Introducing Ti³⁺ defects based on lattice distortion for enhanced visible light photoreactivity in TiO₂ microspheres

Abstract

Defective titanium dioxide (TiO₂) is of much significance due to its improved visible light photoreactivity. Generally, the existence of defects leads to imperfections in the crystal lattice, which in turn affect the dynamics of the evolution of defects and the corresponding physical properties of TiO₂. Until now, how lattice distortion affects the formation of Ti³⁺ defects as well as the corresponding visible light photoreactivity in TiO₂ has remained elusive. Herein, we have successfully introduced Ti³⁺ defects based on lattice distortion in TiO₂ microspheres and found the photocurrent of anatase TiO₂ has been significantly enhanced from 1.78 to 80 μA cm⁻² with an increase in photocatalytic activity of almost three times under visible light irradiation. Furthermore, we show that lattice distortions have minimal contribution to enhancing the visible light photocatalytic activity because the band gap cannot be narrowed due to the absence of Ti³⁺ defects, yet the existence of lattice distortions could suppress the recombination of electron–hole pairs. Moreover, the formation of Ti³⁺ defects is energetically favored in lattice-distorted TiO₂ compared to that of pristine TiO₂. This work highlights the design and development of highly efficient TiO₂ photocatalysts that operate under visible light irradiation.

---

Introduction

Titanium dioxide (TiO₂) has attracted much attention in organic contaminant degradation since Carey et al.¹ successfully degraded polychlorinated biphenyl using TiO₂-based photocatalysts. Since then, enormous effort has been devoted to photocatalyst-based research, photovoltaics, and photo-electrochemical cells. However, the applications of TiO₂ are greatly hindered by its low quantum efficiency in photocatalytic reactions, ineffective utilization of visible light due to the high recombination of photo-induced electron–hole pairs, and relatively large band gap (∼3.2 eV).^2–4 To address these issues, researchers have used various strategies, such as fabricating composites,^5–9 metal anion/cation or non-metal doping,^10–15 surface sensitization by dyes,^16–18 introducing defects^19–28 and so on.

Among the above strategies, the introduction of defects, such as Ti³⁺, oxygen vacancies (OVs), and lattice disorder defects, has been widely studied due to their close relationship with the electronic structure, charge transport, and surface activity of TiO₂.^19–28 Chen et al.²⁵ reported black TiO₂ nanoparticles with significant lattice disorder and Ti³⁺ defects, exhibiting excellent solar-driven photocatalytic activities towards hydrogen generation. Grabstanowicz et al.²⁷ have demonstrated that Ti³⁺-doped TiO₂, accompanied with the formation of OVs, displays a broadened absorbance in the visible light region. These imperfections in the crystal lattice lead to lattice distortion, which further affect the corresponding photochemical properties.^29–31 Still, our understanding with regards to (1) how lattice distortions affect the resulting visible light photoreactivity and (2) the formation of defects remains unclear.

Herein, we have successfully introduced defects based on lattice distortion in TiO₂ microspheres. It is demonstrated that the enhancement of visible-light-driven photocatalysis is mainly attributed to the existence of Ti³⁺ defects, which leads to an elevated valence band edge and a narrower band gap, resulting in higher photoreactivity. The band gap of TiO₂ with lattice distortions could not be narrowed, failing to improve visible light photocatalytic activity without Ti³⁺ defects, but the higher photocurrent generation indicates the migration of recombination of electron–holes pairs. Interestingly, lattice-distorted TiO₂ favours the formation of Ti³⁺ defects. A combined effect of lattice distortion and defects was found to increase the photocatalytic activity nearly three times, and the photocurrent generation increased from 1.78 to 80 μA cm⁻². Such progress is significant for the development of highly efficient photocatalysts and devices for solar fuel generation.

Experimental

Synthesis of TiO₂-based materials

All chemicals were of analytical grade and used without further purification. In order to introduce lattice distortions to TiO₂ microspheres, 0.05 mol Ti(SO₄)₂ and 0.05 mol NaCl were added to 60 mL of deionized water. After 30 min of stirring, the transparent solution was transferred into a high-pressure reactor and heated at 120 °C for 12 h with a heating rate of 5 °C min⁻¹. Then, the resulting mass was cleaned, dried at 100 °C for 12 h, and calcined at 300 °C for 5 h, yielding a white powder. To prepare pure TiO₂, no NaCl was added.

In a typical procedure of introducing defects, 10 mL ethylene glycol (EG) and 0.3 g of as-prepared TiO₂ were added into a high-pressure reactor and were heated at 180 °C for 5 h with a heating rate of 5 °C min⁻¹. The upper transparent solution was discarded and the solid mass was washed and dried at 100 °C for 12 h. The TG test (Fig. S7†) shows that the EG reduction does not leave organic remains in as-prepared samples.

TiO₂ with both lattice distortion and defects was prepared by a combination of experimental procedures described above.

Characterization

X-ray diffraction (XRD) patterns were collected with a PANalytical Empyrean X-ray diffractometer with monochromatic Cu K_α radiation (λ = 1.5418 Å). X-ray photoelectron spectroscopy (XPS) was carried out on a PHI5300 (PerkinElmer) with a monochromatic Mg Kα source to analyze the surface chemistry. Scanning electron microscopy (SEM) images were carried out on a field emission scanning electron microscope (JEOL JSM-7001F; acceleration voltage = 10 kV). Transmission electron microscopy (TEM) images were obtained on a Zeiss LIBRA 200 FEG transmission electron microscope operating at 200 kV. UV-vis diffuse reflectance spectroscopy (DRS) was performed on a Shimadzu UV-2100 spectrophotometer using BaSO₄ as the reference.

Photoelectrochemical characterization

The photoelectrochemical response was measured using a CHI 660B electrochemical workstation with conventional three-electrode setup under visible-light illumination. To fabricate the working electrode, 1 mg of as-prepared pure TiO₂ and chlorine introduced TiO₂ microspheres were coated onto a slice of ITO glass with an area of 1.5 × 1.5 cm². A 0.5 × 0.5 cm² Pt plate and a saturated calomel electrode were used as the counter reference electrodes, respectively. A 0.1 M Na₂SO₄ solution was used as the electrolyte and a 500 W Xe lamp was utilized as the visible light source.

Photocatalytic assessment

The photocatalytic activity of the as-prepared TiO₂ microspheres was evaluated by photocatalytic degradation of a solution containing 20 mg L⁻¹ rhodamine B (RhB) or 15 mg L⁻¹ phenol irradiated with visible light (500 W Xe lamp and a UV cutoff filter (λ > 420 nm)). In a typical process, 70 mg of the as-prepared sample was added to 100 mL of RhB or phenol. Afterwards, the photocatalyst was dispersed in the solution and stirred for 0.5 h in the dark to reach adsorption equilibrium before being exposed to visible-light irradiation. The suspension was sampled every 20 min, and the photocatalyst was removed by centrifugation. The change in RhB or phenol concentration was determined relative to the extinction spectra for RhB by UV-vis spectroscopy.

Results and discussion

XRD patterns (Fig. 1a) show that all samples are composed of the pure anatase phase, which suggests that there was no significant change in the phase structure by the addition of chloride ions during the hydrothermal processing or reduction with ethylene glycol (EG). Nevertheless, it is noted that the (101) and (004) diffraction peaks slightly shift to smaller angles after adding chloride ions, while no obvious shift is observed after EG reduction, implying that the lattice parameters of TiO₂ become larger and lattice distortion is introduced by adding chloride ions. It is further confirmed that the introduction of chlorine ions during the synthetic procedures brings about surface stress in the titania crystals, which results in lattice distortion. All samples are mainly composed of microspheres with average diameters of about 350 nm (Fig. 1b). Our results suggest that reduction with EG has no effect on the morphology, however, the existence of Cl ions during hydrothermal processing has a strong effect on the morphology of TiO₂ (Fig. S1†). The presence of chloride ions leads to more hollow structures in the samples, indicating that the existence of chloride ions may play the role as the surface surfactant^4,32,37 (Fig. S2†). TEM analysis further demonstrates that the hollow microspheres are constructed of numerous nanostructures (Fig. 1c). The lattice in area I matches the (101) crystallographic plane of anatase with d spacing of 0.36 nm, indicating that the exposed facets on the surface hold no significant changes (Fig. 1c). Nonetheless, evident lattice distortion exists in the TiO₂ with the addition of chloride ions, as shown in Fig. 1c and S3,† which is consistent with the XRD results.

Fig. 1 (a) XRD patterns of the as-prepared TiO₂ samples (pristine TiO₂ (Original), pure TiO₂ with lattice with Ti³⁺ defects (Defects), pure TiO₂ with lattice distortion (Distortion), TiO₂ with both lattice distortion and defects (Distortion & Defects)). Patterns specially displaying the (101) and (004) peaks are shown to the right. (b) SEM images of the as-prepared TiO₂ microspheres with both lattice distortion and defects. (c) HRTEM image of TiO₂ with lattice distortion (inset: Fast Fourier Transformation (FFT) for area I) with enlarged HRTEM images of selected areas (I) and (II).

XPS analysis was performed to identify the existence of chloride ions and defects on the surface of the different TiO₂ samples. In all samples, there was no signal for chlorine (Fig. S4†), suggesting that the existence of chloride ions in the hydrothermal processing only leads to lattice distortion in-stead of doping and can be removed with subsequent sample washing. Fig. 2 shows the Ti 2p peaks, where the 2p spin–orbital doublets two peaks center at ∼458.8 eV and ∼464.6 eV correspond to the characteristic Ti 2p_3/2 and Ti 2p_1/2 peaks of Ti⁴⁺, respectively.^33,34 A detectable shoulder was observed at lower binding energies for the samples that contain defects, which can be ascribed to Ti³⁺ ions.^35,36 More importantly, the peak area ratio of Ti³⁺/Ti⁴⁺ increases from 0.256 : 1 to 0.488 : 1 when lattice distortion is also present in TiO₂. This increase in the Ti³⁺ concentration can be ascribed to the presence of lattice distortion, which allows for more accessible sites for Ti⁴⁺ to be reduced to Ti³⁺.

Fig. 2 Ti 2p XPS spectrum of as-prepared TiO₂.

The photocatalytic activity of the different TiO₂ samples was evaluated by the degradation of rhodamine B (RhB) dye as well as phenol (Fig. S8†) under visible light irradiation (λ > 420 nm) (Fig. 3a). Only 37.8% RhB is degraded after 120 min by the original TiO₂. When lattice distortion is present in the TiO₂ microspheres, the degradation rate of RhB slightly decreases, indicating that lattice distortion due to the introduction of chloride ions during the preparation of TiO₂ microspheres fails to enhance the photocatalytic activity of TiO₂. After introducing Ti³⁺ defects, 47.9% RhB is degraded after 120 min using the same reaction conditions, whereas TiO₂ with both distortion and defects achieved a degradation of more than 75%. The degradation rate increases dramatically from 38 × 10⁻⁴ to 95 × 10⁻⁴ s⁻¹ from TiO₂ with just defects to TiO₂ with both distortion and defects, displaying that the activity is nearly three times higher when both lattice distortion and defects exist in TiO₂ microspheres (Fig. S5†). The transient photocurrent response spectra (Fig. 3b) reveal that lattice distortion elevates the photocurrent from 1.78 to 15.56 μA cm⁻² in the TiO₂ microspheres, while Ti³⁺ defects elevate the photocurrent density to 53.78 μA cm⁻². Even though the lattice distortion has no contribution on the improvement of visible light photocatalytic activity, the increase of photocurrent suggests that lattice distortion in TiO₂ could efficiently supress the recombination of electron–hole pairs.³⁸ The existence of both lattice distortion and Ti³⁺ defects elevates the photocurrent density to 80 μA cm⁻². The larger the photocurrent value, the higher the charge separation efficiency. Thus, it can be concluded that the introduction of Ti³⁺ defects facilitates the separation of photoelectrons and holes, which is in accordance with the study of Cai et al.²⁴ Considering the effect of surface area, as listed in Table S1,† the TiO₂ with both defects and lattice distortion possess lower surface areas than the original TiO₂, implying that the surface area is not the major effect on the improvement of visible light photoreactivity. It is thereby deduced that the enhancement in the photocurrent generation can be mainly attributed to the existence of Ti³⁺ defects. The introduction of lattice distortion could not only mitigate the recombination of photo-induced electron–hole pairs,^39,45 but was also found to be advantageous for the generation of Ti³⁺ defects. These results demonstrate that the Ti³⁺ defects play a dominant role in improving the photoreactivity of TiO₂ rather than lattice distortion, and a combined effect of lattice distortion and Ti³⁺ defects is beneficial for the enhancement of visible-light-driven photocatalytic activity and photocurrent generation.

Fig. 3 (a) The degradation of RhB and (b) photocurrent response of the different TiO₂ samples under visible light irradiation.

A schematic is presented in Fig. 4 to illustrate the contribution of surficial lattice distortion to the formation of Ti³⁺ defects. Previous work reported that the effect of adding halogen ions during the sample preparation could tune the surface morphology due to the higher electronegativity of the halogen atoms,^37–39 resulting in lattice expansion on the surface of TiO₂. This increased bond length would energetically favour the generation of oxygen vacancies from TiO₂ due to a lower bonding energy.⁴⁰ Then the oxygen deficiencies transfers their extra two electrons to the adjacent two Ti⁴⁺ atoms to form Ti³⁺.⁴¹ Hence, Ti³⁺ defects appear and the proportion of these defects is increased for TiO₂ with increased amounts of lattice distortion. Finally, the formation of Ti³⁺ defects for TiO₂ with lattice distortion leads to the enhancement of photocurrent generation under visible light while also boosting the photocatalytic activity of TiO₂.

Fig. 4 Schematic illustrating how lattice distortion contributes to the introduction of oxygen vacancies.

Fig. 5a displays the UV-vis diffuse reflectance spectra and the color of the different TiO₂ samples. The color changes from white to yellow, and a noticeable difference in the absorption of visible light is observed when Ti³⁺ defects exist. Valence band (VB) XPS spectra (Fig. 5b) aided in determining the band position for the different TiO₂ samples. The pure TiO₂ shows a VB maximum energy of 3.10 eV below the Fermi level. After Ti³⁺ defects were introduced,^43–45 an obvious upward shift by 0.36 eV occurs in the VB while lattice distortion only brings a slight upward shift by 0.14 eV. When both lattice distortion and defects exist, the VB top shifts to 2.64 eV with a 0.46 eV shift from the original TiO₂. After introducing defects, a VB tail with a difference of 0.38 eV and 0.46 eV appears in TiO₂ with and without lattice distortion, respectively. The corresponding band gaps were calculated⁴² to be 3.06 eV for the TiO₂ with lattice distortion, similar to the band gap for the original TiO₂ The band gap was found to be 2.98 eV after Ti³⁺ defects were introduced (Fig. S6†). It is thus deduced that the Ti³⁺ defects (rather than lattice distortion) lead to the extension of the visible light absorption. The VB edge of TiO₂ with Ti³⁺ defects results in a narrower band gap and higher visible light activity when compared to pristine TiO₂, which is in line with the results of Pan et al.²⁸ A schematic illustration of the energy band structure for the different TiO₂ samples is presented in (Fig. 5c). Due to the contribution of VB tails, the band gap is narrowed to 2.60 eV and further to 2.52 eV as the proportion of Ti³⁺ defects increases. Therefore, lattice distortion has little influence on the visible light photoreactivity of TiO₂, while it benefits the migration and recombination of electron–hole pairs, and both Ti³⁺ defects and lattice distortion, in turn, greatly improve the visible light photoreactivity of TiO₂.

Fig. 5 (a) UV-vis diffuse reflectance spectra. (b) Valence band XPS spectra of the different TiO₂ samples. (c) Schematic illustration of the energy band structure for the different TiO₂ samples.

Conclusions

In conclusion, lattice distortion was introduced to TiO₂ microspheres by introducing chlorine during the hydrothermal processing while defects (Ti³⁺) were introduced by reduction with ethylene glycol. Our results demonstrate the relationship between lattice distortion and Ti³⁺ defects for anatase, which were strongly correlated with their photocatalytic properties. The existence of Ti³⁺ defects was found to strengthen the absorption of light in the UV and visible light regions and further elevates the VB position, leading to overall improved photocatalytic activity towards the degradation of RhB and phenol under visible light irradiation. Our results also demonstrate that defects and lattice distortion enhance the photocurrent generation of anatase TiO₂ under visible light. Ti³⁺ defects are energetically favored over lattice distortion in TiO₂ compared to that in pristine anatase TiO₂. This study sheds light on the mechanism for engineering materials with lattice disorder and opens new opportunities for the design and synthesis of high-performance photocatalysts and solar fuel generation devices that operate under visible light irradiation.

Acknowledgements

This research is financially supported by the Fundamental Research Funds for the Central Universities (No. 106112015CDJXY130010) and the National Natural Science Foundation of China (No. 51302329). We gratefully acknowledge the theoretical calculations processed by Dr Wen Zeng. The preparation of this manuscript was also supported by the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility. ZDH gratefully acknowledges a graduate fellowship from the National Science Foundation under Grant No. DGE-1148903 and the Georgia Tech-ORNL Fellowship.

Notes and references

1. J. H. Carey, J. Lawrence and H. M. Tosine, Photodechlorination of PCB's in the Presence of Titanium Dioxide in Aqueous Suspensions, Bull. Environ. Contam. Toxicol., 1976, 16, 697–701.
2. X. Pan, M. Q. Yang and X. Fu, Defective TiO₂ with Oxygen Vacancies: Synthesis, Properties and Photocatalytic Applications, Nanoscale, 2013, 5, 3601–3614.
3. A. L. Linsebigler, G. Lu and J. T. Yates Jr, Photocatalysis on TiO₂ Surfaces: Principles, Mechanisms, and Selected Results, Chem. Rev., 1995, 95, 735–758.
4. M. Kapilashrami, Y. Zhang and Y. S. Liu, et al., Probing the Optical Property and Electronic Structure of TiO₂ Nanomaterials for Renewable Energy Applications, Chem. Rev., 2014, 114, 9662–9707.
5. W. J. Ong, L. L. Tan and S. P. Chai, Self-assembly of Nitrogen-doped TiO₂ with Exposed {001} Facets on a Graphene Scaffold as Photo-active Hybrid Nanostructures for Reduction of Carbon Dioxide to Methane, Nano Res., 2014, 7, 1528–1547.
6. J. C. Wu, D. S. Liu and A. N. Ko, Dehydrogenation of Ethylbenzene over TiO₂–Fe₂O₃ and ZrO₂–Fe₂O₃ Mixed Oxide Catalysts, Catal. Lett., 1993, 20, 191–201.
7. K. Vinodgopal, I. Bedja and P. V. Kamat, Nanostructured Semiconductor Films for Photocatalysis. Photoelectrochemical Behavior of SnO₂/TiO₂ Composite Systems and its Role in Photocatalytic Degradation of a Textile Azo Dye, Chem. Mater., 1996, 8, 2180–2187.
8. R. Sasikala, A. Shirole and V. Sudarsan, et al., Highly Dispersed Phase of SnO₂ on TiO₂ Nanoparticles Synthesized by Polyol-mediated Route: Photocatalytic Activity for Hydrogen Generation, Int. J. Hydrogen Energy, 2009, 34, 3621–3630.
9. C. Yu, G. Li and S. Kumar, Stable Au25 (SR) 18/TiO₂ Composite Nanostructure with Enhanced Visible Light Photocatalytic Activity, J. Phys. Chem. Lett., 2013, 4, 2847–2852.
10. H. Sun, S. Wang and H. M. Ang, et al., Halogen Element Modified Titanium Dioxide for Visible Light Photocatalysis., Chem. Eng. J., 2010, 162, 437–447.
11. X. Liu, P. Lv and G. Yao, Effection of Chlorine-doped TiO₂ Photocatalysts for Photocatalytic Activity, Asian J. Chem., 2013, 25, 3275.
12. R. Yuan, T. Chen and E. Fei, Surface Chlorination of TiO₂-based Photocatalysts: A Way to Remarkably Improve Photocatalytic Activity in Both UV and Visible Region, ACS Catal., 2011, 1, 200–206.
13. S. Horikoshi, T. Miura and M. A. Kajitani, FT-IR (DRIFT) Study of the Influence of Halogen Substituents on the TiO₂-assisted Photooxidation of Phenol and p-halophenols under Weak Room Light Irradiance, J. Photochem. Photobiol., A, 2008, 194, 189–199.
14. W. Choi, A. Termin and M. R. Hoffmann, The Role of Metal Ion Dopants in Quantum-sized TiO₂: Correlation between Photoreactivity and Charge Carrier Recombination Dynamics, J. Phys. Chem., 1994, 98, 13669–13679.
15. I. Y. Jeon, H. J. Choi and M. Choi, Facile, Scalable Synthesis of Edge-halogenated Graphene Nanoplatelets as Efficient Metal-free Electrocatalysts for Oxygen Reduction Reaction, Sci. Rep., 2013, 3, 1810.
16. J. Wang and Z. Lin, Dye-sensitized TiO₂ Nanotube Solar Cells with Markedly Enhanced Performance via Rational Surface Engineering, Chem. Mater., 2009, 22, 579–584.
17. N. Guijarro, T. Lana-Villarreal and T. Lutz, et al., Sensitization of TiO₂ with PbSe Quantum Dots by SILAR: How Mercaptophenol Improves Charge separation, J. Phys. Chem. Lett., 2012, 3, 3367–3372.
18. W. T. Sun, Y. Yu and H. Y. Pan, et al., CdS Quantum Dots Sensitized TiO₂ Nanotube-array Photoelectrodes, J. Am. Chem. Soc., 2008, 130, 1124–1125.
19. F. Zuo, L. Wang and T. Wu, Self-doped Ti³⁺ Enhanced Photocatalyst for Hydrogen Production under Visible Light, J. Am. Chem. Soc., 2010, 132, 11856–11857.
20. J. Huo, Y. Hu and H. Jiang, In Situ Surface Hydrogenation Synthesis of Ti³⁺ Self-doped TiO₂ with Enhanced Visible Light Photoactivity, Nanoscale, 2014, 6, 9078–9084.
21. S. Wang, L. Pan and J. J. Song, et al., Titanium-defected undoped anatase TiO₂ with p-type conductivity, room-temperature ferromagnetism, and remarkable photocatalytic performance, J. Am. Chem. Soc., 2015, 137, 2975–2983.
22. Y. H. Hu, A Highly Efficient Photocatalyst—Hydrogenated Black TiO₂ for the Photocatalytic Splitting of Water, Angew. Chem., Int. Ed., 2012, 51, 12410–12412.
23. L. Pan, S. Wang and J. Xie, et al., Constructing TiO₂ pn homojunction for photoelectrochemical and photocatalytic hydrogen generation, Nano Energy, 2016, 28, 296–303.
24. J. Cai, K. Lv and J. Sun, et al., Ti powder-assisted synthesis of Ti³⁺ self-doped TiO₂ nanosheets with enhanced visible-light photoactivity, RSC Adv., 2014, 4, 19588–19593.
25. X. Chen, L. Liu and F. Huang, Black Titanium Dioxide (TiO₂) Nanomaterials, Chem. Soc. Rev., 2015, 44, 1861–1885.
26. A. L. Linsebigler, G. Lu and J. T. Yates Jr, Photocatalysis on TiO₂ Surfaces: Principles, Mechanisms, and Selected Results, Chem. Rev., 1995, 95, 735–758.
27. L. R. Grabstanowicz, S. Gao and T. Li, et al., Facile oxidative conversion of TiH₂ to high-concentration Ti³⁺-self-doped rutile TiO₂ with visible-light photoactivity, Inorg. Chem., 2013, 52, 3884–3890.
28. L. Pan, S. Wang and J. J. Zou, et al., Ti³⁺-defected and V-doped TiO₂ quantum dots loaded on MCM-41, Chem. Commun., 2014, 50, 988–990.
29. D. Kim, J.-h. Yang and J. Hong, Ferromagnetism induced by Zn vacancy defect and lattice distortion in ZnO, J. Appl. Phys., 2009, 106, 013908.
30. I. M. Lifshitz and A. M. Kosevich, The dynamics of a crystal lattice with defects, Rep. Prog. Phys., 1966, 29(1), 217–254.
31. M. Trau, N. Yao and E. Kim, et al., Microscopic patterning of orientated mesoscopic silica through guided growth, Nature, 1997, 390, 674–676.
32. E. Andrews and S. M. Larson, Effect of surfactant layers on the size changes of aerosol particles as a function of relative humidity, Environ. Sci. Technol., 1993, 27, 857–865.
33. X. Chen, L. Liu and Y. Y. Peter, Increasing Solar Absorption for Photocatalysis with Black Hydrogenated Titanium Dioxide Nanocrystals, Science, 2011, 331, 746–750.
34. M. S. Lazarus and T. K. Sham, X-ray Photoelectron Spectroscopy (XPS) Studies of Hydrogen Reduced Rutile (TiO_2−x) Surfaces, Chem. Phys. Lett., 1982, 92, 670–674.
35. G. Liu, H. G. Yang and X. Wang, Enhanced Photoactivity of Oxygen-deficient Anatase TiO₂ Sheets with Dominant {001} Facets, J. Phys. Chem. C, 2009, 113, 21784–21788.
36. C. Di Valentin, G. Pacchioni and A. Selloni, Reduced and n-type doped TiO₂: nature of Ti³⁺ species, J. Phys. Chem. C, 2009, 113, 20543–20552.
37. T. R. Gordon, M. Cargnello and T. Paik, Nonaqueous Synthesis of TiO₂ Nanocrystals Using TiF₄ to Engineer Morphology, Oxygen Vacancy Concentration, and Photocatalytic Activity, J. Am. Chem. Soc., 2012, 134, 6751–6761.
38. H. Tan, A. Jain and O. Voznyy, et al., Efficient and stable solution-processed planar perovskite solar cells via contact passivation, Science, 2017, 355, 722–726.
39. M. Lapertot, P. Pichat and S. Parra, Photocatalytic Degradation of p-halophenols in TiO₂ Aqueous Suspensions: Halogen Effect on Removal Rate, Aromatic Intermediates and Toxicity Variations, J. Environ. Sci. Health, Part A: Environ. Sci. Eng., 2006, 41, 1009–1025.
40. L. Ye, L. Zan and L. Tian, et al., The {001} facets-dependent high photoactivity of BiOCl nanosheets, Chem. Commun., 2011, 47, 6951–6953.
41. M. Batzill, E. H. Morales and U. Diebold, Surface studies of nitrogen implanted TiO₂, Chem. Phys., 2007, 339, 36–43.
42. M. A. Butler, Photoelectrolysis and physical properties of the semiconducting electrode WO₂, J. Appl. Phys., 1977, 48, 1914.
43. S. Wang, L. Zhao and L. Bai, Enhancing photocatalytic activity of disorder-engineered C/TiO₂ and TiO₂ nanoparticles, J. Mater. Chem. A, 2014, 2, 7439–7445.
44. A. Sinhamahapatra, J. P. Jeon and J. S. Yu, A New Approach to Prepare Highly Active and Stable Black Titania for Visible Light-assisted Hydrogen Production, Energy Environ. Sci., 2015, 8, 3539–3544.
45. A. Naldoni, M. Allieta and S. Santangelo, Effect of Nature and Location of Defects on Bandgap Narrowing in Black TiO₂ Nanoparticles, J. Am. Chem. Soc., 2012, 134, 7600–7603.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7ra04885h

---