Synthesis and hydrogenation of anatase TiO₂ microspheres composed of porous single crystals for significantly improved photocatalytic activity

Abstract

Anatase microspheres composed of porous single crystals were successfully synthesized via a facile route without preseeding treatment and then further modified using a surface hydrogenation process. The porous materials obtained exhibit excellent photocatalytic activity and good recyclability leading to great potential in practical applications.

---

Anatase TiO₂ is one of the most widely used semiconductors in the field of photocatalysis. Since Carey et al.¹ and Frank and Bard² used a TiO₂ suspension to degrade polychlorobiphenyls (PCBs) and cyanideion in 1976, TiO₂ photocatalysts with nanocrystal morphology have been focused on due to their huge surface area.^3–5 Yet nanocrystals usually form aggregates which are often fragile and unstable under continuous physical impact such as stirring⁶ (Fig. 1a). In 2008, Yang et al.⁷ synthesized large, well-shaped anatase bulk single crystals (BSCs) using TiF₄ as a Ti source and hydrofluoric acid (HF) as a morphology controlling agent. A large size induces improved recyclability in the photocatalytic degradation of aqueous organic pollutants, however the surface area decreases (Fig. 1b). Obviously, the merits of nanocrystals and BSCs are always conflicting each other. One solution is to synthesize a large hierarchical structure, for example, microspheres consisting of 1D or 2D nanostructures;⁸ another effective solution is to make a single crystal with a porous architecture (Fig. 1c), which was exactly achieved by Crossland et al. recently.⁹

Fig. 1 Schematic diagram of the design and modification of an anatase TiO₂ microsphere composed of porous single crystals.

For this, Crossland et al. used close-packed silica nanospheres as hard templates with preseeding in an extremely dilute TiCl₄ solution to grow comparably large anatase crystals throughout the template. The seeding procedure was considered to be crucial because it could eliminate the high energy step in the crystal nucleation-growth sequence.^9,10 However, Zheng et al.¹¹ found that anatase porous single crystals (PSCs) could be synthesized without requiring the time-consuming preseeding treatment when they selected TiCl₄ as the precursor. This finding opens a new route of using appropriate precursors to simplify the synthesis process of anatase PSCs. Nevertheless, elaborate control of the concentration of HF and HCl is often required¹¹ which increases the complexity of the experiments in another way.

Surface hydrogenation has been confirmed as an effective way to improve the photocatalytic performance of TiO₂.¹² There are four major preparation methods: (i) high pressure hydrogen treatment, (ii) low pressure hydrogen treatment, (iii) low pressure mixed-gas treatment, and (iv) hydrogen plasma treatment, all of which have been extensively studied and well documented in a very recent review.¹³ However, these hydrogenation methods are mainly concentrated on nanocrystals and research on the hydrogenation of PSC-based anatase TiO₂ is scarce.

In this work, we selected ammonium fluorotitanate (NH₄)₂TiF₆ as a precursor to synthesize anatase microspheres composed of PSCs (Ms-PSCs), and further investigated the hydrogenation effect of the Ms-PSCs on the photocatalytic degradation of methyl orange (MO). In this synthesis, only (NH₄)₂TiF₆ and water were used besides the silica template. Compared to previous synthesis routes for anatase PSCs,^9,11 this approach shows great advantages due to the avoidance of (i) onerous preseeding treatment, (ii) extremely corrosive and toxic HF as a morphology controlling agent,¹⁴ and (iii) the time-consuming adjustment of the HCl and HF concentration for controlling the hydrolysis and crystallization processes.¹¹ Relative to a single unit of a BSC or PSC, Ms-PSCs have a larger size which benefits their recovery (Fig. 1d); meanwhile, good electron mobility can still be maintained in each single crystal region which is particularly important for applications such as gas sensors and solar cells.¹⁰ The Ms-PSCs obtained exhibited a significantly higher photocatalytic performance than nonporous microspheres composed of bulk single crystals (Ms-BSCs); after hydrogenation, the photocatalytic performance was further improved and good recyclability was retained (Fig. 1e). To the best of our knowledge, this work represents the most convenient synthesis of PSC-based anatase TiO₂ reported to date and the first case of the hydrogenation of PSC-based TiO₂ for improving the photocatalytic activity.

The scanning electron microscopy (SEM) images of the Ms-BSCs and Ms-PSCs are comparatively given in Fig. 2. Without a silica template, the product shows individual particles of around 1.2 μm composed of several truncated bipyramidal TiO₂ single crystals as the primary units. According to the Wulff construction of anatase TiO₂ crystals,¹⁵ the surface of the microspheres is mainly covered by square {001} crystalline facets of single crystal units, which suggests that (NH₄)₂TiF₆ provides both the capping agent and the Ti source. The formation of the anatase microspheres can be attributed to the continuous nucleation and growth of secondary truncated bipyramids on the surface defects of the primary ones introduced through the erosion of HF from the hydrolysis of (NH₄)₂TiF₆, which has been proposed by Wu et al.^16–18 in their morphology regulation of multi-facet anatase spheres. When using a template, the particles display almost unchanged outlines (Fig. S1†) relative to those of the Ms-BSCs and a uniform porous structure with a pore size corresponding to the template spheres (Fig. S2 and S3†); at the same time, many small holes can be found in the bottom of the surface semi-pores (Fig. 2b) which indicate the porous structure obtained using this method is coherent. After hydrogenation, no obvious morphology change can be found in the product (hydrogenated Ms-PSCs, HMs-PSCs). In the transmission electron microscopy (TEM) photograph of a selected particle (Fig. 2c), the light and dark image further confirms the pores are evenly distributed in the sample. Selected area electron diffraction (SAED) of a building unit produces sharp Laue diffraction patterns (Fig. 2d) that are clearly indexed to anatase TiO₂ and indicate the single crystal nature of the porous building unit; the side face of the local unit investigated can be assigned to {001} facets accordingly which further confirms the SEM results. Moreover, a partial M-PSCs is also found (Fig. S4†), in which the right part is the original bulk crystal. This morphology can be attributed to partially wrapping around the silica template when the single crystal is growing, and can be used as direct evidence that crystals are generated in a manner of first forming around silica spheres and then removing the template by alkaline etching.⁹ Taken together with the SEM characterization, the above results indicate that silica indeed plays the role of templating agent and Ms-PSCs are the structural inverse of silica spheres.

Fig. 2 SEM images of Ms-BSCs (a) and Ms-PSCs (b), TEM image of a full particle of a M-PSCs (c) and the SAED pattern (d) of a building unit of the particle in (c).

The N₂ adsorption isotherms (Fig. S5†) of the Ms-PSCs before and after hydrogenation show a type IV isotherm, which is consistent with the typical porous feature demonstrated by the SEM micrographs. The desorption curves of the samples exhibit the hysteresis loop of type E which can be attributed to the inkwell-type pore structure formed by the close-packed template spheres inside the anatase crystal.¹⁹ The Brunauer–Emmett–Teller (BET) surface area of the Ms-PSCs material is obviously higher than that of the Ms-BSCs; after hydrogenation, there is no significant change in the BET value of the product (HMs-PSCs) (Table S1†). The Barrett–Joyner–Halenda (BJH) pore size of the Ms-PSCs is about 65 nm with a narrow distribution which is similar to that of the HMs-PSCs, about 61 nm (Fig. S5†). These results further confirm the above SEM and TEM observation indicating the formation of a porous structure.

The X-ray diffraction (XRD) profiles of the Ms-PSCs and HMs-PSCs exhibit a similar shape to that of the Ms-BSCs (Fig. S6†), in which the diffraction peaks are in good agreement with anatase (JCPDS card no. 21-1272) and no significant impurities can be found. The high-resolution TEM (HRTEM) images indicate that compared to the untreated Ms-PSCs with clear edge lattice fringes shown in Fig. 3a, the HMs-PSCs have an apparent disordered surface layer about 2–3 nm in thickness surrounding the lattice core (Fig. 3b). The UV-visible diffuse reflectance spectra (DRS) in Fig. 3c show that the hydrogenated samples exhibit a higher and broader absorption range than the unhydrogenated samples, extending from the ultraviolet light to visible light regions. This result is in good agreement with the color change of the photocatalysts from white TiO₂ to gray TiO₂. The Raman spectrum shows four peaks at 149.3, 397.5, 515.1 and 639.0 cm⁻¹ for the hydrogenated sample (Fig. 3d), which can be assigned to the typical anatase bands of E_g, B_1g, A_1g and E_g, respectively.²⁰ The significant broadening and shifting of the E_g peak further demonstrates that the original symmetry of the TiO₂ lattice is broken down due to the presence of surface disorder resulting from phonon confinement or nonstoichiometry as a result of oxygen vacancy doping after hydrogenation.^21–24

Fig. 3 HRTEM images of Ms-PSCs (a) and HMs-PSCs (b); UV absorption spectra (c) and Raman spectra (d) of Ms-BSCs, Ms-PSCs and HMs-PSCs.

X-ray photoelectron spectroscopy (XPS) (Fig. S7†) was used to investigate the effect of hydrogenation on the chemical binding of the Ms-PSCs. It can be seen that the Ti 2p spectra of both the Ms-PSCs and HMs-PSCs show two peaks at 458.5 and 464.4 eV corresponding to the Ti 2p3/2 and 2p1/2 peaks of the typical Ti⁴⁺ species in TiO₂.²¹ The O 1s spectra of both samples exhibit a broad O 1s peak at 529.8 eV with a shoulder at a higher energy of 531.3 eV that can be attributed to Ti–OH species and is correlated with the oxygen vacancy directly.^21,25,26 The HMs-PSCs sample shows a larger peak area at 531.3 eV than the Ms-PSCs sample does, which clearly indicates more oxygen vacancies are created through the hydrogenation treatment.

The photocatalytic activity of the samples was evaluated by degrading MO dye under UV light irradiation. Compared to non-porous microspheres, the Ms-PSCs material shows significantly improved photocatalytic activity (Fig. 4a) due to the higher surface area providing more active sites,²⁷ which, however, is slightly lower than that of commercial P25 because of the smaller average size of the P25 nanoparticles.^11,16 After hydrogenation, the photocatalytic performance of the porous product (HMs-PSCs) is further improved, the degradation ratio of MO within 60 min is 4.2 times that of the Ms-BSCs and 1.3 times that of the Ms-PSCs. This improvement indicates that the disordered layer functions together with the increased surface area in the photocatalytic reaction. Vacuum filtration experiments were designed to test the recyclability of the samples. As shown in Fig. 4b, the Ms-PSCs and HMs-PSCs samples only need about ten seconds to separate from aqueous suspension as well as the nonporous Ms-BSCs sample does due to their relatively large crystal size, while commercial P25 nano-TiO₂ needs about 320 seconds. After filtration, the Ms-PSCs and HMs-PSCs both give a recovery ratio of approximately 95% compared to 72% of P25. These results clearly show the good recyclability of these PSC-based materials in the suspension system.

Fig. 4 Photocatalytic decomposition of MO (a), recyclability (b), PL spectra (c) and photocurrent response (d) of the samples.

It is well known that photocatalytic activity is largely affected by the separation capacity of the photo-induced carriers which determine the final quantum yield²⁸ and can be evaluated from the photoluminescence (PL) emission. The lower the PL intensity, the smaller the probability that the photogenerated electron–hole pairs recombine. All the samples show a similar PL pattern shape. A strong peak at about 396 nm is attributed to the emission of the band gap transition with the energy of light approximately equal to the band gap energy of anatase. The PL signals from 450–600 nm are due to excitonic PL, which mainly results from the surface oxygen vacancies and defects of the TiO₂ samples.^29,30 The Ms-PSCs material shows an evidently lower PL intensity than the Ms-BSCs material (Fig. 4c), which can be ascribed to more active sites interacting with absorbed species and hence more separation of electrons/holes. The PL intensity of the HMs-PSCs material further decreases, which is directly related to oxygen vacancies in the newly generated disordered layers which can facilitate charge separation and transport in TiO₂.³¹ In order to further confirm the PL results, photocurrent response tests in several on–off cycles were also carried out (Fig. 4d). It can be seen that all samples display good reproducibility. The current density of the porous samples is obviously higher than that of the nonporous sample and the HMs-PSCs material exhibits the highest current density of 540 μA cm⁻². This result suggests that the effective separation of the photo-induced carriers results in an increase in donor density. Therefore, it can be concluded that both the porous architecture and surface disorder contribute to the enhancement of the photocatalytic activity.

In summary, we have developed a facile synthesis of anatase microspheres composed of porous single crystals in the absence of preseeding treatment, and examined the hydrogenation effect of porous TiO₂ on the photocatalytic activity. The Ms-PSCs material shows significantly improved photocatalytic performance over that of the nonporous bulk microspheres, and better recyclability than nanocrystals. At the same time, the hydrogenation treatment displays a favourable surface modification effect for the Ms-PSCs material on the photocatalytic activity. Overall, this work offers potential for the cost-effective and large scale production of PSC-based TiO₂ with great practicality and gives new insight into the modification of PSC-based TiO₂ with enhanced photocatalytic activity.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (41502032), the Bilateral Project of NSFC-STINT (51611130064), the Natural Science Foundation of Jiangsu Province, China (BK20140212), the Fundamental Research Funds for the Central Universities (2015XKZD01) and a collaborative project from Guangdong Provincial Key Laboratory of Mineral Physics and Materials (GLMPM-007).

Notes and references

1. J. H. Carey, J. Lawrence and H. M. Tosine, Bull. Environ. Contam. Toxicol., 1976, 16, 697–701.
2. S. N. Frank and A. J. Bard, J. Am. Chem. Soc., 1977, 99, 303–304.
3. Y. Wang, Y. He, Q. Lai and M. Fan, J. Environ. Sci., 2014, 26, 2139–2177.
4. Y. Cong, J. Zhang, F. Chen, M. Anpo and D. He, J. Phys. Chem. C, 2007, 111, 10618–10623.
5. Y. Tseng, C. Kuo, C. Huang, Y. Li, P. Chou, C. Cheng and M. Wong, Nanotechnology, 2006, 17, 2490–2497.
6. Y. Yu, C. Cao, W. Li, P. Li, J. Qu and W. Song, Nano Res., 2012, 5, 434–442.
7. H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith, H. M. Cheng and G. Q. Lu, Nature, 2008, 453, 638–641.
8. J. M. Wu, X. M. Song and Y. Ma, J. Hazard. Mater., 2011, 194, 338–344.
9. E. J. W. Crossland, N. Noel, V. Sivaram, T. Leijtens, J. A. Alexander-Webber and H. J. Snaith, Nature, 2013, 495, 215–219.
10. S. So and P. Schmuki, Angew. Chem., Int. Ed., 2013, 52, 7933–7935.
11. X. Zheng, Y. Lv, Q. Kuang, Z. Zhu, X. Long and S. Yang, Chem. Mater., 2014, 26, 5700–5709.
12. X. Chen, L. Liu, Y. Y. Peter and S. S. Mao, Science, 2011, 331, 746–750.
13. X. Chen, L. Liu and F. Huang, Chem. Soc. Rev., 2015, 44, 1861–1885.
14. D. Zhang, G. Li, X. Yang and J. C. Yu, Chem. Commun., 2009, 4381–4383.
15. W. Ong, L. Tan, S. Chai, S. Yong and A. R. Mohamed, Nanoscale, 2014, 6, 1946–2008.
16. J. M. Wu, X. M. Song, L. Y. Ma and X. D. Wei, J. Cryst. Growth, 2011, 319, 57–63.
17. J. M. Wu and M. L. Tang, J. Hazard. Mater., 2011, 190, 566–573.
18. J. M. Wu and M. L. Tang, Nanoscale, 2011, 3, 3915–3922.
19. J. Niu, S. Shen, S. He, Z. Liu, P. Feng, S. Zhang, X. Ou, Y. Qiang and Z. Zhu, Ceram. Int., 2015, 41, 11936–11944.
20. V. V. Yakovlev, G. Scarel, C. R. Aita and S. Mochizuki, Appl. Phys. Lett., 2000, 76, 1107–1109.
21. W. Zhou, W. Li, J. Wang, Y. Qu, Y. Yang, Y. Xie, K. Zhang, L. Wang, H. Fu and D. Zhao, J. Am. Chem. Soc., 2014, 136, 9280–9283.
22. G. Liu, H. G. Yang, X. Wang, L. Cheng, H. Lu, L. Wang, G. Q. Lu and H. Cheng, J. Phys. Chem. C, 2009, 113, 21784–21788.
23. X. Y. Pan and X. M. Ma, J. Solid State Chem., 2004, 177, 4098–4103.
24. T. Leshuk, R. Parviz, P. Everett, H. Krishnakumar, R. A. Varin and F. Gu, ACS Appl. Mater. Interfaces, 2013, 5, 1892–1895.
25. A. Sinhamahapatra, J. Jeon and J. Yu, Energy Environ. Sci., 2015, 8, 3539–3544.
26. W. Wang, Y. Ni, C. Lu and Z. Xu, RSC Adv., 2012, 2, 8286–8288.
27. Z. Liu, H. Ran, J. Niu, P. Feng and Y. Zhu, J. Colloid Interface Sci., 2014, 431, 187–193.
28. Z. Liu, J. Niu, P. Feng, Y. Sui and Y. Zhu, RSC Adv., 2014, 4, 43399–43405.
29. Q. Xiang, K. Lv and J. Yu, Appl. Catal., B, 2010, 96, 557–564.
30. H. Wu, J. Ma, Y. Li, C. Zhang and H. He, Appl. Catal., B, 2014, 152, 82–87.
31. Y. H. Hu, Angew. Chem., Int. Ed., 2012, 51, 12410–12412.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra12053a

---