Novel fabrication of a nitrogen-doped mesoporous TiO₂-nanorod titanate heterojunction to enhance the photocatalytic degradation of dyes under visible light

Abstract

A one-pot synthesis of a mesoporous nitrogen-doped TiO₂-nanorod titanate heterojunction was developed. The reusable prepared heterogeneous photocatalyst exhibited an enhanced photocatalytic degradation of 10 mg L⁻¹ methylene blue dye after 60 min of visible light irradiation, which was attributed to the large surface area, high mesoporosity and effective separation of photo-induced charge carriers.

---

TiO₂, an inexpensive and nontoxic material, has been proposed for use as a photocatalyst in water purification-related fields. However, the intrinsically large band gap energy of TiO₂ (∼3.2 eV for anatase and 3.0 eV for rutile)¹ limits its application as a photocatalyst under visible light. Many efforts have been exerted to extend the application of TiO₂ to the visible region by doping TiO₂ with various foreign non-metallic elements such as C, F and N.^2–4

Heterogeneous photocatalysis has been proved to be a promising method for decomposition of contaminants in wastewater.⁵ However, the rapid recombination of photo-induced charge carriers (e⁻ and h⁺) limits the performance of photocatalysts. Therefore, it is essential to enhance the e⁻/h⁺ separation and charges transfer. Fabrication of heterostructure-based semiconductors is believed to overcome these problems. From the viewpoint of photocatalytic application, it is highly beneficial to couple TiO₂ with another semiconductors. Alkali-metal titanates (A₂Ti_nO_2n+1, A = Li, Na, K and n = 3–8) have been reported to exhibit advantageous characteristics for photocatalytic performance, particularly when existing as a composite with TiO₂. Several studies have been focused on synthesis of titanate–TiO₂ hybrids.^6,7 In those studies, titanates act as a precursor of titanate–anatase hybridization and they are obtained by stoichiometric reaction of TiO₂ and M₂CO₃ or M₂O (M = Na, K), followed by calcination at extremely high temperature (∼900 °C). One can also achieve it by hydrothermal treatment of TiO₂ or titanium alkoxide precursors in high alkaline media. The prepared titanates then undergo an evaporation-induced self assembly (EISA) process⁸ to obtain titanate–anatase composites. Most synthesis methods involve inactive hybrid materials under visible light.

Recently, mesoporous TiO₂ has drawn much attention due to its great performance in terms of photocatalytic activity.^9,10 To bear those in mind, herein, we report a one-pot synthesis of nitrogen-doped mesoporous TiO₂-nanorod titanate heterojunction by using EDTA-Na₂ as a template and titanium alkoxide as a precursor. Briefly, titanium(IV) isopropoxide precursor and isopropanol was added dropwise to the solution of EDTA-Na₂ under vigorous stirring, followed by aging for 24 h, drying and calcination at 400 °C. The obtained powder sample, denoted as C-TNR@N-TiO_2−x, shows a larger surface area, higher mesoporosity and highly enhanced photocatalytic activity under visible light than the pristine TiO₂.

The XRD patterns were examined to study the crystal phases of C-TNR@N-TiO_2−x and pristine TiO₂, as shown in Fig. 1a. The XRD patterns of the pristine TiO₂ indicate only the anatase phase (JCPDS no. 21-1272). After the heat treatment with EDTA-Na₂, formation of a layered titanate Na₂Ti₃O₇ (JCPDS no. 31-1329) with a monoclinic lattice (P2₁/m) was observed.¹¹ The anatase XRD peaks of C-TNR@N-TiO_2−x have a lower intensity than those of the pristine TiO₂, indicating the deterioration in its crystallinity due to the introduction of titanates and carbonaceous species onto the anatase surface in the presence of EDTA-Na₂.¹² The Raman spectra of the pristine TiO₂ shown in Fig. 1b exhibit five typical scattering peaks of anatase phase at 146 (E_g), 198 (E_g), 399 (B_1g), 515 (A_1g + B_1g) and 638 (E_g) cm⁻¹. These peaks are weakened, broadened and obscured for C-TNR@N-TiO_2−x due to the heterogeneity or disorder of surface and intrinsic defects in edge-shared TiO₆ octahedra.^13,14 Other bands at approximately 276, 376, 439, 477 and 906 cm⁻¹, which are indicative of Na₂Ti₃O₇ nanorods,^15,16 were also observed. A weak peak at 1076 cm⁻¹ is related to sodium carbonate.¹⁶ The pristine TiO₂ was made up of clusters of nano-up to micro-scaled particles well distributed on the large surface of TiO₂ crystals (Fig. 2a), which is probably attributed to agglomeration of ultrafine TiO₂ powders into larger particles.¹⁷ EDX analysis further confirms Ti and O are the most abundant elements in the pristine TiO₂ (Fig. 2c). Typical SEM image of C-TNR@N-TiO_2−x (Fig. 2b) clearly shows the formation of clusters of rod-like nanostructures on the TiO₂ surface. Their size distribution is heterogenous: about 140–362 nm in width and 0.7–5.0 μm in length. EDX analysis of the selected area (Fig. 2d) reveals the presence of sodium (Na) in the C-TNR@N-TiO_2−x in comparison with the pristine sample. The Ti/Na ratio was 1.6, indicating nanorods of titanate are mostly in the form of sodium tri-titanate (Na₂Ti₃O₇), which consists of three TiO₆ octahedra in a unit cell and a mixture of Na⁺ and H⁺ in the interlayer space.^15,18 The N₂ adsorption–desorption isotherms and pore size distribution of C-TNR@N-TiO_2−x and pristine TiO₂ are shown in Fig. S1 (ESI†). Both isotherms are type IV adsorption isotherms with hysteresis loops in the relative range of 0.4–1.0, which is indicative of mesoporous structures.¹⁹ The BET area and total pore volume were estimated about 29 m² g⁻¹ and 0.02 cm³ g⁻¹ (pristine TiO₂), and 91 m² g⁻¹ and 0.27 cm³ g⁻¹ (C-TNR@N-TiO_2−x), respectively. The presence of EDTA-Na₂ during sol–gel preparation was responsible for the increase of surface area because the evaporation of dispersed EDTA-Na₂ at high temperature resulted in the formation of porous structure. The average pore size of C-TNR@N-TiO_2−x is 20.4 nm, which is much larger than that of pristine TiO₂ (3.6 nm). The high surface area and porous structure thus benefit the MB adsorption.

Fig. 1 (a) XRD patterns and (b) Raman spectra of C-TNR@N-TiO_2−x and pristine TiO₂.

Fig. 2 FE-SEM and EDX images of the pristine TiO₂ (a, c) and C-TNR@N-TiO_2−x (b, d), respectively.

The compositions and chemical states of elements in C-TNR@N-TiO_2−x and pristine TiO₂ were examined by XPS (Fig. S2, ESI†). In the pristine TiO₂, the XPS spectra of Ti 2p show the contribution of a peak doublet at 459.4 eV and 464.8 eV, which are the typical Ti 2p_3/2 and Ti 2p_1/2 peaks of pure TiO₂, respectively.²⁰ For C-TNR@N-TiO_2−x, a large shift of about 1.5 eV towards lower binding energies was observed for two typical Ti 2p peaks due to the coexistence of Ti⁴⁺ (459.4 eV) and Ti³⁺ (457.9 eV) in which Ti³⁺ chemical state was dominant. Ti³⁺ was reported to have greater photocatalytic activity than Ti⁴⁺. Curve fitting of O 1s spectra reveals four peaks at 529.8 eV, 531.2 eV, 532.0 eV, and 533.6 eV, which are assigned to Ti–O–Ti (lattice O), C=O (oxygen bound species) or oxygen vacancies, Ti–O–N linkage (interstitial N) and NO_x species, respectively.^19,21–24 The C 1s spectra can be deconvoluted into three components, including adventitious C (284.6 eV), oxygen bound species (C–O, 286.4 eV) and carbonyl groups (C=O, 288.6 eV), respectively. The two laters mostly correspond to carbonate species generated during the calcination.²⁵ The formation of carbonate species could facilitate the charge transfer under visible light irradiation.²⁶ When it comes to N 1s spectra, several chemical states of N dopants into TiO₂ lattice were observed at 399.8 eV, 398.7 eV and 404.8 eV, which correspond to interstitial N (Ti–N–O), substitutional N (N–Ti–O) and surface-adsorbed nitrogen species (NO_x).²⁷ The variety of nitrogenated species can be attributed to surface defects resulted from the decomposition of Ti–EDTA complex. So, EDTA-Na₂ functioned not only as a soft template but also as a nitrogen source.

The UV-Vis spectra were analyzed to study the optical properties of the as-prepared sample and pristine material, as shown in Fig. S3 (ESI†). While no significant absorption was observed in visible range for the pristine TiO₂, C-TNR@N-TiO_2−x sample is optically absorbing in the visible range of 400–600 nm. The extended response to visible range of C-TNR@N-TiO_2−x is due to the effect of nitrogen dopants and Ti³⁺. The visible light absorption below 550 nm is related to the effect of nitrogen dopants, while the absorption above 550 nm corresponds to Ti³⁺. Ti³⁺ and nitrogen dopants can narrow the band gap of the hybrid by introducing impurity levels below the conduction band and above the valence band of TiO₂, respectively.²⁸ Since there is negligible absorption above 550 nm, the enhanced visible light harvesting ability is mostly attributed to the introduction of N dopants rather than Ti³⁺. Fig. S3† reveals the color change from white to grey after doping.

The separation of photo-induced charge carriers was also investigated by photoluminescence (PL) spectroscopy (Fig. 3a). The main emission peak for the pristine TiO₂ centered at 387 nm, which can be ascribed to e⁻/h⁺ recombination in the bandgap of anatase. After the titanate–anatase hybridization, it can be observed that C-TNR@N-TiO_2−x has significantly reduced PL intensity in the visible range compared to the pristine TiO₂, indicating an effective separation of photogenerated e⁻/h⁺ due to formation of heterostructures.

Fig. 3 (a) PL spectra of pristine and C-TNR@N-TiO_2−x samples with an excitation wavelength of 325 nm. (b) Photocatalytic degradation of 20 mL of 10 mg L⁻¹ MB⁻¹ under visible light in 2 h by using 20 mg of catalysts. Blank sample (without any catalysts) as comparison.

The photocatalytic activity of the pristine TiO₂ and C-TNR@N-TiO_2−x was investigated via the photodegradation of MB in aqueous solution under visible light irradiation at room temperature. Prior to the irradiation with visible light, the adsorption of all samples was conducted in the dark condition for 30 min. After the adsorption equilibrium, the photocatalytic activity of the samples was tested by examining the degradation of MB under visible light irradiation. In the absence of photocatalysts, there is not much reduction in concentration of MB as a function of time, indicating the negligible degradation of MB under visible light. The same situation was also observed for the pristine TiO₂. The negligible degradation of MB under visible light is ascribed to the photolysis of MB. Compared to pristine TiO₂, C-TNR@N-TiO_2−x exhibits a significantly enhanced photoactivity for abating MB in aqueous solution. About 98% of MB was removed from solution after 1 h visible light irradiation (Fig. 3b).

The reusability of C-TNR@N-TiO_2−x was also tested. After each reaction, the catalyst was collected by centrifugation, and washed with ethanol followed by water, subsequently dried at 105 °C for the next photocatalytic tests. The removal efficiency of 91% was achieved after 4 cycles of reuse (Fig. S4†), indicating reusability feasibility and photo-stability of C-TNR@N-TiO_2−x.

The MB photodegradation by C-TNR@N-TiO_2−x was conducted in the presence of scavengers to elucidate the photocatalytic mechanism. 10.0 mM t-BuOH (˙OH scavenger), BQ (˙O₂⁻ scavenger) and EDTA-Na₂ (h⁺ scavenger) were added into the reaction system. The obtained results were shown in Fig. S5.† h⁺ and ˙OH scavengers showed a little suppression while ˙O₂⁻ scavenger showed great suppression on the photocatalytic activity of C-TNR@N-TiO_2−x, indicating ˙O₂⁻ is the most active radical species while ˙OH and h⁺ are less active for the MB photodegradation.

The proposed mechanism of the photodegradation of MB under visible light irradiation is illustrated in Scheme 1. The enhanced photocatalytic activity of modified samples can be attributed to various possible factors. Firstly, the unique mesoporous structure and large surface area could have significantly contributed to the MB photodegradation on the catalyst surface. Effective adsorption of pollutants on the doped TiO₂ can be a prerequisite for improving the photodegradation of pollutants. Moreover, the porous structure also facilitates the photo-induced charge transfer and charge separation. The incorporation of nitrogen and Ti³⁺ into the hybrid lattice is another reason for the improved photocatalytic activity. N and Ti³⁺ doping can modify both the electronic properties and surface structure of anatase and titanate by introducing impurity energy levels above the VBs and below the CBs, respectively. Ti³⁺ and nitrogen doping levels can also act as trap centers for electrons and holes, respectively, which substantially prevents the charge recombination. Meanwhile, carbonaceous species deposited on the surface of C-TNR@N-TiO_2−x can act as a photosensitizer which is capable of inserting electrons into the CB of TiO₂ and promoting the photocatalytic activity of C-TNR@N-TiO_2−x under visible light irradiation. Lastly, the co-existence of anatase and titanate plays an important role for the photocatalytic activity of the hybrid. Under visible light irradiation, excited electrons in N-doped anatase and titanate heterostructure of C-TNR@N-TiO_2−x can easily shoot off from the N 2p levels to the CB of anatase and Na₂Ti₃O₇, leaving holes in the N 2p states. Due to lower position of the CB of Na₂Ti₃O₇ compared to that of anatase, photo-induced electrons on TiO₂ will migrate from the CB of TiO₂ to that of Na₂Ti₃O₇. These electrons are further transferred to the surface to react with surface-adsorbed oxygen, producing O₂˙⁻ and OH˙ radicals that rapidly attack MB molecules. The synergistic effect of these two phases can cause an effective electron/hole separation, and thus reduce the charge recombination. Those aforementioned factors would result in the fast degradation of MB under visible light.

Scheme 1 The proposed photocatalytic mechanism of mesoporous C-TNR@N-TiO_2−x catalyst.

In summary, a novel nitrogen-doped mesoporous anatase–nanorod titanate composite was successfully synthesized by an one-pot hybridization using EDTA-Na₂ as a template. In comparison with the pristine TiO₂, the prepared heterogeneous photocatalyst exhibited an effectively rapid degradation (98%) of 10 mg L⁻¹ MB⁻¹ after 60 min of visible light irradiation. Its enhanced photocatalytic activity was attributed to high mesoporosity, incorporation of Ti³⁺ and N dopants, and effective electron/hole separation.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Ministry of Science, ICT and Future Planning (2013R1A2A2A03013138).

Notes and references

1. X. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891–2959.
2. Y. Lin, C. Weng, Y. Lin, C. Shiesh and F. Chen, Sep. Purif. Technol., 2013, 116, 114–123.
3. Y. Yu, H. Wu, B. Zhu, S. Wang, W. Huang, S. Wu and S. Zhang, Catal. Lett., 2008, 121, 165–171.
4. J. Senthilnathan and L. Philip, Chem. Eng. J., 2010, 161, 83–92.
5. V. Trevisan, A. Olivo, F. Pinna, M. Signoretto, F. Vindigni, G. Cerrato and C. L. Bianchi, Appl. Catal., B, 2014, 160–161, 152–160.
6. Z. Xiong and X. S. Zhao, J. Am. Chem. Soc., 2012, 134, 5754–5757.
7. H. Zhu, X. Gao, Y. Lan, D. Song, J. Xi and J. Zhao, J. Am. Chem. Soc., 2004, 126, 8380–8381.
8. C. J. Brinker, Y. Lu, A. Sellinger and H. Fan, Adv. Mater., 1999, 11, 579–585.
9. S. Zhan, Y. Yang, X. Gao, H. Yu, S. Yang, D. Zhu and Y. Li, Catal. Today, 2014, 225, 10–17.
10. F. He, J. Li, T. Li and G. Li, Chem. Eng. J., 2014, 237, 312–321.
11. A. L. Sauvet, A. L. Sauvet, S. Baliteau, C. Lopez and P. Fabry, J. Solid State Chem., 2004, 177, 4508–4515.
12. P. Zhang, C. Shao, Z. Zhang, M. Zhang, J. Mu, Z. Guo and Y. Liu, Nanoscale, 2011, 3, 2943–2949.
13. X. Chen, L. Liu, P. Y. Yu and S. S. Mao, Science, 2011, 331, 746–750.
14. S. Sahoo, A. K. Arora and V. Sridharan, J. Phys. Chem. C, 2009, 113, 16927–16933.
15. Y. V. Kolen'ko, K. A. Kovnir, A. I. Gavrilov, A. V. Garshev, J. Frantti, O. I. Lebedev, B. R. Churagulov, G. Van Tendeloo and M. Yoshimura, J. Phys. Chem. B, 2006, 110, 4030–4038.
16. Z. Zhang, J. B. M. Goodall, S. Brown, L. Karlsson, R. J. H. Clark, J. L. Hutchison, I. U. Rehman and J. A. Darr, Dalton Trans., 2010, 39, 711–714.
17. E. Vinodkumar, M. Georg, K. S. Michael, J. H. Steven and C. P. Suresh, ACS Appl. Mater. Interfaces, 2013, 5, 1663–1672.
18. K. Kunlanan, I. Akihide, S. Jason and A. Rose, Chem. Eng. Sci., 2013, 93, 341–349.
19. M. T. Nguyen-Le and B. K. Lee, Chem. Eng. J., 2015, 281, 20–33.
20. S. P. Chenakin, G. Melaet, R. Szukiewicz and N. Kruse, J. Catal., 2014, 312, 1–11.
21. S. H. Lee, E. Yamasue, K. N. Ishihara and H. Okumura, Appl. Catal., B, 2010, 93, 217–226.
22. B. Qiu, Y. Zhou, Y. Ma, X. Yang, W. Sheng, M. Xing and J. Zhang, Sci. Rep., 2015, 5, 8591.
23. R. Jaiswal, N. Patel, D. C. Kothari and A. Miotello, Appl. Catal., B, 2012, 126, 47–54.
24. Z. Luo, A. S. Poyraz, C. H. Kuo, R. Miao, Y. Meng, S. Y. Chen, T. Jiang, C. Wenos and S. L. Suib, Chem. Mater., 2015, 27, 6–17.
25. W. Ren, Z. Ai, F. Jia, L. Zhang, X. Fan and Z. Zou, Appl. Catal., B, 2007, 69, 138–144.
26. L. Zhao, X. Chen, X. Wang, Y. Zhang, W. Wei, Y. Sun, M. Antonietti and M. M. Titirici, Adv. Mater., 2010, 22, 3317–3321.
27. O. Rosseler, M. Sleiman, V. N. Montesinos, A. Shavorskiy, V. Keller, N. Keller, M. I. Litter, H. Bluhm, M. Salmeron and H. Destaillats, J. Phys. Chem. Lett., 2013, 4, 536–541.
28. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.

---

Footnote

† Electronic supplementary information (ESI) available experimental section, nitrogen adsorption–desorption isotherms and pore size distribution, XPS spectra, UV-Vis spectra for pristine TiO₂ and C-TNR@N-TiO_2−x. See DOI: 10.1039/c6ra02826h

---