO adsorption, the surface with V^sur_O becomes more energetically favorable than the surface with V^sub_O. This suggests that the relative stability of the surface with either a V^sub_O or V^sur_O can be reversed, with the surface containing V^sur_O becoming the most stable. We also examine the kinetics of V_O diffusion. We find that the diffusion barriers of V_Os from the subsurface to the surface with CO₂ and CO adsorption are only 0.68 eV and 0.36 eV, which are significantly lower than the clean surface by 0.16 eV and 0.50 eV, respectively. These findings indicate that CO adsorption can significantly enhance the diffusion of V_Os.

Computational methods

The calculations were based on density functional theory (DFT) using the Perdew–Wang 91(PW91) generalized gradient approximation^36,37 and the Vienna ab initio simulation package (VASP) code with projector-augmented wave pseudopotentials.^38,39 An energy cutoff of 500 eV was used for expanding the Kohn–Sham wave functions. The anatase TiO₂ (101) surface was modeled as a (1 × 4) supercell slab. The stoichiometric slab containing three O–Ti–O trilayers (Ti₄₈O₉₆) and a vacuum with a thickness of 20 Å, with a Γ-centered 2 × 2 mesh, has been tested to be well converged. The positions of atoms in the bottom trilayer are fixed to mimic the bulk, and the other atoms were relaxed until the forces converged to 0.01 eV Å⁻¹. To model a reduced anatase TiO₂ (101) surface, an oxygen atom was removed from the slab; the reduced slab contained 48 Ti atoms and 95 O atoms, and the V_O density was 1/4, which was enough for the small molecules to get adsorbed on the surface with V_Os.²⁸ To investigate the reaction kinetics by locating the transition states, we employed the nudged elastic band (NEB) method within the VASP framework.⁴⁰ For each elementary step, the reaction coordinate was represented by five intermediate images constructed through a linear interpolation between the boundary states. Structural relaxations along the reaction pathway were performed using the conjugate gradient scheme, with a force convergence criterion of 0.05 eV Å⁻¹. The reaction barrier was defined as the energy difference between the initial state to the saddle point, identified using the image with the maximum electronic energy along the minimum energy pathway.

The adsorption energy of small molecules on the TiO₂ (101) was calculated using the following equation:⁴¹

E_ads = E_tot − E_sur − E_mol − E_ZPE − PV + TS, (1)

where E_tot is the total energy of a molecule adsorption on the slab, E_sur is the total energy of TiO₂ (101) surface before small molecule adsorption, E_mol is the energy of a molecular on a 15 × 15 × 15 ų vacuum box, E_ZPE is the zero-point energy, V is the volume of the vacuum box, P and T is the relevant temperature and pressure of the box, respectively, and S is the entropy of small molecules. The chemical potentials of small molecules can be referenced to the total energy of the elementary phases at T = 0 K. The formation energies of the defects are always positive or the crystal would be unstable;⁴² positive (negative) energies are intended to be an endothermic (exothermic) reaction.

Results and discussion

Our previous studies showed that the anatase TiO₂ (101) surface system containing a V^sub_O is more stable than that with a V^sur_O as the surface containing V^sur_O is 0.16 eV higher in total energy than the surface containing V^sub_O.²⁸ Upon introducing a V^sub_O into the anatase TiO₂ (101) surface, we considered thirteen potential adsorption sites for small molecules, as shown in Fig. 1a. The CO₂ molecule maintains a linear geometry; in the calculations for all the 13 adsorption sites, it was placed vertically above the surface, and the vertical distance between the lower oxygen atom of the molecule and the surface atom directly beneath the adsorption site was 1.92 Å in each case. After relaxation, the adsorption energy of CO₂ on the anatase TiO₂ (101) surface with a V^sub_O was calculated using eqn (1). As illustrated in Fig. 1b, the most stable adsorption site for CO₂ is the B8 site, with the optimized geometry shown in Fig. 1c and e. The most stable adsorption site corresponded to the next-nearest five-coordinated titanium atom (Ti-5c) site, where the distance between the lower oxygen atom of CO₂ and Ti-5c was 2.59 Å, the angle between CO₂ and the surface was 63.3°, and the adsorption energy at the B8 site was −0.18 eV. Additionally, the adsorption energy of CO₂ at the B3 site was −0.13 eV, where CO₂ got adsorbed on the nearest Ti-5c site. The energy of this site was only 0.05 eV higher than of the B8 site. The optimized geometry for this configuration is shown in Fig. 1d and f; here, the distance between the lower oxygen atom of CO₂ and Ti-5c was 2.47 Å, which was shorter than at the B8 site, and the angle between CO₂ and the surface was 58.9°.

Fig. 1 (a) Top view of a 1 × 4 supercell of the anatase TiO₂ (101) surface with a V^sub_O and thirteen possible adsorption sites for small molecules. (b) Calculated adsorption energies of CO₂ at all the thirteen sites on the V^sub_O-containing surface. (c) Top and (e) side views of the optimized structure of the CO₂-adsorbed B8 site. The black dashed square in (c) outlines the primitive cell of the anatase (101) surface. (d) Top and (f) side views of the optimized structure of the CO₂-adsorbed B3 site. In the structural diagrams, gray, red, cyan, and burgundy spheres represent titanium (Ti), surface lattice oxygen (O), carbon (C), and oxygen atoms (O) from the CO₂ molecule, respectively. The green dashed circles indicate the position of the V^sub_O.

Fig. 2a illustrates the thirteen possible adsorption sites for small molecules on the anatase TiO₂ (101) surface with a V^sur_O. Similarly, for CO₂ adsorption at the 13 adsorption sites on the anatase (101) surface containing a V^sur_O, the initial configurational setup was identical to that used for the surface models with a V^sub_O. When CO₂ was adsorbed on a surface oxygen vacancy, the distance between the bottom oxygen atom of CO₂ and the original oxygen atom at the vacancy was 1.92 Å. The adsorption energies after relaxation are shown in Fig. 2c. Results indicated that the most stable adsorption site for CO₂ was the R2 site, with the optimized geometric structures shown in Fig. 2b and d. The most stable adsorption occurs at the top of the V^sur_O, where the angle between CO₂ and the surface is 77.6°, and the adsorption energy at the R2 site is −0.37 eV.

Fig. 2 (a) Top view of the anatase TiO₂ (101) surface with a V^sur_O, showing the thirteen possible molecule adsorption sites labeled R1–R13. Top view (b) and side view (d) of the optimized atomic structure of the CO₂-adsorbed R2 site. (c) Calculated adsorption energies of CO₂ at the thirteen different sites on the anatase (101) surface with a V^sur_O. In all panels, the V^sur_O sites are represented by black dashed spheres. In the structural diagrams, gray, red, cyan, and burgundy spheres represent titanium (Ti), surface lattice oxygen (O), carbon (C), and oxygen atoms (O) from the CO₂ molecule, respectively. The dashed circles indicate the position of the V^sur_O.

Our previous studies showed that a V^sub_O is more stable than a V^sur_O on a clean anatase TiO₂ (101) surface, with the surface containing a V^sur_O being 0.16 eV higher in total energy than the surface with a V^sub_O. As a result, after CO₂ adsorption on the B3 and B8 sites, the surface with the V^sur_O is 0.08 eV and 0.03 eV lower in total energy than the surface with the V^sub_O.

After CO₂ adsorption, the V_O tends to migrate from the subsurface to the surface. The calculated V_O diffusion pathway and optimized geometrical structures are shown in Fig. 3. Initially, the V_O is located in the subsurface layer, with CO₂ adsorbed on the B3 site (Fig. 3a). The oxygen atom beneath the surface (O-2c) then diffuses to the V^sub_O site (Fig. 3b) and forms a complex (Fig. 3c). Subsequently, the surface O-2c atom diffuses to the site (Fig. 3d), forming a V^sur_O with CO₂ adsorbed at the R2 site (Fig. 3e). The energy profiles are shown in Fig. 3f. Our calculations showed that the diffusion pathway of the V_O from the subsurface to the surface was similar to that of the clean surface.²⁸ The diffusion barrier of V_O on the CO₂-adsorbed surface was 0.68 eV, which was 0.18 eV lower than on the clean surface, suggesting that CO₂ adsorption promotes V_O diffusion.

Fig. 3 Selected optimized structures (a–e) and energy profile (f) during V_O diffusion from the subsurface to the surface layer with pre-adsorbed CO₂. In the structural diagrams, gray, red, cyan, and burgundy spheres represent titanium (Ti), surface lattice oxygen (O), carbon (C), and oxygen atoms (O) from the CO₂ molecule, respectively. The black sphere represents the specific oxygen atom undergoing migration. The green circles indicate the V_O sites involved in the diffusion; in panels (b)–(d), the circles are shown to highlight the migration path. All models maintain the same stoichiometry, containing only one V_O throughout the reaction.

Next, we examined the adsorption energy of CO on the anatase TiO₂ (101) surface with a V^sub_O. Thirteen possible adsorption sites and two adsorption types for CO were considered. In type 1, the CO molecule was oriented vertically with the C atom closest to the surface (Fig. 4a and c), and in type 2, the CO molecule is oriented vertically with the O atom closest to the surface (Fig. 4b). The adsorption energies of CO on the anatase TiO₂ (101) surface with a V^sub_O are shown in Fig. 4d. Results indicated that CO preferred to adsorb via type 1 configuration, where the C atom was closest to the surface, as this configuration had a lower adsorption energy than type 2 configuration. The most stable adsorption site for CO was the B3 site (type 1), and the optimized geometric structure is shown in Fig. 4a and c. The distance between CO and Ti-5c was 2.38 Å, the angle between CO and the surface was 80.1°, and the adsorption energy was −0.25 eV. In contrast, for type 2, the distance between the O atom and Ti-5c was 2.85 Å, the angle between the CO axis and the surface was 86.5°, and the adsorption energy was −0.08 eV. Furthermore, as illustrated in Fig. 4d, the adsorption energy of type 1 is consistently lower than that of type 2 at almost all equivalent adsorption sites. This thermodynamic trend clearly indicated that the CO molecule preferentially adsorbed via type 1 mode, making it the most stable configuration for this system.

Fig. 4 (a) Top and (c) side views of the optimized atomic structure of the CO-adsorbed B3 site on the anatase TiO₂ (101) surface with a V^sub_O in the type 1 configuration. (b) Top view of the corresponding structure in the type 2 configuration. (d) Adsorption energies of CO on the anatase (101) surface with a V^sub_O. The absence of data points at certain sites in (d) indicates that CO placed at these positions is unstable and spontaneously migrates to neighboring stable sites during the relaxation process. In the structural diagrams, gray, red, cyan, and burgundy spheres represent titanium (Ti), surface lattice oxygen (O), carbon (C), and oxygen atoms (O) from the CO molecule, respectively. The green dashed circles indicate the position of the V^sub_O.

We also investigated the adsorption energy of CO on the anatase TiO₂ (101) surface with a V^sur_O. Thirteen potential adsorption sites (Fig. 2a) and two adsorption types for CO were considered. Similar to the surface with a V^sur_O, in type 1, the CO molecule was oriented vertically with the C atom closest to the surface (Fig. 5a and c), and in type 2, the CO molecule was oriented vertically with the O atom closest to the surface (Fig. 5b). The adsorption energies of CO on the anatase TiO₂ (101) surface with a V^sur_O are shown in Fig. 5d. It can be seen that, except for the R7 and R12 sites, the adsorption energies of CO in type 1 configuration were lower than those in type 2 configuration. The most stable adsorption site for CO was the R2 site in type 1 configuration, with the optimized geometry shown in Fig. 5a and c. The most stable adsorption occurred at the V^sur_O site, where the angle between CO and the surface was 57.8° and the adsorption energy was −0.75 eV, which was significantly larger than the energy for the same site in type 2 configuration.

Fig. 5 (a) Top and (c) side views of the optimized atomic structure for the CO-adsorbed R2 site on the anatase TiO₂ (101) surface with a V^sur_O in the type 1 configuration. (b) Top view of the corresponding structure in the type 2 configuration. (d) Adsorption energies of CO on the anatase (101) surface with a V^sur_O for the thirteen different adsorption sites. V^sur_O is represented by black dashed spheres. The absence of data points at certain sites in (d) indicates that CO placed at these positions is unstable and spontaneously migrates to neighboring stable sites during the relaxation process. In the structural diagrams, gray, red, cyan, and burgundy spheres represent titanium (Ti), surface lattice oxygen (O), carbon (C), and oxygen atoms (O) from the CO molecule, respectively. The dashed circles indicate the position of the V^sur_O.

Similar to CO₂, the surface with a V^sur_O was 0.32 eV lower in total energy than the surface with a V^sub_O after CO adsorption. This indicates that CO adsorption can thermodynamically reverse the relative stability of the surface with either a V^sub_O or V^sur_O. After CO adsorption on the anatase TiO₂ (101) surface with a V^sub_O, the relative stability of the configuration with the V^sub_O decreased, and the surface with the V^sub_O became the most stable.

The diffusion of the V_O from the subsurface to the surface with CO adsorption is shown in Fig. 6. The diffusion pathway of the V_O was similar to that observed with CO₂ adsorption. However, the diffusion barrier of the V_O on the surface with CO adsorption was only 0.36 eV, which was significantly lower than that on the clean surface by 0.50 eV. These results suggested that CO adsorption effectively promoted V_O diffusion from the subsurface to the surface layer.

Fig. 6 Selected optimized structures (a–e) and energy profile (f) during the V_O diffusion from the subsurface to the surface layer after CO adsorption. In the structural diagrams, gray and red spheres represent titanium (Ti) and surface lattice oxygen (O) atoms and the cyan and burgundy spheres represent carbon (C) and oxygen (O) atoms from the CO molecule, respectively. The black sphere represents the specific oxygen atom undergoing migration; the green circles indicate the V_O sites involved in the diffusion. In panels (b)–(d), the circles are shown to highlight the migration path. All models have the same stoichiometry, containing only one V_O throughout the process.

Next, we studied the adsorption energy of NO on the anatase TiO₂ (101) surface with a V^sub_O. Similar to CO, thirteen possible adsorption sites and two adsorption types for NO were considered. In type 1, the N atom of NO was the closest to the surface (Fig. 7a and c), while in type 2, the O atom of NO was the closest to the surface (Fig. 7b). The adsorption energy of NO on the anatase TiO₂ (101) surface with a V^sub_O is shown in Fig. 7d. Results indicated that the B3 site was the most stable adsorption site for NO, with an adsorption energy of −0.86 eV, which was much lower than the adsorption energies of CO₂ and CO at the same site by 0.68 eV and 0.61 eV, respectively. The B3 site was located on the nearest Ti-5c site, where the distance between the N atom of NO and Ti-5c was 1.86 Å, and the angle between NO and the surface was 70.8°.

Fig. 7 (a) Top and (c) side views of the optimized atomic structure for the NO-adsorbed B3 site on the anatase TiO₂ (101) surface with a V^sub_O in the type 1 configuration. (b) Top view of the corresponding structure in the type 2 configuration. (d) Adsorption energies of NO on the anatase (101) surface with a V^sub_O. In the structural diagrams, gray, red, orange and burgundy spheres represent titanium (Ti), surface lattice oxygen (O), nitrogen (N), and oxygen atoms (O) from the NO molecule, respectively. The green dashed circles indicate the position of the V^sub_O.

We also studied the adsorption energy of NO on the anatase TiO₂ (101) surface with a V^sur_O. The adsorption sites and the two identified configurations for NO are analogous to those observed for CO. Type 1 and type 2 configurations on the R2 sites are shown in Fig. 8(a–c). The adsorption energy of NO on the anatase TiO₂ (101) surface with a V^sur_O is shown in Fig. 8d. Results indicated that the B3 site was the most stable adsorption site for NO, with an adsorption energy of −2.60 eV, which was significantly lower than the adsorption energies of CO₂ and CO by 2.23 eV and 1.85 eV, respectively. The N atom of NO was positioned on the V^sur_O site, and the angle between NO and the surface was 24.4°, indicating that NO is positioned almost parallel to the surface.

Fig. 8 (a) Top and (c) side views of the optimized atomic structure of the NO-adsorbed R2 site on the anatase TiO₂ (101) surface with a V^sur_O in the type 1 configuration. (b) Top view of the corresponding structure in the type 2 configuration. (d) Adsorption energies of NO on the anatase (101) surface with a V^sur_O for thirteen different adsorption sites. In the structural diagrams, gray, red, orange and burgundy spheres represent titanium (Ti), surface lattice oxygen (O), nitrogen (N), and oxygen atoms (O) from the NO molecule, respectively. The green dashed circles indicate the position of the V^sur_O.

After comparing the total energy following NO adsorption, we found that the surface with the V^sur_O is 1.58 eV lower in energy than the surface with the V^sub_O. This suggested that the relative stability between the surfaces with the V^sub_O and V^sur_O can be reversed through NO adsorption.

To understand the adsorption characteristics of different small molecules on the anatase TiO₂ (101) surface, we calculated the differential charge density for the stable adsorption configurations of CO₂, CO, and NO on the anatase TiO₂ (101) surface with a V^sub_O or V^sur_O, as shown in Fig. 9. When CO₂ was adsorbed on the surface with a V^sub_O (Fig. 9a), the charge density around the oxygen and carbon atoms of CO₂ decreased, while the charge density between the lowest oxygen atom of CO₂ and the surface increased. However, the amount of charge transfer was very small, indicating the weak adsorption of CO₂ on this surface. Conversely, when CO₂ was adsorbed on the surface with a V^sur_O (Fig. 9d), the increase in charge density between the lowest oxygen atom of CO₂ and the surface was greater than that observed on the V^sub_O surface, suggesting the stronger adsorption of CO₂ on the surface with a V^sur_O. When CO was adsorbed on the surface with a V^sub_O (Fig. 9b), similar to the case of CO₂ on this surface, less charge transfer occurred. However, when CO was adsorbed on a surface with a V^sur_O (Fig. 9e), the charge density around the surface atoms near the V^sur_O decreased, and a significant amount of charge accumulated between CO and the surface, indicating the strong bonding between CO and this surface. When NO was adsorbed on the surface with a V^sub_O (Fig. 9c), a high charge density from the vicinity of the surface atoms accumulated between NO and the surface. The amount of charge transfer was greater than that observed for CO₂ and CO adsorption on this surface, and its adsorption energy was consequently larger. Furthermore, when NO was adsorbed on the surface with a V^sur_O (Fig. 9f), the angle between NO and the surface was 24.4°, indicating that NO was positioned almost parallel to the surface. A large amount of charge was accumulated not only between the N atom of NO and the surface but also between the O atom of NO and the surface, leading to a very strong bond between NO and the surface. The close agreement between the charge analysis and adsorption energy results confirmed the robustness of our findings.

Fig. 9 Side view of the charge density difference for CO₂ (a and d), CO (b and e), and NO (c and f) molecules adsorbed on the anatase TiO₂ (101) surface with a V^sub_O (upper) or V^sur_O (lower). The iso-surfaces represent charge accumulation (yellow) and charge depletion (cyan), providing insights into the bonding mechanisms.

Conclusions

We investigated the adsorption sites and energies of CO₂, CO, and NO molecules upon their adsorption on the anatase (101) surface with a V^sub_O or V^sur_O. On the surface with a V^sub_O, the most stable adsorption sites and corresponding energies are B8 (−0.18 eV), B3 (−0.25 eV), and B3 (−0.86 eV) for CO₂, CO, and NO, respectively. On the surface with V^sur_O, the most stable sites and energies are R2 (−0.37 eV), R3 (−0.75 eV), and R3 (−2.60 eV) for CO₂, CO, and NO, respectively. These charge analysis results closely match the adsorption energy findings, which validates the scientific soundness of our work. Importantly, we found that the surface with a V^sur_O is 0.03 eV, 0.32 eV, and 1.58 eV lower in total energy than the surface with a V^sub_O after CO₂, CO, and NO adsorption, respectively. This indicates that small molecule adsorption can thermodynamically reverse the relative stability of the surface with either a V^sub_O or V^sur_O. Additionally, we observed that the diffusion barriers of V_Os on the surface with CO₂ and CO adsorption are only 0.68 eV and 0.36 eV, respectively, which are much lower than those on the clean surface by 0.16 eV and 0.50 eV, respectively. These results suggest that CO adsorption effectively promotes V_O diffusion.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

The data supporting the findings of this study are available within the article. Additional datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Acknowledgements

The calculations were carried out at the High-Performance Computing Center of College of Physics and Electronic Engineering, Zhengzhou Normal University. This work was supported by the National Natural Science Foundation of China (Grant No. 12204430), the Natural Science Foundation of Henan (Grant No. 222300420378), the Technologies Research and Development Program of Henan Province (Grant No. 262102231043, 262102231046), the Key Scientific Research Project of Henan College (Grant No. 26A140016), the Innovative Research Team (in Science and Technology) in University of Henan Province (Grant No. 25IRTSTHN015), the Innovation and Entrepreneurship Training Program for college students in Henan Province (Grant No. 202312949014), and the Scientific Research and Innovation Fund for college students in Zhengzhou Normal University (Grant No. 2024009).

References

1. H. Wang, H. Qi, H. Jia, S. Li and X. Li, High quantum efficiency of hydrogen production from methanol aqueous solution with PtCu-TiO₂ photocatalysts, Nat. Mater., 2023, 22, 619.
2. O. Aldosari and U. Obaid, Unlocking the potential of TiO₂-based photocatalysts for green hydrogen energy through water-splitting: Recent advances, future perspectives and techno feasibility assessment, Int. J. Hydrogen Energy, 2024, 59, 958–981.
3. W. Yuan, B. Zhu, K. Fang, X. Y. Li, J. Yang, T. S. Hansen, Y. Ou, H. Yang, J. B. Wagner, Y. Gao, Y. Wang and Z. Zhang, In situ manipulation of the active Au-TiO₂ interface with atomic precision during CO oxidation, Science, 2021, 371, 517.
4. Z. Jiang, X. Xu, Y. Ma, H. S. Cho, D. Ding, C. Wang, J. Wu, P. Oleynikov, M. Jia and J. Cheng, et al., Filling Metal-Organic Framework Mesopores with TiO₂ for CO₂ Photoreduction, Nature, 2020, 586, 549.
5. F. Xu, K. Meng, B. Cheng, S. Wang, J. Xu and J. S. Yu, Unique, Scheme Hetero Junctions in Self-Assembled TiO₂/CsPbBr₃ Hybrids for CO₂ Photoreduction, Nat. Commun., 2020, 11, 4613.
6. W. Li, A. Elzatahry, D. Aldhayan and D. Zhao, Core-Shell Structured Titanium Dioxide Nanomaterials for Solar Energy Utilization, Chem. Soc. Rev., 2018, 47, 8203–8237.
7. E. J. W. Crossland, N. Noel, V. Sivaram, T. Leijtens, J. A. Alexander-Webber and H. J. Snaith, Mesoporous TiO₂ Single Crystals Delivering Enhanced Mobility and Optoelectronic Device Performance, Nature, 2013, 495, 215–219.
8. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Visible-Light Photocatalysis in Nitrogen-doped Titanium Oxides, Science, 2001, 293, 269–271.
9. M. Gratzel, Photoelectrochemical Cells, Nature, 2001, 414, 338–344.
10. S. Selcuk and A. Selloni, Facet-Dependent Trapping and Dynamics of Excess Electrons at Anatase TiO₂ Surfaces and Aqueous Interfaces, Nat. Mater., 2016, 15, 1107–1112.
11. U. Diebold, The Surface Science of Titanium Dioxide, Surf. Sci. Rep., 2003, 48, 53–229.
12. M. A. Henderson, The Interaction of Water with Solid Surfaces: Fundamental Aspects Revisited, Surf. Sci. Rep., 2002, 46, 1–308.
13. H. Zhang and J. F. Banfield, Understanding Polymorphic Phase Transformation Behavior during Growth of Nanocrystalline Aggregates: Insights from TiO₂, J. Phys. Chem. B, 2000, 104, 3481–3487.
14. L. Kavan, M. Grätzel, S. E. Gilbert, C. Klemenz and H. J. Scheel, Electrochemical and Photoelectrochemical Investigation of Single-Crystal Anatase, J. Am. Chem. Soc., 1996, 118, 6716–6723.
15. M. Lazzeri, A. Vittadini and A. Selloni, Structure and Energetics of Stoichiometric TiO₂ Anatase Surfaces, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 63, 155409.
16. J.-Y. Liu, X.-Q. Gong, R. Li, H. Shi, S. B. Cronin and A. N. Alexandrova, (Photo)Electrocatalytic CO₂ Reduction at the Defective Anatase TiO₂ (101) Surface, ACS Catal., 2020, 10, 4048–4058.
17. M. Setvin, U. Aschauer, P. Scheiber, Y. Li, M. Hou, W. Schmid, Y. Selloni and U. Diebold, Reaction of O₂ with Subsurface Oxygen Vacancies on TiO₂ Anatase (101), Science, 2013, 341, 988–991.
18. Y. Li and Y. Gao, Interplay between Water and TiO₂ Anatase (101) Surface with Subsurface Oxygen Vacancy, Phys. Rev. Lett., 2014, 112, 206101.
19. X. Lang, Y. Liang, L. Sun, S. Zhou and W.-M. Lau, Interplay between Methanol and Anatase TiO₂(101) Surface: The Effect of Subsurface Oxygen Vacancy, J. Phys. Chem. C, 2017, 121, 6072–6080.
20. S. Wendt, R. Schaub, J. Matthiesen, E. Vestergaard, E. Wahlstrom, M. Rasmussen, P. Thostrup, L. Molina, E. Lagsgaard and I. Stensgaard, et al., Oxygen Vacancies on TiO₂(110) and Their Interaction with H₂O and O₂: A Combined High-Resolution STM and DFT Study, Surf. Sci., 2005, 598, 226–245.
21. D. J. Shu, S. T. Ge, M. Wang and N. B. Ming, Interplay between External Strain and Oxygen Vacancies on A Rutile TiO₂(110) Surface, Phys. Rev. Lett., 2008, 101, 116102.
22. J. Zheng, Y. Lyu, R. Wang, C. Xie, H. Zhou, S. P. Jiang and S. Wang, Crystalline TiO₂ Protective Layer with Graded Oxygen Defects for Efficient and Stable Silicon-Based, Nat. Commun., 2018, 9, 3572.
23. S. M. Wu, X. L. Liu, X. L. Lian, G. Tian, C. Janiak, Y. X. Zhang, Y. Z. Lu, H. Z. Yu, J. Hu and H. Wei, et al., Homojunction of Oxygen and Titanium Vacancies and its Interfacial n-p Effect, Adv. Mater., 2018, 30, 1802173.
24. Z. W. Wang and D. J. Shu, Intrinsic Interaction Between In-Plane Ferroelectric Polarization and Surface Adsorption, Phys. Chem. Chem. Phys., 2019, 21, 18680–18685.
25. Z. W. Wang, W. G. Chen, D. Teng, J. Zhang, A. M. Li, Z. H. Li and Y. N. Tang, The Effect of Strain on Water Dissociation on Reduced Rutile TiO₂(110) Surface, RSC Adv., 2021, 11, 8485–8490.
26. Y. He, O. Dulub, H. Cheng, A. Selloni and U. Diebold, Evidence for the Predominance of Subsurface Defects on Reduced Anatase TiO₂(101), Phys. Rev. Lett., 2009, 102, 106105.
27. P. Scheiber, M. Fidler, O. Dulub, M. Schmid, U. Diebold, W. Hou, U. Aschauer and A. Selloni, (Sub)Surface Mobility of Oxygen Vacancies at the TiO₂ Anatase (101) Surface, Phys. Rev. Lett., 2012, 109, 136103.
28. Z. Wang, W. Chen, D. Teng, J. Zhang, A. Li, Z. Li and Y. Tang, Interplay between H₂S and anatase TiO₂(101) surface: The effect of subsurface oxygen vacancy, J. Phys. Chem. C, 2022, 126, 3939–3948.
29. J. Liu, X. Gong and A. N. Alexandrova, Mechanism of CO₂ Photocatalytic Reduction to Methane and Methanol on Defected Anatase TiO₂ (101): A Density Functional Theory Study, J. Phys. Chem. C, 2019, 123, 3505–3511.
30. J. Shi, F. Yang, X. Zhao, X. Ren, Y. Tang and S. Li, Spin-polarized p-block antimony/bismuth single-atom catalysts on defect-free rutile TiO₂(110) substrate for highly efficient CO oxidation, Phys. Chem. Chem. Phys., 2024, 26, 16459.
31. M. Setvin, M. Buchholz, W. Hou and C. Zhang, A Multitechnique Study of CO Adsorption on the TiO₂ Anatase (101) Surface, J. Phys. Chem. C, 2015, 119(36), 21044–21052.
32. K. M. Bulanin, A. V. Emeline, R. V. Mikhaylov, A. Y. Mikheleva, A. V. Rudakova and V. K. Ryabchuk, Dynamics of Adsorbed CO Molecules on the TiO₂ Surface under UV Irradiation, J. Phys. Chem. C, 2023, 127, 11005–11013.
33. C. Fu, H. Xu, L. Wu, Z. Zhang and W. Huang, X-ray-Induced CO₂ Formation via CO Reaction with TiO₂ at Cryogenic Temperature, J. Phys. Chem. Lett., 2021, 12, 9741–9747.
34. Z. Wang, G. Lv, C. Zhang and X. Sun, Catalytic Oxidation Mechanism of NO to HNO₃ on TiO₂ (101) and (001) Surfaces and the Influence Factors on NO Removal: A DFT Study, J. Environ. Chem. Eng., 2021, 9, 104643.
35. L. Mino, M. Cazzaniga, F. Moriggi and M. Ceotto, Elucidating NOx Surface Chemistry at the Anatase (101) Surface in TiO₂ Nanoparticles, J. Phys. Chem. C, 2023, 127, 437–449.
36. J. P. Perdew, K. Burke and M. Ernzerhof, Generalized Gradient Approximation Made Simple, Phys. Rev. Lett., 1996, 77, 3865–3868.
37. F. Allegretti, S. O'Brien, M. Polcik, D. I. Sayago and D. P. Woodruff, Adsorption Bond Length for H₂O on TiO₂(110): A Key Parameter for Theoretical Understanding, Phys. Rev. Lett., 2005, 95, 226104.
38. G. Kresse and D. Joubert, From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758–1775.
39. P. E. Blöchl, Projector Augmented-Wave Method, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
40. G. Mills, H. Jonsson and G. K. Schenter, Reversible Work Transition State Theory: Application to Dissociative Adsorption of Hydrogen, Surf. Sci., 1995, 324, 305–337.
41. V. Wang, N. Xu, J.-C. Liu, G. Tang and W.-T. Geng, Comput. Phys. Commun., 2021, 267, 108033.
42. C. Freysoldt, B. Grabowski, T. Hickel, J. Neugebauer, G. Kresse, A. Janotti and C. G. Van de Walle, Rev. Mod. Phys., 2014, 86, 253–305.

---